Int Urogynecol J (2010) 21:1265–1270
DOI 10.1007/s00192-010-1183-4
ORIGINAL ARTICLE
Ultrasound measurement of vaginal wall thickness: a novel
and reliable technique
Demetri C. Panayi & G. Alessandro Digesu &
Paris Tekkis & Ruwan Fernando & Vikram Khullar
Received: 14 February 2010 / Accepted: 6 May 2010 / Published online: 26 May 2010
# The International Urogynecological Association 2010
Abstract
Introduction and hypothesis The aim of this study was to
validate a technique to measure the vaginal wall thickness
(VWT) using two-dimensional ultrasound.
Methods Women were scanned by two independent operators and by the same operator at two separate visits at the
level of the bladder neck, the apex of the bladder, the
anterior fornix, the anorectal junction, rectum and posterior
fornix. Fresh female cadavers were scanned and ultrasound
thickness of the vagina was compared to histological
thickness.
Results Bland Altman analysis revealed a low mean
difference between operators and between visits by the
same operator. The 95% confidence intervals as a percentage of the mean vaginal wall thickness ranged between
2.8% and 7.4%. There was a low percentage difference
between ultrasound and histological vaginal wall thickness.
Conclusion Ultrasound vaginal wall thickness demonstrated
good intra- and interoperator reliability, as well as consistency
with histological measurement. It is a valid technique.
Keywords Bland Altman analysis . Intraoperator
reliability . Interoperator reliability . Vaginal wall thickness
Abbreviations
VWT vaginal wall thickness
D. C. Panayi : G. A. Digesu : P. Tekkis : R. Fernando : V. Khullar
Imperial College Healthcare NHS Trust,
London, UK
D. C. Panayi (*)
Dept. of Urogynaecology, St. Marys Hospital,
Praed Street,
London W2 1NY, UK
e-mail: drpanayi@gmail.com
Introduction
Bladder and pelvic floor dysfunction have been linked
to anatomical abnormalities detected with new imaging
modalities such as three- and four-dimensional ultrasound and magnetic resonance imaging [1, 2]. Validation
of these techniques is used to demonstrate their effectiveness and, therefore, their value in clinical practice. In
1986, Bland and Altman proposed a method to evaluate a
new clinical technique [3] as a superior alternative to
correlation coefficient, using simple calculation and
graphical representation. This method can be used to
validate new techniques and, therefore, justify their
clinical application.
The vagina is histologically composed of four layers: the
vaginal mucosa is made up of stratified non-keratinised
squamous epithelial tissue. The vaginal submucosa is
vascularised connective tissue, which does not contain
glands or mucosal muscularis and is mainly composed of
collagen and elastin. The muscularis layer of the vagina
consists of smooth muscle and the adventitia is composed
of loose connective tissue. This connective tissue contains
the extracellular matrix containing collagen and elastin, as
well as ground substance [4].
Studies have assessed the histological and biomechanical
properties of the vaginal tissue of women with prolapse.
Altered elastin [5–7], smooth muscle [8], connective tissue
[9] and collagen [10–12] have all been demonstrated in the
vaginal tissue of women with prolapse. Goh asssessed the
biomechanical properties of vaginal tissue in pre- and postmenopausal women with vaginal prolapse. He found the
only difference between the two groups was that the vaginal
tissue of post-menopausal women showed significantly
higher elastic modulus than tissue of pre-menopausal
women [13].
1266
Studies using ultrasound imaging techniques have
examined many aspects of the pelvic floor, including
levator hiatus and the presence of defects or avulsions and
modelling of the pelvic floor musculature [14–17]. There
are no current studies that have used ultrasound imaging to
measure the vaginal wall itself.
Two-dimensional ultrasound has been used previously to
assess the bladder wall thickness. This technique was first
described in 1994 and has been validated [18, 19]. Since
then, the technique has been developed and the bladder
wall has been measured using the transabdominal, transvaginal and transperineal techniques [20–22]. At the time of
writing, we believe this is the first study to measure the
thickness of the vaginal wall using two-dimensional
ultrasound.
The aim of this study is to validate a novel technique that
uses two-dimensional transvaginal ultrasound to assess the
thickness of the vaginal wall as described in the methods.
