See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/353301530
Development of Tuberculosis Spine Instability Score (TSIS): An Evidence-
Based and Expert Consensus-Based Content Validation Study Among Spine
Surgeons
Article in Spine · July 2021
DOI: 10.1097/BRS.0000000000004173
CITATIONS READS
10 1,709
10 authors, including:
Kaustubh Ahuja Pankaj Kandwal
All India Institute of Medical Sciences Rishikesh All India Institute of Medical Sciences Rishikesh
65 PUBLICATIONS 327 CITATIONS 125 PUBLICATIONS 754 CITATIONS
SEE PROFILE SEE PROFILE
Syed Ifthekar Abhay Nene
Aiims rishikesh 78 PUBLICATIONS 966 CITATIONS
47 PUBLICATIONS 190 CITATIONS
SEE PROFILE
SEE PROFILE
All content following this page was uploaded by Abhay Nene on 20 September 2021.
The user has requested enhancement of the downloaded file.
SPINE An International Journal for the study of the spine, Publish Ahead of Print
DOI: 10.1097/BRS.0000000000004173
Development of Tuberculosis Spine Instability Score (TSIS): An evidence-based and
expert consensus-based content validation study among spine surgeons
Kaustubh Ahuja1,* MS, Pankaj Kandwal1,* MS, Syed Ifthekar1 DNB, P. Venkata Sudhakar1
MS, Abhay Nene2,3,4,5,6MS, Saumyajit Basu7,8MS, FRCS, Ajoy Prasad Shetty9 MS, Shankar
Acharya10 MS, MCh, FRCS, Harvinder Singh Chhabra11 MS, Arvind Jayaswal12 MS
1
Department of Orthopaedic Surgery, All India Institute of Medical Sciences, Rishikesh,
Uttarakhand, India
2
Department of Spine, Wockhardt Hospital, Mumbai, India
3
Department of Orthopaedics, Hinduja Healthcare Surgical, Mumbai, India
4
Department of Orthopaedic Surgery, Lilavati Hospital & Research Centre, Mumbai, India
5
Department of Orthopaedics, Breach Candy Hospital Trust, Mumbai, India
6
Department of Orthopedics, Wadia Children’s Hospital, Mumbai, India
7
Department of Spine Surgery, Kothari Medical Center, Kolkata, India
8
Department of Spine Surgery, Park Clinic, Kolkata, India
9
Department of Spine Surgery, Ganga Hospital, Coimbatore, India
10
Department of spine surgery, Sir Gangaram hospital, New Delhi, India
11
Department of Spine Services, Indian Spinal Injuries Center, Vasant Kunj, New Delhi,
India
12
Department of Orthopaedics, Primus Superspeciality Hospital, New Delhi, India
*
Dr. Kaustubh Ahuja and Dr. Pankaj Kandwal share first authorship
Address correspondence and reprint requests:
Dr. Pankaj Kandwal
Additional Professor and Head,
Department of Orthopaedic Surgery,
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
All India Institute of Medical Sciences,
Rishikesh-249203
Phone: +918475000282
E-mail: [email protected]
[email protected]
Mini-abstract
TSIS is a comprehensive scoring criteria to diagnose instability in TB spine and comprises of
age, location of the lesion, pain, degree of kyphosis, vertebral body loss, involvement of
posterior spinal elements, multifocal contiguous disease and presence of intervertebral or
para-spinal abscess.
Abstract
Study Design: An expert-panel consensus-based content validation and case-based clinical
validation study
Objective: To develop a novel scoring system for diagnosing instability in tuberculosis (TB)
spine using an expert-panel consensus followed by clinical validation for validating the
content.
Summary of Background Data: Currently, diagnosis of instability is primarily experience-
based which may lead to considerable variability and misdiagnosis in the hands of a relatively
in-experienced spine surgeon. Considering the potential complications this entity entails, a
universally accepted scoring criteria is very important for accurate and uniform diagnosis of
instability in TB spine.
Methods: The development of TB spine instability score (TSIS) followed a two-step process,
one designing the instrument and the other obtaining judgemental evidence. For judgemental
evidence a panel of experts was appointed to make appropriate modifications and content
validation for finalizing the scoring instrument. This score was applied on 30 patients of TB
spine and ROC curves were drawn for sensitivity and specificity analysis.
Results: The comprehensive scoring criteria to diagnose instability in TB spine was approved
after three rounds of expert panel discussions with an index of content validation more than
0.75 after final round of panel discussion. On case-based validation after plotting ROC
curves, sensitivity and specificity for diagnosing stable and potentially unstable lesions at a
cut-off score of 6 was 92.9% and 86.8% respectively whereas for diagnosing potentially
unstable and unstable lesions at a cut-off score of 10 was 94.3% and 81.9% respectively.
