Anda di halaman 1dari 7

BJOG: an International Journal of Obstetrics and Gynaecology

July 2002, Vol. 109, pp. 821 –827

Perineal ultrasound evaluation of urethral angle and bladder

neck mobility in women with stress urinary incontinence
Roberto Pregazzia,*, Andrea Sartorea, Paolo Bortolib, Eva Grimaldia,
Luigi Troianoa, Secondo Guaschinoa
Objectives 1. To assess the reproducibility of an electronic ultrasonographic technique for the measurement of
urethral angulation; 2. to test the ability of measurement of the urethral angle and bladder neck mobility to
predict genuine stress incontinence; 3. to compare ultrasound variables in stress incontinent women and in
Design Case –control study.
Population Twenty-three incontinent women and 50 controls.
Methods Electronic measurements of the distance between the bladder neck and the symphysis pubis, the
bladder neck and the symphysis pubis line and the midline of the symphysis (alpha angle) and the angle
between the proximal and distal urethra (beta angle) by means of perineal ultrasonography with a
comfortably full bladder at rest, during the Valsalva manoeuvre and during maximal pelvic floor
contraction. The same procedure was performed by a second investigator. Repeatability was evaluated by
the technique described by Bland and Altman. Statistical analysis was performed using Student’s t test and
the two-tailed paired t test.
Main outcome measures To test the possible role of the urethral angle in maintaining female continence.
Results Ultrasound analysis showed good repeatability between the two observers and is not influenced by
vesical volume. Beta angle and urethrovesical mobility are inversely proportional, both in continent and in
incontinent women. Urethral angle identifies genuine stress incontinence better than urethrovesical
mobility (sensitivity 96% vs 87%; specificity 92% vs 68%; positive predictive value 85% vs 55%). There
are significant differences in all ultrasound variables between incontinent women and continent controls.
Conclusions This study suggests a significant role of the urethral angle in maintaining female continence
(in incontinent women it is lower at rest and lowers with straining). Measurement of the urethral angle
can provide useful additional information to that provided by ultrasound evaluation of bladder neck

INTRODUCTION Although the demonstration of mobility of the proximal

urethra is one of the most relevant observations in women
It is widely accepted that pre-operative evaluation of with stress urinary incontinence, the study of the anato-
women with stress urinary incontinence should include an mical and functional relationship of all the different parts of
assessment of urethrovesical mobility. In the last few the urethra (proximal, middle and distal) may clarify
decades, a variety of methods have been used for this important aspects of the mechanism of continence. Perineal
purpose. These include the Q-tip test1, radiographic tech- ultrasonography allows us to visualise and to measure the
niques2 and ultrasonographic studies3,4. angle between the proximal mobile part and the distal fixed
Transvaginal and perineal ultrasonography allow the part of urethra7. In this study, Vierhout and Hol postulated
assessment of bladder neck and urethral mobility at rest that urethral angulation is a normal mechanism of conti-
and during coughing, the Valsalva manoeuvre and pelvic nence that is absent in stress incontinence. The relationship
floor contraction. The technique is simple, not invasive and between the urethral angle and bladder neck mobility has
without discomfort to the women5,6. not been determined. Bladder neck mobility has been
related to the functional integrity of the structures surround-
ing the proximal urethra8, whereas urethral angle has been
related to the functional integrity of both the proximal and
Department of Obstetrics and Gynaecology, University of the distal urethral supports9,10. It might be expected that
Trieste, IRCCS ‘Burlo Garofolo’, Trieste, Italy urethral angle values are not entirely dependent upon
Epidemiology Unit, IRCCS ‘Burlo Garofolo’, Trieste, Italy bladder neck mobility. Therefore, it seems relevant to
* Correspondence: Dr R. Pregazzi, Department of Obstetrics and establish the role of urethral angle measurements as an
Gynaecology, IRCCS ‘Burlo Garofolo’, University of Trieste, Via independent variable in women with and without stress
dell’Istria 65/1, 34137 Trieste, Italy. urinary incontinence.
D RCOG 2002 BJOG: an International Journal of Obstetrics and Gynaecology
PII: S 1 4 7 0 - 0 3 2 8 ( 0 2 ) 0 1 1 6 3 - 1

