Clinical Biomechanics 112 (2024) 106178

Contents lists available at ScienceDirect

Clinical Biomechanics
journal homepage: www.elsevier.com/locate/clinbiomech

A bioengineering investigation of cervical collar design and fit: Implications

on skin health
Laurence J. Russell a, *, Tamara Dodd a, Daniel Kendall a, Amber Lazenbury a, Abigail Leggett a,
Sophie Payton-Haines a, Liudi Jiang b, Davide Filingeri a, Peter R. Worsley a
a
Skin Sensing Research Group, School of Health Sciences, Faculty of Environmental and Life Sciences, University of Southampton, Southampton, UK
b
School of Engineering, Faculty of Engineering and Physical Sciences, University of Southampton, Southampton, UK

A R T I C L E I N F O A B S T R A C T

Keywords: Background: Cervical collars restrict cervical spine movement to minimise the risk of spinal cord injury. Collars
Cervical collar apply mechanical loading to the skin putting it at risk of skin damage. Indeed, cervical collar-related pressure
Pressure ulcer ulcers are unacceptably prevalent, especially at the occiput, mandibles, and chin. Collar design and fit are often
Range of motion
key considerations for prevention.
Interface pressure
Methods: This comprehensive study evaluated four commercial prehospital and acute care cervical collars.
Pressure, microclimate, transepidermal water loss and skin hydration were measured at the interface between
the device and the skin. Range of motion restriction was measured to evaluate effective immobilisation. Head,
neck, and shoulder morphology was evaluated using three-dimensional scans.
Findings: The occiput experienced significantly higher interface pressures than the chin and mandibles for most
collar designs. Interface pressure at the occiput was significantly higher for the Stiffneck extrication collar
compared to the other collar designs. The Stiffneck collar also provided the most movement restriction, though
not significantly more than other designs. Relative humidity at the device skin interface was significantly higher
for the Stiffneck and Philadelphia collars corresponding to closed cell foam padding, in contrast to the open cell
foams lined with permeable fabric used in the other collars. Collar discomfort correlated with both occipital
pressure and skin humidity.
Interpretation: The occiput is at increased risk of cervical collar-related pressure ulcers during supine immobili­
sation, especially for Stiffneck extrication collars. Lined open-cell foams could be used to minimise skin humidity
and increase comfort.

1. Introduction often lead to localised areas of pressure and shear at the device-skin
interface (Sparke et al., 2013). The materials that interface with the
Cervical collars are routinely used to immobilise trauma patients skin can also contribute to a local increase in microclimate temperature
with a suspected spinal injury (Ahn et al., 2011). Immobilisation reduces and humidity, adversely affecting the skin’s tolerance to mechanical
the risk of further damage to the spine that could lead to motor and loads (Kottner et al., 2018). The mechanical loading from interface
sensory impairment and morbidity in the most severe cases. Collars are pressure and shear over time has an associated risk of skin breakdown
worn until spinal clearance can be obtained through clinical examina­ leading to pressure ulcers (PUs), also termed pressure injury. When
tion, often including radiological assessment (National Institute for caused by medical devices, PUs are also termed medical device-related
Health and Care Excellence [NICE], 2016). Cervical collars are also used pressure ulcers (MDRPUs), where skin damage typically conforms
during rehabilitation for musculoskeletal disorders of the cervical spine. with the shape of the overlying device (Black et al., 2010).
This includes weakness resulting from neuromuscular or musculoskel­ The reported incidence of cervical collar-related pressure ulcers
etal disease, as well as from surgical interventions. Immobilisation is varies between observational studies, ranging from 0.7 to 44% (Ackland
achieved through rigid support around the cervical spine with adjust­ et al., 2007; Chan et al., 2013; Davis et al., 1995; Ham et al., 2014;
able straps and height adjustments. The rigid supports and strapping Powers et al., 2006; Wang et al., 2020). Brannigan et al. (2022) report a

* Corresponding author at: AA97, Southampton General Hospital, School of Health Sciences, University of Southampton, Southampton SO16 6YD, UK.
E-mail address: l.j.russell@soton.ac.uk (L.J. Russell).

https://doi.org/10.1016/j.clinbiomech.2024.106178
Received 30 August 2023; Accepted 8 January 2024
Available online 9 January 2024
0268-0033/© 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
L.J. Russell et al. Clinical Biomechanics 112 (2024) 106178

