f

Measurement of forces associated with

compression therapy
D. W e r t h e i m I J. M e l h u i s h 2 R. W i l l i a m s I K. Harding 2
1School of Electronics, University of Glamorgan, Pontypridd, CF37 1DL, UK
2Wound Healing Research Unit, Department of Surgery, University of Wales College of Medicine,
Cardiff CF4 4XN, UK

Abstract--Compression therapy is the principal treatment for leg ulcers associated with
venous disease. The efficacy of compression therapy can be variable, which may in part
be owing to the degree of compression applied. However, if the mechanism of action of
this treatment could be better understood, it might be possible to improve its efficacy. It is
not clear whether assessment of the degree of compression should be made under static
or dynamic conditions, or both. A review of methods used previously suggests the need
for a new method of assessment allowing continuous monitoring, even during movement.
A system for continuous static and dynamic measurements of compression is described.
Using an air chamber and manometer to test the system, agreement within • 3 mmHg is
observed. The system is applied to investigate changes in forces, expressed as pressure,
under bandages and compression stockings. Application of five bandage systems by
experienced nurses to a volunteer shows a marked variation in applied pressure. During
short periods of walking, rapid changes in pressure under compression stockings are
observed, including some transients of less than 0.25s. The method is simple to apply
and may help to understand further the mechanism of action of compression therapy.

Keywords~Bandages, Compression therapy, Forces, Pressure

Med. Biol. Eng. Comput., 1999, 37, 31-34

1 Introduction 1. l Compression therapy

VENOUS DISEASE is the most common condition associated
with chronic ulceration of the leg (BAKER et al., 1992;
HARDING and LEAPER, 1994). The underlying reasons for
the formation of venous ulcers and their slow and unpredict-
able rate of healing are not well understood, and there are
several theories regarding the aetiology of venous leg ulcers
(SHAMI et al., 1992). The principal treatment for venous leg
ulcers is compression therapy, which has been used for
centuries (BLAIR et al., 1988). However, the way compression
therapy assists in healing is not well understood (SARIN et al.,
1992), and varying treatment success rates have been reported
(BLAIR et al., 1988; MOFFATT et al., 1992; FLETCHERet al.,
1997).
It has been suggested that the variable healing rates may be
due to differences in application technique producing varying
degrees of compression (BLAIR et al., 1988; HORNER et al.,
1980). Thus there has been much interest in investigating the
physiological effects of compression therapy.
The site of ulceration in venous leg ulcers is most
commonly in the lower leg area (NELZI~N et al., 1994;
BAKER et al., 1991), which is suggestive of a local vulner-
ability. In patients with venous disease, raised ambulatory
venous pressure, measured in a vein in the foot, has been
found to be associated with an increased incidence of
ulceration (NICOLAIDES et al., 1993).
Correspondence should be addressed to Dr D. Wertheim; email:
wertheim@cardiff.ac.uk
First received 27 March 1998and in final form 6 August 1998
9 IFMBE:1999
It has been suggested that the pressure gradient achieved by
standing in a swimming pool can alleviate the symptoms of
venous ulcers (MONETA et al., 1995). Compression stockings
were therefore designed to achieve a gradient equivalent to the
hydrostatic pressure gradient in the pool. STEMMER et al.
(1980) examined the application of compression stockings
for the treatment or prophylaxis of various conditions and
suggested several pressure profiles that were dependent upon
the medical condition.
However, it is unclear how the degree of compression
required may vary with different disease states and between
different individuals with legs of different sizes and shapes. If
the compression applied is too low, then the efficacy of the
treatment may be impaired (BLAIR et al., 1988). If the applied
'pressure' is too great, then tissue damage may be caused
(CALLUM et al., 1987). It can be expected that the effect of
compression over bony parts of the leg will be different to that
over soft tissue, and this is difficult to assess because of point
loading effects.
Thus there has been much interest in evaluating sub-
bandage pressure or force, as well as pressure and forces
under graduated compression stockings (BLAIR et al., 1988;
RAJ et al., 1980; PARTSCH and HORAKOVA, 1994; SAWADA,
1993; TENNANTet al., 1988; DALE et al., 1983; NELSON et al.,
1995; BARBENEL and SOCKLINGHAM, 1990; SOCKLINGHAM
et al., 1990; STEINBERGand COOKE, 1993; LOGANet al., 1992),
and in evaluating the haemodynamics associated with com-
pression therapy (SARIN et al., 1992; LAWRENCE and
KAKKAR, 1980).