Diagnostic urodynamic evaluation consisted of uroflow-

metry, resting and stress urethral pressure profilometry and
subtracted water cystometry (UroDesk 300, SI.EM Sistemi
Elettromedicali, Milano, Italy). The diagnosis of genuine
stress urinary incontinence was made by a positive stress
test during coughing in the absence of simultaneous detru-
sor contraction or pressure equalisation on the stress
urethral pressure profile.
Perineal ultrasound was performed using a digital
computer machine (Kranzbuhler mod. Sonofritz) with a
6.5 mHz electronic microconvex array (ray 10 mm, sector
angle 180j) placed on the interlabial region of the vulva
in a sagittal orientation in order to obtain views of the
symphysis pubis, bladder and urethra. The test was
performed with the woman in the lithotomy position,
before and after micturition. Ultrasound measurements
included the distance BN – S from the bladder neck
(BN) to the lowest point of the symphysis pubis (S),
the angle between the BN – S line and the midline of the
Fig. 1. The alpha angle between the bladder neck – symphysis line and the
midline of the symphysis. Legend: BN ¼ bladder neck; SP ¼ symphysis symphysis (alpha angle) and the urethral knee (beta
pubis; U ¼ urethra; B ¼ bladder; a ¼ alpha angle. angle). Alpha angle measurements were performed
according to the technique described by Mouritsen and
The aims of this study were: (1) to assess the repro- Rasmussen11. Beta angle was obtained by visualisation of
ducibility of an ultrasonic technique that allows the the urethral knee at the point where the urethra enters the
measurement of the angle between the mobile proximal urogenital diaphragm: this angle is between the proximal
part and the fixed distal part of the urethra; (2) to examine and the distal parts of the urethra. The angles and the
the effect of bladder fullness on these ultrasound measure- BN – S distance were measured electronically by the
ments; (3) to analyse the association between urethral ultrasound software equipment (Figs. 1 and 2). All scans
angulation and bladder neck mobility; (4) to test if assess- were performed by the same examiner at rest, on maximal
ment of the urethral angle and bladder neck mobility is able straining (the Valsalva manoeuvre) and during maximal
to predict genuine stress incontinence compared with pelvic floor contraction (the withholding manoeuvre). All
other urological disorders; (5) to compare the ultrasound measurements were easily performed with the exception
measurements obtained in genuine stress incontinence with of the beta angle during straining and the withholding
other urological disorders.


Seventy-three consecutive women, 40 – 65 years old

(mean age 49 years), referred to the gynaecological urody-
namic clinic with complaints of lower urinary tract symp-
toms, underwent full urodynamic evaluation. All the
women gave informed consent to this study, which was
previously approved by the local ethics committee. All had
a negative culture of urine before evaluation. Each had a
standardised history taken, followed by a general gynae-
cological and neurological examination. The exclusion
criteria were: age greater than 70 years, previous surgery
for prolapse or urinary incontinence, and significant
anterior vaginal wall relaxation, which was defined as
descent to at least the lower one-third of the vagina with
maximal straining in the supine position. For the purposes
of the analysis, all women with urodynamically proven
genuine stress incontinence (n ¼ 23) were compared with Fig. 2. The beta angle between the proximal and the distal parts of the
women who had other urological disorders but proven urethra. Legend: BN ¼ bladder neck; SP ¼ symphysis pubis; U ¼ urethra;
urethral sphincter competence (n ¼ 50). B ¼ bladder; h ¼ beta angle.

D RCOG 2002 Br J Obstet Gynaecol 109, pp. 821 – 827


Table 1. Characteristics of the women. Table 2. Reliability of measurement of the urethral angle.

Genuine stress Other urological Centile Angle (j)