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Conclusions: TSIS is a comprehensive scoring system integrating demographic, anatomical,
clinical and radiological factors aimed at diagnosing instability in TB spine. The
classification determines indications for surgical stabilization in patients with TB spine, with
no or little neurological deficit.
Key Words: Tuberculosis, Spine infection, Instability, Content validation, receiver operating
characteristic, Index of qualitative variation
Level of Evidence: 4
The manuscript submitted does not contain information about medical device(s)/drug(s).
No funds were received in support of this work.
No relevant financial activities outside the submitted work.
Introduction
Surgical management in tuberculosis spine (TB spine) is commonly indicated in a setting of
persistence or worsening of clinical or neurological status despite a trial of Anti Tubercular
Treatment (ATT) or the presence of significant deformity[1, 2]. Additionally, spinal
instability also constitutes an important indication for surgical stabilization[2]. An unstable
spine is at a potential risk for development of a pathological fracture, translation or
dislocation leading to rapid onset neurological deficit[3]. Various clinical and radiological
parameters have been defined as predictors for spinal instability in literature. However, there
is no universally accepted definition for instability in the context of spinal TB.
A number of scoring systems have been described in literature to define instability in various
spinal pathologies such as trauma and spinal metastasis, to aid in surgical decision making[4,
5]. However, instability in TB spine differs from acute traumatic or neoplastic pathologies in
the pattern of spinal involvement, presence of simultaneous destruction and healing processes
and a unique pattern of neurological involvement as well as deformity necessitating a
different and specific scoring system to diagnose instability[6].
Currently, due to lack of well-defined guidelines, spine surgeons rely on clinical experience
to diagnose instability. The diagnosis of an unstable spine in TB is important in a situation
with no or little neurological deficit as these patients have a higher tendency to be managed
conservatively as opposed to patients with neurological deficit. Development of a simple yet
comprehensive system by incorporating clinical and radiological parameters would help in
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
establishing consistent treatment guidelines The current study aims to develop a scoring
system (TB spine instability score: TSIS) with the help of current evidence followed by an
expert consensus for the diagnosis of instability in TB spine. Further, we also seek to evaluate
clinical applicability and validation of the score in a cohort of patients of TB spine without
neurological deficit.
Patients and Methods
Instrument development
The development of TSIS followed a two-step process, one designing the instrument and the
other obtaining judgemental evidence (Fig 1) as described by Stein et al. and Armstrong et al
[7, 8].
Instrument Design- Domain specification and item identification. The authors performed a
systematic review for the first part of the study to identify various clinical and radiological
parameters used by various researchers across PUBMED and Embase to define instability in
TB spine[9]. After a thorough literature search, the authors identified the domains and items
that were deemed crucial in predicting instability in TB spine.
Instrument Construction- Based on the recommendations from Stone, the shortlisted domains
and the items were sent to an institutional advisory committee on TB spine in the form of a
questionnaire-based survey for feedback[10]. The advisory committee comprised of three
fellowship trained spine surgeons with different training backgrounds having substantial
expertise in managing TB spine. After incorporating the necessary modifications from the
responses, a round table meeting with the participation of the members of the advisory
committee was convened to draft a Preliminary Scoring Instrument (PSI) for defining
instability in TB spine.
Judgemental evidence- Expert consensus forms a crucial part of evidence-based medicine
especially in the absence of higher level of evidence[11]. For judgemental evidence, a panel
of experts representing different geographical locations and training backgrounds was
constituted. All the selected experts in the panel had substantial experience and research work
published in the field of TB spine. For the purpose of content validity, the recommended
number of experts range from 2 to 20[8]. To avoid chance agreement, five is the minimum
number of recommended experts[12]. The panel for the content validation in the present
study included eight experts.
The PSI was mailed to all the experts using the platform of google forms (Google Inc.,
California, US). Each sheet comprised of the item description and the proposed scoring
criteria. Each item was rated on a scale of 1 to 10 for relevance in predicting spinal
instability. The score was interpreted as: 0-4 = irrelevant; 5-6 = somewhat relevant; 7-8 =
quite relevant; 9-10 = highly relevant. Further, each expert submitted their recommendation
for the item in the form of: 1 = Accept; 2 = Modify; 3 = Reject. This was followed by an
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
open-ended short question where the experts gave reasons for rejection of the proposed
modifications for each item.
The modified scoring criteria along with suggested modifications was presented to the expert
panel in a web conference. Each item was discussed in a pre-decided format describing the
proposed and modified criteria till a consensus was reached. This was followed by
formulation of the final draft based on the expert consensus. The final scoring instrument was
further mailed to the experts electronically for their approval.