The validation would be from intra- and inter-observer
analysis, as well as comparison of ultrasound measurements
with histological measurement of cadaveric tissue.
Methods
Women were recruited from attenders to the outpatient
department of a tertiary referral centre for urogynaecology.
A history was obtained and each woman underwent pelvic
examination. Women were excluded if they were pregnant,
had undergone previous vaginal surgery or had any
connective tissue, vaginal or any disease that would affect
the vaginal tissue. Women who were postmenopausal and
on hormone replacement therapy (HRT) were also excluded
from the study, and women who intended to commence on
HRT were asked to defer this till after all scans were
completed. Women who were unable to position themselves
in lithotomy due to muscoskeletal or mobility issues were
also excluded from the study. Informed, written consent
was obtained and women were scanned using a 7.5-MHz
transvaginal probe (Voluson I, GE Healthcare, Milwaukee,
WI, USA).
This technique consisted of scanning women in the
lithotomy position whilst performing a valsalva manoeuvre,
within 15 min of emptying their bladder and on confirming
by ultrasound that the bladder contained less than 50 ml of
urine. The vaginal wall thickness (VWT) measurement was
obtained at three anatomical landmarks on the anterior
vaginal wall and three sites on the posterior vaginal wall
with the transvaginal scanner orientated in the sagittal
plane. On the anterior vaginal wall, the vaginal wall
thickness was measured at the level of the bladder neck,
apex of the bladder where it is in contact with the anterior
vaginal wall and in the anterior fornix. On the posterior
Int Urogynecol J (2010) 21:1265–1270
vaginal wall, the anatomical sites were the level of the
anorectal junction, the rectum and at the posterior fornix.
The probe was applied to the vaginal wall with a minimum
amount of pressure required for the probe to be in contact
with the vagina. This was to avoid a distortion or pressure
effect on the measurements. Vaginal wall thickness measurements were defined as the perpendicular thickness of
tissue between the transvaginal probe and the anatomical
site measured. The callipers were placed at the edge of the
vaginal wall closest to the probe and at the point closest to
but not touching the organ adjacent to the vaginal wall at
the relevant anatomical point. On the anterior vaginal wall
measuring the vaginal wall thickness at the bladder neck,
for example, the calliper would be placed at the vaginal
wall edge closest to the ultrasound probe, and then at the
point closest to but not touching the bladder neck. Figure 1
shows measurement of vaginal wall thickness at the level of
the bladder neck and anorectal junction.
Women were scanned at an initial visit by two
independent, blinded operators at all six anatomical sites
using the technique described above. Women were then
asked to return after an interval of at least 4 weeks and then
underwent a repeat transvaginal scan by one of the
operators.
Bland Altman analysis [3] was carried out on the data
and the mean vaginal wall thickness, mean difference and
95% confidence intervals of the mean difference were
determined to demonstrate intra- and interobserver repeatability. The 95% confidence interval of the difference of the
mean was calculated as a percentage of the mean vaginal
wall thickness measurement to give an estimation of error.
The pelvic organs of five female cadavers, which came
from the anatomy department, that, in life, had consented to
being used in medical research and were preserved using
Farr solution (6.25% formalin, 6.25% glycerin, 3.125%
phenol, 84.375% water) were dissected out flush with the
inner surface of the bony pelvis. The pelvic tissues were
Fig. 1 Ultrasound image showing measurement of vaginal wall
thickness at i the level of the bladder neck and ii anorectal junction
Int Urogynecol J (2010) 21:1265–1270
1267
carefully dissected free from the posterior surface of the
symphysis pubis and removed from the cadaver. Using the
transvaginal approach in the sagittal plane, the vaginal
tissue was scanned using a 7.5-MHz transvaginal probe
(Voluson I, GE Healthcare) at each of the six anatomical
sites on the vaginal wall, as previously described. Vaginal
wall thickness measurements were taken from each of the
anatomical sites on the anterior and posterior vaginal walls.
