The Effect of Posterior Wall Support Defects on
Urodynamic Indices in Stress Urinary
Incontinence
DEBORAH L. MYERS, MD, CHRISTINE A. LASALA, MD, JOSEPH W. HOGAN, ScD, AND
PETER L. ROSENBLATT, MD
Objective: To determine if posterior vaginal wall defects
affect urodynamic indices and mask stress urinary inconti-
nence.
Methods: Ninety women with grade 0, 1, 2, or 3 posterior
wall defects were evaluated prospectively by complete uro-
dynamics to assess their urinary complaints. None had
severe anterior or apical support defects. Urethral pressure
profilometry and cough stress test were performed with the
posterior wall in the unretracted position and then with the
posterior wall retracted using a split speculum. Analysis of
covariance was used to compare adjusted mean differences
in maximum urethral closure pressure, functional urethral
length, and units of leakage volume during the cough stress
test in the unretracted and retracted positions among the
posterior wall grade groups.
Results: In women with grade 3 posterior wall defects,
there were significant changes from the unretracted to the
retracted position in maximum urethral closure pressure of
27.0 cm H2O, (99% confidence interval [CI] 212.4, 21.6),
functional urethral length of 20.3 cm (99% CI 20.5, 20.1),
and leak volume units of 10.7 (99% CI 0.4, 1.0) during the
cough stress test. There were four women with grade 3
posterior wall defects who demonstrated potential stress
incontinence when their posterior wall was retracted.
Conclusion: A grade 3 posterior wall defect may artificially
raise maximum urethral closure pressure, increase func-
tional urethral length, and mask urinary stress incontinence
during a cough stress test. Women with grade 3 posterior
wall defects should be tested with the posterior wall re-
tracted during urodynamic evaluation. (Obstet Gynecol
1998;91:710 – 4. © 1998 by The American College of Obste-
tricians and Gynecologists.)
Genuine stress incontinence is a prevalent condition
affecting millions of women throughout the United
From the Department of Obstetrics and Gynecology, Women and
Infants’ Hospital of Rhode Island, Providence, Rhode Island; and the
Center for Statistical Sciences, Brown University School of Medicine,
Providence, Rhode Island.
710 0029-7844/98/$19.00
PII S0029-7844(98)00049-0
States.1,2 Women with severe cystoceles or severe uter-
ine-vaginal vault prolapse may be continent paradoxi-
cally as a result of “kinking” of the urethra. Stress
urinary incontinence may occur when the prolapse is
reduced as a result of a straightening of the urethra.
Thus, it is recommended currently that women with
severe anterior vaginal wall or uterine-vaginal vault
prolapse undergo urodynamic testing with the aid of a
pessary, manual reduction or other devices to unmask
potential stress urinary incontinence.3– 8
The purpose of this study was to determine if the
presence of a posterior vaginal wall support defect in
the absence of severe anterior wall and uterine-vaginal
vault prolapse affects urodynamic indices and also
masks stress urinary incontinence. We theorized that a
severe posterior wall support defect may raise artifi-
cially maximum urethral closure pressure, increase
functional urethral length, and mask urinary leakage
during a cough stress test.
Materials and Methods
Between December 1994 and March 1997, a prospective
observational study of women with urinary complaints
was performed in our tertiary referral urodynamic
laboratory. Each was evaluated by history, voiding
diary, focused neurologic and urogynecologic examina-
tion, cotton swab test, urine culture and sensitivity,
cystoscopy, and complex urodynamics. The grades of
all pelvic support defects were assessed on pelvic
examination using Baden and Walker definitions.9 Only
women with grade 0, 1, or 2 anterior wall defects and
grade 0 or 1 apical defects were included in the study.
Women without posterior wall defects (control group)
and women with all grades of posterior wall defects
were included. All women, whether premenopausal or
postmenopausal, with or without hormone replacement
Obstetrics & Gynecology

Figure 1. Posterior wall in unretracted position.
therapy were eligible. One hundred and one women
entered the study. The complex urodynamic assessment
was performed using a Dantec UD 5500 (Dantec Inc.,
Copenhagen, Denmark), which included uroflowmetry
with determination of postvoid residual urine volume,
multichannel urethrocystometry, urethral pressure pro-
filometry, and cough stress test. All testing was done in
the sitting erect position in a Century birthing chair
(Century Manufacturing Co., Tampa, FL). A Millar 8F
dual microtip transducer (Millar Instruments, Houston,
TX) was used for the multichannel urethrocystometro-
gram and urethral pressure profilometry. The urethro-
cystometrogram was done using sterile water at 22C at
a filling rate of 50 mL/minute. Maneuvers, such as
water stimulation and cough, were used in an effort to
provoke detrusor instability. Urethral pressure pro-
filometry was performed with the dual microtip sensor
positioned at 9 o’clock using a mechanical puller at a
rate of 1 mm/second. Both static and dynamic (cough-
ing at maximum effort) profiles were performed at
cystometric capacity (with a minimum bladder volume
of 200 mL). Maximum urethral closure pressure and
functional urethral length were assessed. Lastly, a
cough stress test was performed at cystometric capacity
with all catheters removed; it was considered positive if
urine loss per urethra was synchronous with the cough.
