Applicability of Perineal Sonography in

Anatomical Evaluation of Bladder Neck

in Women with and without Genuine
Stress Incontinence

Gin-Den Chen, MD,1 Tsaung-Hsien Su, MD,2 Long-Yau Lin, MD1

1
Department of Obstetrics & Gynecology, Chung Shan Medical & Dental College
Hospital, 23, section 1, Taichung-Kang Road, Taichung, Taiwan
2
Urodynamic Unit, Department of Obstetrics & Gynecology, Mackay Memorial Hospital, Taiwan

Received 1 August 1995; accepted 7 November 1996

Abstract: Thirty-seven patients with proven genuine stress incontinence (GSI) un-
derwent ultrasound study to evaluate the bladder neck position relative to the sym-
physis pubis. Sixty-five continent women including 40 parous and 25 nulliparous
women were recruited as a control group. Several parameters were found to be statis-
tically different between incontinent and control groups ( P , 0.01). If 28 degrees or
13 mm were used as the cut-off point for rotational angle (RA) and descent of bladder
neck (DBN), the sensitivity, specificity, and positive and negative predictive values
were 73.0%, 76.9%, 64.3%, and 83.3%, respectively. The specificity and positive pre-
dictive value for GSI increased to 83.1% and 67.6% if these two parameters were used
together. Perineal sonography is valuable in assessing anatomic change of the bladder
neck, but it is not a sensitive tool for predicting GSI.  1997 John Wiley & Sons, Inc.
J Clin Ultrasound 25:189–194, 1997
Keywords: perineal sonography; rotational angle; bladder neck descent; urinary
stress incontinence

Poor anatomic support of the bladder base and
urethrovesical junction, incompetence of the vesi-
cal neck, and urethral function were considered
important factors in uncomplicated genuine
stress urinary incontinence.1 The movement of
the bladder neck beyond the intra-abdominal
pressure transmission zone prevents pressure
transmission to the bladder neck and proximal
urethra to compensate for the pressure transmit-
ted to the bladder in case of a sudden increase of
abdominal pressure which might contribute to
the urine leakage.2
Transrectal,3 transvaginal,4–6 and perineal so-
Correspondence to: G.-D. Chen
 1997 John Wiley & Sons, Inc. CCC 0091-2751/97/040189-006
VOL. 25, NO. 4, MAY 1997
nography7–9 have been used to assess the dy-
namic changes of the urethrovesical junction and
proximal urethra for a decade as an alternative
to conventional radiological techniques. How-
ever, their shortcomings, such as impaired reso-
lution of image, distortion of the anatomical fea-
tures, requirement of urethral catheterization for
detecting the bladder neck and exact determina-
tion of the position of the bladder neck, are still
not clearly resolved.
The purposes of this study were to evaluate
objective parameters of the dynamic changes of
the bladder neck among patients with genuine
stress incontinence (GSI), to develop diagnostic
criteria for differentiating various groups of pa-
tients, and to assess the predictability of GSI us-
ing perineal sonography.
189
 CHEN ET AL

MATERIALS AND METHODS
Thirty-seven patients with proven GSI by urody-
namic study but without marked cystocele or ur-
ethrocele underwent ultrasound study to evalu-
ate the bladder neck relative to the symphysis
pubis. The ages of patients ranged from 24 to 46
(mean: 35.6) years and parity ranged from 1 to 4
(median: 2). Sixty-five continent women, includ-
ing 40 cases with vaginal delivery history and 25
nullipara, were recruited as a control group (con-
trol group 1, C1). The 40 cases with vaginal de-
livery were further defined as control group 2
(C2). The age of the participants with vaginal de-
livery history ranged from 23 to 49 (mean: 37.5)
years and parity ranged from 1 to 4 (median: 2).
The same procedures were performed on the con-
trol groups after they had received the regular
abdominal or transvaginal sonographic exami-
nation.
Perineal sonography was performed by using
a convex probe (3.5 MHz, Aloka 650 scanner).
The probe enclosed in a condom was positioned
just adjacent to the perineum and between the
labia majora at rest. The patient was placed in a
supine position and straightened her legs sponta-
neously. The bladder neck was identified without
urethral catheterization. The bladder capacity
was set at self-comfortably full sensation level,
about 150–250 mL, and the voiding volume was
checked after the procedure was completed.
A sagittal scan was performed to evaluate the
symphysis pubis (SP), the bladder, the vesico-
urethral junction (VUJ), the proximal urethra,
and other pelvic organs. If the bladder neck could
not be shown in the initial sectional plane, the
patient was asked to cough slightly to permit fur-
ther searching for the bladder neck. The patient
was then asked to perform a Valsalva maneuver
or cough, and the picture was frozen at the point
of maximal bladder neck rotational descent. Pic-
tures were taken at rest and during maximal
straining during this sequence of events. All of
the scanning procedures were recorded on vid-
eotape.
The rotational angle (RA) and descent of the
bladder neck (DBN) during maximal straining
were determined by using an arbitrary line
drawn at 120 degrees from midline of SP, drawn
from the inferior margin of SP (Figure 1). The
inferior margin of SP was used as the reference
point for RA. The distance between the inferior
margin of the SP and bladder neck was mea-
sured at rest (Dr) and during maximal straining
(Ds). The distance of the bladder neck from the
arbitrary line (DBN) was described at rest (r)
and during maximal straining (s). The DBN was