Tissue samples were taken from the cadavers from the full
thickness of the anterior vaginal wall and the serosa of the
bladder at the level of the bladder neck, the most cephalad
point of the bladder on the vaginal wall and at the anterior
fornix. Similarly, tissue samples were obtained from the full
thickness of the posterior vaginal wall and to the depth of
the rectal serosa at the level of the anorectal junction,
rectum and posterior fornix. The samples were embedded in
paraffin wax then sectioned and stained with haematoxylin
and eosin. Further sections were taken and the Van Gieson
staining technique was used. This is a connective tissue
stain or trichrome stain. The stained structures were then
measured using a microscope with the 40-times magnification lens and a fixed mark seen while viewing the slide; the
slide was moved to view different parts of the specimen.
Measurements were only carried out in one plane at a time
using the Vernier scale on the slide platform. The vaginal
wall thickness determined by microscopic measurement
was compared to the vaginal wall thickness obtained by
ultrasound measurement of the cadavers.
Ethical approval was obtained from St. Mary’s Local
Research Ethics Committee for this study.
Results
Twenty five women were scanned by the two operators at the
first visit and at a second visit by a single operator as described
in the methods. The mean age of women was 58 years (range:
37–74 years). Mean body mass index (BMI) was 31 (23–42)
The mean and standard deviation of vaginal wall
thickness measurements by the two operators at visit 1
and the mean difference and 95% confidence interval of the
mean difference are shown in Table 1. The 95% confidence
interval of the mean difference as a percentage of the mean
vaginal wall thickness is also shown. This table shows good
consistency between operators for vaginal wall thickness
measurements at each anatomical site, as well as a low
mean difference between operators. The percentage of the
95% confidence interval relative to the mean vaginal wall
thickness ranged from 7.5% to 9.6%, but at all anatomical
sites, the percentage was below 10%.
The mean and standard deviation of vaginal wall
thickness measurements by the single operator at two
separate visits, the mean difference between visits and the
95% confidence interval of the mean difference are shown
in Table 2. The percentage variation of the 95% confidence
interval of the mean difference from the mean is also
shown. Our results show that there was good consistency
between the vaginal wall thickness measurements obtained
by the single operator at each visit. There is a low mean
difference between measurements and the 95% confidence
intervals as a percentage of the mean vaginal wall thickness
ranges between 2.8% and 7.4%.
At the initial visit, both observers were able to
successfully measure the vaginal wall thickness at each
of the anatomical sites on the anterior and posterior
vaginal walls without any missing values. All women
were rescanned at the second visit, and the operator was
able to scan each woman successfully without any
missing values.
Tables 3 and 4 show cadaveric ultrasound and histological measurements of vaginal wall thickness on the anterior
and posterior vaginal wall (n=5). Table 5 shows the
Table 1 The mean and standard deviation of vaginal wall thickness by the two operators at visit 1 (n=25) the mean difference, 95% confidence
interval of the mean difference and the percentage difference from the mean (n=25)
Anatomical
site
Bladder neck
Bladder apex on vagina
Anterior fornix
Anorectal junction
Rectum
Posterior fornix
VWT vaginal wall thickness
Mean VWT in
mm (SD) measured
by operator 1
2.7
2.8
2.9
2.5
2.6
2.9
(0.61)
(0.64)
(0.46)
(0.70)
(0.77)
(0.57)
Mean VWT in mm
(SD) measured by
operator 2
2.7
2.7
2.7
2.5
2.6
2.7
(0.63)
(0.58)
(0.48)
(0.62)
(0.61)
(0.59)
Mean difference
(95% confidence intervals)
0.03
0.07
0.26
−0.20
−0.02
0.20
(−0.14–0.10)
(−0.12–0.14)
(0.14–0.37)
(−0.18–0.04)
(−0.19–0.06)
(0.10–0.31)
Percentage difference of the
95% CI of the mean difference
from the mean (%)
8.9
9.6
8.2
8.8
9.6
7.5
1268
Int Urogynecol J (2010) 21:1265–1270
Table 2 The mean and standard deviation between the vaginal wall thickness measured by operator 1 on two separate visits and 95% confidence
interval of the mean difference between the two visits and the percentage difference from the mean (n=25)
Anatomical
site
Mean VWT in mm
(SD) measured by
operator 1 at visit 1
Bladder neck
Bladder apex on vagina
Anterior fornix
Anorectal junction
Rectum
2.7
2.8
2.9
2.5
2.6
(0.61)
(0.64)
(0.46)
(0.70)
(0.77)
Posterior fornix
2.9 (0.57)
Mean VWT in mm
(SD) measured by
operator 1 at visit 2
2.8
2.9
3.0
2.6
2.7
(0.5)
(0.56)
(0.43)
(0.63)
(0.80)
3.0 (0.52)
Mean difference (95%
confidence intervals)
Percentage difference of the 95%
CI of the mean difference from
the mean (%)
(−0.23–−0.03)
(−0.21–0.03)
(−0.20–−0.01)
(−0.17–0.02)
(−0.18–0.02)
7.1
6.2
6.3
5.8
7.4
−0.12 (−0.23–−0.15)
2.8
−0.1
−0.09
−0.10
−0.07
−0.09
VWT vaginal wall thickness
percentage difference between ultrasound and histological
measurement of vaginal wall thickness at the six anatomical
sites. There was shrinkage of block size measured before
and after staining and processing of each specimen
calculated to be approximately 20%.
