Clinical evidence guidelines for

medical devices

Version 3.1, June 2022

 Therapeutic Goods Administration

Copyright
© Commonwealth of Australia 2022
This work is copyright. You may reproduce the whole or part of this work in unaltered form for your own personal use or, if
you are part of an organisation, for internal use within your organisation, but only if you or your organisation do not use the
reproduction for any commercial purpose and retain this copyright notice and all disclaimer notices as part of that
reproduction. Apart from rights to use as permitted by the Copyright Act 1968 or allowed by this copyright notice, all other
rights are reserved and you are not allowed to reproduce the whole or any part of this work in any way (electronic or
otherwise) without first being given specific written permission from the Commonwealth to do so. Requests and inquiries
concerning reproduction and rights are to be sent to the TGA Copyright Officer, Therapeutic Goods Administration, PO Box
100, Woden ACT 2606 or emailed to <[email protected]>.

Clinical evidence guidelines: Medical devices Page 2 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Contents
About these guidelines __________________________ 6
International activities and alignment _____________________________________ 6
Part 1 – General requirements ___________________ 7
Legislative framework ________________________________________________________ 7
Therapeutic Goods Act --------------------------------------------------------------------- 7
Medical Device Regulations --------------------------------------------------------------- 8
The Essential Principles ______________________________________________________ 9
Summary of the Essential Principles --------------------------------------------------- 9
Demonstrating compliance with the Essential Principles ---------------------- 10
Compliance with standards ------------------------------------------------------------- 15
Clinical evidence requirements ____________________________________________ 16
Key definitions and concepts ----------------------------------------------------------- 17
Why clinical evidence is required ----------------------------------------------------- 19
Who is responsible for providing clinical evidence ------------------------------ 19
When to submit clinical evidence ----------------------------------------------------- 19
Requirements for different device classifications -------------------------------- 19
Direct and indirect evidence------------------------------------------------------------ 20
When there is no or limited clinical data -------------------------------------------- 20
Evidence strategies for different device types ------------------------------------- 20
Sources of clinical data ______________________________________________________ 21
Clinical investigations -------------------------------------------------------------------- 21
Literature review -------------------------------------------------------------------------- 24
Post-Market Data -------------------------------------------------------------------------- 26
Other clinical experience data --------------------------------------------------------- 28
Clinical evaluation ___________________________________________________________ 29
Appraisal of the clinical data ----------------------------------------------------------- 29
Analysis of the clinical data ------------------------------------------------------------- 34
The Clinical Evaluation Report (CER) _____________________________________ 35
Critical analysis and expert opinion -------------------------------------------------- 35
Competent clinical expert --------------------------------------------------------------- 35
CER content---------------------------------------------------------------------------------- 36
Supporting documents ------------------------------------------------------------------- 43
Clinical evidence checklist -------------------------------------------------------------- 46

Clinical evidence guidelines: Medical devices Page 3 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Part 2 – Special topics _________________________ 48

Comparable devices including substantially equivalent devices _____ 48
Comparable devices ----------------------------------------------------------------------- 48
Substantially equivalent devices ------------------------------------------------------ 48
Comparing device characteristics ----------------------------------------------------- 50
Clinical characteristics/intended purpose ----------------------------------------- 50
Technical characteristics ---------------------------------------------------------------- 51
Biological characteristics ---------------------------------------------------------------- 52
Presentation of information regarding comparable devices ------------------ 52
Magnetic resonance imaging (MRI) considerations ____________________ 55
Summary recommendations ----------------------------------------------------------- 56
Defining ‘safety’ in the MR environment -------------------------------------------- 56
Evidence requirements ------------------------------------------------------------------ 57
Defining active implantable medical devices -------------------------------------- 59
Summary of safety and performance data ------------------------------------------ 61
Personalised medical devices (PMDs) ____________________________________ 63
Summary recommendations ----------------------------------------------------------- 64
Compiling the CER ------------------------------------------------------------------------- 75
Defining clinical outcomes -------------------------------------------------------------- 75

Part 3 - Requirements for specific device types ____ 78

In vitro diagnostic (IVD) medical devices ________________________________ 78
Total and partial joint prostheses _________________________________________ 78
Summary recommendations ----------------------------------------------------------- 78
Defining joint prostheses ---------------------------------------------------------------- 79
Clinical evidence --------------------------------------------------------------------------- 80
Compiling the CER ------------------------------------------------------------------------- 83
Measuring clinical success -------------------------------------------------------------- 83
Summary of safety and performance data ------------------------------------------ 85
Post Market Surveillance of Joint Prostheses -------------------------------------- 89
Cardiovascular devices to promote patency or functional flow_______ 91
Summary recommendations ----------------------------------------------------------- 91
Defining CV flow implants --------------------------------------------------------------- 93
Compiling the CER ------------------------------------------------------------------------- 95
Defining clinical success ----------------------------------------------------------------- 97
Summary of safety and performance data ----------------------------------------- 105

Clinical evidence guidelines: Medical devices Page 4 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Implantable pulse generator systems ___________________________________ 113

Summary recommendations ---------------------------------------------------------- 113
Defining implantable pulse generator systems ---------------------------------- 114
Clinical evidence -------------------------------------------------------------------------- 114
Compiling the CER ------------------------------------------------------------------------ 116
Defining clinical success ---------------------------------------------------------------- 117
Summary of safety and performance data ----------------------------------------- 122
Heart valve replacements using a prosthetic valve ____________________ 126
Summary recommendations ---------------------------------------------------------- 126
Defining heart valve prostheses ------------------------------------------------------ 127
Clinical evidence -------------------------------------------------------------------------- 128
Compiling the CER ------------------------------------------------------------------------ 129
Defining clinical success ---------------------------------------------------------------- 132
Summary of safety and performance data ----------------------------------------- 134
Supportive Devices - Meshes, patches and tissue adhesives _________ 149
Summary recommendations ---------------------------------------------------------- 149
Defining supportive devices ----------------------------------------------------------- 150
Clinical evidence -------------------------------------------------------------------------- 151
Compiling the CER ------------------------------------------------------------------------ 152
Defining clinical success ---------------------------------------------------------------- 154
Summary of safety and performance data ----------------------------------------- 158
Software as Medical Device ________________________________________________ 164
Introduction -------------------------------------------------------------------------------- 164
Summary recommendations ---------------------------------------------------------- 164

Glossary and abbreviations ____________________ 164

Glossary ______________________________________________________________________ 164
Abbreviations _______________________________________________________________ 169
Source material ______________________________ 174
Search Method: Identification and selection of clinical studies ______ 174
Identified study designs ___________________________________________________ 176
Version history ______________________________________________________________ 177

Clinical evidence guidelines: Medical devices Page 5 of 178

V3.1 June 2022
 Therapeutic Goods Administration

About these guidelines

These guidelines provide details and guidance on the clinical evidence requirements for medical
devices, including in vitro diagnostic medical devices (IVDs), under Australian legislation. For
IVDs, there is also a supplementary document titled ‘Clinical evidence guidelines supplement: In
vitro diagnostic (IVD) medical devices’, which should be reviewed in conjunction with these
guidelines. The guidelines are intended to be a common reference point for both industry and
the regulator – assisting sponsors and manufacturers to collect, compile and present clinical
evidence in a manner that meets regulatory requirements, while reflecting the TGA’s approach
to how it assesses clinical evidence.
Although it is recognised that other approving regulatory bodies might be comparable, the
evaluation strategies of manufacturers intending to market a medical device in Australia should
be consistent with these guidelines.
These guidelines provide information on:
Legislative framework
This is the relevant Australian legislation for the regulation of medical devices particularly as it
relates to the Essential Principles (EPs) and clinical evidence requirements to demonstrate
compliance with the EPs.
The Essential Principles
The Essential Principles (EPs) are divided into ‘General EPs’ (EPs 1 through 6) and ‘Specific EPs’
(EPS 1 through 15). The Specific EPs include EP 14 which covers the requirement for clinical
evidence, and the impact of clinical evidence on compliance with EPs.
Clinical evidence requirements
This section addresses key concepts and approaches for different device classifications and
types, including different evidence strategies that can be used to demonstrate compliance with
the EPs. Although manufacturers may make assertions about sufficiency of clinical evidence, the
decision regarding sufficiency is taken by the TGA.
Sources of clinical data
Sources of clinical evidence, include clinical investigation studies (and a discussion of multiple
study designs), literature reviews and clinical experience data (including post-market data). The
TGA recognises a hierarchy of clinical evidence and will consider whether the level of the clinical
evidence is commensurate with the risks and benefits posed by the device.
Clinical evaluation and the Clinical Evaluation Report (CER)
This section provides an explanation of the content and format requirements of the CER and
supporting documents, including the role of critical analysis and expert opinion and a useful
checklist.
Comparable devices including substantially equivalent devices
This section gives an explanation of the steps involved in demonstrating substantial equivalence
and information on the applicability of clinical evidence from comparable devices.
Requirements for specific device types
This section provides information on the specific requirements for certain types of devices.

International activities and alignment

Australian regulatory processes are increasingly aligned with other international frameworks,
subject to Australian legislative requirements and other limitations relevant to the Australian
setting. For example, Australia takes part in the International Medical Device Regulators Forum

Clinical evidence guidelines: Medical devices Page 6 of 178

V3.1 June 2022
 Therapeutic Goods Administration

(IMDRF) which is a voluntary group of medical device regulators from around the world
working towards international medical device regulatory harmonisation.
Consequently, some approaches taken by consensus groups such as the IMDRF and by
regulators in other jurisdictions have been incorporated into these guidelines, particularly
insofar as they relate to clinical evidence, with modifications as needed for the Australian
setting. This process is ongoing. Where documentation or guidance from international settings is
inconsistent with Australian legislative requirements, it is the Australian legislative
requirements with which manufacturers and sponsors must comply. Approval by other
regulators does not guarantee approval by the TGA. On subjective matters, the TGA may adopt a
different view from other regulators.

Part 1 – General requirements

Legislative framework
The relevant Australian legislation for the regulation of medical devices is:
• Therapeutic Goods Act 1989 (the Act), particularly Chapter 4
• Therapeutic Goods (Medical Devices) Regulations 2002 (the MD Regulations).

Therapeutic Goods Act

The Act requires ongoing compliance with the EPs (which are detailed in the MD Regulations)
throughout the lifecycle of a medical device. In brief, the Act:
• requires that medical devices comply with EPs about safety and performance characteristics
(Part 4-1)
• provides for compliance with medical device standards to be one way to establish
compliance with EPs, though it is not the only way (Part 4-2)
• requires that conformity assessment procedures (or requirements comparable to
conformity assessment procedures) are applied to medical devices, and empowers the
Secretary to issue TGA conformity assessment certificates (as a means of demonstrating the
application of such procedures) having regard to compliance with the EPs, as relevant (Parts
4-2, 4-3 and 4-4)
• provides for TGA conformity assessment certificates to be subject to the condition that the
manufacturer cooperate in any review to determine whether conformity assessment
procedures having been applied, including those relating to the certification of compliance
with EPs (section 41EJ(2)(b))
• provides for the Secretary to include kinds of medical devices in the Australian Register of
Therapeutic Goods (ARTG), having regard to compliance with the EPs, as relevant (Part 4-5)
• provides that the inclusion of a kind of medical device in the ARTG, is subject to conditions
that relate to ongoing compliance with the EPs. For example, at all times while a kind of
device is included in the ARTG, the sponsor must have available sufficient information to
substantiate compliance with the EPs or have procedures in place with the manufacturer to
ensure that such information can be obtained from the manufacturer. The sponsor must give
this information to the TGA if asked to do so (section 41FN(3)(a) and (c))
• empowers the Secretary, in a broad range of contexts, to seek information relating to
compliance with the EPs (Part 4-8, section 41JA)

Clinical evidence guidelines: Medical devices Page 7 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• Includes offence and civil penalty provisions regarding non-compliance with the EPs -
importing, supplying, or exporting a medical device that does not comply with EPs without
the consent of the Secretary (Part 4-11).
Note: While some medical devices are exempt (under the MD Regulations) from the requirement
to be included in the ARTG, this does not mean that the device is exempt from the requirement
to comply with the EPs (and to apply relevant conformity assessment procedures, or have
comparable procedures applied, to the device: though see further below regarding regulation
3.11 which contains some limited exceptions in relation to the application of clinical evaluation
procedures).

Medical Device Regulations

The Medical Device (MD) Regulations detail the EPs (and the conformity assessment
procedures), and require clinical evidence and the application of clinical evaluation procedures:
• Schedule 1 sets out the EPs. EP 14 provides that every medical device requires clinical
evidence demonstrating that the device complies with the applicable provisions of the EPs.
• Schedule 3 sets out the conformity assessment procedures. In particular, Part 8 of the
Schedule sets out clinical evaluation procedures for manufacturers to obtain and evaluate
clinical data (see further below regarding the clinical evaluation procedures).
• Regulation 3.11 (subject to limited exceptions) provides that the clinical evaluation
procedures, i.e., as set out in Part 8 of Schedule 3, must also be applied to a device for the
purpose of demonstrating that the device complies with the applicable EPs, and in
particular, EPs 1, 3 and 6.
• The limited exceptions, where the clinical evaluation procedures themselves are not
required to be applied, relate to some devices exempt from inclusion in the ARTG (though
not to custom-made medical devices) , and to devices subject to an approval for special or
experimental uses or an authority under the authorised prescriber scheme (under sections
41 HB or 41HC, respectively). However, every medical device still requires clinical evidence
demonstrating compliance with applicable Eps, even where there is no requirement to apply
the clinical evaluation procedures.

Clinical evaluation procedures

The clinical evaluation procedures in Part 8 of Schedule 3 of the MD Regulations provide that the
manufacturer must obtain and evaluate clinical data in relation to the device in the form of
clinical investigation data or a literature review, or both (clause 8.3). See below for requirements
relating to clinical investigation data and literature reviews.
The manufacturer must ensure that:
• the clinical data (whether it is clinical investigation data or a literature review) is evaluated
by competent clinical experts (clause 8.6); and
• clinical evidence demonstrating that the device complies with the applicable provisions of
the EPs is documented in writing (clause 8.6).

Clinical investigation data (clause 8.4) includes:

a) documentation in relation to the design, approval, conduct and results of each investigation
carried out by the manufacturer in relation to the use of the device in or on a human body;
b) a record of qualitative or quantitative information obtained through observation,
measurement, tests, or any other means used to assess the operation of the device; and

Clinical evidence guidelines: Medical devices Page 8 of 178

V3.1 June 2022
 Therapeutic Goods Administration

c) a written report by an expert in the relevant field, being a report that contains a critical
evaluation of all the clinical investigation data held in relation to the device.
If clinical investigation data is collected in Australia, the investigation must have been conducted
in accordance with the ethical standards set out in the relevant ‘National Statement’ relating to
ethical conduct in human research published by the National Health and Medical Research
Council (NHMRC), as in force from time to time (clause 8.4(4)).
If clinical investigation data is collected outside Australia, the investigation must have been
conducted in accordance with the principles of the Declaration of Helsinki, as in force at the time
and place where the investigation was conducted (clause 8.4(5)).
See Compliance with standards for further information regarding the relevant NHMRC ethical
standards and the principles of the Declaration of Helsinki.

A literature review (clause 8.5) includes:

a) a compilation, prepared using a documented methodology, of published literature and
unpublished scientific literature, both favourable and unfavourable, relating to medical
devices of that kind, including the following:
i. expert opinion;
ii. information about the hazards and associated risks arising
from the use of the device for its intended purpose, and the
foreseeable misuse of the device;
iii. information about the performance of devices of that kind,
including a description of the techniques used to examine
whether devices of that kind achieve their intended purpose;
and
b) a written report by an expert in the relevant field, being a report that contains a critical
evaluation of the compilation of literature mentioned in paragraph (a).

The Essential Principles

A medical device must comply with the Essential Principles (EPs) in the MD Regulations, which
set out requirements relating to device safety and performance.

Summary of the Essential Principles

There are 6 general and 10 specific EPs.

General:
• Principle 1: Use not to compromise health and safety
• Principle 2: Design and construction to conform with safety principles
• Principle 3: Must perform the way the manufacturer intended
• Principle 4: Must be designed and manufactured for long-term safety
• Principle 5: Must not be adversely affected by transport or storage
• Principle 6: Benefits must outweigh any undesirable effects

Clinical evidence guidelines: Medical devices Page 9 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Specific:
• Principle 7: Chemical, physical and biological properties
• Principle 8: Infection and microbial contamination
• Principle 9: Construction and environmental properties
• Principle 10: Principles for medical devices with a measuring function
• Principle 11: Protection against radiation
• Principle 12: Medical devices connected to or equipped with an energy source
• Principle 13: Information to be provided with a medical device
• Principle 13A: Patient implant cards and patient information leaflets
• Principle 14: Clinical evidence
• Principle 15: Principles applying to IVDs only

Demonstrating compliance with the Essential Principles

Compliance with the applicable EPs is required for all devices (whether included in the ARTG or
exempt under the Regulations from inclusion). The intended purpose, risk profile, classification,
and other specific features of a device will be relevant to the type of evidence required to
substantiate compliance with the EPs.
EPs 1, 2, 3, 4, 6, 13, 13A and 14 are particularly relevant to meeting clinical evidence
requirements. Essential Principle 14 is the overarching principle, and is addressed immediately
below, followed by principles 1 through 4, 6, 13, and 13A.

Principle 14: Clinical evidence

EP 14 states that every medical device requires clinical evidence, appropriate for the use and
classification of the device, demonstrating that the device complies with the applicable
provisions of the EPs. In addition to other procedures, manufacturers must apply clinical
evaluation procedures to the medical devices they supply (regulation 3.11 of the MD
Regulations).
The way in which EP 14 is applied must take into consideration the recognised hierarchy of
clinical evidence. Where evidence is not of the highest order, particularly for high-risk devices,
robust justification should be provided.
These clinical evaluation procedures must be implemented in accordance with the requirements
specified in Part 8 of Schedule 3 of the MD Regulations. Part 8 requires the manufacturer to:
• obtain clinical data, in the form of ‘clinical investigation data’ (clause 8.4) and/or ‘literature
review’ (clause 8.5)
• ensure that the clinical data held in relation to the device is critically evaluated by competent
clinical experts in the relevant field, and that the clinical evidence demonstrating that the
device complies with the applicable provisions of the EPs is documented in writing (clause
8.6).

Regulation 3.11 of the MD Regulations, when addressing that clinical evaluation procedures
must be applied (for the purpose of demonstrating compliance with the applicable EPs) focuses
on demonstrating compliance, in particular, with EP 1, 3 and 6. In addition, all other EPs should
also be considered because of their interaction with EP 14.

Clinical evidence guidelines: Medical devices Page 10 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Thus, while it is expected that clinical evidence will primarily demonstrate that the device
complies, in particular, with EPs 1, 3 and 6, manufacturers and sponsors should also consider
other EPs as necessary. EPs 1, 2, 3, 4, 6 and 13 and 13A are addressed further in the guidelines
below.

Evidence must be a true and complete account of available scientific knowledge and the sponsor
and manufacturer must apply due diligence.

Principle 1: Use not to compromise health and safety

A medical device must be designed and produced in a way that does not compromise the clinical
condition or safety of patients, the safety and health of users or, where applicable, other persons.
Risks associated with the use of the device must be acceptable when weighed against the
intended benefit to the patient, and compatible with a high level of protection of health and
safety. Clinical evidence must be available to support this.
Key considerations from a clinical perspective include:
• The context of how the device is to be used. For example, whether it is to be used by
specialist medical practitioners only, or by the general public. This is relevant to the safety
assessment for many devices.
• How the device is used. For example, the type of treatment administered, or procedure or
testing undertaken, and any inherent dangers that have implications for the safety of the
device.
• Any inherent dangers in the proposed treatment setting should also be taken into account.
The patient, user and any other person in the vicinity of the device may need to be
considered.
• The number of patients exposed to the device and whether this sample is large enough to
ensure that all health and safety issues have been described and quantified accurately.

Principle 2: Design and construction to conform with safety principles

The design and construction of a medical device must conform with safety principles, having
regard to the generally acknowledged state of the art. This requires that any risks associated
with the use of the device are identified and minimised.
Manufacturers are required to mitigate risk to the lowest possible level. Manufacturers must
establish, implement, document and maintain a quality management system (QMS) to ensure the
ongoing safety of a medical device. Risk management is a continuous process throughout the
lifecycle of a medical device that requires regular updating.
Manufacturers should:
• establish and document a risk management plan for the device
• identify and analyse hazards and risks arising from the use or foreseeable misuse of the
device – this includes ensuring that the clinical evidence is sufficient to reliably inform the
nature and frequency of such hazards;
• ensure confidence in the quantification of hazards posing the highest level of risk through
measurement in a sufficient sample size
• eliminate, reduce or control the identified risks - for example, by design considerations
• inform users of residual risks - for example, by warnings in the Instructions for Use (IFU)

Clinical evidence guidelines: Medical devices Page 11 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• evaluate the impact of new information generated over the lifecycle of the device (for
example, previously unrecognised hazards or changes to the state of the art) on risk
acceptability and (if necessary) amend control measures accordingly
• document the risk management process - for example, by a risk management report and
failure mode & effects analysis (FMEA).
The TGA will examine risk management documents and have regard to the current state of the
art to determine whether:
• all risks and hazards have been identified in accordance with current knowledge
• the risks are eliminated or reduced as far as possible by the design and construction of the
device
• any residual risks have been mitigated to the lowest possible level through the information
provided with the device (labels, IFU, patient information leaflet, patient implant card) and
other risk mitigation strategies.

Principle 3: Must perform the way the manufacturer intended

Medical devices must achieve the performance intended by their manufacturer and be
designed and manufactured in such a way that, during intended conditions of use, they are
suitable for their intended purpose. The intended purpose(s) is determined from:
• labelling
• instructions for use
• any advertising material relating to the device and/or technical documentation
describing the mechanism of action of the device.
The TGA will examine and judge whether there is sufficient clinical evidence to demonstrate
that the device performs as intended on a case-by-case basis. Whilst a manufacturer may
make assertions about sufficiency, the TGA is the arbiter in determining sufficiency.
Performance should be measured using validated mortality and morbidity measures.
Where multiple intended uses are claimed, each use should be separately substantiated by
clinical evidence relevant to that use.
If the range of indications is broad and diverse, it may be reasonable to provide evidence of
safety and performance for the higher risk and most common indications, with a
justification as to why these were selected as ‘worst case scenarios’ and/or common
indications and how these results can be extrapolated to other indications.

Principle 4: Must be designed and manufactured for long-term safety

A medical device must be designed and manufactured in a way that ensures its characteristics
and performances (as mentioned in EP 1, 2 and 3) are not adversely affected if the device is used
within the period in which it can be safely used (as indicated by the manufacturer), is not
subjected to stresses outside of normal conditions of use, and is regularly maintained and
calibrated (in accordance with the manufacturer’s instructions).
The characteristics in EP 1 and 2 relate to health and safety. In considering EP 4 in relation to
health and safety, the design and production of the device must ensure these characteristics are
not affected to such a degree that the health or safety of the patient, user or other persons are
compromised (during the expected life of the device, when it is subjected to the stresses which
can occur during normal conditions of use).

Clinical evidence guidelines: Medical devices Page 12 of 178

V3.1 June 2022
 Therapeutic Goods Administration

EP 3 focuses on the medical device performing as intended. In considering EP 4 in relation to

performance, the TGA will have regard to the intended purpose of the device. The clinical
evidence must demonstrate that the device, as designed and produced, performs as intended for
the length of time appropriate to the intended purpose.
For some devices, it may be difficult to demonstrate performance and safety through clinical
investigations for the intended permanent duration. In such cases, the duration of follow-up in
clinical investigations should be a reasonable surrogate of the device’s intended permanent
duration. In many circumstances, this will require follow up data collection from an adequate
number of patients to prove a suitable end point. Post-market data for the device may be used to
supplement the clinical evidence on long term safety and performance, noting the well
documented issues in post-market data such as under-reporting of adverse events.
The broad framework in the Act (see above) relating to compliance with the EPs provides for
compliance to be a requirement that must be met throughout the lifecycle of devices. The
importance of continuing to obtain clinical evidence, including in the post-market environment,
is emphasised.

Principle 6: Benefits must outweigh any undesirable effects

The benefits to be gained from the use of a medical device for the performance intended by the
manufacturer must outweigh any undesirable effects arising from its use. As noted above, under
the Australian regulatory framework, medical devices must have clinical evidence that provides
assurance of safety and performance.
Typically, where a state of the art (for established device types) or the current standard of care
(for novel devices) exists, a randomised controlled trial will be expected to provide assurance of
safety and performance. In practical terms, the level of assurance required will vary according to
the risk of the device. Essentially, the greater the risk for the subject device relative to the state
of the art/standard of care, the greater the benefit that needs to be demonstrated to ultimately
demonstrate that the benefits outweigh any undesirable effects.
Where use of the device augments an existing treatment pathway, the additive risks should be
considered and informed by an adequately powered study. The treatment pathways will vary in
different geographies; the risks and benefits posed by the device should reflect the standard of
care in Australia’s health system.
It should be clearly stated if the proposed use of the device is in addition to or instead of the
current standard of care when demonstrating that the benefits outweigh the risks of use, In
doing so, consideration should be given to the type of trial that is required. Typically, where the
device is proposed for use in addition to the standard of care, a superiority trial is necessary to
demonstrate that any additive risks, however small, are balanced by a measurable benefit.

Example: A device is proposed for treatment of non-healing wounds to improve

oxygenation at the wound base. The device is applied to the wound via a spray and
is intended to be used in conjunction with the existing standard of care such as
dressings. It is claimed that, when used in conjunction with the existing standard
of care, the device is superior to the standard of care only. There are some minor
risks associated with the use of the device, as well as some risks that are difficult
to quantify. The submitted clinical evidence did not demonstrate a statistically
significant improvement in mean time to complete wound healing. As the device
introduces some risks without clear benefits, the claim of superiority was not
substantiated. The requirements of EP 6 were not satisfied.

Clinical evidence guidelines: Medical devices Page 13 of 178

V3.1 June 2022
 Therapeutic Goods Administration

For more general information on balancing considerations of benefit and risk refer to FDA,
Factors to Consider When Making Benefit-Risk Determinations in Medical Device Premarket
Approval and De Novo Classifications, 2016. Note that in the Australian context, the relevant
assessment is made against the wording of EP 6.
In developing the device (and consistent with EP 2), all possible methods to minimise hazards
identified in the risk assessment should have been incorporated into the device design and risk
mitigation strategies. Users are to be informed of residual risks. Such risks will also be
considered in the context the benefits of using the device. Manufacturers should be able to
demonstrate that any residual risks are acceptable.
Clinical investigations should be appropriately designed to provide an assessment of the benefit-
risk profile for the medical device when it is used for its intended purpose(s). A safety profile can
be established via clinical investigations, literature reviews and clinical experience (from post-
market data, adverse event data and special access use). It may also be appropriate, on occasion,
to rely on data from comparable devices to support the safety of a device.
The ISO 14971:2019 standard is recommended as a guide when making benefit-risk
determinations.

Principle 13 and 13A: Information to be provided with a medical

device/Patient implant cards and patient information leaflets
Certain information must be provided with a medical device (or certain medical devices). This
includes information about:
• the intended purpose of the device
• risks or undesirable side effects
• contraindications, warnings, restrictions or precautions
• installation, calibration and maintenance during its intended life
• instructions for use and other instructions.
The type of information required by EP 13 and 13A is critical to understanding and complying
with EPs 1, 2, 3, 4 and 6.
A clear intended purpose is required. The TGA will determine the intended purpose of the
device from documentation provided with or in relation to the device. Any
claims/statements in relation to the performance and safety of the device provided on the
labelling and/or packaging, instructions for use, patient or clinician cards, leaflets,
manuals, brochures etc., must be consistent with each other and supported by the clinical
evidence available for the device.
The patient groups for whom the device has a positive benefit-risk balance need to be well
defined. This information should explain how to insert, implant or use the device safely. It must
highlight any potential hazards, with appropriate contraindications, warnings or precautions.
Requirements for handling or storage, and any risks associated with the disposal of the device
should also be outlined. Other information provided with the device must also be consistent and
supported by the evidence. This should be reflected in the wording of the information provided.
EP 13A provides for a patient implant card and patient information leaflet to be provided with
certain medical devices. Patient information leaflets include very specific information about how
to use the device safely, side effects and risks, and the expected device lifetime (along with
information on intended purpose and performance).

Clinical evidence guidelines: Medical devices Page 14 of 178

V3.1 June 2022
 Therapeutic Goods Administration

The information provided with a medical device has an impact on the risks and therefore the
safety of the device. Unclear or ambiguous terms or statements, poor grammar and spelling,
foreign words or poor diagrams can all negatively impact on the ability of a patient or person to
safely use the device as intended and therefore negatively affect the benefit-risk profile of the
device.

Compliance with standards

Compliance with recognised standards published by an Australian or International Standards
Agency may be used to support compliance with the relevant EPs, especially for devices based
on technologies with well-established safety and performance characteristics. Compliance with
such standards is not a legislative requirement, but it is a typical part of an evidence strategy
(this can be contrasted with requirements to meet ethical standards or principles, which are
requirements in the MD Regulations, see below). Where relevant standards are not met,
explanation should be provided for why this has not occurred.
If a manufacturer chooses to use other standards or solutions (for example, an internal safety
test), this must be consistent with the state of the art, and they must demonstrate the relevance
and adequacy of this approach for supporting performance and safety requirements.
Compliance with one or more relevant standards does not equate to ongoing compliance with
safety and performance requirements contained in the EPs. A broader process for continuous
compliance is required that incorporates clinical evidence, a robust quality management system
and appropriate risk management processes.
There are three main International Standards Organization (ISO) documents relevant to general
clinical evidence requirements and ongoing compliance for medical devices:
• ISO 13485:2016 - Quality Management Systems (QMS)
• ISO 14971:2019 - Application of risk management to medical devices
• ISO 14155:2020 - Good Clinical Practice

ISO 13485:2016 Quality Management Systems

The primary objective of this standard is to facilitate harmonised medical device regulatory
requirements for quality management systems (QMS), with an emphasis on meeting national
regulatory requirements. The standard is based on the ISO 9000 family of standards that define,
establish and maintain a quality assurance system for manufacturing and service industries.

Manufacturers are expected to continue to monitor the performance and safety of devices,
including IVDs, via a surveillance program as part of their QMS once the device is marketed.
These programs should be appropriate to the use and risks of the device. Data from safety and
adverse event reports and complaints, newly identified risks, literature, any updated or new
clinical investigations, significant regulatory actions and formal surveillance activities such as
registries should be used to review the performance, safety and benefit-risk assessment of the
device. This data should be evaluated and the CER updated in line with this new information.

The CER should be updated every 1-5 years depending on the novelty of the device and risk (as
per MEDDEV 2.7/1 revision 4). As this information is incorporated into the ongoing risk
analysis, it may result in changes to the IFU and other information supplied with the device.

Compliance with ISO 13485:2016 is not mandatory in Australia. However, under the Conformity
Assessment Standards Order (Standard for Quality Management Systems and Quality Assurance
Techniques) 2019, compliance with ISO 13485:2016 is considered to satisfy the QMS
requirements specified in the legislation.

Clinical evidence guidelines: Medical devices Page 15 of 178

V3.1 June 2022
 Therapeutic Goods Administration

ISO 14155:2020 Good clinical practice

ISO 14155:2020 provides guidance on the design and conduct of clinical investigations involving
medical devices. It can also be used by regulatory bodies and ethics committees when reviewing
clinical investigational plans. Thirteen principles are included, such as adherence to ethical
principles (as per the Declaration of Helsinki), subjects’ rights, a determination that benefits
outweigh risks and oversight by an independent ethics committee.
Clinical investigation data collected in Australia is subject to the ethical standards set out in the
National Health and Medical Research Council’s (NHMRC) National Statement of Ethical Conduct
in Human Research (clause 8.4(4) of Part 8 of Schedule 3 of the MD Regulations). The
investigation must have been conducted in accordance with these ethical standards, including
that research needs to meet the requirements of ISO 14155:2020. For clinical investigation data
collected outside of Australia, clause 8.4(5) states that:
If clinical investigation data is collected outside Australia, the investigation must have been
conducted in accordance with the principles of the Declaration of Helsinki, as in force at the
time and place where the investigation was conducted.

The clinical evaluation procedures also refer to the obtaining of clinical data. The manufacturer
must ensure that the clinical data obtained takes account of any standards that may apply to the
device (clause 8.3(2) of Part 8 of Schedule 3 of the MD Regulations).

ISO 14971:2019 Application of risk management to medical devices

ISO 14971:2019 Application or risk management to medical devices specifies a process for a
manufacturer to identify the hazards associated with medical devices, including IVDs, to
estimate and evaluate the associated risks, to control these risks, and to monitor the
effectiveness of the controls. It addresses assessing benefit, and benefit-risk determinations. The
requirements of ISO 14971:2019 are applicable to all stages of the lifecycle of a medical device.

Standards applicable to specific medical devices

Examples of device types that have specific ISO standards outlining requirements for
demonstrating clinical evidence are the current editions of the series of standards:
• ISO 11979-7:2018 - Ophthalmic implants - intraocular lenses
• ISO 5840-1:2021, ISO 5840-2:2021 and ISO 5840-3:2021 - Cardiovascular implants- cardiac
valve prostheses
• ISO 14708:2020 - Implants for surgery - Active implantable medical devices
• ISO 14117:2019 - Electromagnetic compatibility test protocols for active implantable
medical devices.

There is also a technical specification ISO/TS 10974:2018 titled ‘Assessment of the safety of
magnetic resonance imaging for patients with an active implantable medical device’, which
refers to non-clinical testing of AIMDs in an MR environment.

Clinical evidence requirements

Clinical evidence comprises clinical data and its evaluation pertaining to a medical device. It
should provide the TGA with a current and accurate picture of both the state of scientific
knowledge in relation to the treatment modality to which a device relates, and in relation to the
subject device specifically. From this information, an acceptable benefit-risk profile may be
demonstrated for a medical device, by showing that it performs as intended and that all
identified undesirable effects and hazards, having been minimised during the design and
development process, are outweighed by the benefits.
Clinical evidence guidelines: Medical devices Page 16 of 178
V3.1 June 2022
 Therapeutic Goods Administration

This section outlines the critical role of clinical evidence in establishing the safety and
performance of a medical device, requirements regarding the submission of clinical evidence,
and the expectations regarding the detail and extent of evidence required for different medical
devices.

Key definitions and concepts

For the purpose of these guidelines, the terms below have the following meanings. In so far as is
possible (and subject to the Australian regulatory framework) these terms align with the IMDRF
document Clinical Evidence - Key Definitions and Concepts (IMDRF MDCE WG/N55
FINAL:2019).

Clinical investigation: Systematic investigation or study in or on one or more

human subjects, undertaken to assess the safety and/or performance of a medical
device.
Note: 'Clinical trial' or 'clinical study’ is synonymous with 'clinical investigation'
and these terms are used interchangeably in this document.
Clinical data: Safety and/or performance information that is generated from the
clinical use of a device.
Note: Under the clinical evaluation procedures in Part 8 of Schedule 3 of the MD
Regulations, the manufacturer must obtain clinical data in relation to the device in
the form of clinical investigation data, or a literature review, or both.
Clinical evaluation: A set of ongoing activities that use scientifically sound
methods for the assessment and analysis of clinical data to verify the safety
and/or performance of a medical device when used as intended by the
manufacturer.
Note: The clinical evaluation procedures (in the MD Regulations) set out
requirements in relation to the obtaining and evaluation of clinical data.
Clinical evidence: The clinical data and the clinical evaluation pertaining to a
medical device.
Note: EP 14 provides that every medical device requires clinical evidence
demonstrating that the device complies with applicable EPs.

Clinical evidence guidelines: Medical devices Page 17 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Definitions of additional terms used throughout these guidelines are as follows:

Adverse event: Any untoward medical occurrence in patients/subjects, users or

other persons. In the context of clinical investigation, for patients/subjects, this
would include all untoward medical occurrences, whether or not related to the
device that is the subject of the investigation that occurred in the course of the
investigation. In the context of clinical experience, this would only include
untoward medical occurrences that may be related to the medical device.
Clinical use: Use of a medical device in or on living human subjects.
Note: This includes use of a medical device that does not have direct patient contact.
Clinical Investigation Plan: Document that states the rationale, objectives, design
and pre-specified analyses, methodology, monitoring, conduct and record-keeping
of the clinical investigation.
Competent clinical expert: Generally expected to be someone with relevant
medical qualifications and direct clinical experience in the use of the device or
device type in a clinical setting.
Note: The clinical evaluation procedures (in the MD Regulations) require the
manufacturer to ensure that the clinical data is evaluated by competent clinical
experts.
Critical analysis: The process of the careful and systematic examination, appraisal
and evaluation of both favourable and unfavourable data.
Predicate: A previous iteration of the device, within the same lineage of devices,
with the same intended purpose and from the same manufacturer, in relation to
which a manufacturer may seek to demonstrate substantial equivalence.
Comparable device: A medical device with related function chosen by the
manufacturer to inform the clinical evaluation of the device in question.
Note: A ‘comparable device’ is distinct from a ‘comparator’, which is the state of the
art/standard of care against which a medical device may be compared (for example,
in a clinical study).
Serious Adverse Event: An adverse event that led to a death or led to a serious
deterioration in health (one that results in a life-threatening illness or injury;
results in a permanent impairment of a body structure or body function; requires
inpatient hospitalisation or prolongation of existing hospitalisation; results in
medical or surgical intervention to prevent permanent impairment to body
structure or a body function; led to foetal distress, foetal death or a congenital
abnormality/ birth defect).
Substantial equivalence: A finding that comparable devices are similar to such an
extent that there would be no clinically significant difference in safety and
performance, taking into account the intended purpose and clinical, technical and
biological characteristics of the devices.

Clinical evidence guidelines: Medical devices Page 18 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Why clinical evidence is required

All medical devices supplied in Australia must have clinical evidence sufficient to demonstrate
an appropriate level of safety and performance when used for their intended purpose(s). This is
to maintain the quality of the Australian healthcare system, and to help ensure the health of the
Australian population. Clinical evidence is an important component of the technical
documentation of a medical device which, along with other documentation (such as the device
description, labelling, risk analysis and manufacturing information), is necessary for a
manufacturer to demonstrate that the device complies with the EPs, including EP 14.

Who is responsible for providing clinical evidence

The manufacturer and sponsor must provide, or have available, clinical evidence to demonstrate
compliance of the devices with the EPs, if requested. The obligation to collect and compile
clinical evidence (typically in the form of a CER) lies with the manufacturer, who provides this to
the sponsor. The sponsor must be able to provide information to the TGA to demonstrate such
compliance. This applies to all medical devices regardless of classification.
For devices that are included in the ARTG, the legislation requires, pursuant to conditions
specified in section 41FN of the Act, that:
• the sponsor must have available sufficient information to substantiate compliance with the
EPs
OR
• have procedures in place with the manufacturer that will allow them to obtain such
information from the manufacturer.
Further, devices included in the ARTG are subject to the condition that the sponsor will, if asked
by the Secretary (or Delegate), give such information to the TGA.
For a medical device to be supplied in Australia, it must be able to be demonstrated that the
applicable EPs (as set out in Schedule 1 of the MD Regulations) be met to ensure the device is
safe and performs as intended. The Act (under Part 4-11) provides for offence and civil penalty
provisions for importing, supplying or exporting a medical device that does not comply with
applicable EPs.

When to submit clinical evidence

Clinical evidence is required to be available throughout the lifecycle of a device, though is
typically submitted to the TGA at the time of application for a conformity assessment certificate
or application for inclusion in the ARTG, and as part of post-market surveillance or reviews. It
should be evaluated and updated periodically as new information on safety and performance is
obtained from clinical studies, literature or clinical experience in relation to the subject device
and/or comparable devices.
The TGA may request and review this clinical evidence at any time. The clinical evidence
requirements described in these guidelines apply in each of these circumstances.

Requirements for different device classifications

Medical devices are classified according to their level of risk:
• Medical devices are classified under Schedule 2 of the MD Regulations from lowest to
highest risk into Classes I (which includes Im, being devices with a measuring function, and

Clinical evidence guidelines: Medical devices Page 19 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Is, being devices supplied in a sterile state), IIa, IIb, III and Active Implantable Medical
Devices (AIMDs). Refer to Medical devices overview for more information.
• IVDs are classified under Schedule 2A of the MD Regulations from lowest to highest risk
Classes 1 to 4. Refer to Classification of IVD medical devices for more information.
Some EPs do not apply to certain classes of devices in certain circumstances (for example, EP
13.4(2) in relation to when instructions for use need to be provided). Further, the EPs may
impose requirements subject to whether or not the device has a measuring function or whether
the device is intended to be supplied in a sterile state, a non-sterile state, or both. Clinical
evidence requirements must be met for applicable provisions of the EPs (noting many aspects of
the EPs are applicable to all medical devices).
Greater scrutiny will be given by the TGA to higher classification devices as part of ensuring
safety and performance. Further, the classification, design and use of the device are relevant
factors when considering the nature, type and range of evidence appropriate to being able to
demonstrate compliance with applicable provisions of the EPs. EP 14 itself notes that every
medical device requires clinical evidence, appropriate for the use and classification of the
device, demonstrating that the device complies with the applicable provisions of the EPs.

Direct and indirect evidence

The TGA recognises the following types of clinical evidence for the purpose of substantiating
compliance with the EPs:
• Direct clinical evidence - this is derived from an evaluation of clinical data pertaining to the
subject device.
• Indirect clinical evidence - this is derived from an evaluation of clinical data pertaining to a
comparable device with which substantial equivalence has been demonstrated.
Evidence from comparable devices that are not substantially equivalent may support or
supplement direct or indirect clinical evidence. However, it will not generally constitute
sufficient clinical evidence for substantiating compliance with the EPs (except for certain low
risk, well established technologies).

When there is no or limited clinical data

In some instances, it may be difficult to collect direct clinical data for a device due to very small
numbers of eligible patients, high risk procedures limiting use, or practical or ethical
considerations that limit the feasibility of conducting a high quality clinical investigation. If there
is no (or limited) clinical data for the specific device, depending upon the nature of the device,
you may be able to provide a clinical justification for why clinical evidence is either not required
or only partially required.

Evidence strategies for different device types

Evidence strategies refer to the mix of clinical evidence sources (see below regarding Sources of
clinical data) that are used, together with critical analysis and expert opinion, to demonstrate
compliance with the EPs. The following points should be noted:
• Evidence strategies will be scrutinised more for higher risk devices and for those with
greater novelty, with greater expectations around direct evidence and/or high-quality
clinical investigation data.
• The clinical evidence must distinguish the safety and efficacy of the device from that of the
procedure itself. The clinical merits of established procedures are not under assessment by
the TGA.

Clinical evidence guidelines: Medical devices Page 20 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• Where a novel device necessitates the introduction of an invasive procedure and/or

modification of an established treatment algorithm, the clinical evidence must consider the
safety of the procedure or the modification in the overall risk/benefit balance.
• New components within established device systems should consider the safety and
performance of the device system. However, new clinical investigation data involving the use
of the new component may not be required. Instead, conformance with the relevant
standards, together with a reasoned clinical argument regarding why this new component
would not adversely affect safety and performance, may be sufficient.
As stated by IMDRF, a number of factors are relevant when considering the type of evidence
required to substantiate compliance with regulatory requirements:
Clinical evaluation of medical devices that are based on existing, established technologies
and intended for an established use of the technology is most likely to rely on compliance
with recognised standards and/or literature review and/or clinical experience of
comparable devices. High risk devices, those based on technologies where there is
little or no experience, and those that extend the intended purpose of an existing
technology (i.e. a new clinical use) are most likely to require clinical investigation
data. The manufacturer will need to give consideration to the advantages and limitations
of each data type.
Possible sources of clinical data that form the basis of the clinical evidence pertaining to device
safety and performance are discussed below.

Sources of clinical data

Clinical data (meaning safety and performance information that is generated from the clinical
use of a medical device) may be generated for either the subject device or a comparable device
(including substantially equivalent devices). It includes:
• data from clinical investigations (synonymous with trials and/or studies)
• literature reviews
• post-market data
• other clinical experience data (also known as Real World Data).
The manufacturer is responsible for identifying relevant data and determining the extent of data
needed for a complete clinical evaluation (as per IMDRF MDCE WG/N56FINAL:2019 Clinical
Evaluation section 6).
The following section provides further guidance on the sources of clinical data and how they
may be used to demonstrate compliance with the EPs to establish the safety and performance of
the medical device for its intended purpose(s). As appropriate, the guidance is aligned with the
publicly available IMDRF and related MEDDEV documents (in particular, MEDDEV 2.7/1
Revision 4).

Clinical investigations
A clinical investigation is any systematic investigation or study in or on one or more human
subjects, undertaken to assess the safety, clinical performance and/or effectiveness of a medical
device. Clinical investigation data is further explained in clause 8.4 of Part 8 of Schedule 3 of the
MD Regulations. Clinical investigations are further discussed in the IMDRF MDCE
WG/N57FINAL:2019 Clinical Investigation document.

Clinical investigations include feasibility studies, studies conducted for the purpose of gaining
market approval, and those conducted following market approval (see IMDRF MDCE
Clinical evidence guidelines: Medical devices Page 21 of 178
V3.1 June 2022
 Therapeutic Goods Administration

WG/N55FINAL:2019 Clinical Evidence – Key Definitions and Concepts). Clinical investigation

data sourced directly from the device produces a higher level of confidence in its relevance and
capacity to inform the safety and performance characteristics of the device and is the preferred
option for fulfilling clinical evidence requirements.

It should be clearly indicated if the subject device has been modified since the clinical data were
gathered, to clarify the device version and the nature of the changes.