Content validity- For establishing content validity, index of content validity (CVI) was
calculated by dividing the number of quite relevant and highly relevant responses by total
number of responses for each item. CVI more than 0.75 was considered as cut-off for
excellent level of content validity. An item with the CVI value from 0.6 to 0.75 was
considered relevant but needing revisions whereas items with CVI less than 0.6 were
eliminated. A modified Kappa value was calculated with Kappa describing the agreement of
relevance using the formula κ = (CVI- Pc)/(1- Pc) where Pc denotes the probability of chance
occurrence[13]. Pc was calculated using the formula Pc = [N!/A!(N-A)!]* 0.5N where N =
number of experts and A = Number agreeing on quite or highly relevant, ! = a mathematical
symbol for the product of all positive integers less than or equal to N, for example 5! to mean
5x4x3x2x1. Relevance of an item was calculated using the guidelines as proposed by as
proposed by Cicchetti and Sparrow: Fair = k of 0.40 to 0.59; Good = k of 0.60 to 0.74; and
Excellent = k of 0.75 to 1.00 [14].
Clinical Validation
For clinical validation of the score, 30 patients of TB spine with no neurological deficit were
presented to 30 experts of TB spine across several hospitals in the form of clinical vignettes.
Each vignette comprised of patient’s clinical information including age, sex, Oswestry
disability index (ODI), visual analogue scale (VAS) and neurological status (Tuli and Kumar
grade 0 or 1)[15] and radiological images including representative radiographs and an MRI
comprising of dedicated sagittal and axial sections of the region of interest.
For selection of the experts, the national directory for spine surgeons was sought and the
experts were selected for the study only if they fulfilled the following criteria: (1) Five or
more publications pertaining to TB spine in indexed journals, (2) Chairpersons or active
members of various national and international spine societies, (3) Actively involved in
managing at least 100 TB spine patients annually and (4) Fellowship trained with at least 8
years of experience in spine surgery.
The cases were classified as either surgical or conservative by each expert on the basis of
stability of each lesion. The responses were further analysed using index of qualitative
variation (IQV). Cases with IQV less than 0.8 for surgical management were classified as
unstable, IQV less than 0.8 for conservative management were classified as stable whereas
cases with IQV more than 0.8 were classified as potentially unstable. The same was
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
considered as the gold standard. All the cases were further scored using TSIS by a two
blinded reviewers (one fellowship trained spine surgeon and one musculoskeletal radiologist
) and receiver operating characteristic (ROC) curves were plotted using average TSIS score
against the gold standard. Further, area under the ROC curve (AUC) were used to calculate
cut-off scores, sensitivity and specificity for the three categories.
Results
Instrument development
Based on the results of the systematic review and inputs obtained from the questionnaire-
based survey filled up by the institutional advisory committee, four major domains
(Demographic, Anatomical, Clinical and Radiological) and 10 items were shortlisted to
constitute the PSI. The scoring for radiological domain was done with help of MRI and
upright X rays. This PSI was subject to three rounds of discussions and two rounds of
revisions by the panel of experts to obtain a consensus and evaluate content validity. Of the
10 invited panellists, eight participated in all the three rounds of discussions and approved the
final score. At the end of round one survey, involvement of posterior spinal elements
(average score = 8.1), kyphotic deformity (average score = 7.9) and anatomical location of
the lesion (average score = 7.8) scored the maximum for the relevance in predicting
instability whereas patient expectations (average score = 5.1) scored the least and was
subsequently eliminated (Table 1). At the end of the final round, the approved scoring
instrument comprised of four domains and 8 items. Each item included in the expert panel
discussion is discussed below.
Domain: Demographic
Item: Age
With respect to the age of the patient, the PSI suggested two points for a patient with age less
than 15 and no points for age more than 15 years. The CVI for this item was found to be 0.63
(κ=0.53) and two modifications were suggested. After the second-round meeting, the scoring
criteria was modified to: less than 5 years = 3 points; 5-10 years = 2 points; 10-15 years = 1
point; more than 15 years = 0 point. At the end of round three, the CVI was found to be 0.87
(κ = 0.86) and the scoring criteria was accepted.
Domain: Anatomical
Item: Location of the lesion
The item takes into consideration the differences in the relative stability of different regions
of the spine. PSI proposed two points for junctional areas of the spine (Occiput to C2, C7-
T2, T10- L2, L5- S1), one point for semi-rigid spine (C3-C7 and L2-L5) and no points for the
rest of the spine. The CVI for this item was found to be 0.87 (κ=0.86) and the scoring criteria
was accepted without any further modifications.