pooled average of 7%. This high variability is likely a result of differ­ approved by the University of Southampton Ethics Committee (ERGOII –
ences in preventative measures, setting, and reporting procedures. 18,511.A1).
Several studies suggest under-reporting as a potential limitation in less
accessible locations, such as the occiput, where the device and hair may 2.1. Test equipment
block visible signs of skin damage. Indeed, common sites of skin
breakdown include the occiput, chin, shoulders, and clavicle (Ackland Four different cervical collars were used for this study (Fig. 1). The
et al., 2007). Amongst the risk factors, collar application duration leads Miami J standard collar with sternal pad (Össur UK Ltd., Stockport, UK),
to the most significant increase in PU risk (Brannigan et al., 2022). These made of 2 mm thick LDPE (Low Density Polyethylene) panels with 10
observational studies do not report the effects of collar design and fit and mm thick fabric-lined open-cell foam (Sorbatex). The Stiffneck Select
often do not report the type of collar used. collar (Laerdal Medical Ltd., Orpington, UK), made of 2 mm thick HDPE
Experimental studies suggest that different collar designs introduce (High Density Polyethylene) shell with 6 mm thick closed-cell foam
varied interface pressure, indicating variable PU risks. Studies on padding. The Philadelphia Adjustable Tracheotomy collar (Össur UK
interface pressure and range of motion (Tescher et al., 2007; Tescher Ltd., Stockport, UK), made of 3 mm thick PP (Polypropylene) supports
et al., 2016; Whitcroft et al., 2011; Worsley et al., 2018) demonstrate with 10 mm thick formed closed-cell Plastazote foam. The Aspen Vista
these differences between collar designs. However, interface pressure collar (Aspen Medical Products, Irvine, CA, USA), made of 1–2 mm thick
alone provides a limited view of ulcer risk and may not directly relate to polymer panels with 9 mm thick cotton-lined open-cell foam padding.
its ability to restrict cervical movement (Oomens et al., 2010). Several Interface pressures were monitored using a commercial system
intrinsic and extrinsic factors could contribute to ulceration (Bader and (Tactilus Free Form Sensor System, Sensor Products Inc., Madison, NJ,
Worsley, 2018). In particular, individual head shape and size may affect USA). Their V Series 10 × 10 mm square sensors have an operating range
collar fit, yet have not been fully evaluated in collar designs. Thus, a of 0 and 258 mmHg, reported accuracy of ±10%, repeatability of ±2%,
more holistic approach, combining biomechanical and physiological and a sampling frequency of 49 Hz (Sensor Products Inc., 2023).
assessments, is required to evaluate PU risks from collar usage. A recent Microclimate at the device-skin interface was measured using combined
study has contributed to this approach, investigating the use of bio­ sensors (SHT75; Sensirion AG, Stäfa, Switzerland), recording relative
markers to evaluate the skin’s response to loading from cervical collars humidity and temperature at 0.5 Hz with report accuracy of ±1.8% RH
(Worsley et al., 2018). However, this study was limited to two-collar and ± 0.3 ◦ C, respectively. The skin’s physiological status was assessed
designs, restricting a full comparison across common designs and ma­ by measuring transepidermal water loss (TEWL) and skin hydration.
terials of collars. TEWL and skin hydration were measured with commercial sensors
The design and fit of cervical collars in the context of skin health Tewameter, MPA9 and Corneometer, MPA9, respectively (Courage &
have yet to be widely reported. This study investigated collar function, Khazaka, Köln, Germany). Surface scans of individuals’ heads, necks and
collar-skin interface pressure, physiological skin response, and patient shoulders were taken using a commercial hybrid structured light scan­
comfort across four common cervical collar designs in healthy volun­ ner in infrared mode with an accuracy of 0.6 mm (EinScan H; Shining
teers. A combination of measures previously used to evaluate medical 3D, Hangzhou, China). Scans were processed using proprietary software
device-related pressure ulcers and anthropometrics was used to compare (EXScan H v1.0.5.3; Shining 3D, Hangzhou, China). Range of motion
the effects of different collar designs. The association between collar was measured using a handheld digital inclinometer (Digital Levelmeter
function, interface pressure, skin response and comfort were also 1700, SOAR, Japan). With the collars in situ, the participants were also
studied. asked to report their perceived discomfort using an 11-point numeric
rating scale. The lowest score (0) represents no discomfort at any point,
2. Methods and the highest (10) means extreme discomfort.

A convenience sample of healthy participants was recruited via 2.2. Test protocol
poster advertisement from the local university population. Exclusion
criteria included age below 18 or above 60 years, illness, infection, skin A randomised crossover design was used for this study. Participants
conditions affecting the neck area, conditions affecting the cervical were randomised to two of the four commercial collar designs being
spine, cardiovascular insufficiencies affecting soft tissue health and re­ tested.
covery, and inflammatory conditions. Informed consent was received Testing was performed in a biomechanics laboratory with a
from all participants before participation in the study. The study was controlled ambient temperature of 20–22 ◦ C and relative humidity of

Fig. 1. Collar designs; A) Miami J standard collar with sternal pad (Össur UK Ltd., UK), B) Stiffneck Select collar (Laerdal Medical, UK), C) Philadelphia Adjustable
Tracheotomy collar (Össur UK Ltd., UK), D) Aspen Vista collar (Aspen Medical Products, USA).

