‘Do anti-embolism stockings fit our legs? Leg survey and data analysis’
Abstract
Background
Anti-embolism stockings are commonly used worldwide to prevent the development of
thrombosis in hospitalised patients. Patients are typically measured for, and fitted with,
anti-embolism stockings during extended periods of non-ambulation. Anti-embolism
stockings must critically fit the leg to achieve optimum blood flow and thus success of
prophylaxis. Therefore, hospitals endeavour to maintain stock of anti-embolism stockings
that fit the majority of their patients.
Objectives
The objective of this study was to establish whether popular styles/brands of anti-embolism
stockings “fitted” the legs of randomly selected volunteers.
Design/Methods
Volunteer’s legs were measured at ankle, calf and thigh following guidance from British
nurses and in accordance with brand instructions. Leg measurements were subsequently
compared to the size charts of 10 anti-embolism stocking styles made by 4 different
manufacturers. Where a volunteer’s leg measurements matched all measurement points
for any size, in style’s range, “fit” was achieved.
Settings
Volunteers were measured in different settings around Scotland, including private homes,
work places and shopping centres.
Participants
A convenience sample of 471 volunteers (283 female, 188 male) were recruited on the basis
of willingness to participate and being over 16 years old. Volunteers ranged from 17 to 82
years old with an average age of 35.
Results
The 10 different styles of anti-embolism stockings, made by 4 different brands, examined for
this paper had a “fit” coefficient ranging from 0% to 100% for our volunteer’s legs. The fit
coefficient is strongly influenced by the Brand’s sizing policy.
Conclusions
The proportion of legs that “fit” a particular brand of anti-embolism stockings can be
increased through: 1. the reduction of the number of leg measurement points that need to
be matched to the size chart of the stockings; 2. the use of open-ended size ranges; 3. the
use of increased size range width and 4. the use of increased overlap between sizes.
However, all but the last of these measures can have a potentially deleterious impact on the
ability of the stocking to deliver the optimum graduated pressure profile to all legs that “fit”
the stocking, resulting in important implications to the efficacy of prophylaxis.
Key words: Anti-embolism stockings; fit; leg shape; leg size; mechanical
thromboprophylaxis; Sigel profile.
What is already known about this topic?

Anti-embolism stockings are commonly used worldwide to prevent thrombosis
development in hospitalised patients.

Fitting the correct size of anti-embolism stockings to patients is critical to patient
compliance and the success of thromboprophylaxis.

Incorrect use or fitting of anti-embolism stockings may cause skin problems, ischaema,
and increased risk of thrombosis development.
What this paper adds?

The proportion of our sample that “fitted” different AES brands and styles ranged from
0% to 100%.

Good “fit” was significantly easier to achieve with knee length anti-embolism stockings
compared to thigh length anti-embolism stockings.

Open-ended sizing and reduction of the number of measurement points on the size
charts increase the proportion of patients that appear to “fit” anti-embolism stockings.
However, these measures may have implications on the ability of the stockings to deliver
optimum pressure profile (Sigel profile) to those patients.
Introduction
Pulmonary embolism is believed to be the most common cause of preventable death in
hospitalised patients [Geerts et al, 2008] and approximately 25,000 people die annually as a
result of venous thromboembolism in England alone [Dumbleton and Clift, 2008]. Venous
thromboembolism is the collective term for pulmonary embolism and deep vein thrombosis,
a pulmonary embolism often results from deep vein thrombosis [SIGN, 2010] when part of
the blood clot breaks off and occludes a vein [Dumbleton and Clift, 2008]. Deep vein
thrombosis has been found in between 10 and 80% of hospitalised patients not receiving
prophylaxis [Dumbleton and Clift, 2008, Geerts et al, 2008] and approximately 0.1% of the
general population present per year with clinical symptoms of deep vein thrombosis in the
UK [SIGN, 2010]. Therefore, the majority of surgical patients and many medical patients are
routinely given thromboprophylaxis in order to prevent or reduce the occurrence of venous
thromboembolism [Guyatt et al, 2012, NICE, 2010, SIGN, 2010].
A range of thromboprophylaxis are available including pharmacologic and mechanical
methods. The recommended thromboprophylaxis for each patient depends on their
condition and risk factors [Guyatt et al, 2012, NICE, 2010, SIGN, 2010]. There are 3 different
types of mechanical thromboprophylaxis: anti-embolism stockings (AES); intermittent
pneumatic compression devices and venous/pneumatic foot pumps. The use of mechanical
thromboprophylaxis has been shown to reduce the incidence of deep vein thrombosis by
40% and combined mechanical and pharmaceutical thromboprophylaxis reduce the
incidence of deep vein thrombosis by up to 80% [Dumbleton and Clift, 2008]. This paper
examines the “fit” of anti-embolism stockings (AES), the commonly used form of
thromboprophylaxis for hospitalised non-ambulatory, recumbent patients.