Medical & Biological Engineering & Computing 1999, Vol. 37 31

 Several studies of compression therapy applied by health
care professionals have reported that sub-bandage pressure can
vary markedly (BLAIR et al., 1988; RAJ et al., 1980; PARTSCH
and HORAKOVA, 1994; SAWADA, 1993; TENNANTet al., 1988;
DALE et al., 1983; NELSON et al., 1995; LOGAN et al., 1992).
These studies have been on both human and artificial legs.
However, only a few studies have investigated the effect of
movement on the forces applied by compression therapy
(SOCKLINGHAMe t al., 1990; STEINBERG and CooKE, 1993).
The reasons for the variability in sub-bandage pressure are
unclear but may be due to the application techniques, the
variability of the patients or volunteers, or the physical
properties of the bandage systems.
1.2 Measurement o f pressure and force
Pressure is a scalar quantity that is generally used for fluids,
where the force per unit area is constant at any given cross-
section; this assumption is clearly not applicable for bandages
on legs, and so the term force is preferable. In addition to
normal forces, shear forces may be important, as they may
induce a tearing or bringing together effect on vulnerable
tissue.
The law of Laplace has been found to be useful in a number
of areas in medicine (STILLWELL,1973). It has been suggested
that Laplace's equation P = T/R, where R = radius, P = sub-
bandage pressure, and T = tension, can be used to estimate the
sub-bandage pressure P, if the tension T required to produce a
given extension is known, using a modified form
p- TN x constant
CW
where N = number of layers, c = limb circumference, w =
width of bandage, and, in this case, T = tension, in Newtons
(THOMAS, 1996; NELSON, 1996).
However, the suitability of the application of the equation in
this situation is not clear, as the leg is clearly neither
cylindrical, nor a fluid, and the bandage may not be at uniform
tension around the leg. In addition, in the modified form, the
term included for the number of layers as a simple multiple
appears questionable.
Transducers used for measurement of sub-bandage forces
tend to be of low profile to minimise interference in the
measurement. The force applied on a transducer is likely to
vary in many positions on the leg. If an average 'pressure' is
monitored with a large-area transducer, then information about
local high-force or pressure-areas may be missed, and a low-
area transducer is likely to have greater sensitivity to position.
Thus different measuring systems may produce differing
results.
Air-filled sensor systems usually allow only intermittent
monitoring (PARTSCH and HORAKOVA, 1994; SAWADA, 1993;
STEINBERG and COOKE, 1993), and liquid-filled sensor sys-
tems are likely to be sensitive to movement artefact, because
of movement o f the liquid column (NELSON et al., 1995;
BARBENEL and SOCKLINGHAM,1990; SOCKLINGHAMet al.,
1990).
We feel that studies during movement may be useful, in
view of the importance of the calf muscle pump. Furthermore,
it has been found that sequential-gradient, intermittent pneu-
matic compression can enhance venous ulcer healing (COLER-
IDGE SMITH et al., 1990; MCCULLOCH et al., 1994), and a
limited range of motion was found in limbs of patients with
chronic venous insufficiency (CVI), with the degree of limita-
tion appearing to be related to the degree of CVI (BACK et al.,
1995). Thus, to enable continuous monitoring during moue-
32 Medical & Biological Engineering & Computing
ment, we developed a system in which a transducer can be
placed directly on the leg, with no air or liquid connecting
link.
2 Aim
The aim of this study was to develop and apply a system for
continuous monitoring of forces under bandages and gradu-
ated compression stockings, both under stationary conditions
and during movement.
3 Method
A Fontanometer sensor*, based on the strain gauge
principle, was used to assess sub-bandage pressure
(WERTHEIM et al., 1996). The sensor appears as a rigid disc,
approximately 3 m m thick and 12.6mm in diameter. The
sensor is temperature compensated from 15 ~ to 40~ The
sensitive area of the sensor is rectangular and centrally placed
on one side of the disc. The longer side of the sensitive area is
placed on the leg, such that it is parallel to the long axis of the
leg, which has a large effective radius of curvature. Because
there is no liquid column, both static and dynamic measure-
ments can be made.
Calibration was achieved using a water column and also an
air chamber connected to a mercury manometer.
Five extensible bandage systems were applied by eight
nurses to the leg of a sitting healthy volunteer. Four of the
bandage systems consisted o f two bandages applied at two
different extensions, as described in the manufacturers'
instructions. The fifth bandage system was a multilayer
bandage system. The position of the leg was adjusted as
required by the nurse. The forces expressed as pressure were
monitored, using the sensor placed 4.5 cm above the medial
malleolus. The circumference of the leg at this position was
25 cm.
The system was also used to investigate changes in forces,
associated with movement, under graduated compression
stockings. Three sensors placed on the medial aspect of the
left leg were used to monitor forces on six healthy volunteers.
The sensors were connected to amplifiers that were connected
to a 12-bit analogue-to-digital converter in a notebook com-
puter. The signal from the amplifiers was sampled at 400 Hz
per channel. The data were stored on hard disk and transferred
to a spreadsheet program for further analysis. The sensors
were placed at three positions: 9 cm above the medial mal-
leolus (lower); at the level of greatest prominence of the calf
muscle (mid); and at the upper border of the calf muscle
(upper). Class 2 (European/Continental classification) com-
pression stockings were then applied. The volunteers sat with
both legs horizontal. Data were recorded during plantar flexion
and dorsiflexion of the left foot. Data were also collected
during short periods of walking.
4 Results
Using a water column or an air chamber with mercury
manometer for calibration of the sensors, agreement within
4-3 mmHg was observed. An example of comparison with an
air chamber and manometer is shown in Fig. 1.
A marked variation in recorded pressure under compression
bandages applied to the volunteer was observed, as shown in
Fig. 2. Bandage system 2 used the same bandage as system 1.
However, for system 2, the nurses were asked to apply the
*Gaeltec ktd, Scotland
1999, Vol. 37
 10 60