2
L.J. Russell et al. Clinical Biomechanics 112 (2024) 106178

30–40%. Participants were given at least 10 min to acclimatise to the

laboratory environment before baseline measurements were taken, fol­
lowed by collar application, intervention data collection, collar removal
and post-intervention data collection, as shown in Fig. 2. Participants
attended one session, during which two collars were tested according to
block randomisation. Participants were asked to have clean, washed skin
and have shaved where appropriate at least 48 h before participation.
Demographic data, including age, gender, height, and weight, were
collected.
Thereafter, the participant lay supine on a standard viscoelastic
mattress for each collar test (Medstrom Ltd., Castle Donington, UK).
Baseline physiological measurements (hydration and TEWL) were taken
on the underside of the chin (Fig. 3). The collar was then applied ac­
cording to the manufacturer’s instructions, with feedback on tension
provided by the participant. After 10 min of application, the interface
pressure sensors were inserted between the collar and the skin at the
occiput, chin, right, and left mandible (Fig. 3) and a reading of static
loading was taken. The pressure sensors continued recording while the
range of motion was measured. The participant was asked to perform
flexion, extension, and right and left rotation, stopping when they
experienced resistance from the collar. Neutral and maximum flexion
and extension angles were recorded with the inclinometer resting on the
forehead and touching the nose without compressing it (sagittal plane).
Maximum right and left rotation were recorded with the inclinometer at
90◦ resting on the forehead (transverse plane). The pressure sensors
were then removed and replaced with the temperature and humidity Fig. 3. Collar application with sensor locations indicated, occiput (square),
sensors at the right and left mandible and chin (Fig. 3)—these sampled right and left mandibles (circles), and chin (hexagon).
data for ~30 s until a stable recording was achieved. Before collar
removal, participants were asked to rank their perceived discomfort collar differences, respectively. Nonparametric statistics (median,
using the 11-point verbal rating scale. The collar was then removed, and interquartile range) were applied to comfort scores, interface pressure,
the physiological skin response measures were taken per the baseline and range of motion. Statistical differences were evaluated using a
measurements in the same location under the chin (hydration and Wilcoxon signed-rank test for within-collar differences and a Mann-
TEWL). 3D surface scans were then taken of the participants with and Whitney U test for between-collar differences. Spearman correlation
without the collar whilst sitting upright. A 10-min refractory period was was used to analyse the relationship between non-parametric variables.
imposed between the first collar being removed and baseline measure­ The statistical significance level was set to p ≤ 0.05 for all outcomes. To
ments being taken for the second collar test to ensure adequate soft account for multiple comparisons performed during the study, a Bon­
tissue recovery. ferroni correction was applied to the Alpha value, such that the statis­
tical significance was p < 0.0083 where six pairs of collars were
2.2.1. Data analysis compared.
Statistical analysis was performed using Python (SciPy library). Data
from each test were examined for normality using Shapiro-Wilk and 3. Results
Anderson-Darling tests. Parametric statistics (mean ± standard devia­
tion) were found to be appropriate for analysing hydration and TEWL. 3.1. Participant demographics
Statistical differences between parametric variables were evaluated
using a student’s t-test and paired t-test for between-collar and within- A sample of 25 healthy volunteers (10/15, Male/Female) were
recruited. The participants had a median age of 21 years (range 18–38),
mean height of 173 ± 10 cm, and mean weight of 70.5 ± 16.6 kg, with a
corresponding body mass index (BMI) of 23.3 ± 3.3 kg/m2. Table 1
shows the breakdown of these demographics for each collar after ran­
domisation. There were no significant differences between height,
weight, and corresponding BMI between collars. However, some distinct
differences in gender distribution were noted, with more females testing
the Philadelphia and Aspen collars.

3.2. Interface pressure

Fig. 4 illustrates the interface pressure values across the four mea­
surement sites for each collar during the static lying posture. Significant
differences were observed between measurement locations for all collars
apart from the Miami J collar, with median pressure values ranging
between 51.0 and 66.0 mmHg across the four locations. The Stiffneck
collar revealed significantly higher interface pressure at the occiput than
the mandibles (p < 0.001), with some readings exceeding the measur­
able range of the sensors at 258 mmHg. The Philadelphia and Aspen
Fig. 2. Test protocol timeline schematic for one collar application. collars also had significantly higher interface pressure at the occiput

3
L.J. Russell et al. Clinical Biomechanics 112 (2024) 106178

Table 1
Participant demographics and anthropometrics.
Collar n Age Height ± SD [m] Weight ± SD [kg] BMI ± SD [kg/m2] M/F Neck Height [mm] Neck Circumference [mm]

Miami J 13 25 (18–38) 1.73 ± 0.12 74.2 ± 18.7 24.4 ± 3.1 7/6 103 ± 19 371 ± 48
Stiffneck 13 24 (20–38) 1.74 ± 0.10 72.5 ± 16.2 23.7 ± 3.1 7/6 104 ± 21 375 ± 44
Philadelphia 13 21 (18–26) 1.71 ± 0.09 65.0 ± 14.2 22.0 ± 3.3 3/10 113 ± 21 345 ± 30
Aspen 11 21 (20–27) 1.74 ± 0.12 71.6 ± 17.1 23.1 ± 3.5 3/8 116 ± 21 352 ± 28

Fig. 4. Plot of interface pressure for four collars across four measurement locations (Occiput, right mandible, left mandible, and chin), pressure sensor upper limit
(258 mmHg) indicated with horizontal line.

when compared to the chin (p < 0.008).