Discussion
Validation can be achieved by demonstrating effectiveness
of a tool against a gold standard or showing the
repeatability of a technique. This may be achieved by
demonstrating agreement between observers when exposing the patient to a tool or technique independently without
awareness of the other's findings. This would show interobserver repeatability. Alternatively, one can show consistency with a single observer at separate exposures of a
patient to a technique or evaluation tool for intra-observer
repeatability. In this way, test–retest or intra-observer
repeatability can be demonstrated.
In this study, measurement of mean vaginal wall
thickness between two separate blinded operators has
good repeatability. This is demonstrated by the low
mean difference and 95% confidence intervals between
observers; in particular, the vaginal wall thickness at the
level of the bladder neck, bladder and rectum showed
very low mean differences. The 95% confidence
intervals of the mean difference when expressed as a
percentage of the mean vaginal wall thickness were less
than 10% for all three anatomical sites on the anterior
and posterior vaginal walls. This represents good
clinical inter-observer reliability of this technique.
When considering intra-observer repeatability, we have
shown that the mean difference between visits by the
same operator was low, with narrow 95% confidence
intervals. Specifically, the mean differences of mean
vaginal wall thickness at the bladder, rectum and
anorectal junction were particularly low. The 95%
confidence interval of mean difference when expressed
as a percentage of the mean vaginal wall thickness was
lower than that of the inter-observer data and well below
the 10% threshold for reliability. These inter-observer data
show good repeatability and clinical reliability. Measurement of the vaginal wall thickness at the posterior fornix
showed a percentage variation of only 2.8%, which
represents excellent clinical reliability.
We are unable to explain why different anatomical sites
have different mean differences in the intra-observer or
Table 3 Ultrasound and histological measurements of vaginal wall thickness, in female cadavers A to E at three anatomical points on the anterior
vaginal wall: bladder neck, bladder apex and anterior fornix
Cadaver
A
B
C
D
E
USS VWT
bladder neck
Histology VWT
bladder neck
USS VWT
bladder apex
Histology VWT
bladder apex
USS VWT
anterior fornix
Histology VWT
anterior fornix
2.3
2.4
2.6
3.2
2.8
1.9
2.0
2.3
2.8
2.4
2.6
3.2
3.2
2.6
3.0
2.3
2.9
3.0
2.3
2.6
3.1
3.7
2.9
3.0
3.2
2.8
3.2
2.5
2.7
2.9
All measurements in millimetres (n=5)
USS ultrasound, Hist histological, VWT vaginal wall thickness
Int Urogynecol J (2010) 21:1265–1270
1269
Table 4 Ultrasound and histological measurements of vaginal wall thickness, in female cadavers A to E at three anatomical points on the
posterior vaginal wall: anorectal junction, rectum, and posterior fornix
Cadaver
USS VWT anorectal
junction
Histology VWT
anorectal junction
USS VWT
rectum
Histology VWT
rectum
USS VWT
posterior fornix
Histology VWT
posterior fornix
3.0
2.3
2.6
3.0
3.2
2.7
2.0
2.1
2.5
2.8
3.5
2.5
2.9
3.4
3.6
3.2
2.2
2.6
3.1
3.3
3.3
2.9
3.2
3.1
3.3
2.9
2.6
2.9
3.0
2.8
A
B
C
D
E
All measurements in millimetres
USS ultrasound, Hist histological, VWT vaginal wall thickness, ARJ anorectal junction, PF posterior fornix
inter-observer data. One would expect that measuring the
vaginal wall thickness at specific anatomical sites such as
the bladder neck and the anorectal junction would produce
closer inter- and intra-observer agreement when compared
to measurements at the anterior and posterior fornices, for
example, which have more scope for variation in probe
placement, and there is no specific anatomical landmark.
