625404

research-article2015
JHS0010.1177/1753193415625404Journal of Hand SurgerySassi and Giddins

Full Length Article

JHS(E)
The Journal of Hand Surgery

Gender differences in carpal tunnel (European Volume)

XXE(X) 1­–5
© The Author(s) 2016
relative cross-sectional area: a possible Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav

causative factor in idiopathic carpal DOI: 10.1177/1753193415625404

jhs.sagepub.com

tunnel syndrome

S. A. Sassi and G. Giddins

Abstract
Previous research has not established a consistent difference in hand size or carpal tunnel cross-sectional
area between patients with and without carpal tunnel syndrome. We tested the hypothesis that there would be
no difference in relative carpal tunnel sizes between men and women. We defined relative carpal tunnel size as
the cross-sectional areas at the inlet (level of the pisiform) and outlet (level of the hook of the hamate) of the
carpal tunnel divided by the length of the capitate (as a measure of hand size). We made the measurements on
the magnetic resonance imaging scans of 50 men and 50 women taken for symptoms unrelated to carpal tunnel
syndrome. The mean relative cross-sectional area was appreciably smaller in women than men (p < 0.05).
This suggests that the carpal tunnel cross-sectional area relative to the size of the hand is constitutionally
smaller in women than in men. This could in theory be a significant factor in patients developing carpal tunnel
syndrome.
Level of evidence: V

Keywords
Carpal tunnel syndrome, carpal tunnel cross-sectional area, carpal tunnel cross-sectional area ratio

Date received: 3rd June 2014; revised: 4th December 2015; accepted: 8th December 2015

Introduction
By definition the cause of idiopathic carpal tunnel
syndrome (CTS) is unknown. There are many recog-
nized associations with CTS, notably pregnancy, dia-
betes, some occupations (Kim et al., 2004) and obesity
(Bland, 2005). The strongest association is with
female gender. Various causes for the much higher
incidence in women have been proposed, including
hormone balance (Song et al., 2014); and in particu-
lar, wrist and hand size (Farmer and Davis, 2008).
Some studies have found an association with the
cross-sectional area and shape of the carpal canal
(Boz et al., 2004; Chiotis et al., 2013) but most have
not, whether measuring external hand size (Bleecker
et al., 1985) or with investigations such as ultrasound
(Kim et al., 2012) or magnetic resonance imaging
(MRI) (Pierre-Jerome et al.,1997).
Chiotis et al. (2013) examined hand size and carpal
tunnel cross-sectional area with ultrasound, in
patients with and without CTS. They calculated carpal
tunnel ratios by dividing the depth of the cross-sec-
tion of the canal by the width. They found a correlation
between the carpal tunnel inlet and outlet ratios and
CTS. Specifically they found that patients with CTS
had larger carpal tunnel cross-sectional areas, but
offered no explanation why. In essence their carpal
tunnel ratio measured whether the canals were more
elliptical or more circular. They showed a link between
more circular canals and CTS, and suggested this
may alter the effect on the median nerve of raised
carpal canal pressure. The reasons for this associa-
tion remain unclear.
Cobb et al. (1997) studied carpal tunnel cross-sec-
tional area on MRI scans and compared them with
the volumes of the contents of the carpal tunnel. They
found that in patients with CTS, the contents of the
Orthopaedic Department, Royal United Hospital, Bath, UK
Corresponding author:
S. A. Sassi, Orthopaedic Department, The Royal United Hospital,
Bath BA1 3NG, UK.
Email: sassi.sassi@nhs.net

Downloaded from jhs.sagepub.com at EMORY UNIV on March 28, 2016

2 The Journal of Hand Surgery (Eur)

Table 1.  Reasons for the MRI scan request.

Reason for MRI No of men No of women

Ligament injury?
Carpal bone fracture?
Non traumatic wrist pain
TFCC injury?
Miscellaneous
7 6
22 18
7 14
8 7
6 5

MRI: magnetic resonance imaging; TFCC: triangular fibrocartilage

complex.

tunnel took up a greater percentage of the available

cross-sectional area. This implies that the contents
are greater in patients with CTS, which would fit with
Figure 1. Measurement of carpal tunnel surface area at
a higher pressure. It is not clear whether this is cause the level of the pisiform.
or effect, and does not establish the significance of
the cross-sectional area of the carpal tunnel in CTS.
The increased incidence of CTS in women com-
pared with men is well recognized. If there was a link
between idiopathic CTS and the relative size of the
carpal canal, then it would be expected that the canal
sizes in women would be smaller relative to the size
of the hand than in men.
The aim of this study was to test the null hypothe-
sis that the relative cross-sectional areas (RCSAs) of
the carpal tunnel in men and women would not be
different.