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Domain: Clinical
Item: Pain
PSI proposed two points for patients experiencing mechanical pain on loading and movement
or 'instability catch' with relief on recumbency, one point for occasional pain and no points
for no pain. The CVI for pain was found to be 0.63 (κ=53). Based on the suggestions of the
first and second round meeting, a new category in the item was added scoring three points for
patients experiencing mechanical pain on loading and movement or 'instability catch' with no
relief even on recumbency. The consensus was reached at the end of the third round, CVI was
found to be 1 (κ=1) and the scoring criteria was accepted.
Domain: Radiological
Item: Kyphosis deformity
PSI proposed three points for kyphosis greater than 60º(measured with Cobb’s method), two
points for kyphosis between 30- 60º and one point for kyphosis between 10- 30ºand no points
for kyphosis less than 10º. Although this item had a high average relevance score (7.9),
however CVI for this item was found to be 0.63 (κ=0.53). Incorporation of the physiological
lordosis of the diseased segment into the deformity was suggested. Based on it, the item was
renamed as ‘Adjusted kyphosis deformity’ and was described as the calculated kyphosis
deformity adjusted after adding the expected physiological lordosis at the diseased segment
(five degrees per segment from C2 to C7 and L1 to L4 and 15 degrees per segment from L4-
S1)[16]. At the end of the third round CVI was found to be 1 (κ=1) and modified scoring
criteria was accepted.
Item: Vertebral Body Loss
PSI proposed two points for vertebral body loss more than 50 percent on either side of the
paradiscal lesion, one point for loss less than 50 percent and no points for no loss. The CVI
for vertebral body loss was found to be 0.63 (κ=0.53) at the end of round one. The suggested
modifications revolved around the ambiguity in the calculation of vertebral body loss. Based
on the discussions, the authors used a previously described technique to calculate vertebral
body loss by Rajasekaran et al.[17] and modified the scoring criteria to: more than 1.5 = 3
points; 0.75 to 1.5 = 2 points; less than 0.75 = 0 points. The CVI at the end of third round was
found to be 1(κ=1) and the scoring criteria was accepted.
Item: Involvement of posterior spinal elements
According to the proposed scoring criteria in the PSI, three points were given for pan-
vertebral involvement, two points for bilateral facet-joint involvement, one point for
unilateral facet joint involvement and no points for spared posterior elements. The item
scored the highest average relevance score (8.1). Modifications were suggested by two
experts and CVI was found to be 0.63 (κ=0.53). The accepted scoring criteria included three
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
points for bilateral facet joint involvement as evidenced directly on MRI or indirectly with
the help of radiological indicators, two points for unilateral facet joint destruction and no
points for spared posterior elements. Facet joint involvement in MRI was defined as presence
of joint destruction or soft tissue/ abscess around the joint or marrow edema of adjacent
bones. Scoliosis and antero-posterior or lateral translation on X rays were the radiological
indicators of facet joint destruction. At the end of third round, CVI was found to be 1 (κ=1).
Item: Multilevel involvement
With respect to multilevel involvement of the disease, PSI proposed three points for
involvement of three or more-disc spaces or spinal segments, two points for two-disc spaces
or spinal segments and one and zero points for one and no disc space involvement
respectively. Of eight, seven experts accepted the scoring criteria. CVI for the same was
found to be 0.75 (κ=0.72) and the scoring criteria was accepted without any modifications.
Item: Presence or absence of intraspinal/paraspinal abscess
The item was included to differentiate the impact of the nature of lesion on spinal stability.
PSI proposed one point for wet lesion as opposed to no points for a dry lesion. CVI for the
current item was found to be 0.63 (κ=0.53). The terminology for the current item was
changed to presence or absence of intervertebral or paraspinal abscess instead of wet/dry
lesion. At the end of third round meeting, the CVI was found to be 0.87 (κ=0.86) and
modified scoring criteria was accepted.
Clinical Validation
Of 30 patients, 6 were classified as unstable, 10 as stable whereas 14 were classified as
potentially unstable based on consensus expert opinion and the same was considered as
reference standard. The area under the ROC curve plotted to determine cut-off between stable
and potentially unstable lesions was found to be 0.96 (p<0.05) with a sensitivity and
specificity of 92.9% and 86.8% respectively at a cut-off score of 6 (Fig. 2). Similarly, for
ROC curve plotted including potentially unstable and unstable lesions, AUC was found to be
0.96 (p<0.05) with a sensitivity and specificity of 94.3% and 81.9% respectively at a cut-off
score of 10 (Fig. 3). All the 10 stable patients were found to have a TSIS score of 6 or less
(range – 3-6). All but two patients in the potentially unstable group were found to have scores
ranging from 7 to 10 (both false negatives had a score of 6). In the unstable group five out of
six patients had a TSIS score of 11 or more (one false negative with a TSIS score of 10).
Subsequently 6 and 10 were decided to be the cut-off scores for stable and potentially
unstable groups respectively.