Leakage volume during the cough stress test was mea-
sured in a plastic beaker. Leakage volume was catego-
VOL. 91, NO. 5, PART 1, MAY 1998
rized as follows: 0 for no leak, 11 for up to 5 mL, 21 for
6 –15 mL, and 31 for greater than 16 mL.
All women had static and dynamic urethral pressure
profilometry and cough stress test done with the pos-
terior vaginal wall in the unretracted and retracted
position, using the bottom blade of a split Pederson
speculum, regardless of the grade of posterior wall
defect (Figures 1 and 2). Diagnoses of urinary inconti-
nence (ie, genuine stress incontinence, unstable detru-
sor or detrusor instability, overflow incontinence, sen-
sory urgency, and urinary retention) were made
according to International Continence Society defini-
tions.10 We considered women to have intrinsic sphinc-
ter deficiency if maximum urethral closure pressure
was at most 20 cm H2O, and if functional urethral
length was at most 1.4 cm or large leakage volume was
seen.11–14 Women who demonstrated severe detrusor
instability that precluded accurate urethral pressure
profilometry were excluded.
Groups were assigned according to the grade of
posterior wall defect, as defined by Baden and Walker9:
grade 0 (control group) indicating no posterior wall
descent, grade 1 indicating maximal posterior wall
descent no further than halfway down the vaginal
canal, grade 2 indicating maximal posterior wall de-
scent from halfway down the vaginal canal to just
above the hymeneal ring, and grade 3 indicating max-
imal posterior wall descent at or beyond the hymeneal
ring.
Within each group, we summarized age by using
means and standard deviations (SDs), parity by using
percentiles, and estrogen status by using proportions.
The distributions of age and parity across groups were
compared using analysis of variance and the Kruskal-
Figure 2. Posterior wall retracted with split speculum.
Myers et al Rectocoele and Urodynamics 711
Wallis test, respectively. The proportions of women Table 2. Number of Urodynamic Diagnoses by Posterior
estrogenized were compared by using Fisher exact test. Wall Grade
We recorded maximum urethral closure pressure, Grade GSI DI ISD DI 1 GSI DI 1 ISD Other*
functional urethral length, and leak volume with the 0 (n 5 20) 4 5 0 3 2 6
posterior wall in both the retracted and unretracted 1 (n 5 23) 9 4 4 3 2 1
positions. For each urodynamic index, our primary 2 (n 5 19) 11 1 3 3 1 0
outcome variable was the difference between the mea- 3 (n 5 28) 13 4 1 5 1 4

surements taken in the two positions. We estimated
grade-specific mean difference for each urodynamic
index; we also tested whether differences varied signif-
icantly by grade. To test for a grade effect on mean
difference, we used an F statistic from an analysis of
covariance model, treating the difference between re-
tracted and unretracted measurements as the response,
and the measurement in the unretracted position as the
covariate. When testing for a group effect in studies of
change, analysis of covariance controls for bias induced
by correlation between changes and baseline measure-
ments, also known as regression to the mean.15 All
analyses were performed using the program STATA 5.0
(STATA Corp., College Station, TX).
For sample size calculations, we fixed the type I error
rate (alpha) at 1% and power at 80%. We used sample
size formulas for analysis of variance, which requires
specification of within-grade and between-grade vari-
ances. For each outcome, the within-grade variance
(SD2) was estimated from pilot data. Between-group
variance refers to the variance of the grade-specific
mean changes and is determined by the effect size. For
each outcome, we assumed zero change in women with
grades 0 –2 posterior wall defects, so that effect size
refers to change in women with grade 3 posterior wall
defects.
For change in maximum urethral closure pressure, to
detect an effect of 10 cm H2O, assuming an SD of 10 cm
H2O, 23 women were needed in each group to realize
80% power. To detect a change of 0.5 cm in functional
Table 1. Demographic Characteristics Among Posterior
Wall Grade Groups
Posterior wall defect grade
0 1 2 3
Number 20 23 19
Age (y)*
Mean 52 56 59
SD 16 12 15
Parity†
25th percentile 0 2 2 2
50th percentile 3 3 3 3
75th percentile 3 3 4 4
% estrogenized‡ 55 48 47
SD 5 standard deviation.