AES apply graduated compression to the leg, which increases blood flow [Dumbleton and
Clift, 2008], reduces stasis [Geerts et al, 2008] and venous distension [NICE, 2010] in the
legs, thereby lowering the risk of thrombosis formation and embolism [Dumbleton and Clift,
2008]. The ideal compression profile is generally agreed to be the ‘Sigel profile’ [Dumbleton
and Clift, 2008, NICE, 2007, Sigel et al, 1975] and most manufacturers design their AES to
deliver this pressure profile to legs properly “fitted” with their stockings [Arjo Huntleigh,
2012, Carolon, 2009, Preventex, 2012]. A stocking that exerts the Sigel profile would exert
18mmHg to the ankle, 14mmHg to calf, 8mmHg at the knee (popliteal break), 10mmHg at
the lower thigh and 8mmHg at the upper thigh. This level of compression has been shown
to increase blood flow in the deep veins by 75% [Sigel et al, 1975].
The pressure exerted by AES is rarely measured in a clinical environment due to lack of time
and availability of an appropriate pressure sensor. Instead stockings are tested using
laboratory methods through the measurement of tension and subsequent calculation of
pressures that would be exerted on a range of limb circumferences. A number of test
methods and devices exist for measuring anti-embolism stockings [BSI, 1993, RAL, 2008] but
all measure the tension of the stocking at the ankle, calf and thigh area (if present) for
particular extensions or circumferences. The pressures that would be exerted by that
stocking at particular circumference of ankle, calf and or thigh are then calculated using an
equation based on the Laplace Law [BSI, 1993]. The simple form of the Laplace Law states
that pressure is equal to tension divided by the radius of curvature (which is easily
calculated from the circumference). Therefore, if a particular stocking is put on a leg that is
thicker or thinner than that for which it was designed, both the tension and circumference
change and it is likely that a different pressure profile will be exerted. Thomas, Toyick and
Fisher (2000) measured changes in pressure (referred to as % stiffness) of up to 36% when
stockings were tested in accordance with the standard BS 7672, “Specification for
Compression, stiffness and labelling of anti-embolism hosiery”, at one size larger and
smaller than that recommended by the manufacturer. Moreover, MacLellan (2002) showed
that the pressure exerted on the maximum recommended leg measurement was frequently
4 to 5mmHg higher than on the minimum recommended leg measurement for commercially
available AES. He further demonstrated that the pressure profiles exerted on the minimum
and maximum recommended leg dimensions were frequently different [MacLellan, 2002].
Therefore, the match between stocking size and leg size is critical to achieving the ideal
pressures at ankle, calf and thigh and the ideal positive pressure gradient from ankle to the
proximal part of the leg.
Guidelines on thromboprophylaxis stress the importance of measuring legs carefully and
correctly, and fitting the correct size of AES in order to ensure good fit, optimum benefit and
compliance with anti-embolism stockings [Geerts et al, 2008, NICE, 2010, SIGN, 2010] as
well as minimising the risks associated with their use [Dumbleton and Clift, 2008, NICE,
2010]. It is also recommended that staff be trained in the use and fitting of AES and that
stocking use be “closely monitored” and patients’ legs be re-measured and re-fitted if
oedema develops [NICE, 2010] or leg size changes. Further, the NICE guideline 92 cautions
against fitting AES to “unusual leg size or shape” due to the risks associated with poorly
fitted stockings. They cite the risks of potential tourniquet effects at the proximal part of
the leg leading to ischaemia and increased risk of thrombosis development if stockings are
the wrong size or shape for the leg [NICE, 2010]. However, the judgement of what
constitutes an “unusual leg size or shape” is normally left to the nurse fitting the stockings
and may be influenced by the size charts of AES stocked by that hospital.
Each brand has its own size ranges at a variety of measuring points and its own methods of
communicating the sizes. Some brands offer fewer measurement points than others and
some have open-ended sizing (where the upper or lower limit of a size range is not
specified), both of which give the impression that they are appropriate for use on a wider
range of patients than brands with narrower or more specific size ranges and or more
measurement points. This can result in confusion as to which products are most suitable for
an individual patient.
Given the critical nature of the fit of AES on the leg in achieving optimum venous return and
maximum protective benefit we turn to the question of whether the AES currently available
in our hospitals is available in the correct sizes for our patient’s legs. To the best of the
author’s knowledge no survey of patient’s leg sizes could be found and the British Standards
committee members for compression hosiery were unaware of a survey having been
conducted. Therefore, since any member of the public could be admitted to hospital for
many of the surgical and medical conditions for which AES is recommended a survey of
randomly selected volunteer’s legs around Scotland was undertaken. These measurements
were compared to the size charts of 10 styles of AES produced by 4 major brands in order to
determine how many of our volunteer’s legs would fit a standard AES and how many had
"unusual leg size or shape” according to the current AES provision.
Method
Measures: The circumferential leg measurements, also called girths, at the ankle, calf and
thigh of 471 volunteers were measured in a variety of locations around Scotland. Both legs
of each volunteer were measured to the nearest cm at the ankle, largest calf girth and
largest thigh girth as indicated by the measurement instructions for anti-embolism stockings
(AES) and following training provided by 2 Scottish National Health Service nurses.
