Acra Obsrer G ~ ~ w r m

S l~ m d1997; 76: 667 672 cup&P”& 0 A<fu ObJret Ginr<olS‘wtd I997

Primed in Drnniark ail ri,Iirs reservrd
Acta Obstetricia et
Gynecologica Scandinavica
ISSN 0001-6349

ORIGINAL ARTICLE

Urinary retention in the post-partum period

The relationship between obstetric factors and the post-partum post-void
residual bladder volume
SHING-KAIYIP, GEOFFREY
BRIEGER,LIN-YEEHIN AND TONYCHUNG
From the Department of Obstetrics and Gynecology, Chinese University of Hong Kong, Hong Kong

Acfu Obstet Gyneco/ Sand 1997; 76: 667-672. 0 Acta Obstet Gynecol Scand I997

Objective. The three objectives of this study are: to investigate the incidence of post-partum
urinary retention after vaginal delivery, to investigate the relationship between various obstet-
ric parameters and the post-partum post-void residual bladder volume and to study the natu-
ral progression of the post-void residual bladder volume in patients with covert post-partum
urinary retention.
Methodv. Women who had a vaginal delivery (n=691) in a teaching hospital during a 2-
month period were studied. They were classified into three groups: normal patients, those
with overt urinary retention, and covert urinary retention. Their day 1 post-partum post-void
residual bladder volume were recorded and analyzed with respect to the obstetric parameters.
Patients with covert retention were followed up daily with ultrasound to monitor their post-
void residual volume.
Results. The incidences of overt and covert retention in our unit were 4.9‘5, and 9.7%, respec-
tively. The overall incidence of post-partum urinary retention after vaginal delivery was
14.6%. The duration of the first and second stages of labor were significantly associated with
the post-partum post-void residual bladder volume. In all patients with covert retention, their
post-void residual volume returned to normal within 4 days.
Conclusion. Post-partum urinary retention is a common phenomenon that may be related to
the process of parturition. Covert retention is a self-limiting phenomenon and specific treat-
ment is unnecessary.