* In test of equal means via analysis of variance, P 5 .288.
†
In test of equal medians via Kruskal-Wallis test, P 5 .217.
‡
In test of equal proportions via Fisher exact test, P 5 .957.
GSI 5 genuine stress incontinence; DI 5 detrusor instability; ISD 5
intrinsic sphincter deficiency.
* Overflow incontinence, sensory urgency, or urinary retention.
urethral length, assuming an SD of 0.4 cm, 15 women
were needed in each group for 80% power. Finally, to
detect a change in leakage volume of one unit, assum-
ing an SD of 0.7, 12 women were needed in each group
for 80% power. Sample sizes were calculated using
nQuery Advisor release 2.0 (Statistical Solutions Ltd.,
Boston, MA).
Results
Eleven women were excluded because of severe detru-
sor instability. Thus, of the 90 women included in the
study, 20 had grade 0 posterior wall defects (control
group); 23, grade 1; 19, grade 2; and 28, grade 3. There
were no significant differences in age, parity, or estro-
gen status among the grade groups (Table 1). All
women were white. The urodynamic diagnoses made
for the 90 women categorized by each posterior wall
grade are listed in Table 2.
Table 3 presents grade-specific adjusted mean differ-
ences and 99% confidence intervals (CIs) for each of the
urodynamic indices. We report adjusted differences
because measurements made in the unretracted (base-
line) position were associated highly with grade, and
unadjusted mean differences would reflect both grade
effects and baseline effects. The adjusted mean differ-
ences (also known as least squares means) were calcu-
lated by holding the baseline value constant across all
grades.
We found no significant between-grade variation in
mean differences of maximum urethral closure pressure
(F [3,85] 5 1.47; P 5 .229); however, women with grade
3 posterior wall defects did have the largest estimated
28 mean difference in closure pressure of 27.0 cm H2O
(99% CI 212.4, 21.6). We found no significant between-
59
grade variation in mean differences in functional ure-
11
thral length (F [3,85] 5 0.45; P 5 .716); however, women
with grade 3 posterior wall defects did have a statisti-
cally significant difference of 20.3 cm (99% CI 20.5,
20.1). Mean differences in leakage volume did vary
54
significantly by grade of posterior wall defect (F [3,85]
5 4.49; P 5 .006), and those with grade 3 defects
exhibited the largest difference of 0.7 units (99% CI 0.4,
1.0).

712 Myers et al Rectocoele and Urodynamics Obstetrics & Gynecology

Table 3. Adjusted Differences in Urodynamic Indices in Unretracted and Retracted Positions
Position

UDE Grade Unretracted Retracted D adj. 99% CI

MUCP (cm H2O) 0 63.7 (6.1) 59.3 (5.8) 22.2 (28.8, 4.4)
1 47.8 (5.4) 46.8 (5.2) 21.1 (26.8, 4.6)
2 47.4 (6.0) 43.3 (5.8) 24.3 (210.7, 2.2)
3 39.8 (5.0) 33.9 (4.8) 27.0 (212.4, 21.6)
FUL (cm) 0 2.3 (0.1) 2.2 (0.1) 20.1 (20.4, 0.1)
1 2.0 (0.1) 1.8 (0.1) 20.2 (20.4, 0.0)
2 2.1 (0.1) 1.9 (0.1) 20.2 (20.4, 0.0)
3 2.1 (0.1) 1.9 (0.1) 20.3 (20.5, 20.1)
Leakage volume (units) 0 0.7 (0.2) 0.8 (0.3) 0.1 (20.3, 0.5)
1 1.0 (0.2) 1.4 (0.2) 0.4 (0.0, 0.7)
2 1.3 (0.2) 1.5 (0.3) 0.1 (20.3, 0.5)
3 0.9 (0.2) 1.6 (0.2) 0.7 (0.4, 1.0)
UDE 5 urodynamic indice; D adj. 5 adjusted mean difference; CI 5 confidence interval; MUCP 5 maximum urethral closure pressure; FUL 5
functional urethral length.
Values given are mean (standard error).