Determination of “fit”: The girths of each volunteer’s left and right legs were separately
compared to the size charts of 4 major brands of AES: Covidien’s (Kendall) T.E.D.; Arjo
Hunteigh’s Flowtron; Urgo’s Preventex; H&R Healthcare’s Carolon. For below knee AES,
each volunteer’s ankle girth was matched to the size chart and size matches were noted;
this was repeated for the calf girth. Where both ankle and calf matched the same size of
below knee stocking the volunteer’s leg was said to match, or “fit”, a particular size. If no
ankle girth range was specified then the match to the brand was determined based on calf
girth only. If the ankle and calf did not match the same product size, e.g. the ankle was
‘small’ but the calf was ‘medium’, then no stocking in that range would be deemed to “fit”
the volunteer. This process was repeated for thigh length stockings using ankle, calf and
thigh girths. The percentage of our volunteers’ legs to “fit” each size of stocking in the
range is presented, as is the percentage of volunteers’ legs that “fitted” no size in the range.
The “fit” coefficient for a brand is defined as the percentage of our volunteers’ legs that
fitted any one or more sizes in a range at all relevant measurement points. Legs may not
“fit” any size in a brands’ range because the leg is too thin or too thick for the sizes available
or because the leg has a different profile to the size charts, being either too conical (e.g.
larger calves) or too cylindrical (e.g. larger ankles), for example.
Participants and recruitment: Volunteers were recruited to the survey using convenience
sampling methods in a variety of locations and settings around Scotland including: private
homes, work places, shopping centres and a nursing home. Attempts were made to cover
different social/economic demographics in the choice of measurement locations and
measurements were made on all days of the week and at different times of day in an
attempt to avoid bias. However, measurement setting was largely determined by the
researchers’ ability to obtain permission to measure thus the sample may not be
representative of the wider population.
In shopping centres, researchers set up a screen to enable private measurements. During
quiet periods all passers-by were approached until someone agreed to participate, at busy
periods the researchers approached as many passers-by as possible. Once a volunteer
showed interest in participation the terms and purpose of the research and the
measurement process were explained using a standardised script. All willing volunteers
were taken behind the screen in order to have both legs measured for this study, the only
exclusion criteria were absence of legs or being younger than 17 years old. All volunteers
signed an ethical consent form prior to being measured. Their sex was noted by the
researcher on the measurement form and each participant was asked their age, no other
demographic information was requested or noted.
In work places and the nursing home, all available employees or residents were asked
whether they wished to participate in the study, if they were willing to participate they were
measured in a private room. In the nursing home a member of staff was also present. The
same exclusion criteria, script, ethical consent form and measurement form were used for
all measurements in all locations and settings.
Data analysis: All leg measurements were recorded in a Microsoft Excel spreadsheet.
Separate copies of this spreadsheet were made for each AES brand and style evaluated in
this study. Each leg measurement was manually matched to sizes according to the size
chart of each brand and style at ankle, calf and thigh. Microsoft excel functions were then
used to determine whether all measurement points for each leg “fitted” all measurement
points for each size of stocking. Where a leg matched the same size of stocking at ankle and
calf for below knee AES or at ankle, calf and thigh for above knee AES it was deemed to “fit”.
The unit of measurement for analysis was legs and 942 legs were measured and analysed in
total.
Note: Each brand of AES evaluated in this study provided a range of sizes in 2 or 3 leg
lengths for each of their products. Three of the 4 brands examined provided open-ended
leg length ranges, such as a short thigh length AES being suitable for patients with a leg
length “less than 74cm”; therefore all of our volunteers would be said to “fit” these AES in
terms of length. Carolon was the only brand to set upper and lower limits for its AES
lengths; this may have further reduced the proportion of our volunteers to “fit” their
products. However, leg length was not included in this evaluation as it is less critical in
determining the pressure gradient delivered to legs than circumferential measurements and
most brands utilise an open-ended sizing policy. All brands require that leg length and a
number of circumferential measurements be taken before selecting the correct size of AES
for a patient. For the rest of this paper however, only the circumferential measurements
required will be discussed.
Results
Survey volunteers ranged from 17 to 82 years of age, with an average age of 35 and
standard deviation of 14. 283 of the volunteers were female and 188 were male.
Table 1 shows that different brands of anti-embolism stockings (AES) vary significantly in
terms of the number of sizes available and the proportion of our volunteers whose
measurements “fit” a particular size. The size ranges and reasons for these differences of
percentage “fit” are explored in later in this paper. Note that the percentage of volunteers
“fitting” each size for some styles adds up to more than 100% because some of the
volunteers “fitted” 2 sizes. The “fit” coefficient is defined as the percentage of our
volunteers who “fitted” any one, or more, sizes in a range of AES at all relevant
measurement points and is found by subtracting the percentage of volunteers who did not
fit any size of AES in Table 1 from 100%.
Table 1 shows that, according to the manufacturer’s size charts, Covidien’s below knee
T.E.D. stockings had a “fit” coefficient of 100%, meaning that they “fitted” all of our
volunteer’s legs, with 10% (n=95) “fitting” 2 sizes of below knee stocking. Urgo’s Preventex
brand below knee stockings had a “fit” coefficient of 82% (n=776), Flowtron AES “fitted”
74% (n=701) and Carolon “fitted” 62% (n=583) of our sample.
Table 1 – Number and percentage of legs that match each AES size chart at all relevant
measurement points
brand
T.E.D.
Preventex
Flowtron
AES
Carolon
n=
%
n=
%
n=
%
n=
%
Brand and
style
T.E.D.