Table 2
There were also significant differences between the collar designs for
Mean range of motion for all collars in flexion, extension, rotation to the right,
pressure at the occiput. The occiput pressure of the Stiffneck was
rotation to the left.
significantly higher than the Miami J and Aspen collars (p < 0.008),
with a more than two-fold increase compared to the Miami J collar. Miami J Stiffneck Philadelphia Aspen
Select
There were no significant differences between pressures at the chin.
However, there were some instances of very high chin pressure for the Flexion (mean ± SD) 12.1 ± 10.9 ± 7.5 11.6 ± 6.6 12.8 ±
[deg] 6.3 6.5
Miami J and Stiffneck collars and very low or absent chin pressure for
Extension (mean ± SD) 15.1 ± 10.9 ± 7.2 13.2 ± 4.9 16.5 ±
the Miami J, Stiffneck and Philadelphia collars. This is evidenced by [deg] 8.5 8.8
outliers in the box and whisker plots. Rotation Right (mean ± 26.7 ± 20.3 ± 12.4 23.7 ± 10.0 28.0 ±
SD) [deg] 10.9 12.7
Rotation Left (mean ± SD) 30.2 ± 22.0 ± 12.6 31.0 ± 9.5 32.3 ±
3.3. Cervical range of motion [deg] 13.0 14.8
Total Flexion Extension 27.2 ± 21.8 ± 13.1 24.8 ± 8.4 29.2 ±
No significant differences in range of motion were observed between (mean ± SD) [deg] 12.5 11.6
the different collars. However, differences in mean value indicate that Total Transverse Rotation 56.9 ± 42.3 ± 24.3 54.7 ± 18.3 60.3 ±
(mean ± SD) [deg] 23.1 27.0
the Stiffneck collar was the most restrictive, and the Aspen and Miami J
collars were the least restrictive. Across all collar designs, the most re­
striction was observed in the sagittal plane, namely flexion and exten­ %RH mean difference, 95% CI [11,20], p < 0.001 (chin)) and Aspen
sion. By contrast, transverse rotation was higher, with mean values just collars (23 %RH mean difference, 95% CI [17,29], p < 0.001 (chin)).
over 2-fold higher than flexion/extension. A high degree of inter-subject Additionally, the Miami J saw increased relative humidity at the chin
variability was observed in all four collars; for example, in left rotation, compared to the Aspen collar (p < 0.008). (See Fig. 5.)
values ranged from 6.5 to 54.0◦ for the Aspen collar. (See Table 2.) Mean skin temperatures were lowest in the Aspen Vista
(34.2–34.8 ◦ C) collar and highest for the Philadelphia (35.0–35.5 ◦ C)
3.4. Microclimate collar across each measurement site. Though temperatures in the Phil­
adelphia collar were significantly higher than the Aspen Vista collar
Skin temperature and relative humidity did not vary significantly across at the chin (p < 0.008), the mean differences were small, 0.8 ◦ C
between the three test sites. Significant differences in relative humidity (95% CI [0.1,1.5]) at the chin, 0.8 ◦ C (95% CI [0.3,1.3]) at the right
were observed between collars. Mean relative humidities were lowest mandible, and 0.6 ◦ C (95% CI [0.1,1.2]) at the left mandible. The
for the Aspen Vista collar. Relative humidity was significantly increased Stiffneck also experienced significantly higher skin temperatures than
for the Stiffneck compared to the Miami J (20 %RH mean difference, the Aspen Vista at the right mandible (p < 0.008), with a mean differ­
95% CI [16,24], p < 0.001 (chin)) and Aspen collars (27 %RH mean ence of 0.6 ◦ C (95% CI [0.2,1.0]). Across all collars, the mean skin
difference, 95% CI [23,33], p < 0.001 (chin)). Relative humidity was temperatures were above normal physiological ranges for a resting adult
significantly increased for the Philadelphia compared to the Miami J (15 male (30–34 ◦ C) (Gagge et al., 1967; Winslow et al., 1937).

4
L.J. Russell et al. Clinical Biomechanics 112 (2024) 106178

Fig. 6. Plot of perceived discomfort for each of the collar where 10 in most
discomfort and 0 is no discomfort.

J collar with significant correlations between neck circumference and

range of motion in extension-flexion range (r(10) = 0.81, p < 0.01) and
transverse rotation range (r(10) = 0.73, p < 0.01).
There was a moderate positive correlation between the perceived
discomfort and pressure at the occiput, r(46) = 0.41, p < 0.01. No
correlation was observed between the perceived discomfort and other
interface pressure measurements. Perceived discomfort also had a
moderate positive correlation with the relative humidity at all three
measurement sites, r(46) = 0.56 (chin), 0.56 (right), 0.49 (left), p <
Fig. 5. Plot of microclimate parameters (mean (± 1 std) temperature and 0.01. Range of motion measures were most notably moderately nega­
relative humidity) at the collar-skin interface for each collar at the chin, right
tively correlated with the pressure at the occiput r(46) = − 0.48, − 0.50
mandible, and left mandible.
(extension, rotation), p ≤ 0.001.

3.5. Skin response 4. Discussion

There were limited changes in the biophysical skin parameters at the This study compared the biomechanical, physiological, and percep­
chin, with a high degree of inter-subject variability (Table 3). TEWL and tual responses between four cervical collar designs in healthy volun­
hydration did not differ significantly from baseline (p > 0.5). No sig­ teers. The interface conditions varied significantly between collars and
nificant variations in these measures existed between the different collar between locations on the collar, implying effects from collar designs and
designs (p > 0.4). Most TEWL values were within normative ranges materials used within each device. The effectiveness of restricting
(8–14 g/hm2), irrespective of collar type. movement in the cervical spine varied between collars and depended on
the plane of motion. Pressure and relative humidity values observed in
3.6. Perceived discomfort this study for specific collar conditions could put skin and underlying
soft tissues at risk of damage if sustained for prolonged periods. The
Fig. 6 shows the perceived discomfort was significantly higher for the results from this study could inform the design of new collars that
Stiffneck than the Miami J and Aspen collars (p < 0.001). Although non- combine effective movement restriction with optimised skin health for
significant, the Stiffneck collar was also more uncomfortable than the vulnerable patients.
Philadelphia collar (p = 0.030). The Stiffneck had a median discomfort Interface pressures were significantly higher at the occiput than at
score of 6 (range 5–8), compared to Aspen and Miami J, which had the mandible and chin for the Stiffneck, Philadelphia and Aspen collars.
median scores of 3 (range 1–5 for both collars). This agrees with the literature indicating an increased incidence of
pressure ulcers at the occiput (Powers et al., 2006). The Stiffneck had the
3.7. Correlation analysis highest median occipital interface pressure (155, IQR = 122,258
mmHg), which could be associated with its limited conformity to the
The relationships between anthropometric measurements and shape of the head, especially around the rear portion of the collar. This
interface pressure were limited (Fig. 7). However, for specific collars, increased interface pressure for the Stiffneck collar aligns with the re­
statistically significant correlations between anthropometrics and range ported high incidence of pressure ulcers using this design of extrication
of motion were observed. This was particularly the case with the Miami