Among those with grade 3 posterior wall defects, five
women demonstrated significant decreases in maxi-
mum urethral closure pressure, decreases in functional
urethral length, and increases in leakage volume when
the posterior wall was retracted with the split Pederson
speculum. These changes were significant enough to
convert their diagnoses from genuine stress inconti-
nence to intrinsic sphincter deficiency. The largest indi-
vidual decrease in maximum urethral closure pressure
when the posterior wall was retracted in any of the
women with grade 3 posterior wall defects was 15 cm
H2O.
Four women with grade 3 posterior wall defects had
a change in leakage volume from no leakage to some
leakage with retraction of the posterior vaginal wall,
thus their urodynamic diagnosis was changed from no
evidence of genuine stress incontinence to the presence
of genuine stress incontinence. These four women dem-
onstrated potential stress urinary incontinence when
the posterior wall was retracted. One of the four dem-
onstrated not only potential stress urinary incontinence,
but also intrinsic sphincter deficiency. No women with
grade 0 –2 posterior wall defects demonstrated potential
stress urinary incontinence.
Discussion
The results of this study demonstrate the importance of
reducing severe posterior wall support defects in the
absence of other severe support defects. In patients with
grade 3 posterior wall defects, we found significant
changes in maximum urethral closure pressure, func-
tional urethral length, and leakage volume when the
posterior wall is retracted. We theorize that these
changes in urodynamic indices occur because severe
posterior wall defects act to compress and to support
the anterior vaginal wall, thus keeping patients more or
completely continent.
We found a significant decrease in maximum urethral
closure pressure when a grade 3 posterior wall defect
was retracted. This finding is clinically relevant for a
woman with a grade 3 posterior wall defect who has
complaints of stress urinary incontinence and is found
on testing to have a maximum urethral closure pressure
in the range of 25–35 cm H2O. In this clinical situation,
reducing the posterior wall defect may reveal intrinsic
sphincter deficiency and ultimately may change the
treatment of this patient. If the maximum urethral
closure pressure is well above 35 cm H2O, it is unlikely
that the maximum urethral closure pressure will de-
crease more than 15 cm H2O; therefore, an additional
set of urethral pressure profiles with the posterior wall
retracted is probably not necessary.
It has been suggested by many authors16 –20 that
surgical treatment of intrinsic sphincter deficiency re-
quires either a suburethral sling operation or periure-
thral collagen injections. For the patients in our study
who revealed intrinsic sphincter deficiency in the re-
tracted position who ultimately went on to have sur-
gery, a suburethral sling of Mersilene (Ethicon Inc.,
Sommerville, NJ) mesh was performed. However, one
woman who demonstrated potential stress urinary in-
continence and intrinsic sphincter deficiency was not a
sling candidate; she only underwent a posterior repair.
Postoperatively, this woman did demonstrate the po-
tential stress urinary incontinence that was shown pre-
operatively and later underwent a periurethral GAX-
Collagen (Contigen, C.R. Bard Inc., Covington, GA)
injection.
Both a significant decrease in functional urethral
length and a significant increase in leakage volume
were found when the posterior wall was retracted in

VOL. 91, NO. 5, PART 1, MAY 1998 Myers et al Rectocoele and Urodynamics 713
women with grade 3 posterior wall defects. Shortened
functional urethral length and large leakage volumes,
although not diagnostic by themselves, can be indica-
tors of intrinsic sphincter deficiency when used along
with other criteria.11–14
Four women with grade 3 posterior wall defects
demonstrated “potential stress urinary incontinence.”
Therefore, all women with severe posterior wall defects
should undergo a preoperative urodynamic evaluation
to uncover potential stress urinary incontinence, even in
the absence of symptoms of urinary incontinence.
Again, no women with grade 0, 1, or 2 posterior wall
defects demonstrated potential stress urinary inconti-
nence.
An error that may have been introduced into the
results obtained in the group of women with grade 3
posterior wall defects is that the split Pederson specu-
lum may have been retracting the posterior wall beyond
its normal axis and consequently prohibiting the pelvic
floor musculature from contracting effectively, thus
decreasing maximum urethral closure pressure and
functional urethral length, increasing leakage volume,
and demonstrating potential stress urinary inconti-
nence incorrectly. However, if such artifact were
present, significant changes would have been seen in
women with grade 0 –2 posterior wall defects, and none
of these changes were observed. In fact, the bottom
blade of a split Pederson speculum was used intention-
ally, as opposed to a Graves speculum, to ensure that
overstretching of the vaginal introitus did not occur.
It is recommended currently that changes in urinary
incontinence should be sought when severe anterior
wall or apical defects are present. On the basis of our
findings, we recommend that in women with severe
posterior wall support defects, a urodynamic evalua-
tion be performed with the posterior wall retracted to
evaluate stress urinary incontinence and to reveal pos-
sible changes in maximum urethral closure pressure,
functional urethral length, and in leakage volume dur-
ing a cough stress test or to unmask potential stress
urinary incontinence.