incontinence disorders
(n ¼ 23) (n ¼ 50) 97.7 28
97 28
Age (years) [SD] 49.1 [6.7] 49.7 [6.5] 95 26
Nulliparae (number) 5 6 90 24
Multiparae (number) 18 44 75 21
Weight (kg) [SD] 64.3 [3.5] 63.3 [4.1] 50 18 3j to þ3j ¼ 15 – 21j
Body mass index [SD] 24.8 [1.8] 24.6 [1.4] 25 14
10 11
5 10
3 8
manoeuvres. The beta angle was visualised and measured 2.3 7 3j to þ3j ¼ 4 – 10j
in all women at rest, in 50 of the 73 women (20
Mean [SD] ¼ 17.65 [5.31].
incontinent, 30 continent) during straining and only in
a few women during the withholding manoeuvre. Diffi-
culties in obtaining the beta angle were caused by the
considerable urethral distortion due to maximal straining All terminology conforms to that proposed by the
and the withholding manoeuvres. Values of the beta angle International Continence Society unless otherwise stated14.
obtained during the withholding manoeuvre were not
considered in this study.
The same procedure was separately performed by a RESULTS
second investigator in order to investigate the repeatability
of the method. The two investigators were blinded to the The 73 women in this study were divided into two
continence status of the women and to each other’s categories: (1) 23 women with a urodynamic diagnosis of
results. Repeatability was evaluated following the tech- genuine stress incontinence; (2) 50 control women with
nique described by Bland and Altman12. The definition of other urologic symptoms but proven urethral sphincter
repeatability was obtained according to the coefficient competence (6 women with detrusor instability, 9 with
adopted by the British Standards Institution13, where it sensory urgency, 35 with voiding disorders not associated
is expected that at least 95% of the differences between with a specific urodynamic diagnosis: 16 affected by
two different observers are less than two standard devia- strangury and 19 by dysuria).
tions. To choose the cutoff limits of urethral angle and The baseline characteristics of the women are shown in
bladder neck mobility discriminating between continent Table 1. The controls were similar to the women with stress
and incontinent women, we constructed receiver – operator incontinence with respect to age, parity, body weight and
characteristic curves (ROC curves). Student’s t test and body mass index. Six women with genuine stress inconti-
the two-tailed paired t test were also employed in the statis- nence were premenopausal (26%) compared with 18 of the
tical analysis. controls (36%).


Mean + 2SD

Differences 0
-5 Mean - 2SD

5 10 15 20 25 30

Mean of resting beta angle measurements

Fig. 3. Plot of differences of rest beta angle measurements between the two observers in the 73 women. Only 43 points are represented because 30 are

D RCOG 2002 Br J Obstet Gynaecol 109, pp. 821 – 827


Mean + 2SD
2 Mean
Difference 0
-4 Mean - 2SD
5 10 15 20 25 30
Mean of resting beta angle measurements
Fig. 4. Plot of differences of rest beta angle measurements before and after micturition in the 73 women. Only 41 points are represented because 32 are

The results of the ultrasound analyses by two indepen- differences was 2.4j to þ2.4j. Bladder fullness does not
dent investigators are reported in Fig. 3. This figure shows affect the reliability of measurements of the beta angle.
the plot of the differences between the two observers of Figure 5 shows the association between the mobility of
measurement of the beta angle at rest. The mean difference the bladder neck and urethral angle in the whole population
was 0j and the 95% data interval of the differences was by means of a regression line. We regressed bladder neck
3j to þ3j. This shows good agreement and shows that mobility on urethral angle; that is, the response variable is
measurement of the beta angle is a reliable clinical test. In urethral angle and the explanatory variable is bladder neck
order to demonstrate the reliability of the technique, we mobility. The regression coefficient is 0.375 and its 95%
placed these limits of agreement against a centile chart of confidence interval (CI) ranges from 0.486 to 0.264.
beta angle (Table 2). Table 3 shows the results of the ultrasound measurements
The effect of bladder fullness upon measurement of the in the incontinent women and in the controls. There is a
beta angle is reported in Fig. 4. The mean difference significant difference in all ultrasound variables between
between the beta angle when the bladder was empty and stress incontinent women and controls. The stress inconti-
when it was full was 0.01j and the 95% data interval of the nent women demonstrated significantly lower values of the

Urethral angle

0 10 20 30 40 50
Bladder neck mobility
Fig. 5. Relationship between the mobility of bladder neck and urethral angle. Only 63 points are represented because 10 are superimposed.

D RCOG 2002 Br J Obstet Gynaecol 109, pp. 821 – 827


Table 3. Ultrasonographic parameters in women with stress incontinence and in controls.