However, this was not borne out by our data.
The technique is also able to obtain measurements
without difficulty, which is supported by our findings that
neither operator was unable to measure the vaginal wall
thickness at any anatomical site or in any individual
woman. The single observer was also able to obtain repeat
measurements without any missing values. We found that
transvaginal ultrasound measurement of vaginal wall
thickness was applicable to all women in the study
irrespective of demographic factors, including body mass
index. The technique was well tolerated, easy to learn and
quick to perform and is therefore suitable to the outpatient
setting. We acknowledge that a weakness of the technique
is that it was not possible to standardise the pressure
applied by the vaginal probe to the vaginal walls, and
therefore, we could not account for distortion effect on the
measurements. However, the technique was to apply the
minimal amount of pressure possible and the reliability of
the technique demonstrated in this study suggests that the
possible distortion effect is minimal.
The cadaveric data show a close relationship between the
ultrasound measurements of vaginal wall thickness at each
of the anatomical sites on the anterior and posterior vaginal
wall. Taking into account the shrinkage associated with
staining and processing of the tissue onto slides, there was a
close relationship between the findings on histological
measurement and the measurements taken using ultrasound
on cadavers. This further supports the validity of this
technique.
We believe that this technique is therefore valid and has
applications in clinical practice. The vaginal wall has been
evaluated using other techniques in the assessment of
women with vaginal wall prolapse, and ultrasound measurement of the vaginal wall may be useful in these
patients. We believe it may also be valuable preoperatively to surgeons contemplating vaginal wall repair
in women with prolapse, especially in post-menopausal
women or women who have undergone previous repair.
Conclusion
Transvaginal ultrasound measurement of the vaginal wall
thickness demonstrated good inter- and intraobserver
reliability and is a valid technique. The measurements
obtained also related closely with histological measurements of vaginal wall thickness in cadavers. This method of
Table 5 The percentage difference between ultrasound and histological measurements of vaginal wall thickness at the six anatomical
sites measured
Cadaver
A
B
C
D
E
Bladder neck (%)
Bladder apex (%)
Anterior fornix (%)
Anorectal junction (%)
Rectum (%)
Posterior fornix (%)
21
20
13
14
17
13
10
7
13
10
11
15
24
20
1
11
15
24
20
14
9
14
12
10
9
14
12
10
3
14
1270
measuring vaginal wall thickness can therefore be used in
future studies of the vaginal wall. This is a new technique
for evaluating vaginal wall thickness which could be used
for assessing vaginal wall thickness in women with
prolapse.
Int Urogynecol J (2010) 21:1265–1270
9.
10.
11.
Conflicts of interest D. C. Panayi: Funded by Pfizer travel and
accommodation: International Continence Society Cairo 2008.
P. Tekkis: No disclosures.
R. Fernando: Travel and accommodation expenses: Pfizer.
V. Khullar: Paid consultant to: Astellas, Lilly, Allergan, Pfizer,
Gynecare, Cook and Bioxell.
References
1. DeLancey JO, Miller JM, Kearney R, Howard D, Reddy P, Umek
W, Guire KE, Margulies RU, Ashton-Miller JA (2007) Vaginal
birth and de novo stress incontinence: relative contributions of
urethral dysfunction and mobility. Obstet Gynecol 110:354–362
2. Dietz HP, Steensma AB (2006) The prevalence of major
abnormalities of the levator ani in urogynaecological patients.
