154 Burns.
10, 1 54-l 63 Printed in Great Britain
Pressure therapy in the treatment of
post-burn hypertrophic scar-A critical look
into its usefulness and fallacies by pressure
monitoring
J. C. Y. Cheng, J. H. Evans, K. S. Leung, J. A. Clark,T. T. C. Choy, f?. C. Leung
Department of Orthopaedic and Traumatic Surgery, Faculty of Medicine, The Chinese
lJn>versity of Hong Kong
Summary
Pressure therapy is generally accepted as the best non-
invasive means of preventing and controlling hyper-
trophic scarring after burn injury. Most studies in the
past have failed to correlate clinical response with
magnitude of the garment-scar interface pressure.
This study looked critically at our usual techniques
of pressure therapy using pressure ‘sensors’ manu-
factured locally and based on electro-pneumatic prin-
ciples. Many pitfalls, such as large variations of
pressure at different geometric sites on the body, elastic
deterioration in garments, problems of garment manu-
facture, and the unfavourable properties of the Lycra
garments, were observed. Recommendations on press-
ure treatment were made based on our experiences, to
improve the present technique of pressure therapy.
These included the standardization of measurement
techniques and garment tailoring, the regular checking
of pressure at the garment-scar interface using pressure
transducers, the appropriate garment adjustments, a
strict regimen for garment wearing, and the intelligent
use of pressure-padding and reinforcement. Areas of
further research are also discussed.
INTRODUCTION
HYPERTROPHIC scarring is a well-known compli-
cation following bum injury to the human body.
It causes serious functional and cosmetic disable-
ment and considerable socio-economic impli-
cations to the community. Methods oftreatment
available and preventive measures for this con-
dition are, on the whole, unsatisfactory and inef-
fective for large scar areas. The use of external
pressure as a non-invasive means of prevention
and treatment of hypertrophic scars has received
growing recognition in different parts of the
world in recent years (Fujimori et al., 1968;
Larson et al., 197 1; Larson et al., 1974; Kischer
et al., 1975; Tolhurst, 1977; Leung and Ng,
1980; Naismith, 1980; Robertson et al., 1980).
Pressure therapy is currently applied by an
elasticated garment acting on the scar surface
with or without additional pressure-paddings.
Lycra, a spandex fibre material consisting of
80-85 per cent polyurethane is the material com-
monly used, The general belief is that pressure
therapy should be started early, preferably with-
in two weeks after the burn wound or the skin-
grafted area has healed (Larson et al., 1974;
Thomson, 1974; Leung and Ng, 1980; Robertson
et al., 1980); the garment should be worn twenty-
four hours a day with short rest periods for meals
and hygienic measures; the treatment should be
continued for at least nine months and relapses
of the healing process should call for a prompt
resumption of treatment. A pressure of at least
24 mmHg was considered necessary for effective
therapy (Larson et al., 1974; Baur et al., 1976).
Commercially available pressure garments
were plainly accepted as effective without alter-
ation and actual garment-scar interface press-
ures were seldom measured. Measuring and
monitoring of the pressure applied by a garment
on a hypertrophic scar was just considered un-
necessary and it was only recently that
special pressure ‘sensors’ were available for
Cheng at al.: Pressure therapy treatment monitored 155

clinical use. These pressure transducers opened

up new prospectives for more scientific and
quantitative studies on the application and effec-
tiveness of pressure therapy (Isherwood et al.,
I
Zcm
-AIRSAC

-ELECTRODES
1975; Naismith, 1980; Robertson et al., 1980).
This manuscript first analyses the problems
1
associated with the present techniques of press-
ure therapy, then identifies areas for further
improvement and research.