Table 3
TEWL and hydration measurement before and after (pre and post) collar application for each collar.
Outcome measure Miami J Stiffneck Philadelphia Aspen

Pre Post Pre Post Pre Post Pre Post

TEWL (g/hm2) mean ± SD 14.4 ± 5.5 13.5 ± 3.6 16.0 ± 7.6 14.8 ± 4.4 15.0 ± 3.0 16.5 ± 3.3 14.0 ± 3.2 14.3 ± 2.8
Hydration (AU*) mean ± SD 49.7 ± 19.2 48.7 ± 18.6 55.9 ± 14.3 57.3 ± 15.4 59.2 ± 13.8 59.2 ± 14.8 53.1 ± 14.9 52.3 ± 14.5
*
Arbitrary units.

5
L.J. Russell et al. Clinical Biomechanics 112 (2024) 106178

Fig. 7. Spearman correlation coefficient between each of the test parameters. Statistical significance is indicated by bold fonts, where p < 0.05.

collar (Ham et al., 2016). By contrast, the Miami J had very similar and collar users, further investigation is needed into the causes of this high
lower median pressure values and thus better-distributed load across the variability. This research area would benefit from an objective measure
sites (64, 51, 66, and 51 mmHg for the occiput, right mandible, left of the movement restriction.
mandible, and chin, respectively), which may indicate higher confor­ This study identified significant differences in the relative humidity
mity of fit. Across all collars, individuals experienced interface pressure and temperature at the device-skin interface between the different collar
values exceeding 100 mmHg, particularly over the occiput, which ex­ designs. Though limited differences in skin temperature were observed
ceeds the 90 mmHg interface pressure reported to occlude microcircu­ between collars, all designs contributed to an increased skin tempera­
lation (Worsley et al., 2020). If applied for extended periods, even ture when compared to normal physiological ranges for a resting adult
pressures below this threshold can significantly damage the soft tissue male (30–34 ◦ C) (Gagge et al., 1967; Winslow et al., 1937). The Stiffneck
(Gefen, 2009a, 2009b). These interface pressure values at the occiput and Philadelphia collars had higher relative humidity than the Aspen
and mandible for the Aspen Vista, Stiffneck Select and Miami J collars Vista and Miami J collars. Black et al. (1998) also found increased skin
recorded are similar to those recorded in the literature (Plaisier et al., humidity in a Philadelphia collar compared to an Aspen collar. The
1994;Tescher et al., 2007; Tescher et al., 2016; Worsley et al., 2018). padding materials likely contribute significantly to temperature and
Recorded values here fall in the range between mean and peak values of humidity increases. The Stiffneck and Philadelphia both use an uncov­
26.9 mmHg (mean) and 64.5 mmHg (peak) at the occiput and 24.1 ered, closed-cell foam. In contrast, the Aspen Vista and Miami J collars
mmHg (mean) and 110.0 mmHg (peak) at the mandible recorded by use an open cell foam with a layer of fabric covering the surfaces that
Tescher et al. (2016). This is as expected, given their study’s higher contact the skin. Compared to open-cell foam and permeable fabrics,
sensor area and resolution. In several cases, it was observed that the closed-cell foam reduces heat and moisture transfer to the surrounding
collar did not make complete contact with the chin portion of the collar, environment (Glicksman, 1994). Humidity and temperature play an
resulting in very low or zero pressure. In these cases, the jaw shape important but not fully understood role in the risk of pressure ulcer
meant most contact was carried on the mandibles. development (Kottner et al., 2018). Changes in the microclimate around
The Stiffneck was the most restrictive compared to the other three the skin affect its mechanical properties and ability to withstand
collars aligning with a previous study comparing the Stiffneck and extended mechanical loading (Kottner et al., 2018). Animal models
Aspen Vista collars (Worsley et al., 2018). There was substantial inter- demonstrate that temperatures above 35 ◦ C significantly impact the
subject variability in the range of motion for each collar design. This viability of loaded soft tissues (Kokate et al., 1995). The elevated tem­
demonstrates a high variance in the risk of motion-related injury to the peratures and high relative humidity for specific collar designs observed
spine. Additionally, range of motion was correlated with some anthro­ in this study likely put the underlying soft tissue at increased risk of
pometric measurements, indicating that patient morphology influences pressure ulcers. Skin hydration and TEWL are also affected by the
the collar’s efficacy. The range of motion evaluated in this study com­ device-skin interface. However, this study revealed no changes in TEWL
pares well with those reported in the literature (Tescher et al., 2007; or hydration at the chin following collar application. This could have
Tescher et al., 2016; Worsley et al., 2018). Though the transverse been due to the limited contact at the chin site for some participants and
rotation range is generally higher than in the literature, Tescher et al. the relatively short application period of the device. Indeed, other
(2016) reported transverse rotation to be half that of this study for the studies with more prolonged device application have demonstrated
Aspen collar. However, Worsley et al. (2018) reported a similar trans­ changes in TEWL and hydration (Abiakam et al., 2023; Kottner et al.,
verse rotation range to the present study. To better meet the needs of all 2015).