TSIS score is calculated by adding up individual scores from each criterion. The minimum
and maximum possible scores are 0 and 21 respectively. Scores less than 7 indicate a stable
spine, between 7 to 10 denote ‘impending instability’ and scores greater than 10 denote
unstable spine and the possible need for surgical fixation (Table 2).
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
The authors have described the application of TSIS score in three cases (Fig.4, 5A-C, 6AB,
7A-C).
Discussion
Spinal instability in TB spine is particularly worrisome due to the risk of development of
rapid onset neurological deficit and the prognosis associated with it[3]. However, the concept
of spinal instability in TB spine is remarkably underdiscussed in literature [9, 18].
The current scoring system considers young age as an important predictor of spinal
instability. The immaturity and flexibility of a young spine renders spine unstable even in the
presence of minor deformity or destruction[17, 19, 20]. This instability may manifest as
gradual progression of deformity with age with eventual development of neurological deficit
or as sudden and rapid onset neurological deficit. Both the scenarios are known to have bad
prognosis with respect to neurological recovery[2, 21]. Moreover, paediatric patients with TB
spine usually present with higher degree of deformity and a greater tendency to collapse
during active phase of the disease[17, 20, 22].
Junctional regions of the spine are subject to transitional forces, placing them at a relatively
higher risk of instability and deformity progression[23]. Therefore, junctional regions are
given the highest score in the current classification followed by semi-rigid lumbar and
cervical non-junctional regions of spine. Dorsal spine which is relatively well supported by
the rib-cage and sacral segments below S1 were given the least score according to this
classification.
Mechanical back pain is a consistent feature of instability in any spinal pathology. Severe
back pain, paraspinal muscle spasm, painful restriction of movements and an ‘instability
catch’ is commonly used to clinically diagnose instability in TB spine[24, 25]. The current
classification gives a higher weightage to mechanical pain with no relief on recumbency as
compared to mechanical pain that gets relieved on recumbency as the former represents
higher severity and destruction due to the disease process.
Kyphotic deformity in TB spine constitutes an important indication for surgical stabilization.
Progressive failure of anterior column due to the disease process and repetitive loading may
eventually lead to the failure of a healthy posterior column resulting in spinal instability[20,
22]. The authors have given the physiological lordosis or cervical and lumbar vertebrae
special emphasis while calculating the degree of deformity and the entity was called ‘adjusted
kyphotic deformity’. For the ease of calculation and applicability, the current scoring criteria
adds five degrees for each involved cervical and lumbar vertebral segment or disc space till
L4. For L4-L5 and L5-S1 involvement 15 degrees was added to the calculated kyphosis to
calculate the adjusted kyphosis deformity[16].
Vertebral bodies play a major role in the axial load transmission during physiological loading
of the spine. TB spine, primarily being an anterior disease may lead to significant vertebral
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
body loss which may eventually lead to instability, deformity progression and neurological
complications.
Involvement of posterior elements or destruction of facet joints is a well-established
determinant for instability in all spinal pathologies. Destruction of facet joints leads to
excessive motion under physiological loads, thus placing spine and neural structures at risk.
Involvement of posterior elements can be directly visualized in MR imaging or indirectly by
radiological signs like scoliosis and translation.[26, 27].
Nature of the lesion is considered a determinant for spinal instability[28]. Wet lesions, as
determined by the presence of intervertebral or paraspinal abscess imparts more instability to
the spine as compared to dry or granulomatous presentation as seen in MR films[29].
The desirable characteristics of a good scoring or classification system should include ability
to aid in surgical decision making, reproducibility, content validity, reliability and
comprehensiveness[4]. Although TSIS comprehensively encompasses all the objective
criteria that predict instability in TB spine, it does not include subjective criteria which may
also play a role in surgical decision making. Presence of various medical co-morbidities, pre-
existing osseous disorders such as ankylosing spondylitis and osteoporosis and patient
expectations should be given emphasis before taking a decision for surgery[30]. TB spine
differs from other spinal disorders such as trauma, degenerative or neoplastic pathologies in
having concomitant healing response despite destruction by the tubercular pathology[31].
Therefore, it is prudent to keep in mind that TSIS is a dynamic score and may change with
time and healing of the disease. Finally, in cases of multilevel non-contiguous disease, TSIS
for each lesion should be calculated separately and should be treated as separate lesions. TSIS
is a tool designed to define presence of impending or frank instability in patients with TB
spine. Further studies aimed at evaluating inter- and intra-observer reliability in larger
multicentre trials are necessary to test the reproducibility and acceptability of the score
among spine surgeons.
Key Points
Spinal instability in tuberculosis (TB) spine constitutes an important indication for
surgical stabilization. However, there is no universally accepted definition for
instability in the context of spinal TB.