PATIENTS AND METHODS

Pressure therapy clinic L TO LIGHT + SOUND
SIGNALLING DEVICE
Since 1976, over 400 patients have attended the
Pressure Therapy Clinic at Princess Margaret
Hospital in Hong Kong. The majority (60 per
cent were new patients referred shortly after
5
TO MANOMETER
bum wound healing while the rest (40 per cent)
were late cases of well-formed hypertrophic
scars which had been present from a few months
to three years.
After initial assessments, suitable patients
were started on pressure therapy. The elasticated
Lycra garments were all custom-made by the
Occupational Therapists. The technique of tail-
oring and patient measurement has been de-
scribed in a previous article (Leung and Ng,
1980). Patients with pressure garments were fol- ELECTRODES SEPARATED
lowed up in the Clinic regularly every two to
four weeks. They were seen jointly by the Sur-
geon and the Occupational Therapist and the Interface pressure recorded
symptoms of pain, itchiness, blistering, and the when electrodes separate.
state of the scar which included colour, hardness
and thickness were noted. The tension in the gar-
ment was assessed subjectively by ‘feeling’ at the
seams and checking on the comfort and toler- - AIR
ance of the patient wearing it. The garments
were then adjusted if they appeared ‘slack- ELECTRODES IN CONTACT
ened’, usually at monthly intervals. Appropriate
splints and physiotherapy were provided should Fig. I. Electra-pneumatic pressure transducer.

there be joint tightness. Signs of clinical im-

provement in hypertrophic scars were defined as
decreased vascularity, thinning and softening of
the scar and an associated decrease in the itchi-
ness and pain. Based on these criteria, we ana-
lysed 100 patients in 1979 and 72.5 per cent
were found to show good response (Leung and
Ng, 1980). This previous study, like many others
of similar nature, has an inherent limitation of
assuming that the garments applied the necess-
ary external pressure and no actual measure-
ments of the garment-scar interface pressures
were taken.
Pressure measurement
As the pressure provided by a Lycra garment is
strongly dependent on the local geometry of the
body surface, it is imperative that this geometry
must not be significantly disturbed by the press-
ure measurement technique (Naismith, 1980).
This implies that an acceptable pressure sensor
or transducer must be small, thin, flexible, and
provides a pressure reading at a small localized
area. Specially constructed capacitance trans-
ducers or sensors utilizing electro-pneumatic
principles would satisfy these requirements. The
latter type is more suitable for routine clinical
use as it requires no calibration, can be handled
easily, is robust, and provides a direct reading of
the externally applied pressure on the scar (Fig.
1). Commercially available electro-pneumatic
pressure transducers, however, are not suitable
156 Burns Vol. 1 O/No. 3

LIGHT-SOUND
SIGNALLING DEVICE

TRANSDUCER

AIR FLOW MANOMETEF

(mmHg measurement)

0
ONE-WAY PUMP
(hand operated)

Fig. 2. Garment-scar interface pressure measuring apparatus.

for our needs as they are usually designed to
measure much higher pressures than that
normally applied by pressure garments.
In this study, a reliable system of pressure
‘sensor’ utilizing electro-pneumatic principles
and developed in the Bioengineering Depart-
ment of Strathclyde University, Glasgow,
were used for pressure measurement. They were
tested repeatedly and found to have an accuracy
of f 3 mmHg within the range of O-50 mmHg
pressure. The pressure was read off from a
special Air-flow Measuring Manometer with sen-
sitivity of f 1 mm water gauge pressure (Figs. 2
and 3).
Methodology
Instead of adopting the subjective ‘feeling’
method of assessing the tension in the garment
the pressure ‘sensors’ gave quantitative garment-
scar interface pressure measurements. Our gen-
eral understanding and appreciation is enhanced
if the following information can be obtained:
(i) The actual magnitude of pressure at the
garment-scar interface,
(ii) The pressure variation over different
parts of the body in regions of different
geometry,
(iii) The changes in pressure when the gar-
ment was worn continuously for long
periods of time, i.e., due to garment
deterioration,
(iv) Pressure differences related to differ-
ences in tailoring techniques among
different therapists, i.e. manufacturing
differences.
Starting early in 1982, 50 patients having press-
ure garment treatment for hypertrophic scarring
affecting different sites of the body were selected
Cheng et al.: Pressure therapy treatment monitored 157

Fig. 3. Garment-scar interface pressure measuring apparatus.