T.E.D. with
belt
Preventex
Flowtron
AES
Carolon
standard
Carolon ATS
Below knee anti-embolism stockings
legs "fitting" size at ankle and calf
legs
that fit
1 size
xs
s
m
l
xl
xxl
xxxl
7
556 305 143
23
3
847
n/a
1% 59% 32% 15% 2%
0%
90%
1
178 406 150 113
704
n/a
n/a
0% 19% 43% 16% 12%
75%
7
138 425
89
96
18
2
627
1% 15% 45% 9% 10% 2%
0%
67%
7
81
355 106
51
566
n/a
n/a
1%
9% 38% 11% 5%
60%
Thigh length anti-embolism stockings
legs "fitting" size at ankle, calf and thigh
legs
xs
n=
%
n=
%
n=
%
n=
%
n=
%
n=
%
n/a
0
0%
n/a
7
1%
n/a
n/a
s
m
l
xl
xxl
xlp/
xxxl
7
1%
7
1%
1
0%
59
6%
4
0%
0
0%
556
59%
556
59%
178
19%
120
13%
13
1%
0
0%
290
31%
290
31%
399
42%
42
4%
29
3%
0
0%
15
2%
15
2%
34
4%
32
3%
14
1%
0
0%
0
0%
50
5%
n/a
n/a
n/a
n/a
1
0%
0
0%
n/a
n/a
0
0%
n/a
that fit
1 size
728
77%
678
72%
544
58%
257
27%
60
6%
0
0%
legs
that fit
2 sizes
95
10%
72
8%
74
8%
17
2%
legs that
don't fit
any size
0
0%
166
18%
241
26%
359
38%
legs
that fit
2 sizes
95
10%
95
10%
34
4%
2
0%
0
0%
0
0%
legs that
don't fit
any size
119
13%
169
18%
364
39%
683
73%
882
94%
942
100%
Table 1 also shows that all thigh length AES brands studied had lower “fit” coefficients than
the below knee AES. Further, 3 of the 6 styles examined “fitted” significantly fewer than
half of our volunteers’ legs and Carolon’s Adjustable Thigh Stockings (ATS) had a “fit”
coefficient of 0%, i.e. they “fitted” none of our volunteers’ 942 legs.
Covidien (Kendall) T.E.D. anti-embolism stockings
Figure 1 shows the T.E.D. stocking size ranges above plots of the distribution of our
volunteers’ leg girths. In this and subsequent Figures:

The plots show the total number of volunteers’ ankles, calves and thighs at each
circumference.

The rectangles above the plots indicate closed AES size ranges with defined lower
and upper limits

Single headed arrows indicate open-ended sizing: an arrow facing left indicates a
‘less than’ size range while an arrow facing right indicates a ‘more than’ size range.

The double headed arrow indicates a size range that is ‘unspecified’, i.e. the absence
of a measurement point. This could be interpreted to mean that the AES will “fit”
any girth at that measurement point.

The vertical ends of each rectangle, left or right facing arrow are in-line with the
circumferential limits of that size as shown on the horizontal axis.
Figure 1 clearly shows that all of our volunteers could be assumed to “fit” a below knee
T.E.D. stocking, because the AES calf girths extend beyond the plot of volunteer’s calf girths.
(The small below knee T.E.D. stocking ‘“fits” any leg with a calf smaller than 30.5cm and the
generous extra extra extra large (xxxl) size extended well beyond the largest calf girth in our
sample.) The ankle girth is unspecified, therefore no T.E.D. stocking size will be rejected for
any leg on the basis of ankle girth, the potential implications of this will be discussed at the
end of this paper. Figure 1 and Table 1 show that medium sized below knee T.E.D. stockings
“fitted” most of our volunteers and that almost all our volunteers could be “fitted” with a
medium, large or extra large size.
400
l
m
s
350
300
xs
250
number of legs with specified girth
xlp
xl
2s
1s
s
200
xl
l
m
m
l
xxl
xlp
xl
xxl
xxxl
xxl
thigh girths for TED AES
calf girths for thigh length TED AES
calf girths for knee length TED AES
ankle girths for TED AES are unspecified
150
volunteer's
ankle girths
100
volunteer's
calf girths
50
0
16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91
volunteer's
thigh girths
leg and AES girth in cm at ankle, calf and thigh
Figure 1 – Plot showing volunteers’ leg girths and Covidien T.E.D. AES size ranges at ankle,
calf and thigh
Covidien manufacture 2 styles of thigh length AES; with belt and without belt, the later
being the ‘standard’ thigh length stocking. Figure 1 shows the size ranges of both styles of
thigh length T.E.D. stocking above the plots of our volunteers’ ankle, calf and thigh girths.
The calf size range for small ‘without belt’ style is labelled as 1s while the small size range
for the ‘with belt’ style is labelled as 2s in Figure 1. The thigh length T.E.D. with belt is
available in extra small (xs) to extra large (xl) sizes, while the thigh length T.E.D. without belt
is available in small to extra extra large (xxl) sizes. The only deviation in shared size ranges
between these styles is in the calf measurement.