Stress incontinence (n ¼ 23) Controls (n ¼ 50) P

Alpha angle (°)

Rest 100 [5] 92 [6] <0.001
Valsalva manoeuvre 120 [8] 100 [8] <0.001
Pelvic floor contraction 90 [9] 80 [8] <0.001

Beta angle (°)

Rest 12 [2] 20 [4] <0.001
Valsalva manoeuvre 5 [2] 21 [5] <0.001

BN – S distance (mm)*
Rest 21 [2] 25 [3] <0.001
Valsalva manoeuvre 18 [2] 23 [3] <0.001
Pelvic floor contraction 17 [2] 24 [4] <0.001

Maximal mobility** 30 [6] 20 [8] <0.001

Data are presented as mean [SD].

* Distance between bladder neck and symphysis pubis.
** Alpha angle during straining  alpha angle during withholding.

beta angle. Beta angle lowers with straining in incontinent predictive value (85% vs 55%) and greater negative pre-
women, while it increases with straining in the controls. dictive value (98% vs 92%); the likelihood ratio of a
In Figs. 6 and 7, we constructed the ROC curves, positive test was 12 vs 2.72 and the likelihood ratio of a
respectively, for urethral angle and bladder neck mobility negative test was 0.04 vs 0.19.
to choose the cutoff values discriminating between con-
tinent and incontinent women. These are a plot of the true
positive rate against the false positive rate for the different DISCUSSION
possible cutpoints of a diagnostic test. In the first case, we
obtained an ROC area of 0.93 (95% CI 0.87– 0.99) and we Modern ultrasound examination allows adequate evalu-
chose 14j as the cutoff point: the value of the area shows ation of the dynamics of the bladder neck and the proximal
excellent accuracy of the test. In the second case, we urethra in women with stress urinary incontinence.
obtained an ROC area of 0.85 (95% CI 0.76– 0.94) and Ultrasonography can be used as an alternative to conven-
we chose 26 mm as the cutoff point: the value of the area tional radiological techniques for pre- and post-operative
shows good accuracy of the test. evaluation of women with incontinence15. Bladder neck
Table 4 shows the efficiency of measurement of the mobility can be demonstrated by perineal or vaginal
urethral angle compared with bladder neck mobility in ultrasound and measured using the symphysis pubis as
identification of genuine stress incontinence. The measure- the immobile reference point11. Moreover, perineal ultra-
ment of urethral angle shows greater sensitivity (96% vs sonography allows us to visualise and to measure the angle
87%), greater specificity (92% vs 68%), greater positive between the proximal mobile part and the distal fixed part




0 0.2 0.4 0.6 0.8 1

Fig. 6. ROC curve for urethral angle.

D RCOG 2002 Br J Obstet Gynaecol 109, pp. 821 – 827



True positive rate


0 0.2 0.4 0.6 0.8 1
False positive rate (1-specificity)
Fig. 7. ROC curve for bladder neck mobility.