In some circumstances direct clinical investigation data are not available for the subject device
or are insufficient in quantity or quality. In this situation clinical investigation data from a
comparable device may be used to support the safety and performance of the device under
assessment (the subject device). The approach taken to determine if a comparable device is
substantially equivalent, and hence can be used as a source of indirect evidence, is described in
Comparable devices including substantially equivalent devices.
As per MEDDEV 2.7/1 revision 4, June 2016 p35-35, the manufacturer should perform a detailed
gap analysis to decide if additional clinical investigations need to be carried out:
The gap analysis should determine whether the existing data are sufficient to verify that the device
is in conformity with all the Essential Requirements (equivalent, but not identical to, the EPs)
pertaining to clinical performance and clinical safety.
Special attention should be given to aspects such as:
• new design features, including new materials
• new intended purposes, including new medical indications, new target populations (age,
gender, etc.)
• new claims the manufacturer intends to use
• new types of users (e.g. lay persons)
• seriousness of direct and/or indirect risks
• contact with mucosal membranes or invasiveness
• increasing duration of use or numbers of re-applications
• incorporation of medicinal substances
• use of animal tissues (other than in contact with intact skin)
• issues raised when medical alternatives with lower risks or more extensive benefits to patients
are available or have become newly available
• issues raised when new risks are recognised (including due to progress in medicine, science and
technology)
• whether the data of concern are amenable to evaluation through a clinical investigation.
Data on the safety and performance of other devices and alternative therapies, including
benchmark devices and equivalent devices, should be used to define the state of the art or identify
hazards due to substances and technologies. This will allow the clinical data requirements to be
established more precisely in relation to the intended purpose of a device. Precision in this analysis
and the choice of selected medical indications and target populations may reduce the amount of
clinical data needed from additional clinical investigations.
Note: the EU term ‘equivalence’ is very similar to the Australian term ‘substantial equivalence’
(in the case of high risk devices, the EU term is considered stricter) and is discussed in the EU

Clinical evidence guidelines: Medical devices Page 22 of 178

V3.1 June 2022
 Therapeutic Goods Administration

document MDCG 2020-5 Clinical Evaluation – Equivalence. See also these guidelines’ chapter on
Comparable devices including substantially equivalent devices.

Conducting clinical investigations

A properly conducted clinical investigation will comply with both the clinical investigation plan
and the laws and requirements in the location it is conducted in, along with any additional
requirements imposed by Australian legislation, to ensure the protection of human subjects and
the integrity of the data. Clinical investigations may be undertaken in Australia or outside of
Australia:
• When clinical investigation data is collected in Australia, the investigation must comply with
the NHMRC National Statement of Ethical Conduct in Human Research (clause 8.4(4) of Part
8 of Schedule 3 of the MD Regulations). The NHMRC National Statement refers to further
requirements that research needs to meet. As updated, these include the requirements of the
International Council for Harmonisation (ICH) Guideline for Good Clinical Practice and ISO
14155:2020 Clinical Investigation of Medical Devices for Human Subjects – Good clinical
practice. Trials should comply with the requirements in both of these documents.
• When clinical investigation data is collected outside Australia, the investigation must comply
with the principles of the Declaration of Helsinki (clause 8.4(5) of Part 8 of Schedule 3 of the
MD Regulations). The Declaration refers to the need to take into account applicable
international laws and standards. Trials should also comply with the International
Conference on Harmonisation (ICH) Guideline for Good Clinical Practice and ISO 14155:2020
Clinical Investigation of Medical Devices for Human Subjects – Good clinical practice.
The clinical investigation report should note if the clinical investigation was carried out in
accordance with the relevant standards or principles (and name the regulatory authority or
ethics committee(s) giving approval). Clinical investigations not conducted in accordance with
the applicable ethical standards or principles (as required by clauses 8.5(4) and (5)) should not
be relied upon by the manufacturer or sponsor and the reasons for this noted in the Clinical
Evaluation Report (CER).
Clinical trials should be registered in line with Declaration of Helsinki and World Health
Organisation recommendations. Since 1 July 2005, the International Committee of Medical
Journal Editors (ICMJE) has required (and recommended that all medical journal editors
require) registration of clinical trials in a public trials registry at or before the time of first
patient enrolment as a condition of consideration for publication. For more information refer to
ICMJE clinical trial registration requirements.
Registries include clinicaltrials.gov or any registry participating in the WHO International
Clinical Trials Registry Platform, such as the Australian New Zealand Clinical Trials Registry
(ANZCTR).
Clinical trials can be conducted within Australia under either the Clinical Trial Notification (CTN)
or Clinical Trial Approval (CTA) schemes for devices not currently included in the ARTG, or to
extend the intended purpose of a medical device beyond the current market approval.

Reporting standards for clinical investigations

International guidance on reporting standards for clinical trials can be found in ISO 14155:2020
Clinical investigation of medical devices for human subjects – Good clinical practice. Annex D of
this ISO provides useful information on what should go into a clinical trial report.
In addition to high-level guidance on how to structure a full clinical trial report, the reporting
requirements for specific trial designs are also included, as outlined below.

Clinical evidence guidelines: Medical devices Page 23 of 178

V3.1 June 2022
 Therapeutic Goods Administration

The following checklists are intended to inform reporting standards for peer-
reviewed publications and should be viewed as minimum requirements
only for full clinical trial reports.

Reporting standards for randomised controlled trials

The Consolidated Standards of Reporting Trials (CONSORT) statement provides an evidence-
based set of minimum guidelines for reporting parallel group randomised-controlled trials. The
statement provides a 25-item checklist and flow diagram displaying the progress of all
participants through randomised clinical trials. The focus is on transparent reporting of how the
trial was designed, analysed and interpreted.

Reporting standards for observational studies

The Strengthening the Reporting of Observational studies in epidemiology (STROBE) statement
is used for reporting observational studies, including case series and surveys. The statement
provides a 22-item checklist for reporting criteria, and the use of a flow diagram is suggested but
no official format is given. The STROBE statement provides guidance on how to report
observational research well and is endorsed by leading journals.

Reporting standards for diagnostic accuracy studies

The Standards for the Reporting of Diagnostic accuracy studies (STARD) statement is used for
the reporting of in vivo diagnostic accuracy studies. The statement provides a 25-item checklist
and flow diagram describing the design of the study and the flow of patients through the study.
The focus of the statement is on identifying the quality of reporting.

Reporting standards for systematic literature reviews

Guidelines for reporting systematic literature reviews are outlined in the Preferred Reporting
Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. It is recommended that
the PRISMA be followed closely when compiling a literature review as part of a submission for
pre- and post-market reviews. The statement includes a 27-item checklist and flow diagram
describing the study selection process in systematic literature reviews. Guidelines for Meta-
analysis of Observational studies in Epidemiology (MOOSE) may also be used for meta-analyses
of observational studies.

Literature review
Conducting a literature review is useful for identifying clinical data that is not in in the
possession of the manufacturer. A literature review may be presented in addition to clinical
investigation data described above, or on its own. Studies identified in a literature review that
do not pertain directly to the subject device or a substantially equivalent device may be used to
present the state of the art and inform risk management. If such studies are being relied on to
help meet clinical evidence requirements, a reasoned justification is necessary as to why any
data obtained for another device may be used to support the safety and performance of the
subject device (see chapter on Comparable devices including substantially equivalent devices).
A literature review in relation to a medical device includes a compilation, prepared using a
documented methodology, of published and unpublished scientific literature, both favourable
and unfavourable, relating to the medical device (Part 8 of Schedule 3 of the MD Regulations
(clause 8.5)). This includes expert opinion, information about the hazards and associated risks
arising from the use of the device for its intended purpose and the foreseeable misuse of the
device, and information about the performance of the device, including a description of the
techniques used to examine whether the device achieves its intended purpose. A written report

Clinical evidence guidelines: Medical devices Page 24 of 178

V3.1 June 2022
 Therapeutic Goods Administration

must be prepared by an expert in the relevant field containing a critical evaluation of the
compilation of literature.
The manufacturer of the medical device must ensure that the clinical data is evaluated by a
competent clinical expert and that clinical evidence demonstrating that the device complies with
the applicable provisions of the EPs is documented in writing (clause 8.6).
In brief, therefore, a literature review involves the systematic identification, synthesis and
analysis of the literature on the device or device type (culminating in a written report by an
expert). The highest standard of literature review is a systematic review with meta-analysis. In
all cases, it is critical that the methods used to conduct the literature review are transparent and
reproducible in order for the clinical assessor to evaluate objectivity (lack of bias) and quality.
A literature review should consist of the following components.

Search protocol
Prior to conducting a literature review, a protocol should be developed to identify, select and
collate relevant literature. The protocol should include the search aim(s) and outline the
population, intervention, comparator(s) and outcome(s) (PICO) criteria for the review. A record
must be kept of databases searched with justification, search terms used (including key words
and MeSH headings), date searched, period covered by the search, search limits applied
(including language, study design, etc.) and inclusion and exclusion criteria. This must contain
enough detail for a clinical assessor to reproduce the search. The search protocol should
describe the method used to extract data from included studies and any processes for
confirming data extracted by investigators.

Selection strategy
The selection criteria applied to the resulting list of studies should be defined in enough detail to
enable the clinical assessor to understand how the list of studies included in the review was
compiled. When selecting papers, the study design, quality of the data reported, quality of
analysis and the clinical significance of the results should be considered. Any weighting criteria
applied to the included studies should be detailed. Variables for data extraction should be listed
and defined.
A flow diagram should detail each step in the screening process, including total numbers of
studies screened, assessed for eligibility and included in the review. Objective, non-biased,
systematic search and review methods should be used such as PRISMA (Preferred Reporting
Items for Systematic reviews and Meta-Analyses) or Meta-analysis of Observational Studies in
Epidemiology (MOOSE) guidelines in accordance with the section Reporting standards for
clinical trials. The report should also summarise how each citation did or did not fit the selection
criteria for inclusion in the review. This may be presented as an appendix of excluded studies
with justification for the decision.

Review and critical analysis

It is preferred that the study characteristics and results of individual studies are summarised in
tabular format. This should include, for all outcomes considered (including safety and
performance measures), an effect size estimate and confidence interval for each study. Where
relevant, the range found across all studies for outcomes (e.g. adverse event rates for different
types of adverse events) should be presented. Then critical analysis of the literature should be
undertaken. This is not a simple summary of the individual study results, but a critique and
discussion of the study method, results and outcomes and how these apply to the device when
used for its intended purpose.

Clinical evidence guidelines: Medical devices Page 25 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Literature report
A report must be provided, analysed and endorsed (evidenced by signature and date) by a
competent clinical expert, containing a critical appraisal of this compilation, as per the legislative
requirements. Reviews should be prepared by researchers skilled in systematic review methods
in conjunction with a clinical expert. Where the review relies in part or wholly on literature for a
substantially equivalent or comparable device, the report should also clearly justify how the
device described in the compiled literature is relevant to the safety and performance of the
subject device. It is important that the published literature be able to establish the clinical
performance and safety of the device and demonstrate a favourable risk profile.
For further guidance on performing a literature review see MEDDEV 2.7/1 revision 4 (section 9
and appendices 5 and 6).

Post-Market Data
Post-market data should be provided for pre-market and post-market TGA assessments and
reviews. Post-market data may be collected by manufacturers, sponsors, regulatory agencies or
others. All post-market data available to the sponsor should be reported. Examples include:
• The number of units sold (or unit demand) worldwide since launch stratified by year and by
country (particularly if numbers are small) or geographic region. Note: this may not always
be appropriate for high use devices, those with several components or those on the market
for many years.
• The number and types of complaints to the manufacturer regarding the device, both as
reported and as confirmed on analysis and, in the case of new devices, stratified by year of
occurrence of complaint.
• The total number of adverse events (including serious adverse events) and vigilance data
reported to regulatory agencies, both as reported and as confirmed on analysis and
categorised by type (e.g. device malfunction, use error, inadequate design or manufacture)
and clinical outcome (e.g. death, amputation, surgical procedure required, no harm to
patient). These should be stratified by year of supply and/or year of occurrence of event.
• Any regulatory actions such as voluntary or mandatory recalls, including recalls for product
correction, removals, suspensions, withdrawals or other corrective actions occurring in the
market for IFU changes or other reasons and cancellations of the device anywhere in the
world, or any other corrective and preventive actions (CAPAs).
• Any data from Post-Market Clinical Follow-up (PMCF) studies, including interim and final
reports, and where relevant, interim datasets.

The manufacturers should clearly indicate whether the data reported is for the device or a
comparable device. The data should be compiled into a complaints, adverse events, and vigilance
report(s) that will allow the clinical assessor to better evaluate the benefit-risk profile of the
device. The CER should include an analysis and commentary on the profile, severity and
frequency (rate) of events reported. Adverse event and complaint data and rates should be
discussed and critiqued to enable an understanding of the safety profile of the device in a ‘real-
world’ setting. As the time since first approval worldwide lengthens, the relevance of post-
market data for comparable device(s) diminishes and should be replaced by data for the device
itself.
Post-market data can support the substantiation of the safety and performance claims of the
device, and guide risk identification, assessment and mitigation. It is useful for identifying less
common but serious device-related adverse events and it provides valuable long-term
information about the safety and performance of a device. Post-market data is particularly
important where there may be a paucity of clinical data from other sources, or when data from

Clinical evidence guidelines: Medical devices Page 26 of 178

V3.1 June 2022
 Therapeutic Goods Administration

other sources is not sufficiently robust to establish a favourable benefit-risk profile for the
device.
When updated post-market data is provided to the TGA for any regulatory purpose, any new
regulatory actions and any new serious adverse events should be identified and described.

Adverse events and complaints

Adverse events are required to be reported to the governing bodies of the countries in which the
device is used when the event leads to or may lead to death or serious injury. Adverse events
and complaints data are available in the manufacturer’s own internal complaint handling log and
in publicly available databases such as the FDA’s Manufacturer and User Facility Device
Experience (MAUDE) database or TGA Incident Reporting and Investigation Scheme (IRIS).
One of the serious limitations of post-market adverse event and complaint reports is under-
reporting by end-users. This limitation should be considered in any post-market analysis. More
information on reporting adverse events (and complaints) can be found on the Database of
Adverse Event Notifications - medical devices page on the TGA website. Serious adverse events
should be subject to particular scrutiny as part of post-market surveillance and reporting.

Post-market regulatory actions

Information about recall actions and suspension or cancellation of marketing approval (in any
jurisdiction) is also valuable. Recall actions generally take place to resolve a problem with a
device for which there are deficiencies or other issues concerning safety, quality or performance.
There are generally two key types of recall action (a) correction, which may involve temporary
removal from use for example, for changes to the IFU, and (b) permanent removal of deficient,
defective or unsafe medical devices from use. In the case of implanted devices, hazard alerts may
apply. The full range of recall actions are described in the TGA’s Uniform recall procedure for
therapeutic goods (URPTG). More information about recall actions can be found on the System
for Australian Recall Actions (SARA) page on the TGA website.
The occurrence of and reasons for suspensions, removals, withdrawals, cancellations or other
corrective actions in any jurisdiction, should be reported.

Post-Market Clinical Follow-up (PMCF) studies

A PMCF study is a study carried out following marketing authorisation intended to answer
specific questions (uncertainties) relating to safety, clinical performance and/or effectiveness of
a device when used in accordance with its labelling. Data obtained from PMCF studies are a
subset of post-market data. PMCF studies can be used to collect additional clinical data to
address the remaining uncertainties about the potential benefits and residual risks of the device.
Further guidance on PMCF studies can be obtained from the IMDRF document titled Post-Market
Clinical Follow-Up Studies, which examines:
• circumstances where a PMCF study may be indicated
• elements of a PMCF study including objectives, design and implementation
• use of information from PMCF studies.
When PMCF studies are planned as part of a risk management strategy, including as part of pre-
market applications, then a Clinical Investigation Plan should be provided. PMCF studies
conducted post approval in other jurisdictions may also be used as clinical investigation studies
for pre-market applications in Australia.

Clinical evidence guidelines: Medical devices Page 27 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Other clinical experience data

Clinical experience data encompasses data generated through any clinical use of the device that
is not related to clinical investigation. This may include post-market surveillance reports, sales
and complaints data, vigilance reports and clinically relevant field corrective safety actions (all
part of post-market data, above), and other sources of clinical experience data.
Other sources of clinical experience data are often referred to as Real World Data (RWD) and
may come from the following sources:
• Electronic Health Records (EHRs)
• claims and billing activities
• product and disease registries
• patient-generated data including in home-use settings
• data gathered from other sources that can inform on health status, such as mobile devices.
Real-world evidence (RWE) is the clinical evidence regarding the usage and potential benefits or
risks of a medical product derived from analysis of RWD.
The FDA document titled Use of Real-World Evidence to Support Regulatory Decision-Making
for Medical Devices provides further guidance. In some situations, RWD may be of sufficient
quality to help inform the benefit-risk profile of devices at various points in their lifecycle and
hence inform regulatory decisions.
Clinical experience data may be relevant and useful regardless of whether it is used as direct or
indirect clinical evidence to demonstrate compliance with the EPs. However, as the time since
first approval worldwide lengthens, the importance of direct device data increases.

Device registries
Device registries are systematic collections of data of medical outcomes following use of medical
devices. They play a unique and important role in medical device surveillance. These can provide
additional detailed information about patients, procedures, and devices not routinely collected
by other means. Registries can provide valuable information on device performance in terms of
functional outcomes and quality of life of patients. Registries using multiple device types may
provide a suitable in-built comparator, such as the average of a particular performance and/or
safety marker across the device category (for example, joint registries may provide average
revision rates across different types of joint prosthesis). In other instances (for example, single
device registries), comparators derived from the literature will be required. Use of registries
should take appropriate account of data limitations, variation across registries with respect to
data structure and analysis and populations covered. Examples of Australian device registries
include the Australian Breast Device Registry, the Australian National Orthopaedic Association
National Joint Replacement Registry (AOANJRR) and the Victorian Cardiac Outcomes Registry.

Regulatory approval in other jurisdictions

If the device is approved for use in another jurisdiction the manufacturer should provide
regulatory status, including the certificate number, date of issue and name under which the
device is marketed. The exact wording of the intended purpose and any specific conditions in
other jurisdictions should be provided. For example, if magnetic resonance imaging (MRI)
designation in other jurisdictions is provided, this will improve the efficiency of the assessment.
The sponsor should specify whether the approval process included a clinical assessment, and
may wish to provide copies of clinical assessments and other relevant documents.

Clinical evidence guidelines: Medical devices Page 28 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Clinical evaluation
The IMDRF document IMDRF MDCE WG/N56FINAL:2019 Clinical Evaluation is an international
consensus document on clinical evaluation. Clinical evaluation is a set of ongoing activities that
use scientifically sound methods for the assessment and analysis of clinical data to verify the
safety and clinical performance of the device when used as intended by the manufacturer. The
clinical evaluation should enable conclusions to be drawn on the balance of risks and benefits of
the device.
The three steps in clinical evaluation are (i) data identification, (ii) appraisal, and (iii) analysis.
The first step - identification of relevant clinical data – is described in the previous section. The
next step is appraisal of each data set to determine the limitations and merits of the clinical data
in terms of relevance, clinical significance and quality. The final step is to analyse the data to
draw a conclusion on the balance of benefits and risks. This section covers steps (ii) and (iii).

• Clinical evaluation is an ongoing process conducted throughout the

lifecycle of a medical device. Manufacturers must periodically review the
performance, safety and benefit-risk profile of the device and update the
clinical evidence accordingly.
• Over the lifecycle of the device the clinical evaluation will change. For
instance, when the device has been on the market for a number of years,
the relevance of comparable device data is less significant, and direct
clinical experience data is likely to be of greater relevance.

Appraisal of the clinical data

The clinical data should be appraised to elucidate its merits and limitations. Appraisal involves:
• assessing each piece of data to determine its quality, its relevance to the subject device or
comparable device and its clinical significance, considering the target population and
intended purpose
• determining the contribution of each dataset to the overall performance and safety profile of
the subject device, considering the data generation/collection methods and potential
sources of confounding or bias that may influence results.
Assessment of evidence quality should consider study type, size and design, and also
comparability (to standard of care or alternative treatments). The quality and relevance of
clinical evidence provided is a significant consideration in determining whether the
requirements of the EPs have been met. The following should be noted:
• Although there is no rule regarding study size, those involving a sample size that is not
statistically-powered will generally be considered poor quality evidence.
• Where possible, studies should be statistically-powered to demonstrate non-inferiority
against the established standard of care.
• Single arm studies (and other study designs) with no comparator arm are generally
considered inadequate evidence.
• Comparisons of datasets obtained through different methodologies (for example, a case
series using the subject device with standard of care outcomes established from a literature
search) are generally considered poor quality evidence and may be subject to greater
scrutiny, as necessary, when assessing whether that data supports compliance with the EPs.

Clinical evidence guidelines: Medical devices Page 29 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• Clinical safety and performance should generally be expressed in terms of person-centred

outcomes, such as mortality, morbidity, adverse events, and patient reported outcome
measures (PROMs).
• Where study findings are expressed in terms of markers or intermediate measures of safety
and performance, a clinically reasoned argument should be provided linking the study
findings with patient centred outcomes.
Refer also to the matters raised below in relation to MEDDEV 2.7/1 Rev 4 appendix 6, Appraisal
of clinical data.
Studies conducted by manufacturers or sponsors, or those who have received funding or
support from manufacturers or sponsors, will be considered on their merits. Peer reviewed
articles should clearly identify any conflicts of interest (actual or perceived). It is accepted that
certain studies require support from manufacturers (such as large-scale pre-market approval
studies) or will be conducted by manufacturers (such as PMCF studies). A discussion of the
extent of involvement of manufacturers or sponsors should form part of the study report and the
critical analysis contained in the CER.
An important part of the clinical evaluation is determining the overall strength of the evidence
presented. A widely accepted tool for ranking different types of study design is the National
Health and Medical Research Council’s (NHMRC) levels of evidence. The levels of evidence rank
different study designs into a hierarchy according to their potential to adequately answer a
particular research question (e.g. diagnostic, intervention, screening etc.). The hierarchy is based
on the level of bias inherent in the study design. Using this hierarchy, systematic reviews of
randomised controlled trials represent the strongest level of evidence, followed by individual
randomised controlled trials, pseudo randomised controlled trials, non-randomised
comparative trials, and case series. The level (or sufficiency) of evidence ultimately affects the
confidence that can be placed in the study results. Manufacturers should source the highest level
of evidence available that demonstrates the safety and performance of the device for the
intended purpose(s).
Several appraisal tools are available for assessing the quality, suitability and contribution of the
clinical data, noting that it is preferable to use tools that have been validated. The evaluator
should choose tools that are appropriate for the data set in question, and indicate which ones
were used, along with checklists and other relevant information in appendices.
The following table includes commonly used quality appraisal tools.
Table 1: Commonly used quality appraisal tools

Tool Applicable study designs Source

Jadad
Randomised studies https://www.ncbi.nlm.nih.gov/pubmed/8721797
Score

Downs & Randomised & non- http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1

Black randomised studies 756728/

Studies of diagnostic http://www.bristol.ac.uk/population-health-

QUADAS
accuracy sciences/projects/quadas/

AMSTAR Systematic reviews http://amstar.ca/

Clinical evidence guidelines: Medical devices Page 30 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Additional guidance on quality appraisal tools is provided by the Scottish Intercollegiate

Guidelines Network (SIGN), the Centre for Evidence-Based Medicine (CEBM), and the Cochrane
Collaboration’s Handbook for Systematic Reviews of Interventions.
With regards to appraisal tools to determine the suitability and weighting contribution of each
dataset, IMDRF provides examples of possible appraisal criteria in their Clinical Evaluation
document, as follows:
Table F1: IMDRF Sample Appraisal Criteria for Suitability

Suitability Criteria Description Grading System

Appropriate device Were the data generated from D1 Actual device

the device in question? D2 Comparable device
D3 Other medical device
Appropriate device Was the device used for the A1 Same use
application same intended use (e.g. A2 Minor deviation
methods of deployment, A3 Major deviation
application, etc.)?

Appropriate patient group Were the data generated from a P1 Applicable

patient group that is P2 Limited
representative of the intended P3 Different population
treatment population (e.g. age,
sex, etc.) and clinical conditions
(i.e. disease, including state and
severity)?

Acceptable report/data Do the reports or collations of P1 High quality

collation data contain sufficient P2 Minor deficiencies
information to be able to P3 Insufficient information
undertake a rational and
objective assessment?

Table F2: IMDRF Sample Appraisal Criteria for Data Contribution

Data Contribution Criteria Description Grading System

Data source type Was the design of the study T1 Yes

appropriate? T2 No

Outcome measures Do the outcome measures O1 Yes

reported reflect the intended O2 No
performance of the medical
device?

Follow up Is the duration of follow-up F1 Yes

long enough to assess F2 No
treatment effects and identify
complications?

Clinical evidence guidelines: Medical devices Page 31 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Data Contribution Criteria Description Grading System

Statistical significance Has a statistical analysis of S1 Yes

the data been provided and is S2 No
it appropriate?

Clinical significance Was the magnitude of the C1 Yes

treatment effect observed C2 No
clinically significant?

The appraisal process should be described in sufficient detail to allow the clinical assessor to
undertake a rational and objective assessment of the information provided. This may take the
form of a table, in which each study is assessed in terms of its quality, suitability and weighting
contribution, though the layout and presentation of this information may vary depending on the
tools used for evaluation. The manufacturer should present data on the risk of bias in each study
and outcome level assessments. The results of any assessment of risk of bias across studies
(such as publication bias and selective reporting within studies) should also be presented where
such information is available. Funding sources should be included if it is one of the variables for
data extraction.
Manufacturers and sponsors are also referred to MEDDEV 2.7/1 Rev 4 appendix 6, Appraisal of
clinical data for examples of studies that may lack scientific validity for demonstration of
adequate clinical performance and/or clinical safety.
a. Lack of information on elementary aspects
This includes reports and publications that omit disclosure of
• the methods used
• the identity of products used
• numbers of patients exposed
• what the clinical outcomes were
• all the results the clinical study or investigation planned to investigate
• undesirable side-effects that have been observed
• confidence intervals/ calculation of statistical significance
• if there are intent-to-treat and per protocol populations, definitions and results for the two
populations.
b. Numbers too small for statistical significance
Includes publications and reports with inconclusive preliminary data, inconclusive data from
feasibility studies, anecdotal experience, hypothesis papers and unsubstantiated opinions.
c. Improper statistical methods
This includes
• results obtained after multiple subgroup testing, when no corrections have been applied for
multiple comparisons
• calculations and tests based on a certain type of distribution of data (e.g. Gaussian distribution
with its calculations of mean values, standard deviations, confidence intervals, t-tests, other

Clinical evidence guidelines: Medical devices Page 32 of 178

V3.1 June 2022
 Therapeutic Goods Administration

tests), while the type of distribution is not tested, the type of distribution is not plausible, or the
data have not been transformed. Data such as survival curves, e.g. implant survival, patient
survival, symptom-free survival, are generally unlikely to follow a Gaussian distribution.
d. Lack of adequate controls
In the following situations, bias or confounding are probable in single arm-studies and in other
studies that do not include appropriate controls.
• when results are based on subjective endpoint assessments (e.g. pain assessment)
• when the endpoints or symptoms assessed are subject to natural fluctuations (e.g. regression to
the mean when observing patients with chronic diseases and fluctuating symptoms, when
natural improvement occurs, when the natural course of the disease in a patient is not clearly
predictable)
• when effectiveness studies are conducted with subjects that are likely to take or are foreseen to
receive effective co-interventions (including over-the-counter medication and other therapies)
• when there may be other influencing factors (e.g. outcomes that are affected by variability of
the patient population, of the disease, of user skills, of infrastructure available for planning/
intervention/ aftercare, use of prophylactic medication, other factors)
• when there are significant differences between the results of existing publications, pointing to
variable and ill-controlled influencing factors.
In the situations described above, it is generally not adequate to draw conclusions based on direct
comparisons with external or historic data (such as drawing conclusions by comparing data from a
clinical investigation with device registry data or with data from published literature).
Different study designs may allow direct comparisons and conclusions to be drawn in these
situations, such as randomised controlled design, cross-over design, or split-body design.
e. Improper collection of mortality and serious adverse events data
Demonstration of adequate benefits and safety is sometimes based on mortality data or occurrence
of other serious outcomes that limit a subject’s ability to live in his home and be available for
follow-up contacts. In this type of study,
• consent of the subjects for contacting reference persons/ institutions for retrieval of medical
information should be obtained during recruitment; when subjects can no longer be found,
outcomes should be investigated with the reference persons/ institutions
• the consequences of missing data on the results should be analysed (e.g. with a sensitivity
analysis); alternatively, when patients can no longer be found and their outcomes cannot be
identified, they should be considered to meet the SAE endpoint under investigation (e.g. the
mortality endpoint of a study).
In mortality studies (and other studies addressing serious outcomes) procedures for investigating
serious patient outcomes, numbers of subjects lost to follow-up, reasons why subjects leave the
study, and the results of sensitivity analysis should be fully disclosed in reports and publications.
f. Misinterpretation by the authors
Includes conclusions that are not in line with the results section of the report or publication, such as
• reports and publications not correctly addressing lack of statistical significance/confidence
intervals that encompass the null hypothesis
• effects too small for clinical relevance.

Clinical evidence guidelines: Medical devices Page 33 of 178

V3.1 June 2022
 Therapeutic Goods Administration

g. Illegal activities
Includes clinical investigations not conducted in compliance with local regulations. Clinical
investigations are generally expected to be designed, conducted and reported in accordance with
EN ISO 14155 or to a comparable standard, and in compliance with local regulations and the
Declaration of Helsinki.

Analysis of the clinical data

The objective of analysis is to make a benefit-risk determination regarding whether the
appraised data sets available for a medical device collectively demonstrate the safety, clinical
performance and/or effectiveness of the device in relation to its intended use.
Following the appraisal of all the clinical data, the manufacturer should provide a well-reasoned
and documented analysis of the foreseeable risks that could occur with the use or misuse of the
device, and compare these with an analysis of the expected benefits that may be provided to the
end user. The nature, extent, probability and duration of benefits should be considered. This
analysis should be clearly supported by evidence, including appropriate references. In
demonstrating whether the expected benefits of the device outweigh the undesirable effects, the
analysis may consider (but should not be limited to) the following criteria:
• the strengths and limitations of the clinical data presented in support of the safety and
performance of the device for the intended purpose(s) e.g. level and nature of evidence, bias,
confounders, length of follow-up
• the clinical significance of the benefits of the device for the intended purpose(s) as
demonstrated by the clinical data
• based on the clinical data provided and on a sound statistical approach, a reasonable
prediction of the proportion of ’responders’ out of the target group or subgroups should be
made
• the safety issues identified in the clinical investigation data and/or literature review and
post-market data (clinical experience) for the intended purpose(s), as well as reasonably
foreseeable hazards associated with the clinical use of the device that the data may not have
captured e.g. misinterpretation or misuse of the device
• the probability of patients experiencing a harmful event, that is, the proportion of the
intended population that would be expected to experience a harmful event and whether an
event occurs once or repeatedly may be factored into the measurement of probability
• the duration and severity of adverse events caused by the device or the procedure
• whether there are mitigation strategies that have been implemented to address real or
theoretical safety issues i.e. risk management documentation and IFU/labelling
• any issues of uncertainty surrounding the application of the device for its intended purpose,
e.g. limitations in the statistical analysis, generalisability of results to an Australian
population.
The clinical expert should comment on the risk analysis and risk management approach by the
manufacturer and make a determination of the benefit-risk profile of the use of the device in the
intended target groups for the indications sought. The Clinical Evaluation Report (CER) should
clearly demonstrate a favourable profile based on current knowledge and the state of the art in
the relevant medical fields, considering the totality of the clinical data on the device.

Clinical evidence guidelines: Medical devices Page 34 of 178

V3.1 June 2022
 Therapeutic Goods Administration

The Clinical Evaluation Report (CER)

After the completion of the clinical evaluation process, a CER should be compiled that includes:
• scope and context of the evaluation
• clinical data
• data appraisal and analysis
• conclusions reached about performance, safety and presentation (including labelling, patient
information and IFU) of the medical device when used for the intended purpose(s)
• a benefit-risk determination.
The clinical evaluation report should contain sufficient information to be read as a standalone
document by an independent party (including a regulatory authority) for the purposes of
assessing legislative compliance regarding clinical evidence requirements.

The following section provides an overview of the recommended content and format of the
clinical evaluation report (CER), which is a standard component of pre-market applications and
may also be required to be provided to the TGA in relation to post-market matters. The CER
should be updated periodically through the lifecycle of the device to incorporate new evidence
including clinical experience data and updated benefit-risk analyses. A record of reviews and
amendments should be kept (along with a copy of each historical version and the most recent
version).

Critical analysis and expert opinion

All clinical data in the CER must be critically evaluated by a competent clinical expert, who
arrives at a reasoned conclusion on the benefit-risk profile of the subject device and provides
their written endorsement and/or signature. Critical analysis requires consideration of all
relevant clinical evidence regarding the device, including evidence that is less favourable to the
device. The manufacturer must show due diligence in ensuring all relevant clinical evidence is
identified and discussed.

Competent clinical expert

A competent clinical expert is generally someone with relevant medical qualifications and direct
clinical experience in the use of the device or device type in a clinical setting. For a novel, high
risk device, the clinical expert is expected to have current or recent clinical experience with the
device type (preferably within the past two years).

The selection of a clinical expert will therefore depend on the device type and its intended
purpose(s). For example, for a coronary stent submission the clinical expert should be a
cardiologist or equivalent. For a lower risk device that is not typically used by a medical
practitioner, another health practitioner who uses the device or similar devices in a clinical
setting may be considered an appropriate clinical expert. In order for the clinical assessor to
determine whether an appropriate clinical expert has been chosen, the full curriculum vitae of
the clinical expert should be included with any convergence of interests or potential for conflict
with the manufacturer or sponsor noted.

Clinical evidence guidelines: Medical devices Page 35 of 178

V3.1 June 2022
 Therapeutic Goods Administration

CER content
The content and format of the CER should be as follows, to facilitate timely review by the TGA.
Manufacturers may also refer to the IMDRF document Clinical Evaluation (Appendix G) and
MEDDEV 2.7/1 revision 4.
The content of the CER should include all of the following.
a) General details
b) Description of the medical device and its intended application
c) Intended therapeutic and/or diagnostic indications and claims
d) Context of the evaluation and choice of clinical data types
e) Summary of relevant pre-clinical data
f) Discussion regarding comparable devices including substantially equivalent devices
g) Summary of the clinical data and appraisal
h) Data analysis
i) Conclusions
j) Name, signature and curriculum vitae of the clinical expert and date of report

General details
The subject device should be identified by its proprietary name (and any code names assigned
during its development), and its manufacturer.

Description of the medical device and its intended purpose

A description of the subject device should be provided, including the following information,
cross-referenced to relevant sections of the manufacturer’s technical information where
applicable:
• models and sizes
• the device group to which the device belongs (e.g. biological artificial aortic valve)
• materials used, including whether it incorporates a medicine (new or existing), biological
tissues and/or blood products
• the device components (including software and accessories)
• mechanical characteristics
• how the device functions
• any other relevant information relating to the device such as sterility and radioactivity.
Diagrams or photographs of the device including steps for assembly and use are helpful. The
description should be detailed enough to allow for a valid evaluation of compliance with EPs,
retrieval of meaningful literature and, if applicable, assessment of equivalence to other devices
described in the literature, or alternatively, assessment of the novelty of the design, features or
mechanism of the device. If the application is for a multi-component procedure pack, each
component in the system must be adequately described.

Clinical evidence guidelines: Medical devices Page 36 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Intended therapeutic and/or diagnostic indications and claims

It is important to describe the indications for use, including the clinical condition being managed
(with reference to the severity and stage of disease), the intended patient population, and the
intended application of the device (whether single use/reusable; invasive/non-invasive,
implantable, duration of use or contact with the body, and the organs, tissues or body fluids
contacted by the subject device).
The magnetic resonance (MR) status of the device (MR unsafe, MR conditional, MR safe or ‘safety
in magnetic resonance environment not evaluated’) should be provided for all implantable
devices (and components of these devices that may be taken into the MR scanner room).
Additionally, any contraindications, clinical performance/safety claims and warnings should also
be documented. Particular attention should be paid to whether the indications and claims are
supported by the clinical data.

Context of the evaluation and choice of clinical data types

Outline the developmental context for the medical device. The information should include
whether the medical device is based on a new technology, a new clinical application of an
existing technology, or the result of incremental change of an existing technology. The amount of
information will differ according to the history of the technology.
Where a completely new technology has been developed, this section will need to give an
overview of the developmental process and the points in the development cycle at which clinical
data have been generated. For long standing technology, a shorter description of the history of
the technology (with appropriate references) could be used. Clearly state if the clinical data used
in the evaluation are for a comparable device. Identify the comparable device(s) and provide a
justification of the comparability, cross-referenced to the relevant nonclinical documentation
that supports the claim.
State the EPs relevant to the device in question, in particular, any special design features that
pose special performance or safety concerns (e.g. presence of medicinal, human or animal
components) that were identified in the device risk management documentation and that
required assessment from a clinical perspective. Outline how these considerations were used to
choose the types of clinical data used for the evaluation. Where published scientific literature
has been used, provide a brief outline of the search and retrieval process, cross-referenced to the
literature search protocol and reports.
The CER should describe the developmental and regulatory context for the subject device. The
developmental context (often referred to as the ‘state of the art’) includes whether the device is
based on a new technology, a new clinical application of an existing technology, or the result of
incremental change of an existing technology. If the device has evolved from predicate(s) over
time, the number and dates of certificates for these may be useful in exploring the history of the
device.
The regulatory context includes a list of the countries in which the subject device has been
marketed and the dates of introduction into each country. The exact wording of the intended
purpose in other jurisdictions should also be provided. It is preferable that certificates of
conformity in other regulatory jurisdictions (e.g. CE marking, FDA, Health Canada) be provided
including the number and date of issue of international certificates, as these allow verification of
post-market data (e.g. through searches of FDA’s Manufacturer and User Facility Device
Experience (MAUDE)), and may increase confidence in performance and safety claims.
The trade name(s) of the device in other regulatory jurisdictions should also be clearly stated, if
different from the name used in Australia. A description of the notification or approval process
undertaken by the overseas regulator is also helpful (specifically, if this involved clinical
assessment). Information on concurrent applications for registration in the other jurisdictions,
particularly Europe, the USA, Japan and Canada, is helpful, if available.
Clinical evidence guidelines: Medical devices Page 37 of 178
V3.1 June 2022
 Therapeutic Goods Administration

Summary of relevant pre-clinical data

Pre-clinical data may be provided where it adds to the clinical evaluation or risk management
strategy. It should be provided where pre-clinical testing (e.g. bench testing including
verification and validation, animal testing) is being relied upon as evidence of device safety
and/or performance, including to verify claims made in the device labelling not adequately
substantiated by the clinical data.

Discussion regarding comparable devices including substantially

equivalent devices
In some circumstances, the safety and performance of the subject device may be substantiated
by presenting evidence from a substantially equivalent device (indirect clinical evidence) or
supported by evidence from a comparable device. Information to help manufacturers determine
the extent to which clinical evidence from comparable devices (including substantially
equivalent devices) can be relied on to meet the requirements of the EPs can be found in the
section Comparable devices including substantially equivalent devices. Where indirect evidence
is presented, the CER should summarise the clinical, technical and biological differences
between the subject and comparable devices and provide a critical analysis regarding the impact
of these differences on clinical outcomes (see Comparing device characteristics).
Note: If the evidence strategy relies on clinical evidence for a comparable or substantially
equivalent device, this evidence must be provided as part of the submission. Under the
Australian legislative framework, it is not sufficient to demonstrate substantial equivalence to an
ARTG-included device – the clinical evidence for the substantially equivalent device needs to be
evaluated in order to demonstrate compliance of the subject device with the EPs.

Summary of the clinical data and appraisal

What constitutes appropriate clinical data will vary depending on the type of device under
assessment and its state of development, but may include clinical investigation data, literature
review data and/or clinical experience data pertaining to the subject device, substantially
equivalent device and/or comparable device. The CER should include a summary of all clinical
data used in the evaluation accompanied by a critical analysis outlining how this data supports
device safety and performance.
For key clinical investigations, either the full clinical investigation report and/or copies of full
text journal articles should be provided in the CER, or as supporting documents. Brief
summaries of studies with insufficient detail to enable a thorough assessment of the study
methodology are not acceptable. Details of literature searches should be included in the CER or
provided in the supporting documents. An outline of the data appraisal methods used in the
evaluation (such as quality, suitability and weighting assessments) and a summary of the key
results should be provided. For further information refer to Appraisal of clinical data and
MEDDEV 2.7/1 Rev 4 appendix 6, Appraisal of clinical data.

Data analysis
A competent clinical expert should evaluate all the clinical data and provide a well-reasoned
argument as to how the clinical data demonstrate the performance and safety of the subject
device when used for the intended purpose(s), and hence a positive benefit-risk profile. It
typically involves a discussion of the following performance, safety and labelling aspects:
• Performance: key data sets that contribute to the demonstration of the performance of the
subject device and (where useful) particular performance characteristics; the consistency of
the results; statistical and clinical significance of effects.

Clinical evidence guidelines: Medical devices Page 38 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• Safety: safety issues identified in the clinical investigation data, literature review and/or
clinical experience data; the total experience with the subject device to date; a summary of
device-related adverse events (with a focus on serious adverse events).
• Product labelling: the consistency of the product labelling with the clinical data and whether
residual risks associated with the use of the device are adequately conveyed in the IFU.
Emphasis should be placed on explaining the links between the clinical evidence and the
intended purpose, indications, contraindications, warnings and precautions, and actual and
potential adverse effects of the device on health.

• It is critical that a CER (which serves to detail the clinical evidence as required
by the legislation) is not simply a summary of the data, followed by a
statement that the data demonstrate safety and performance. This approach
does not represent an adequate clinical evaluation.
• It must be explicitly clear to the clinical assessor whether direct clinical
evidence (pertaining to the device) or indirect clinical evidence (pertaining to
a substantially equivalent device) are provided.
• It is important to identify any changes made to the device since the clinical
data were gathered and if so to document the changes and to clarify the exact
version of the device.

The CER should include an evaluation of the post-market data presented in the submission and
any other data from clinical experience (special access schemes etc.) and comment on its clinical
significance. The detailed datasets can be provided in the supporting documents. In assessing
the post-market data, the clinical expert should comment on adverse events, vigilance reports
and complaint rates and any recalls, withdrawals, removals, suspensions and cancellations for
any reason in any jurisdiction and discuss the implications for the safety of the device. The
evaluation of the post-market data should clearly indicate whether the data reported is for the
subject device or a comparable device.

Conclusions
The conclusion of the CER should clearly outline key findings from the evaluation regarding the
performance and safety of the subject device, with respect to its intended purpose. Statements
that address the following should be included. Whether the:
• clinical evidence demonstrates compliance with EP 14 and the other EPs
• clinical evidence on the device and/or substantially equivalent device is supportive of the
safety and performance of the subject device
• residual risks have been adequately mitigated with appropriate justification, for example,
inclusion of relevant statements in the IFU and risk management documentation, and
through post-market clinical follow up studies
• risks associated with the use of the subject device are acceptable when weighed against the
benefits to the patient.

Name, signature and curriculum vitae of clinical expert and date of report
As stated in clause 8.6 of Part 8 of Schedule 3 of the MD Regulations:
The manufacturer of a kind of medical device must ensure that the clinical data is
evaluated by competent clinical experts.
Clinical evidence guidelines: Medical devices Page 39 of 178
V3.1 June 2022
 Therapeutic Goods Administration

The manufacturer must ensure that clinical evidence demonstrating that the device
complies with the applicable provisions of the EPs is documented in writing.
For further information refer to Critical analysis and expert opinion.

Clinical evidence guidelines: Medical devices Page 40 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Overview for constructing a CER

The following diagram outlines the components that comprise clinical evidence for a medical
device and the process to compile a CER.

Clinical evidence pathway:

Market Authorisation and
Monitoring & Compliance

Demonstrate YES
substantial equivalence
to a comparable device
NO

Compile the clinical evidence

Include copies of all source documents*

Clinical data may comprise:

· Full reports of pivotal Additional information:

clinical studies
Relevant pre-clinical data
· Comprehensive literature
review Overseas regulatory status
· Clinical experience
Comprehensive information on
the device

Risk analysis
Compilation, and management document
synthesis and
critical analysis by a clinical
expert

Collate full CV of the

Clinical Evaluation Report clinical expert
NO Does the evidence comply with
the applicable provisions of the
Essential Principles?

YES

Compile CER and supporting

documents and
submit to TGA

Figure 1. Overview of the process of constructing a CER

* Source documents for clinical data may not initially be required for a clinical assessment requested as part of an
audit of an application for inclusion based on EU certification, provided that the CER contains sufficient detail for the
TGA assessor to appreciate how the clinical expert was able to demonstrate compliance with the EPs.

Clinical evidence guidelines: Medical devices Page 41 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Common errors in the CER

There are a number of common errors or deficiencies in CER submissions that can be avoided,
including:
• absence of the required components of the CER and/or referenced attachments and
appendices missing
• intended purpose(s), indication and claims inconsistent between documents - for example,
the application, IFU and CER list different intended purpose(s)
• intended purpose(s), indication and claims not supported by clinical data
• lack of information about the regulatory history of the device in other countries, for example
recalls, withdrawals, removals from market, suspensions and cancellations and the reasons
for these in any jurisdiction
• information on comparable devices not included and/or substantial equivalence not
demonstrated (if relevant)
• insufficient or incomplete clinical investigation(s) data, literature and post-market data with
the device or comparable device (if relevant)
• in submissions where a literature review is provided:
– no documented method and/or no demonstrated comprehensive literature review
– insufficient information and/or poor-quality search protocol that result in inability to
reproduce or understand the literature review strategy
– provision of a multitude of publications with little or no explanation as to why they are
of relevance
– no identification of device used or indication for use in articles reviewed
– no summary of study characteristics and findings for each included article
• little or no synthesis and critical evaluation of the clinical investigation data, results of the
literature review and post-market data:
– no discussion of relative strengths of the data, for example randomised controlled
trials, case control studies, case series
– substantial equivalence covering technical characteristics, biological characteristics
and clinical use not established to validate the data for a different device (i.e.
comparable device) to the device under review
– lack of discussion of the validity or otherwise of outcome measures used
– no endorsement by the clinical expert that the differences will not adversely affect the
safety or performance of the device
• inadequate critique and summary of the totality of evidence provided for the device
• no post-market data including adverse events, vigilance reports, complaints, failures in cases
where this information is available
• CER not endorsed/signed by clinical expert and/or CER not dated or out-dated
• inappropriate selection of clinical expert/s
• CV of clinical expert/s not provided.