6
L.J. Russell et al. Clinical Biomechanics 112 (2024) 106178

Patient discomfort has not been fully reported in studies evaluating 5. Conclusions
the efficacy of cervical collars. Discomfort negatively contributes to
patients’ experience and quality of life, adding to the already traumatic This study investigated several key bioengineering factors with a
situation that warrants collar use. This potentially contributes signifi­ view to preventing medical device-related pressure ulcers which have
cantly to device abandonment in the rehabilitation setting. Patient- been identified by Bader et al. (2019). By exploring a range of factors at
reported discomfort has only been recorded for the Aspen Vista and the device-skin interface this study has enabled a better understanding
Stiffneck Select collars in the literature; discomfort for the Aspen Vista is of the most significant differences between collars and the influence of
similar to that recorded in the present study, but discomfort associated collar design and materials on key outputs. The data collected in the
with the Stiffneck Select is higher than previously reported (Worsley present study demonstrate apparent differences in the interface pressure
et al., 2018). In an observational study, Ham et al. (2016) found a high and humidity across common collar designs. A more than twofold in­
incidence of pain (63.2%), occurring most frequently at the occiput and crease in pressure at the occiput was observed in the Stiffneck collar
for female patients, but the collar design was not mentioned. Discomfort compared to the MiamI J collar. Collars that used closed-cell foam
was correlated with the pressure at the occiput and the relative humidity padding had 20 %RH higher humidity at the interface compared to those
at the interface. Elevated temperature, in combination with the build-up that used lined open-cell foams. Both factors are known to play a sig­
of moisture, is associated with increased discomfort (Gagge et al., 1967). nificant role in the increased risk of pressure ulcers. This was also
This study revealed elevated temperatures across all collar designs and evident in perceived discomfort for the end-users.Improving the design
significantly elevated interface humidity for specific designs, the com­ and fit of cervical collars will reduce the incidence of collar-related
pounding effects of which likely contribute to the discomfort levels pressure ulcers, improve the patient’s quality of life, and reduce the
observed. Further work investigating the dominant factor contributing cost of care.
to perceived discomfort would help focus research on collar designs for
improved overall user experience. CRediT authorship contribution statement

Laurence J. Russell: Conceptualization, Data curation, Formal

4.1. Limitations analysis, Investigation, Methodology, Project administration, Visuali­
zation, Writing – original draft, Writing – review & editing. Tamara
The present study recruited young, healthy participants. This limits Dodd: Data curation, Writing – original draft, Investigation. Daniel
the generalisability of the findings to the population of collar users, who Kendall: Data curation, Investigation, Writing – original draft. Amber
often have comorbidities. The collars were fitted according to the re­ Lazenbury: Data curation, Investigation, Writing – original draft.
searcher’s interpretation of the manufacturer’s guidelines and not by a Abigail Leggett: Data curation, Investigation, Writing – original draft.
trained orthotist which may also contribute to the high variability in Sophie Payton-Haines: Data curation, Investigation, Writing – original
some results. However, collars often need to be donned and doffed by draft. Liudi Jiang: Supervision, Writing – review & editing. Davide
the user or their family members; it was thus relevant to involve lay Filingeri: Supervision, Writing – review & editing. Peter R. Worsley:
persons. Conceptualization, Methodology, Supervision, Writing – original draft,
Pressure was measured at discrete points at locations specifically Writing – review & editing.
susceptible to pressure injury. This does not fully describe the interface
pressure distribution nor the mechanical characteristics of the under­
lying tissue. Additionally, the pressure sensors were limited to reading Declaration of competing interest
values below 258 mmHg, which was exceeded in several instances,
especially for the Stiffneck collar at the occiput. Therefore, there was a The authors declare that they have no conflict of interest.
ceiling effect on the observed pressures at this site. The true extent of the
mechanical interactions need to be explored with sensors incorporating
Acknowledgements
a wider operating range. Finite element modelling could be used to
further estimate interface pressure distribution and the stress and strain
This work was supported by the UK Engineering and Physical Sci­
within the underlying tissue, whereby potential outcomes could also be
ences Research Council (EPSRC) grant EP/S02249X/1 for the Centre for
verified by the sensor measurements in this study.
Doctoral Training in Prosthetics and Orthotics. For the purpose of open
The short collar application could have led to the minimal physio­
access, the author has applied a Creative Commons Attribution (CC BY)
logical response of the skin. Prolonged donning periods could reveal
licence to any Author Accepted Manuscript version arising.
more about the skin’s response. Additionally, the physiological response
measurements were limited to the chin site as this was the most
convenient. A high degree of variability in interface pressure was References
observed between participants at this site, including instances of zero
Abiakam, N.S., Jayabal, H., Filingeri, D., Bader, D.L., Worsley, P.R., 2023. Spatial and
pressure, further limiting interpretation of findings. Future research temporal changes in biophysical skin parameters over a category I pressure ulcer.
focussing on the occiput site could also be valuable. Int. Wound J. https://doi.org/10.1111/iwj.14194.
Participant discomfort was measured with an 11-point numerical Ackland, H.M., Cooper, J.D., Malham, G.M., Kossmann, T., 2007. Factors predicting
cervical collar-related decubitus ulceration in major trauma patients. Spine 32 (4),
rating scale consistently across all collar applications. This method has 423–428. https://doi.org/10.1097/01.brs.0000255096.52871.4e.
not been validated for the present study setting, limiting the interpre­ Ahn, H., Singh, J., Nathens, A., MacDonald, R.D., Travers, A., Tallon, J., Fehlings, M.G.,
tation of results. Despite not being validated, this method has been used Yee, A., 2011. Pre-hospital care management of a potential spinal cord injured
patient: a systematic review of the literature and evidence-based guidelines.
previously to successfully identify significant differences in collar com­ J. Neurotrauma 28 (8), 1341–1361. https://doi.org/10.1089/neu.2009.1168.
fort (Worsley et al., 2018). Bader, D.L., Worsley, P.R., 2018. Technologies to monitor the health of loaded skin
Although the parameters measured in this study are known to in­ tissues. Biomed. Eng. Online 17 (1), 40. https://doi.org/10.1186/s12938-018-0470-
z.
fluence the risk of pressure ulcers, the exact relationships with pressure
Bader, D.L., Worsley, P.R., Gefen, A., 2019. Bioengineering considerations in the
ulcer development – especially in combination – are unclear. Addi­ prevention of medical device-related pressure ulcers. Clin. Biomech. (Bristol, Avon)
tionally, there are many other intrinsic and extrinsic factors that will 67, 70–77. https://doi.org/10.1016/j.clinbiomech.2019.04.018.
influence the risk of collar-related pressure ulcers (Coleman et al., Black, C.A., Buderer, N.M., Blaylock, B., Hogan, B.J., 1998. Comparative study of risk
factors for skin breakdown with cervical orthotic devices: Philadelphia and Aspen.
2014). Therefore, the significance of collar design considerations re­ J. Trauma Nurs. 5 (3), 62–66. https://doi.org/10.1097/00043860-199807000-
mains uncertain. 00002.