Lack of clear guidelines and good quality evidence may lead to considerable
misdiagnosis and variability in the management of unstable spinal TB.
In the present study, the authors developed a novel scoring system for diagnosing
instability in tuberculosis (TB) spine using evidence from pre-existing literature
followed by an expert-panel consensus for validating the content.
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
References
1. Tuli SM (2013) Historical aspects of Pott’s disease (spinal tuberculosis) management.
Eur Spine J 22:529–538. https://doi.org/10.1007/s00586-012-2388-7
2. Tuli SM (1975) Results of treatment of spinal tuberculosis by “middle-path” regime. J
Bone Joint Surg Br 57:13–23
3. Jain AK, Kumar J (2013) Tuberculosis of spine: neurological deficit. Eur Spine J
22:624–633. https://doi.org/10.1007/s00586-012-2335-7
4. Fisher CG, DiPaola CP, Ryken TC, et al (2010) A novel classification system for spinal
instability in neoplastic disease: an evidence-based approach and expert consensus from
the Spine Oncology Study Group. Spine 35:E1221-1229.
https://doi.org/10.1097/BRS.0b013e3181e16ae2
5. Lee JY, Vaccaro AR, Lim MR, et al (2005) Thoracolumbar injury classification and
severity score: a new paradigm for the treatment of thoracolumbar spine trauma. J
Orthop Sci 10:671–675. https://doi.org/10.1007/s00776-005-0956-y
6. Jain AK, Jain S (2012) Instrumented stabilization in spinal tuberculosis. Int Orthop
36:285–292. https://doi.org/10.1007/s00264-011-1296-5
7. Stein KF, Sargent JT, Rafaels N (2007) Intervention research: establishing fidelity of the
independent variable in nursing clinical trials. Nurs Res 56:54–62.
https://doi.org/10.1097/00006199-200701000-00007
8. Armstrong TS, Cohen MZ, Eriksen L, Cleeland C (2005) Content validity of self-report
measurement instruments: an illustration from the development of the Brain Tumor
Module of the M.D. Anderson Symptom Inventory. Oncol Nurs Forum 32:669–676.
https://doi.org/10.1188/05.ONF.669-676
9. Ahuja K, Ifthekar S, Mittal S, et al (2021) Defining mechanical instability in
tuberculosis of the spine: a systematic review. EFORT Open Rev.
https://doi.org/10.1302/2058-5241.6.200113
10. Stone DH (1993) Design a questionnaire. BMJ 307:1264–1266.
https://doi.org/10.1136/bmj.307.6914.1264
11. Guyatt G, Vist G, Falck-Ytter Y, et al (2006) An emerging consensus on grading
recommendations? BMJ Evid-Based Med 11:2–4. https://doi.org/10.1136/ebm.11.1.2-a
12. Zamanzadeh V, Ghahramanian A, Rassouli M, et al (2015) Design and Implementation
Content Validity Study: Development of an instrument for measuring Patient-Centered
Communication. J Caring Sci 4:165–178. https://doi.org/10.15171/jcs.2015.017
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
13. Rodrigues IB, Adachi JD, Beattie KA, MacDermid JC (2017) Development and
validation of a new tool to measure the facilitators, barriers and preferences to exercise
in people with osteoporosis. BMC Musculoskelet Disord 18:540.
https://doi.org/10.1186/s12891-017-1914-5
14. Cicchetti DV, Sparrow SA (1981) Developing criteria for establishing interrater
reliability of specific items: applications to assessment of adaptive behavior. Am J Ment
Defic 86:127–137
15. Jain AK, Kumar S, Tuli SM (1999) Tuberculosis of spine (C1 to D4). Spinal Cord
37:362–369. https://doi.org/10.1038/sj.sc.3100833
16. Damasceno LHF, Catarin SRG, Campos AD, Defino HLA (2006) Lordose lombar:
estudo dos valores angulares e da participação dos corpos vertebrais e discos
intervertebrais. Acta Ortopédica Bras 14:193–198. https://doi.org/10.1590/S1413-
78522006000400003
17. Rajasekaran S, Shanmugasundaram TK (1987) Prediction of the angle of gibbus
deformity in tuberculosis of the spine. J Bone Joint Surg Am 69:503–509
18. Murugappan K, Khandwal P, Upendra B, Jayaswal A (2009) Comment on the new
classification of surgical treatment of spinal tuberculosis. Int Orthop 33:291–292.
https://doi.org/10.1007/s00264-008-0582-3
19. Rajasekaran S, Soundararajan DCR, Shetty AP, Kanna RM (2018) Spinal Tuberculosis:
Current Concepts. Glob Spine J 8:96S-108S.