Clinical evidence guidelines: Medical devices Page 42 of 178

V3.1 June 2022
 Therapeutic Goods Administration

It must be explicitly clear to the clinical assessor whether direct clinical evidence (pertaining to
the subject device) or indirect clinical evidence (pertaining to a comparable device that has been
demonstrated to be substantially equivalent) are provided for assessment. Further, it is
important to clarify if any changes have been made to the device since the clinical data were
gathered.
The CER must be in English and in an easily readable form. Avoid unclear or ambiguous terms,
poor grammar and spelling, and poorly organised information.
Avoiding these common errors will help to ensure that submissions for pre-and post-market
clinical assessments are processed efficiently, thereby reducing the time required to report back
to the applicant.

Supporting documents
The following supporting documents add to the evidence provided in the CER (they may be
provided separately or as part of the CER):
• risk assessment and risk management documents
• Instructions for Use (IFU), labelling, product manual and all other documents supplied with
the device
• additional information on the device
• pre-clinical data (if relevant)
• clinical investigation reports (full study reports or peer reviewed journal articles)
• literature search and selection strategy
• pivotal articles from the literature review
• post-market surveillance reports.
Further guidance is set out below.

Risk assessment and risk management documents

A well-reasoned and comprehensively documented risk analysis outlining the potential hazards
related to the device is necessary in order to demonstrate compliance with the EPs. The
manufacturer should ensure that all risks identified in the clinical data are included in the risk
assessment; that is, risks relating to patient treatment, method of operation of the device
including potential device failures, and risks relating to usability i.e. harm to the patient that
results from use of the device but is not caused by the device itself.
Device-related hazards include, but are not limited to, chemical, mechanical, thermal, electrical,
radiation, and biological hazards. Use-related hazards 1 refer to hazards associated with user
interactions with the device and include, but are not limited to, hazards that occur when the
device is used as intended by appropriately trained clinicians but there are inherent risks
associated with the procedure or use of the device, when the device is not used as intended,
users are not suitably trained or equipped to use the device, users are not capable of using the
device, or when the user’s expectations about the device are not consistent with the intended
use of the device.
All ongoing safety concerns (risks) should be specified as to the potential causes, nature,
probability, extent, duration, frequency and severity of occurrence. This type of analysis should

1 FDA guidance: Applying Human Factors and Usability Engineering to Medical Devices (Feb 2016), page 5
Clinical evidence guidelines: Medical devices Page 43 of 178
V3.1 June 2022
 Therapeutic Goods Administration

commence before beginning product development as it generates the safety requirements for
the design specification. Once all potential hazards arising from the use of the device for its
intended purpose(s) in the target population have been identified, the manufacturer is expected
to implement a Quality Management System (QMS) to mitigate and monitor these undesirable
effects and hazards. Expected rate of occurrence/frequency of hazards is especially useful in
post-market when attempting to determine whether a particular adverse event is occurring
more than it ought to.
Strategies to mitigate and minimise these risks such as contraindications or warnings in the IFU,
check lists, educational initiatives, patient cards and any others documents supplied with the
device should be discussed, including the expected impact of these risk mitigation and
minimisation strategies. Residual risks that remain after the implementation of risk mitigation
strategies should be identified – where required these should be addressed through post-market
surveillance, including post-market clinical follow-up (PMCF) studies. Documentation of the risk
management and QMS is necessary to allow the clinical expert to comment on whether these
have been mitigated adequately and to draw conclusions on the overall benefit-risk profile of the
device. ISO 14971:2019 can provide further guidance on this.
The risk management documentation should be reviewed and updated throughout the lifecycle
of the device. When new risks or issues arise post-market, these should be incorporated into risk
management documents.

Instructions for use (IFU), labelling and other documents supplied with the
device
The IFU, product manuals, patient cards, labelling and promotional materials, surgical
technique/instructions and other documents should be provided. These must highlight the risks
and ensure that they are appropriately communicated to user. They should be commented on in
the CER, as they relate to identification of relevant issues regarding safety and performance, and
in some instances it may be useful to include particular documents in the CER itself (i.e., as
appendices). The IFU should take into account who may use the device. For example, self-use
devices may require an IFU that is aimed at a different audience compared with devices
intended to be used by a medically qualified person.
The clinical evaluation should discuss the supporting documents, including whether these are
consistent with the clinical evidence, with particular attention paid to indications for use, target
population, contraindications and adverse events. The IFU should include all identified hazards
and other clinically relevant information that may impact on the use of the device and sufficient
warnings to mitigate risks where possible. Foreseeable safety or performance concerns that may
arise from the hazards identified in the IFU, labelling and other documents should have been
identified and incorporated into the overall benefit-risk analysis.

When available, the clinical assessment report from a European Union notified
body may aid timely clinical review of the submission.

Additional information on the device

Further description of the device may be required. As a guide, sufficient detail would generally
be provided through satisfying the requirements of Appendix 3 of MEDDEV 2.7.1 Rev 4 on
Device description – typical contents.

Pre-clinical data (if relevant)

Medical devices may contain elements that cannot be assessed solely through clinical testing, but
which are critical to the safety or performance of the device. In such cases, a concise summary of

Clinical evidence guidelines: Medical devices Page 44 of 178

V3.1 June 2022
 Therapeutic Goods Administration

the pre-clinical data may be required to establish the safety and performance profile for the
device. In some cases, it may be relevant to include a summary of the following in the supporting
documents when recommended for the device type by relevant ISO technical specifications,
standards or by other international regulatory agencies such as the US FDA:
• physical and chemical analyses
• engineering assessment
• sterilisation and stability
• microbiology
• in vivo and in vitro testing
• engineering studies under simulated conditions of use
• modelling data
• Good Laboratory Practices (GLP) testing

For applications for inclusion based on EU certification which are selected for
clinical audit only, pre-clinical data should be presented in a manner relevant to
the TGA clinical assessment.

Clinical investigation reports

This may be provided in the form of full study reports or peer reviewed publications. Full study
reports should include significantly more detail than peer-reviewed publications and may be the
most appropriate evidence form in some circumstances. Clinical investigation reports should
include the design, subject selection and inclusion/exclusion criteria, population demographics,
duration, safety and performance data, adverse events and complications, patient
discontinuation, device failures and replacements, tabulations of data from all individual subject
reporting forms and copies of such forms for each subject who died during a clinical
investigation or who did not complete the investigation, results of statistical analyses of the
clinical investigations, contraindications and precautions for use of the device, and other
information, as appropriate.

Literature search and selection strategy

It is recommended that the full electronic search strategy for at least one database searched, and
the strategy for selecting studies which were included in the review, are covered in this part of
the supporting documents as a way to demonstrate the rigour of the search and selection
strategy. The search strategy should include a summary justification as to how each citation did
or did not fit the selection criteria for inclusion.

Pivotal articles from the literature review

The full text of pivotal articles in the literature review contributing to the clinical evidence base
used to demonstrate compliance with the EPs should be provided.

Post-market surveillance reports

Post-market surveillance (PMS) reports (or equivalent) are one way to present the post-market
data relevant to a device. Where reports cover a range of devices, it may be necessary to stratify
data by device type to provide satisfactory clinical evidence for the subject device/s. Similarly,
data stratification by year, geographical region, or clinical indication may also be required to

Clinical evidence guidelines: Medical devices Page 45 of 178

V3.1 June 2022
 Therapeutic Goods Administration

meet regulatory requirements. Where a PMS report does not contain all the post-market data
required, then additional datasets and critical analysis may be required.

Clinical evidence checklist

The following checklist sets out the recommended heading structure for the CER.
Indicate that each of the relevant recommended sections has been included, who authored each
section, and on which page(s) they can be located within the CER.

Section Included Author(s) Page(s)

1. General details and device Yes No N/A

description -including
GMDN, UDI, lineage and
version (if applicable)

2. Intended purpose, Yes No N/A

indications and claims

3. Developmental context Yes No N/A

and state of the art

4. Regulatory status in other Yes No N/A

countries (including
evidence and supporting
documents)

5. Summary of relevant pre- Yes No N/A

clinical data (if applicable)

6. Demonstration of Yes No N/A

substantial equivalence or
comparability (if applicable)

7. Summary and appraisal of Yes No N/A

clinical data

8. Data analysis and benefit- Yes No N/A

risk analysis

9. Conclusions Yes No N/A

10. Name, signature and Yes No N/A

curriculum vitae of clinical
expert and date of report

11. Risk assessment and Yes No N/A

management documents

12. IFU, labelling and other Yes No N/A

documents supplied with
the device

Clinical evidence guidelines: Medical devices Page 46 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Section Included Author(s) Page(s)

13. Full clinical investigation Yes No N/A

reports

14. Literature search and Yes No N/A

selection strategy

15. Full text of pivotal articles Yes No N/A

from the literature review

16. Post-market surveillance Yes No N/A

reports

17. Additional relevant Yes No N/A

information on the device
(if applicable)

Clinical evidence guidelines: Medical devices Page 47 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Part 2 – Special topics

Comparable devices including substantially equivalent

devices
The International Medical Device Regulators Forum (IMDRF) defines a comparable device as: ‘A
medical device with related function chosen by the manufacturer to inform the clinical
evaluation of the device in question’.
The first step in the comparability process is for the manufacturer to select an appropriate
comparable device. The comparable device should have (or include) the same intended purpose.
Substantially equivalent devices are a subset of comparable devices where the devices are
similar to such an extent that there would be no clinically significant difference in safety and
performance.
Note: it is not a requirement that a comparable or substantially equivalent device be included in
the ARTG. However, regulatory approval in Australia and/or other jurisdiction(s) will be taken
into account by the TGA alongside the quality and applicability of the evidence.

Comparable devices
The extent to which clinical evidence for a comparable device can be used to support safety
and performance of the subject device will depend on how similar the devices are. Comparable
devices should be considered with respect to clinical, technical and biological characteristics.
For more information refer to Comparing characteristics.
To inform the clinical evaluation, these characteristics should be broadly similar, but
consideration should be given to how differences may affect the clinical safety and performance
of the device.
Comparable devices should also belong to the same generic family in terms of their intended use
and commonality of technology.
Clinical evidence for comparable devices may provide important background information on the
course of a disease, current state of the art and treatment options and the evolution of device
technologies. The risks identified for comparable devices may help inform the risk management
of the subject device. Scientific literature on comparable devices may assist in establishing
performance and safety measures and benchmarks, including rates of adverse events.
Clinical evidence for comparable devices that are not substantially equivalent may form part of
clinical evaluation and may support or supplement direct clinical data. It will not typically, in
itself, constitute sufficient clinical evidence for the purpose of demonstrating compliance with
the EPs - except for certain categories of lower risk devices, whose clinical effects are well
understood and characterised.

Substantially equivalent devices

Substantially equivalent devices are those that are the most similar to the subject device, to such
an extent that there would be no clinically significant difference in safety and performance.
Clinical data for substantially equivalent devices can be used as indirect evidence for the subject
device – it may constitute the sole or major clinical data source for demonstrating compliance
with the EPs.

Clinical evidence guidelines: Medical devices Page 48 of 178

V3.1 June 2022
 Therapeutic Goods Administration

The TGA approach to substantial equivalence differs slightly from other jurisdictions, partly due
to differences in legislation.
For example, European guidance requires that equivalence claims for implantable devices and
class III devices not from the same manufacturer ‘must have a contract in place that allows full
access to the technical documentation on an ongoing basis’. 2
The FDA (via the 510(k) pathway) requires that the new device is as safe and effective as the
predicate 3 though the FDA and TGA have different definitions of the term ‘predicate’.
Our position for determining substantial equivalence is that:
• all relevant clinical, technical and biological characteristics of the devices should be
compared (preferably with the aid of tables) and discussed
• a sufficiently detailed critical analysis should demonstrate that the devices are similar to
such an extent that there would be no clinically significant difference in safety and
performance
• a suitable clinical expert must endorse the above.
Where the manufacturer’s technical data is not available for a comparable device, a robust
method of analysis that quantifies this may be provided as an alternative. Reference to an
applicable ISO standard may be of assistance. If a comparable regulator has previously
determined the devices to be substantially equivalent, we will consider this factor in our
assessment.
In general, if substantial equivalence between two devices cannot be demonstrated then
direct clinical evidence for the subject device will be required to demonstrate compliance
with the EPs.

Predicates
Predicates are comparable devices that represent a logical starting point for gathering clinical
evidence. They can often be demonstrated to be substantially equivalent.
A predicate:
• is a previous iteration of the subject device
• has the same intended purpose as the subject device
• is within the same lineage of devices as the subject device
• is from the same manufacturer as the subject device.
Where there are multiple devices in a lineage, you may demonstrate substantial equivalence
between the subject device and any device whose clinical investigation data is being used as
indirect evidence in the clinical evaluation, even if that device was several iterations earlier in
the lineage.
However, be cautious when claiming substantial equivalence to a predicate very early in the
device lineage. You should consider the relevance of the comparison and the potential impact of
multiple incremental changes on safety and performance; such devices may be more

2 EU MDCG 2020-5 Clinical Evaluation – Equivalence. A guide for manufacturers and notified bodies.

April 2020
3 US FDA, ‘The 510(k) Program: Evaluating Substantial Equivalence in Premarket Notifications [510(k)].

Guidance for Industry and Food and Drug Administration Staff. Document issued on July 28, 2014.
Clinical evidence guidelines: Medical devices Page 49 of 178
V3.1 June 2022
 Therapeutic Goods Administration

appropriately viewed as comparable devices. Direct clinical evidence for the subject device and
revision of the CER to include this data should occur as early as is practicable.

Example: Devices A, B and C are all in the same lineage of orthopaedic prostheses.
Device C is being assessed. Device B was the most recent predicate, but the main
clinical investigation data was derived from an earlier iteration, Device A.
Substantial equivalence must be demonstrated between Device C and A to enable
this clinical data to be accepted as indirect evidence for demonstrating compliance
with the EPs. Substantial equivalence could also be claimed for Device B – this
would strengthen the evidential value of its post-market data.

Comparing device characteristics

To determine the degree of comparability of devices, we consider the clinical, technical and
biological characteristics. Higher risk devices require a more thorough and comprehensive
evaluation of these characteristics. Once each characteristic and its associated elements (see
below) have been compared, we can determine the degree of similarity. To demonstrate
substantial equivalence, the devices must be similar to such an extent that there would be no
clinically significant difference in clinical safety and performance.
Comparisons are ideally based on a single device or lineage.

Comparisons that consider multiple devices

In general, you should not provide comparisons that consider multiple devices (unless these are
from the same lineage). For instance, it would not be appropriate to compare technical
characteristics of a subject device to multiple different devices, without considering that the
overall combination of characteristics of the subject device may be novel and hence have new
safety and performance implications.
Comparisons may address multiple devices in the following circumstances:
• Where a subject device is part of a system of devices. In this instance, comparisons may
address the multiple devices within the system, including how the individual components of
the system interact and the safety and performance of the system as a whole.
• Where technical differences between devices are present. Discussion of the impact of
the technical difference may be supported by a relevant example from a third device.
In these instances, the clinical evidence for the comparable devices must allow sufficient
conclusions to be drawn regarding the safety and performance of the subject device.
You must identify any safety or performance concerns regarding a comparable device, including
matters subject to regulatory review or action (in Australia or another jurisdiction). Whether
design and usage issues have been appropriately addressed in the subject device will be an
important consideration in determining compliance with the EPs. Sponsors must provide
detailed, reasoned arguments supported by appropriate evidence.

Clinical characteristics/intended purpose

Firstly, consider and compare the clinical characteristics which relate to the intended purpose of
the devices.

Clinical evidence guidelines: Medical devices Page 50 of 178

V3.1 June 2022
 Therapeutic Goods Administration

The intended purpose should refer to the clinical condition being treated, the severity and stage
of disease, the site of application to/in the body and the patient population. You should clearly
state this information in your submission.
The TGA will also assess whether the intended purpose is consistent with information provided
with the device, in the IFU, labelling, any advertising material for the devices and technical
documentation.
Elements to consider include, but are not limited to, the comparability of the subject device and
comparable device with respect to:
• indications for use, including the disease or condition the medical device will diagnose, treat,
prevent, cure or mitigate
• patient population (for example, age, gender, anatomy, physiology)
• the site of application to/in the body (organs, parts of the body, tissues or body fluids
contacted by the medical device)
• type of contact (for example, contact with mucosal membranes, invasiveness, implantation)
• duration of use or contact with the body
• environment of use (for example, healthcare facility, home)
• intended user (for example, use by health care professional, lay person)
• repeat applications, including any restrictions as to the number or duration of applications.
In general, the intended purpose must be the same for substantial equivalence to be considered.
However, if the intended purpose differs (for example if it is narrower than, but encompassed
by, the intended purpose for the comparable device), clinical evidence demonstrating safety and
performance for the comparable device specific to the intended purpose for the subject
device may be considered as part of a substantial equivalence claim.

Technical characteristics
Technical characteristics should be broadly similar between the comparable devices, including
but not limited to:
• design, for example:
– dimensions and design tolerances
– how the different components of the device system work together
• material, for example:
– chemical formulation
– additives
– processing such as forged
– state such as crystalline
• specifications and properties, for example:
– physiochemical properties such as type and intensity of energy
– wavelength

Clinical evidence guidelines: Medical devices Page 51 of 178

V3.1 June 2022
 Therapeutic Goods Administration

– porosity
– particle size
– viscosity
– nanotechnology
– specific mass
– atomic inclusions such as nitrocarburising
– oxidability
– tensile strength and degradation characteristics
• deployment methods (where applicable)
• critical performance requirements
• principles of operation
• software algorithms

Biological characteristics
Biological characteristics should be broadly similar between the comparable devices, including
but not limited to:
• biocompatibility of materials in contact with body fluids/tissues
• biological action
• degradation mechanism and profile
• biological response, for example:
– inflammatory response
– immune response
– tissue integration

Presentation of information regarding comparable devices

To make a comprehensive comparison, it is best that you:
• use tables which provide a description of the characteristics for the two devices and note
both their similarities and differences (see example below)
• clearly and explicitly state all differences between the devices
• critically analyse these differences to determine their likely impact on safety and
performance.
We highly recommend including illustrations of the devices highlighting similarities and
differences.
Your conclusions regarding the impact of differences on device safety and performance must be
endorsed by a suitable clinical expert who has:
• relevant medical qualifications

Clinical evidence guidelines: Medical devices Page 52 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• direct clinical experience in the use of the device or device type in a clinical setting.
A full curriculum vitae of the clinical expert should be provided.
If it can be established that the clinical, technical and/or biological differences would result in no
clinically significant difference in device safety and performance, then the comparable device
may be considered ‘substantially equivalent’ to the subject device.
In order to establish substantial equivalence, the differences between the two devices will need
to be minimal. You need to provide evidence to substantiate your claim, such as pre-clinical
(bench testing or in vivo studies) and/or clinical (clinical investigation or post-market) data.
Multiple and/or major differences will compromise claims of substantial equivalence. In these
scenarios, we may consider the devices ‘comparable’ subject to the facts.
It is your responsibility to ensure that all relevant information relating to the comparable device
is provided for clinical assessment – in particular the clinical data that demonstrates its safety
and performance.

Clinical evidence guidelines: Medical devices Page 53 of 178

V3.1 June 2022
 Therapeutic Goods Administration

The following flowchart provides guidance on how to demonstrate substantial equivalence:

Figure 2. Identifying substantially equivalent and comparable devices

Clinical evidence guidelines: Medical devices Page 54 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Table 2: Example summary table

The following table template provides an example of how to present information regarding
comparable devices.

Evidence presented The device Comparable Impact of difference

device on safety and
performance

Clinical characteristics

{e.g. clinical condition

treated, intended
purpose/ indications,
site in body, operational
procedures, target
population including
age, anatomy,
physiology}

Technical characteristics

{e.g. materials, design,

function, energy source
etc.}
{e.g. deployment
methods}

Biological characteristics

{e.g. biocompatibility}

Magnetic resonance imaging (MRI) considerations

Addressed in this section are the clinical and pre-clinical evidence requirements to demonstrate
the safety and performance of Implantable Medical Devices (IMDs) in the Magnetic Resonance
(MR) environment. Active IMDs (AIMDs) are implanted devices that depend on a source of
energy for their operation and convert energy, whilst passive IMDs (PIMDs) are those that do
not have such a requirement. The evidence considered in this section applies to:
• Active Implantable Medical Devices (AIMDs), including but not limited to:
– implantable permanent pacemakers (PPM)
– implantable cardioverter defibrillators (ICD)
– cardiac resynchronisation therapy (CRT) devices
– implantable loop recorders (ILR); and
– the associated leads.

Clinical evidence guidelines: Medical devices Page 55 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• Passive Implantable Medical Devices (PIMDs), including but not limited to:
– orthopaedic implants such as hip or knee implants
– cardiovascular stents
– heart valves
– neurovascular aneurysm clips or coils
– interventional guidewires or catheters
Each unique type of IMD system has its own associated risk-benefit profile that needs to be
addressed by the manufacturer.

Summary recommendations
• AIMDs and many PIMDs, for example orthopaedic implants, are complex medical devices
forming systems of multiple independent components. The unique configuration of
components for each device system may have consequences for the safety of the device
system in the MR environment. Therefore, manufacturers are advised to provide
appropriate evidence to support the safety and identify the risks and hazards of each unique
device system separately. Due to the nature of their materials, currently available AIMDs can
only be marked as ‘MR conditional’ or ‘MR unsafe’. PIMDs can be marked as ‘MR safe’, ‘MR
conditional’ or ‘MR unsafe’.
• For IMDs claimed to be ‘MR conditional’ under specified conditions of use, these conditions
must be clearly articulated in the submission and in the IFU, and/or other supporting
documents with evidence supporting any reported thresholds.
• For PIMDs, the use of non-clinical data alone suffices to meet the requirements for the
applicable EPs. Clinical data are not required.
• A well-documented risk analysis and management system and quality management system
should be provided with the CER.
• Provision of clinical data for AIMDs if applicable:
– Post-market data or clinical investigations from another jurisdiction where the device
is already approved can provide useful clinical evidence and are acceptable. This
includes clinically indicated MRIs provided that potential sources of bias have been
minimised. Studies should be appropriate to inform on the safety and performance of
the device for its intended purpose in relation to MR conditional use.
– examples of appropriate safety outcomes are provided in Table 26 - Safety of active
implantable medical devices in the MR environment.
– when submitting a comprehensive literature review, full details of the method used
should be included in the CER in sufficient detail to ensure the literature review can be
reproduced.
– for guidance on the presentation of clinical evidence and conduct of comprehensive
literature reviews manufacturers are directed to relevant sections.

Defining ‘safety’ in the MR environment

The specific terminology used to define the safety of medical devices in the MR environment is
outlined in ASTM Standard F2503-13, “Standard Practice for Marking Medical Devices and Other
Items for Safety in the Magnetic Resonance Environment”. In this context, the term “MR

Clinical evidence guidelines: Medical devices Page 56 of 178

V3.1 June 2022
 Therapeutic Goods Administration

environment” refers to the physical space surrounding a MR magnet, which is affected by the
static, gradient and radiofrequency (RF) electromagnetic fields. Standard F2503-13 defines
three terms to classify the safety of medical devices in the MR environment:
• MR safe: An item that poses no known hazards resulting from exposure to any MR
environment. A medical device can only be classified as MR safe if it is composed of materials
that are electrically non-conductive, non-metallic, and non-magnetic (e.g. glass, plastic,
silicone). Such devices may be determined to be MR safe based on scientific rationale rather
than test data;
• MR conditional: An item with demonstrated safety in the MR environment within defined
conditions. Minimum requirements for demonstrating conditional MR safety requires
consideration of the possible interactions between the device and the static, gradient and
radiofrequency fields present in the MR environment, and consideration of MR image
artefacts from the implants. Known potential hazards related to the use of AIMDs in the MR
environment that should be addressed in order to demonstrate conditional safety are
outlined in Table 3 (below).
• MR unsafe: An item that poses unacceptable risks to patients, medical staff or other persons
in the MR environment.
Table 3: Known potential hazards for active implantable medical devices in the MR
environment related to the static, gradient and radiofrequency fields

Static Gradient Radiofrequency

MR hazard/clinical impact
field field field

Force and torque/discomfort, dislodgement Ÿ

Vibration/discomfort, device damage Ÿ Ÿ

Device interactions/therapy delivery, device

Ÿ Ÿ Ÿ
reset, device damage

Device case heating/discomfort, tissue necrosis Ÿ Ÿ

Unintended cardiac stimulation/arrhythmia

Ÿ Ÿ
induction, asystole

Lead electrode heating/therapy delivery,

Ÿ
sensing

MR = magnetic resonance. Table source: Gold et al 2015.

Evidence requirements
Evidence requirements to demonstrate the safety of an IMD system in the MR environment will
vary depending on whether the device is labelled as ‘MR safe’, ‘MR conditional’, or ‘MR unsafe’:
• Device systems claimed to be ‘MR safe’ must be shown to be non-conducting, non-metallic,
and non-magnetic in order to satisfy the applicable EPs. A scientifically based rationale to
demonstrate that the device poses no known hazards in all possible MR imaging
environments may be sufficient. It is unlikely that any AIMD systems currently available
would be designated as MR safe.

Clinical evidence guidelines: Medical devices Page 57 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• Device systems claimed to be ‘MR conditional’ must be shown to pose no known hazards in
the MR environment under specific conditions. For ‘MR conditional’ PIMD systems, the
requirements may be satisfied with non-clinical data alone. In any case, the data should be
accompanied by appropriate warnings and specified conditions of use, outlined in the
instructions for use (IFU) and/or manual and other easily accessible documents.
Other information that should be provided for IMDs includes:
• the technical specification of the device(s)
• the components to which the device is paired when used clinically, for example the pulse
generator with its lead(s)
• scanning exclusion zones implemented
• a risk analysis and management document.

Requirements for PIMDs

For PIMDs claimed to be ‘MR conditional’, the following experimental data are required using
non-clinical testing methods specified in the standards below or equivalent methods.
• Magnetically Induced Displacement Force: ASTM F2052-14, Standard Test Method for
Measurement of Magnetically Induced Displacement Force on Medical Devices in the
Magnetic Resonance Environment
• Magnetically Induced Torque: ASTM F2213-06 (Reapproved 2011), Standard Test Method
for Measurement of Magnetically Induced Torque on Medical Devices in the Magnetic
Resonance Environment
• Heating by RF Fields: ASTM F2182-11a, Standard Test Method for Measurement of Radio
Frequency Induced Heating Near Passive Implants During Magnetic Resonance Imaging
• Image Artifact: ASTM F2119-07 (Reapproved 2013), Standard Test Method for Evaluation of
MR Image Artifacts from Passive Implants
If the testing does not include all sizes of the device, a size or combination of sizes that represent
the worst-case scenario for each test should be included in the testing. A rationale should be
included for determining why the selected size(s) represent the worst-case scenario for each
test.
All testing protocols should be described with the following elements:
• test objective
• equipment used
• acceptance criteria
• rationale for test conditions
• rationale for the acceptance criteria
• number of devices tested
• description of devices tested, including device size
• description of any differences between test sample and final product, and justification for
why differences would not impact the applicability of the test to the final product
• results (summarised and raw form).

Clinical evidence guidelines: Medical devices Page 58 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Regulatory status in other jurisdictions

If the IMD or comparable device is approved for use in another jurisdiction, the manufacturer or
sponsor should provide regulatory status, including the certificate number, date of issue and
name under which the device is marketed, exact wording of the intended purpose, MR status in
key jurisdictions, for example the US, EU, Japan and Canada and IFU used in other jurisdictions.

Post-market data
Information arising from product experience in Australia or other jurisdictions where a device is
already in use adds to the clinical evidence for pre- and post-market reviews. The following
information should be provided if available:
• all product recalls, including for product correction, suspensions, removals, cancellations
and withdrawals, whether withdrawals of indications or the device(s), amendments to the
IFU or other key documents such as product manuals, or any other corrective actions in any
jurisdiction
• distribution numbers of the device(s) including by country and/or geographical region for
every year since launch. It is accepted that this may not always be appropriate for high
volume devices, those with many components or those on the market for many years
• the number of years of use
• for every year since launch data from post-market vigilance and monitoring reports, adverse
events and complaints for IMDs and comparable devices categorised by type (e.g. device
reset, device failure, induced arrhythmia, etc.) and clinical outcomes (e.g. death or serious
harm, etc.) as reported to regulatory bodies
• post-market data from other jurisdictions can be used to support an application for MR
conditional use only if the MR status and MR conditions of use in the other jurisdictions are
fully specified including the device combinations used
• explanted devices returned to manufacturers should be accounted for with an explanation of
device failures and corrective measures.

Defining active implantable medical devices

An active medical device is a device that uses and converts energy in a significant way in order
to operate. Active devices may use any form of energy except for gravitational or direct human
energies. Active medical devices can be broadly characterised to serve two main purposes, as
defined in the Therapeutic Goods (Medical Devices) Regulations 2002:
• Active medical devices for diagnosis are intended by the manufacturer to be used on a
human being, either alone or in combination with another medical device, to supply
information for the purpose of detecting, diagnosing, monitoring or treating physiological
conditions, states of health, illnesses or congenital deformities.
• Active medical devices for therapy are intended by the manufacturer to be used on a
human being, either alone or in combination with another medical device, to support,
modify, replace or restore biological functions or structures for the purpose of treating or
alleviating an illness, injury or handicap.
Active implantable medical devices are further defined in the Regulations as:
Active implantable medical devices
An active medical device, other than an implantable medical device, that is intended by the
manufacturer:
a) either:

Clinical evidence guidelines: Medical devices Page 59 of 178

V3.1 June 2022
 Therapeutic Goods Administration

i. to be, by surgical or medical intervention, introduced wholly,

or partially, into the body of a human being; or
ii. to be, by medical intervention, introduced into a natural
orifice in the body of a human being; and
b) to remain in place after the procedure.
Implantable permanent pacemakers (PPM), implantable cardioverter defibrillators (ICD),
cardiac resynchronisation therapy (CRT) devices, implantable loop recorders (ILR); and their
leads are a subclass of active implantable medical devices that are used to monitor and/or
regulate cardiac rhythm.
In serving this purpose these devices may simultaneously function as both therapeutic and
diagnostic devices. While there are subtle differences in the design and purpose of these
different cardiac devices, they typically include:
• circuitry that controls the timing and intensity of electrical impulses delivered to the heart
• a battery used to generate electrical impulses and power the circuitry
• a case that encloses the circuitry and battery
• pacing lead(s) that deliver electrical impulses between the circuitry and the chambers of
the heart
• a connector block that connects the pacing lead(s) to the case.
Different configurations of the above design characteristics are used to treat different medical
conditions:
Permanent pacemakers (PPM) are pacing devices used to regulate abnormal heart rhythm.
PPMs deliver low-energy electrical impulses to treat bradyarrhythmias. They may include one
pacing lead for single-chamber right ventricular pacing, or two pacing leads for right ventricular
and right atrial pacing.
Implantable cardioverter defibrillators (ICD) are capable of delivering both low-energy
impulses for pacing, and high-energy impulses for defibrillation. ICDs are typically implanted in
patients at risk of life-threatening ventricular arrhythmias, in whom a high-energy impulse is
required to restore normal rhythm. ICDs typically have a larger battery than a PPM, and include
one lead for right ventricular pacing and defibrillation, +/- another lead for right atrial pacing.
Cardiac resynchronisation therapy (CRT) devices are pacing devices used to regulate the lack
of synchrony between the left and right ventricles. CRT devices are typically used to treat
patients with advanced heart failure. They include either two or three pacing leads for right
ventricle, left ventricle, +/- right atrial pacing. CRT devices may also deliver high-energy
impulses to correct life-threatening arrhythmias (CRT-Ds).
Implantable loop recorders (ILR) are single-lead cardiac monitoring devices. They can be
used as a temporary tool to diagnose patients with unexplained palpitations or syncope, or for
long-term monitoring of patients with unresolved syncope who may be at risk of atrial
fibrillation. Unlike other classes of active implantable cardiac devices, they are not capable of
pacing or defibrillation.
Regardless of the type of AIMD, it is recommended that manufacturers provide the following
information regarding the physical and chemical characteristics of the device. These
characteristics include, but are not limited to:
• the materials from which the device components are made, including the chemical
composition

Clinical evidence guidelines: Medical devices Page 60 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• the dimensions and geometry of the device components

• the list of other devices that are likely to be used in conjunction with the device.

Summary of safety and performance data

Selection of included studies

Table 4: Summary of primary studies report in narrative reviews on the safety of AIMDs
in the MR environment

Characteristics of Evidence reported in narrative reviews

included studies

Dominant design of 3 RCTs, 1 case-control and 38 case series investigations were

included studies included in narrative review articles

Sample size range RCTs: 263-466

for included study
designs Case-control: 65
Case series: 1 to 272

Patient follow-up Range 0-12 months (median 3 months)

Safety outcomes Force and torque

reported
• Generator movement
• Lead dislodgement
• Lead damage
• Force (Newtons)
Vibration
• Generator movement
• Patient discomfort due to vibration
Device interactions
• Reed switch activation/deactivation
• Diminished battery voltage (≥ 0.04 V)
• Power-on-reset
• Temporary communication failure with device
• Device reprogramming
• Pause in pacing
• Signal (image) artefacts
Device case heating

Clinical evidence guidelines: Medical devices Page 61 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Characteristics of Evidence reported in narrative reviews

included studies

• Detectable heat increase near generator

Lead electrode heating
• Increase in pacing capture threshold (≥ 0.5 V)
• Increase in cardiac enzyme level (Troponin-I)
• Decrease in atrial sensing amplitude ≥50%, or amplitude lower
than 1.5 mV 255
• Decrease in ventricular sensing amplitude ≥ 50%, or amplitude
lower than 5.0 mV 256
• Change in pacing lead impedance (≥ 50 Ω)
Unintended cardiac stimulation
• Inappropriate pacing
• Induction of arrhythmia
• Heart palpitations

References
ASTM F2503-13, Standard Practice for Marking Medical Devices and Other Items for Safety in
the Magnetic Resonance Environment, ASTM International, West Conshohocken, PA, 2013
Shellock FG, Woods TO, Crues JV, 3rd. MR labeling information for implants and devices:
explanation of terminology. Radiology. 2009;253(1):26-30
Gold MR, Kanal E, Schwitter J, Sommer T, Yoon H, Ellingson M, et al. Preclinical evaluation of
implantable cardioverter-defibrillator developed for magnetic resonance imaging use. Heart
Rhythm. 2014
U.S. FDA. Establishing safety and compatibility of passive implants in the magnetic resonance
(MR) environment: Guidance for industry and Food and Drug Administration staff.
RockvilleMD2014 [updated 2014]
American Society for Testing and Materials International. Designation: ASTM F2052-14,
standard test method for measurement of magnetically induced displacement force on medical
devices in the magnetic resonance environment. West Conshohocken, Pa: American Society for
Testing and Materials International, 2014.
American Society for Testing and Materials International. Designation: ASTM F2213-06
(Reapproved 2011), standard test method for measurement of magnetically induced torque on
medical devices in the magnetic resonance environment. West Conshohocken, Pa: American
Society for Testing and Materials International, 2011.
American Society for Testing and Materials International. Designation: ASTM F2182-11a,
standard test method for measurement of radio frequency induced heating on or near passive
implants during magnetic resonance imaging. West Conshohocken, Pa: American Society for
Testing and Materials International, 2011.
American Society for Testing and Materials International. Designation: ASTM F2119-07
(Reapproved 2013), standard test method for evaluation of MR image artifacts from passive

Clinical evidence guidelines: Medical devices Page 62 of 178

V3.1 June 2022
 Therapeutic Goods Administration

implants. West Conshohocken, Pa: American Society for Testing and Materials International,
2013.
Patel KH, Lambiase PD. The subcutaneous ICD-current evidence and challenges. Cardiovasc
Diagn Ther. 2014;4(6):449-59
Linde C, Ellenbogen K, McAlister FA. Cardiac resynchronization therapy (CRT): clinical trials,
guidelines, and target populations. Heart Rhythm. 2012;9(8 Suppl):S3-s13
Krahn AD, Klein GJ, Yee R, Skanes AC. The use of monitoring strategies in patients with
unexplained syncope--role of the external and implantable loop recorder. Clin Auton Res.
2004;14 Suppl 1:55-61
Ahmed FZ, Morris GM, Allen S, Khattar R, Mamas M, Zaidi A. Not all pacemakers are created
equal: MRI conditional pacemaker and lead technology. J Cardiovasc Electrophysiol.
2013;24(9):1059-65
Ainslie M, Miller C, Brown B, Schmitt M. Cardiac MRI of patients with implanted electrical cardiac
devices. Heart. 2014;100(5):363-9
Beinart R, Nazarian S. Effects of external electrical and magnetic fields on pacemakers and
defibrillators: from engineering principles to clinical practice. Circulation. 2013;128(25):2799-
809
Chow GV, Nazarian S. MRI for patients with cardiac implantable electrical devices. Cardiol Clin.
2014;32(2):299-304
Ferreira AM, Costa F, Tralhao A, Marques H, Cardim N, Adragao P. MRIi-conditional pacemakers:
Current perspectives. Medical Devices: Evidence and Research. 2014;7(1):115-24
Nazarian S, Beinart R, Halperin HR. Magnetic resonance imaging and implantable devices. Circ
Arrhythm Electrophysiol. 2013;6(2):419-28
Shinbane JS, Colletti PM, Shellock FG. Magnetic resonance imaging in patients with cardiac
pacemakers: era of "MR Conditional" designs. J Cardiovasc Magn Reson. 2011;13:63
van der Graaf AWM, Bhagirath P, Gotte MJW. MRI and cardiac implantable electronic devices;
current status and required safety conditions. Neth Heart J. 2014;22(6):269-76

Personalised medical devices (PMDs)

This section provides guidance on expectations in relation to the nature and type of clinical
evidence generally necessary for demonstrating the performance and safety of personalised
medical devices (PMDs) and meeting the relevant provisions of Essential Principles (EPs). The
following section is intended to supplement the general and device-specific recommendations
outlined in other parts of these guidelines and any applicable standards. PMDs will generally
follow the same regulatory requirements and submission expectations as non-PMDs of the same
classification; however, there are some additional considerations that apply to PMDs.
The TGA uses three specific terms to describe the current range of PMDs. These terms are
defined in the Therapeutic Goods (Medical Devices) Regulations 2002:
• Patient-matched medical devices;
• Adaptable medical devices; and
• Custom-made medical devices.

Clinical evidence guidelines: Medical devices Page 63 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Note: Medical device production systems (MDPS) are a new regulatory concept (not currently in
effect) designed to provide options for healthcare facilities wanting to produce in-house medical
devices for use in treating their patients. While a definition of MDPS is in the MD Regulations,
further processes (including a legislative instrument) are required to be put in place to
implement the concept. Further discussion of MDPS is not covered in this section.
The overall framework for the regulation of PMDs is explained in more detail on the TGA
website including in relation to the future implementation of MDPS.
Personalised medical devices range from low-risk devices such as personalised shoe inserts and
dental aligners to high-risk devices such as permanent orthopaedic implants. A wide range of
technologies may be employed in the manufacture of PMDs, and this helps inform the clinical
evidence strategy. Since each PMD is manufactured and/or adapted to address unique
requirements (e.g. anatomical and/or physiological features of a particular individual), no two
PMDs are likely to be identical. The uniqueness in design, manufacture, and/or point-of-care
modification of PMDs poses additional challenges in designing and conducting clinical
investigations, and for manufacturers conducting clinical evaluation to demonstrate
performance and safety and hence compliance with the EPs throughout the device lifecycle.
In particular, this section provides guidance on the clinical evidence requirements for PMDs
with reference to their particular type. PMD manufacturers are encouraged to use these
recommendations as a guide when developing their strategy for generating clinical evidence.

Summary recommendations

Guiding principle
Whilst generating clinical evidence for PMDs poses additional challenges when designing and
conducting clinical studies, due to device heterogeneity, the same principles that apply to clinical
study design for non-PMD devices should still be applied to PMDs. For both pre-market
approvals, and in addressing post market issues, a legitimate and reasoned approach to clinical
data generation, alongside a critical analysis regarding its limitations, will be viewed more
favourably than a paucity or absence of clinical data. Whilst post market clinical follow up
studies and real world data may form a significant element of a manufacturer’s strategy for
generating clinical evidence throughout the device lifecycle, especially in relation to addressing
residual risks, they do not lessen the need for well-designed clinical investigation studies for
pre-market approval of higher risk devices.

Types of PMDs
PMDs are a heterogenous group of devices and are required to comply with the relevant
provisions of the EPs. They consist of:
• Patient-matched medical devices, which are defined in terms of the ‘specified design
envelope’ (which is also defined in the MD Regulations). Manufacturers should consider
generalisability – the extent to which available clinical data can be extrapolated to all
potential device specifications within the design envelope – in addition to the clinical
evidence requirements discussed in Part 1 – General Requirements. This should include
discussions about worst-case and common-use scenarios.
• Adaptable medical devices, which are defined in terms of point-of-care modifications.
Similar to patient-matched medical devices, the nature and extent of personalisation needs
to be considered and the clinical evidence must substantiate the safety and performance of
the device as modified, in accordance with the manufacturer’s instructions, after the device
has been supplied.

Clinical evidence guidelines: Medical devices Page 64 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• Custom-made medical devices, which are produced on a small scale. Whilst these devices
are exempted from inclusion on the ARTG (and hence pre-market submission of clinical
evidence), consideration should be given to whether unique design parameters are expected
to affect the device performance and safety. Custom-made medical devices must still comply
with all relevant provisions of the EPs and records must be maintained in relation to
performance and safety.
Note: a ‘low-volume’ exemption from ARTG inclusion also exists for patient-matched medical
devices where 5 or less such devices of a kind are manufactured in any financial year. Such
devices must still comply with all relevant provisions of the EPs, and records relating to
performance and safety be maintained.

Clinical Evidence
Clinical evidence, as discussed in Part 1 – General Requirements, remains an essential aspect of
design validation for medical devices and forms an important component of the technical
documentation to demonstrate conformity with the EPs.
Given the unique aspects of their design, the Total Product Life Cycle (TPLC) approach is
particularly important for PMDs and the clinical evidence should be reviewed and updated
periodically throughout the lifecycle of a PMD to ensure continued acceptability of the benefit-
risk determination. Claims made by the manufacturer about performance and safety of PMDs,
must be supported by the clinical evidence and include consideration for all personalised
elements.
In Part 1 of these guidelines, the TGA has provided information regarding key definitions and
concepts, clinical evidence requirements, sources of clinical data and clinical evaluation in
relation to medical devices, which are also applicable to PMDs.
The depth and extent of the clinical evidence should be appropriate to the risk-based
classification, novelty, and parameters involved with personalisation of the device.

Clinical evidence considerations for patient-matched medical devices

A patient-matched medical device is manufactured to match the anatomical and/or physiological
features, or to address a pathological condition, of a particular individual. The device is
manufactured within the parameters of a specified design envelope using production processes
that are capable of being either or both validated and verified, and of being reproduced.
A characteristic feature of a patient-matched medical device type is the specified design
envelope.

Specified design envelope

Regardless of the risk-based classification of a medical device, the concept of specified design
envelope is applicable to devices coming under the definition of patient-matched medical
devices (for example patient-matched plagiocephaly helmets, or patient-matched 3D printed
orthognathic surgical plates). A specified design envelope can be conceived of as a set of all
relevant parameters (minimum and maximum dimensions, performance limits or other relevant
factors) that characterize a patient-matched medical device for production purposes. These
parameters or factors may be based on a standard device template (Figure 1 refers). The
manufacturer should unequivocally identify all relevant parameters that constitute the specified

Clinical evidence guidelines: Medical devices Page 65 of 178

V3.1 June 2022
 Therapeutic Goods Administration

design envelope and explicitly establish the boundaries 4 (reference intervals/categories) for
each parameter.

Figure 1. A template for depicting potential parameters within a design envelope schema

Parameters that characterise a design envelope may be divided broadly into the six categories
set out below. Given the variety of technologies, materials and processes used in the
manufacturing of medical devices, not all parameters may apply to every patient-matched
medical device.

i. Structural parameters
The manufacturer should establish explicit boundaries for the dimensions, area, volume,
shape, angles, relative positions, screw hole sizing and number, and other geometrical
parameters for the device. In this category, the manufacturer should also include any
patient-imaging data used in the device design process. Where the surface morphology
of the anatomy is used in the device design process, the manufacturer should specify
anatomical landmarks or margins to establish the geometrical limits on the device
design.

In addition to the external structural parameters for the device, where applicable, the
manufacturer should also establish design limits on the internal structural features of
the device, such as porosity, lattice strut size, wall thickness, etc.

ii. Material parameters

The manufacturer should identify all raw materials used in the production of the device
and their characteristics (biological, physical, chemical), and adhere to relevant material

4For the purposes of this document, boundaries mean the reference intervals (for a parameter that only
accepts numerical data) and categories (for a parameter that only accepts categorical data).