FIGURE 1. Diagram showing the relationships of symphysis pubis and the location of the bladder neck and
bladder at rest and during maximal straining. SP 5 symphysis pubis, BN 5 bladder neck, Dr 5 distance of
the bladder neck and the inferior margin of the symphysis pubis at rest, Ds 5 distance of the bladder neck
and the inferior margin of the symphysis pubis during maximal straining, r 5 distance of the bladder neck
and the arbitrary line at rest, s 5 distance of the bladder neck and the arbitrary line during maximal straining,
120 degrees 5 the angle between the middle line of the symphysis pubis and the arbitrary line.

190 JOURNAL OF CLINICAL ULTRASOUND

 BLADDER NECK IN WOMEN
defined as r minus s. This arbitrary line was
compared with the line drawn from the lower
margin of SP to the inferior margin of the 5th
sacral vertebra, which serves as a reference line
in the lateral chain-cystogram.10
The parameters Dr, Ds, r, s, RA, and DBN
were compared between the incontinent and con-
trol groups. The Student t test was used in the
statistical analysis to compare the parameters
between the incontinent and continent groups.
The significant level was set at P , 0.01. The
cut-off points of Dr, Ds, RA, and DBN were eval-
uated through the receiver operating characteris-
tic (ROC) curves to determine the highest sensi-
tivity and the lowest false predictive value. The
ability of perineal sonography to predict GSI was
expressed in sensitivity, specificity, and pre-
dictive values.
RESULTS
The dynamic changes of bladder neck and the in-
formation of the anatomic relationships of blad-
der neck, proximal urethra, and other pelvic
organs during straining could be clearly demon-
strated in sonographic scanning as shown in Fig-
ure 2.
FIGURE 2. Ultrasound images showing the relationships of the symphysis pubis and the location of the bladder at rest and during maximal
straining. SP 5 symphysis pubis, BL 5 bladder, BN 5 bladder neck, U 5 urethra, Cx 5 cervix of uterus, V 5 vagina, R 5 rectum. Oblique dotted
line 5 the middle line of the symphysis pubis, transverse dotted line 5 an arbitrary line drawn at 120 degrees from the inferior margin of the
symphysis pubis.
VOL. 25, NO. 4, MAY 1997
There were no significant differences in the
measured parameters between continent women
in the two selected control groups (C1 and C2).
The parameters of Dr, Ds, r, s, RA, and DBN
were significantly different between the GSI
group and each of the selected control groups
(Table 1).
Descent of bladder neck was statistically sig-
nificantly larger during the Valsalva maneuver
or coughing in the GSI group (17.6 6 7.3 vs
9.0 6 6.9; 17.6 6 7.3 vs 11.2 6 5.4 mm, all p ,
0.01). During maximal straining, the RA of blad-
der neck in the GSI group was statistically sig-
nificantly larger than that in the continent group
(42.9 6 19.7 vs 17.7 6 15.0; 42.9 6 19.7 vs
22.7 6 13.4 degrees, all p , 0.01).
The cut-off points for RA and DBN of bladder
neck were found to be 28 degrees and 13 mm
from their receiver operating characteristic
(ROC) curves, respectively (Figure 3). The cut-off
points for evaluating Dr and Ds were found to be
meaningless through ROC analysis. If either 28
degrees or 13 mm was used as a cut-off point for
RA and DBN, the sensitivity, specificity, and pos-
itive and negative predictive values were 73.0%,
76.9%, 64.3%, and 83.3%, respectively. Combin-
ing these two parameters, the specificity and pos-
191
 CHEN ET AL