Methods and materials

We reviewed the wrist MRI scans of 50 adult
(>18 years old) men and 50 adult women scanned at
our hospital. Exclusion criteria were: scans of inade-
quate quality to make reliable measurements; scans
not including the full extent of the carpal canal;
masses in the canal that might affect the carpal tun-
nel cross-sectional area; carpal or distal radius
abnormalities that might affect the carpal tunnel
cross-sectional area; scans for acute trauma; and
evidence of previous carpal tunnel surgery. We also
checked the operative records in our hospital and at
other providers to ensure that none of these patients
had undergone carpal tunnel release that might
affect the carpal tunnel cross-sectional area.
All of the scans we reviewed were performed using
1.5T Siemens Avanto scanner on the routine wrist pro-
tocol (using dedicated wrist coil). This protocol includes:
T1 coronal, T2 medic three-dimensional coronal, Short
Tau Inversion Recovery (STIR) coronal, proton density
sagittal and STIR axial. The slice thickness obtained
using this protocol is 3 mm, apart from the T2 medic
three-dimensional coronal which produces 1 mm slices.
We recorded the age and gender of each patient,
the hand that was scanned and the reason for the
scan as written on the request form (Table 1). We
Figure 2. Measurement of carpal tunnel surface area at
the level of the hook of hamate.
made measurements using the area measurement
software built into the Picture Archiving and Com-
munication System (PACS) system. We measured the
cross-sectional areas at the inlets and outlets of the
carpal canals, i.e. the level of the pisiform and the level
of the hook of the hamate (Figures 1 and 2). These are
recognized sites of measurement (Allmann et al.,
1997) and are easily defined on the scans. We also
measured the lengths of the capitate bone (the long-
est vertical distance between the proximal and distal
articulating surfaces of the capitate – Figure 3). We
used the capitate because the wrist MRI scans typi-
cally do not include the entire length of the middle
finger metacarpal (the most recognized surrogate for
hand size). Capitate length appears to be a reasona-
bly reliable alternative indicator of hand size (Nattrass
et al. 1994). In their article, Nattrass et al. showed
that a modified carpal height ratio (measured as the
carpal height divided by the length of the capitate)
is nearly as accurate as the standard carpal height
ratio defined by Youm et al. (1978) (measured as the

Downloaded from jhs.sagepub.com at EMORY UNIV on March 28, 2016

Sassi and Giddins 3
Figure 3.  Measurement of the length of the capitate.
carpal height divided by the middle finger metacarpal
length). By implication, the length of the capitate is a
reasonably reliable surrogate measure of hand size,
similar to the length of the middle finger metacarpal.
The RCSAs were calculated by dividing the carpal tunnel
cross-sectional areas by the length of the capitate.
The measurements were repeated in 20 cases
with a delay of 3 months to check for intra- and inter-
observer reliability.
Statistical analysis
The data were normally distributed so we analysed
them using student’s t-test.
Results
We assessed 50 men and 50 women. The men had a
mean age of 36 years (range 19–79). The women had
a mean age of 36 years (range 18–75) (Appendix 1,
available online). The right wrist was scanned in 32 of
the men and 43 of the women.
At the carpal tunnel inlet (the level of the pisiform),
the mean RCSA in men was 11.3 (range 7.3–15.6)
and in women it was 9.0 (range 5.8–12.1) (p < 0.05)
(Figure 4). At the carpal tunnel outlet (the level of the
hook of hamate), the mean RCSA in men was 10.9
(range 7–14.9) and in women it was 8.5 (range 5.4–
11.6) (p < 0.05) (Figure 5).
Twenty measurements were repeated by the first
author, 3 months apart to test for intra-observer
variability; the intra-class correlation coefficient was
Downloaded from jhs.sagepub.com at EMORY UNIV on March 28, 2016
0.97. Measurements were also made by the second
author on 20 scans; the intra-class correlation coef-
ficient, for inter-observer variability, was 0.86.
Discussion
Previous studies have looked at carpal canal size and
hand and wrist morphology to assess whether there
is a difference between men and women to account
for the gender difference in the prevalence of CTS.
Previous studies have not found a difference. No pre-
vious studies have calculated carpal canal RCSAs as
a measure of relative canal size, i.e. compared with
the size of the patient’s hand. Our hypothesis is that
patients who develop CTS do so in part because their
carpal canal cross-sectional area is relatively smaller
than average. This trait could help explain the famil-
ial link to CTS as relative canal size could be inher-
ited. Furthermore, we hypothesize that the sizes of
the contents of the carpal canal will be proportionate
to the size of the hand, i.e. the cross-sectional areas
of the median nerve and flexor tendons will be pro-
portionate to hand size. If so, then the contents would
take up a larger part of the cross-sectional area of
the canal in patients with relatively smaller carpal
canals and so render them more susceptible to
developing CTS. We were not able to measure the vol-
umes of the carpal canal structures sufficiently accu-
rately to assess this.
We tested the null hypothesis that there was no
difference in relative canal size between men and
women. The data clearly show this can be rejected;
thus women have significantly smaller carpal tunnel
RCSAs than men. Women have smaller carpal tun-
nels than men; that is because they have smaller
hands. We have shown that, relative to the size of
their hands, women have disproportionately smaller
carpal tunnels. This correlation suggests that carpal
canal size may be important in the development of
CTS. This is however not proven, as none of the
patients in this study had clinical evidence of CTS,
although some may develop CTS in future. The data
suggest that if there is a strong link with canal size
and CTS then there may be a cut-off at a RCSA of
around 8–9, where smaller ratios are more common
in women than men by a factor of at least 4; i.e. at or
below that ratio women would be four or more times
more likely to develop CTS, which fits with clinical
experience. As yet this is clearly not proven, but can
only be regarded as speculation. It is however the
potential basis for further investigation.
There are weaknesses in this study. We assessed
100 patients; this is a large sample size but could be
bigger. The patients may have had MRI scans for rea-
sons that predisposed them to smaller canal ratios,
4 The Journal of Hand Surgery (Eur)