S 84

O:

a~ 30 r ~ ~ ~ ~ ~ ~ _
= 20
-10 ~" 10 ~ ~l~ . . . . .
0 20 40 60 80 1O0
manometer,mmHg I dorsiflex plantar flex
0 I I I I I I ] I I I I I I I I I I I I I
Fig. 1 Example of comparison of sensor using air chamber con- 1 2 3 4 5 6 7 8 9 10
nected to manometer; (manometer reading-sensor system time, s
reading) against manometer reading
Fig. 3 Example of changes in pressure under compression stock-
ings associated with dorsiflexion and plantar flexion of left
bandage with greater extension. Even so, there is little differ-
foot
ence in the median pressure applied. Similarly, bandage
system 4 used the same bandage as system 3. However, for
system 4 the nurses were asked to apply the bandage with 70-
greater extension. Fig. 2 also shows the approximate overall standing walking
range of pressure of the bandage systems; this anticipated 6o. n~,d /~ /1
range is based on manufacturers' data for the single bandage
systems and published data for the multilayer bandage system 5040-'i ...................-..-. /~ ~
and assumes a limb circumference of 25 cm.
In the studies with graduated-compression stockings, plan- 30-
tar flexion and dorsiflexion of the left foot with respect to the upper
Q.
resting position were associated with changes in recorded
pressure that could vary in the different positions monitored 1o L L L
on the leg, as shown in the example in Fig. 3. The changes in
0 ~ I I I I I I I I I t I I I
pressure also varied in duration from transient changes to the 1 2 3 4 5 6 7
period of the manoeuvre. Changes in pressure during walking time, s
were also found to be dependent on where they were mon- Fig. 4 Example of changes in pressure under compression stock-
itored and of variable duration, including some transients of ings when walking. 'L ' marks show when highestforces were
less than 0.25 s in duration, as seen in the example in Fig. 4. recorded under left Joot when walking (using two Force
During walking, the median of the greatest increase in pres- Sensing Resistors)
sure recorded by the mid-placed sensor on the six volunteers
was 22.5 mmHg (range 4-68 mmHg).
during movement. There is, however, limited evidence as to
the appropriate level of compression and'whether and how this
5 Discussion and conclusions should vary in different people and in different disease states.
Thus the implications of the observed variation in bandage
The pressure applied with compression therapy is likely to
application are difficult to assess, although it may account, at
vary around the perimeter and length of the leg. Graduated
least in part, for the previously observed variation in efficacy.
compression is thought to help venous return but, as there may
In addition, it is clear that compression achieved varies over
be a decrease in pressure through tissue (SHAW and MURRAY,
the leg, varies with position and during movement, and may
1982) to the veins, it may be important to take account of the
vary with time.
type and bulk of tissue as well as the skin surface pressure. For
The system is suitable for clinical use and may thus be of
example, as the perimeter at the level of maximum circum-
value in comparing treatment strategies and in comparing the
ference of the calf muscle is greater than the perimeter below
way in which they are applied by different people. The system
that level, it may be that the pressure reduction from the
is being further developed to detect shear forces.
surface of the leg to some veins is greater at the level of
The results of this study suggest that the forces applied by
maximum calf muscle circumference.
experienced nurses using conventional compression bandages
The system described above, which uses a small tempera-
can vary markedly. Furthermore, this study has demonstrated
ture compensated sensor, has allowed continuous direct mea-
that forces under graduated compression stockings can vary
surement of forces associated with compression therapy, even
with movement and with foot position. Thus this system may
help to further understanding of the mechanism of action of
compression therapy.
80 84