Key words; post-partum; ultrasound; urinary retention

Submitted 23 August. 1996

Accepted 28 December, 1996

Urinary retention in the post-partum period is a leave it in situ for a few days. Some obstetric units
common event and the reported incidence ranges adopt a very vigorous management protocol for
from 1.7 to 17.9%(1). The pathophysiology of post- the detection of post-partum urinary retention in-
partum urinary retention is poorly understood. The cluding early catheterization of the bladder to re-
obstetric parameters commonly associated with lieve the retention ( I , 3). Some even advocate the
urinary retention are nulliparity, instrumental deliv- use of early suprapubic catheterization to hasten
ery and epidural analgesia (2). However, Kerr-Wil- bladder recovery (4). Such management may not
son et al. have reported that the current manage- always be necessary and are not without risk to
ment of labor does not predispose women to post- the patients. In-dwelling catheters have been as-
partum bladder hypotonia and urinary retention ( 3 ) sociated with patient discomfort, urinary tract in-
but there are few population-based studies that have fections, urethral mucosal irritation and sub-
investigated post-partum urinary retention and sequent urethral stricture formation. Suprapubic
post-void residual bladder volume. catheters may cause serious complications such as
The usual management of post-partum urinary bowel perforation, intra-peritoneal leakage of
retention is to insert an in-dwelling catheter and urine or rarely necrotizing fasciitis (5).
0 Actu Obstet Gynecol Scund 76 (1997)
668 S.-K. Yip et al.
The purpose of the present study is to investi- lapping with the overt retention group. An Aloka
gate the association between various obstetric par- SSD2000 real-time ultrasound machine with a 3.5-
ameters and post-partum urinary retention as MHz curvi-linear abdominal probe was used for
measured by the post-partum post-void residual the examinations. All patients were asked to empty
bladder volume. A minimal intervention manage- their bladder 15 minutes before the scan. The
ment for patients with asymptomatic urinary re- amount of the post-void residual bladder volume
tention was also tested in this study. in the bladder was estimated by measuring the 3
orthogonal diameters (dl, d2, d3) and calculated
using the formula for approximation of the ellip-
Material and methods
soid ( 6 ) :
A prospective descriptive study was conducted be-
tween March 12 to May 12, 1996 and a total of x x d l Xd2Xd3
volume =
707 women delivered vaginally in the authors’ in- 6
stitution were eligible for the study. Informed ver-
bal consent had been obtained from all patients where d 1=widest diameter in the transverse scan
recruited into the study. The exclusion criteria d2=anterior-posterior diameter in the
wcrc: longitudinal scan
1. All deliveries by cesarean section. d3=cephalo-caudal diameter in the longi-
2. All patients with in-dwelling catheterization tudinal scan
during labor, or patients with in-dwelling cath- If the post-void residual bladder volume was
eterization inserted after delivery for reasons greater than or equal to 150 ml on post-partum
other than acute urinary retention, e.g. severe day 1, further daily ultrasound examinations
pre-eclampsia. would be performed until the post-void residual
3 . Multiple pregnancies. bladder volume was less than 150 ml. There was
The patients were divided into three categories: no other intervention.
Group I (overt urinury retention): Patients with in- The following obstetric parameters for each pa-
dwelling catheterization after delivery because of tient were examined: age, parity, duration of first,
acute post-partum urinary retention. This group second and third stage of labor, mode of delivery
of patients had trouble voiding and presented with (normal vaginal delivery or instrumental delivery),
strain to void, urgency, frequency, and strangury. birth weight, presence of unanticipated birth canal
All patients in this group had indwelling catheter trauma like vaginal or perineal lacerations, epi-
inserted within 9 hours of delivery. dural analgesia, and episiotomy. The first stage of
labor is defined as the onset of cervical dilatation
Group 2 (covert uriniiry retention): Patients who associated with uterine contractions to full cervical
had no urinary symptoms but with a post-void re- dilatation. The second stage of labor is defined as
sidual bladder volume of more than or equal to the period from full cervical dilatation to the full
150 ml on post-partum day 1. expulsion of the fetus.
Statistical analyses were performed using com-
Group 3 (normal patients): Patients who had no puter software package Statistical Package for the
urinary symptoms and their post-void residual Social Sciences (SPSSTM)for Windows version 6.0.
bladder volume was less than I50 ml on post-part- x2 test, Student’s t test and multiple linear re-
um day I . gression were used to analyze the data.
As there is no general agreement on volume
definition of pathological urinary retention, an ar- Results
bitrary volume of 150 ml was chosen as the normal
upper limit of post-void residual bladder volume During the 2-month study period, 707 women were
and it is a commonly used criterion in clinical delivered vaginally in the authors’ institution and
practice. 691 of these agreed to participate in the study. Six-
Ultrasound examination was used to identify teen patients were excluded because they had in-
patients with covert urinary retention. A trans-ab- dwelling catheter inserted during labor or in the
dominal ultrasound scan was performed in the post-natal ward for reasons unrelated to urinary
morning of the first day of delivery to measure the retention such as severe pre-eclampsia or post-par-
post-void residual bladder volume. All day 1 post- tum hemorrhage.
partum patients who had not had an in-dwelling In the 691 patients, there were 34 with overt re-
catheter were scanned. The ultrasound examin- tention (Group l), 67 with covert retention (Group
ations were performed by one of the authors 2), and 590 normal patients (Group 3). On post-
(SKY) at least 9 hours after delivery to avoid over- partum day 1 the incidence of overt retention was
0 Actri Obstrt Gynecol Scanrl 76 (1997)
 Urinary retention in the post-partum period 669
4.9%, and the incidence of covert retention was The mean duration of first stage of labor was
9.7%. The overall incidence of patients with post- statistically longer in both retention groups
void residual bladder volume of 2-150 ml was (Groups 1 and 2) than that of Group 3. There was
14.6%. The post-void residual bladder volume no difference in the duration of labor between
(meants.d.) of the 3 groups were 834.05359.0 ml Groups 1 and Group 2.
(Group l), 359.71223.2 ml (Group 2), and The mean duration of second stage of labor was
39.9k30.2 ml (Group 3). There were no statisti- statistically longer in both retention groups
cally significant differences in mean age, duration (Groups 1 and 2) than that in Group 3 but not
of third stage of labor and birth weight of the three statistically different between Groups 1 and 2.
groups. There was also no statistically significant Instrumental delivery was associated with statis-
difference in the number of genital tract trauma in tically more overt or covert urinary retention (x2=
the three groups (Table I). 23.01, p<O.OOOOl). The same holds for multiparity

Table I. Obstetric characteristics of the three groups of patients

Group 1 Group 2 Group 3

Study population (n=34) (n=67) (n=590)
n=691 mean 2s.d. mean+-s.d. meants.d. Significancea