As indicated earlier, Figure 1 shows that the ankle girth of all thigh length sizes is
unspecified, so all T.E.D. stockings are deemed to “fit” all legs at the ankle. The small
standard thigh length stockings have open-ended size ranges for calf (labelled 1s) and thigh
measurement points and therefore “fit” any leg smaller than the upper calf and thigh limits
for that size. The medium and large T.E.D. stockings have open-ended size ranges for thigh
dimensions, while the xl, extra large plus (xlp) and xxl sizes have closed size ranges for both
calf and thigh. In this case the open-ended sizing on small, medium and large thigh length
T.E.D. stockings meant that all our volunteers who “fitted” these sizes in below knee
stockings also “fitted” these (or xl) thigh length T.E.D. stockings as shown in Table 1, leading
to a “fit” coefficient of 87% (n=823) for standard thigh length T.E.D. stockings. The potential
implications of open-ended sizing are discussed at the end of this paper. Figure 1 shows
that most of our volunteers thigh girths are covered by the small, medium, large size and
that the medium AES calf size range matched more of our volunteers’ calves than any other,
with the large and xl size matching a significant minority. The reason that the xl thigh length
T.E.D. stocking matched less than 2% (n=15) of our volunteers’ legs was because it had the
same sized calf as the large T.E.D. stocking but a larger thigh circumference (63.5 – 81.3cm).
In our study the volunteers who had larger (xl or xlp) thigh girth tended also to have a larger
(xlp) calf girth; 5% (n=50) of our volunteers “fitted” an xlp T.E.D. stocking. None of our
volunteers with xxl calves had thighs big enough to “fit” the xxl thigh size range.
Figure 1 also shows that the medium, large and extra large sizes of thigh length T.E.D.
stocking with belt are identical to the standard thigh length T.E.D. stockings. The extra small
(xs) thigh length with belt and small T.E.D. stockings in both ranges share the same thigh
size (less than 63cm) and unspecified ankle girth. However, the calf girth range is subdivided in the thigh length with belt T.E.D. stocking: the xs calf size is less than 25cm and the
small (2s) calf size ranges from 25 to 30.5cm. None of our volunteers “fitted” the xs thigh
length with belt T.E.D. stocking as the smallest calf we measured had a girth of 29cm.
Urgo Preventex
Urgo manufacture Preventex brand AES in 3 styles: below knee, thigh length and thigh
length with belt. Figure 2 shows the 5 sizes of below knee and 4 sizes of thigh length
Preventex AES (both styles are produced in identical sizes) above the plots of our
volunteers’ leg sizes. The calf size ranges are identical in all styles for small to xl stockings
but the xxl size is only available in below-knee style, the ankle size is identical for ‘small’ to
‘large’ stockings in all styles. However, the ankle size for both xl thigh length styles is
greater than 26cm, while in below knee stockings the xl ankle range is 25-28cm and the xxl
size (below knee only ) “fits” ankles with a circumference greater than 28cm.
Table 1 showed that 82% (n=776) of our volunteers’ legs “fitted” at least one size of
Preventex below knee AES; this was the second highest “fit” coefficient for our sample.
17% (n=159) of our volunteers had ankles and calves that matched different AES sizes at the
ankle and calf measurement points, while 1% (n=7) had calves that were bigger than the
upper limit for the below-knee xxl size. Preventex AES medium to xl ankle and calf sizes
were specified in narrow bands that did not overlap each other significantly, as shown in
Figure 2. This limited the variety of volunteers’ leg profiles that would “fit” these sizes.
However, the open-ended ankle sizing and wider calf girth range on the xxl Preventex AES
size meant the level of “fit” for this size was better than it would have been if entirely closed
and narrow size ranges had been used.
400
l
m
s
350
300
s
number of legs with specified girth
250
200
s
m
l
xl
l
m
xxl
xl
xl
xxl
xl
thigh girths for Preventex AES
calf girths for Preventex AES
ankle girths for knee length Preventex AES, small to xxl
ankle girths for thigh length Preventex AES, small to xl
volunteer's
ankle girths
150
100
volunteer's
calf girths
50
0
16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91
volunteer's
thigh girths
leg and AES girth in cm at ankle, calf and thigh
Figure 2 – Plot showing volunteers’ leg girths and Preventex AES size ranges at ankle, calf
and thigh
Figure 2 shows that the thigh size range of Preventex AES was wide enough to
accommodate all our volunteers’ thighs, with the majority (89%, n=839) having a girth less
than 63cm and therefore “fitting” small, medium and large AES at the thigh. As with T.E.D.
stockings nearly all our volunteers who “fitted” a small, medium or large below knee
Preventex stocking also “fitted” a thigh length Preventex stocking as shown in Table 1. The
potential implications of open-ended sizing are discussed at the end of this paper.
Figure 2 shows that large Preventex AES matched more of our volunteer’s ankles (62%,
n=585) and calves (53%, n=500) than any other size and that the small size matched very
few of our volunteers’ ankles (0.1%, n=1) and calves (3%, n=30). However, Figure 2 also
shows that Preventex AES size ranges do not provide adequate provision for our volunteers’
calf girths. In fact 1% (n=7) of our volunteers’ calves were larger than the below knee xxl
size and 6% (n=56) of our volunteers’ calves were larger than the thigh length xl size.
Therefore, patients with large calf girths could only be “fitted” with below knee, if any,
Preventex AES. Furthermore, and of significance to the thigh length products, 29% (n=277)
of our volunteers’ calf girths were between 40 and 46cm matching the xl size range,
whereas only 11% (n=103) had thigh girths of 63cm or more, matching the xl thigh size
range. Therefore, there may be scope to further increase the “fit” coefficients of Preventex
AES by adjusting the size range boundaries on the basis of a larger scale study.