~60 Acknowledgments--This study was funded by a grant from the Higher

E Education Funding Council for Wales.
~ 4~ We are grateful for the help given to us by the nursing staff of the
Wound Healing Research Unit, University of Wales College of
~. 20 Medicine.

1 t
2 I
3 I
4 I
5 I

bandage system References

Fig. 2 Variation in pressure observed on human leg with different BACK, T. L., PADBERG, F. T., ARAKI, C. T., THOMPSON, P. N., and
bandage systems. Broken lines indicate anticipated range of HOBSON, R. W. (1995): 'Limited range of motion is a significant
pressure from bandage data. (--) rain, max, (0) median f ~ ulceration', J. Vasc. Surg., 22, pp. 519-523

Medical & Biological Engineering & Computing 1999, Vol. 37 33

BAKER, S. R., STACEY, M. C., JOPP-MCKAY, A. G., HOSKIN, S. E.,
and THOMPSON, P. J. (1991): 'Epidemiology of chronic venous
ulcers', Dr. J. Surg., 78, pp. 864-867
BAKER, S. R., STACEY, M. C., SINGH, G., HOSKIN, S. E., and
THOMPSON, P. J. (1992): 'Aetiology of chronic leg ulcers', Eur.
J. Vasc. Surg., 6, pp. 245-251
BARBENEL,J. C., and SOCKLINGHAM,S. (1990): 'Device for measur-
ing soft tissue interface pressures', J. Biomed. Eng., 12, pp. 519-
522
BLAIR, S. D., WRIGHT, D. D. 1., BACKHOUSE,C. M., RIDDLE,E., and
McCOLLUM, C. N. (1988): 'Sustained compression and healing of
chronic venous ulcers', Dr. Med. J., 297, pp. 1159-1161
CALLUM, M. J., RUCKLEY, C. V., DALE, J. J., and HARPER, D. R.
(1987): 'Hazards of compression treatment of the leg: an estimate
from Scottish surgeons', Br. Med. J., 295, p. 1382
COLERIDGE SMITH, P. D., SARIN, S., HASTY, J., and SCURR, J. H.
(1990): 'Sequential gradient pneumatic compression enhances
venous ulcer healing: A randomized trial', Surgery, 108, pp.
871-875
DALE, J., CALLUM,M., and RUCKLEY,C. V. (1983): 'How efficient is
a compression bandage?', Nursing Times, November 16, pp. 49-51
FLETCHER, A., CULLUM, N., and SHELDON,T. (1997): 'A systematic
review of compression treatment for venous leg ulcers', Dr. Med.
J., 315, pp. 576-580
HARDING, K. G., and LEAPER, D. J. (1994): 'The hospital manage-
ment of venous ulceration', Curr. Practice Surg., 6, pp. 8-11
HORNER, J., FERNANDES E FERNANDES, J., NICOLAIDES, A. N.
(1980): 'Value of graduated compression stockings in deep
venous insufficiency', Dr. Med. J., 34, pp. 81-87
LAWRENCE, D., and KAKKAR, V. V. (1980): 'Graduated, static,
external compression of the lower limb: a physiological assess-
ment', Dr. J. Surg., 67, pp. 119-121
LOGAN, R. A., TtlO/VIAS,S., HARDING, E. F., and COLLYER, C. J.
(1992): 'A comparison of sub-bandage pressures produced by
experienced and inexperienced bandagers', J. Wound Care, 1,
pp. 23-26
MCCULLOCH, J. M., MARLER, K. C., NEAL, M. B., and PHIFER,T. J.
(1994): 'Intermittent pneumatic compression improves venous
ulcer healing. Advances in Wound Care', 7, pp. 22-26
MOFFATT, C. J., FRANKS, P. J., OLDROYD, M., BOSANQUET, N.,
BROWN, P., GREENHALGH,R. M., and MCCOLLUM, C. N. (1992):
'Community clinics for leg ulcers and impact on healing', Br. Med.
J., 305, pp. 1389-1392
MONETA, G. L., NEHLER, M. R., CHITWOOD,R. W., and PORTER, J.
M. (1995): 'The natural history, pathophysiology and nonoperative