Clinical evidence guidelines: Medical devices Page 66 of 178

V3.1 June 2022
 Therapeutic Goods Administration

standards. For example, additively manufactured orthopaedic implants may utilize Ti-
6Al-4V Grade 5 and Grade 23 (extra-low interstitial) materials.

iii. Manufacturing parameters

The manufacturer should identify all manufacturing parameters that can be varied
during the manufacturing processes and establish explicit boundaries for each
parameter. This should include parameters associated with production, post-production
processing, fabrication, assembly, cleaning, sterilization (if required), packaging and
labelling of the device. For example, a manufacturer may produce two variants of a
spinal interbody cage using PEEK (polyetheretherketone), one model with and the other
without Ti coating on the superior and inferior surfaces.
iv. Clinical environment parameters
The manufacturer should identify all parameters relating to the clinical environment in
which the device is intended to be used, and establish explicit boundaries for each
parameter. For example, a manufacturer may produce two different patient-matched
maxillofacial bone plates in the same specified design envelope, one intended to be used
in the upper jaw and the other intended to be used in the lower jaw where it withstands
greater chewing forces.

v. Performance parameters
The manufacturer should identify all parameters relating to the performance of the
device when the device is used as intended, and establish explicit boundaries for each
parameter. For example, a manufacturer may produce three variants of a spinal
interbody cage (for patients with normal bone quality, osteopenia, and osteoporosis) to
reduce the risk of subsidence, each with different porosities and compressive stiffness
characteristics.

vi. Other parameters

If a parameter is not captured in any of the above categories but will characterize the
device for production purposes, the manufacturer should include the parameter in the
specified design envelope under this category and establish explicit boundaries for the
parameter.

Where the parameter is represented using categorical data, the manufacturer should establish
all of the categories that the parameter can accept. Where the parameter is represented using
numerical data (continuous or discrete), the manufacturer should establish the reference
interval, minimum increment, and unit of measurement for the variable.
Many patient-matched medical devices require the use of imaging data such as 3D printing from
computed tomography, magnetic resonance images and other scans (see example below). In
these cases, part of defining the parameters of the design envelope should involve particular
attention to imaging requirements. This includes but is not limited to a description of all
acceptable imaging modalities, minimum imaging quality, maximum timeframe between image
acquisition and first use of the device, and software used. These parameters are of clinical
importance to ensure patient-matched medical devices are of consistent quality and safety to
meet individualised patient care needs.
Clinical evaluation of a patient-matched medical device should consider all devices that can be
produced within the specified design envelope and how they relate to their intended patients. In
addition, these parameters should be evaluated and discussed by the clinical expert in a manner
relevant to the TGA clinical assessment. Given the heterogeneity in the design of patient-
matched medical devices, the clinical evidence provided for such devices must demonstrate
safety and clinical performance, and acceptability of benefit-risk profile, across the entire design

Clinical evidence guidelines: Medical devices Page 67 of 178

V3.1 June 2022
 Therapeutic Goods Administration

envelope (or that part of it covered by the intended purpose), in line with the usual
requirements for non-PMD devices.

Example – defining the design envelope:

Deformational plagiocephaly is a very common
craniofacial problem in babies, which results in a
misshapen, flat or asymmetrical head shape. Patient-
matched plagiocephaly helmets are a treatment option for
this condition in infants as young as six months (and can
be used up to 14 months of age).
Michael is an orthotist who manufacturers plagiocephaly helmets using an
advanced digital scanning and 3D printing technology. Michael uses validated
processes for the design and manufacture of plagiocephaly helmets for head
circumference between 38cm and 49 cm. Patient anatomical features are
captured via 3D digital scanning to an accuracy of 50 microns. A maximum
timeframe of 2 weeks is set between scanning and intended start date for use of
the helmet. During device use, further digital scans are to be obtained on at least a
monthly basis to ensure the helmet is still appropriate.
Using a dedicated software, the 3D scans are used for digital reconstruction of the
patient’s cranium. The digitally reconstructed 3D cranium along with the
treatment plan provided in the clinician’s prescription is used for generating the
final design of the helmet, which includes generic but scaled features to apply
gradual pressure on the skull in order to achieve the end-state cranial profile.
Michael has described all the design features including minimum and maximum
helmet dimensions, pressure applied and contour limits.
The final design file is 3D printed using validated production equipment,
processes and qualified raw materials. Post-production (cleaning, surface
smoothing, finishing, and packaging) is also completed using equipment and
processes within validated parameters.
Michael has also conducted a systematic review of the literature to establish the
current standards of care for deformational plagiocephaly. Apart from lifestyle
measures and higher risk surgical options, he determines that patient-matched
helmet therapy is the established state of the art treatment for persistent
deformity. Given this device is considered low risk, he decides to conduct a single-
arm clinical investigation that will enrol a small number of patients and include
helmets that span the entire 38-49cm range covered by the specified design
envelope boundaries. The study outcomes will include changes in head shape
(with the regular 3D scans) and side effects, such as adverse skin reactions, over a
6-month follow-up period (typical helmet duration of use).
Michael will provide the TGA with all the relevant information relating to the
structural, material, manufacturing, clinical, performance and 3D printing
parameters in order to clearly articulate the specified design envelope for
plagiocephaly helmets. He will also provide evidence from the clinical
investigation to establish safety and performance outcomes for this low risk
device.

Clinical evidence guidelines: Medical devices Page 68 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Generalisability of devices within the design envelope (external validity)

A suitably qualified clinical expert/s who has endorsed the CER should determine the extent to
which clinical investigation data can be extrapolated to all potential device specifications within
the design envelope. Justification of generalisability may involve the use of clinical data, bench
testing and/or computer modelling. Each difference in the clinical, technical and biological
characteristics between the trialled devices and the extent of the specified design envelope
should be discussed, with appropriate assessment of how the differences may impact the safety
and performance of the subject device. The clinical expert should comment on worst-case and
common-use scenarios in regard to clinical risk(s) when determining the extent of extrapolation,
noting that they are not necessarily mutually exclusive (it is possible for worst-case scenarios to
also be common-use scenarios).

· Worst-case scenario(s): identified devices with the highest risk features within the
design envelope (more than one may exist for patient-matched medical devices within a
specified design envelope).

Lower risk devices may include a discussion from the clinical expert as to why a certain
subtype is the worst-case scenario, whereas for higher risk devices, manufacturers
should generally provide evidence for worst-case scenarios (with clinical data, bench
testing and/or computer modelling).

· Common-use scenario(s): identified devices within the design envelope with the most
frequently used design parameters. When defining common-use scenario(s), it may be
useful to consider anthropometric differences (e.g. sex, age, ethnicity), as common-use
scenario(s) may vary between sub-populations.
Therefore, when compiling and reporting clinical data for a patient-matched medical device,
attention should also be given to how the data support the performance and safety of worst-case
and common-use scenarios within the design envelope. This may be achieved by reporting data
specific to these scenarios, or by providing a robust justification as to why data from other
scenarios can be extrapolated to these scenarios (see example below).

Example – generalisability of clinical evidence

Jane is the clinical expert evaluating a clinical evidence report for a 3D-printed
patient-matched mandibular advancement splint, used to treat mild sleep apnoea.
In the CER, Jane discusses and critically analyses a clinical study of the device.
Even though every patient in the study is treated with a unique patient-matched
device, Jane provides a scientific rationale for why the results of the study are
generalisable to every patient-matched device potentially produced within the
specified design envelope, based on bench testing and/or computer modelling.
This includes a consideration of “worst-case” scenarios captured in the clinical
study (the most vulnerable to adverse events and reduced performance) and
“common-use” scenarios (the most common device design parameters based on
common anatomy and pathology).
The TGA assessor accepts that generalisability has been established.
Consequently, this clinical investigation provides support for the safety and
performance of the subject device even though every patient receives a slightly
different product.

Clinical evidence guidelines: Medical devices Page 69 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Substantial equivalence
A manufacturer may use clinical data for a comparable medical device (either PMD or non-PMD)
to support safety and clinical performance claims for the subject device. The extent to which
such evidence may be acceptable will depend on how similar the devices are for relevant
aspects, including the intended use and other clinical, technical, and biological characteristics,
and manufacturing processes. Consideration should be given to how the differences may affect
the safety or clinical performance of the subject device. Where claims of substantial equivalence
are made with a device that is either a non-PMD or has a different design envelope, there should
be a reasoned argument provided as to why the clinical evidence for the claimed equivalent
device can be applied to the entire design envelope (or if this is not claimed, what subset of the
design envelope the data is relevant to). This should include discussion of worst-case and
common-use scenarios.
Whilst claims of substantial equivalence may be useful (in order to provide indirect clinical
evidence for a subject device), in most circumstances this should only represent one component
of a strategy to generate sufficient clinical evidence for PMDs. There remains a need for direct
clinical evidence, commensurate with the risk of the device, to provide additional assurance that
modifications do not affect device safety and/or performance. This is of particular importance
when substantial equivalence claims involve a non-PMD device.

Example – substantial equivalence

Jane is the clinical expert evaluating a 3D-printed patient-matched mandibular
advancement splint for treatment of mild sleep apnoea. She claims substantial
equivalence to a similar patient-matched mandibular advancement splint
produced by a different manufacturer.
In the CER, Jane describes all parameters of the design envelope for the subject
device, including structural (e.g. minimum and maximum dimensions, attaching
mechanism for the maxillary and mandibular components, dental alignment
specifications), material, manufacturing, clinical environment, performance and
3D printing parameters (e.g. CT imaging requirements).
From a literature review she determines that mandibular advancement splints are
the current standard of care for mild sleep apnoea and provides support for
generalisability of devices made within the design envelope through the
identification of worst-case and common-use scenarios (from the clinical
literature).
She then compares all aspects of the specified design envelope with that of the
comparable device and provides a critical analysis describing how any differences
could impact the safety and/or performance of the subject device, citing published
and original preclinical data.
The TGA assessor accepts that the subject device is substantially equivalent to the
proposed comparable device.
The submitted CER contains a mixture of clinical evidence that relates to both the
comparable device (indirect evidence) and the subject device (direct evidence),
which when combined, is sufficient to support the safety and performance of the
3D-printed patient-matched mandibular advancement splint.

Clinical evidence guidelines: Medical devices Page 70 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Clinical Investigation
A key challenge in the design of clinical investigation for PMDs, is that the PMD device
intervention will be heterogeneous with regards to several of its parameters (compared to non-
PMD studies where features such as materials, structure and dimensions are either constant or
subject to minor variations). The comparability of subjects within the group receiving the
subject device should be optimised through study design, and uncertainties addressed through
critical analysis (as discussed in Part 1 of these guidelines).
Despite the uncertainties and limitations in clinical investigations for patient-matched medical
devices, it is preferable to undertake a feasible clinical investigation and discuss the limitations
(which can then be subject to a risk management framework), rather than seeking approval of
devices with a lack of clinical data (which would likely mean non-compliance with the essential
principles).
For high-risk devices and those based on technologies where there is little to no prior clinical
experience, direct clinical evidence from the use of the patient-matched medical device in
humans will generally be required to demonstrate conformity with Essential Principles in the
pre-market application. When designing clinical investigation for such devices, consideration
should be given to the:
• prevalence and incidence of clinical condition in the general population;
• availability of evidence relating to comparable devices for the same intended purpose;
• standard of care for the clinical condition (based on a literature review);
• availability of evidence relating to comparator devices for the same intended purpose;
• meaningful, measurable, patient-relevant clinical outcome(s);
• follow-up duration and study endpoints to allow for objective assessment of the claimed
benefits;
• procedures for recording distinct design and manufacturing features of each device used in
the investigation; and
• subgroup analyses of worst-case and common-use scenarios.
The clinical data from a clinical investigation should be collected in a way which permits
subgroup analyses of the various parameters of the specified design envelope.
For high-risk devices, if a comparator medical device (PMD or non-PMD) exists for the same
intended use, the clinical investigation should be designed on sound scientific principles and
methodology, including an appropriate statistical plan, with the comparator device as a positive
control. If a single-arm study is the appropriate design for a particular clinical condition or
device use, data should be collected in a way that allows for objective comparison with the
standard of care. If no treatment exists for the clinical condition, clinical investigation data
should be collected in a way that allows for comparison with the natural clinical course of the
condition and objective assessment of benefit-risk profile for the device. Clinical investigations
should be conducted following relevant standards (ISO 14155) and/or applicable regulatory
requirements.

Literature review
Similar to clinical investigations, a literature review will generally only provide appropriate
clinical evidence to the extent that the studied devices investigate use over the breadth of the
design envelope. Again, the external validity of the trialled devices should be compared to all
potential device specifications, with use of both worst-case and common-use scenario analysis.

Clinical evidence guidelines: Medical devices Page 71 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Post-market data
The TPLC approach is particularly important for PMDs and should be conducted to ensure
ongoing acceptability of the residual risks and to identify any new or emerging risks.
Manufacturers of patient-matched medical devices are generally expected to submit detailed
post-market surveillance (PMS) plans, proportionate to the risk class and the type of device, as
part of a pre-market application.
PMS plans should describe PMCF activities to proactively and systematically collect, categorise,
and analyse data relevant to the performance and safety of the device periodically throughout its
lifecycle. For each PMCF activity, details on the aims, methods of data collection and analyses
(including rationales for their appropriateness) should be provided. Data should be collected in
a way that allows for subgroup analyses of parameters in the specified design envelope (in
particular, worst-case and common-use scenarios) and patient characteristics, to facilitate an
objective assessment of claims made by the manufacturer regarding performance and safety.

Other clinical experience data

The systematic collection of other clinical experience data relevant to the PMD (for example,
product and/or disease registries) can add to the evidence base. Development of high-quality
registries for specific products is encouraged. Data collection should be designed to enhance
quality and comparability with other sources of clinical experience data (such as data generated
from clinical investigations) and include an explanation of the processes in place to safeguard
data quality and integrity.
Ongoing collection and analysis of both post-market and other clinical experience data for a PMD
may, over time, support widening of the design envelope.

Clinical evidence considerations for adaptable medical devices

Adaptable medical devices are mass-produced and intended to be assembled or adapted after
supply, according to the manufacturer’s instructions, to address anatomical and/or physiological
features of a particular individual, or a pathological condition (for example devices such as
craniofacial implants, dental implants and orthotics that are mass-produced, but adapted by a
health professional at the point-of-care). The manufacturer of an adaptable medical device is
generally required to supply the device with instructions for use and patient information leaflets
(where applicable, e.g. for implantable or active implantable devices) that will ensure the final
device complies with all relevant EPs after it has been assembled or adapted.
There should be clear articulation and consideration of the risks relating to usability and
appropriateness of the instructions for use and patient information leaflets (where applicable)
within the risk management report. A manufacturer of an adaptable medical device should
always consider conducting a usability study (consistent with IEC 62366-1) to validate the
instructions provided in the IFU for adaptation/assembly of the device.
Similar to patient-matched medical devices, due to heterogeneity in the final form of adaptable
medical devices, there are challenges associated with meeting clinical evidence requirements
when compared to other non-PMD devices, that should be considered when collecting and
discussing clinical data. The principles regarding worst-case and common-use scenarios will
generally be applicable to adaptable medical devices and should be considered and discussed.
Clinical trials should be designed to consider these variables and allow for subgroup analysis. A
well-reasoned and comprehensive risk analysis, including risks relating to parameters for device
adaptation, potential device failure modes, and device usability should form the basis for
generating the clinical evidence.
Clinical evidence may include data from a comparable device, but direct clinical evidence,
particularly with regard to usability, should be provided for higher risk and more novel devices
in most circumstances. The design of clinical investigations, including study objectives, use of a

Clinical evidence guidelines: Medical devices Page 72 of 178

V3.1 June 2022
 Therapeutic Goods Administration

comparator arm, and statistical considerations, should be such that it provides sufficient clinical
data to address residual risks and aspects of clinical performance not addressed through the
current state of the art for the condition being treated.
In the PMS plan, the manufacturer should include adequate details on PMCF activities to collect,
categorise, and analyse the data on the performance and safety of such devices throughout the
device lifecycle and update the clinical evidence periodically. Data from PMCF activities should
be collected in a way that allows for subgroup analyses of parameters relating to device
adaptation, for objective assessment of claims made by the manufacturer on the safety and
clinical performance of the devices.

Example – Adaptable medical device

Louise is a dentist and clinical expert for a manufacturer of temporary stainless
steel crowns that are used as an interim measure whilst patients await fitting of
patient-matched dental crowns. One new device is a mass-produced stainless steel
molar crown that comes in a range of sizes. Dentists are required to trim the edges
with crown scissors and “crimp” the device with crimping pliers for the individual
patient. The device is then cemented to the tooth.
The IFU contains detailed instructions for acceptable point-of-care manipulation
of the subject device with respect to sizing. The greatest extent of trimming and
crimping supported by the IFU are considered worst-case scenarios (justified
through computer modelling).
Louise has conducted a state of the art literature review and from the literature is
satisfied that temporary stainless steel and acrylic resin dental crowns are the
accepted standard of care when patients are awaiting a more permanent, patient-
matched crown.
The manufacturer has conducted detailed useability testing for the subject device
with a sample of dentists. Subsequently a clinical investigation was conducted to
investigate performance and safety outcomes when compared with another state
of the art temporary stainless steel dental crown (with specific regard to worst-
case scenarios). Outcomes included validated tools for dental function
(performance) as well as adverse events (e.g. infection, migration, early
extraction, pain).
Louise has provided a detailed risk management report to appropriately mitigate
residual risks, especially those associated with the intended device modifications
(trimming and crimping). The residual risks are clearly articulated in the
instructions for use.
A detailed ongoing PMCF is planned which will continue to monitor residual risks
throughout the device lifecycle.
The clinical evidence provided is considered sufficient to support the safety and
performance of the subject device, and the temporary stainless steel molar crown
is included on the ARTG.

Clinical evidence guidelines: Medical devices Page 73 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Clinical evidence considerations for custom-made medical devices

Custom-made medical devices are for the sole use of a particular patient, or health professional
(in the course of their practice), and are manufactured in accordance with particular design
characteristics specified by a health professional in a written request to the manufacturer. The
design characteristics for custom-made medical devices must be intended to address the
anatomical and/or physiological features, or the pathological condition, of the intended
recipient.
The requesting health professional must also have determined that the device is necessary to
address the features or condition of the intended recipient because there is no kind of medical
device included in the Australian Register of Therapeutic Goods (ARTG) to address those
matters either wholly or to an appropriate level.
These devices are typically required when device specifications are outside the design envelope
of a patient-matched medical device. An example is a knee-replacement system required for a
rare anatomical variation for which there are no suitable non-PMDs, and which is also beyond
the boundaries of specified design envelopes of existing patient-matched medical devices.
Custom-made medical devices are exempt from the requirement to be included in the ARTG,
however they are not exempt from the requirement to comply with all applicable Essential
Principles including EP 14. They are also not exempt from the need to follow clinical evaluation
procedures in Part 8 of Schedule 3 of the MD Regulations.
While clinical evidence for a custom-made medical device may be limited due to the unique
nature of the device, the manufacturer should consider the following factors:
• The reasons why a custom-made medical device was requested, including the determination
by the requesting health professional that there is no kind of medical device available to
address the condition of the intended recipient;
• The design inputs and outputs; and
• Pre-clinical and clinical data to support the claims on safety and clinical performance. Whilst
the presence of any preclinical and clinical data should be considered prior to the issue of a
custom-made device, this data will be of particular importance to maintain in the post-market
context.
There should be clear articulation and consideration of risks relating to usability and
appropriateness of the instructions for use and patient information leaflet (where applicable)
within the risk management report. This is particularly important if the custom-made medical
device is based on a non-PMD predicate and/or where a non-PMD version of the device is
available.

Example – Custom-made medical devices

Michael, the orthotist that makes plagiocephaly helmets, receives a written
request from a paediatrician for a helmet for an 8-month old infant with a head
circumference of 50cm. This is outside the reference interval boundary for the
specified design envelope, however, there are no devices included on the ARTG
that could meet these specifications. A unique device is required to meet the
anatomical characteristics of the individual patient and Michael considers the
potential effects the larger size may have on the safety and performance of the
subject device. Given the low assessed risk and the current clinical need, Michael
agrees to make the custom-made medical device. Michael will keep detailed

Clinical evidence guidelines: Medical devices Page 74 of 178

V3.1 June 2022
 Therapeutic Goods Administration

records of the device specifications, clinical outcomes and adverse events

associated with the device use.

Compiling the CER

The recommended approach to clinical evaluation and to compiling the CER (outlined earlier in
Part 1 of the guidelines) and principles underpinning the TGA clinical assessment of devices
apply to PMDs. The principles around hierarchy of evidence and quality of evidence based on
study size and methodology set out in Part 1 are relevant to PMDs. A clinical evaluation should
similarly be formed with reference to the manufacturer’s risk management strategy.
There should be clear articulation and consideration of the risks relating to usability and
appropriateness of the instructions for use and patient information leaflet (where applicable)
within the risk management report.

Considerations for patient-matched medical devices

Clinical evaluation of patient-matched medical devices will centre around sufficiency of the
clinical evidence to support safety and performance over the entire breadth of the design
envelope. Therefore, the design envelope should be characterised in depth with particular
reference to worst-case and common-use scenarios. Due to the heterogenous nature of PMDs,
the risk management report, instructions for use and patient information leaflets (where
applicable) will also be critical.

Considerations for adaptable medical devices

Similar to patient-matched medical devices, clinical evidence should support safety and
performance of adaptable medical devices and include an evaluation of point-of-care assembly
or modifications. Risk management report, instructions for use and patient information leaflets
(where applicable) will carry additional emphasis in assurance for these devices.

Considerations for custom-made medical devices

Whilst custom-made devices are exempt from ARTG inclusion and therefore from pre-market
submission of clinical evidence, they are still required to comply with the Essential Principles.
Manufacturers must maintain records which include custom-made device specifications, clinical
evidence and risk management reports as part of their quality management system.
Additionally, custom-made medical devices have the same post-market surveillance
requirements as other PMDs.

Defining clinical outcomes

PMDs are inherently a heterogenous group of devices that cover the spectrum of risk. The
nature, type and range of clinical evidence that is expected will therefore reflect this spectrum,
with greater scrutiny given by the TGA to higher classification devices and devices with novel
features. Nevertheless, manufacturers should note the following points:
• Generalisability of PMDs (external validity) within clearly defined parameters will generally
need to be established in order to correlate available clinical data to all PMD subtypes.
• Outcome measures used to substantiate safety and performance of a PMD should be the
same patient-centred outcomes that define clinical treatment success for a particular
condition (i.e. surrogate markers will not generally be sufficient). It is expected that these

Clinical evidence guidelines: Medical devices Page 75 of 178

V3.1 June 2022
 Therapeutic Goods Administration

outcomes be compared to existing standards of care (or the natural clinical course of the
condition where no treatment exists).
• Risk management reports with appropriately detailed Instructions for Use and Patient
Information Leaflets (where applicable) and ongoing post-market surveillance are critical to
support clinical utility of such devices across their lifecycle.
In situations where uncertainty remains in relation to the acceptability of a strategy for
generating clinical evidence, the manufacturer may consider engaging with the TGA for a pre-
submission meeting to obtain general feedback.

Example – Premarket application for a subperiosteal mandibular implant

John is an oral surgeon and clinical expert evaluating a new 3D-printed patient-
matched subperiosteal titanium mandibular implant that is intended to be used in
the treatment of patients with an atrophic mandible (for whom endosseus dental
implants are not suitable).
In the CER, John provides a detailed description of the specified design envelope
that includes structural, material, manufacturing, performance, clinical
environment, and 3D printing parameters. He has discussed how the above
parameters relate to individual patient needs. Generalisability of clinical evidence
for the subject device across the entire design envelope has been justified through
the identification of worst-case and common-use scenarios (with the use of
computer modelling and state of the art literature review, respectively).
John claims that the subject device is substantially equivalent to another patient-
matched 3D-printed subperiosteal titanium mandibular implant. He compares the
clinical, technical, and biological characteristics of the devices across the design
envelopes. Clinical reasoning is provided to explain why minor differences are not
expected to alter the performance and safety of the subject device (supported by
computer modelling).
John discusses the clinical evidence that he has for the comparable device. The
manufacturer of the comparable device conducted pre-market clinical
investigations for worst-case and common-use scenarios and demonstrated that
the comparable device has superior effectiveness for dental function when
compared to bone reconstructive treatments used in the dental management of
patients with an atrophic mandible (standard of care), and minimal adverse
events. Post-market experience with the comparable device demonstrates very
low complaint rates. John has provided indirect clinical evidence in support of the
safety and performance of subject device within worst-case and common-use
scenarios.
John has also conducted a short-term, single-arm clinical study using the subject
device for a small sample of patients. Similar positive outcomes (dental function)
have been described without any significant adverse events (e.g. pain, infection,
explantation).
John has provided a detailed risk management report with appropriate risk
mitigation and residual risks are clearly articulated in the instructions for use. He
has also included a detailed plan for PMCF to provide ongoing evaluation of the
subject device.
Based on the classified risk of the subject device, the TGA assessor accepts that the
clinical evidence provided is sufficient to substantiate the safety and performance

Clinical evidence guidelines: Medical devices Page 76 of 178

V3.1 June 2022
 Therapeutic Goods Administration

of this patient-matched medical device across the specified design envelope. The
subject device is included on the ARTG.

Clinical evidence guidelines: Medical devices Page 77 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Part 3 - Requirements for specific device types

Part 3 provides guidance to assist industry and clinical researchers to understand the TGA’s
expectations in relation to clinical evidence for particular types of medical devices.
The guidance articulates the nature, type and/or range of evidence that the TGA considers, at a
minimum, will generally facilitate an adequate assessment of the benefit-risk profile of the
device to be determined, taking into account safety, performance and patient health outcomes.
This assessment is part of the process by which the TGA considers compliance of kinds of
medical devices against the EPs set out in Schedule 1 of the MD Regulations.

Specific device types currently covered are:

• IVDs
• Total and partial joint prostheses
• Cardiovascular devices to promote patency or functional flow
• Implantable pulse generator systems
• Heart valve replacements using a prosthetic valve
• Supportive Devices - Meshes, Patches and Tissue Adhesives

In vitro diagnostic (IVD) medical devices

In vitro diagnostic (IVD) medical devices (or IVDs) have their own classification system, and
must comply with all the EPs, including EP 15 which relates specifically to IVDs.
The Clinical evidence guidelines supplement: In vitro diagnostic (IVD) medical devices provides
specific guidance in relation to IVDs.

Total and partial joint prostheses

Joint prostheses include devices used in hip, knee, shoulder, ankle, elbow and wrist joint
replacements. Joint replacement (also called arthroplasty) is a commonly performed
orthopaedic operation with the objective of relieving pain and improving mobility.
This section outlines what is generally expected regarding appropriate clinical evidence to
demonstrate that a joint prosthesis is safe and performs as intended through compliance with
the applicable Essential Principles (EPs) of safety and performance in Schedule 1 of the MD
Regulations. The section is intended to supplement the general recommendations outlined in
other parts of these guidelines.
Note this section provides guidance for all joint prostheses including, but not limited to a ‘joint
replacement medical device’ as defined in the MD Regulations.

Summary recommendations
Joint prostheses are complex medical devices that can be used in combination with other devices
or components. Manufacturers are advised to list the common combinations and provide clinical
data to support the safety and performance of the device for these nominated configurations.
Joint prostheses pose a significant regulatory challenge because these devices need to have a
long in vivo life without exposing the patient to unduly high risks of adverse events or
undesirable effects. In summary, the following is recommended:
Clinical evidence guidelines: Medical devices Page 78 of 178
V3.1 June 2022
 Therapeutic Goods Administration

• Clinical data is in the form of DIRECT evidence (pertaining to the subject device/system
only) or INDIRECT evidence (data pertaining to a substantially equivalent device/system).
• For clinical evidence based on an evaluation of comparable device data, manufacturers are
advised to submit all relevant documents with a supporting justification by a clinical expert
to:
– establish substantial equivalence between the device and the nominated comparable
device, and
– confirm that any identified differences will not adversely affect safety and performance
of the device.
• Manufacturers should provide details of the clinical context within which the clinical data
were obtained. The clinical context of the data should be congruent with the indications for
use.
• Provision of clinical data:
– manufacturers who intend to conduct clinical trials should design trials to the highest
practical NHMRC level of evidence and trials should be appropriate to inform on the
safety and performance of the device for its intended purpose
– it is recommended that the minimum period for patient follow-up for clinical trials is
two years
– the main clinical outcomes that determine safety and performance are ‘time to first
revision’ and patient scores such as the Harris Hip Score:
▪ for revision data, the manufacturers are advised to benchmark the device against
devices of the same class and against a similar patient population as reported by
an international joint registry (e.g. OA vs patients prone to dislocation vs trauma vs
osteoporosis vs bone resection for tumour etc.)
▪ for patient performance data, manufacturers are advised to define the anticipated
improvement in patient scores post-surgery (ideally, these should be
internationally recognised assessment tool(s) used to measure clinical success)
– to assess the risk of delayed need for revision surgery (that is in vivo times greater than
two years), the manufacturers should consider using surrogate markers that are
predictive of prosthesis failure - alternatively, manufacturers may use post-market data
if the device is approved and marketed in Australia or elsewhere.
• For guidance on the conduct of comprehensive literature reviews and on the compilation
and presentation of clinical evidence, manufacturers are directed to the relevant sections in
this document.

Defining joint prostheses

For the purposes of this guidance document a joint prosthesis is an implantable medical device,
irrespective of its configuration, that is intended by the manufacturer to replace in full or in part
a section of the joint.
Joint prostheses can consist of either monoblock or modular designs. There are practical
advantages to modular designs as they allow tailoring of the prosthesis to the patient’s anatomy.
However, modular devices with multiple components are more complex and may have a
different benefit-risk profile when compared with monoblock designs. Each combination is
unique and may have its own associated benefit-risk profile that needs to be addressed by the
manufacturer.

Clinical evidence guidelines: Medical devices Page 79 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Limb-preserving devices may also include joint implants. These devices are designed for
functional limb reconstructions for patients with significant bone loss usually around the knee
and hip. Such bone loss can occur following treatment of malignant bone tumours, aggressive
benign bone tumours, infection, multiple revised and failed joint replacements or massive
trauma.

Clinical evidence
In Part 1 of these guidelines, the TGA has provided information regarding key definitions and
concepts, clinical evidence requirements, sources of clinical data and clinical evaluation in
relation to medical devices, which are also applicable to joint prostheses. It is acknowledged that
the volume of clinical evidence available may vary between the types of devices under review.
For instance, the evidence available for a revision joint replacement system may be limited
compared to a primary joint replacement system, and this is considered during the assessment
process.
Direct clinical evidence on the actual device is preferred. Otherwise, indirect clinical evidence on
a comparable device may be used after substantial equivalence has been established through a
comparison of the clinical, technical and biological characteristics as described in Comparable
devices including substantially equivalent devices.

Substantial equivalence
A manufacturer may use clinical data from a comparable device to support the performance and
safety of the subject device once substantial equivalence has been established. In addition to the
guidance provided in Comparable devices including substantially equivalent devices,
manufacturers should consider the following when presenting substantial equivalence claims
for joint prostheses:
• The sub-classification of the subject device/system. For example:
– A cemented device will generally not be considered substantially equivalent to an
uncemented device given the widely accepted biological and technical differences in
fixation mode.
– A modular device will generally not be considered substantially equivalent to a
monoblock device given that modularity introduces additional potential failure modes.
– A revision device will generally not be considered substantially equivalent to a primary
device given differences in clinical characteristics and outcomes, as well as key
technical differences.
– For knee systems, a posterior stabilised knee system will generally not be considered
substantially equivalent to a cruciate retaining knee system given the widely accepted
biomechanical differences resulting from the design principles of each system, which
presents in the form of key technical differences.
• Biological/technical characteristics which should be considered include (but are not limited
to) the material of the prostheses, coating, coating thickness, coating porosity, rigidity,
fatigability, torsional strength, tensile strength, dimensions, geometry, weight, intended
fixation methods, components to which the joint prosthesis may be paired and combinations
which may be deployed. It is emphasised that large technical differences, or multiple small
technical differences, can negatively impact a substantial equivalence claim.
• Side-by-side technical diagrams of the subject and comparable from multiple perspectives
are encouraged as they facilitate the TGA’s assessment of substantial equivalence.

Clinical evidence guidelines: Medical devices Page 80 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• For applications that include multiple device components, substantial equivalence claims
presented separately for each device component are preferred (for example, for a knee
system application, three separate substantial equivalence claims comparing the femoral
components, the tibial baseplates and the tibial inserts).
It is also important to clarify if any changes have been made to the device since the clinical data
were gathered and, if so, to document the changes and to clarify the exact version of the device.
Where the device and the predicate share a common design origin, particularly when the device
is part of a modular system, the lineage of devices with the same intended purpose should be
provided as well.

Clinical investigation(s)
Clinical investigations will generally be required for orthopaedic devices with novel features.
The design of the clinical investigation(s) should be appropriate to generate valid measures of
clinical performance and safety. The preferred design is a randomised controlled clinical trial
and conditions should ideally represent those in clinical practice in Australia. The eligible
patient groups should be clearly defined with exclusion/inclusion criteria.
Manufacturers are advised to justify the patient numbers recruited according to sound scientific
reasoning through statistical power calculation. Some examples of Randomised Controlled Trials
(RCTs) involving joint prostheses include the UK Knee Arthroplasty Trial (KAT) 5 and the A
JOINTs Canada Project. 6
The duration of the clinical investigation should be appropriate to the device and the patient
population and medical conditions for which it is intended. Duration should always be justified,
taking into account the time-frame of expected complications. Clinical trials must be
independently audited at key stages throughout the trial to document that the integrity of the
trial was maintained. Analysis of clinical events should be blinded and independently
adjudicated wherever possible.
Additional resources regarding the design and conduct of clinical investigation(s) are available
on the clinical trial pages of the TGA and FDA websites. These guides inform on appropriate
numbers of patients to be recruited as well as the necessary patient follow-up for statistically
significant and clinically meaningful results. Guidance on the recommended reporting
requirements for clinical investigation reports is provided in Reporting standards for clinical
investigations.

Literature review
A literature review involves the systematic identification, synthesis and analysis of all available
published and unpublished literature, favourable and unfavourable on the device, or, if relying
on indirect evidence, the comparable device to which substantial equivalence has been
established as described in Comparable devices including substantially equivalent devices.
Data on the materials used to construct the prosthesis, its dimensions and geometry, the number
and type of paired components for modular devices and the intended purpose will define the
construction of search strategies as well as study selection. This ensures that the searches are
comprehensive and the included studies are relevant to the device and/or comparable device.
The selection of a comparable device should be made prior to performing the literature

5 Murray DW, MacLennan GS, Breeman S et al. A randomised controlled trial of the clinical effectiveness
and cost-effectiveness of different knee prostheses: the Knee Arthroplasty Trial (KAT). Health Technol
Assess. 2014;18:1-235, vii-viii
6 Litchfield RB, McKee MD, Balyk R et al. Cemented versus uncemented fixation of humeral components in

total shoulder arthroplasty for osteoarthritis of the shoulder: a prospective, randomized, double-blind
clinical trial-A JOINTs Canada Project. J Shoulder Elbow Surg. 2011;20:529-36.
Clinical evidence guidelines: Medical devices Page 81 of 178
V3.1 June 2022
 Therapeutic Goods Administration

selection, extraction of the clinical data and analysis of the pooled results. A full description of
the device used in any given study or adequate information to identify the device (e.g.
manufacturer name and model number) should be extractable from the study report. If this is
not possible, the study should be excluded from the review. Manufacturers are referred to
Literature review for further information.

Post-market data
Post-market data should be provided for the actual device or for the comparable device to
which substantial equivalence has been established, refer to Post-market data.
It is particularly important to include the following:
• information about the regulatory status of the device (or comparable device if relying on
this), including name under which the device is marketed in key jurisdictions such as
Canada, USA and Japan, certificate number, date of issue, the exact wording of the intended
purpose/approved indication and other relevant details such as MRI designation in other
jurisdictions
• any regulatory action including withdrawals, recalls, including recalls for product correction
(and the reason for these, such as IFU changes) cancellations or any other corrective actions
occurring in the market in any jurisdiction as reported or required by regulatory bodies
• distribution numbers 7 of the device(s) including distribution by country and/or
geographical region for every year since launch. It is acknowledged that this may not always
be appropriate for high volume devices, those with several components and those which
have been on the market for many years.
• number of years of use
• for every year since launch, the number of complaints, vigilance and monitoring reports and
adverse events categorised by type and clinical outcome (e.g. death, serious harm, revision
due to loosening, fracture, implant breakage, etc.)
• explanted joint prostheses returned to manufacturers should be accounted for with an
explanation of failures and corrective measures.
Publicly available post-market data such as adverse event reporting on the FDA MAUDE
database and the TGA IRIS should be provided including for devices from other manufacturers
when demonstrating substantial equivalence with comparable devices.
For reports of adverse events, revisions and complaints to be a useful adjunct to other forms of
clinical evidence, the manufacturer should make an active, concerted effort to collect the reports
and to encourage users to report incidents. Experience shows that merely relying on
spontaneous reports leads to underestimation of the incidence of problems and adverse events.
The post-market data should be critically evaluated by a competent clinical expert to enable an
understanding of the safety and performance profile of the device in a ‘real-world’ setting.
Other clinical experience data (including registry data)
National joint registries have been established in Canada, Denmark, England and Wales, Finland,
New Zealand, Norway, Romania, Scotland, Slovakia and Sweden as well as Australia. The quality
of data extracted from joint registries may vary. This may be influenced, for example, by the
structure of the registries.

8 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6554112/
Clinical evidence guidelines: Medical devices Page 82 of 178
V3.1 June 2022
 Therapeutic Goods Administration

The AOANJRR is highly regarded amongst the international orthopaedic community. So is the
UKNJR. These are bodies that collect a large variety of data including but not limited to
‘cumulative revision rates (CRR)’ or ‘cumulative percent revision (CPR)’ and ‘patient reported
outcomes measures (PROMs)’. Good quality registry data (inclusive of these types) is therefore
considered a robust source of post-market data.
When utilising registry data, manufacturers are also encouraged to present the data alongside
an in-built comparator (for example, the average revision rate across the device type).

Compiling the CER

The approach to clinical evaluation and compiling the CER outlined in Part 1 – General
Requirements also applies to joint prostheses. In compiling the clinical evidence the
manufacturers must ensure that a competent clinical expert critically evaluates all the clinical
data from clinical investigation(s), literature review and/or post-market data and endorses the
CER (evidenced by signature and date), when seeking to demonstrate that the clinical evidence
is sufficient to comply with the applicable EPs and that the device is safe and performs as
intended.

Supportive data and information

The following information on the device must also be provided:
• risk assessment and management document
• IFU, labelling, product manual, PIC/PILs, and all other documents supplied with the device.
These must highlight the residual risks and ensure that they are appropriately
communicated to the user.
Additional information should be provided as applicable. This may include (but is not limited
to):
• additional specifications of the device(s)
• surgical technique guides
• the materials from which the device is made including chemical composition
• other devices that may be used in conjunction with the device
• any aspects of non-clinical testing results that inform the design of the clinical trial should be
included in the supporting documents
• biocompatibility testing, bench testing and animal studies where applicable
• specific testing of any adjuvant medicinal components may be necessary especially if these
are new chemical entities in the Australian context. This should cover interactions between
the device and the medicine, pharmacodynamics and time-release profiles.

Measuring clinical success

Safety
For safety, the primary outcome measure is revision. It is acknowledged that some reoperations
which are considered a revision in one registry are not considered a revision in another registry,
therefore comparisons of implant performance using data from different registries have to be
undertaken with caution. Typically revision is reported as the Cumulative Percent Revision
(CPR) based on the time to the first revision. The Australian Orthopaedic Association National

Clinical evidence guidelines: Medical devices Page 83 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Joint Replacement Registry (AOANJRR) provides annual reports on the performance of joint
prostheses for hip, knee and shoulder and provides the CPR for joint prostheses.

The AOANJRR is a comprehensive database providing manufacturers with

detailed revision data for devices that are available and used in Australia.
Manufacturers should demonstrate that CPRs for a device or comparable
device, if used to substantiate the safety and performance of the device, are
equal to or better than published CPRs for joint prostheses of the same class as
defined by the AOANJRR or another international joint registry (such as the
National Joint Registry [England and Wales]), within the first two years of
implantation at a minimum.

If clinical investigations are conducted, it is recommended that the minimum patient follow-up is
two years: this is based on the internationally accepted consensus of orthopaedic surgeons and
editors of orthopaedic journals. The AOANJRR analysis methods can identify devices that are
prone to early failure as indicated by a higher than expected CPR within the first two years of
implantation. This supports the concept of the two year minimum patient follow-up in clinical
trials. However, manufacturers should be aware that this is the minimum and will not capture
information relating to the late failure of a prosthesis. In this situation, manufacturers can assist
the clinical assessors by providing adjunct data from surrogate markers. The choice of markers
and a justification that these are predictive of future prosthesis failure should be clinically
justified.
To assess performance based on rates of revision the manufacturer should:
• identify the published early CPR as documented in the AOANJRR (or other national
registries) for devices that are in the same class as the device
• determine whether the device or the comparable device is performing as expected for that
class of device as compared to the reference CPR reported by an international joint registry
• document the reason for revision; reasons include, but are not limited to:
– aseptic and septic loosening for hip, knee and shoulder prostheses
– dislocation and fracture for hip and shoulder prostheses
– postoperative alignment for hip and knee arthroplasty
– wear/erosion for shoulder arthroplasty
• where appropriate provide adjunct data for surrogate markers that may assist in predicting
late failure of the device. Examples of surrogate markers:
– radiological findings e.g. radiolucent lines for hip and knee procedures
– radiostereometric analysis (RSA) to determine early (within two years) migration of
joint components. RSA may be a viable surrogate to identify prostheses that would
require early revision due to aspect loosening
– in the case of metal-on-metal devices, appropriate monitoring of metal ion
concentrations in body fluids are a measure of metal exposure and may have merit as a
surrogate marker of excessive wear.

Clinical evidence guidelines: Medical devices Page 84 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Manufacturers, in selecting and reporting surrogate markers of safety, should

provide a clinical justification for the selection and where possible should use
validated measurement tools.

Performance
Performance related parameters reported in the peer reviewed literature for hip, knee and
shoulder prostheses are provided in Table 7.
Clinical success is evaluated by patient-oriented assessment tools that determine functional
outcomes. Functional scores provide an aggregate of patient reported domains (e.g. pain, need
for support device) with an objective measure of joint motion (e.g. degree of flexion or abduction
and alignment) and represent a clinically meaningful grading of joint performance. However, for
joint arthroplasty, the short-term performance of a device may be dominated by procedure
variables therefore sufficient time should lapse to isolate device specific improvements.
The recommended two year minimum patient follow-up is congruent with the reported time to
a stable output for two validated patient scores (these being the Harris Hip Score (HHS) and the
Short Form-36 Health Survey (SF 36)). These scores have the greatest change in the first six
months post-surgery for patients that have received a unilateral primary total hip replacement
and peak or plateau at 18 months and remain high for 5 years.

When documenting patient performance scores, it is recommended that

manufacturers provide data with a minimum of two years follow-up post-
surgery to reduce the risk of confounding due to procedure variables.

Ideally, manufacturers should define both a Minimum Clinically Important Difference (MCID)
and the success margin that can be used to evaluate clinical success. Indicative MCIDs and the
expected improvement in function score post-operatively, as well as standardised rating scores
are provided for some but not all functional scores, refer to Table 8. When available, these values
can inform the design of clinical trials and provide a minimum effect size to determine the
necessary statistical power as well as the clinical interpretation of the data.

Summary of safety and performance data

Reported clinical outcomes of hip, knee and shoulder prostheses

Table 7: Summary of performance data extracted from systematic reviews and primary
research reports on the safety and performance of hip, knee or shoulder arthroplasty

Performance Hip Knee Shoulder

parameter

Revision/reoperation
(time to first revision a a a
and revision rates)

Clinical evidence guidelines: Medical devices Page 85 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Performance Hip Knee Shoulder

parameter

Function scores a a a
Harris Hip Score Hospital for Special Western Ontario
(HHS) Surgery Score (HSSS) osteoarthritis of the
Shoulder (WOOS)
Western Ontario and
McMaster Oxford Shoulder
osteoarthritis index Score (OSS)
(WOMAC)
American Shoulder
Bristol Knee Score and Elbow Surgeons
(BKS) Scale (ASESS)
Oxford Knee Score Constant score
(OKS)
Knee Society Score
(KSS)

Quality of Life (QoL) a a

scores
EuroQoL 5D SF36
SF12

Minimum Clinical a a a
Important Difference
(MCID) identified in HHS OKS WOOS
collating evidence for Oxford Hip Score SF
this guidance report. (OHS)
SF 12
WOMAC
WOMAC
EQ-5D
SF 12

Minimum clinically important differences (MCIDs)

If validated MCIDs are available, manufacturers should provide full documentation and justify
their utility when assessing the safety of the device. Alternatively, meaningful MCIDs can be
established using either an anchor-based or distribution-based approach. In this case, the
manufacturer should provide details of the method and assumptions used in determining the
MCIDs in the submission.

MCIDs can be used to establish the size of the trial that is necessary to allow
statistical verification of clinically meaningful outcomes. These also provide a
margin within which a joint prosthesis can be assessed to be as safe as and to
perform as well as a currently available device(s).