0
0

0
0

0 I 0

0
0

VENTRAL

Fig. 4. Specific sites for pressure monitoring on upper limb.

b
DORSAL
158 Burns Vol lo/No. 3

Table /. Garment-scar interface pressures (mmHg) at different sites (50 patients)

Range of pressures Average pressure

Area (minimum & maximum) (mmHgl

Upper Limb
Mid-arm Medial O-10 5
Lateral 5-l 5 10
Dorsal 5-10 8
Ventral 5-15 10
Mid-forearm Medial 5-l 5 10
Lateral 5-20 15
Dorsal 5-l 5 10
Ventral 3-l 5 8
Lower Limb
Mid-thigh Medial O-10 8
Lateral 5-20 15
Posterior o-1 5 7
Anterior 5-l 5 10
Mid-calf Medial O-15 10
Lateral 5-20 12
Posterior 5-20 15
Anterior 1040 25
Trunk
Front Presternal area o- 2 0
Lateral chest 5-25 15
Abdomen o- 3 0
Back Interscapular area o- 3 0
Scapular prominence 5-25 15
Lumbosacral area o- 5 0
Buttock o-15 8
Head c!?Neck
Front Chin 1040 25
Cheek O-10 5
Malar region 5-25 20
Forehead/neck O-10 5
Neck o- 5 2
Back Occiput 5-20 15
Neck o- 5 2

for study. The range of pressures at different The measuring sites were located over the ven-
areas were recorded with the pneumatic pressure tral and dorsal sides (4 each) of the arm as indi-
transducer. Children were excluded from the cated in Fig. 4.
study because of their unreliable behaviour Using the same measuring and tailoring tech-
which would affect the results. All the pressure niques, different garments were manufactured by
measurements were made at the interface be- different therapists according to their experience
tween the hypertrophic scar and the garment and on similar areas of scar undergoing pressure
which had been manufactured locally and worn treatment. The pressures applied by these gar-
by the patients and no special adjustments on ments were compared using the pneumatic
the garments were made prior to measurements. transducers.
Another group of 10 patients with hyper-
trophic scars on the upper limbs were selected RESULTS
for more detailed studies. They were all fitted Pressure variation at different sites of
with newly manufactured pressure garments application
and the pressures at eight specific sites were Standard sites over the head and neck, trunk and
measured immediately and thereafter at weekly extremities were chosen for the pressure measure-
intervals with no garment adjustments made. ments. The pressure readings were grouped
Cheng et al.: Pressure therapy treatment monitored 159

:30 -
B :
E 01
&2%b +
iti
E
p .20-
z
f
E
E 15.

E
k
y lo-
3
u)
5-

I I I I I 1 b
0 1 2 3 4 5 6 7 6
Time (Weeks)

Fig. 5. Graphs showing the decay of garment-scar interface pressure with

time from data of 10 patients.