https://doi.org/10.1177/2192568218769053
20. Rajasekaran S (2001) The natural history of post-tubercular kyphosis in children.
Radiological signs which predict late increase in deformity. J Bone Joint Surg Br
83:954–962. https://doi.org/10.1302/0301-620x.83b7.12170
21. Rajasekaran S (2012) Kyphotic deformity in spinal tuberculosis and its management. Int
Orthop 36:359–365. https://doi.org/10.1007/s00264-011-1469-2
22. Rajasekaran S (2013) Natural history of Pott’s kyphosis. Eur Spine J 22:634–640.
https://doi.org/10.1007/s00586-012-2336-6
23. An HS, Vaccaro A, Cotler JM, Lin S (1994) Spinal Disorders at the Cervicothoracic
Junction. Spine 19:2557–2564
24. Nene A, Bhojraj S (2005) Results of nonsurgical treatment of thoracic spinal
tuberculosis in adults. Spine J Off J North Am Spine Soc 5:79–84.
https://doi.org/10.1016/j.spinee.2004.05.255
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
25. Bhojraj S, Nene A (2002) Lumbar and lumbosacral tuberculous spondylodiscitis in
adults. Redefining the indications for surgery. J Bone Joint Surg Br 84:530–534.
https://doi.org/10.1302/0301-620x.84b4.12363
26. Rajasekaran S (2001) The natural history of post-tubercular kyphosis in children.
Radiological signs which predict late increase in deformity. J Bone Joint Surg Br
83:954–962. https://doi.org/10.1302/0301-620x.83b7.12170
27. Kandwal P, Garg B, Upendra B, et al (2012) Outcome of minimally invasive surgery in
the management of tuberculous spondylitis. Indian J Orthop 46:159–164.
https://doi.org/10.4103/0019-5413.93680
28. Yadav G, Kandwal P, Arora SS (2020) Short-term outcome of lamina-sparing
decompression in thoracolumbar spinal tuberculosis. J Neurosurg Spine 1–8.
https://doi.org/10.3171/2020.1.SPINE191152
29. Jain AK, Sreenivasan R, Saini NS, et al (2012) Magnetic Resonance evaluation of
tubercular lesion in spine. Int Orthop 36:261–269. https://doi.org/10.1007/s00264-011-
1380-x
30. Shetty AP, Viswanathan VK, Kanna RM, Shanmuganathan R (2017) Tubercular
spondylodiscitis in elderly is a more severe disease: a report of 66 consecutive patients.
Eur Spine J 26:3178–3186. https://doi.org/10.1007/s00586-017-5157-9
31. Jain AK, Dhammi IK (2007) Tuberculosis of the spine: a review. Clin Orthop 460:39–
49. https://doi.org/10.1097/BLO.0b013e318065b7c3
Figure Legends
Figure 1 – A schematic work flowchart depicting various stages involved in the formulation
of TB spine instability score.
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Figure 2- Calculation of Vertebral Body Loss. Each partially destroyed vertebral body as seen
in a mid-sagittal T2 weighted MRI image is represented in the form of a rectangle assuming
its height to be the average of the anterior vertebral body height of the unaffected cranial and
caudal vertebra. Further, the lateral height of each vertebral body is divided in 10 equal parts
and the height loss measured in tenths from each vertebra is summed to obtain vertebral body
loss (VBL). Note that for each completely destroyed vertebra, 1 is added to the calculated
VBL.
Figure 3– ROC curve of stable and potentially unstable TB spine cases diagnosed by
TSIS. AUC = 0.96; cut-off value = TSIS score 6. AUC = area under the ROC curve;
ROC = receiver operative characteristic.
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Figure 4 – ROC curve of potentially unstable and unstable TB spine cases diagnosed by
TSIS. AUC = 0.96; cut-off value = TSIS score 10. AUC = area under the ROC curve;
ROC = receiver operative characteristic.
Figure 5 - 13-year old girl with dorso-lumbar TB spine with involvement of posterior
elements and 30 degree pre-op kyphosis with no measurable neurological deficit. Fig. 5(A) -
Pre-op radiographs showing lateral subluxation at D11 over D12 with complete destruction of
D10 vertebra. Fig 5(B) – MRI showing a prevertebral collection with posterior element
involvement. Fig 5(C) – Post-operative radiographs showing posterior instrumentation at 18
month follow-up. TSIS score in this patient was calculated as follows: Age (13 years) = 1;
Location – junctional (D9-D12) = 2; Pain – present at rest with no relief on recumbency = 3,
Adjusted kyphotic deformity – 40 degrees = 2; Vertebral body collapse – 1.3 = 2; Additional
involvement of posterior elements – bilateral facet joint involvement with lateral translation =
3; Multifocal contiguous disease – 3 disc spaces = 3; Abscess – present =1. The final tally of
the component scores yields a score of 17 out of 21. This score indicates that the patient has
an unstable spine. She was managed with posterior instrumentation for stabilization and
reported a good clinical and radiological outcome at 18-month follow-up.