Clinical evidence guidelines: Medical devices Page 86 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Table 8: Example MCID and success margins for performance scores identified from
systematic reviews and primary research reports on the safety and performance of hip,
knee or shoulder arthroplasty

Score Grading Success margin Minimum Clinical

post-surgery important
Difference (MCID)

Hip

Harris Hip Score Scale 0 to100 > 20 points range: 7 to 10

(HHS)
poor <70 + radiographically
stable implant
fair 70 to 79,
+ no additional
good 80 to 89, femoral
excellent 90 to 100 reconstruction

Oxford Hip Score Scale 0 to 48 e.g. patients with a range: 5 to 7

(OHS) pre-surgery score of
0 to 19 may indicate 0 to 19 and receiving
severe hip arthritis a total hip
20 to 29 may indicate replacement
moderate to severe Absolute change at
hip arthritis 6mo post-surgery
30 to 39 may indicate 19 (95% CI 16.6 to
mild to moderate hip 21.4)
arthritis
40 to 48 may indicate
satisfactory joint
function

Western Ontario and 8

McMaster
Osteoarthritis Index
(WOMAC)

Clinical evidence guidelines: Medical devices Page 87 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Score Grading Success margin Minimum Clinical

post-surgery important
Difference (MCID)

Knee

Oxford Knee Score Scale 0 to 48 e.g. patients with a 5 [95% CI 4.4 to 5.5]
(OKS) pre-surgery score of
0 to 19 may indicate 0 to 19 and receiving
severe knee arthritis a total knee
20 to 29 may indicate replacement (39)
moderate to severe Absolute change at
knee arthritis 6mo post-surgery
30 to 39 may indicate 14 (95% CI 12.7 to
mild to moderate 15.3)
knee arthritis
40 to 48 may indicate
satisfactory joint
function

Western Ontario and for TKR: ~15

McMaster
Osteoarthritis index
(WOMAC)

Shoulder

Western Ontario Primary Shoulder

Osteoarthritis of the replacement: ~ 10%
Shoulder Index
(WOOS)

Constant Shoulder Ratings;

Score
>30 poor
21 to 30 fair
11 to 20 good
<11 excellent

Quality of life

EQ 5D Hip: 0.074

SF12 4.5 [95% CI 3.9 to

5.2]

SF36 Multiple MCIDs for

specific SF 36
domains

Clinical evidence guidelines: Medical devices Page 88 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Post Market Surveillance of Joint Prostheses

Post market surveillance aims to ensure that compliance with the Essential Principles (EPs) as
required by the Therapeutic Goods Act 1989 is ongoing throughout the lifecycle of a medical
device, and to identify any issues that require investigation. The EPs of particular relevance to
the clinical assessment are 1, 2, 3, 4, 6 and 14.
However, it should be noted that although a device may have demonstrated compliance with the
EPs at the time of premarket assessment, this can change over time, for example, due to the
emergence of new safety concerns. If it appears that the benefit risk balance of a device has
become unfavourable, the TGA may take action to mitigate the risk to public health.
One source of data the TGA utilises to conduct routine post market surveillance of ARTG-
included joint prostheses is the AOANJRR Annual Report.
The Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) is a
government-funded organisation that collects data on joint replacement (hip, knee and
shoulder) and revision procedures performed in Australia. Specifically, the AOANJRR has
collected data on 98 to 100% of hip and knee joint replacement procedures performed in
Australia since 2003. 8 The information that is collected for the primary joint replacement
procedure includes, but is not limited to, de-identified baseline patient demographics, the name
and brand of the device implanted, method of prosthesis fixation, surgical technique, the
indication for the joint replacement and the date and location of the procedure. When a revision
occurs, the same information is collected.
The Registry collates and analyses this data using statistical methods to produce an estimate of
the revision rate for each individual prosthesis, enabling comparison between different
prostheses to occur. The revision rate is currently the key performance and safety indicator for
joint prostheses. The Registry publishes an Annual Report in October that contains detailed
information about revision rates and reasons for revision, and also identifies prostheses with a
“Higher Than Anticipated Revision Rate”.
The TGA utilises this information, in conjunction with other sources of data, for example, post
market surveillance data collected by companies and published literature, to monitor the real-
world performance and safety of joint prostheses. In addition to the publicly available
information from the AOANJRR, the TGA (like manufacturers, hospitals and researchers) can
request additional analyses from the AOANJRR where required as part of routine post market
surveillance of individual joint prostheses.
Outcomes from post market surveillance may indicate the need for intervention to mitigate the
increased risk of revision with a particular prosthesis. There are a number of possible actions
that can be initiated by a sponsor or required by the TGA . These include, but are not limited to:
• cancellation from the ARTG
• A product recall
• Product Defect Correction
• Hazard Alert
The Recalls page on the TGA website 9 contains further information about the types and levels of
recall and non-recall actions.

8 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6554112/
9 https://www.tga.gov.au/recalls
Clinical evidence guidelines: Medical devices Page 89 of 178
V3.1 June 2022
 Therapeutic Goods Administration

Below are some hypothetical illustrative examples of the post market surveillance process for
joint prostheses, facilitated by the analysis of registry data:

Example 1:
Company A makes a total knee joint replacement ‘ABC’ that was identified by the AOANJRR as
having a higher than anticipated revision rate. . An investigation by the TGA and Company A
determined that the revision rate was higher when the patella was not being resurfaced at the
time of primary procedure. The revision rate when the patella was resurfaced was within
normal limits. The company, in conjunction with the TGA recalls section, issued a Hazard Alert to
remind surgeons to resurface the patella at the time of primary procedure to mitigate the risk of
revision.

Example 2:
Company X makes a femoral stem ‘G’ and a compatible acetabular cup ‘H’. Company Y makes a
femoral stem ‘E’ and a compatible acetabular cup ‘R’. The AOANJRR identified the combination of
stem G with cup R as having a higher than anticipated revision rate. A joint investigation by the
TGA and Companies X and Y determined that neither company condoned the use of their femoral
stem/acetabular cup with those manufactured by different companies. The revision rates of
femoral stem G and acetabular cup R were observed to be within normal limits when used
appropriately with the same company’s respective components. The TGA issued a Medical
Device Safety Update that was also distributed to specialty colleges to remind surgeons to avoid
combining femoral stem G with acetabular cup R.

Example 3:
Company X , Company Y, Company Z all manufacture a certain type of patellar resurfacing
device. The indications for using the device in question have been very broad, covering both
primary and revision surgery. An end user has flagged concerns to the TGA regarding the
medium to long term performance of this class of device. The 5 to 15 yr CRRs are between 2 to 3
times that of the rest of the class. The TGA contacts the AOANJRR and requests all data
pertaining to these devices. Analysis of medium to long term data demonstrated significantly
elevated CRRs for these devices compared to the rest of the class. The TGA contacts the
manufacturers involved. The manufacturer is asked to withdraw the devices in question or to
refine the intended use to very specific clinical indications.

Patient Reported Outcome Measures (PROMs) are expected to become

increasingly relevant as a performance and safety indicator, in addition to revision
rates. A patient reported outcome is defined as any report of a patient’s health
status that comes directly from the patient without interpretation by others. 10
PROMs evaluate the patient’s perspective in terms of pain, function, health-related
quality of life and complications. In this capacity, PROMs may represent a means
of identifying patients with joint replacements who have not proceeded to a
revision surgery, however have not obtained any benefit in terms of improvement
in pain and function.

10 https://aoanjrr.sahmri.com/documents/10180/681914/AOANJRR+PROMs+Pilot+Final+Report
Clinical evidence guidelines: Medical devices Page 90 of 178
V3.1 June 2022
 Therapeutic Goods Administration

Cardiovascular devices to promote patency or functional

flow
This section provides an overview of the clinical evidence that can be used to establish the safety
and performance of cardiovascular (CV) devices to promote patency or functional flow (‘CV flow
implants’).
It provides information on:
• the minimum levels of evidence that are appropriate and useful in assessing the safety and
performance of CV flow implants
• the minimum clinical outcomes that define clinical success and demonstrate that a CV flow
implant performs as intended.

Summary recommendations
• The CV flow implants discussed here, namely arterial stents-carotid, coronary and
peripheral, implants for abdominal aortic aneurysms (AAA) repair, implants for patent
ductus arteriosus (PDA) repair, and inferior vena cava (IVC) filters to prevent pulmonary
embolism (PE) are complex medical devices that may be used in combination with other
devices or components. Manufacturers are advised to list the likely combinations and
provide clinical evidence to support the safety and performance of the new device(s) for
these nominated configurations.
• For submissions reliant on comparable device data, manufacturers are advised to submit all
relevant documents with a supporting justification by a clinical expert to:
– establish substantial equivalence between the device and the nominated comparable
device, and
– confirm that any identified differences will not adversely affect safety and performance
of the device.
• Manufacturers should provide details of the clinical context within which the clinical data
were obtained. The clinical context of the evidence base should be congruent with the
indication(s) for use.
– Patient details are critical when comparing pre- and post-market data. Patient selection
may differ in these scenarios and result in patients of different risk profiles for failure
or adverse events. Risk of such bias should be identified and addressed in the CER.
• Provision of clinical data
– Manufacturers who intend to conduct clinical trials should design trials to the highest
practical NHMRC Level of Evidence. Trials should be appropriate to inform on the
safety and performance of the device for its intended purpose
– Use of the acute (< 48h), sub-acute (< 30days), late (< 1year) or very late (> 1 year)
timeline should be considered. However, for temporary devices the timeline should be
congruent with the in vivo dwell time
– The main clinical outcomes that determine safety and performance of CV flow implants
vary significantly by device type; for example, (a) a common primary outcome measure
for carotid stent studies is a composite of death or stroke (or death, stroke or
myocardial infarct (MI)); (b) a common primary outcome measure for coronary stents
is target lesion revascularisation (TLR) and/or total vessel revascularisation (TVR);
and (c) common primary outcome measures for IVC filters are PE (fatal and non-fatal),

Clinical evidence guidelines: Medical devices Page 91 of 178

V3.1 June 2022
 Therapeutic Goods Administration

deep vein thrombosis (DVT) and occurrence of a venous thromboembolism (VTE)

distal to the filter.
▪ It is advised that a clinical justification is provided to support the selection of the
primary outcomes and if necessary the use of secondary outcomes or surrogate
markers
▪ The manufacturer is advised to benchmark the device against devices of the same
class as reported in appropriate registers (if available) or provide direct
comparative data with a comparable device
▪ For patient performance data, manufacturers are advised to define the anticipated
improvement in patient scores post-surgery. Ideally, these should be
internationally recognised assessment tool(s) used to measure clinical success, e.g.
QoL or exercise stress test
– The manufacturers should consider using surrogate markers that are predictive of
implant failure when in vivo times are longer than one year. For example, use of
endoleak type II with aneurysm expansion to predict late failure of AAA. However, a
clinical justification is needed to support the selection of surrogates and the predicative
power of surrogates should be validated
– It is recommended that the manufacturer supply post-market data if the device is
approved and marketed in another jurisdiction to demonstrate long-term safety and
performance outcomes
– When submitting a comprehensive literature review, full details of the search method
used should be included in the CER with detail sufficient to enable the review process
to be repeated by clinical assessors
– Risks identified in the clinical data should be appropriately mitigated and/or included
in the IFU and other information supplied with the device.
• Compilation of the CER
– in compiling the clinical evidence for a supportive device the manufacturer must
ensure that an appropriate clinical expert, that is, someone with relevant medical
qualifications and direct clinical experience in the use of the device or device type in a
clinical setting, critically evaluates all the clinical data that informs on the safety and
performance of the device
– the clinical expert must then endorse the CER (evidenced by signature and date)
containing the clinical evidence to demonstrate that the evidence meets the
requirements of the applicable EPs and the device is safe and performs as intended.

Clinical evidence guidelines: Medical devices Page 92 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Defining CV flow implants

The guidance in this section applies to the following CV flow implants:
• Arterial stents (carotid, coronary and peripheral)
• Implants for abdominal aortic aneurysms (AAA) repair
• Implants for patent ductus arteriosus (PDA) repair
• Inferior vena cava (IVC) filters to prevent pulmonary embolism

Arterial stents-carotid, coronary and peripheral

Arterial stents are metal mesh devices used to correct the pathological narrowing of an artery
and to maintain patency e.g., in the neck, heart or vessels of the leg. The aim of a stent is to act as
a scaffold to keep the artery open to maintain blood flow and prevent re-stenosis. Using an
endovascular approach, a fine wire is inserted into the femoral artery (or other suitable vessel)
and passed through the blood vessels into the artery with the blockage. The stent is passed along
the wire, often after pre-dilation of the narrowing using a balloon catheter. Stents come in
varying diameters, lengths, and shapes and may be self-expandable. They may be “bare metal”
(without any coating, often made of stainless steel or cobalt chromium alloy) or “drug eluting”
(coated with a drug such as sirolimus or paclitaxel to help prevent restenosis).

Implants for abdominal aortic aneurysm (AAA) repair

While open surgical repair remains the treatment of choice for abdominal aortic repair
endovascular repair is becoming more frequently used. AAA grafts have been developed by a
number of manufacturers and are generally woven polyester, some with a nitinol exoskeleton.
These come in different shapes such as straight, bifurcated and fenestrated devices with various
inbuilt systems to attach the device to the patient’s aorta.

Implants for patent ductus arteriosus (PDA) repair

Minimally-invasive transcatheter closure of PDAs has become the preferred method of
treatment for children beyond the neonatal period, versus surgical closure with ligation or
division of the ductus arteriosus through a thoracotomy incision. PDA implants have been
developed by a number of manufacturers with treatment choice based on the size of the PDA,
e.g. stainless steel coils which may be used for small PDAs; devices such as a self-expanding
device made of nitinol wire mesh and polyester for larger PDAs.

Inferior vena cava (IVC) filters

IVC filters are intended to prevent pulmonary embolism. The filters are metal alloy devices,
generally in an umbrella shape, that are inserted into the inferior vena cava in order to
mechanically trap fragmented clots from the deep leg veins to prevent their movement to the
pulmonary circulation. Filters are designed to be introduced percutaneously. The latest
generation of filters are temporary or ’retrievable’ and are designed to be removed 2 to 12
weeks after insertion (as specified by the manufacturer) if their use is no longer required.

Clinical evidence
The clinical evidence can be derived from clinical investigation(s) data, a comprehensive
literature review and/or clinical experience (generally post-market data) from the use of the
device and/or comparable device. The intended purpose, clinical indications, claims and
contraindications must be supported by the clinical data.
It is important to clarify if any changes have been made to the device since the clinical data were
gathered and if so to document the changes and to clarify the exact version of the device. Direct

Clinical evidence guidelines: Medical devices Page 93 of 178

V3.1 June 2022
 Therapeutic Goods Administration

clinical evidence on the actual device is preferred. Otherwise indirect clinical evidence may be
used after substantial equivalence has been demonstrated through a comparison of the clinical,
technical and biological characteristics as described in Comparable devices including
substantially equivalent devices.
Where the device and the predicate share any common design origin, the lineage between the
devices should be provided as well as a list of other devices that may be used in conjunction with
the new device for example the delivery system, such as the catheter system for stents, including
any balloons. Manufacturers should refer to Clinical evidence requirements for more
information.

Clinical investigation(s)
The design of the clinical investigation should be appropriate to generate valid measures of
clinical performance and safety. The preferred design is a randomised controlled clinical trial
and conditions should ideally represent clinical practice in Australia. All device characteristics
and the intended purpose(s) must be specified when designing clinical investigations including
for devices using data from a comparable device as these will determine the criteria for a full
and reasoned clinical justification for the selection. The eligible patient groups should be clearly
defined with exclusion/inclusion criteria. Manufacturers are advised to justify the number of
patients recruited according to sound scientific reasoning through statistical power calculation.
The duration of the clinical investigation should be appropriate to the device and the patient
population and medical conditions for which it is intended to be used. Duration should always
be justified, taking into account the timeframe of expected complications. CV flow implants must
have long in vivo lives without exposing recipients to unduly high risks. Medication which may
affect outcomes, for example anticoagulant treatment must be taken into account when
determining all endpoints. Analysis of clinical events should be blinded and independently
adjudicated wherever possible.

Literature review
A literature review involves the systematic identification, synthesis and analysis of all available
published and unpublished literature, favourable and unfavourable, on the device, or, if relying
on indirect evidence, comparable device to which substantial equivalence has been established
as described in Comparable devices including substantially equivalent devices.
Data on the materials used to construct the device, its dimensions and geometry, the
components with which it will be used and the intended purpose will define the construction of
search strategies as well as study selection. This ensures that the searches are comprehensive
and the included studies are relevant to the device and/or comparable device. The selection of a
comparable device should be made prior to performing the literature selection, extraction of the
clinical data and analysis of the pooled results. A full description of the device used in any given
study must be extractable from the study report or adequate information to identify the device
(e.g. manufacturer name and model number). If this is not possible, the study should be excluded
from the review.
Literature review describes the process of performing a literature review, summarised briefly
below. As a minimum a literature review should include:
• a search protocol: determined prior to implementing the search, that details the aim, search
terms, planned steps, inclusion and exclusion criteria
• selection strategy: the citations should be assessed against clearly defined selection criteria
documenting the results of each search step with clear detail of how each citation did or did
not fit the selection criteria for inclusion in the review.
• a review and critical analysis: the selected literature should be synthesised and critiqued

Clinical evidence guidelines: Medical devices Page 94 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• a literature report: a report should be prepared which must be critically evaluated and
endorsed (evidenced by signature and date) by a competent clinical expert, containing a
critical appraisal of the compilation.
It is important that the published literature is able to establish the clinical performance and
safety of the device, and demonstrate a favourable benefit-risk profile.

Post-market data
Post-market data can be provided for the actual device or for a comparable device, refer to
Clinical evidence requirements. It is particularly important to include the following:
• information about the regulatory status of the device (or comparable device if relying on
this), including the certificate number, date of issue and name under which the device is
marketed, the exact wording of the intended purpose/approved indication(s) and other
details such as MRI status in other jurisdictions
• any regulatory action including CE mark withdrawals, recalls, including recalls for product
correction, suspensions, removals, cancellations, voluntary recalls in any jurisdiction (and
the reason for these i.e. IFU changes) or other corrective actions occurring in the market as
reported to or required by regulatory bodies
• distribution numbers of the device(s) including distribution by country and/or geographical
region for every year since launch. It is accepted that this may not always be appropriate for
high volume devices, those with many components or those on the market for many years
• the number of years of use
• for every year since launch, the number of complaints, vigilance and monitoring reports and
adverse events categorised by type and clinical outcome
• explanted devices returned to manufacturers should be accounted for with an explanation of
device failures and corrective measures.
For further details refer to Post-market data. Publicly available post-market data such as
adverse event reporting on the FDA MAUDE database and the TGA IRIS should be provided for
all devices including those from other manufacturers. The manufacturers should include post-
market surveillance data from national jurisdictions where the device is approved for clinical
use.
For reports of adverse events and complaints and restenosis, for example, to be a useful adjunct
to other forms of clinical evidence, the manufacturer should make an active, concerted effort to
collect the reports and to encourage users to report incidents. Experience shows that merely
relying on spontaneous reports leads to underestimation of the incidence of problems and
adverse events.
The post-market data should be critically evaluated by a competent clinical expert to enable an
understanding of the safety and performance profile of the device(s) in a ‘real-world’ setting.

Compiling the CER

Clinical outcomes to define the safety and performance of the CV flow devices were identified
from clinical studies published in the peer reviewed literature. In compiling the clinical evidence
the manufacturer should ensure that a clinical expert in the relevant field critically evaluates all
the clinical data from clinical investigation(s), literature review and/or post-market data
(clinical experience) and endorses the CER (evidenced by signature and date), to demonstrate
that the clinical evidence is sufficient to comply with the applicable EPs and that the device is
safe and performs as intended.

Clinical evidence guidelines: Medical devices Page 95 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Previous sections outline the components that may comprise clinical evidence for a medical
device and the recommended process of compiling a CER. These guidance documents apply
whether the manufacturer is using direct clinical evidence or relying on indirect clinical
evidence from a comparable device. Guidance on defining a comparable device is provided in
Comparable devices including substantially equivalent devices.
As per The Clinical Evaluation Report the CER should include the following:
a) General details
b) Description of the medical device and its intended application
c) Intended therapeutic and/or diagnostic indications and claims
d) Context of the evaluation and choice of clinical data types
e) Summary of relevant pre-clinical data
f) Discussion regarding comparable devices including substantially equivalent devices
g) Summary of the clinical data and appraisal
h) Data analysis
i) Conclusions
j) Name, signature and curriculum vitae of clinical expert and date of report

Supportive data and information

The following information on the device must also be provided:
• risk assessment and management document
• IFU, labelling, product manual and all other documents supplied with the device. These must
highlight the risks and ensure that they are appropriately communicated to user.
Additional information should be provided as applicable. This may include (but is not limited
to):
• additional specifications of the device(s)
• the materials from which the device is made including chemical composition
• other devices that may be used in conjunction with the device
• any aspects of non-clinical testing results that inform the design of the clinical trial should be
included in the supporting documents
• biocompatibility testing, bench testing and animal studies where applicable
• specific testing of any adjuvant medicinal components may be required especially if these
are new chemical entities in the Australian context. This should cover interactions between
the device and the medicine, pharmacodynamics and time-release profiles.
• any further details of post-market data
When relying on a comparable device for CV flow implants with the same intended purpose, a
comparison of the technical and physical characteristics of the device and comparable device
should be demonstrated through direct testing in order to establish substantial equivalence.

Clinical evidence guidelines: Medical devices Page 96 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• the technical characteristics of the device include, but are not limited to; the material of the
implant including chemical composition; dimensions; geometry; weight; coating; mechanical
properties such as tensile strength; integrity including fatigue testing; biocompatibility and
behaviour and effects and appearance of the device with magnetic resonance imaging
• the technical characteristics of required delivery systems such as the delivery systems for
stents (including balloons). In such cases, sample specifications would cover, for example:
diameter and profile; bonding pressure at bonded junctions; maximum pressure for
balloons; balloon inflation and deflation times; and stent diameter versus balloon inflation
pressure
• a supporting justification by a clinical expert is required to establish substantial equivalence
between the device and the comparable device, and confirm that any identified differences in
the technical and physical characteristics will not adversely affect safety and performance of
the device
• the use of more than one comparable device is discouraged; however, these may be used if
each is a valid comparable device and each is found to be substantially equivalent to the new
device under consideration
• a clinical justification should be presented as to why direct clinical data are either not
required, or only partially required.
The comparable device must have clinical data to support its safety and performance and all
supporting data must be provided with the CER. As time since first approval lengthens
comparable device data becomes less relevant and should be replaced by data derived from
clinical experience with the device.

Defining clinical success

For the selected CV flow devices, the literature did not generally separate outcomes into those
related to safety and those related to performance. For that reason, all outcomes are reported
together here, separated into the four types of flow devices. Outcomes were often a mix of final
outcomes such as MI, stroke and death, and surrogate outcomes such as restenosis, TVR and
clinical improvement.

Arterial stents
Table 9 (below) provides a summary of the clinical outcomes used to assess safety and
performance of coronary, carotid and peripheral stents as reported in clinical trials included in
the identified systematic reviews. These data are indicative of outcome measures commonly
reported for these three devices but should not be considered exhaustive.

Clinical evidence guidelines: Medical devices Page 97 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Table 9: Clinical outcomes for three classes of arterial stents reported in the clinical trials
included in the systematic review evidence base

Outcomes reported in studies Carotid* Coronary Peripheral

Composite of death or stroke OR a a**

death or stroke or MI (1* outcome)

TVR and/or TLR a a

(1* outcome) (TLR)

Restenosis a a a

Stroke (disabling / major) a

TIA a

MI a a
(recurrent)

Facial neuropathy / cranial nerve a

palsy

Death a a a

Stent thrombosis (definite or a

probable; also early or late)

MACE a

Technical / procedural success a a

Vessel patency assessed via duplex a

US and/or angiography

Reintervention a

Amputation a

Clinical improvement as per the a

Rutherford Scale

Hemodynamic improvement a

Clinical evidence guidelines: Medical devices Page 98 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Outcomes reported in studies Carotid* Coronary Peripheral

Length of follow-up in included SRs 1 month to 4 6 months to 6 6 months to 2 or 3

years (one to 11 years (most 3-5 years (one to 8
years) or 6 years) years)
The CREST Late events up
study109: to 1 year but
longer
Baseline (pre- timelines may
procedure) then be required**
18 & 54h post-
procedure then
1, 6 and 12
months then
annually
thereafter

KEY: MI=myocardial infarction, TLR=target lesion revascularisation, TVR=total vessel revascularisation,

TIA=transient ischemic attack, MACE= major adverse cardiac events, US=ultrasound, SR=systematic
review
* Outcomes were often divided into <30 day (peri-procedural) or >30 day outcomes
** Outcomes defined in the European Commission MEDDEV 2.7/1 and Academic Research Consortium

Coronary stents
Outcomes were often divided into <30 day (peri-procedural) or >30 day outcomes. Adverse
events within the peri-procedural periods may be related to the procedure while those
occurring after 30 days are more likely to represent device-related events. Adverse events for
coronary stents and the timing of these may be described differently in the literature.
Manufacturers are advised to use standardised definitions for clinical endpoints for coronary
stents as defined by the Academic Research Consortium (ARC), in 2007. The ARC nominated
clinical outcomes have been adopted by the European Commission in their guidance MEDDEV
2.7/1. These include, but are not limited to, outcomes listed in Table 9 (above). The MEDDEV
2.7/1 and ARC also address criteria for collecting clinical data and the use of composite clinical
outcomes. These include:
• Composite adverse events divided into device–oriented (cardiac death, MI, TLR) and patient-
oriented (all-cause mortality, any MI, any repeat revascularisation)
• Composite acronyms such as MACE (major adverse cardiac events) should be used with
caution because of the varied definitions of MACE used clinically and in research
• If MACE is the nominated clinical endpoint, manufacturers are advised to provide a clear
definition with clinical justification for the elements included in this composite measure.
Manufacturers should also provide evidence of clinical device success. Typically this will include
the successful delivery and deployment of the device, removal of the stent delivery system and
final residual stenosis of <50% of the target lesion as assessed by Quantitative Coronary
Angiography. Clinical procedural success includes the previous measures associated with stent
deployment and stenosis reduction with the additional parameter that there are no ischemia
driven adverse events to a maximum of seven days post procedure.
Patient follow-up should be reported for acute (0 – 2 hours), sub-acute (> 24 hours to 30 days),
late (> 30 days to 1 year) and very late (> 1 year) events. This timeline is in line with reported
patient follow-up times in the peer-reviewed literature (Table 9 & 11).

Clinical evidence guidelines: Medical devices Page 99 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Carotid stents
Outcomes were divided into <30 day (peri-procedural) or >30 day outcomes, with the main
primary outcomes being a composite of meaningful endpoints such as:
• death or stroke or MI
• secondary outcomes included a mix of surrogate and final outcomes such as restenosis,
stroke, disabling/major stroke, transient ischemic attack (TIA), MI, facial neuropathy/cranial
nerve palsy, and death

Manufacturers are advised to use a validated stroke assessment tool e.g. the
National Institute of Health Stroke Scale to evaluate patients pre- and post-
procedure.

Across the research literature the rates at which adverse events occur are highly variable. The
diversity is due to differences in patient groups (symptomatic vs. asymptomatic), operator
experience and technique, medical management goals and the primary study endpoints.
All will affect the rate at which adverse events occur and whether these rates may be considered
clinically acceptable for a given patient cohort.
Examples of indicative rates for death, stroke and MI events are reported for the CREST clinical
trial.109 These are reported as % ± SD:
• Peri-procedure (< 30days)
– Death; 0.7% ± 0.2
– Stroke (any) ; 4.1% ± 0.6
– MI; 1.1 ± 0.3
• After 4 years including peri-procedural period
– Death; 11.3% ± 1.2
– Stroke (any) ; 10.2% ± 1.1
However manufacturers are advised to provide a clinical justification of the event rates deemed
to be acceptable for the target patient population in which the carotid stent is to be used.
Procedural success requires a successful deployment of stent and withdrawal of delivery system
with a < 30% residual stenosis.
Similar to coronary stents, patient follow-up should be reported for acute, sub-acute, late and
very late time points as indicated. This timeline is in line with patient follow-up reported in the
studies included in the systematic reviews examined for this report and ranged from 1 month to
at least 4 years with one study extending to 11 years.

Peripheral stents
Peripheral stents are used for the treatment of peripheral artery disease (PAD). Outcomes
included a mix of surrogate and final outcomes including:
• Technical success, vessel patency assessed via duplex ultrasound and/or angiography, TLR,
restenosis, reintervention, amputation, clinical improvement as per the Rutherford Scale,
hemodynamic improvement, and death (Table 9, 10 & 11).

Clinical evidence guidelines: Medical devices Page 100 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Examples of safety and performance values for some parameters include, but are not limited to,
the following:
• Primary success of 95% with a 5% restenosis at 1 year has been report for nitinol stents.
However, restenosis rates at 1 year range from 5% to 25%, depending on lesion length and
location;
• For patients included in the Excellence in Peripheral Arterial Disease (XLPAD) registry for
the treatment of symptomatic infrainguinal PAD adverse events at 1 year follow-up include:
– Amputation of target limb; 4.6%
– MI; 1.9%
– Target vessel thrombosis; 4.1%
– Need for surgical revasculisation; 5.9%
• Technical success has been report to be greater than 95%
• Given the physical dimensions of this class of stent, stent fracture may occur at rates in
excess of 30% of treated legs.115 Stent fracture significantly impacts primary patency rates
and manufacturers are advised to report these rates
• Patency at 1 and 3 years are reported to be 69 to 79% and 59 to 70% respectively.
Generalised safety and performance values cannot be provided because of the heterogeneity in
lesion anatomy and location, stent size, materials and associated stent technologies. Therefore
manufacturers are advised to:
• define the patient cohort and provide a clinical justification for selected safety and
performance parameters
• define the lesion anatomy according to a recognised classification system e.g. TransAtlantic
Inter-Society Consensus.
Follow-up in the studies included in the systematic reviews examined for this report ranged
from 6 months to 2 or 3 years with one study extending to 8 years. These are in line with patient
follow-up based on the acute (< 48h), sub-acute (< 30days), late (< 1year) or very late (> 1 year)
timeline.

Implants for AAA repair

Much of the evidence focussed on adverse events (AEs) and post-operative complications, as
well as mortality (30-day, aneurysm-related and all-cause) - Table 11. Additional outcomes were
a mix of surrogate and final outcomes and include:
• Reintervention rates (including conversion from endovascular aneurysm repair [EVAR] to
an open procedure), MI, stroke, renal failure and aortic rupture
• Secondary outcomes focussed on practical and logistical issues such as procedure time,
blood loss, fluoroscopy time, contrast load, recovery time, need for blood transfusion, days in
an intensive care unit (ICU) and length of hospital stay (LOHS).
Clinical success is defined by a consideration of both clinical and radiological criteria and
standards. These include:
• Deployment of the device at the intended location without death as a result of the
intervention.
• Absence of Type I and Type III endoleaks.

Clinical evidence guidelines: Medical devices Page 101 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• Aneurysm expansion of ≤ 5mm in diameter or ≤ 5% volume.

• Absence of aneurysm rupture or need to convert to open surgery.
In contrast clinical failure is defined as:
• Graft dilation of > 20% in diameter or persistent increase in aneurysm size.
• Graft migration or failure of device to integrate.
• Type II endoleak with an aneurysm expansion.
Manufacturers should specify the time period for clinical success. Life table or Kaplan Meier
estimates should not have standard deviations of greater than 10%.
Any changes in lesion anatomy during follow-up should be referenced to measures taken
immediately post-procedure.
Technical success is defined as the successful deployment and removal of the delivery device
without the need for surgical conversion or mortality. Chaikof et al further qualified technical
success to include:
• Access to arterial system using a remote site (e.g. femoral artery) with or without a
permanent conduit to access the site
• Deployment of endoluminal graft with secure proximal and distal fixation
• Absence of type I or type III endoleak
• Patent endoluminal graft without twists, kinks, or obstruction (> 30% stenosis or pressure
gradient of > 10 mmHg).
• The need for additional modular components, stents and adjunctive surgical procedures
should be reported.
Follow-up in the studies included in the systematic reviews examined for this report ranged
from 30 days (peri-procedural) to 9 years. Again these are in line with patient follow-up based
on the acute (< 48h), sub-acute (< 30days), late (< 1year) or very late (> 1 year) timeline.

Implants for PDA repair

Outcomes of primary interest were adverse events and the surrogate outcomes of primary
success, residual shunt and need for blood transfusion. Manufacturers need to provide clear
patient characteristics and lesion anatomy. Clinical evidence should be provided for all lesion
types that are included in the indication(s) for use of the implant. The diversity of lesion size and
heterogeneity of currently marketed devices for PDA repair limits the generation of generalised
safety and performance values. Manufacturers are advised to provide a justification for the
selected clinical outcomes and values that define clinical and technical success.
The following values have been reported in the literature and serve as a guide to acceptable
safety and performance for a PDA device:
• Clinical success based on the absence of non-trivial residual angiographic shunt is report to
be 90 to 96% for two commercially available devices
• Manufacturers are advised to demonstrate PDA closure rate at implant, 24 hours post-
procedure and at appropriate clinical follow-up. Follow-up has been reported at 1, 2 and 5
years. Patient follow-up and assessment method should be supported with a clinical
justification

Clinical evidence guidelines: Medical devices Page 102 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• Major adverse events (e.g. device embolization, device malposition) have been reported to
occur at 2.2% (95% CI 1.0 to 3.7).
Follow-up in the studies included in the systematic review examined for this report was unclear
but was possibly 6 months. However, manufacturers are advised that follow-up should be
reported for the peri–procedure period as well as late (≤1 year) and very late (≥ one year) time
points.

IVC filters to prevent PE

Of primary interest were adverse events, PE (including fatal PE), DVT, and occurrence of a VTE
distal to the filter. Manufacturers are advised to provide details of target patient baseline risk for
PE, operator experience and technique, medical management goals and the primary study
endpoints. These have been shown to be independently associated with adverse events.
The following safety and performance values are indicative and are provided to assist the
manufacturer in the preparation of submissions. The list is not exhaustive and should be
considered as a guide only.
• Fatal PE is not frequently reported and manufacturers should use appropriate study designs
with sufficient power to detect such events when possible. If meta-analysis is performed,
then the Peto Odds methods for rare events should be considered.
• Based on the IVC filter registry maintained by British Society of Interventional Radiology
(BSIR) more than 96% of filters were deployed as intended. However, manufacturers should
report the filter orientation on deployment (i.e. centralised, tilted or abutting the IVC wall).
• Manufacturers should report the dwell time for the device and the impact on retrieval for
temporary devices.
• Any structural failure should be reported.
• Manufacturers are advised that DVT was reported to be lower than the 1% in BSIR registry
data. However, the clinical profile of the patient cohort may affect this adverse event.
Therefore, manufacturers are advised to provide a clinical justification for expected DVT
rates in the target population.
• Perforations are the most common long-term adverse event occurring in 0.3 to 14% of filter
deployments; the range may reflect differences in IVC filter type.
• The BSIR IVC registry requires notification of filter migration of > 10mm. Manufacturers are
advised to report any filter migrations.
• Mortality rates reported for the BSIR IVC registry ranged from 4.3 to 12.3% depending on
filter type, dwell time and clinical condition of the patient. Manufacturers are advised to
provide a clear clinical context for the use of the IVC filter to assist the clinical assessor to
determine whether the device has a favourable benefit-risk profile.
Similar to other CV devices, technical success is based on the successful deployment of the IVC
filter in the correct orientation and location as well as the removal of the delivery system.

Clinical evidence guidelines: Medical devices Page 103 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Follow-up in the studies included in the systematic reviews examined for this report ranged
from in-hospital only to 8 years. Follow-up periods should be congruent with the in vivo life
span for temporary devices. For permanent devices the acute (< 48h), sub-acute (< 30days), late
(< 1 year) or very late (> 1 year) timeline should be considered.

· Manufacturers, in selecting and reporting surrogate markers of safety and

performance (as described in the previous section) should provide a
clinical justification for the selection and, where possible, should use
validated measurement tools.

· When documenting patient performance scores, it is recommended that

manufacturers provide data with a minimum of one year follow-up post-
surgery to reduce the risk of confounding due to procedure variables.

Clinical evidence guidelines: Medical devices Page 104 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Summary of safety and performance data

Characteristics of clinical studies of CV flow implants

Table 10: Study characteristics extracted from systematic reviews and primary research reports on the safety and performance of selected CV
flow implants

Characteristics Arterial stents: Implants for AAA Implants for PDA IVC filters
of included repair repair
studies Carotid (6 SRs) (2 SRs)
Coronary(6 SRs) (4 SRs) (1 SR)
Peripheral (5 SRs) (1 RCT)
(1 retrospective (1 retrospective
comparative cohort) cohort study)
Carotid Coronary Peripheral

Number of 11 to 41 10 to 28 4 to 14 5 to 32 7 2 and 8
included studies
per SR

Clinical evidence guidelines: Medical devices Page 105 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Characteristics Arterial stents: Implants for AAA Implants for PDA IVC filters
of included repair repair
studies Carotid (6 SRs) (2 SRs)
Coronary(6 SRs) (4 SRs) (1 SR)
Peripheral (5 SRs) (1 RCT)
(1 retrospective (1 retrospective
comparative cohort) cohort study)
Carotid Coronary Peripheral

Dominant design 3 SRs were 5 SRs were 3 SRs were 2 SRs were limited to SR: All Level IV SRs: Levels II-IV
of included limited to RCTs; limited to RCTs; limited to RCTs; RCTs; 1 included RCTs &
studies 3 included a mix 1 included RCTs 1 included SRs & registries; 1 included Primary study: Level RCT=Level II
of MAs, RCTs, & observational RCTs; 1 included RCTs, observational IV
cohort studies, studies RCTs & case cohort studies &
case series & series registries
registry studies

Clinical evidence guidelines: Medical devices Page 106 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Characteristics Arterial stents: Implants for AAA Implants for PDA IVC filters
of included repair repair
studies Carotid (6 SRs) (2 SRs)
Coronary(6 SRs) (4 SRs) (1 SR)
Peripheral (5 SRs) (1 RCT)
(1 retrospective (1 retrospective
comparative cohort) cohort study)
Carotid Coronary Peripheral

Sample size 3 SRs with RCTs: 5 SRs with RCTs: 3 SRs with RCTs: 2 SRs with RCTs: total SR 2014: n=259 SR 2010: 2 RCTs of
(range) for total enrolled = total enrolled = total enrolled = enrolled = 1,594 to patients in device 129 and 400
included studies 4,796 to 7,572 6,298 to 14,740 627 to 1,387 3,194 patients group; n=551 in patients (division
patients patients patients control group between arms NR)
1 SR with RCTs &
3 SRs with 1 SR with RCTs 1 SR with SRs registries; total enrolled Primary study: Level SR 2014: n=432 in
various study and and RCTs; total = 52,220 patients III-2 retrospective filter groups;
designs: total observational enrolled = cohort with concurrent n=4160 in
enrolled = up to studies: total unclear 1 SR with RCTs, controls; n=51 in historical control
575,556 enrolled = observational studies & device group; n=130 in groups
10,447 1 SR with RCTs registries: total enrolled control group
and case series: = 72,114 RCT 2012: total
total enrolled = n=141 (70 in
1,628 Primary study: total device group, 71 in
enrolled = 2,198 control group)

Reported Carotid artery 4 assessed DES Balloon Primarily EVAR versus Implanted device IVC filter versus
comparisons stenting vs. versus BMS; 2 angioplasty with open repair; also EVAR versus surgical closure no filter
endarterectomy assessed DES stents (BMS or versus watchful waiting
(one study also versus BMS or DES) versus in candidates deemed
included medical another type of balloon not fit for surgery
therapy) DES angioplasty
alone (one
compared BMS
versus DES)

Clinical evidence guidelines: Medical devices Page 107 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Characteristics Arterial stents: Implants for AAA Implants for PDA IVC filters
of included repair repair
studies Carotid (6 SRs) (2 SRs)
Coronary(6 SRs) (4 SRs) (1 SR)
Peripheral (5 SRs) (1 RCT)
(1 retrospective (1 retrospective
comparative cohort) cohort study)
Carotid Coronary Peripheral

Quality of 2 SRs did not 1 SR did not All 5 SRs SRs assessed via Jadad SR: With the NOS, SR 2010: With
included report quality report quality assessed study or Cochrane assessed studies as D&B, assessed
evidence as assessment; 1 assessment; 1 quality using a Collaboration tool. Other having low-risk bias; studies as low
reported developed a developed a variety of tools study types used NOS. funnel plot for primary quality
custom tool but custom tool but (e.g., Cochrane RCT quality usually outcome showed no
did not report did not report Collaboration, high; others low to obvious publication SR 2014: With the
results; 3 used a results; the other Jadad, custom); moderate bias Jadad scale,
tool developed 4 used various quality was assessed studies
by the Cochrane tools and generally as scoring 2/5 &
Collaboration determined assessed as 3/5 (low)
and found risk of studies were moderate to high
bias generally generally high
low quality with low
risk of bias

Patient Follow- From 1 month to Generally 3 to 5 6 months to 8 From post-op course in SR: 6 months SR 2010: NR
up 5 years years years; generally hospital up to 9.1 years
6-24 months Primary study: 24 SR 2014: 34 days
months to 8 years
RCT 2012: 15 (±
SD 2) months

KEY: SD=Standard deviation; SR=Systematic review; RCT=randomized controlled trial KEY: AAA=Abdominal aortic aneurysm; BMS=Bare metal stents; D&B=Downs &
Black; DES=Drug eluting stents; EVAR=endovascular aneurysm repair; IVC=Inferior vena cava; MA=Meta-analysis; NOS=Newcastle-Ottawa scale; NR=not reported;
PDA=Patent ductus

Clinical evidence guidelines: Medical devices Page 108 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Reported clinical outcomes on selected CV flow implants

Table 11: Summary of types of safety and performance data extracted from SRs and
additional primary research on CV flow implants

Type of CV flow Outcomes reported in included research

implant

Arterial stents: • Carotid: often divided into <30 day (peri-procedural) or >30 day
outcomes
• Carotid (6 SRs)
– Primary: Composite of (a) death or stroke OR (b) death or
• Coronary stroke or MI
(6 SRs) – Secondary: Death, stroke / disabling / major stroke, TIA,
• Peripheral MI, facial neuropathy / cranial nerve palsy

(5 SRs) – Restenosis
• Coronary
– TVR and / or TLR
– Death
– Recurrent MI
– Stent thrombosis (definite or probable; also early or late)
– Various composite endpoints such as MACE
• Peripheral
– Death, reintervention, amputation
– Technical success, vessel patency, TLR, restenosis
– Clinical improvement as per Rutherford Scale,
hemodynamic improvement, QOL

Implants for AAA • AEs / postop complications, e.g., MI, stroke, renal failure, aortic
repair rupture
(4 SRs) • Mortality (30-day, aneurysm-related, all-cause)
(1 retrospective • Reintervention rates including conversion from EVAR to open
comparative cohort) procedure
• Secondary endpoints, e.g., QOL, procedure time, blood loss,
blood transfusion, fluoroscopy time, contrast load, recovery
time, days in ICU & LOHS

Clinical evidence guidelines: Medical devices Page 109 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Type of CV flow Outcomes reported in included research

implant

Implants for PDA • AEs

repair
• Primary success
(1 SR)
• Residual shunt
(1 retrospective cohort
study) • Blood transfusion
• LOHS

IVC filters • AEs

(2 SRs) • DVT
(1 RCT) • Fatal PE
• PE
• VTE distal to the filter

KEY: AAA=Abdominal aortic aneurysm; AE=Adverse events; CTA=computed tomography angiography;

DVT=Deep vein thrombosis; EVAR=Endovascular aneurysm repair; ICU=Intensive care unit; IVC=Inferior
vena cava; LOHS=Length of hospital stay; MACE=Major adverse cardiac events; MI=myocardial infarction;
NR=not reported; PE=Pulmonary embolus; PDA=Patent ductus arteriosus; QOL=Quality of life;
SD=Standard deviation; SR=Systematic review; TIA=transient ischemic attack; TLR=target lesion
revascularisation; TVR=total vessel revascularisation; VTE=Venous thromboembolism

References
NICE. Interventional procedure overview of carotid artery stent placement for symptomatic
extracranial carotid stenosis. 2010:IP 008_2.
Canadian Agency of Drugs and Technologies in Health (CADTH). Carotid artery stenting versus
carotid endoarterrectormy, a review of the clinical and cost-effectiveness. Health Technology
Inquiry Service 2010 [cited December 2014]
Raman G, Kitsios GD, Moorthy D et al. AHRQ Technology Assessments. Management of
Asymptomatic Carotid Stenosis. Rockville (MD): Agency for Healthcare Research and Quality
(US) 2012.
Chen Z, Chen L, Wu L. Transcatheter amplatzer occlusion and surgical closure of patent ductus
arteriosus: comparison of effectiveness and costs in a low-income country. Pediatr Cardiol.
2009;30(6):781-5.
Wang K, Pan X, Tang Q, Pang Y. Catheterization therapy vs. surgical closure in pediatric patients
with patent ductus arteriosus: a meta-analysis. Clin Cardiol. 2014;37(3):188-94.
Boehm W, Emmel M, Sreeram N. The Amplatzer duct occluder for PDA closure: indications,
technique of implantation and clinical outcome. Images Paediatr Cardiol. 2007;9(2):16-26.
Huang TC, Chien KJ, Hsieh KS, Lin CC, Lee CL. Comparison of 0.052-inch coils vs. amplatzer duct
occluder for transcatheter closure of moderate to large patent ductus arteriosus. Circ J.
2009;73(2):356-60.
Young T, Tang H, Hughes R. Vena caval filters for the prevention of pulmonary embolism.
Cochrane database of systematic reviews (Online). 2010(2):CD006212.