Arjo Huntleigh Flowtron anti-embolism stockings
Arjo Huntleigh produce Flowtron AES in 3 styles: below knee; thigh length and belted/full
length stockings. Each style of stockings is available in 7 sizes and 3 lengths. The girth range
for each stocking size at ankle, calf and thigh do not vary between the different styles or
lengths of stockings and are shown in Figure 3.
Table 1 showed that, despite having more sizes than any other manufacturer in this study,
Flowtron below-knee stockings had a fit coefficient of only 74% (n=701), with 26% (n=241)
of our volunteer’s legs not “fitting” any of the same Flowtron sizes at both measurement
points. This relatively low level of “fit” for below- knee stockings is principally due to their
use of exclusively closed (i.e. defined upper and lower limits) and relatively narrow size
ranges with little overlap between sizes at the ankle, as shown in Figure 3. Further, when
the number of measurement points increased from 2 for below knee products to 3 for thigh
length products, the number of legs with the required profile to match the stockings
decreased significantly. Only 27% (n=257) of the legs we measured matched the Flowtron
AES size chart at all 3 measurement points, with 73% (n=683) not “fitting” a Flowtron thigh
length size. This low “fit” coefficient was largely due to their exclusive use of closed size
ranges at 3 measurement points. The flowtron sizing structure means that a limited leg
shape is catered for in 7 different sizes. Legs that were more conical or more cylindrical
than the Flowtron size provision are not catered for by these stockings.
400
350
s
xs
300
xs
number of legs with specified girth
250
200
xs
s
m
l
xl
s
xxl
m
xxxl
l
xl
xxl
m
xxxl
l
xl
xxl
xxxl
thigh girths for
Flowtron AES
calf girths for
Flowtron AES
ankle girths for
Flowtron AES
volunteer's
ankle girths
150
volunteer's
calf girths
100
50
volunteer's
thigh girths
0
16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91
leg and AES girth in cm at ankle, calf and thigh
Figure 3 – Plot showing volunteers’ leg girths overlaid with Flowtron AES size ranges at
ankle, calf and thigh
Despite the relatively low “fit” coefficients, less than 1% of our sample had leg girths that
didn’t match any ‘Flowtron AES’ size at a single measurement point: 2 ankles and 7 calves
required a stocking larger than Flowtron’s xxxl and 7 thighs were smaller than the xs size
range.
H&R Healthcare Carolon anti-embolism stockings
Carolon produce below knee AES in 5 sizes and standard thigh length AES in 4 sizes, both
styles have the same ankle and calf girth ranges for sizes small to xl, however no xxl
standard thigh length size is available. The girth ranges of these sizes for below knee and
standard thigh length AES are shown in black above plots of our volunteers’ legs in Figure 4.
Figure 4 shows that the ankle girths of Carolon below knee and standard thigh length AES
are designed to accommodate a narrow range of ankle girths with very little overlap (0.5cm)
between the sizes. In our sample, 62 ankles (7%) were bigger than the 30.5cm upper limit
of Carolon’s below knee xxl AES stocking and 182 (19%) of our sample ankles were bigger
than the 28cm upper limit of the standard thigh length xl size. Similarly, 2% (n=20) and 13%
(n=119) of our volunteers’ calves were too large for the xxl below knee and xl thigh length
AES, respectively. However, 5% (n=46) of our volunteers’ thighs were smaller than
Carolon’s small thigh length stocking; none of our sample had thighs that were too large for
the xl size. Therefore, Carolon’s sizing was both too limited to accommodate the measured
range of ankle, calf and thigh girths but they had also anticipated different shapes of leg at
ankle, calf and thigh to the majority of those we measured. In addition they used
(inconsistent) narrow size ranges with limited overlap that provide limited flexibility in
accommodating a range of leg shapes. These issues with Carolon sizing for our sample are
reflected in Table 1, which shows that Carolon below knee AES had a “fit” coefficient of 60%
(n=566) and their standard thigh length AES had a “fit” coefficient of only 6% (n=60).
450
s
400
s
350
300
s
s
number of legs with specified girth
250
l
200
xl
m
s
l
m
s
xxl
xxl
xl
m
m
xxl
xl
l
l
xl
m
m
l
l
xl
xl
xxl
thigh girths for
Carolon ATS AES
thigh girths for standard
Carolon AES
calf girths for Carolon ATS AES
xxl
calf girths for standard Carolon AES
ankle girths for Carolon ATS AES
ankle girths for standard Carolon AES
volunteer's
ankle girths
150
volunteer's
calf girths
100
volunteer's
thigh girths
50
0
16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91
leg and AES girth in cm at ankle, calf and thigh
Figure 4 – Plot showing volunteers’ leg girths and Carolon AES size ranges at ankle, calf and
thigh
Carolon also produce a range of Adjustable Thigh length Stockings (ATS) available in 5 sizes;
these are shown in grey above our volunteers’ leg girths in Figure 4. The ATS stockings are
intended for bariatric patients, those with oedema and over bandages following surgery.