7
L.J. Russell et al. Clinical Biomechanics 112 (2024) 106178

Black, J.M., Cuddigan, J.E., Walko, M.A., Didier, L.A., Lander, M.J., Kelpe, M.R., 2010. National Institute for Health and Care Excellence [NICE], 2016. Spinal Injury: Assessment
Medical device related pressure ulcers in hospitalized patients. Int. Wound J. 7 (5), and Initial Management. [Nice Guideline No. 41]. https://www.nice.org.uk/guid
358–365. https://doi.org/10.1111/j.1742-481X.2010.00699.x. ance/ng41.
Brannigan, J.F.M., Dohle, E., Critchley, G.R., Trivedi, R., Laing, R.J., Davies, B.M., 2022. Oomens, C.W., Loerakker, S., Bader, D.L., 2010. The importance of internal strain as
Adverse events relating to prolonged hard collar immobilisation: a systematic review opposed to interface pressure in the prevention of pressure related deep tissue injury.
and Meta-analysis [review]. Global Spine J. 12 (8), 1968–1978. https://doi.org/ J. Tissue Viability 19 (2), 35–42. https://doi.org/10.1016/j.jtv.2009.11.002.
10.1177/21925682221087194. Plaisier, B., Gabram, S.G.A., Schwartz, R.J., Jacobs, L.M., 1994. Prospective evaluation of
Chan, M., Al-Buali, W., Charyk Stewart, T., Singh, R.N., Kornecki, A., Seabrook, J.A., craniofacial pressure in four different cervical orthoses. J. Trauma 37 (5), 714–720.
Fraser, D.D., 2013. Cervical spine injuries and collar complications in severely https://doi.org/10.1097/00005373-199411000-00004.
injured paediatric trauma patients [review]. Spinal Cord 51 (5), 360–364. https:// Powers, J., Daniels, D., McGuire, C., Hilbish, C., 2006. The incidence of skin breakdown
doi.org/10.1038/sc.2013.6. associated with use of cervical collars. J. Trauma Nurs. 13 (4), 198–200. https://doi.
Coleman, S., Nixon, J., Keen, J., Wilson, L., McGinnis, E., Dealey, C., Stubbs, N., org/10.1097/00043860-200610000-00016.
Farrin, A., Dowding, D., Schols, J.M., Cuddigan, J., Berlowitz, D., Jude, E., Sensor Products Inc, 2023. Tactilus Free Form Sensor System. https://www.sensorprod.
Vowden, P., Schoonhoven, L., Bader, D.L., Gefen, A., Oomens, C.W., Nelson, E.A., com/freeform.php.
2014. A new pressure ulcer conceptual framework. J. Adv. Nurs. 70 (10), Sparke, A., Voss, S., Benger, J., 2013. The measurement of tissue interface pressures and
2222–2234. https://doi.org/10.1111/jan.12405. changes in jugular venous parameters associated with cervical immobilisation
Davis, J.W., Parks, S.N., Detlefs, C.L., Williams, G.G., Williams, J.L., Smith, R.W., 1995. devices: a systematic review [review]. Scand. J. Trauma Resuscitation Emergency
Clearing the cervical spine in obtunded patients: the use of dynamic fluoroscopy. Med. 21, 81. https://doi.org/10.1186/1757-7241-21-81.
J. Trauma 39 (3), 435–438. Tescher, A.N., Rindflesch, A.B., Youdas, J.W., Jacobson, T.M., Downer, L.L., Miers, A.G.,
Gagge, A.P., Stolwijk, J.A.J., Hardy, J.D., 1967. Comfort and thermal sensations and Basford, J.R., Cullinane, D.C., Stevens, S.R., Pankratz, V.S., Decker, P.A., 2007.
associated physiological responses at various ambient temperatures. Environ. Res. 1 Range-of-motion restriction and craniofacial tissue-interface pressure from four
(1), 1–20. https://doi.org/10.1016/0013-9351(67)90002-3. cervical collars. J. Trauma 63 (5), 1120–1126. https://doi.org/10.1097/
Gefen, A., 2009a. Reswick and Rogers pressure-time curve for pressure ulcer risk. Part 1. TA.0b013e3180487d0f.
Nurs. Stand. 23 (45) https://doi.org/10.7748/ns2009.07.23.45.64.c7115, 64, 66, 68 Tescher, A.N., Rindflesch, A.B., Youdas, J.W., Terman, R.W., Jacobson, T.M., Douglas, L.