Clinical evidence guidelines: Medical devices Page 110 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Brott TG, Hobson RW, 2nd, Howard G et al. Stenting versus endarterectomy for treatment of
carotid-artery stenosis. N Engl J Med. 2010;363(1):11-23
Cutlip DE, Windecker S, Mehran R et al. Clinical end points in coronary stent trials: a case for
standardized definitions. Circulation. 2007;115(17):2344-51.
European Commision - Directives on medical devices. Evaluation of Clinical Data: A Guide for
Manufacturers and Notified Bodies. 2008.
Kip KE, Hollabaugh K, Marroquin OC, Williams DO. The problem with composite end points in
cardiovascular studies: the story of major adverse cardiac events and percutaneous coronary
intervention. J Am Coll Cardiol. 2008;51(7):701-7.
O'Brien M, Chandra A. Carotid revascularization: risks and benefits. Vasc Health Risk Manag.
2014;10:403-16.
Schillinger M, Minar E. Percutaneous treatment of peripheral artery disease: novel techniques.
Circulation. 2012;126(20):2433-40.
Kudagi VS, White CJ. Endovascular stents: a review of their use in peripheral arterial disease. Am
J Cardiovasc Drugs. 2013;13(3):199-212.
Norgren L, Hiatt WR, Dormandy JA, Nehler MR, Harris KA, Fowkes FG. Inter-Society Consensus
for the Management of Peripheral Arterial Disease (TASC II). J Vasc Surg. 2007;45 Suppl S:S5-67.
Chaikof EL, Blankensteijn JD, Harris PL et al. Reporting standards for endovascular aortic
aneurysm repair. J Vasc Surg. 2002;35(5):1048-60.
Brunetti MA, Ringel R, Owada C et al. Percutaneous closure of patent ductus arteriosus: a
multiinstitutional registry comparing multiple devices. Catheter Cardiovasc Interv.
2010;76(5):696-702.
El-Said HG, Bratincsak A, Foerster SR et al. Safety of percutaneous patent ductus arteriosus
closure: an unselected multicenter population experience. J Am Heart Assoc.
2013;2(6):e000424.
Uberoi R, Tapping CR, Chalmers N, Allgar V. British Society of Interventional Radiology (BSIR)
Inferior Vena Cava (IVC) Filter Registry. Cardiovasc Intervent Radiol. 2013;36(6):1548-61.
Meier P, Knapp G, Tamhane U, Chaturvedi S, Gurm HS. Short term and intermediate term
comparison of endarterectomy versus stenting for carotid artery stenosis: systematic review
and meta-analysis of randomised controlled clinical trials. BMJ. 2010;340:c467.
Bonati LH, Lyrer P, Ederle J, Featherstone R, Brown MM. Percutaneous transluminal balloon
angioplasty and stenting for carotid artery stenosis. Cochrane Database Syst Rev.
2012;9:Cd000515.
Liu ZJ, Fu WG, Guo ZY, Shen LG, Shi ZY, Li JH. Updated systematic review and meta-analysis of
randomized clinical trials comparing carotid artery stenting and carotid endarterectomy in the
treatment of carotid stenosis. Ann Vasc Surg. 2012;26(4):576-90.
Al-Damluji MS, Nagpal S, Stilp E, Remetz M, Mena C. Carotid revascularization: a systematic
review of the evidence. J Interv Cardiol. 2013;26(4):399-410.
Gahremanpour A, Perin EC, Silva G. Carotid artery stenting versus endarterectomy: a systematic
review. Tex Heart Inst J. 2012;39(4):474-87
Piccolo R, Cassese S, Galasso G, De Rosa R, D'Anna C, Piscione F. Long-term safety and efficacy of
drug-eluting stents in patients with acute myocardial infarction: a meta-analysis of randomized
trials. Atherosclerosis. 2011;217(1):149-57.

Clinical evidence guidelines: Medical devices Page 111 of 178

V3.1 June 2022
 Therapeutic Goods Administration

De Luca G, Dirksen MT, Spaulding C et al. Drug-eluting vs bare-metal stents in primary

angioplasty: a pooled patient-level meta-analysis of randomized trials. Arch Intern Med.
2012;172(8):611-21; discussion 21-2.
Wallace EL, Abdel-Latif A, Charnigo R et al. Meta-analysis of long-term outcomes for drug-eluting
stents versus bare-metal stents in primary percutaneous coronary interventions for ST-segment
elevation myocardial infarction. Am J Cardiol. 2012;109(7):932-40.
Bangalore S, Amoroso N, Fusaro M, Kumar S, Feit F. Outcomes with various drug-eluting or bare
metal stents in patients with ST-segment-elevation myocardial infarction: a mixed treatment
comparison analysis of trial level data from 34 068 patient-years of follow-up from randomized
trials. Circ Cardiovasc Interv. 2013;6(4):378-90.
Palmerini T, Biondi-Zoccai G, Della Riva D et al. Clinical outcomes with drug-eluting and bare-
metal stents in patients with ST-segment elevation myocardial infarction: evidence from a
comprehensive network meta-analysis. J Am Coll Cardiol. 2013;62(6):496-504.
Kalesan B, Pilgrim T, Heinimann K et al. Comparison of drug-eluting stents with bare metal
stents in patients with ST-segment elevation myocardial infarction. Eur Heart J. 2012;33(8):977-
87.
Medical Advisory Secretariat. Stenting for peripheral artery disease of the lower extremities: an
evidence based analysis. Ont Health Technol Assess Ser [Internet]. 10(18)
Acin F, de Haro J, Bleda S, Varela C, Esparza L. Primary nitinol stenting in femoropopliteal
occlusive disease: a meta-analysis of randomized controlled trials. J Endovasc Ther.
2012;19(5):585-95
Chambers D, Epstein D, Walker S et al. Endovascular stents for abdominal aortic aneurysms: a
systematic review and economic model. Health Technol Assess. 2009;13(48):1-189, 215-318, iii.
Simpson EL, Kearns B, Stevenson MD, Cantrell AJ, Littlewood C, Michaels JA. Enhancements to
angioplasty for peripheral arterial occlusive disease: systematic review, cost-effectiveness
assessment and expected value of information analysis. Health Technol Assess. 2014;18(10):1-
252.
Chowdhury MM, McLain AD, Twine CP. Angioplasty versus bare metal stenting for superficial
femoral artery lesions. Cochrane Database Syst Rev. 2014;6:Cd006767.
Stather PW, Sidloff D, Dattani N, Choke E, Bown MJ, Sayers RD. Systematic review and meta-
analysis of the early and late outcomes of open and endovascular repair of abdominal aortic
aneurysm. Br J Surg. 2013;100(7):863-72.
Paravastu SC, Jayarajasingam R, Cottam R, Palfreyman SJ, Michaels JA, Thomas SM. Endovascular
repair of abdominal aortic aneurysm. Cochrane Database Syst Rev. 2014;1:Cd004178.
van Beek SC, Conijn AP, Koelemay MJ, Balm R. Editor's Choice - Endovascular aneurysm repair
versus open repair for patients with a ruptured abdominal aortic aneurysm: a systematic review
and meta-analysis of short-term survival. Eur J Vasc Endovasc Surg. 2014;47(6):593-602.
Edwards ST, Schermerhorn ML, O'Malley AJ et al. Comparative effectiveness of endovascular
versus open repair of ruptured abdominal aortic aneurysm in the Medicare population. J Vasc
Surg. 2014;59(3):575-82.
Haut ER, Garcia LJ, Shihab HM et al. The effectiveness of prophylactic inferior vena cava filters in
trauma patients: a systematic review and meta-analysis. JAMA Surg. 2014;149(2):194-202.
Sharifi M, Bay C, Skrocki L, Lawson D, Mazdeh S. Role of IVC filters in endovenous therapy for
deep venous thrombosis: the FILTER-PEVI (filter implantation to lower thromboembolic risk in
percutaneous endovenous intervention) trial. Cardiovasc Intervent Radiol. 2012;35(6):1408-13.
Clinical evidence guidelines: Medical devices Page 112 of 178
V3.1 June 2022
 Therapeutic Goods Administration

Implantable pulse generator systems

Implantable pulse generator systems are active medical devices that produce electrical
discharges. This section specifically covers cardiac active implantable devices and implantable
electrical nerve stimulation devices.

Summary recommendations
• Implantable pulse generator systems (pacemakers including cardiac resynchronisation
therapy with or without defibrillation (CRT, CRT-D), implantable cardiac defibrillators
(ICDs) and implantable electrical nerve stimulation devices), are complex medical devices
that may be used in combination with other devices or components. Manufacturers are
advised to list all components and combinations and provide clinical evidence to support the
safety and performance of the new device for these nominated configurations.
• Provision of clinical investigation data: Manufacturers who intend to conduct clinical
investigations should use study designs to the highest practical NHMRC Level of Evidence,
and trials should be appropriately designed to inform on the safety and performance of the
device for its intended purpose.
– For Active Implantable Cardiac Devices (AICDs), patient follow-up in clinical trials
should include the peri-operative, acute (≤ 3 months) and chronic (> 3 months) phases,
with the patient then monitored during yearly follow-up visits. Follow-up time should
be sufficient to identify late adverse events. The nominated follow-up periods should be
supported by clinical justification.
– For implantable devices for pain and other neurological symptom control, patient
follow-up for clinical trials should include the peri-operative, acute (≤ 3 months) and
chronic (> 3 months) phases. Due to the chronicity of pain and other neurological
symptoms, performance should be studied for 1 year or longer post device
implantation.
• The clinical outcomes that determine safety and performance of implantable pulse generator
systems vary significantly by device type:
– The manufacturer is advised to benchmark the new device against devices of the same
class as reported by an international registry, if available.
– Nominated values that indicate safety and performance should be appropriate to
patient health status and indicated use and justified by a clinician who is an expert in
the field.
– For patient performance data manufacturers are advised to define the anticipated
improvement in patient scores post-surgery or post-treatment. Ideally, these should be
by an internationally recognised assessment tool(s) used to measure clinical success
e.g. pain assessment via a visual analogue scale.
– When submitting a comprehensive literature review, full details of the method used
should be included in the CER in sufficient detail to ensure the literature review can be
reproduced.
– A well-documented risk assessment and management system should also be provided.
All clinical risks identified in the clinical investigation data, literature review and post-
market clinical experience should inform and be reflected in the risk assessment
documentation. These risks should be appropriately rated and quantified, before
assigning risk reduction activities such as statements in the IFU and training materials
to reduce inherent risks.

Clinical evidence guidelines: Medical devices Page 113 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• For guidance on the conduct of comprehensive literature reviews and presentation of

clinical evidence, manufacturers are directed to the relevant sections and appendices.
– In compiling the clinical evidence for an implantable pulse generator system, the
manufacturer should ensure that a clinical expert, that is, someone with relevant
medical qualifications and direct clinical experience in the use of the device or device
type in a clinical setting, conducts a critical evaluation of all the clinical data that
informs the safety and performance of the device.
– The clinical expert must determine whether the clinical evidence is sufficient to
demonstrate that the device meets the requirements of the applicable EPs, including
that it is deemed to be safe and to perform as intended, and that there is a positive
benefit-risk ratio with regard to its use. The clinical expert should then endorse the CER
(by signature and date).
• A full curriculum vitae of the clinical expert should be included in the CER.

Defining implantable pulse generator systems

These are active medical devices that produce electrical discharges as required for a variety of
treatments, and include (but are not limited to) the following two categories.
• Active Implantable Cardiac Devices (AICD) including:
– single and dual chamber pacemakers
– cardiac resynchronisation therapy pacemakers, with or without defibrillation (i.e. CRT-
D and CRT respectively)
– implantable cardiac defibrillators (ICDs)
• Electrical nerve stimulation devices
– only implantable electrical nerve stimulation devices will be covered in this guidance;
transcutaneous electrical nerve stimulation (TENS) devices are not included.
Implantable pulse generator systems can pose a significant regulatory challenge as they are
active devices that must have long in vivo lives without exposing recipients to unduly high risks
of adverse events.

Clinical evidence
The clinical evidence can be derived from clinical investigation(s) data, a comprehensive
literature review and/or clinical experience (generally post-market data) from the use of the
device (direct evidence) and/or a comparable device (indirect evidence). The intended purpose,
clinical indications, claims and contraindications must be supported by the clinical data.
Manufacturers should refer to Clinical evidence requirements for further information.
Direct clinical evidence on the actual device is preferred. Otherwise indirect clinical evidence
may be used after substantial equivalence has been demonstrated through a comparison of the
clinical, technical and biological characteristics as described in Comparable devices including
substantially equivalent devices.
It is important to indicate if any changes have been made to the device since the clinical data
were gathered and to document these changes and clarify the exact version of the device. The
manufacturer should ensure that combinations of components that are to be included in the IFU
are tested.

Clinical evidence guidelines: Medical devices Page 114 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Clinical investigation(s)
Regardless of design, clinical studies should provide unbiased results that allow an objective
comparison of implantable pulse generators with respect to their safety and performance. To
achieve this for new device applications based on direct clinical data the manufacturers should
ensure that clinical trials are conducted according to internationally recognised standards for a
given trial design, e.g., follow the ISO standard 14155.
Clinical trials must be independently audited at key stages throughout their conduct to
document that the integrity of the trial(s) was maintained. Clinical trial data should be reported
using an internationally recognised standard for a given study design, e.g., the CONSORT
reporting standards for RCTs.
For AICDs patient follow-up in clinical trials should include the peri-operative, acute (≤ 3
months) and chronic (> 3 months) phases, with the patient then monitored during yearly follow-
up visits. Follow-up time should be sufficient to identify late adverse events. The nominated
follow-up periods should be supported by clinical justification.
For implantable devices for pain and other neurological symptom control, patient follow-up for
clinical trials should include the peri-operative, acute (≤ 3 months) and chronic (> 3 months)
phases. Due to the chronicity of pain and other neurological symptoms, performance should be
studied for 1 year or longer post device implantation.
For applications based on clinical data from a comparable device, the manufacturer should
demonstrate that clinical data are derived from methodologically sound clinical studies and
describe any direct relationship that exists between the comparable device and the new device
with respect to the clinical data. Where the device and the predicate share any common design
origin, the lineage between the devices should be provided. Manufacturers are advised to
provide all relevant documents with a justification by a clinical expert to establish substantial
equivalence and to confirm that any identified differences between the device and the
nominated comparable device will not adversely affect the safety and performance of the device.
For further information on demonstrating substantial equivalence refer to Comparable devices
including substantially equivalent devices.

Literature review
The manufacturer should ensure that an internationally recognised method is followed when
conducting a systematic literature review. A literature review involves the systematic
identification, synthesis and analysis of all available published and unpublished literature,
favourable and unfavourable, on the device when used for its intended purpose as outlined in
the literature review section. The data can be generated from the use of the device or, if relying
on indirect evidence, the comparable device to which substantial equivalence has been
established. All included studies on the device and/or comparable device should have been
appraised for reporting quality and potential bias.
If the literature review is to include equivalent device/s, such devices should be identified
beforehand after substantial equivalence has been demonstrated. Clinical evidence provided in
the form of a literature review will be in support of safety and performance for the subject
device only if the reviewed studies relate to the device itself or device/s demonstrated to be
substantially equivalent. However, a literature review relating to a class of device, i.e. relating to
similar but not substantially equivalent devices, may provide supporting evidence of safety and
performance for the device type, to which the data for the subject device or substantially
equivalent device/s may be compared. For each study included in the literature review, the
device used must be clearly identified by manufacturer name and model, and studies relating to
the subject device or devices demonstrated to be substantially equivalent should be identified as
such and analysed separately to those for other devices.

Clinical evidence guidelines: Medical devices Page 115 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Post-market data
Post-market data should be provided where available for the device itself, as well as for the
comparable device. For implantable pulse generators, the regulatory status of the device should
include the MR designation in each jurisdiction where it is approved for use. It is particularly
important to include the following:
• distribution numbers of the device(s) by country and/or geographical region for every year
since launch. It is accepted that this may not always be appropriate for high volume devices,
those with many components or those on the market for many years
• safety data including medical device vigilance reports, adverse events, and complaints
categorised by type and clinical outcome for every year since launch should be reported,
including all deaths (all cause, cardiac and sudden cardiac death). Mortality data should
include clear definitions of patient death categories and overall mortality rate, and all patient
deaths should be supported by sufficient documentation.
• the number of years of use
• Examples of registry data for implantable pulse generator systems have been reported in
peer reviewed studies from Spain, Denmark, Sweden, France, Italy, China, Germany, Poland,
the United States, and Australia.
• Any explanted pulse generators returned to manufacturers should be accounted for with an
explanation of failures and corrective measures.
For reports of adverse events (AEs) and complaints etc., to be a useful adjunct to other forms of
clinical evidence, the manufacturer must make a positive, concerted effort to collect the reports
and to encourage users to report incidents. Experience shows that merely relying on
spontaneous reports leads to an underestimation of the incidence of complaints, vigilance and
adverse event reports.

Compiling the CER

In compiling the clinical evidence the manufacturer should ensure that an expert in the relevant
field critically evaluates all the clinical data from clinical investigation(s), literature review
and/or post-market data (clinical experience). The clinical expert should demonstrate
substantial equivalence for comparable devices where applicable and then endorse the CER
(evidenced by signature and date) that establishes whether the clinical evidence is sufficient to
demonstrate the requirements of the applicable EPs, in particular that the device is safe,
performs as intended, and has a favourable risk-benefit profile.
Previous sections outline the components that may comprise clinical evidence for a medical
device and the recommended process of compiling a CER. These guidance documents apply
whether the applicant is using direct clinical evidence or relying on indirect clinical evidence
from a comparable device. Guidance on defining a comparable device is provided in Comparable
devices including substantially equivalent devices.
As per The Clinical Evaluation Report the CER should include the following:
a) General details
b) Description of the medical device and its intended application
c) Intended therapeutic and/or diagnostic indications and claims
d) Context of the evaluation and choice of clinical data types
e) Summary of relevant pre-clinical data

Clinical evidence guidelines: Medical devices Page 116 of 178

V3.1 June 2022
 Therapeutic Goods Administration

f) Discussion regarding comparable devices including substantially equivalent devices

g) Summary of the clinical data and appraisal
h) Data analysis
i) Conclusions
j) Name, signature and curriculum vitae of clinical expert and date of report

Supportive data and information

The following information on the device must also be provided:
• risk assessment and management document
• IFU, labelling, product manual and all other documents supplied with the device. These must
highlight the risks and ensure that these are appropriately communicated to user.
Additional information should be provided as applicable. This may include (but is not limited
to):
• additional specifications of the device(s)
• the materials from which the device is made including chemical composition
• the components to which the device is paired when used clinically
• the technical characteristics of the leads and electrodes
• other devices that may be used in conjunction with the device
• any aspects of non-clinical testing results that inform the design of the clinical trial
• biocompatibility testing, bench testing and animal studies where applicable
• specific testing of any adjuvant medicinal components may be required especially if these
are new chemical entities in the Australian context. This should cover interactions between
the device and the medicine, pharmacodynamics and time-release profiles.

Defining clinical success

General
Safety and performance data should be provided for the peri-operative, acute (≤ 3 months post-
implant) and chronic phases (> 3 months post-implant). Ideally, patients should be assessed
with planned yearly follow-up visits. Given the long-term in vivo life of these implantable
devices and the potential permanent implantation of some components e.g. leads, manufacturers
are advised that long-term follow-up is required. According to peer reviewed literature, typical
follow-up periods are three or more years.
Manufacturers are advised to consult ISO 14708 “Implants for surgery – Active implantable
medical devices”, part 2 (pacemakers), part 3 (neurostimulators) and part 6 (ICDs). These ISO
standards detail requirements that must be met to provide basic assurance of safety for both
patients and users, by ensuring protection from:
• unintended biological effects
• external energy sources for example: electric currents, electrostatic discharge
• external cardiac defibrillators

Clinical evidence guidelines: Medical devices Page 117 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• temperature and pressure

• electromagnetic fields including MR environment
• ionising radiation
Novel features or pacing modes not previously evaluated in comparable devices should be
allocated more extensive study and assessment in the submitted clinical evidence to
demonstrate safety and performance.
Irrespective of their placement, implantable pulse generators can be affected by electromagnetic
interference (EMI). The risks of altered device function on exposure to electromagnetic fields
that are produced either intentionally or as by-products of use of other devices should be
assessed. Typical EMI sources include cardioversion, RF ablation, electrosurgery, radiotherapy,
use of TENS devices, metal detectors, wireless services (including cellular phones) and MRI
environments. Manufacturers are advised to refer to MRI considerations and the current version
of ISO 14117 (electromagnetic compatibility test protocols for active implantable medical
devices) in conjunction with this section.
The American Society of Anaesthesiologists, in collaboration with American Heart Association
and the Society of Thoracic Surgeons, have provided a consensus statement on postoperative
evaluation of AIMDs following procedures that expose patient to EMI (excluding MRI) and
appropriate recommendations should be included in the IFU.

· Manufacturers should define the electromagnetic fields and the duration

of exposure to such fields within which the device performs as intended
i.e. the tolerance to electromagnetic field exposure.

· This information is necessary to inform the content of IFU and manuals

provided with the device.

Active implantable cardiac devices

Safety
Systematic reviews on single, dual-chamber and CRT pacemaker systems either with or without
defibrillation capability and ICD systems included the following peri-procedure events and
longer term outcomes that were tracked as safety measures:
• procedural complications e.g. pneumothorax, haemothorax, pocket haematoma and infection
• device pocket erosion
• coronary sinus dissection or perforation, damage to arteries and nerves, air embolism,
venous thrombosis, cardiac perforation
• pericardial effusion
• device migration
• toxic or allergic reaction, e.g. nickel allergy, silicone allergy
• CRT-D and ICDs; arrhythmia and inappropriate shocks
– A Health Canada guidance report (Health Canada. Guidance Document: Medical Device
Applications for Implantable Cardiac Leads. File No. 11-113340-236; 2011) also lists
changes to defibrillation thresholds and lead impedances
• device-related problems

Clinical evidence guidelines: Medical devices Page 118 of 178

V3.1 June 2022
 Therapeutic Goods Administration

– leads: dislodgement, reposition, difficult placement, malfunction or fracture

– sensing problems (loss, oversensing or undersensing)
– loss of capture
• extracardiac stimulation
• CRT and CRT-D: progression to pacemaker syndrome, atrial fibrillation, heart failure or
stroke
• hazards related to use in the MRI environment (refer to Section10: Demonstrating the safety
of Implantable Medical Devices (IMDs) in the Magnetic Resonance (MR) environment)
• death

Performance
In guidance documents on pacemakers and their associated leads issued by Health Canada and
US FDA, and systematic reviews (SRs) related to CRT-D and ICD evidence, the key performance
outcomes were listed as:
• implantation success
• sensing characteristics
• battery longevity
• QoL measures using a validated tool e.g. the New York Heart Association Classification or SF-
36 scores
• reduced mortality (all cause, cardiac and sudden cardiac deaths)
– mortality data should include clear definitions of patient death categories and overall
mortality rate, and all patient deaths should be supported by sufficient documentation
• avoidance of rehospitalisation (for any reason) after device placement, including heart
transplant
• for CRT and CRT-D devices the pacing impedances (low [< 200 ohms] or high [> 3000 ohms]
measured using a recognised standard method [ISO 14708-2]) are within the ranges
specified by manufacturer
• voltage stimulation threshold (CRT, CRT-D)
• improved cardiac function (CRT, CRT-D) e.g. left ventricle ejection fraction (LVEF), reduced
incidences of atrial fibrillation (AF), stroke, heart failure
• improvement in clinical symptoms

Implantable electrical nerve stimulation devices

Implantable electrical nerve stimulators (including such devices as deep brain and vagal nerve
stimulators) are a treatment modality for patients who suffer chronic pain e.g. neuropathic,
nociceptive and non-cancerous pain and other disabling neurological symptoms.
The different aetiologies of pain and other neurological symptoms can impact on the
performance of neurostimulators. Therefore manufacturers are advised to clearly define the
target symptom and stimulation loci to assist clinical assessors to evaluate the safety and
performance of implantable neurostimulators for pain or the management of other neurological
symptoms. Devices can be categorised as either intracranial (e.g. deep brain stimulation) or
extracranial (e.g. spinal cord, vagal nerve or peripheral nerve stimulators).

Clinical evidence guidelines: Medical devices Page 119 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Safety: intracranial neurostimulators

Adverse events are variously reported and include:
• usual risks associated with major surgery
• infection
• intracerebral or extra-axial haematomas
• seizure (intraoperative or trial stimulation period)
• seizure long-term
• neurological deficit (short-term < 1 mo)
• neurological deficit long-lasting
• local pain/headache
• hardware maintenance e.g. shortened battery life, failed leads
• MRI environment safety concerns including heating (which has been reported to have
caused permanent neurological impairment and is of greatest concern for various
neurostimulator devices)

Safety: extracranial neurostimulators

Adverse events are variously reported and include:
• device-related complications e.g. electrode migration, lead fracture
• distorted or loss of sensation (paraesthesia or numbness)
• dural puncture (spinal cord stimulators)/CSF leak
• infection
• discomfort or pain
• undesired stimulation
• hardware maintenance e.g. shortened battery life, failed leads
• MRI environment safety concerns - including heating (which has been reported to create the
greatest concern for various neurostimulator devices)

Performance: intracranial and extracranial neurostimulators

The evidence reviewed reported on various outcomes including:
• pain (pain reduction, pain intensity scores, pain coping ability, reduction or cessation in use
of pain medication, pressure pain threshold, time to first reduction in pain, and maximum
reduction in pain) as well as anxiety score
– measured using validated scales e.g. visual analogue scales (VAS) or numerical rating
scales
– reported success criterion e.g. more than 50% of patients achieve a greater than 50%
reduction in VAS of pain intensity on follow-up, usually at 6 to 24 months

Clinical evidence guidelines: Medical devices Page 120 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• symptom reduction or improvement for non-analgesic neurostimulator indications (e.g.

movement disorders such as Parkinsonian tremor, essential tremor, dystonia; urinary or
faecal incontinence; epilepsy)
• patient function e.g. QoL, mood, sleep and function scores should be assessed using validated
tools such as:
– Oswestry Disability Index and the Low Back Pain Outcome Scale
– SF-36
– Zung Self-Rating Depression Scale
• return to work
• hospital attendance
• patient satisfaction and experience

Manufacturers are advised that ranges for stimulation parameters of

frequency (Hz), Amplitude (V) and pulse-width (ms) should be provided and
included in IFU documentation.

Clinical evidence guidelines: Medical devices Page 121 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Summary of safety and performance data

Studies from the peer reviewed literature

Table 12: Study characteristics extracted from SRs on the safety and performance of
selected implantable pulse generators

Characteristics Pacemakers ICDs Pain management

of included (including CRT) (5 SRs) devices
studies (2 SR) (5 SRs or narrative
reviews)

Number of Dominant design RCT 4 SRs / MAs only Mixed evidence base
included studies total included studies included RCTs: range 3 with the number of
per SR n = 45 to 8; 1 SR only included included studies
cohort studies: n=18 ranging from 11 to 62

Clinical Dual-chamber versus (a) Primary prevention (a) Complex regional

situation(s) single chamber of SCD in patients w/ pain syndrome (b)
pacemakers for CKD at risk of life- neuropathic or
bradycardia due to threatening ventricular ischaemic (c) low-back
atrioventricular arrhythmias; (b) disorders (d)
block or sick sinus patients w/ HF; (c) nociceptive or
syndrome patients w/ ARVD/C; neuropathic pain (e)
(d) primary prevention headaches
of SCD in older patients

Dominant design 1 SR including 4 RCTs 4 SRs included only Case series and RCT
of included of parallel group RCTs; 1 SR included
studies design and 28 only observational
randomised studies
crossover
comparisons

Sample size RCTs: 58 to 2568 Total N in SRs ranged Total N in the SR

(range) for Crossover studies: 8 from 610 to 5674 ranged from 210 to 509
included studies to 48

Reported Dual-chamber versus (a) Usual medical Medical and/or surgical

comparisons single chamber therapy, placebo or treatment (appropriate
ventricular pacing amiodarone; (b) CRT-D to condition) that does
(ICD + CRT); (c) not include SCS.
“appropriate control”
(not specified but could
not include ICD or CRT-
D)

Patient follow- RCTs: 1.5 to 5 years Means of 3 months to Ranged from 1 month
up Crossover studies: 48 3.8 years to 7.2 years
hours to 8 weeks
KEY: ARVD/C= arrhythmogenic right ventricular dysplasia / cardiomyopathy; CKD=chronic kidney
disease; CRT=cardiac resynchronisation therapy; CRT-D=cardiac resynchronisation therapy plus ICD;
HF=heart failure; ICD=implantable cardiac defibrillator; MA=meta-analysis; RCT=randomised controlled
trial; SCD=sudden cardiac death; SR=systematic review; w/=with

Clinical evidence guidelines: Medical devices Page 122 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Table 13: Reported clinical outcomes in the peer reviewed literature on selected
implantable pulse generators

Type of pulse Outcomes reported in the included research or resources

generator

Pacemakers Safety: implantation success, lead fracture, lead dislodgement, conductor

(including CRT) failure, extracardiac stimulation, insulation failure, loss of capture,
sensing problems (loss, oversensing or undersensing), perforation and
(2 SR) other lead-related AEs, including death
• Voltage stimulation thresholds
• Sensing characteristics
• Pacing impedances (Low or high)
• Battery longevity

ICDs Safety (AEs / postop complications): pneumothorax, haemothorax,

pocket haematoma, lead dislodgement or reposition or difficult
(5 SRs) placement or malfunction or fracture, ICD migration, impending ICD
pocket erosion, infection, ICD-related infection, pericardial effusion or
tamponade, coronary sinus dissection or perforation, damage to arteries
and nerves, air embolism, venous thrombosis, cardiac perforation,
arrhythmia, inappropriate shocks
• Mortality (all-cause and ICD-related)
• Rehospitalisation (for any reason) after ICD placement including
heart transplant
• Improvement in clinical conditions
• QoL
• From Health Canada: defibrillation thresholds and lead impedances
(since the device is designed for cardioversion or defibrillation)

Pain • Safety intracranial (AEs / postop complications): l risks associated

management with major surgery, infection, intracerebral or extra-axial
haematomas, subdural or epidural haemorrhage, seizure
(5 SRs) (intraoperative or trial stimulation period), seizure long-term,
neurological deficit (short-term < 1 mo), neurological deficit long-
lasting, local pain/headache, hardware maintenance e.g. shorten
battery life, failed leads, MR environment safety concerns e.g.
heating leading to neurological damage
• Safety extracranial (AEs / postop complications): device-related
complications e.g. electrode migration, lead fracture, loss of
paraesthesia, dural puncture (spinal cord stimulators), infection,
hardware maintenance e.g. shortened battery life, failed leads, MR
environment safety concerns
• Pain (pain reduction, pain intensity scores, pain coping, pressure
pain threshold, time to first reduction in pain, and maximum
reduction in pain) as well as anxiety score

Clinical evidence guidelines: Medical devices Page 123 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Type of pulse Outcomes reported in the included research or resources

generator

• Patient function e.g. QoL, mood, sleep and site specific function
scores should be assessed using validated tools such as:
– return to work
– patient satisfaction and experience
– analgesic consumption
– hospital attendance

KEY: AE=adverse events; FVC=forced vital capacity; ICD=implantable cardiac defibrillator; ROM = range of
motion; QOL=quality of life; SR=systematic review

References
Coffey RJ, Lozano AM. Neurostimulation for chronic noncancer pain: An evaluation of the clinical
evidence and recommendations for future trial designs. J Neurosurg. 2006;105(2):175-89
US Food & Drug Administration (FDA). Guidance for the submission of research and marketing
applications for permanent pacemaker leads and for pacemaker lead adaptor 510(k)
submissions; 2000
Coma Samartin R, Cano Perez O, Pombo Jimenez M. Spanish Pacemaker Registry. Eleventh
official report of the Spanish Society of Cardiology Working Group on Cardiac Pacing (2013). Rev
Esp Cardiol. 2014;67(12):1024-38.
Moller M, Arnsbo P. [The Danish Pacemaker Registry. A database for quality assurance]. Ugeskr
Laeger. 1996;158(23):3311-5.
Gadler F, Valzania C, Linde C. Current use of implantable electrical devices in Sweden: data from
the Swedish pacemaker and implantable cardioverter-defibrillator registry. Europace.
2015;17(1):69-77.
Mouillet G, Lellouche N, Yamamoto M et al. Outcomes following pacemaker implantation after
transcatheter aortic valve implantation with CoreValve devices: Results from the FRANCE 2
Registry. Catheter Cardiovasc Interv. 2015.
Benkemoun H, Sacrez J, Lagrange P et al. Optimizing pacemaker longevity with pacing mode and
settings programming: Results from a pacemaker multicenter registry. Pacing Clin
Electrophysiol. 2012;35(4):403-8.
Proclemer A, Zecchin M, D'Onofrio A et al. [The pacemaker and implantable cardioverter-
defibrillator registry of the Italian Association Arrhythmology Cardiac Pacing and cardiac pacing
- annual report 2013]. G Ital Cardiol (Rome). 2014;15(11):638-50.
Chen KP, Dai Y, Hua W et al. Reduction of atrial fibrillation in remotely monitored pacemaker
patients: results from a Chinese multicentre registry. Chin Med J. 2013;126(22):4216-21.
Markewitz A. [Annual Report 2009 of the German Cardiac Pacemaker Registry: Federal Section
pacemaker and AQUA - Institute for Applied Quality Improvement and Research in Health Ltd].
Herzschrittmacherther Elektrophysiol. 2011;22(4):259-80.
Przybylski A, Derejko P, Kwasniewski W et al. Bleeding complications after pacemaker or
cardioverter-defibrillator implantation in patients receiving dual antiplatelet therapy: Results of
a prospective, two-centre registry. Neth Heart J. 2010;18(5):230-5.

Clinical evidence guidelines: Medical devices Page 124 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Poole JE, Gleva MJ, Mela T et al. Complication rates associated with pacemaker or implantable
cardioverter-defibrillator generator replacements and upgrade procedures: results from the
REPLACE registry. Circulation. 2010;122(16):1553-61.
Reid CM, Brennan AL, Dinh DT et al. Measuring safety and quality to improve clinical outcomes--
current activities and future directions for the Australian Cardiac Procedures Registry. Med J
Aust. 2010;193(8 Suppl):S107-10.
Nielsen JC, Thomsen PE, Hojberg S et al. A comparison of single-lead atrial pacing with dual-
chamber pacing in sick sinus syndrome. Eur Heart J. 2011;32(6):686-96
ISO Standard 14117:2019 - Active implantable medical devices -- Electromagnetic compatibility
-- EMC test protocols for implantable cardiac pacemakers, implantable cardioverter
defibrillators and cardiac resynchronization devices
Crossley GH, Poole JE, Rozner MA et al. The Heart Rhythm Society (HRS)/American Society of
Anesthesiologists (ASA) Expert Consensus Statement on the perioperative management of
patients with implantable defibrillators, pacemakers and arrhythmia monitors: Facilities and
patient management: Executive summary this document was developed as a joint project with
the American Society of Anesthesiologists (ASA), and in collaboration with the American Heart
Association (AHA), and the Society of Thoracic Surgeons (STS). Heart Rhythm. 2011;8(7):e1-18.
Castelnuovo E, Stein K, Pitt M, Garside R, Payne E. The effectiveness and cost-effectiveness of
dual-chamber pacemakers compared with single-chamber pacemakers for bradycardia due to
atrioventricular block or sick sinus syndrome: Systematic review and economic evaluation.
Health Technol Assess. 2005;9(43):iii, xi-xiii, 1-246.
Kong MH, Al-Khatib SM, Sanders GD, Hasselblad V, Peterson ED. Use of implantable cardioverter-
defibrillators for primary prevention in older patients: A systematic literature review and meta-
analysis. Cardiol J. 2011;18(5):503-14
Colquitt JL, Mendes D, Clegg AJ et al. Implantable cardioverter defibrillators for the treatment of
arrhythmias and cardiac resynchronisation therapy for the treatment of heart failure:
Systematic review and economic evaluation. Health Technol Assess. 2014;18(56):1-560
Chen S, Ling Z, Kiuchi MG, Yin Y, Krucoff MW. The efficacy and safety of cardiac
resynchronization therapy combined with implantable cardioverter defibrillator for heart
failure: A meta-analysis of 5674 patients. Europace. 2013;15(7):992-1001
Chen S, Yin Y, Krucoff MW. Effect of cardiac resynchronization therapy and implantable
cardioverter defibrillator on quality of life in patients with heart failure: A meta-analysis.
Europace. 2012;14(11):1602-7
Pun PH, Al-Khatib SM, Han JY et al. Implantable cardioverter-defibrillators for primary
prevention of sudden cardiac death in CKD: A meta-analysis of patient-level data from 3
randomized trials. American Journal of Kidney Disease. 2014;64(1):32-9
Schinkel AF. Implantable cardioverter defibrillators in arrhythmogenic right ventricular
dysplasia/cardiomyopathy: patient outcomes, incidence of appropriate and inappropriate
interventions, and complications. Circulation: Arrhythmia and Electrophysiology.
2013;6(3):562-8
Health Canada. Guidance Document: Medical Device Applications for Implantable Cardiac Leads.
File No. 11-113340-236; 2011
American Heart Association – Classes of Heart Failure – New York Heart Association Functional
Classification system

Clinical evidence guidelines: Medical devices Page 125 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Plow EB, Pascual-Leone A, Machado A. Brain stimulation in the treatment of chronic neuropathic
and non-cancerous pain. The Journal of Pain. 2012;13(5):411-24
Taylor RS, Van Buyten JP, Buchser E. Spinal cord stimulation for complex regional pain
syndrome: A systematic review of the clinical and cost-effectiveness literature and assessment of
prognostic factors. European Journal of Pain. 2006;10(2):91-101
Fontaine D, Hamani C, Lozano A. Efficacy and safety of motor cortex stimulation for chronic
neuropathic pain: Critical review of the literature. J Neurosurg. 2009;110(2):251-6
Simpson EL, Duenas A, Holmes MW, Papaioannou D, Chilcott J. Spinal cord stimulation for
chronic pain of neuropathic or ischaemic origin: Systematic review and economic evaluation.
Health Technol Assess. 2009;13(17):iii, ix-x, 1-154
Looi KL, Gajendragadkar PR, Khan FZ et al. Cardiac resynchronisation therapy: pacemaker
versus internal cardioverter-defibrillator in patients with impaired left ventricular function.
Heart. 2014;100(10):794-9.

Heart valve replacements using a prosthetic valve

Heart valve replacement using a prosthetic valve is performed to reduce the morbidity and
mortality associated with native valvular disease or to replace a malfunctioning prosthetic valve.

Summary recommendations
• Prosthetic heart valves are complex medical devices which are currently made of either
synthetic material (mechanical valves) or biological tissues (bioprosthesis) or a combination
of both and inserted via open surgery or percutaneously. Manufacturers are advised to
provide clinical evidence to support the safety and performance of the particular device and
any accessories used to deliver the device.
• Provision of clinical investigation data:
– manufacturers who intend to conduct clinical trials should design trials to the highest
practical NHMRC level of evidence and trials should be appropriate to inform on the
safety and performance of the device for its intended purpose
– to comply with ISO 5840, clinical trials should continue until the minimum number of
patients with each valve type have each been followed for a minimum of one year and
there are at least 400 valve years of follow-up of each valve type. For modification of an
existing valves already on the ARTG the patient years deemed acceptable may in some
circumstances be adjusted based on a risk analysis of the changes
– for evaluating the performance of prosthetic heart valves it is recommended that the
Objective Performance Criteria (OPC) as listed in ISO 5840 (and updates) be reported
including early (within 30 days post implantation), mid- term outcomes (after 30 days
post implantation) and at one year (or two years for reimbursement). The selection
should be supported by a clinical justification
– typical safety and performance values are provided in Table 15, Table 16, Table 17,
Table 18 and Table 19 and Table 20.
• Pre-clinical data demonstrating the mechanical and physical characteristics should be
consistent with the intended purpose and anticipated in vivo lifespan of the heart valve
replacement.
• Documentation demonstrating biocompatibility of the device should be provided.

Clinical evidence guidelines: Medical devices Page 126 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• For submissions reliant on comparable device data, manufacturers are required to submit all
relevant documents with a supporting clinical justification by the clinical expert that
establishes substantial equivalence between the device and the nominated comparable
device.
• When submitting a comprehensive literature review full details of the method, search
strategy, inclusion/exclusion criteria for selection of studies and analysis should be included
in the CER with sufficient detail to ensure the search can be reproduced.
• In addition, a well-documented risk analysis and management system must be provided with
the CER. The clinical investigation data, literature review and post-market clinical
experience should inform the risk assessment documentation. All clinical risks identified in
the clinical data should be reflected in the risk assessment documentation. These risks
should be appropriately rated and quantified and ideally be presented as risk matrices,
before assigning risk reduction activities such as statements in the IFU and training
materials to reduce residual risks. The residual risk following risk mitigation
implementation should be estimated.
• Manufacturers should provide details of the clinical context within which the clinical data
was obtained. The clinical context of the evidence should be consistent with the indications
for use.
• Compilation of the clinical evidence
– in compiling the clinical evidence for a prosthetic heart valve the manufacturer should
ensure that a competent clinical expert critically evaluates all the clinical data that
informs on the safety and performance of the device
– the competent clinical expert must then endorse the CER (evidenced by signature and
date) which demonstrates that the clinical evidence is sufficient to meet the
requirements of the applicable EPs and the device is deemed to be safe and to perform
as intended
• The full CV of the clinical expert should be provided

Defining heart valve prostheses

This section includes both conventional heart valves (those that are implanted using open heart
surgery) and percutaneous heart valves (those that are collapsed into a catheter and are
expanded at the time of implantation). The guidance also applies to ‘sutureless’ (meaning heart
valves with fewer sutures, not without sutures) valve technology whereby the valve is mounted
on a self-expanding nitinol frame that is implanted into the aortic annulus following resection of
the diseased tissue. Each type of valve has its own associated risk benefit profile that needs to be
addressed by the manufacturer.
Currently there are three main types of prosthetic heart valves, mechanical, biological and valves
that combine mechanical and biological components (using hybrid valve technology).
The main designs of mechanical (synthetic) valves include:
• the caged ball valve
• the tilting disc (single leaflet) valve
• the bileaflet valve.
Biological valves (bioprosthesis or tissue valves) are classified into two major categories:
• xenografts made from bovine, porcine, or equine tissue

Clinical evidence guidelines: Medical devices Page 127 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• homografts obtained from cadaveric donors.

Xenografts may have a supporting frame (stent) or no supporting frame (stentless).
Manufacturers and applicants are advised to read this guidance section in conjunction with
other relevant sections and ISO documentation, ISO 5840:2021 and ISO 5840-3:2021.

Clinical evidence
The clinical evidence can be derived from clinical investigation(s) data, a comprehensive
literature review and/or post-market data (clinical experience) from the use of the device
(direct) and/or comparable device (indirect). Direct clinical evidence on the actual device is
preferred. It is important to clarify if any changes have been made to the device since the clinical
data were gathered and if so to document the changes and to clarify the exact version of the
device. Otherwise indirect clinical evidence from a comparable device may be used after
substantial equivalence has been demonstrated through a comparison of the clinical, (intended
purpose) technical and biological characteristics as described in Comparable devices including
substantially equivalent devices. Where the device and the predicate share any common design
origin, the lineage between the devices should be provided as well.
The intended purpose, clinical indications, claims and contraindications must be supported by
the clinical data and documented in the IFU and other information supplied with the device.
Manufacturers should refer to Clinical evidence requirements for more information.

Clinical investigation(s)
The design of the clinical investigation(s) should be appropriate to generate valid unbiased
measures of clinical performance and safety. If clinical studies on cardiac valve prostheses are
conducted it is recommended that manufacturers refer to ISO 5840-1:2021; ISO 5840-2:2021
and ISO 5840-3:2021 as guides to study design.
Additional resources regarding clinical study design and conduct are available on the TGA and
FDA websites. The preferred design is a randomised controlled clinical trial and conditions
should ideally represent clinical practice in Australia. The eligible patient groups should be
clearly defined with exclusion/inclusion criteria.
It is recommended that the clinical study continue until the minimum number of patients of each
valve type has each been followed for a minimum of one year (two years if seeking
reimbursement). There must be at least 400 valve years of follow-up of each valve type. This is
based on guidance in ISO 5840:2021. For modification of an existing valve on the ARTG the
patient years deemed acceptable may in some circumstances be adjusted based on a risk
analysis of the changes. The manufacturer is responsible for providing justification of the study
protocol. The number of patient years should also be documented.
Medication which may affect outcomes, for example anticoagulant treatment, must be taken into
account when determining all endpoints. Analysis of clinical events should be blinded and
independently adjudicated wherever possible.

Literature review
A literature review involves the systematic identification, synthesis and analysis of all available
published and unpublished literature, favourable and unfavourable, on the device when used for
its intended purpose or, if relying on indirect evidence, the comparable device to which
substantial equivalence has been established.
Data on the materials used to construct the prosthesis, its dimensions and geometry and the
intended purpose and population will define the construction of search strategies as well as
study selection when conducting a comprehensive literature review. This ensures that the

Clinical evidence guidelines: Medical devices Page 128 of 178

V3.1 June 2022
 Therapeutic Goods Administration

searches are complete and the included studies are related to the device and/or comparable
device. The search strategy should be made prior to performing the literature review, extraction
of the clinical evidence and analysis of the pooled results. A full description of the device used or
adequate information to identify the device (e.g. manufacturer name and model number) in any
given study must be extractable from the study report. If this is not possible, the study should be
excluded from the review.

Post-market data
Post-market data can be provided for the actual device or for the comparable device. It is
particularly important to include the following:
• information about the regulatory status of the device(s) (or comparable device(s) if relying
on this), including the certificate number, date of issue and name under which the device is
marketed, the exact wording of the intended purpose/approved indication(s), any
conditions and other information which may be relevant such as MRI designation in other
jurisdictions.
• any regulatory action including CE mark withdrawals, recalls, including recalls for product
correction (and the reason for these i.e. IFU changes), removals, suspensions and
cancellations and any other corrective actions anywhere in the world
• distribution numbers of the device(s) including distribution by country and/or geographical
region for every year since launch. It is accepted that this may not always be appropriate for
high volume devices, those with many components or those on the market for many years
• the number of years of use
• for every year since launch, the number of complaints, vigilance and monitoring reports and
adverse events categorised by type and clinical outcome
• explanted devices returned to manufacturers should be accounted for with an explanation of
device failures and corrective measures.
Publicly available post-market data such as adverse event reporting on the FDA MAUDE
database and the TGA IRIS may be used for devices from other manufacturers. The manufacturer
should include post-market surveillance data from national jurisdictions where the device is
approved for clinical use. Registries for different prosthetic heart valves have been established
in Belgium, France, Germany, Italy, New Zealand and the United Kingdom as well as Australia.
For reports of adverse events and device failures to be useful clinical evidence, the manufacturer
must make a positive, concerted effort to collect the reports and to encourage users to report
incidents. Experience shows that merely relying on spontaneous reports leads to an
underestimation of the incidence of failures and adverse events.
The post-market data should be critically evaluated by a competent clinical expert to enable an
understanding of the safety and performance profile of the device(s) in a ‘real-world’ setting.