TABLE 1 the hazard of radiation exposure. Also, it is more

Comparison of Parameters Between GSI and
reproducible than radiological measurements.
Control Groups
The vaginal probe was proven to reduce the
Control likelihood of detecting incontinence in women
GSI (C1 5 65) with GSI or reduce the severity of GSI diag-
TCH (N 5 37) (C2 5 40) P Value
nosed. It might be inappropriate in assessing the
Dr (mm) 26.8 6 4.0 29.3 6 3.3 P , 0.01 descent and mobility of the bladder neck and
29.5 6 3.6 P , 0.01
Ds (mm) 19.4 6 3.5 23.0 6 3.8 P , 0.01
urogenital prolapse, because the vaginal probe
22.4 6 3.9 P , 0.01 may directly contact, and thus compress, the va-
RA (degrees) 42.9 6 19.7 17.6 6 14.9 P , 0.01 ginal wall.11 On the contrary, by using the convex
22.7 6 13.6 P , 0.01 probe adjacent to the perineum and between the
r (mm) 18.2 6 5.6 21.3 6 5.4 P , 0.01
22.3 6 4.6 P , 0.01 labia majora, normal pelvic floor anatomy is not
S (mm) 0.6 6 6.8 12.3 6 8.5 P , 0.01 distorted nor the bladder neck mobility impeded.
11.4 6 6.0 P , 0.01 Therefore, the dynamic changes of the bladder
DBN (mm) 17.6 6 7.3 9.0 6 6.9 P , 0.01
11.2 6 5.4 P , 0.01
neck and the relative changes of other pelvic or-
gans can be clearly shown in the sonographic ex-
Dr: distance between bladder neck and lower margin of symphysis
pubis at rest.
amination.
Ds: distance between bladder neck and lower margin of symphysis Schaer et al8 found that some landmarks re-
pubis during straining. lating to the bladder neck could not be demon-
RA: rotational angle of bladder neck.
r: distance between bladder neck and arbitrary line at rest.
strated satisfactorily at rest and during the Val-
s: distance between bladder neck and arbitrary line during salva maneuver, especially in measuring the
straining. funneling of the bladder neck and severe bladder
DBN: descent of bladder neck. base descent. They considered these results to be
P value , 0.01 was statistically significantly different by Student
t test. The parameters of C1 and C2 group were not statistically signifi- a typical phenomenon illustrating the difference
cantly different by Student t test. between ultrasound and x-ray techniques. Ultra-
sound portrays a thin cut through the female pel-
vis, whereas x-ray assessment scans all layers of
itive predictive value for GSI could be increased the pelvic region with the potential risk of a mis-
to 83.1% and 67.6% (Table 2). leading assessment of the anatomy when not per-
formed correctly.9
Because our cases had proven GSI by urody-
DISCUSSION namic study without marked pelvic relaxation,
the above phenomenon could be ruled out in this
Ultrasound has been advocated as an alterna- study. When the position of the bladder neck
tive to videocystourethrography in several could not be clearly shown in the image during
studies.2–6,12,13 There were some benefits, such as straining, the patient was asked to cough and
no x-ray exposure and less expense. Perineal so- perform the Valsalva menuever repeatedly. Thus
nography could be performed repeatedly without the point of maximal bladder neck rotational de-

FIGURE 3. Diagram of cut-off point for rotational angle (left) and descent of the bladder neck (right) as shown in receiver operating characteristics
curve analysis.