passim. L., Miers, A.G., Austin, C.M., Delgado, A.M., Zins, S.M., Lahr, B.D., Pichelmann, M.
Gefen, A., 2009b. Reswick and Rogers pressure-time curve for pressure ulcer risk. Part 2. A., Heller, S.F., Huddleston, P.M., III., 2016. Comparison of cervical range-of-motion
Nurs. Stand. 23 (46), 40–44. https://doi.org/10.7748/ns2009.07.23.46.40.c7169. restriction and craniofacial tissue-Interface pressure with 2 adjustable and 2
Glicksman, L.R., 1994. Heat transfer in foams. In: Hilyard, N.C., Cunningham, A. (Eds.), standard cervical collars [randomized controlled trial]. Spine 41 (6), E304–E312.
Low Density Cellular Plastics: Physical Basis of Behaviour. Springer, Netherlands, https://doi.org/10.1097/BRS.0000000000001252.
pp. 104–152. https://doi.org/10.1007/978-94-011-1256-7_5. Wang, H.-R.N., Campbell, J., Doubrovsky, A., Singh, V., Collins, J., Coyer, F., 2020.
Ham, H.W., Schoonhoven, L.L., Galer, A.A., Shortridge-Baggett, L.L., 2014. Cervical Pressure injury development in critically ill patients with a cervical collar in situ: a
collar-related pressure ulcers in trauma patients in intensive care unit [comparative retrospective longitudinal study. Int. Wound J. 17 (4), 944–956. https://doi.org/
study]. J. Trauma Nurs. 21 (3), 94–102. https://doi.org/10.1097/ 10.1111/iwj.13363.
JTN.0000000000000046. Whitcroft, K.L., Massouh, L., Amirfeyz, R., Bannister, G.C., 2011. A comparison of neck
Ham, W.H.W., Schoonhoven, L., Schuurmans, M.J., Leenen, L.P.H., 2016. Pressure movement in the soft cervical collar and rigid cervical brace in healthy subjects.
ulcers, indentation marks and pain from cervical spine immobilization with J. Manipulative Physiol. Ther. 34 (2), 119–122. https://doi.org/10.1016/j.
extrication collars and headblocks: an observational study. Injury 47 (9), jmpt.2010.12.007.
1924–1931. https://doi.org/10.1016/j.injury.2016.03.032. Winslow, C.E.A., Herrington, L.P., Gagge, A.P., 1937. Physiological reactions of the
Kokate, J.Y., Leland, K.J., Held, A.M., Hansen, G.L., Kveen, G.L., Johnson, B.A., Wilke, M. human body to varying environmental temperatures. Am. J. Phys. 120 (1), 1–22.
S., Sparrow, E.M., Iaizzo, P.A., 1995. Temperature-modulated pressure ulcers: a https://doi.org/10.1152/ajplegacy.1937.120.1.1.
porcine model. Arch. Phys. Med. Rehabil. 76 (7), 666–673. https://doi.org/ Worsley, P.R., Stanger, N.D., Horrell, A.K., Bader, D.L., 2018. Investigating the effects of
10.1016/s0003-9993(95)80637-7. cervical collar design and fit on the biomechanical and biomarker reaction at the
Kottner, J., Dobos, G., Andruck, A., Trojahn, C., Apelt, J., Wehrmeyer, H., Richter, C., skin. Med. Dev. Evidence Res. 11, 87–94. https://doi.org/10.2147/MDER.S149419.
Blume-Peytavi, U., 2015. Skin response to sustained loading: a clinical explorative Worsley, P.R., Crielaard, H., Oomens, C.W.J., Bader, D.L., 2020. An evaluation of dermal
study. J. Tissue Viability 24 (3), 114–122. https://doi.org/10.1016/j. microcirculatory occlusion under repeated mechanical loads: implication of
jtv.2015.04.002. lymphatic impairment in pressure ulcers. Microcirculation 27 (7), e12645. https://
Kottner, J., Black, J., Call, E., Gefen, A., Santamaria, N., 2018. Microclimate: a critical doi.org/10.1111/micc.12645.
review in the context of pressure ulcer prevention. Clin. Biomech. (Bristol, Avon) 59,
62–70. https://doi.org/10.1016/j.clinbiomech.2018.09.010.