together and expressed as the ‘maximum and
minimum’ and the ‘average pressure’ at each
specific chosen site, and they are listed in
Table I.
From Table I, wide variations of pressure
were found over different parts of the body. Very
low pressures were detected over areas like the
back, the abdomen, and the prestemal area (sites
of large radii of curvature); much higher press-
ures were noted over the shin and chin region
while the pressures over the limbs were medium
in range.
It was also obvious that the majority of the 50
patients showed an average garment-scar inter-
face pressure of 5-l 5 mmHg and only a few had
applied pressures greater than 20 mmHg. We
noted with great interest that all of these gar-
ments had been previously checked by the thera-
pists and passed as being suitable for supplying
satisfactory garment tension. And what was
more, actual good clinical responses were
recorded.
Time-dependent changes in garment-scar
interface pressure
The applied pressures on all the eight specific
sites of the 10 chosen patients were measured
immediately after a satisfactory initial fitting,
and then at weekly intervals. Only a single gar-
ment was given to each patient so that he had
to wear it throughout the testing period of 8
weeks.
The pressure data obtained for each site was
averaged from the measurements on 10 patients,
and these pressure readings were plotted against
the time intervals of measurement in a graph
(Fig. 5).
The graph indicates a significant drop in
garment-scar interface pressure which amounted
to around 50 per cent of the initial pressure after
an interval of four weeks. All the eight sites
under study gave the same results indicating the
gradual decline in interface pressure.
Tailoring techniques of manufacture and
pressure variation
There were significant differences in the press-
ures recorded (greater than 5 mmHg) for the
same area of scar of the same patient treated
with different garments made by different thera-
pists although they were directed to follow the
same manufacturing techniques. This could be
related to differences in tailoring measurements,
cutting and sewing techniques (Table II), and no
therapist could guarantee the reproduction of
garments with identical tension.
160
Tab/e //. Variations in applied pressures (mmHg) by 3 different occupa-
tional therapists on the same patient using separately manufactured
garments
Therapist Therapist
Type of garment 1 2
Gloves 9 12
Forearm Tubes 15
12
Pants 5 6
12
Mask 15
10
DISCUSSION
Pressure therapy has been demonstrated to be ef-
fective in inducing earlier remodelling of hyper-
trophic scars both clinically (Fujimori et al.,
1968; Larson et al., 1971; Larson et al., 1974;
Thomson, 1974; Tolhurst, 1977; Leung and Ng,
1980; Robertson et al., 1980) ultrastructurally
(Larson et al., 1971; Kischer et al., 1975;
Baur et al., 1976; Tully, 1980), biochemically
(Kischer et al., 1975; Naismith, 1980) and bio-
mechanically (Tully, 1980). The exact range
of effective external pressures, though usually
taken as around the arterial capillary pressure of
25 mmHg (Larson et al., 1974; Baur et al.,
1976) has actually not been scientifically
proven.
Our previous analysis of 100 patients (Leung
and Ng, 1980) did show good clinical response
in 72.5 per cent despite the fact that the average
pressure in this group was found retrospectively
to be in the range of 5-l 5 mmHg only. This im-
plies that pressure therapy of a low pressure
range can still be effective as a therapeutic
measure in certain types of patients. However,
recent studies have shown that within the ‘safety
margin’ of 35-40 mmHg, higher pressures ap-
parently induce more rapid maturation of
hypertrophic scars (Naismith, 1980). It seems,
therefore, logical to assume that if the pressure
can be maintained more constant at or above an
‘optimal’ level, one should be able to achieve
even better clinical results.
Our analysis of the present techniques of
pressure therapy using Lycra garments has indi-
cated obvious pitfalls. These include the large
variation of pressure over different sites on the
body, the elastic deterioration or extensibility
of garment, and the problems associated with
tailoring techniques.
Burns Vol. 1 O/No. 3
Therapist
3 Site of injury
8 Dorsum of
Hand
20 22 Dorsal
18 18 Ventral
3 Dorsal
15 10 Ventral
10 18 Chin
6 9 Forehead
The variations in applied pressure over differ-
ent parts of the body appear to be related to the
compliance of the underlying tissue and the local
geometry of the area. The interface pressure
tends to be very high over bony prominences
such as shin regions, in contrast to the soft abdo-
men and buttock where the pressure may be as
low as zero. Geometrically, the pressure on a
convex area of small radius of curvature was
higher than that on an area with large radius of
curvature although the garment tension ap-
peared much the same. Furthermore, concave
surfaces like the presternal area, the interscapu-
lar area, the sacral area and all surfaces where
there is a change of curvature from convex to
concave tend to show zero pressures no matter
how tight the garment is stretched. These obser-
vations in fact are adequately explained by the
Laplace Law which states that PRESSURE=
TENSION/RADIUS OF CURVATURE. Con-
cave surfaces possess negative radii of curvature
and therefore will not be able to produce any
interface pressure despite the higher tension
provided by the garment.
The continuous fall-off in pressure at the
garment-scar interface is the result of the
marked visco-elastic property of the garment
material and appears to be unavoidable unless a
different material is used with more favourable
extensibility properties.
While realizing the pitfalls in our routine press-
ure treatment, which we believe will be present
in any other treatment centre in the absence of
regular pressure monitoring, we would make the
following recommendations:
(i) The techniques of measuring, tailoring
and sewing of the garments should be
standardized to give adequate pressure.
(ii) Regular checks on the pressure under
Cheng et al.: Pressure therapy treatment monitored 161