Compiling the CER

Previous sections outline the components that may comprise clinical evidence for a medical
device and the recommended process of compiling a CER. This guidance applies whether the
applicant is using direct clinical evidence or relying on indirect clinical evidence from a
comparable device. As time since first approval lengthens comparable device data becomes less
relevant and should be replaced by data derived from clinical experience with the device.

Clinical evidence guidelines: Medical devices Page 129 of 178

V3.1 June 2022
 Therapeutic Goods Administration

As per The Clinical Evaluation Report the CER should include the following:
a) General details
b) Description of the medical device and its intended application
c) Intended therapeutic and/or diagnostic indications and claims
d) Context of the evaluation and choice of clinical data types
e) Summary of relevant pre-clinical data
f) Discussion regarding comparable devices including substantially equivalent devices
g) Summary of the clinical data and appraisal
h) Data analysis
i) Conclusions
j) Name, signature and curriculum vitae of clinical expert and date of report

Supportive data and information

The following information on the device must also be provided:
• risk assessment and management document
• IFU, product manual and all other documents supplied with the device. The clinical evidence
must highlight the risks and ensure that these are appropriately communicated to user.
Additional information should be provided as applicable. This may include (but is not limited
to):
• additional specifications of the device(s)
• the materials from which the device is made including chemical composition
• other devices that may be used in conjunction with the device
• any aspects of non-clinical testing results that inform the design of the clinical trial should be
included in the supporting documents
• biocompatibility testing, bench testing and animal studies where applicable
• specific testing of any adjuvant medicinal components may be required especially if these
are new chemical entities in the Australian context. This should cover interactions between
the device and the medicine, pharmacodynamics and time-release profiles.
Current heart valve prostheses vary in their composition, method of insertion and way in which
they are fixed.
In submissions to the TGA, it is recommended that manufacturers of heart valve prostheses refer
to ISO documents for guidance on the type of information that should be provided with respect
to the characteristics of the device, for example 5840-1: 2015, Cardiovascular implants --
Cardiac valve prostheses -- Part 1: General requirements; 5840-2:2015 Cardiovascular implants
-- Cardiac valve prostheses -- Part 2: Surgically implanted heart valve substitutes and 5840-
3:2013 Cardiovascular implants -- Cardiac valve prostheses -- Part 3: Heart valve substitutes
implanted by transcatheter techniques.

Clinical evidence guidelines: Medical devices Page 130 of 178

V3.1 June 2022
 Therapeutic Goods Administration

For mechanical heart valve prostheses these include, but are not limited to:
• the materials used in the valve
• the design of the valve
• the size of the valve
• assembly technique
• testing and quality control procedures
• haemodynamic properties
• packaging and sterilisation procedures.
For biological heart valve prostheses these include, but are not limited to:
• the material used in the valve
• the design of the valve
• the size of the valve
• assembly technique
• testing and quality control procedures
• haemodynamic properties
• tissue preservation and/or cross-linking technique(s)
• anticalcification treatment(s)
• packaging and sterilisation procedures.
All device characteristics and the intended purpose(s) are essential prerequisites for the design
of clinical studies to demonstrate the clinical safety and performance of devices with no
equivalent comparable device(s).
If a comparable device is available and data from that device is used to support a submission, the
device characteristics and intended purpose will determine the criteria for a full clinical
justification for the selection of the comparable device. The following should be included when
relying on a comparable device for heart valve prostheses:
• A comparison of the technical and physical characteristics of the new and comparable
device(s) should be demonstrated through direct testing in order to establish substantial
equivalence
– direct comparisons of the technical and physical characteristics include, but are not
limited to; the composition of the prostheses, hydrodynamic performance,
biocompatibility, accessories such as implantation tools, corrosion resistance, shelf life,
fatigability, durability, dimensions, geometry and weight. Refer to ANNEX D and I in ISO
5840:2021 for a more comprehensive list
– any differences in the technical and physical characteristics should be addressed in the
clinical justification to determine whether the difference will affect the benefit-risk
profile when the device is used for its intended purpose
– the use of more than comparable device is discouraged; however, these may be used if
each comparable device is found to be substantially equivalent to the device under
consideration

Clinical evidence guidelines: Medical devices Page 131 of 178

V3.1 June 2022
 Therapeutic Goods Administration

– a clinical justification should be presented when using a comparable device as to why

direct clinical data are either not required, or are only partially required
• The comparable device(s) must have clinical data to support safety and performance.
• The clinical expert should critically evaluate all the clinical data for the device and
comparable device and then endorse the CER (evidenced by signature and date) that
establishes whether the clinical evidence is sufficient to demonstrate the requirements of
the applicable EPs and that the device is safe and performs as intended.

Defining clinical success

The studies identified for these guidelines identified appropriate clinical outcomes to establish
the safety and performance of prosthetic heart valves however outcomes were sometimes
classified differently. For example, mortality and stroke were referred to as safety outcomes in
some studies and performance outcomes in others, or included under both headings. For this
reason outcomes are reported together here, separated into early and late outcomes post
treatment.
It is recommended that early outcomes are reported at 30 days post treatment and include the
following:
• all-cause mortality
• valve related mortality
• thromboembolism
• valve thrombosis
• all cause reoperation
• explant
• all stroke (disabling and non-disabling)
• life threatening bleeding (note: bleeding should be classified as either ‘all haemorrhage’ or
‘major haemorrhage’)
• acute kidney injury (stage 2 or 3, including need for haemodialysis)
• peri-procedural myocardial infarction
• endocarditis
• major vascular complication
• coronary obstruction requiring intervention
• valve-related dysfunction (note: valve regurgitation should be reported as ‘all paravalvular
leaks’ and ‘major paravalvular leaks’)
In addition, it is recommended the following outcomes be reported after 30 days:
• all-cause mortality
• all stroke (disabling and non-disabling)
• hospitalisation for valve-related symptoms or worsening congestive heart failure
• a quality of life measure e.g. the New York Heart Association Classification (NYHA) or the
Minnesota Living with Heart Failure Questionnaire (MLHF)

Clinical evidence guidelines: Medical devices Page 132 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• prosthetic valve endocarditis

• prosthetic valve thrombosis
• bleeding, unless unrelated to valve therapy (e.g. trauma) (note: bleeding should be classified
as either ‘all haemorrhage’ or ‘major haemorrhage’ ‘anticoagulant-related haemorrhage’
• reoperation
• thromboembolic events (e.g. stroke)
• structural valve deterioration
• non-structural valve dysfunction/valve related dysfunction (note: valve regurgitation should
be reported as ‘all paravalvular leaks’ and ‘major paravalvular leaks’ and it should be noted if
the dysfunction required a repeat procedure)
At one year the following should be reported:
• Structural valve deterioration
• Thromboembolism
• Major, reversible ischemic neurological deficit (RIND)
• Valve thrombosis
• Anticoagulant-related haemorrhage
• Prosthetic valve endocarditis
• Non-structural valve dysfunction/paravalvular leak
• Re-operation
It is recommended that the following outcomes; valve related dysfunction, prosthetic valve
endocarditis, prosthetic valve thrombosis, thromboembolic events and bleeding, be reported in
a time-related manner as described in Guidelines for reporting mortality and morbidity after
cardiac valve interventions.
The outcomes listed above are a recommended minimum based on a consensus report produced
by the Valve Academic Research Consortium. For appropriate definitions, diagnostic criteria and
measurement of the above outcomes manufacturers should consult the following documents:
• the Valve Academic Research Consortium Consensus Documents on standardised endpoint
definitions for transcatheter aortic valve implantation
• guidelines by Akins et al (2008) for reporting mortality and morbidity after cardiac valve
interventions
• guidelines on the evaluation of prosthetic valves with echocardiography
• the update of objective performance criteria for clinical evaluation of new heart valve
prostheses by ISO (Wu et al 2014)
For valve function, including transcatheter and surgically implanted valves, indicative values on
what is considered a normal functioning valve and what is considered a dysfunctional valve are
reported in documents by VARC and guideline documents on the evaluation of prosthetic valves
with echocardiography (Table 15, Table 16, Table 17, Table 18 and Table 19).
For surgically implanted valves other than those implanted through the transcatheter technique,
specific objective performance criteria (OPC) for thromboembolism, valve thrombosis, all and

Clinical evidence guidelines: Medical devices Page 133 of 178

V3.1 June 2022
 Therapeutic Goods Administration

major haemorrhage, all and major paravalvular leaks and endocarditis have been determined by
ISO and reported in Wu et al (2014) (Table 20). A new valve should have complications rates
lower than twice the OPC. For transcatheter valves the number of events for each of the listed
outcomes should be similar to or less than those reported in studies published in peer reviewed
journals or heart valve registries for a similar type of prosthetic heart valve in the same valve
position. Values that are reported need to be supported by clinical justification.
Manufacturers should report early (within 30 days post implantation) and late valve outcomes
(after 30 days post implantation) with a follow-up of one year or more (two years if seeking
reimbursement) and a minimum of 400 valve years of follow-up for each valve type.
Outcomes are comprised of the most relevant patient endpoints as defined by the Valve
Academic Research Consortium (VARC).
For surgically implanted valves, manufacturers should refer to the objective performance
criteria determined by the ISO for what is considered an acceptable number of events for
different outcomes.
For transcatheter valves the number of events for each outcome should be similar to or less than
those reported in studies published in peer reviewed journals or heart valve registries for a
similar type of prosthetic heart valve in the same valve position.

Summary of safety and performance data

Reported clinical outcomes on prosthetic heart valves

Table 14: Summary of outcome data extracted from health technology assessments on
prosthetic heart valves

Safety parameter Surgical Aortic Transcatheter Sutureless valve

Valve Replacement Aortic Valve replacement
Implantation

Death (any cause) a a a

Death (cardiovascular cause) a a

Repeat hospitalisation a

Myocardial infarction a

Strokes a a

Transient ischemic attack a

kidney injury/need for

a a
haemodialysis

Vascular complications a

Bleeding/haemorrhage a a a

Endocarditis a a a

Clinical evidence guidelines: Medical devices Page 134 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Safety parameter Surgical Aortic Transcatheter Sutureless valve

Valve Replacement Aortic Valve replacement
Implantation

Atrial fibrillation a a

Tamponade/pericardial
a
effusion

Life threatening
arrhythmias/arrhythmias a
requiring intervention

Haemodynamic
collapse/need for a
haemodynamic support

New pacemaker a a

Device malfunction,
a a
misplacement or migration

Non-structural dysfunction a

Structural valvular
a
deterioration

Injury to valve or
a
myocardium

Valve-in-valve or second
a
valve required

Conversion to sutured valve a

Conversion to surgical valve

a
replacement

Thromboembolism a a

Valve thrombosis a

Reintervention/reoperation
a a
or freedom from reoperation

Aortic
regurgitation/paravalvular a a
regurgitation

Atrioventricular block a

Clinical evidence guidelines: Medical devices Page 135 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Safety parameter Surgical Aortic Transcatheter Sutureless valve

Valve Replacement Aortic Valve replacement
Implantation

Cross-clamp time a a

Bypass time a a

Left ventricular mass

a
regression index

Life expectancy based on

a
microsimulation

Event-free life expectancy

a
based on microsimulation

Successful implantation a

Length of stay in intensive

a
care

Length of hospital stay a

Haemodynamic parameters

Post-operative mean and

peak aortic pressure a a a
gradient

Effective orifice area index a a

Left ventricular ejection

a
fraction

Mean aortic valve area a a

Change in NYHA* class a a a

6-minute walk test a

*NYHA: New York Heart Association

Clinical evidence guidelines: Medical devices Page 136 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Table 15: Parameters used to assess transcatheter valve function and a guide to what are
considered normal values as defined by the Valve Academic Research Consortium

Parameter Prosthetic Aortic Valve Stenosis

Normal Mild Stenosis Moderate/Severe

Stenosis

Quantitative parameters (flow dependent)†

Peak velocity (m/s) <3m/s 3–4 m/s >4m/s

Mean gradient (mm/Hg) <20 mm Hg 20–40 mm Hg >40 mm Hg

Quantitative parameters (flow-independent)

Doppler velocity index‡ >0.35 0.35–0.25 <0.25

Effective orifice area§ >1.1 cm2 1.1–0.8 cm2 <0.8 cm2

Effective orifice area║ >0.9 cm2 0.9–0.6 cm2 <0.6 cm2

Parameter Prosthesis-Patient Mismatch

Insignificant Moderate Severe

Indexed effective orifice >0.85 cm2/m2 0.85–0.65 cm2/m2 <0.65 cm2/m2

area¶ (cm2/m2)

Indexed effective orifice >0.70 cm2/m2 0.90–0.60 cm2/m2 <0.60 cm2/m2

area# (cm2/m2)

Parameter Prosthetic Aortic Valve Regurgitation

Mild Moderate Severe

Semi-quantitative parameters

Diastolic flow reversal Absent or brief intermediate Prominent,

in the descending aorta- early diastolic holodiastolic
PW

Circumferential extent <10% 10–29% ≥30%

of prosthetic valve
paravalvular
regurgitation (%)**

Quantitative parameters‡

Clinical evidence guidelines: Medical devices Page 137 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Parameter Prosthetic Aortic Valve Regurgitation

Mild Moderate Severe

Regurgitant volume <30 mL 30–59 ml ≥60 ml

(mL/beat)

Regurgitant fraction (%) >30% 30–49% ≥50%

EROA (cm2) 0.10 cm2 0.10–0.29 cm2 ≥0.30 cm2

†These parameters are more affected by flow, including concomitant aortic regurgitation
‡For left ventricular outflow tract (LVOT) >2.5 cm, significant stenosis criteria is <0.20
§Use in setting of Body Surface Area (BSA) ≥1.6 m2 (note: dependent on the size of the valve and the size
of the native annulus).
║Use in setting of BSA <1.6 m2, ¶ Use in setting of BMI <30 kg/m2, # Use in setting of BMI ≥30 kg/m2
**not well-validated and may overestimate the severity compared with the quantitative Doppler
EROA: effective regurgitant orifice area; PW: pulsed wave
Table 16: Guide to normal values, intermediate values for which stenosis may be possible
and values that usually suggest obstruction in mechanical and stented-biological
prosthetic aortic valves* from Zoghbi et al (2009)

Parameter Normal Possible stenosis Suggests significant

stenosis

Peak velocity (m/s)† <3 3-4 >4

Mean gradient (mm Hg)† <20 20-35 >35

DVI ≥0.30 0.29-0.25 <0.25

EOA (cm2) >1.2 1.2-0.8 <0.8

Contour of the jet Triangular, early Triangular to Rounded,

velocity through the peaking intermediate symmetrical contour
PrAV

AT (ms) <80 80-100 >100

AT: acceleration time; DVI: Doppler velocity index; EOA: effective orifice area; PrAV: prosthetic aortic
valve;
*In conditions of normal or near normal stroke volume (50-70 mL) through the aortic valve
†These parameters are more affected by flow, including concomitant aortic regurgitation

Clinical evidence guidelines: Medical devices Page 138 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Table 17: Parameters for evaluation of the severity of prosthetic aortic valve
regurgitation from Zoghbi et al (2009)

Parameter Mild Moderate Severe

Valve structure and motion

Mechanical or bioprosthetic Usually normal Abnormal† Abnormal†

Structural parameters

LV size Normal Normal or Dilated‡

mildly dilated‡

Doppler parameters (qualitative or semiquantitative)

Jet width in central jets (% LVO Narrow (≤25%) Intermediate Large (≥65%)
diameter): colour* (26-64%)

Jet density: CW Doppler Incomplete or faint Dense Dense

Jet deceleration rate (PHT, Slow (>500) Variable (200- Steep (<200)
ms):CW doppler§ 500)

LVO flow vs. pulmonary flow: Slightly increased Intermediate Greatly increased
PW Doppler

Diastolic flow reversal in the Absent or brief Intermediate Prominent,

descending aorta: PW Doppler early diastolic holodiastolic

Doppler parameters (quantitative)

Regurgitant volume (mL/beat) <30 30-59 >60

Regurgitant fraction (%) <30 30-50 >50

CW: continuous wave; LV: left ventricular; LVO: left ventricular outflow; PHT: pressure half-time; PW:
pulsed wave
*Parameter applicable to central jets and is less accurate in eccentric jets: Nyquist limit of 50-60 cm/s.
†Abnormal mechanical valves, for example, immobile occlude (valvular regurgitation), dehiscence or
rocking (paravalvular regurgitation); abnormal biologic valves, for example, leaflet thickening or prolapse
(valvular), dehiscence or rocking (paravalvular regurgitation).
‡Applies to chronic, late postoperative AR in the absence of other aetiologies.
§Influenced by LV compliance.

Clinical evidence guidelines: Medical devices Page 139 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Table 18: Doppler parameters for assessment of stenosis in prosthetic mitral valves from
Zoghbi et al (2009)

Parameter Normal Possible stenosis Suggests significant

stenosis

Peak velocity (m/s) <1.9 1.9-2.5 ≥2.5

Mean gradient (mm ≤5 6-10 >10

HG)

VTIPrMv/VTILVO†§ <2.2 2.2-2.5 >2.5

EOA (cm2) ≥2.0 1-2 <1

PHT (ms) <130 130-200 >200

PHT: pressure half time; PrMV: prosthetic mitral valve.

*Best specificity for normality or abnormality is seen if the majority of the parameters listed are normal or
abnormal, respectively.
†Slightly higher cut off values than shown may be seen in some bioprosthetic valves.
‡values of the parameters should prompt a closer evaluation of valve function and/or other
considerations such as increased flow, increased heart rate, or prosthesis-patient mismatch.
§These parameters are also abnormal in the presence of significant prosthetic mitral regurgitation.

Clinical evidence guidelines: Medical devices Page 140 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Table 19: Echocardiographic and Doppler criteria for severity of prosthetic mitral valve
regurgitation using findings from transthoracic echocardiograms and transesophogeal
echocardiogram from Zoghbi et al (2009)

Parameter Mild Moderate Severe

Structural parameters

LV size Normal* Normal or dilated Usually dilated‡

Prosthetic valve║ Usually normal Abnormal¶ Abnormal¶

Doppler parameters

Colour flow jet area║# Small, central jet Variable Large central jet (usually
(usually < 4 cm2 or >8 cm2 or >40% of LA
<20% of LA area) area) or variable size
wall-impinging jet
swirling in left atrium

Flow convergence** None or minimal Intermediate Large

Jet density: CW Incomplete or faint Dense Dense

Doppler║

Jet contour: CW Parabolic Usually parabolic Early peaking, triangular

Doppler║

Pulmonary venous Systolic dominance Systolic blunting§ Systolic flow reversal†

flow║

Quantitative parameters††

VC width (cm) ║ <0.3 0.3-0.59 ≥0.6

R vol (mL/beat) <30 30-59 ≥60

RF (%) <30 30-49 ≥50

EROA (cm2) <0.20 0.20-0.49 ≥.50

EROA: effective regurgitant orifice area; LA: left atrial; RF: regurgitant fraction; R vol: regurgitant volume;
VC: vena contracta.
*LV size applied only to chronic lesions.
†Pulmonary venous systolic flow reversal is specific but not sensitive for severe MR.
‡In the absence of other aetiologies of LV enlargement and acute MR.
§Unless other reasons for systolic blunting (e.g., atrial fibrillation, elevated LA pressure).
║Parameter may be best evaluated or obtained with TEE, particularly in mechanical calves.
¶Abnormal mechanical valves, for example, immobile occlude (valvular regurgitation), dehiscence or
rocking (paravalvular regurgitation); abnormal biologic valves, for example, leaflet thickening or prolapse
(valvular), dehiscence or rocking (paravalvular regurgitation).
#At a Nyquist limit of 50 to 60 cm/s.
**Minimal and large flow convergence defined as a flow convergence radius<0.4 and ≥0.9 cm for central
jets, respectively, with a baseline shift at a Nyquist limit of 40 cm/s; cut-offs for eccentric jets may be
higher.
†† These quantitative parameters are less well validated than in native MR.

Clinical evidence guidelines: Medical devices Page 141 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Table 20: Objective performance criteria (OPC) from the ISO for valve-related
complications for new valves or newly modified valves implanted surgically (% per
patient-year)*

Mechanical Valve Bioprosthetic Valve

Adverse event Aortic Mitral Aortic Mitral

Thromboembolism 1.6 2.2 1.5 1.3

Valve thrombosis 0.1 0.2 0.04 0.03

Major haemorrhage 1.6 1.4 0.6 0.7

Major paravalvular 0.3 0.5 0.3 0.2

leak

Endocarditis 0.3 0.3 0.5 0.4

*Not for transcather valves. A new valve is required to have complication rates lower than twice the OPC

Clinical evidence guidelines: Medical devices Page 142 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Characteristics of clinical studies on heart valve prostheses

Table 21: Summary of study characteristics of six systematic reviews on surgical aortic
valve replacement identified in a health technology assessment

Review Design of Numbers of Follow-up Comparison

included studies and
studies patients

Kassai et al 2000 RCTs 3 studies (2 in Mean of 11–12 Aortic and/or

adults) years for adults mitral:
mechanical vs.
1,229 patients bioprosthetic
(1,011 adult)

Kunadian et al RCTs 11 studies NR Aortic: Stented

2007 vs. non-stented
919 patients bioprosthetic

Lund and Bland, Observational 32 articles Mean 6.4 years Aortic:

2006 describing 38 for mechanical Mechanical vs.
case series (range, 3.9 to bioprosthetic
10.8) and 5.3
17,439 patients years (2.6 to
10.1 for
bioprosthetic)

Puvimanasghe Observational 22 studies Total follow-up Aortic: St. Jude

et al 2004 in patient-years mechanical vs.
13,281 patients was 25,726 for porcine
Puvimanasinghe St Jude bioprosthetic
et al 2003 mechanical and
54,151 for
porcine
bioprosthesis

Puvimanasinghe Observational 13 studies 18 years for Aortic:

et al 2006 Carpentier- Carpentier-
6,481 patients Edwards Edwards
pericardial pericardial
valves and up to aortic vs.
20 years for Carpentier-
Carpentier- Edwards supra-
Edwards annular
porcine bioprosthetic
supraanular
valves

Rizzoli et al Observational 11 studies Mean duration: Tricuspid:

2004 6.8 years Bioprosthetic vs.
1,160 patients mechanical
valves

NR: not reported

Clinical evidence guidelines: Medical devices Page 143 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Table 22: Summary of study characteristics of 57 RCTs* on surgical aortic valve

replacement identified in a Health Technology Assessment.
Total number of patients: 12,379

Valve types Valve comparisons Average follow-up time

studied

Aortic (n=43) Most common comparison was 1 year or sooner (69% of studies)
bioprosthetic stented vs.
bioprosthetic unstented (n=15) >1 to 5 years (24% of studies)
> 5 to 10 years (7% of studies

Aortic and Homograft vs. mechanical (n=1) >1 to 5 years (36% of studies)
mitral
(n=11) Mechanical vs. mechanical (n=7) > 5 to 10 years (45% of studies)
Mechanical vs. bioprosthetic (n=2) >10 years (18% of studies)
Bioprosthetic vs. bioprosthetic (n=1)

Mitral (n=3) All compared mechanical valves Mean of 5 years

*Note: Sixteen of the 57 trials were included in the systematic reviews in Table 21

Clinical evidence guidelines: Medical devices Page 144 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Table 23: Summary of study characteristics of two Health Technology Assessments on

transcatheter aortic valve implantation

Study Study details Numbers of Follow-up Comparison

patients

NICE (2011) HTA including 1 Systematic Systematic Level II studies:

systematic review (all review: n=2,375 review: greater TAVI vs. standard
Level IV studies)*, 2 than 1 year in 7 therapy and TAVI
Level II studies:
level II studies, 1 case series and vs. surgical
n=358 and n=699
Level III study and 6 30 days in 22 implantation
Level IV studies Level III study: case series
n=175
Level II studies:
Level IV studies: maximum of 2.8
n=ranged from years and 1.4
70 to 1,038 years (median)
Level III study:
median of 466
days
Level IV studies:
ranged from 30
days to a median
of 3.7 years

Tice HTA including 2 Level Level II studies: Level II studies: Level II studies:
(2014)205 II studies†, 10 Level n=358 and n=699 19 months and TAVI vs. standard
III studies‡ and 16 24 months therapy and TAVI
Level III studies:
Level IV studies§ vs. surgical
ranged from Level III studies:
placement
n=51 to n=8,536 ranged 1 month
to 24 months Level III studies:
Level IV studies:
all TAVI vs.
ranged from Level IV studies:
surgical
n=130 to ranged from 1
implantation
n=10,037 month18 months
except one TAVI
vs. surgical
implantation vs.
medical therapy

Registries NA 132 to 4,571 Major events NA

generally
reported at 30
days and then
yearly after that.
Maximum follow-
up of 3 years for
the registries
identified
HTA: Health Technology Assessment; NA: not applicable
*Note: given the systematic review is not on Level II studies it does not meet the Level I study
classification as prescribed by the NHMRC
†Same Level II studies as included in NICE (2011)
‡ Includes one Level III study which is a meta-analyses
§ Includes two Level IV studies which are meta-analyses

Clinical evidence guidelines: Medical devices Page 145 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Table 24: Summary of study characteristics of two Health Technology Assessments and
one multicentre case series on sutureless aortic valve replacement

Study Study details Numbers of Follow-up Comparison

patients

NICE (2012) HTA including 7 Range from 30 to Range from 1 Level III study
studies* (1 Level III 208 duration of compared S-AVR
and 6 Level IV) hospital stay (NR) to TA-TAVI
to 16 months

Sinclair et al HTA including 6 Range from 6 to Range from a NA

(2013) studies† (all Level IV) 140 mean of 313 days
to up to 3 years

Englberger et Single Level IV 141 5 years NA

al (2014) (multicentre) study

HTA: Health Technology Assessment; NA: not applicable; S-AVR: sutureless aortic valve replacement; TA-
TAVI: transapical-transaortic valve implantation
*This Health Technology Assessment also included one case report which was not included in data
extraction
†The Health Technology Assessment included nine case series in total but three were only in abstract
form so were not included in data extraction. One of the six case series in this Health Technology
Assessment was also included in the Health Technology Assessment by NICE 2012

References
Kappetein AP, Head SJ, Genereux P et al. Updated standardized endpoint definitions for
transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus
document. Eur Heart J. 2012;33(19):2403-18.
Williams JW, Coeytaux R, Wang A, Glower DD. Percutaneous Heart Valve Replacment. Technical
Brief No. 2. . Rockville, MD: Prepared by Duke Evidence-based Practice Center under Contract
No. 290-02-0025. Agency for Healthcare Research and Quality; 2010.
Health Policy Advisory Committee on Technology, Sutureless aortic valve replacement in
patients with severe aortic valve stenosis
ISO Standard 5840-1:2021: Cardiovascular implants -- Cardiac valve prostheses -- Part 1:
General requirements
ISO Standard 5840-2:2021: Cardiovascular implants - Cardiac valve prostheses - Part 2:
Surgically implanted heart valve substitutes
ISO Standard 5840-3:2021: Cardiovascular implants - Cardiac valve prostheses - Part 3: Heart
valve substitutes implanted by transcatheter techniques
Avanzas P, Munoz-Garcia AJ, Segura J et al. Percutaneous implantation of the CoreValve self-
expanding aortic valve prosthesis in patients with severe aortic stenosis: early experience in
Spain. Rev Esp Cardiol. 2010;63(2):141-8
Buellesfeld L, Gerckens U, Schuler G et al. 2-year follow-up of patients undergoing transcatheter
aortic valve implantation using a self-expanding valve prosthesis. J Am Coll Cardiol.
2011;57(16):1650-7

Clinical evidence guidelines: Medical devices Page 146 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Eltchaninoff H, Prat A, Gilard M et al. Transcatheter aortic valve implantation: early results of the
FRANCE (FRench Aortic National CoreValve and Edwards) registry. Eur Heart J. 2011;32(2):191-
7
Godino C, Maisano F, Montorfano M et al. Outcomes after transcatheter aortic valve implantation
with both Edwards-SAPIEN and CoreValve devices in a single center: the Milan experience. JACC
Cardiovasc Interv. 2010;3(11):1110-21
Haussig S, Schuler G, Linke A. Worldwide TAVI registries: What have we learned? Clin Res
Cardiol. 2014;103(8):603-12
Moat NE, Ludman P, de Belder MA et al. Long-term outcomes after transcatheter aortic valve
implantation in high-risk patients with severe aortic stenosis: the U.K. TAVI (United Kingdom
Transcatheter Aortic Valve Implantation) Registry. J Am Coll Cardiol. 2011;58(20):2130-8
Moynagh AM, Scott DJ, Baumbach A et al. CoreValve transcatheter aortic valve implantation via
the subclavian artery: comparison with the transfemoral approach. J Am Coll Cardiol.
2011;57(5):634-5
Tamburino C, Capodanno D, Ramondo A et al. Incidence and predictors of early and late
mortality after transcatheter aortic valve implantation in 663 patients with severe aortic
stenosis. Circulation. 2011;123(3):299-308
Thomas M, Schymik G, Walther T et al. One-year outcomes of cohort 1 in the Edwards SAPIEN
Aortic Bioprosthesis European Outcome (SOURCE) registry: the European registry of
transcatheter aortic valve implantation using the Edwards SAPIEN valve. Circulation.
2011;124(4):425-33
Thomas M, Schymik G, Walther T et al. Thirty-day results of the SAPIEN aortic Bioprosthesis
European Outcome (SOURCE) Registry: A European registry of transcatheter aortic valve
implantation using the Edwards SAPIEN valve. Circulation. 2010;122(1):62-9
Walters DL, Sinhal A, Baron D et al. Initial experience with the balloon expandable Edwards-
SAPIEN Transcatheter Heart Valve in Australia and New Zealand: the SOURCE ANZ registry:
outcomes at 30 days and one year. Int J Cardiol. 2014;170(3):406-12
Zahn R, Gerckens U, Grube E et al. Transcatheter aortic valve implantation: first results from a
multi-centre real-world registry. Eur Heart J. 2011;32(2):198-204
Akins CW, Miller DC, Turina MI et al. Guidelines for reporting mortality and morbidity after
cardiac valve interventions. Ann Thorac Surg. 2008;85(4):1490-5
European Journal of Cardio-Thoracic Surgery 42 (2012) S45–S60 - Updated standardized
endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research
Consortium-2 consensus document
Leon MB, Piazza N, Nikolsky E et al. Standardized endpoint definitions for Transcatheter Aortic
Valve Implantation clinical trials: a consensus report from the Valve Academic Research
Consortium. J Am Coll Cardiol. 2011;57(3):253-69
Zamorano JL, Badano LP, Bruce C et al. EAE/ASE recommendations for the use of
echocardiography in new transcatheter interventions for valvular heart disease. J Am Soc
Echocardiogr. 2011;24(9):937-65
Zoghbi WA, Chambers JB, Dumesnil JG et al. Recommendations for Evaluation of Prosthetic
Valves With Echocardiography and Doppler Ultrasound. A Report From the American Society of
Echocardiography's Guidelines and Standards Committee and the Task Force on Prosthetic
Valves, Developed in Conjunction With the American College of Cardiology Cardiovascular

Clinical evidence guidelines: Medical devices Page 147 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Imaging Committee, Cardiac Imaging Committee of the American Heart Association. J Am Soc
Echocardiogr. 2009;22(9):975-1014
Wu Y, Butchart EG, Borer JS, Yoganathan A, Grunkemeier GL. Clinical Evaluation of New Heart
Valve Prostheses: Update of Objective Performance Criteria. Ann Thorac Surg. 2014
Kassai B, Gueyffier F, Cucherat M, Boissel JP. Comparison of bioprosthesis and mechanical valves,
a meta-analysis of randomised clinical trials. Cardiovasc Surg. 2000;8(6):477-83
Kunadian B, Vijayalakshmi K, Thornley AR et al. Meta-analysis of valve hemodynamics and left
ventricular mass regression for stentless versus stented aortic valves. Ann Thorac Surg.
2007;84(1):73-8
Lund O, Bland M. Risk-corrected impact of mechanical versus bioprosthetic valves on long-term
mortality after aortic valve replacement. J Thorac Cardiovasc Surg. 2006;132(1):20-6.e3
Puvimanasinghe JP, Takkenberg JJ, Edwards MB et al. Comparison of outcomes after aortic valve
replacement with a mechanical valve or a bioprosthesis using microsimulation. Heart.
2004;90(10):1172-8
Puvimanasinghe JP, Takkenberg JJ, Eijkemans MJ et al. Choice of a mechanical valve or a
bioprosthesis for AVR: does CABG matter? Eur J Cardiothorac Surg. 2003;23(5):688-95;
discussion 95
Puvimanasinghe JP, Takkenberg JJ, Eijkemans MJ et al. Comparison of Carpentier-Edwards
pericardial and supraannular bioprostheses in aortic valve replacement. Eur J Cardiothorac Surg.
2006;29(3):374-9
Rizzoli G, Vendramin I, Nesseris G, Bottio T, Guglielmi C, Schiavon L. Biological or mechanical
prostheses in tricuspid position? A meta-analysis of intra-institutional results. Ann Thorac Surg.
2004;77(5):1607-14
National Institute for Health and Clinical Excellence. Interventional procedure overview of
transcatheter aortic valve implantation for aortic stenosis.: National Institute for Health and
Clinical Excellence; 2011
Tice JA, Sellke FW, Schaff HV. Transcatheter aortic valve replacement in patients with severe
aortic stenosis who are at high risk for surgical complications: Summary assessment of the
California Technology Assessment Forum. J Thorac Cardiovasc Surg. 2014;148(2):482-91.e6
National Institute for Health and Clinical Excellence. Interventional procedure overview of
sutureless aortic valve replacement for aortic stenosis. ; 2012. Report No.: IPG456
Sinclair A, Xie X, McGregor M. Surgical aortic valve replacement with the ATS Enable sutureless
aortic valve for aortic stenosis. Montreal, Canada: Technology Assessment Unit (TAU) of the
McGill University Health Centre (MUHC); 2013
Englberger L, Carrel TP, Doss M et al. Clinical performance of a sutureless aortic bioprosthesis:
Five-year results of the 3f Enable long-term follow-up study. J Thorac Cardiovasc Surg.
2014;148(4):1681-7

Clinical evidence guidelines: Medical devices Page 148 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Supportive Devices - Meshes, patches and tissue

adhesives
Supportive devices act as scaffolds, reinforcement or buttressing and include all devices that
hold, fix or sustain body organs or incisions. The majority of supportive devices are surgical
meshes for hernia and gynaecological repair, central nervous system (CNS) patches, and tissue
adhesives, but sheeting of various origins is also included.
These devices can be made from biologic and non-biologic materials and be permanent or
absorbable in various combinations. Each type of supportive device has its own associated
benefit-risk profile that needs to be addressed by the manufacturer.

Summary recommendations
• Manufacturers are advised that pre-clinical data demonstrating that the mechanical,
biocompatibility and physical characteristics of the device are congruent with the intended
purpose and anticipated in vivo lifespan of the surgical support.
• Provision of clinical investigational data:
– manufacturers who intend to conduct a clinical trial should design the trial using the
highest practical NHMRC Level of Evidence and trials should be appropriate to inform
on the safety and performance of the device for its intended purpose
– it is suggested that the minimum period for patient follow-up for clinical trials is 24
months for permanent and biological meshes. At the time of writing there is no agreed
recommended follow up for patches or tissue adhesives.
– across the surgical supports the main clinical outcomes that determine safety and
performance for hernia repair are recurrence rate, reoperation rate, function and QoL
scores, adhesions (particularly for intraperitoneal mesh), mesh degradation, seroma
and pain, and for pelvic organ prolapse (POP) and stress urinary incontinence (SUI),
cure of stress incontinence and patient scores such as the Pelvic Organ Prolapse
Quantification System (POP-Q)
▪ for revision data, the manufacturer is advised to benchmark the device against
devices of the same class as reported by an international registry, if available
▪ for patient performance data, manufacturers are advised to define the anticipated
improvement in patient scores post-surgery. Ideally, these should be
internationally recognised assessment tool(s) used to measure clinical success, e.g.,
QoL or cough stress test
– when submitting a comprehensive literature review, full details of the method used
should be included in the CER in sufficient detail to ensure the literature review can be
reproduced.
– for guidance on the conduct of comprehensive literature reviews and presentation of
clinical evidence manufacturers are directed to relevant sections in this document.
• For submissions reliant on comparable device data, manufacturers and sponsors are advised
to submit all relevant documents with a supporting clinical justification that establishes
substantial equivalence between a device and the nominated comparable device(s).
• In addition, a well-documented risk analysis and management system should also be
provided with the CER.

Clinical evidence guidelines: Medical devices Page 149 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• Manufacturers should provide details of the clinical context within which the clinical data
were obtained. The clinical context of the evidence base should be congruent with the
indications of use for which the manufacturer seeks TGA approval.
• Compilation of the CER:
– in compiling the clinical evidence for a supportive device the manufacturer should
ensure that a clinician who is an expert in the field and experienced in the use of the
device critically evaluates all the clinical data that informs on the safety and
performance of the device
– the clinical expert must then endorse the CER containing the clinical evidence
(evidenced by signature and date) to demonstrate that the evidence meets the
requirements of the applicable EPs and the device is deemed to be safe and to perform
as intended.

Defining supportive devices

The TGA describes supportive devices as devices in the following sub-groups.
• Surgical mesh: this is a textile-based sheet (typically knotted or warp knitted) used as a
temporary or permanent support for organs or other tissues. It is used for hernia repair,
POP, SUI and many other purposes. The main classes of surgical mesh are biological and
synthetic or a combination of these. Types of mesh include bio-mesh, polypropylene,
expanded polytetrafluoroethylene (ePTFE), composite polypropylene-PTFE, polyester,
composite meshes that combine permanent and absorbable materials such as collagen,
polyglactin, polylactic acid and polyglycolic acid and in combination with materials such as
titanium. More than one type can be used at once and they can be absorbable, semi-
absorbable and non-absorbable. The configuration of mesh varies. Fixation methods include
staples, sutures, tackers and glue.
• Patches: specifically CNS patches, both absorbable and non-absorbable, are impermeable
adhesive membranes used in intradural neurosurgical procedures, as an alternative to using
autologous grafts or cadaveric implants. These patches are used to reinforce dural closure
when there is the risk of postoperative cerebrospinal fluid (CSF) leak.
• Tissue Adhesives: these are an alternative to sutures and staples used for closure of
wounds and fixation of devices such as surgical mesh, patches and scaffolding to tissues.
They may also be used as a sealant for closure, for example, of colostomies. Tissue adhesives
are defined as any substance with characteristics that allow for polymerization. This
polymerization must either hold tissue together or serve as a barrier to leakage or to control
bleeding. Fibrin sealants are the most commonly used adhesives. Other adhesives include
cyanoacrylates, albumin-based compounds, collagen-based compounds, glutaraldehyde
glues and hydrogels. Tissue adhesives can act as a barrier to microbial penetration as long as
the adhesive film remains intact.
Any of the supportive devices can include biocompatible coated materials such as silver coating,
titanium dioxide, hydroxyapatite, hyaluronate, monocryl, paclitaxel and many other materials.

Clinical evidence guidelines: Medical devices Page 150 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Clinical evidence
The clinical evidence can be derived from clinical investigation(s) data, a comprehensive
literature review and/or post-market data (clinical experience) on the device (direct) and/or the
comparable device (indirect). Direct clinical evidence on the actual device is preferred. It is
important to clarify if any changes have been made to the device since the clinical data were
gathered and if so to document the changes and to clarify the exact version of the device.
Otherwise indirect clinical evidence on a comparable device may be used after substantial
equivalence has been demonstrated through a comparison of the clinical, technical and
biological characteristics as described in Comparable devices including substantially equivalent
devices.
Where the device and the predicate share any common design origin, the lineage of the devices
should be provided as well. The intended purpose, clinical indications, claims and
contraindications must be supported by the clinical data. Manufacturers should refer to Clinical
evidence requirements for more information.

Clinical investigation(s)
The design of the clinical investigation(s) should be appropriate to generate valid measures of
clinical performance and safety. The preferred design is a randomised controlled clinical trial
and conditions should ideally represent clinical practice in Australia.
The eligible patient groups should be clearly defined with exclusion/inclusion criteria, patient
profiles and morbidity as well as specific indications. In addition the risks, techniques, design of
implants and accessories and experience of users should be taken into account. Manufacturers
are advised to justify the patient numbers recruited according to sound scientific reasoning
through statistical power calculation. Registry data from jurisdictions where the device is
marketed may provide useful clinical evidence.
The duration of the clinical investigation should be appropriate to the device, the patient
population and medical conditions for which it is intended. Duration should always be justified,
taking into account the time-frame of expected complications. Analysis of clinical events should
be blinded and independently adjudicated wherever possible.

Literature review
A literature review involves the systematic identification, synthesis and analysis of all available
published and unpublished literature, favourable and unfavourable, on the device or comparable
device when used for its intended purpose(s).
The literature search protocol should be determined prior to implementing the search, detailing
the aim, search terms, planned steps and inclusion and exclusion criteria. Data on the materials
used to construct the device, their biocompatibility, the device dimensions and geometry and the
intended purpose will determine the construction of search strategies as well as study selection.
The selection of comparable device should be made prior to performing the literature selection,
extraction of the clinical evidence and analysis of the pooled results. The search output should
be assessed against clearly defined selection criteria documenting the results of each search step
with clear detail of how each citation does or does not fit the selection criteria for inclusion in
the review. This ensures that the searches are comprehensive and the included studies are
related to the device in question or substantially equivalent device(s).
A full description of the device used or adequate information to identify the device (e.g.
manufacturer name and model number) must be extractable from study report. If this is not
possible, the study should be excluded from the review. The overall body of evidence from the
literature should be synthesised and critically evaluated by a competent clinical expert and a
literature report prepared containing a critical appraisal of this compilation. The full details of
the search can be provided in the supporting documents and should be sufficient to allow the
search to be reproduced.

Clinical evidence guidelines: Medical devices Page 151 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Post-market data
Post-market data can be provided for the actual device or for the comparable device.
It is particularly important to include the following:
• information about the regulatory status of the device(s) or comparable device, including the
certificate number, date of issue and name under which the device is marketed, the exact
wording of the intended purpose/approved indication(s) and any conditions in other
jurisdictions
• any regulatory action such as CE mark withdrawals, recalls (including recalls for product
correction, and the reason for these i.e. IFU change), suspensions, removals, cancellations,
any other corrective action ) anywhere in the world as reported to or required by regulatory
bodies
• distribution numbers of the device(s) including by country and/or geographical region for
every year since launch. It is accepted that this may not always be appropriate for high
volume devices, those with many components or those on the market for many years
• the number of years of use
• for every year since launch, adverse events, complaints and vigilance data categorised by
type and clinical outcome (adhesion, tissue damage (erosion, dehiscence etc.), chronic pain,
bacterial infection and toxicity due to chemical components of the device)
• the post-market surveillance data from national registries in jurisdictions where the device
is approved for clinical use if available
• explanted devices returned to manufacturers should be accounted for with an explanation of
device failures and corrective measures.
Publicly available post-market data such as adverse event reporting on the FDA MAUDE
database and the TGA IRIS may be used for devices from other manufacturers.
For reports of adverse events and device failures to be useful clinical evidence, the manufacturer
must make a positive, concerted effort to collect the reports and to encourage users to report
incidents. Experience shows that merely relying on spontaneous reports leads to an
underestimation of the incidence of devices failures and adverse events.
The post-market data should be critically evaluated by an appropriately qualified clinical expert,
that is, someone with relevant medical qualifications and direct clinical experience in the use of
the device or device type in a clinical setting. The CER should then be endorsed by the clinical
expert (evidenced by signature and date) to enable an understanding of the safety and
performance profile of the device(s) in a ‘real-world’ setting.

Compiling the CER

In compiling the clinical evidence the manufacturer should ensure that a clinical expert in the
relevant field critically evaluates all the clinical data from clinical investigation(s), literature
review and/or post-market data (clinical experience) and provides a written report, the CER, to
allow the clinical assessor to determine whether the clinical evidence is sufficient to
demonstrate that the requirements of the applicable EPs have been met and the device is safe
and performs as intended.

Clinical evidence guidelines: Medical devices Page 152 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Sources of clinical data and The Clinical Evaluation Report outline the components that may
comprise clinical evidence for a medical device, and the process to compile a CER, respectively.
These apply whether the manufacturer is using direct clinical evidence or relying on indirect
clinical evidence based on a comparable device. Guidance on defining a comparable device is
provided in Comparable devices including substantially equivalent devices.
The device description should include sufficiently detailed information to satisfy the
requirements of Appendix 3 of MEDDEV 2.7.1 Rev 4 on “Device description – typical contents”.
For supportive devices this may include, but is not limited to; the material type, chemical
composition, biological compatibility testing, coating, porosity, flexibility, tensile strength,
durability and dimensions. If biological actives are impregnated the in vitro activity should be
demonstrated and documented in the submission.
The design of clinical studies to demonstrate the clinical safety and performance of devices that
have no equivalent comparable device must include all device characteristics and all intended
uses. If a comparable device is available and data from that device is used to support a
submission, the device characteristics and intended purpose will determine the criteria for a full
and reasoned clinical justification for the comparable device selection.
As per The Clinical Evaluation Report the CER should include the following:
a) General details
b) Description of the medical device and its intended application
c) Intended therapeutic and/or diagnostic indications and claims
d) Context of the evaluation and choice of clinical data types
e) Summary of relevant pre-clinical data
f) Discussion regarding comparable devices including substantially equivalent devices
g) Summary of the clinical data and appraisal
h) Data analysis
i) Conclusions
j) Name, signature and curriculum vitae of clinical expert and date of report

Supportive data and information

The following information on the device must also be provided:
• risk assessment and management document
• IFU, labelling, product manual and all other documents supplied with the device. The clinical
evidence must highlight the risks and ensure that these are appropriately communicated to
user.
Additional information should be provided as applicable. This may include (but is not limited
to):
• additional specifications of the device(s)
• the materials from which the device is made including chemical composition
• other devices that may be used in conjunction with the device
• any aspects of non-clinical testing results that inform the design of the clinical trial should be
included in the supporting documents
Clinical evidence guidelines: Medical devices Page 153 of 178
V3.1 June 2022
 Therapeutic Goods Administration

• biocompatibility testing, bench testing and animal studies where applicable

• specific testing of any adjuvant medicinal components may be required especially if these
are new chemical entities in the Australian context. This should cover interactions between
the device and the medicine, pharmacodynamics and time-release profiles.
• any further details of post-market data

Defining clinical success

Meshes
Hernia repair surgery is the most common application for surgical meshes followed by
reconstructive surgery for POP and SUI.
Meshes can be used for either a primary or secondary repair or as suture line reinforcement
material. It is imperative that the clinical evidence reflects the indication for use of the mesh
under review. Measures such as de novo or worsening prolapse in a non-treated compartment
and urinary symptoms may be reported as both safety and performance measures.