Therefore, Carolon ATS cater for very large thighs, as a result of which 17% (n=163) of our
volunteers had thighs that were too small for Carolon ATS stockings and 75% (n=702) of our
volunteers matched the ‘small’ thigh size, as might be expected. However, as Figure 4
shows, the sizing of Carolon ATS stockings at ankle and calf was not significantly larger than
their standard stockings, with 7% (n=62) of our volunteers’ ankles and 2% (n=26) of their
calves being too large for Carolon ATS stockings. This massive offset between the belowknee and above-knee measurements meant that none of our volunteers would “fit” a
Carolon ATS stocking at all measurement points (“fit” coefficient of 0%). It may be that
Carolon ATS cater for an important minority of patients that we did not encounter but it
seems odd that the ankle and calf girths should be so restricted when oedema and
bandages are commonly found in the lower leg and bariatric patients often have large lower
legs as well as thighs.
Discussion
Number of measurement points
As might be expected, when the number of circumferential measurement points that need
to be compared to size charts increases from 1 (calf only) to 2 (ankle and calf or calf and
thigh) or 3 (ankle, calf and thigh) the match or “fit” rate decreased. The only style in this
study with 1 measurement point , the below-knee T.E.D. stocking, had a “fit” coefficient of
100%. Products with 2 measurement points, Carolon, Preventex and Flowtron below-knee
and T.E.D. thigh length AES, had “fit” coefficients ranging from 62% to 87%; while products
with 3 measurement points, Carolon, Preventex and Flowtron thigh length AES, had “fit”
coefficients ranging from 0 to 61%.
By not specifying a measurement at all, e.g. T.E.D. stockings unspecified ankle girth, the
process of deciding which stocking to “fit” is much simpler, since fewer leg measurements
are required and a match between below knee AES and patient dimensions is almost
guaranteed. However, the erroneous implication to nurses fitting these stockings is that the
ankle girth does not matter and/or that the stockings will “fit” any ankle. All AES
manufacturers strive to design and manufacture stockings with stiffness profiles that
minimise the difference in pressure delivered to different sized legs. However, it is unlikely
that stockings would be able to deliver the same level of compression to all legs within an
un-specified size range. Therefore, legs may receive more or less compression than is
desirable and in some instances negative pressure gradients are likely to occur (MacLellan,
2002), that decrease blood flow rather than enhance it. Both MacLellan (2002) and Thomas,
Toyick and Fisher (2000) showed that the majority of T.E.D. stockings tested provided
significantly lower pressures at the ankle than the accepted ideal 18mmHg when tested at
‘in-proportion’ ankle circumferences. Thomas, Toyick and Fisher showed that these ankle
pressures varied by 7 to 24% when the ankle circumference was changed by between only
1.5 and 2.3cm. Figure 1 shows that the ankles we measured ranged from 19 to 44cm, a
difference of 25cm.
Open-ended sizing
Figures 1 and 2 show that Preventex and T.E.D. brand AES both use open-ended sizing (as
indicated by a single headed arrow) in their products, this meant that no ankle, calf or thigh
was considered too small for these brands and no ankle was too large. This led to
considerably higher “fit” coefficients for Preventex and T.E.D. stockings, 82% and 100% for
below-knee stockings and 61 to 87% for thigh length stockings respectively, than for Carolon
and Flowtron in both below-knee and thigh length styles.
For example, both T.E.D. and Preventex specify that their small, medium and large thigh
length stockings “fit” thighs smaller than 63.5 or 63cm respectively. If these stockings were
able to deliver the same pressure to all these thighs, this open-ended sizing would be very
convenient as it maintains high levels of “fit” to a wide range of leg shapes and sizes.
However, this is very difficult to achieve in practice as demonstrated by MacLellan (2002)
and Thomas, Toyick and Fisher (2000). In our study the thighs that would “fit” these
stockings ranged from 36 to 63 cm in circumference and it is likely that if the stockings were
designed not to produce a tourniquet effect on larger thighs they would tend to slip down
the smaller thighs.
As a matter of interest, if Flowtron were to adopt the open-ended thigh sizing used by T.E.D.
and Preventex in their 3 smallest sizes, they would increase the proportion of our volunteers
who “fitted” their thigh length stockings from 27% (n=257) to 64% (n=603). However,
Flowtron’s existing narrow size ranges, with defined upper and lower limits, means there is
a greater likelihood that their stockings will achieve the ideal pressure profile on legs that
“fit” their sizes.
Adjustment of size bands based on leg measurement data
When our volunteers’ legs did not match an AES size at all measurement points it was
because their legs had a different shape profile to the brand’s size range. In some instances
this was due to the normal variation in human shape, which would be difficult to address in
a standardised product, but in some instances the brand’s size ranges appeared to be out of
alignment with the majority of our volunteers. For example, 62% (n=585) of our volunteers’
ankles and 56% (n=528) of their calves matched Flowtron’s medium size range, as indicated
in Figure 3. However, 61% (n=572) of our volunteer’s thighs matched Flowtron’s small size
range at the thigh. If Arjo Huntleigh were to re-assign the thigh size ranges of Flowtron AES,
and re-design the thigh portion of their stockings accordingly, so that what is currently their
xs size became their small, what is currently their small became their medium, etc., then the
“fit” coefficient of this sample for their thigh size AES would increase from 27% (n=257) to
60% (n=565). Clearly no such changes should be made on the basis of this relatively small
sample, but these results indicate that significant improvements could be made to product
sizing with access to the results of a large scale survey.