of the urethra. In our study, urethral angulation was angle and urethrovesical mobility are not entirely inter-
measured electronically by the ultrasound software. The dependent. The two variables seem to represent different
technique gives reliable results with different investigators mechanisms maintaining continence. The static measure-
and is not influenced by bladder fullness. On the other ment of urethral angle, when compared with measurement
hand, the angle can be easily visualised and exactly of urethrovesical mobility, shows a greater ability to
measured only at rest: Valsalva manoeuvre and pelvic distinguish genuine stress incontinence from other urolo-
floor contraction cause urethral distortion, which makes gical disorders.
the measurements difficult. Our data suggest that urethral angulation plays a signifi-
Is it possible to correlate the static observations of the cant role in female continence. With genuine stress incon-
urethra with the dynamic function of urethrovesical support tinence, the urethral angle is lower at rest and lowers with
structures? The proximal urethra is mobile, with a posterior straining. We also suggest that measurement of the urethral
movement of the vesical neck seen at the onset of mictu- angle can provide useful additional information to that
rition and an elevation noted when the woman is instructed found by ultrasound evaluation of bladder neck mobility.
to contract her levator ani muscles16,17. The distal urethra is
surrounded by two arches of striated muscle (compressor
urethrae and the urethrovaginal sphincter) in the region References
of the urogenital diaphragm. Activity of these muscles
causes the increase in urethral pressure that precedes and 1. Karram MM, Bhatia M. The Q-tip test: standardization of the
exceeds the increase in abdominal pressure during a technique and its interpretation in women with urinary incontinence.
Obstet Gynecol 1988;71(6 Pt 1):807 – 811.
2. Mc Guire EJ, Lytton B, Pepe V, Kohorn EI. Stress urinary inconti-
It seems obvious that with the loss of urethrovesical nence. Obstet Gynecol 1976;47(3):255 – 264.
support the urethral angle becomes less evident; therefore, 3. Koelbl H, Strassegger H, Riss PA, Gruber H. Morphologic and
urethrovesical mobility and urethral angle are inversely functional aspects of pelvic floor muscles in patients with pelvic
proportional. Nevertheless, bladder neck mobility seems relaxation and genuine stress incontinence. Obstet Gynecol 1989;
74(5):789 – 795.
dependent upon anatomical and functional integrity of the
4. Benson JT, Sumners JE, Pittman JS. Definition of normal female
supports of the proximal urethra (endopelvic fascia and its pelvic floor anatomy using ultrasonographic techniques. J Clin Ultra-
lateral attachments to the arcus tendineus fasciae pelvis and sound 1991;19(5):275 – 282.
pelvic diaphragm), whereas urethral angulation seems 5. Johnson JD, Lamensdorf H, Hollander IN, Thurman AE. Use of
dependent upon anatomic and functional status of both transvaginal endosonography in the evaluation of women with stress
urinary incontinence. J Urol 1992;147(2):421 – 425.
proximal and distal supports. We suggest that urethral
6. Kohorn EI, Scioscia AL, Jeanty P, Hobbins JC. Ultrasound cystour-
ethrography by perineal scanning for the assessment of female stress
urinary incontinence. Obstet Gynecol 1986;68(2):269 – 272.
Table 4. Efficiency of measurement of the urethral angle compared with 7. Vierhout ME, Hol M. The knee of the urethra: its possible role in the
bladder neck mobility in identification of genuine stress incontinence. female continence mechanism. Urogyn Int J 1994;8(2):81 – 84.
8. DeLancey JOL. Structural support of the urethra as it relates to stress
Urethral angle Bladder neck mobility
urinary incontinence: the hammock hypothesis. Am J Obstet Gynecol
14 26
1994;170(6):1713 – 1720 [discussion 1720 – 3].
Incontinent women (n ¼ 23) 22 20 9. Petros P, Ulmesten U. An integral theory and its method for the diag-
Controls (n ¼ 50) 4 16 nosis and management of female urinary incontinence. Scand J Urol
Nephrol Suppl 1993;153:1 – 93.

D RCOG 2002 Br J Obstet Gynaecol 109, pp. 821 – 827


10. Westby M, Asmussen M, Ulmsten U. Location of maximum intra- 14. Bates P, Bradley WE, Glen E, et al. The standardization of terminol-
urethral pressure related to urogenital diaphragm in the female subject ogy of lower urinary tract function. J Urol 1979;121(5):551 – 554.
as studied by simultaneous urethrocystometry and voiding urethrocys- 15. Schaer GN, Koechli OR, Schuessler B, Haller U. Perineal ultrasound
tography. Am J Obstet Gynecol 1982;144(4):408 – 412. for evaluating the bladder neck in urinary stress incontinence. Obstet
11. Mouritsen L, Rasmussen A. Bladder neck mobility evaluated by Gynecol 1995;85(2):220 – 224.
vaginal ultrasonography. Br J Urol 1993;71(2):166 – 171. 16. Jeffcoat TNA, Roberts H. Observations on stress incontinence of
12. Bland JM, Altman DG. Statistical methods for assessing agreement urine. Am J Obstet Gynecol 1952;64:721 – 738.
between two methods of clinical measurement. Lancet 1986:301 – 310 17. Muellner SR. Physiology of micturition. J Urol 1951;65:805 – 810.
(February 8). 18. DeLancey JOL. Structural aspects of the extrinsic continence mech-
13. British Standards Institution. Precision of Test Methods I: Guide for anism. Obstet Gynecol 1988;72(3 Pt 1):296 – 301.
the Determination and Reproducibility for a Standard Test Method
(BS 5497, Part 1). London: BSI, 1979. Accepted 9 April 2002

D RCOG 2002 Br J Obstet Gynaecol 109, pp. 821 – 827