BJOG 113:225–230
3. Bland JM, Altman DG (1986) Statistical methods for assessing
agreement between two methods of clinical measurement. Lancet
1:307–310
4. Delancey JO, Starr RA (1990) Histology of the connection
between the vagina and levator ani muscles. Implications for
urinary tract function. J Reprod Med 35:765–771
5. Chen B, Wen Y, Polan ML (2004) Elastolytic activity in women
with stress urinary incontinence and pelvic organ prolapse.
Neurourol Urodyn 23:119–126
6. Drewes PG, Yanagisawa H, Starcher B, Hornstra I, Csiszar K,
Marinis SI, Keller P, Word RA (2007) Pelvic organ prolapse in
fibulin-5 knockout mice: pregnancy-induced changes in elastic
fiber homeostasis in mouse vagina. Am J Pathol 170:578–589
7. Karam JA, Vazquez DV, Lin VK, Zimmern PE (2007) Elastin
expression and elastic fibre width in the anterior vaginal wall of
postmenopausal women with and without prolapse. BJU Int
100:346–350
8. Badiou W, Granier G, Bousquet PJ, Monrozies X, Mares P, de
Tayrac R (2008) Comparative histological analysis of anterior
vaginal wall in women with pelvic organ prolapse or control
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
subjects. A pilot study. Int Urogynecol J Pelvic Floor Dysfunct
19:723–729
Moalli PA, Shand SH, Zyczynski HM, Gordy SC, Meyn LA
(2005) Remodeling of vaginal connective tissue in patients with
prolapse. Obstet Gynecol 106:953–963
Chen BH, Wen Y, Li H, Polan ML (2002) Collagen metabolism
and turnover in women with stress urinary incontinence and pelvic
prolapse. Int Urogynecol J Pelvic Floor Dysfunct 13:80–87
Wong MY, Harmanli OH, Agar M, Dandolu V, Grody MH (2003)
Collagen content of nonsupport tissue in pelvic organ prolapse
and stress urinary incontinence. Am J Obstet Gynecol 189:1597–
1599
Jackson SR, Avery NC, Tarlton JF, Eckford SD, Abrams P, Bailey
AJ (1996) Changes in metabolism of collagen in genitourinary
prolapse. Lancet 15-6(347):1658–1661
Goh JT (2002) Biomechanical properties of prolapsed vaginal
tissue in pre- and postmenopausal women. Int Urogynecol J
Pelvic Floor Dysfunct 13:76–79
Dietz HP, Shek C, Clarke B (2005) Biometry of the pubovisceral
muscle and levator hiatus by three-dimensional pelvic floor
ultrasound. Ultrasound Obstet Gynecol 25:580–585
Dietz HP, Shek KL (2009) Levator defects can be detected by 2D
translabial ultrasound. Int Urogynecol J Pelvic Floor Dysfunct
20:807–811
Lanzarone V, Dietz HP (2007) Three-dimensional ultrasound
imaging of the levator hiatus in late pregnancy and associations
with delivery outcomes. Aust N Z J Obstet Gynaecol 47:176–180
Valsky DV, Yagel S (2007) Three-dimensional transperineal
ultrasonography of the pelvic floor: improving visualization for
new clinical applications and better functional assessment. J
Ultrasound Med 26:1373–1387
Khullar V, Cardozo L, Salvatore S, Hill S (1996) Ultrasound: a
noninvasive screening test for detrusor overactivity. Br J Obstet
Gynaecol 103:904–908, Ref Type: Generic
Khullar V, Salvatore S, Cardozo L, Bourne TH, Abbott D,
Kelleher C (1994) A novel technique for measuring bladder wall
thickness in women using transvaginal ultrasound. Ultra Obstet
Gynecol 4:220–223, 2007. Ref Type: Generic
Lekskulchai O, Dietz HP (2008) Detrusor wall thickness as a test
for detrusor overactivity in women. Ultrasound Obstet Gynecol
32:535–539
Oelke M, Hofner K, Wiese B, Grunewald V, Jonas U (2002)
Increase in detrusor wall thickness indicates bladder outlet
obstruction (BOO) in men. World J Urol 19:443–452
Robinson D, Anders K, Cardozo L, Bidmead J, Toozs-Hobson P,
Khullar V (2002) Can ultrasound replace ambulatory urodynamics
when investigating women with irritative urinary symptoms?
BJOG 109:145–148