Conclusions
Different brands and styles of anti-embolism stockings “fitted” different proportions of our
volunteers’ legs, ranging from 100% at best to 0% at worst. Each brand studied here
manufactured below-knee stockings that “fitted” more than half of our volunteers’ legs but
only 2 brands produced thigh length stockings that “fitted” more than half of our
volunteers’ legs. The difference in the proportion of our volunteers that “fitted” each brand
of stockings was largely due to different sizing policies by different brands:
1. Reducing the number of measurement points on the leg increases the likelihood that the
leg measurements provided will match the stocking’s size chart and a “fit” can be
achieved. However, by not providing a size range for an important leg dimension, such
as the ankle, the quality of “fit” is unknown and the pressure profile exerted by the
stocking may not be the same as that measured in the laboratory during product
development or quality control.
2. Open-ended size ranges will increase the number of patients that a brand of AES “fits”
however they may not exert the same pressure or pressure profile on all legs that
appear to “fit” the stocking.
3. Wide size ranges that overlap between sizes are more likely to “fit” more people than
narrow size ranges with little overlap, but again this may give a wider range of pressures
on the leg.
4. Improvements could be made to brand’s “fit” coefficients with access to a large sample
of patient’s leg measurements.
These results are based on an analysis of a relatively small random sample of UK resident’s
legs. However, we hope that the significant variation in the “fit” of different AES brands to
these legs demonstrates the need for a large scale, international survey of leg sizes and
similar analysis to be conducted before hospitals select the brands they wish to stock in
order to “fit” the widest possible number of patients. Ideally such a survey of leg sizes
would be conducted on patients as “unusual leg shapes” may be more prevalent in
hospitalised patients than in the general population. The results of such a study should be
used to inform brand’s sizing and product developments in addition to hospital’s choice of
brands and sizes.
References
Arjo Huntleigh, 2012. http://www.arjohuntleigh.com/Page.asp?PageNumber=3980
BSI, 1993. British Standards Institute. BS 7672: 1993 Specification for Compression,
stiffness and labelling of anti-embolism hosiery. 1993. 1-9.
Carolon, 2009. http://www.hrhealthcare.co.uk/downloads/carolon/Test-Results-vs-SiegelProfile.pdf
Dumbleton, T., Clift, M., 2008. Buyers Guide: Compression Hosiery. CEP08036. NHS
Purchasing and Supply Agency. Accessed at http://www.derbyatec.co.uk/buyers-guidecompression-hosiery-now-published/ on 7-6-11. p1-73
Geerts, W.H., Bergqvist, D., Pineo, G. F., Heit, J.A., Samama, C.M., Lassen, M.R., Colwell,
C.W., 2008. Prevention of Venous Thromboembolism*: American College of Chest
Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). CHEST 2008; 133; 381453S. Downloaded from
http://chestjournal.chestpubs.org/content/133/6_suppl/381S.full.html
Guyatt, G.H., Akl, E.A., Crowther, M., Gutterman, D.D., Schuunemann, H.J., and for the
American College of Chest Physicians Antithrombotic Therapy and Prevention of Thrombosis
Panel, 2012. Executive Summary: Antithrombotic Therapy and Prevention of Thrombosis,
9th Edition: American College of Chest Physicians Evidence-Based Clinical Practice
Guidelines. Chest 2012; 141; 7S-47S. Downloaded from:
http://chestjournal.chestpubs.org/content/141/2_suppl/7S.full.html
MacLellan, D.G., 2002. Compression Profiles of Antiembolic Stockings. Australian and New
Zealand Journal of Phlebology. 6 (1), pp9-14.
NICE, 2007. NICE guideline 46: Venous thromboembolism Reducing the risk of venous
thromboembolism (deep vein thrombosis and pulmonary embolism) in inpatients
undergoing surgery. April 2007. accessed at
http://www.spitalmures.ro/_files/protocoale_terapeutice/chirurgie/reducerea_riscului_de_
embolie_la_pacientii_chirurgicali.pdf on 20-9-11
NICE, 2010. National Clinical Guideline Centre – Acute and Chronic Conditions. Venous
thromboembolism: reducing the risk of venous thromboembolism (deep vein thrombosis
and pulmonary embolism) in patients admitted to hospital. 2010. Accessed at
www.nice.org.uk/guidance/CG92/Guidance on 13-9-11. p45-62 and p144-68
Preventex, 2012. ‘Preventex anti-embolism stockings’ product leaflet. Urgo Medical, Urgo
Limited, Loughborough.
RAL , 2008. RAL Deutsches Institut fur Gutesicherung und Kennzeichnung e.V.. RALGZ_387/1. Medical Compression Hosiery. Sankt Augustin.
Sigel, B., Edelstein, A., Savitch, L., et al., 1975. Type of compression for reducing venous
stasis . A study of lower extremities during inactive recumbency. Archives of Surgery. 110:
171-175.
SIGN, 2010. Scottish Intercollegiate Guidelines Network (SIGN). Prevention and
management of venous thromboembolism. Edinburgh. (SIGN publication number 122).
Thomas, S., Toyick, N., Fisher, B., 2000. Graduated External Compression and the
Prevention of Deep Vein Thrombosis. Surgical Materials Testing Laboratory, Bridgend.