Figure 4.  Combined graph for male and female inlet ratios, i.e. at the level of pisiform.
RCSA: relative cross-sectional areas.

Figure 5.  Combined graph for male and female outlet ratios, i.e. at the level of hook of hamate.
RCSA: relative cross-sectional areas.

but there is no evidence of this. We only measured
the canal ratios at the inlet and outlet of the carpal
tunnel. The data were very consistent, but physical
factors important in the development of CTS may also
include canal shape, canal length and dynamic
changes in the canal. We did not measure the size of
the contents of the canals as the images and soft-
ware made this difficult to achieve. Finally, the MRI
slices were every 3 mm, which would potentially
introduce a measurement error as it makes it more
difficult to measure the CSA at the same site. This
should, in theory, reduce the differences in the data
rather than increase them.
In conclusion, this study appears to show a clear
normal distribution of canal RCSA at the inlet and
outlet of the carpal canal with canal RCSA in women
significantly smaller than in men. If this is clinically
relevant to CTS, then patients with CTS should

Downloaded from jhs.sagepub.com at EMORY UNIV on March 28, 2016

Sassi and Giddins 5
typically have smaller canal RCSAs than non-affected
subjects. Investigation of this possible correlation
requires further study.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of
this article.
Funding
The authors received no financial support for the research,
authorship, and/or publication of this article.
Supplementary material
Supplementary material is available at jhs.sagepub.com/
supplemental.
References
Allmann KH, Horch R, Uhl M et al. MR imaging of the carpal tun-
nel. Eur J Radiol. 1997, 25: 141–5.
Bland JD. The relationship of obesity, age, and carpal tunnel syn-
drome: more complex than was thought? Muscle Nerve. 2005,
32: 527–32.
Bleecker ML, Bohlman M, Morland R, Tipton A. Carpal tun-
nel syndrome: role of carpal canal size. Neurology. 1985, 35:
1599–604.
Boz C, Ozmenoglu M, Altunayoglu V, Velioglu S, Alioglu Z. Individual
risk factors for carpal tunnel syndrome: an evaluation of body
Downloaded from jhs.sagepub.com at EMORY UNIV on March 28, 2016
mass index, wrist index and hand anthropometric measure-
ments. Clin Neurol Neurosurg. 2004, 106: 294–9.
Chiotis K, Dimisianos N, Riqopoulou A, Chroni E. Role of anthro-
pometric characteristics in idiopathic carpal tunnel syndrome.
Arch Phys Med Rehabil. 2013, 94: 737–44.
Cobb TK, Bond JR, Cooney WP, Metcalf BJ. Assessment of the
ratio of carpal contents to carpal tunnel volume in patients with
carpal tunnel syndrome: a preliminary report. J Hand Surg Am.
1997, 22: 635–9.
Farmer JE, Davis TRC. Carpal tunnel syndrome: a case-control
study evaluating its relationship with body mass index and
hand and wrist measurements. J Hand Surg Eur. 2008, 33:
445–8.
Kim HS, Joo SH, Han ZA, Kim YW. The nerve/tunnel index: a new
diagnostic standard for carpal tunnel syndrome using sonog-
raphy: a pilot study. J Ultrasound Med. 2012, 31: 23–9.
Kim JY, Kim JI, Son JE, Yun SK. Prevalence of carpal tunnel syn-
drome in meat and fish processing plants. J Occup Health.
2004, 46: 230–4.
Nattrass G, King GJ, McMurty RY, Brant RF. An alternative method
for determination of the carpal height ratio. J Bone Joint Surg
Am. 1994, 76: 88–94.
Pierre-Jerome C, Bekkelund S, Mellgren S, Nordstrom R.
Quantitative MRI and electrophysiology of preoperative carpal
tunnel syndrome in a female population. Ergonomics. 1997, 40:
642–9.
Song CH, Gong HS, Bae KJ, Kim JH, Nam KP, Baek GH. Evaluation
of female hormone-related symptoms in women undergoing
carpal tunnel release. J Hand Surg Eur. 2014, 39: 155–60.
Youm Y, McMurthy RY, Flatt AE, Gillespie TE. Kinematics of the
wrist.I. An experimental study of radial-ulnar deviation and
flexion extension. J Bone Joint Surg Am. 1978, 60: 423–31.