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Figure 6 - 28-year old male with TB spine of dorsal region (D6-D7) with intact neurology but
signs of upper motor neuron lesion (spasticity). Fig. 6(A) – Pre-operative radiographs
showing 20 degree kyphosis with paravertebral collection. Fig 6(B) – MRI showing a 50%
paradiscal destruction of D6 and D7 with prevertebral abscess and compression over the cord.
The calculation of TSIS score proceeded as follows: Age – 28 years = 0; Location – Rigid
(D6-D7) = 0; Pain – present on loading with relief on recumbency = 2; Kyphosis – 10-30
degrees = 1; Vertebral body loss - <0.75 = 0; Additional involvement of posterior elements –
absent = 0; Multifocal contiguous disease – 1disc space = 1; Abscess – present = 1. The tally
of the component scores add up to 5 out of 21. The lesion was deemed stable and suitable for
conservative management with anti-tubercular therapy.
Figure 7 – 20-year old lady with TB spine of L1-L2 level with intact neurology and severe pain. Fig 7(A) –
Radiographs demonstrating near complete destruction of L2 vertebra with an adjusted kyphotic angle of 40
degrees. Fig 7(B) – Sagittal MRI cuts confirms the findings of an X-ray and show compression of the cord at the
conus level. Fig 7(C) – Axial MR cuts confirm the absence of posterior element involvement in the involved
vertebra. The TSIS score was calculated as follows: Age – 20 = 0; Location – junctional = 2; Pain – present on
loading with no relief on recumbency = 3; Adjusted kyphotic deformity – 40 degrees = 2; Vertebral body loss -
0.75-1.5 = 2; Additional involvement of posterior elements – absent = 0; Multifocal contiguous disease – 1 disc
= 1; Abscess – absent = 0. The tally of the component scores yielded a score of 10. According to the scoring,
this patient has an impending instability and should be counselled regarding the possible need for surgical
fixation.
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Table 1. Various included items in the preliminary scoring instrument and their index
of content validity and k values after first and third rounds
Item Round 1 Round 3
CVI Κ value CVI K value
Age 0.63 0.53 0.87 0.86
Location 0.87 0.86 1 1
Pain 0.63 0.53 1 1
Kyphosis 0.63 0.53 1 1
Vertebral body loss 0.63 0.53 1 1
Additional involvement of 0.63 0.53 1 1
posterior elements
Multifocal contiguous 0.75 0.72 1 1
disease
Intervertebral/Paravertebral 0.63 0.53 0.87 0.86
abscess
Expected mobility after 0.50 Eliminated
surgery
Osteoporosis 0.50 Eliminated
Table 2. Final scoring system with all elements of TSIS
Age
<5 years 3
5-10 years 2
10-15 years 1
>15 years 0
Location
Junctional (occiput-C2, C7-T2, T10-L2, L5-S1) 2
Mobile spine (C3-C6, L2-L5) 1
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Rigid (Rest of the spine) 0
Pain
Pain even at rest, Pain on loading and 3
movement or 'instability catch' with no relief on
recumbency
Pain on loading and movement or 'instability 2
catch' with relief on recumbency
Occasional pain or not mechanical 1
Pain-free lesion 0
Adjusted kyphotic deformity*
>60 degrees 3
30-60 degrees 2
10-30 degrees 1
<10 degrees 0
§
Vertebral body loss
>1.5 3
0.75-1.5 2
<0.75 0
Additional involvement of posterior
spinal elements
Bilateral facet joint involvement and destruction 3
OR radiographically scoliosis, AP or lateral
translation
Unilateral facet joint involvement 2
None 0
Multifocal contiguous disease
3 or more disc space/ spinal segments 3
2 disc spaces/ spinal segments 2
1 disc space/ spinal segments 1
No disc space (central disease) 0
Intervertebral/Paravertebral abscess
Presence 1
Absence 0
Total score**:
< 7: Stable
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
7-10: Impending instability
>10: Unstable
*measured kyphosis plus five degrees for each involved cervical or lumbar disc/vertebra and
15 degrees for L4-L5 and L5-S1 region each
§
calculated by dividing each vertebra into 10 equal parts on a lateral radiograph and adding
fractional loss of the height of each vertebra
**The current scoring system is a dynamic guide for spinal stability and a higher score
represents instability at a particular stage of the disease which may change with time.
Although the score provides a systematic approach to determine spinal instability, a number
of subjective criteria should also be considered before final surgical decision making which
are not included in the score such as co-morbidities, osteoporosis, activity status and
expectations of the patient.
View publication stats
Copyright © 2021 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