Safety
Post-operative complications and/or reoperation are the primary safety outcome measures
although subjective measures of success should also be included.
Complications associated with surgical mesh for hernia repair reported in the literature include
adhesions, fistula, bowel obstruction, mesh erosion, bleeding, infection, haematoma, seroma and
chronic pain. Bowel obstruction is not seen in extra peritoneal mesh placement. Some of these
complications may occur with surgery and are not due to the mesh per se.
Complications associated with surgical mesh for POP and SUI reported in the literature include
pain, bleeding, organ perforation (such as bladder and urethral perforation), dyspareunia,
visceral injury, urinary issues (including retention, voiding dysfunction, urge incontinence,
overactive bladder) as well as late events such as mesh erosion and exposure. A summary of the
safety data extracted from systematic reviews is provided in Table 25. Clinical experts have
reported additional complications associated with the use of surgical mesh for POP and SUI
which include inflammation, seroma, haematoma, infection, fistula, urinary tract infection, bowel
dysfunction, nerve injury, chronic pain and de novo or worsening prolapse in a non-treated
compartment.
The manufacturer should report all post-surgical complications and serious adverse events or
failures that have been found with the use of the mesh or comparable device(s) if used for
comparison. Registers also collect valuable information on surgical outcomes and some public
measures of performance and adverse outcomes.
One direct register for meshes used in POP repair was identified:
• Austrian Urogynecology Working Group registry for transvaginal mesh devices for POP
repair

Clinical evidence guidelines: Medical devices Page 154 of 178

V3.1 June 2022
 Therapeutic Goods Administration

In addition a number of registers for surgeries that involve meshes for hernia repair were
identified:
• Swedish hernia register
• Herniamed, a German internet-based registry for outcome research in hernia surgery
• Americas Hernias Society Quality Collaborative (AHSQC) in the USA
• European Registry of Abdominal Wall Hernias (EuraHS)
• ClubHernie
The Environmental Protection Agency’s Integrated Risk Information System (IRIS) is a US safety
database for toxicology and human effects data from chemical substances which may in some
cases provide information on products used in or with meshes.
Based on the literature reviewed for these guidelines, if clinical studies are conducted, the
minimum patient follow-up should be 24 months for hernia and gynaecological repair. However,
manufacturers should be aware that late adverse events of a device can occur many years after
implantation.
Safety parameters should be established a priori with nominated values clinically justified by a
clinical expert experienced in the use of the device.

Performance
It is useful to divide success into objective success measures and subjective success measures,
such as clinician reported outcomes and patient-reported outcomes. Performance related
parameters reported in the peer reviewed literature for surgical meshes include recurrence
rates, reoperation rates, functional scores, quality of life scores and pain. For absorbable devices,
clearance and metabolism times are also provided in Table 25. Other measures for performance
are objective success measures (including anatomic success measure such as POP-Q) and
subjective success measures such as quality of life outcomes. An important outcome is de novo
or worsening prolapse in a non-treated compartment and, specifically in regards to SUI, de novo
or worsening urinary symptoms should be included as a measure of performance.

Primary repair
Recurrence and reoperation rates can be used to measure clinical success in primary repair
surgery.
Recurrence rates of 15-25% are frequently reported after mesh repair of a hernia. The rates of
reoperation vary based on the indication, patient characteristics and surgical procedure
undertaken, therefore, depending on these characteristics, rates within this range may be
considered acceptable. A satisfactory result of biologic mesh application is a recurrence rate of
18% or below and seroma formation of 12% or less.
Importantly, patient follow-up periods must be comparable to accurately compare recurrence
rates as a function of supportive devices.

Primary and secondary outcomes

Clinical success is often evaluated by patient-oriented assessment tools that determine
functional outcomes. It can also be evaluated by primary outcomes or secondary outcomes, and
it is important to make a distinction between these two. Functional scores provide an aggregate
of patient reported domains (e.g. pain) with an objective measure of mesh success (e.g. current
size of hernia) and represent a clinically meaningful grading of mesh performance. However, for
procedures using surgical mesh, the short-term performance of a device may be dominated by
procedural variables; therefore sufficient time should lapse to isolate device-specific
improvements.

Clinical evidence guidelines: Medical devices Page 155 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Measures of performance that may be of use include the Ventral Hernia Working Group (VHWG)
grading system and the Pelvic Organ Prolapse Quantification System (POP-Q). POP-Q is a
validated staging system for pelvic organ prolapse and currently the most quantitative, site-
specific system with high reported inter-observer reliability. The VHWG has a staging system
which predicts both risk and likely outcome in terms of both recurrence and SSO. It is made up
of the VHWG grading system plus a defect size component to predict SSO and recurrence and has
been validated for clinical application.
Where validated measurement tools are not used, manufacturers can assist the clinical assessor
by providing data based on surrogate markers. The choice of surrogate markers and the
validation of these to predict future complications or failure should be clinically justified and
consistent with the proposed therapeutic indications.
Examples of surrogate markers for mesh performance are:
• Reoperation for recurrence in hernia surgery
• For hiatal hernia, radiological or endoscopic absence of a recurrent hernia (defined as >2cm
in size)211
• For POP, examples of surrogate markers of performance include: recurrent prolapse,
ongoing pain including dyspareunia, de novo urinary or bowel symptoms.
• For SUI, de novo or worsening urinary symptoms

Manufacturers should, where possible, use validated measurement tools.

When selecting and reporting surrogate markers of performance
manufacturers should provide a clinical justification for the selection.

Minimum benchmarks that need to be reached to demonstrate the device is performing as

expected and is equivalent to already marketed products should be used. For prolapse, at one
year POP-Q stage II or greater is considered to be surgical failure and POP-Q stage I was
considered a surgical cure. For hernia, at the time of writing, there are no benchmarks for
performance.

Patches
Central Nervous System (CNS) patches, both bioabsorbable and non-absorbable, are
impermeable adhesive membranes used in (intradural) neurosurgical procedures, as an
alternative to using autologous grafts or cadaveric implants. These patches are used to reinforce
dural closure when there is the risk of postoperative cerebrospinal fluid (CSF) leak.

Safety
For safety, the primary outcome measures are CSF leak, CSF fistula and deep wound infection.
Other complications associated with CNS patches (studies reviewed tested for these effects but
their occurrence was very rare) include adverse or allergic effects, hydrocephalus, brain tissue
scarring, new epileptic seizures and mortality, refer to Table 26. The manufacturer should
report all of the above and any other serious post-surgical events for the patch or comparable
device if used for comparison.
Based on the literature reviewed for these guidelines, the minimum possible patient follow-up
for studies conducted on CNS patch surgery is three months. However, manufacturers should be
aware that 3 months is the minimum and will not capture information relating to the late failure
of a patch. At the time of writing there are no benchmarks for CNS patches. Manufacturers
should define a minimum performance marker based on the literature and clinical expertise,
providing a clinical justification for the parameters and values that have been selected.
Clinical evidence guidelines: Medical devices Page 156 of 178
V3.1 June 2022
 Therapeutic Goods Administration

Performance
Performance related parameters reported in the peer reviewed literature for patches are
provided in Table 27.
Clinical success is often evaluated by patient-oriented assessment tools that determine
functional outcomes. With regards to mesh, functional scores provide an aggregate of patient
reported domains (e.g. pain) with an objective measure of mesh success (e.g. improvement in
POP-Q stage) and represent a clinically meaningful grading of mesh performance. No such tool
has been found for application of CNS patch. The most useful functional measure for CNS patches
is the existence of cerebrospinal fluid leakage. Manufacturers should define a minimum
performance marker based on the literature and clinical expertise, providing a clinical
justification for the parameters and values that have been selected.

Tissue adhesives
Safety
Chronic pain, infection, inflammation, tissue damage, bleeding and leakage of bile and other
fluids are primary outcome measures for tissue adhesive surgeries, refer to Table 26. Chronic
pain can be measured with Visual Analogue Score (VAS) as mild, moderate or severe persisting
from 3 months to 1 year. Secondary outcomes reported in the literature are numbness,
discomfort, patient satisfaction, QoL (measured with SF12), length of hospital stay, and time to
return to normal activities. The manufacturer should report any post-surgical complications and
failure of the adhesive or comparable adhesive device.
Articles reporting on tissue adhesives rarely report follow up times, rather they refer to post-
operative outcomes. Recurrence rates considered acceptable for surgeries using tissue
adhesives are important in measuring success. In the literature, recurrence was found to be
1.5% at 17.6 months in a study on hernia repair using fibrin glue. Another study found a
recurrence rate of 2.3% at 15 months. Thus a recurrence rate <2.3% in 15-18 months may be
acceptable. Rates for tissue adhesives other than those containing fibrin glue are not readily
evident, at time of writing. Patient follow-up periods must be comparable when using
recurrence rates as a measure of performance of tissue adhesives. Nominated recurrence rates
need to have a rigorous clinical justification provided by a clinical expert with experience in the
use of the device or device types who takes into account current research when evaluating all of
the clinical data in the CER.

Performance
Recurrence is one performance related parameter reported in the peer reviewed literature for
tissue adhesives (Table 27).
Clinical success of surgery is often evaluated by patient-oriented assessment tools that measure
functional outcomes. Functional scores would provide an aggregate of patient-reported domains
(e.g. pain) with an objective measure of success (e.g. fluid leakage) and represent a clinically
meaningful grading of performance. A functional measure for tissue adhesives is wound closure.
It is recommended that the manufacturers define a minimum performance marker based on the
literature and clinical expertise and provide a clinical justification for the parameters and values
that have been selected.

When documenting patient performance scores for tissue adhesives, it is

recommended that manufacturer provide a clinical justification for the follow-
up period used. At the time of writing 15-18 months follow-up has been
reported in the literature.

Clinical evidence guidelines: Medical devices Page 157 of 178

V3.1 June 2022
 Therapeutic Goods Administration

As assessment tools of device performance may not be available, manufacturers can assist the
clinical assessors by providing data on direct markers.
Examples of direct markers for performance of adhesives are:
• achievement of haemostasis/ increased number of patients reaching haemostasis –
measured as no evidence of bleeding from exposed surfaces
• presence of haematoma/ seroma during study, visual perception of oedema 1-7 days post-
operatively
• fluid drainage 24h post-operatively, volume of blood loss or transfusion, and resection
surface complications such as intra-abdominal fluid collections detected by CT scan
• reduction in drainage volume
• morbidity defined as all complications arising directly related to the procedure
• mortality defined as death within 30 days of the procedure or within the same hospital
admission

Manufacturers should, where possible, use validated measurement tools. If

selecting and reporting surrogate markers of performance manufacturers
should provide a clinical justification for the selection and validation of these
to predict device complications or failure.

Summary of safety and performance data

Characteristics of clinical studies on supportive devices

Table 25: Summary of study characteristics extracted from systematic reviews and
primary research reports on safety and performance of supportive devices

Characteristic of Meshes - Hernia Meshes - Patches Tissue

included studies Gynaecological Adhesives

Systematic 11 5 0 4
reviews

Number of 4 - 40 20 - 45 NA 4 - 10
included studies
per systematic
review

Sample size 14 - 1120 63 - 95 NA 20 - 255

(range) for
included studies

Dominant design RCT, observational, RCTs NA RCTs,

of included case control, observational
studies prospective cohort studies

Clinical evidence guidelines: Medical devices Page 158 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Characteristic of Meshes - Hernia Meshes - Patches Tissue

included studies Gynaecological Adhesives

Reported Lightweight v. heavy Mesh v. NA Fibrin sealant

comparisons mesh conventional v. staples
repair
Lichtenstein repair v. Fibrin sealant
mesh plugs Mesh v. vaginal v. Tranexamic
colpopexy acid
Sutures v. glue for
mesh fixation Mesh v. anterior Fibrin sealant
or posterior v. control
Sublay v. onlay for colporrhaphy
mesh position
Laporascopic v. open
surgery
Comparing mesh
materials
Biologic v. non
biologic mesh
Human-derived v.
porcine-derived
biologic mesh
Self-gripping mesh or
suture fixation

Quality of Poor to satisfactory Low to high NA Inadequate to

included evidence good
as reported

Patient Follow-up 1 month to 10 years 3 months to 3 NA 7 months to 4

years years
Comparative trials
e.g. RCTs

Clinical evidence guidelines: Medical devices Page 159 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Reported clinical outcomes in the peer reviewed literature

Table 26: Summary of safety data extracted from systematic reviews on supportive
devices

Vaginal Hernia surgical Tissue

Safety parameter Patches
surgical mesh mesh adhesives

Death ü ü

Urinary issues ü N/A N/A

Pain ü ü

Chronic pain ü N/A ü

Infection ü ü ü

Bleeding ü ü ü

Organ perforation ü N/A

Dyspareunia ü N/A N/A

Material exposure ü N/A N/A

Visceral injury ü N/A N/A

Mesh erosion ü ü N/A N/A

Haematoma ü N/A ü

Seroma ü N/A ü

Bile leak N/A N/A N/A ü

Cytotoxicity ü N/A ü

CSF leakage N/A N/A ü

Adhesions N/A ü N/A N/A

Fistula N/A ü N/A N/A

Bowel obstruction N/A ü N/A N/A

Hydrocephalus N/A N/A ü N/A

Greyed cells (N/A) indicate that the safety parameter is not applicable to that device class

Clinical evidence guidelines: Medical devices Page 160 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Table 27: Summary of performance data extracted from systematic reviews, RCTs and
primary research reports on the safety and performance of supportive devices

Performance Surgical Mesh - Surgical Mesh - Absorbable Patches Tissue

parameter Gynaecological Hernia devices Adhesives

Revision/
reoperation
ü ü ü ü ü
(recurrence
rates)

Function Pelvic Organ Existence

scores Prolapse of CSF
Quantification leakage
System (POP-Q)

Incontinence Impact
Questionnaire

Short-form
prolapse/Urinary
Incontinence Sexual
Questionnaire
(PISQ-12)

Patient Global
Impression of
Change (PGIC)

Pelvic Floor Distress

Inventory (PFDI-20)

Pelvic Floor Impact

Questionnaire
(PFIQ-7)

Surgical Satisfaction
Questionnaire (SSQ)

Quality of SF-36
Life (QoL)
scores SHS
SF-12
EuroQol EQ-5D

Clinical evidence guidelines: Medical devices Page 161 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Performance Surgical Mesh - Surgical Mesh - Absorbable Patches Tissue

parameter Gynaecological Hernia devices Adhesives

Pain VAS
post-
herniorrhaphy
pain
questionnaire
McGill pain
Questionnaire
Inguinal Pain
Questionnaire
Cunningham
classification of
post-
herniorrhaphy
pain

Clearance Days to
clear the
body, days
metabolised,
excretion
route

References
US Food & Drug Administration. Hernia Surgical Mesh Implants; 2018.
US Food & Drug Administration. Urogynecologic Surgical Mesh Implants; 2021.
Persu C, Chapple CR, Cauni V, Gutue S, Geavlete P. Pelvic Organ Prolapse Quantification system
(POP–Q) – a new era in pelvic prolapse staging. Journal of Medicine and Life. 2011;4(1):75-81.
Furnee E, Hazebroek E. Mesh in laparoscopic large hiatal hernia repair: a systematic review of
the literature. Surgical Endoscopy. 2013;27:3998–4008.
Reece TB, Maxey TS, Kron IL. A prospectus on tissue adhesives. The American Journal of Surgery.
2001;182(2, Supplement 1):40S-4S.
Providing yearly figures allows the estimation of cumulative percent incidence, or incidence as a
fraction of observation years. It is good practise for manufacturers to perform these calculations
for themselves on a regular basis. For implants such as surgical meshes, Kaplan - Meier analysis
is preferred. This sort of analysis is often performed with revision surgery (i.e. % requiring
revision at 1, 2, 3, etc. years from the time of primary surgery as an end point), but it can also be
performed on many other endpoints.
Brown CN, Finch JG. Which mesh for hernia repair? Annals of The Royal College of Surgeons of
England. 2010;92(4):272-8
Dwyer P. Evolution of biological and synthetic grafts in reconstructive pelvic surgery.
International Urogynecological Journal. 2006;17:S10-S5

Clinical evidence guidelines: Medical devices Page 162 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Bjelic-Radisic V, Aigmueller T, Preyer O et al. Vaginal prolapse surgery with transvaginal mesh:
results of the Austrian registry. International Urogynecology Journal. 2014;25(8):1047-52
Nilsson EK, Haapaniemi S. The Swedish hernia register: an eight year experience. Hernia.
2000;4:286-9
Stechemesser B, Jacob DA, Schug-Pass C, Kockerling F. Herniamed: an internet-based registry for
outcome research in hernia surgery. Hernia. 2012;16(3):269-76
Abdominal Core Health Quality Collaborative Foundation. Abdominal Core Health Quality
Collaborative; 2021.
EuraHS. European Registry of Abdominal Wall Hernias (EuraHS).
Club Hernie. ClubHernie; 2021.
Jia X, Glazener C, Mowatt G et al. Systematic review of the efficacy and safety of using mesh in
surgery for uterine or vaginal vault prolapse. Int Urogynecol J. 2010;21(11):1413-31
Aslani N, Brown CJ. Does mesh offer an advantage over tissue in the open repair of umbilical
hernias? A systematic review and meta-analysis. Hernia. 2010;14:455-62
Hobart WH. Clinical Outcomes of Biologic Mesh. Surgical Clinical of North America.
2013;93(5):1217-25
Bellows CF, Smith A, Malsbury J, Helton WS. Repair of incisional hernias with biological
prosthesis: a systematic review of current evidence. American Journal of Surgery.
2013;205(1):85-101.
Bump RC, Mattiasson A, Bø K et al. The standardization of terminology of female pelvic organ
prolapse and pelvic floor dysfunction. American Journal of Obstetrics and Gynecology.
1996;175(1):10-7
Lamb ADG, Robson AJ, Nixon SJ. Recurrence after totally extra-peritoneal laparoscopic repair:
implications for operative technique and surgical training. Surgeon. 2006;4(5)
Salamon CG, Culligan PJ. Subjective and objective outcomes one year after robotic-assisted
laparoscopic sacrocolpopexy. Journal of Robotic Surgery. 2012;7(1):35-8
Ferroli P, Acerbi F, Broggi M et al. A novel impermeable adhesive membrane to reinforce dural
closure: a preliminary retrospective study on 119 consecutive high-risk patients. World
Neurosurgery. 2013;79(3-4):551-7
von der Brelie C, Soehle M, Clusmann HR. Intraoperative sealing of dura mater defects with a
novel, synthetic, self adhesive patch: application experience in 25 patients. British Journal of
Neurosurgery. 2012;26(2):231-5.
Kaul A, Hutfless S, Le H et al. Staple versus fibrin glue fixation in laparoscopic total
extraperitoneal repair of inguinal hernia: a systematic review and meta-analysis. Surgical
Endoscopy. 2012;26:1269–78.
Topart P, Vandenbroucke F, Lozac'h P. Tisseel vs tack staples as mesh fixation in totally
extraperitoneal laparoscopic repair of groin hernias. Surgical Endoscopy. 2005;19:724-7.
Schwab R, Willms A, Kröger A, Becker HP. Less chronic pain following mesh fixation using a
fibrin sealant in TEP inguinal hernia repair. Hernia. 2006;10(3):272-7.
Sanjay P, Watt DG, Wigmore SJ. Systematic review and meta-analysis of haemostatic and
biliostatic efficacy of fibrin sealants in elective liver surgery. Journal of Gastrointestinal Surgery.
2013;17(4):829-36
Therapeutic Goods Administration. AusPAR: Fibrin adhesive/sealant. Australian Public
Assessment Report 2014 [cited 29 Oct 2014]

Clinical evidence guidelines: Medical devices Page 163 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Software as Medical Device

Introduction
Software use is increasingly common in medical devices. The guidance provided in this chapter
describes the clinical evidence that should be provided for devices in which the claim of clinical
benefit is attributed to the software itself. ( Software that is used to support the intended
purpose of the device is not covered in this chapter. )
In the context of software, clinical benefit may be considered slightly differently from that for
pharmaceuticals or other medical devices. The clinical benefit of software as a medical device
most commonly lies in obtaining or collating clinical information which assists with clinical
decision making. If a software device is making the claim of clinical benefit, this must be
supported by clinical evidence including safety, effectiveness and performance data.

Summary recommendations
The TGA is a member for the International Medical Device Regulators forum and follows the
recommendations for evaluation outlined at
Software as a Medical Device (SaMD): Clinical Evaluation (IMDRF/SaMD
WG/N41): https://www.imdrf.org/documents/software-medical-device-samd-clinical-
evaluation
and
Guidance on Clinical Evaluation (MDR) / Performance Evaluation (IVDR) of Medical Device
Software (MDCG 2020-1): https://ec.europa.eu/health/system/files/2020-
09/md_mdcg_2020_1_guidance_clinic_eva_md_software_en_0.pdf

Glossary and abbreviations

Glossary
Adverse event: Any untoward medical occurrence in patients/subjects, users or other persons.
In the context of clinical investigation, for patients/subjects, this would include all untoward
medical occurrences, whether or not related to the device that is the subject of the investigation,
that occurred in the course of the investigation. In the context of clinical experience, this would
only include untoward medical occurrences that may be related to the medical device.
Application audit: The Act enables the Regulations to prescribe certain kinds of applications
that are to be selected for audit. These kinds of applications must be selected for audit by the
Secretary. However, the Secretary may also select for auditing any other application under
section 41FH of the Act. The TGA has established two levels of application audit, Level 1 and
Level 2:
• Level 1: Targeted for completion within 30 days - The TGA will consider:
– the original or correctly notarised copy of the manufacturer’s Australian Declaration of
Conformity
– Copy of the latest and current conformity assessment evidence for the medical device
– Information about the device, including copies of the label, instructions for use and
advertising material such as brochures, web pages and advertisements

Clinical evidence guidelines: Medical devices Page 164 of 178

V3.1 June 2022
 Therapeutic Goods Administration

• Level 2: Targeted for completion within 60 days – The TGA will consider all of the
documentation considered in a Level 1 audit. In addition, the TGA will consider:
– the risk management report
– the clinical evaluation report
– efficacy and performance data for medical devices that disinfect including those that
sterilise other medical devices.
Assessor: A medically qualified person who reviews the clinical evaluation report and
supporting documents provided to the TGA with applications for inclusion, review of conformity
assessment procedures and post-market reviews of medical devices.
Australian Register of Therapeutic Goods (ARTG): The ARTG is the register of information
about therapeutic goods for human use that may be imported, supplied in or exported from
Australia. All medical devices, including Class I, must be included in the ARTG before supply in
Australia. There are limited exceptions to this requirement specified in the legislation.
Biological characteristics: Relate to use of the materials or substances in contact with the same
human tissues or body fluids. Evaluators should consider biological safety (e.g. in compliance to
ISO 10993) as well as other aspects necessary for a comprehensive demonstration of
equivalence. A justification explaining the situation should be provided for any difference.
Clinical data: Safety and/or performance information that is generated from the clinical use of a
device.
Note: Under the clinical evaluation procedures in Part 8 of Schedule 3 of the MD Regulations, the
manufacturer must obtain clinical data in relation to the device in the form of clinical
investigation data or a literature review, or both.
Clinical evaluation: A set of ongoing activities that use scientifically sound methods for the
assessment and analysis of clinical data to verify the safety and/or performance of a medical
device when used as intended by the manufacturer.
Note: the clinical evaluation procedures (in the MD Regulations) set out requirements in relation
to the obtaining and evaluation of clinical data.
Clinical evidence: The clinical data and the clinical evaluation pertaining to a medical device.
Note: EP 14 provides that every medical device requires clinical evidence demonstrating that the
device complies with applicable EPs.
Competent clinical expert: Generally expected to be someone with relevant medical
qualifications and direct clinical experience in the use of the device or device type in a clinical
setting.
Note: the clinical evaluation procedures (in the MD Regulations) require the manufacturer to
ensure that the clinical data is evaluated by competent clinical experts.
Clinical Evaluation Report (CER): A report by an expert in the relevant field outlining the
scope and context of the evaluation, the inputs (clinical data), appraisal and analysis stages, and
conclusions about the safety and performance of the device. The clinical evaluation report
should be signed and dated by the clinical expert.
Clinical investigation: Systematic investigation or study in one or more human subjects,
undertaken to assess the safety and/or performance of a medical device.
Note: 'clinical trial' or ' clinical study' is synonymous with 'clinical investigation’ and these terms
are used interchangeably in this document.

Clinical evidence guidelines: Medical devices Page 165 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Clinical investigation data: Safety and/or performance information that is generated from the
use of a medical device (based on the definition above this information is generated in or on one
or more human subjects).
Clinical investigation plan: Document that states the rationale, objectives, design and pre-
specified analyses, methodology, monitoring, conduct and record-keeping of the clinical
investigation.
Clinical performance: The ability of a medical device to achieve its intended clinical purpose as
claimed by the manufacturer.
Clinical safety: Acceptability of risks as weighed against benefits, when using the medical
device according to the manufacturer’s labelling.
Clinical use: Use of a medical device in or on living human subjects.
Note: Includes use of a medical device that does not have direct patient contact.
Comparable device: A medical device with related function chosen by the manufacturer to
inform the clinical evaluation of the device in question.
Note: A ‘comparable device’ is distinct from a ‘comparator’, which is the state of the art/standard
of care against which a medical device may be compared (for example, in a clinical study).
Conformity Assessment: The systematic examination of evidence generated and procedures
undertaken by the manufacturer, under requirements established by the Regulatory Authority,
to determine that a medical device is safe and performs as intended by the manufacturer and,
therefore, conforms to the Essential Principles.
Conformity assessment is the name given to the processes that are used to demonstrate that a
device and manufacturing process meet specified requirements. In Australia this means that the
manufacturer must be able to demonstrate that both the medical device and the manufacturing
processes used to make the device conform to the requirements of the therapeutic goods
legislation.
Conformity assessment is the systematic and ongoing examination of evidence and procedures
to ensure that a medical device complies with the Essential Principles. It provides objective
evidence of the safety, performance, benefits and risks for a specified medical device and also
enables regulatory bodies to ensure that products available in Australia conform to the
applicable regulatory requirements.
The Conformity Assessment Procedures allow risk based premarket assessment for devices. All
manufacturers of all medical devices are required to meet manufacturing standards and all
manufacturers, except those manufacturing the lowest risk devices, are audited and are required
to have their systems certified. The level of assessment is commensurate with the level and
nature of the risks posed by the device to the patient, ranging from manufacturer self-
assessment for low risk devices through to full TGA assessment with respect to high-risk
devices.
Conformity assessment certificate: A certificate to demonstrate that the conformity
assessment procedure has been assessed.
Critical analysis: The process of the careful and systematic examination, appraisal and
evaluation of both favourable and unfavourable data.
Essential Principles: The Essential Principles (EPs) provide the measures for safety and
performance and are set out in Schedule 1 of the MD Regulations. For a medical device to be
supplied in Australia, it must be demonstrated that the relevant EPs have been met.
Hazard: Potential source of harm.
Clinical evidence guidelines: Medical devices Page 166 of 178
V3.1 June 2022
 Therapeutic Goods Administration

Incident: Any malfunction or deterioration in the characteristics and/or performance of a

device, as well as any inadequacy in the labelling or the instructions for use which, directly or
indirectly, might lead to or might have led to the death of a patient, or user or of other persons or
to a serious deterioration in their state of health.
Indications for use: The disease or condition the device will diagnose, treat, prevent, cure or
mitigate, including a description of the patient population for which the device is intended.
Intended purpose: The objective intent of the manufacturer regarding the use of a product,
process or service as reflected in the specifications, instructions and information provided by
the manufacturer.
In-Vitro Diagnostic (IVD): A medical device is an IVD if it is a reagent, calibrator, control
material, kit, specimen receptacle, software, instrument, apparatus, equipment or system,
whether used alone or in combination with other diagnostic goods for in vitro use. It must be
intended by the manufacturer to be used in vitro for the examination of specimens derived from
the human body, solely or principally for the purpose of giving information about a physiological
or pathological state, a congenital abnormality or to determine safety and compatibility with a
potential recipient, or to monitor therapeutic measures. The definition of an IVD does not
encompass products that are intended for general laboratory use that are not manufactured,
sold or presented for use specifically as an IVD.
Kind of medical device: A single entry in the ARTG may cover a range of products that are of
the same kind rather than individual devices. At present, medical devices (with the exception of
Class III and Active Implantable Devices (AIMDs) and Class 4 IVDs and Class 4 in-house IVDs) are
included as a group in the ARTG under a single entry if they: have the same sponsor; have the
same manufacturer; have the same medical device classification; have the same nomenclature
system code (GMDN code).
Manufacturer: Refer to section 41BG of the Act.
Medical device: A medical device is:
(a) any instrument, apparatus, appliance, material or other article (whether used alone or in
combination, and including the software necessary for its proper application) intended, by
the person under whose name it is or is to be supplied, to be used for human beings for the
purpose of one or more of the following:
i. diagnosis, prevention, monitoring, treatment or alleviation of disease;
ii. diagnosis, monitoring, treatment, alleviation of or compensation for an injury or
disability;
iii. investigation, replacement or modification of the anatomy or of a physiological
process or state;
iv. control or support of conception;
v. in vitro examination of a specimen derived from the human body for a specific
medical purpose;
and that does not achieve its principal intended action in or on the human body by
pharmacological, immunological or metabolic means, but that may be assisted in its
function by such means; or
(aa) any instrument, apparatus, appliance, material or other article specified under subsection
(2A); or

Clinical evidence guidelines: Medical devices Page 167 of 178

V3.1 June 2022
 Therapeutic Goods Administration

(ab) any instrument, apparatus, appliance, material or other article that is included in a class of
instruments, apparatus, appliances, materials or other articles specified under subsection
(2B); or
(b) an accessory to an instrument, apparatus, appliance, material or other article covered by
paragraph (a), (aa) or (ab).
Refer to section 41BD of the Act for remainder of the definition.
Medical device classifications: Medical devices are classified according to their risk. The
device classifications are determined using a set of rules contained in the Regulations that take
into account the degree of invasiveness in the human body, the duration and location of use and
whether the device relies on a source of energy other than the body or gravity. There are two
sets of classification rules; one based on the above and the other based on IVDs as medical
devices.
Predicate: A previous iteration of the device, within the same lineage of devices, with the same
intended purpose and from the same manufacturer, in relation to which a manufacturer is
seeking to demonstrate substantial equivalence.
Post-market surveillance: Once a device has been included in the ARTG, the sponsor has
ongoing responsibilities. These include monitoring and reporting to the TGA adverse events,
vigilance reports, complaints, performance issues and regulatory actions in other jurisdictions.
Please refer to Sections 22 and 23 of the ARGMD.
Risk: Combination of the probability of occurrence of harm and the severity of that harm.
Risk management: Systematic application of management policies, procedures and practices to
the tasks of analysing, evaluating, controlling and monitoring risk.
Serious adverse event: An adverse event that led to a death or led to a serious deterioration in
health (one that results in a life-threatening illness or injury; results in a permanent impairment
of a body structure or body function; requires inpatient hospitalisation or prolongation of
existing hospitalisation; results in medical or surgical intervention to prevent permanent
impairment to body structure or a body function; led to foetal distress, foetal death or a
congenital abnormality/ birth defect).
Sponsor: Refer to Section 3 of the Act.
Substantial equivalence: A finding that comparable devices are similar to such an extent that
there would be no clinically significant difference in safety and performance, taking into account
the intended purpose and clinical, technical and biological characteristics of the devices.
Technical characteristics: These relate to the design, specifications, physicochemical
properties including energy intensity, deployment methods, critical performance requirements,
principles of operation and conditions of use.

Clinical evidence guidelines: Medical devices Page 168 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Abbreviations
Abbreviation Meaning

AAA Abdominal aortic aneurysm

AE Adverse event

AICD Active implantable cardiac device

AIMD Active implantable medical device

AMSTAR Assessing the Methodological Quality of Systematic Reviews

ARGMD Australian Regulatory Guidelines for Medical Devices

ARTG Australian Register of Therapeutic Goods

ASERNIP-S Australian Safety and Efficacy Register of New Interventional Procedures –

Surgical (Royal Australasian College of Surgeons)

AOANJRR Australian Orthopaedic Association National Joint Replacement Registry

BMS Bare metal stent

BSIR British Society of Interventional Radiology

CAPA Corrective and preventive action

CE Conformité Européenne (European Conformity)

CEBM Centre for Evidence-Based Medicine

CDMSNet Canadian Medical Devices Sentinel Network

CDRH Center for Devices and Radiological Health [USA]

CER Clinical Evaluation Report

CNS Central Nervous System

CONSORT Consolidated Standards of Reporting Trials

CPR Cumulative Percent Revision

CRT Cardiac Resynchronisation Therapy

CSR Clinical Study Report

Clinical evidence guidelines: Medical devices Page 169 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Abbreviation Meaning

CTA Computed tomography angiography

D&B Downs & Black [quality assessment tool]

DES Drug-eluting stent

DVT Deep vein thrombosis

EHR Electronic Health Record

EPs Essential Principles

EU European Union

EUDAMED European Databank on Medical Devices

FDA Food and Drug Administration [USA]

GCP Good Clinical Practice

GHTF Global Harmonization Task Force

GMDN Global Medical Device Nomenclature [System]

HBD Harmonisation By Doing

HDE Humanitarian device exemption

ICMJE International Committee of Medical Journal Editors

ICD Implantable Cardioverter Defibrillator

ICU Intensive Care Unit

IDE Investigational Device Exemption

IDEAL Innovation, Development, Exploration, Assessment, Long-term study

[Collaboration]

IFU Instructions For Use

ILR Implantable Loop Recorder

IMDRF International Medical Device Regulators Forum

IRIS Medical device Incident Reporting and Investigation Scheme (TGA)

Clinical evidence guidelines: Medical devices Page 170 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Abbreviation Meaning

ISO International Standards Organization

IVD In vitro diagnostic

KAT Knee Arthroplasty Trial

LOHS Length of hospital stay

MA Meta-analysis

MACE Major adverse cardiac events

MAUDE Manufacturer and User Facility Device Experience database

MCID Minimum clinically important difference

MDPS Medical device production systems

MD Therapeutic Goods (Medical Devices) Regulations 2002

Regulations

MDR Medical Device Reporting (Program) [USA]

MedSun Medical Device Surveillance Network [USA]

MHLW Ministry of Health, Labour & Welfare [Japan]

MHRA Medicines and Healthcare Products Regulatory Authority [UK]

MI Myocardial infarction

MLHF Minnesota Living with Heart Failure Questionnaire

MPMDB Marketed Pharmaceutical and Medical Devices Bureau [Canada]

MR Magnetic Resonance

MRI Magnetic Resonance Imaging

NB Notified Body [EU]

NCAR National Competent Authority Report

NHMRC National Health and Medical Research Council

NHS National Health Service [UK]

Clinical evidence guidelines: Medical devices Page 171 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Abbreviation Meaning

NICE National Institute for Heath and Care Excellence

NOS Newcastle-Ottawa scale [quality assessment tool]

NR Not Reported

NYHA New York Heart Association Classification

OPC Objective Performance Criteria

PAL Pharmaceutical Affairs Law [Japan]

PCT Pacing Capture Threshold

PDA Patent Ductus Arteriosus

PE Pulmonary Embolus

PMA/PMAS Pre-Market Approval or Pre-Market Approval Supplement [USA]

PMCF Post-Market Clinical Follow-up

PMD Personalised medical device

PMDA Pharmaceuticals and Medical Devices Agency [Japan]

PPM Permanent Pacemaker

PRISMA Preferred Reporting Items for Systematic reviews and Meta-Analyses

PMS Post-market Surveillance

POP-Q Pelvic Organ Prolapse Quantification System

QMS Quality Management System

QOL Quality Of Life

QUADAS Quality Appraisal of Diagnostic Accuracy Studies

RANZCR Royal Australian and New Zealand College of Radiologists

RCT Randomized controlled trial

RF Radiofrequency

Clinical evidence guidelines: Medical devices Page 172 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Abbreviation Meaning

RIND Reversible Ischemic Neurological Deficit

RSA Radiostereometric analysis

RWD Real World Data

RWE Real World Evidence

SAR Specific Absorption Rate

SD Standard Deviation

SIGN Scottish Intercollegiate Guidelines Network

SSO Surgical Site Occurrence

SR Systematic Review

STARD Standards for Reporting of Diagnostic Accuracy

STED Summary Technical Document

STROBE Strengthening the Reporting of Observational Studies in Epidemiology

TIA Transient Ischemic Attack

TGA Therapeutic Goods Administration

TLR Target Lesion Revascularisation

TPLC Total product life cycle

TVR Total Vessel Revascularisation

UK United Kingdom

USA United States of America

VARC Valve Academic Research Consortium

VHWG Ventral Hernia Working Group

VTE Venous Thromboembolism

WHO World Health Organisation

Clinical evidence guidelines: Medical devices Page 173 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Source material

Search Method: Identification and selection of clinical

studies
The search method used by ASERNIP-S to identify, retrieve and review the evidence that
supports this guidance document was a pragmatic adaption of a rapid systematic review. 11
Selection criteria were established a priori and include publication type, type of medical device
under review, intended purpose of the medical device, adverse events (safety), and clinical
outcomes related to device performance.
Using the PubMed Clinical queries tool, representative articles were identified through scoping
searches. These results informed the selection of appropriate text words and subject headings.
All searches were executed using the Ovid platform for Medline, Embase and the Cochrane
library and Evidence-based medicine databases. Through the application of search filters for
study type (Systematic Reviews, Randomised Controlled Trials and Registry trials) search
results were restricted to appropriate level evidence.
For joint prostheses - Search title: Total and partial joint arthroplasty: Search terms: Shoulder
surgery[MeSH], Shoulder Joint/surgery[MeSH]; Knee surgery [MeSH]; ‘knee Joint’
surgery[MeSH]; hip Joint surgery [MeSH]; arthroplasty[MeSH]; ((shoulder OR hip OR Knee) adj2
replacement). [text word]; ((shoulder OR hip OR Knee) adj2 joint).[text word]; ((shoulder OR
hip OR Knee) adj3 arthroplasty)[text word]; ((shoulder OR hip OR Knee) adj3 surger?)[text
word].
For cardiovascular devices for patency and functional flow- Search title: Cardiovascular devices
for patency and functional flow: Search terms: Heart [MeSH]; aneurysm [MeSH]; aorta[MeSH];
Venae cavae[MeSH]; ‘Ductus Arteriosus, Patent’[MeSH]; vascular.[text word]; endovascular.[text
word]; cardiovascular. [text word]; heart.[text word]; cardiac.[text word]; ; ‘vena cava’.[text
word]; aorta.[text word]; ‘Patent ductus arteriosus’.[text word] ; aneurism.[text word]
Selected CV flow implants included the following types:
• Arterial stents (carotid, coronary and peripheral)
• Implants for AAA repair
• Implants for PDA repair
For implantable pulse generators - Search title: electrical impulse generators: Search terms:
Pacemaker, Artificial[MeSH]; Biological Clocks[MeSH]; Tachycardia, Ectopic Atrial[MeSH];
implantable cardioverter-defibrillators[MeSH]; Defibrillators, Implantable[MeSH]; Tachycardia,
Ventricular[MeSH]; Ventricular Fibrillation[MeSH]; Pain Management[MeSH]; Postoperative
pain[MeSH]; Analgesia, Patient-Controlled[MeSH]; Magnetic Field Therapy[MeSH]
Selected implantable pulse generators of the following types:
• Active Implantable Cardiac Devices (AICD) including:
– single and dual chamber pacemakers
– cardiac resynchronisation therapy pacemakers, with or without defibrillation (i.e. CRT-
D and CRT respectively)

11 Watt A, Cameron A, Sturm L et al. Rapid reviews versus full systematic reviews: an inventory of current

methods and practice in health technology assessment. Int J Technol Assess Health Care. 2008;24:133-9.
Clinical evidence guidelines: Medical devices Page 174 of 178
V3.1 June 2022
 Therapeutic Goods Administration

– implantable cardiac defibrillators (ICDs)

• Electrical nerve stimulation devices
For heart valve prostheses- Search title: Heart valve replacement using a prosthetic valve:
Search terms: Heart valve prosthesis [MeSH]; heart valve prosthesis implantation[MeSH]; (valv$
adj3 prosthe$).[text word]; (valv$ adj3 bioprosthe$).[text word]; (artificial adj3 valv$).[text
word]; (mechanical adj3 valv$).[text word]; (bioprosthe$ OR prosthe$ OR mechanical).[text
word]; (aortic adj3 valv$).[text word]; (mitral adj3 valv$).[text word]; (pulmon$ adj3valv$).[text
word].
For supportive devices- Search title: Supportive devices – meshes, patches and tissue adhesives:
Search terms: Surgical mesh [MeSH]; Bioabsorbable Implants; Absorbable Implants [MeSH];
Coated Materials. Biocompatible [MeSH]; Tissue scaffolds [MeSH]; Tissue adhesives [MeSH];
Fibrin Tissue Adhesive [MeSH]; Blood patch, Epidural [MeSH].
For active implantables in the magnetic resonance environment - Search title: safety of active
implantables in the magnetic resonance environment: Search terms: Magnetic Resonance
Imaging[MeSH]; magnetic resonance [text word]; MRI [text word]; MR [text word]; Cardiac
Pacing, Artificial[MeSH]; Pacemaker, Artificial[MeSH]; defibrillators, implantable[MeSH];
safe*[text word]; performance [text word]; efficacy [text word]; heat* [text word]; scar*[text
word]; burn*[text word]; artefact* [text word]; ;dislodge*[[text word]; interference [text word];
ICD [text word]; Defibrillator [text word]; pacemaker* [text word] ; resynch* [text word];
cardiac monitor [text word] ; loop recorder [text word]; ICM [text word].
A focused internet search was conducted to identify recent and relevant legislation, current
guidance documents and other standards/documents to assist in the compilation and
presentation of clinical evidence. Only documents that are publicly available to manufacturers
were included. For regulatory documents, the scope of the search was confined to Australia and
the comparable jurisdictions of Canada, the EU/UK, Japan and the USA.
Evidence from both the targeted internet searches and peer reviewed literature focused on
study designs that are based on solid scientific principles which generate clinical evidence on the
safety and performance of the device. Such evidence sources include, but are not limited to,
controlled clinical trials, case control studies, case series and post-market registry data.
Summaries of exemplar articles documenting clinical research on the safety and performance of
the device types have also been presented. Reports were selected based on recency and
relevance and to be representative of those currently used in clinical practice in Australia.
Searches were restricted to English language articles published between January 2009 and June
2014 with updates for some topics to January 2015. All citations were retrieved and initial
selection was based on title and abstract with potentially relevant articles retrieved in full text
for final selection.

Clinical evidence guidelines: Medical devices Page 175 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Identified study designs

Based on the NHMRC levels of evidence study, designs used to evaluate the safety and
performance of high risk medical devices range from systematic reviews of RCTs to case-series
reports (Level IV). Irrespective of level of evidence the quality of reporting varied from low to
high as assessed by validated quality tools.
In summary, the clinical evidence in this document includes:
• systematic reviews of RCTs, comparative cohort trials and cases-series
– given the diversity of included evidence these systematic reviews do not meet the Level
I classification as prescribed by the NHMRC
• RCTs (Level II)
– when practical, this should be the preferred study design
– clinical trials of a RCT design are reported for the high risk devices and included in the
evidence base
• observational studies (Level III)
– these are a valid alternative to RCTs 12 provided appropriate matching of treatment
groups is performed, e.g. through the application of propensity scores 13
• case series (Level IV)
– these can inform on the safety and performance of the high risk devices and have a high
sensitivity for adverse events
• post-market registries
– these are established for some of the high risk devices and provide a valuable resource
for post-market safety and performance data from other jurisdictions that can be used
to support a pre- or post-market review of safety and performance of a high risk device

12 Mann CJ. Observational research methods. Research design II: cohort, cross sectional, and case-control

studies. Emergency medicine journal. 2003;20:54-60

13 Yue LQ. Statistical and regulatory issues with the application of propensity score analysis to

nonrandomized medical device clinical studies. J Biopharm Stat. 2007;17:1-13; discussion 5-7, 9-21, 3-7
passim
Yue LQ. Regulatory considerations in the design of comparative observational studies using propensity
scores. J Biopharm Stat. 2012;22:1272-9
Clinical evidence guidelines: Medical devices Page 176 of 178
V3.1 June 2022
 Therapeutic Goods Administration

Version history
Version Description of change Author Effective date

V1.0 Original publication Medical Devices Branch, February 2017

Therapeutic Goods
Administration

V1.1 Minor updates to reflect CTA Biological Science Section November 2020
name change

V2.0 Updated section on Comparable Devices Clinical Section March 2021

devices including substantially
equivalent devices.
Updated links and footnotes.

V3.0 Updated Part 1 – General Devices Clinical Section November 2021

Requirements and overall
structure.
Updated links.

V3.1 Updated Part 1 – General Devices Clinical Section June 2022

Requirements – The Essential
Principles
New chapter - Personalised
medical devices (PMDs)
Updated - Total and partial joint
prostheses
New chapter - Software as
Medical Device
Updated - Abbreviations

Clinical evidence guidelines: Medical devices Page 177 of 178

V3.1 June 2022
 Therapeutic Goods Administration

Therapeutic Goods Administration

PO Box 100 Woden ACT 2606 Australia

Email: [email protected] Phone: 1800 020 653 Fax: 02 6203 1605
https://www.tga.gov.au

D21-3241725