References
1. Department of Health and Human Services. Urinary incontinence
in adults. National Institutes of Health consensus development
conference statement. Vol. 7. Bethesda, Maryland: U.S. Depart-
ment of Health and Human Services, 1988.
2. Diokno AC, Brock BM, Brown MB, Herzog AR. Prevalence of
urinary incontinence and other urological symptoms in the non-
institutionalized elderly. J Urol 1986;136:1022–5.
3. Rosenzweig BA, Pushkin S, Blumenfeld D, Bhatia NN. Prevalence
of abnormal urodynamic test results in continent women with
severe genitourinary prolapse. Obstet Gynecol 1992;79:539 – 42.
4. Stanton SL, Hilton P, Norton C, Cardozo L. Clinical and urody-
714 Myers et al Rectocoele and Urodynamics
namic effects of anterior colporrhaphy and vaginal hysterectomy
for prolapse with and without incontinence. Br J Obstet Gynaecol
1982;89:459 – 63.
5. Fianu S, Kjaeldgaard A, Larsson B. Preoperative screening for
latent stress incontinence in women with cystocele. Neurourol
Urodyn 1985;4:3–7.
6. Richardson DA, Bent AE, Ostergard DR. The effect of uterovaginal
prolapse on urethrovesical pressure dynamics. Am J Obstet Gy-
necol 1983;146:901–5.
7. Hertogs K, Stanton SL. Mechanism of urinary incontinence after
colposuspension: Barrier studies. Br J Obstet Gynaecol 1985;92:
184 – 8.
8. Bump RC, Fantl JA, Hurt WG. The mechanism of urinary conti-
nence in women with severe uterovaginal prolapse: Results of
barrier studies. Obstet Gynecol 1988;72:291–5.
9. Baden WF, Walker TA. Genesis of the vaginal profile: A correlated
classification of vaginal relaxation. Clin Obstet Gynecol 1972;15:
1048 –54.
10. Abrams P, Blaivas JG, Stanton SL, Andersen JT. The standardiza-
tion of terminology of the lower urinary tract function recom-
mended by the International Continence Society. Int Urogynecol J
1990;1:45–58.
11. American Uro-Gynecologic Society. Intrinsic sphincteric defi-
ciency: A clinical opinion. AUGS quarterly report no 2. Chicago:
American Uro-Gynecologic Society, 1995.
12. Horbach NS, Ostergard DR. Predicting intrinsic urethral sphincter
dysfunction in women with stress urinary incontinence. Obstet
Gynecol 1994;84:188 –92.
13. McGuire EJ, Fitzpatrick CC, Wan J, Bloom D, Sanvordenker J,
Ritchey M, et al. Clinical assessment of urethral sphincter function.
J Urol 1993;50:1452– 4.
14. Theofrastous JP, Bump RC, Elser DM, Wyman JF, McClish DK.
Correlation of urodynamic measures of urethral resistance with
clinical measures of incontinence severity in women with pure
genuine stress incontinence. The Continence Program for Women
Research Group Am J Obstet Gynecol 1995;173:407–14.
15. Fleiss JL. The design and analysis of clinical experiments. New
York: Wiley and Sons, 1986.
16. Nichols DH. The mersilene mesh gauze-hammock for severe
urinary stress incontinence. Obstet Gynecol 1977;138:88 –93.
17. McGuire EJ. Urodynamic findings in patients after failure of stress
incontinence operations. Prog Clin Biol Res 1981;78:351– 60.
18. Appell RA. Collagen injection therapy for urinary incontinence.
Urol Clin N Am 1994;21:177– 82.
19. McGuire EJ, Bennett CJ, Konnak JA, Sonda LP, Savastano JA.
Experience with pubovaginal slings for urinary incontinence at the
University of Michigan. J Urol 1987;138:525– 6.
20. Horbach NS, Blanco JS, Ostergard DR, Bent AE, Cornella JL. A
suburethral sling procedure with polytetrafluoroethylene for treat-
ment of genuine stress incontinence with low urethral closure
pressure. Obstet Gynecol 1988;71:648 –9.
Address reprint requests to:
Deborah L. Myers, MD
100 Dudley Street
Providence, RI 02905
Received October 6, 1997.
Received in revised form December 29, 1997.
Accepted January 23, 1998.
Copyright © 1998 by The American College of Obstetricians and
Gynecologists. Published by Elsevier Science Inc.
Obstetrics & Gynecology