DORSAL ZERO PRESSURE

CROSS-SECTION

VENTRAL
PADDING

PADDING

Fig. 7. Pressure distribution around biconcave surfaces

(e.g. upper trunk): its modification with padding to
change a negative radius of curvature to positive one.

CROSS-SECTION

VENTRAL

Fig. 6. Pressure distribution around the palm: its

modification with padding to decrease the radius of Fig. 8. Recommendation for the attachment of padding
curvature. to garment.

the garment at the initial and sub-
sequent fittings are necessary and this
should be done objectively using elec-
tro-pneumatic pressure transducers.
(iii) Close follow-up and documentation
of pressures and clinical response is
important.
(iv) Tailoring adjustment of the garments
should be done routinely and regularly
when found slackened or when the
applied pressure drops.
(v) Every patient should be given two sets
of garments with 12-hourly changes
advisable to minimize the time-depen-
dent drop in pressure. A completely
new set of garments is advisable every
month.
(vi) Garments should be worn twenty-four
hours a day except for meals and
hygienic measures and the therapy
should be continued for at least one
year and until full clinical maturation
of the scars are observed.
(vii) Appropriate use of pressure-paddings,
for example, to increase the effective
convexity of scarred area, and thus
increase the pressure exerted on the
scars. A practical example is supplied
in the treatment of scars over the
dorsum of the hand (Fig. 6).
For concave surfaces like the prestemal
area, interscapular area, pressure-paddings
to change the effective radius of curvature
from a negative to a positive convex one is
absolutely necessary if positive pressure is
desired (Fig. 7).
Many types of material are available for
making the pressure-paddings. Soft material
like polyurethane foam (Fujimori et al.,
1968). harder materials such as plastozote.
162
Fig. 9. Double-layered garment.
sansplints and orthoplast (Larson et al.,
1971; Thomson, 1974; Tolhurst, 1977) are
all useful. For an uneven scarred surface,
elastomer works well by filling up all the
crevices and building up the contour.
These materials have all been used for our
patients, either singly or in combination.
They are inserted into the garment-scar
interface and held there to the garment by
velcro attachments (Fig. 8).
In order to produce a pressure of above
25 mmHg over the limbs, one has to apply
a very tight garment and usually the patient
finds it too difficult to put on. By rolling
another layer of Lycra garment in the form
of a sleeve on top of the-original layer (that
is double layered), a consistent high press-
ure of above 25 mmHg can be achieved
(Fig. 9). This manoeuvre is simple to apply
and is most effective.
FURTHER RESEARCH
The pressure measurements at the garment-scar
interface presently are all ‘static’ measurements,
that is, they are taken with the patient relaxed
and resting. Physical activities and movements
would predictably alter the pressures signifi-
cantly. One is not certain whetherthese ‘dynamic’
pressure effects could have more important re-
modelling potential on the scars than the static
pressures.
Further research, in the author’s opinion,
could be directed towards the development of
more effective ways of monitoring the static
pressures as well as the instantaneous interface
pressure variations with limb movements, estab-
Burns Vol. 1 O/No. 3
lishing simple and reliable objective means of
quantifying the progress and maturation of
hypertrophic scars, prospective trials on the
delineation of the optimal pressure range for
therapy, and last, but not least, preventing the
growth of hypertrophic scars at the biochemical
and pharmaceutical levels. The most currently