treatment of chronic venous insufficiency', in RUTHERFORD,R. B.
(Ed.): 'Vascular surgery' (W. B. Saunders, Philadelphia) 4th edn.
NELSON, E. A., RUCKLEY, C. V., and BARBENEL, J. C. (1995):
'Improvement in bandaging technique following training', J.
Wound Care, 4, pp. 181-184
NELSON, E. A. (1996): 'Compression bandaging in the treatment of
venous leg ulcers', J. Wound Care, 5, pp. 415-418
NELZI~N, O., BERGQVIST, D., and LINDHAGEN,A. (1994): 'Venous
and non-venous leg ulcers: clinical history and appearance in a
population study', Dr. J. Surg., 81, pp. 182-187
34 Medical & Biological Engineering & Computing
NICOLAIDES,A. N., HUSSEIN, M. K., SZENDRO,G., CHRISTOPOULOS,
D., VASDEKIS,S., and CLARKE, H. (1993): 'The relation of venous
ulceration with ambulatory venous pressure measurements', J.
Vase. Surg., 17, pp. 414-9
PARTSCH, H.,. and HORAKOVA, M. A. (1994): 'Compression stock-
ings for the treatment of venous leg ulcers', Wien. Med. Wschr.,
144, pp. 242-249
RAJ, T. B., GODDARD, M., and MAKIN, G. S. (1980): 'How long do
compression bandages maintain their pressure during ambulatory
treatment of varicose veins?', Dr. J. Surg., 97, pp. 122-124
SARIN, S., SCURR, J. H., and COLERIDGE SMITH, P. D. (1992):
'Mechanism of action of external compression on venous func-
tion', Dr. J. Surg., 79, pp. 499-502
SAWADA,Y. (1993): 'Alterations in pressure under elastic bandages:
Experimental and clinical evaluation', J. Dermatol., 20, pp. 767-
772
SHAMI, S. K., SHIELDS,D. A., SCURR, J. H., and COLERIDGESMITH,
P. D. (1992): 'Leg ulceration in venous disease', Postgrad. Med. J.,
68, pp. 779-785
SHAW, J. A., and MURRAY, D. G. (1982): 'The relationship between
pressure and soft-tissue pressure in the thigh', J. Bone Joint Surg.,
64, pp. 1148-1151
SOCKLINGHAM,S., BARBENEL,J. C., and QUEEN, D. (1990): 'Ambu-
latory monitoring of the pressures beneath compression bandages',
Care Sci. Pract., 8, pp. 75-78
STEINBERG, M. D., and COOKE, E. D. (1993): 'Design and evaluation
of a device for measurement of interface pressure', J. Biomed.
Eng., 15, pp. 464-468
STEMMER, R., MARESCAUX, J., and FURDERER, C. (1980): "Die
Kompressionbehandlung der unteren Extremitaten speziell durch
Kompressionstriimpfe', Der Hautarzt, 31, pp. 355-365
STILLWELL, G. K. (1973): 'The law of Laplace', Mayo Clin. Proc.,
48, pp. 863-869
TENNANT, W. G., PARK, K. G. M., and RUCKLEY, C. V. (1988):
'Testing compression bandages', Phlebology, 3, pp. 55-61
THOMAS, S. (1996): 'High-compression bandages', J. Wound Care, 5,
pp. 40-43
WERTHEIM, D., MELHUISH, J., TRENARY, K., SHUTLER, S., WIL-
LIAMS, R., and HARDING, K. (1996): 'Assessment of forces
associated with compression therapy under static and dynamic
conditions', Proc. 18th Ann. Int. Conf. IEEE Eng. Med. Biol. Soc.
Amsterdam, October
Authors' biographies
The authors are developing non-invasive methods for objective
assessment of compression therapy. This research is part of a
collaborative multidisciplinary project between the Medical Elec-
tronics and Signal Processing Research Unit (MESPRU), School
of Electronics, University of Glamorgan and the Wound Healing
Research Unit (WHRU), Department of Surgery, University of Wales
College of Medicine.
1999, Vol. 37