192 JOURNAL OF CLINICAL ULTRASOUND

 BLADDER NECK IN WOMEN
Sensitivity, Specificity, and Positive and Negative Predictive Values for
DBN . 13 mm
(%)
Sensitivity
Specificity
Positive
predictive value
Negative
predictive value
DBN: descent of bladder neck.
RA: rotational angle of bladder neck.
scent could be reached. Conclusively, perineal so-
nography can provide a reproducible result for
determining the dynamic change of the bladder
neck area.
Gordon et al9 have found similar results be-
tween perineal ultrasound scanning and x-ray
chain urethrocystography in evaluating the blad-
der neck position in GSI patients. Furthermore,
the experience of Mackay Memorial Hospital in-
dicated that an arbitrary line drawn at 120 de-
grees from the lower margin of the symphysis
pubis axis in introital ultrasound could replace
the pubo-sacral line in chain cystourethrography.
A close correlation was found between this arbi-
trary line and the pubo-sacral line in chain cys-
tourethrography in a previous report.10 By using
this arbitrary line, the mobility of bladder neck
and proximal urethra could be shown in perineal
sonography just as in chain cystourethrography.
This procedure has been suggested as a suitable
alternative to conventional radiological tech-
niques in the evaluation of mobility of the blad-
der neck. Our method could easily and directly
provide information on the dynamic changes of
the bladder neck and the anatomic relationships
of the bladder neck in the absence of urethral
catheterization, which might introduce an iatro-
genic infection and distort the urethral profile.
In this study, the movement of the bladder
neck was shown as a semicircular movement
with the tip of the symphysis pubis as the center
and a line from the tip to the bladder neck as
the radius (BS). This finding was similar to the
description of Mouritsen et al.12
These parameters in the two selected control
groups are not significantly different. In conti-
nent women, parity seems to make little contri-
bution to the mobility of the bladder neck (Table
1).
The Dr and Ds were significantly different be-
tween the GSI and continent groups (Dr: 26.8 6
VOL. 25, NO. 4, MAY 1997
TABLE 2
Diagnosis of GSI
DBN . 13 mm and
RA . 28 Degree RA . 28 Degree
(%) (%)
73.0 73.0 62.2
76.9 76.9 83.1
64.3 64.3 67.6
83.3 83.3 79.4
4.0 vs 29.3 6 3.3 or 29.5 6 3.6. Ds: 19.4 6 3.6 vs
23.0 6 3.8 or 22.4 6 3.9), but there was consider-
able overlap. Nevertheless, Dr was significantly
longer than Ds in the GSI compared with the se-
lected control groups.
If two of three criteria (angle at rest .95 de-
grees, BS ,2.3 cm, and mobility .20 degrees)
are regarded as the diagnostic criteria of inconti-
nence, the sensitivity was 84% and the specificity
82% using vaginal ultrasonography in the study
of Mouritsen et al.12 The sensitivity of transrectal
ultrasonic evaluation of women with stress uri-
nary incontinence (when 1 cm drop of the
urethrovesical junction is considered as the up-
per normal limit) was 86% and the specificity
was 91% in the report of Bergman et al.13
In our study, the diagnostic cut-off points for
RA and DBN were set as 28 degrees and 13 mm
in diagnosing the GSI. The sensitivity, specific-
ity, and positive and negative predictive values
were 73.0%, 76.9%, 64.3%, and 83.3%, respec-
tively. The specificity and positive predictive
value for GSI could be increased to 83.1% and
67.6% only if these two criteria were considered
together.
In conclusion, perineal sonography can detect
the realistic movement of the bladder neck with-
out restricting it by the presence of the probe.
This offers real-time dynamic scanning of the re-
lationship between pelvic anatomic changes at
rest and during straining. It is valuable in as-
sessing anatomic change of the bladder neck, in
GSI patients. However, there is not sufficient ev-
idence to use perineal sonography in predicting
GSI.
REFERENCE
1. Summitt RL Jr, Bent AE, Ostergard DR: The
pathophysiology of genuine stress incontinence.
Int Urogynec J 1:12–17, 1990.
2. Johnson JD, Lamensdorf H, Hollander IN, et al:
193
 CHEN ET AL
Use of transvaginal endosonography in the evalu-
ation of women with stress urinary incontinence.
J Urol 147:421–425, 1992.
3. Bergman A, Mckenzie CJ, Richmond J, et al:
Transrectal ultrasound versus cystography in the
evaluation of anatomical stress urinary inconti-
nence. Br J Urol 62:228–234, 1988.
4. Richmond DH, Suther J: Transrectal ultrasound
scanning in urinary incontinence: the effect of the
probe on urodynamic parameters. Br J Urol
64:582–585, 1989.
5. Quinn MJ, Beynon J, Mortenson NJMcC, et al:
Transvaginal endosonography: a new method to
study the anatomy of the lower urinary tract in
urinary stress incontinence. Br J Urol 62:414–
418, 1988.
6. Koebl H, Bernaschek G: A new method for sono-
graphic urethrocystography and simultaneous
pressure-flow measurements. Obstet Gynecol
74:417–422, 1989.
7. Creighton SM, Pearce JM, Stanton SL: Perineal
video-ultrasonography in the assessment of vagi-
nal prolapse: early observations. Br J Obstet Gyne-
col 99:310–313, 1992.
194
8. Schaer GN, Koechli OR, Schuessler B, et al. Peri-
neal ultrasound for evaluating the bladder neck in
urinary stress incontinence. Obstet Gynecol
85:220–224, 1995.
9. Gordon D, Pearce M, Norton P, et al: Comparison
of ultrasound and lateral chain urethrocystogra-
phy in the determination of bladder neck descent.
Am J Obstet Gynecol 160:182–185, 1989.
10. Hwu YM, Su TS, Yang JM, et al: In comparison of
vaginal introital sonography and chain cys-
tourethrography in patients with genuine stress
incontinence. 1992 Annual Convention of the Soci-
ety of Ultrasound in Medicine of the Republic of
China. JUMROC 9(3):190, 1992.
11. Wise BG, Burton G, Cutner A, et al: Effect of vagi-
nal ultrasound probe on lower urinary tract func-
tion. Br J Urol 70:12–16, 1992.
12. Mouritsen L, Rasmussen A: Bladder neck mobility
evaluated by vaginal ultrasonography. Br J Urol
71:166–171, 1993.
13. Bergman A, Ballard CA, Plat LD: Ultrasonic eval-
uation of urethrovesical junction in women with
stress urinary incontinence. J Clin Ultrasound
16:295–300, 1989.
JOURNAL OF CLINICAL ULTRASOUND