needed requirement, perhaps, should be related
to the standardization of a pressure range for
better clinical results.
Undesirable features of Lycra material in-
clude: the anisotropic property resulting from
the unidirectional lay of the elastic element pro-
ducing different elastic properties at right angles,
the non-linear response of the Lycra to stress,
the marked visco-elastic property giving rise
to time-dependent fabric elastic deterioration
characteristics, and the abrasive and non-
absorbing nature of the nylon knit component
causing intolerance and blistering problems over
the fragile epidermis in certain patients and es-
pecially in hot-humid weather. A more suitable
material with more desirable elastic and mech-
anical properties with biocompatibility could
perhaps be manufactured.
The achievement of all these objectives would
require the collaboration of multiple disci-
plines involving clinicians, bioengineers and
technologists.
Acknowledgements
We are most grateful to the British Council who
sponsored one of the authors of this paper, Dr
John H. Evans from Strathclyde, Glasgow, on a
special one month’s visit to Hong Kong for the
main supervision of this study. Professor R. M.
Cheng et al.: Pressure therapy treatment monitored
Kenedi, the Associate Director of the Hong
Kong Polytechnic, has given us most appropri-
ate technical advice and encouragement. We
would like to thank also Mr H. W. Hui of the
Textile Division of the Hong Kong Polytechnic
for advising us on Lycra material, Mr S. M. Chu,
Technician in the Electronics Department ofthe
Chinese University of Hong Kong, for manufac-
turing our pneumatic pressure transducers, all
colleagues and occupational therapists who have
treated our bum patients and Mrs A. Kwong
who typed the manuscript for us.
REFERENCES
Ram P. S., Larson, D. L., Stacey T. R. et al., (1976)
Ultrastructural analysis of pressure-treated human
hypertrophic scars. J. Trauma, 16, 958.
Fujimori R., Hiramoto M. and Ofuji S. (1968)Sponge
fixation method for treatment of early scars. Plnst.
Reconstr. Surg 42, 322.
Isherwood P. A., Robertson J. C. and Rossi A. (1975)
Pressure measurements beneath below-knee ampu-
tation stump bandages: elastic bandaging, the Puddi
foot dressing and a pneumatic bandaging techniques
compared. Br. .I. Surg. 62,982.
(‘orre.~pondmcr should he addressed lo: Professor P. C. Leung, Department of Orthopaedic
of Medicine, The Chinese University of Hong Kong. Shatin, N.T., Hong Kong.
163
Kischer C. W., Shetlar M. R. and Shetlar C. L. (1975)
Alteration of hypertrophic scars induced by mech-
anical pressure. Arch. Dermatol. 111, 60.
Larson D. L., Abston S., Evans E. B. et al. (I 97 I )
Techniques for decreasing scar formation and con-
tracture in the burned patient. J. Trauma, 11, 807.
Larson D. L.. Abston S.. Willis B. et al. (1974) Con-
tracture and scar formation in the burned patient.
Clin. Plast. Surg. 1,653.
Leung P. C. and Ng M. (1980) Pressure treatment for
hypertrophic scars. Burns 6, 244.
Naismith R. S. (1980) Hypertrophic scar therapy:
Pressure induced remodelling and its determinents.
Ph.D. Thesis, Bioengineering Unit, University of
Strathclyde, Glasgow.
Robertson J. C. and Druett J. E. (1980) Pressure
therapy for hypertrophic scarring preliminary
communication. J. Roy. Sot. Med. 73, 348.
Thomson W. G. (1974) Hypertrophic scarring. Proc.
Roy. Sot. Med. 6, 256.
Tolhurst D. E. (1977) Hypertrophic scarring pre-
vented by pressure: A case report. Br. J. Plast. Surg.
30,218.
Tully A. E. (1980) Hypertrophic scar tissue: Its micro-
.structure and mechanical property and the Qjlects qf
pressure therapy Ph.D. Thesis, Bioengineering
Unit, University of Strathclyde. Glasgow.
Paper accepted 22 December 1982
and Traumatic Surgery. Faculty