Incidence (Yo) 4.9 9.7 85.4

Post-partum day 1 post-void residual bladder
volume (ml) 834.02359.0 359.72223.2 39.9230.2

Age 28.825.0 28.7?5.2 28.6?5.0 1 vs 2 NS

2 vs 3 NS
1 vs 3 NS
Parity average 0.5-tO.6 0.720.9 0.920.9 1 vs 2 NS
2 vs 3 NS
1 vs3 <0.05
First stage duration (minutes) 400.1 2232.9 380.1 2253.8 291.1 +184.8 1 vs 2 NS
2 vs 3 40005
1 vs 3 <0.001
Second stage duration (minutes) 42.7t39.5 30.2231 .O 21.6224.4 1 vs 2 NS
2 vs 3 <0.01
1 vs 3 <0.0001
Third stage duration (minutes) 10.6+4.9 11.I k 6 . 0 11.8212.1 1 vs 2 NS
2 vs 3 NS
1 vs3 NS
Birth weight (grams) 3278.5?349.9 3140.42 383.1 3225.42487.1 1 vs 2 NS
2 vs 3 NS
1 vs 3 NS

Group 1 Group 2 Group 3 x2 (2df) Significanceb

Instrumental delivery
No 20 54 517 23.01 <0.00001
Yes 14 13 73
Epidural analgesia
No 29 63 578 19.68 <0.0005
Yes 5 4 12
Episiotomy
No 2 13 150 7.57 <0.05
Yes 32 54 440
Genital tract trauma
No 27 52 446 0.37 NS
Yes 7 15 144
Parity
Nulliparous 19 34 220 8.57 <0.05
Muitiparous 15 33 370
~~~

Group 1 =overt retention Group 2=covert retention Group 3=normal patients Genital trauma=unantrcipated genital tract lacerations during labor a=Student's
ttest comparing the 3 groups b=x2test with 2 degrees of freedom NS=not statistically significant

0 Actu Obstet Gynecol Scund 76 (1997)

670 S.-K. Yip ef nl.
Table I0 Multiple regression analysis of patient characteristics and the post- had been assessed from post-partum day 2 or day
void residual bladder volume on post-partum day 1 3 onwards (2, 3). We chose to assess the post-void
Coefficient Standard error residual bladder volume from post-partum day 1
Variable b t Significance- because this most closely reflected the effect of lab-
or and enabled us to chart the course of bladder
First stage
duration 0 138 0047 2 942 <0.005
function from shortly after delivery.
Second stage Ultrasound assessment of post-void residual
duration 1003 0347 2888 <0.005 bladder volume had been shown to be accurate
(2). The trans-abdominal ultrasound technique
*-Backward stepwise multiple analysis
was employed because it is associated with mini-
mal patient discomfort and is as accurate in esti-
mating volumes of intra-abdominal structures as
(x2=X.57, p<0.05), epidural analgesia (x2=19.68, the more invasive transvaginal route (7). The
p<0.0005), and episiotomy (x2=7.57, p<0.05). formula for the volume of an ellipsoid was
There was no difference in the incidence of birth chosen to estimate the post-void residual bladder
canal trauma betwcen the three groups (x2=0.37,
y>0.05). n x d l xd2Xd3
Backward stepwise multiple regression analysis for its practi-
controlling for epidural analgesia was performed 6
to test the obstetric parameters (age, parity, dur-
ation of first, second and third stage of labor, birth cality and accuracy. It requires only the measure-
weight, mode of delivery, episiotomy, genital tract ment of the three orthogonal diameters of the
trauma, and parity) in order to determine which bladder, namely the cephalo-caudal and the an-
obstetric parameters have the strongest association terior-posterior diameters in the longitudinal
with the post-void residual bladder volume on scan, and the widest diameter in the transverse
post-partum day 1. The duration of first and sec- scan, all of which are easy to measure using the
ond stage of labor were significantly related with trans-abdominal ultrasound, Griffiths et al. have
the post-void residual bladder volume on post-par- demonstrated the bladder volume thus estimated
tum day 1 (Table 11). The other variables did not has an accuracy of 0.84 and the repeatability of
show any significant relationship. the measurement was 2 10.2% (6).
A progressive decrease in the number of patients In the initial analyses of our data, urinary reten-
with post-void residual bladder volume 2 150 ml tion was shown to be related to several obstetric
was documented (Table 111). None of the patients parameters, namely parity, duration of first and
had a post-void residual bladder volume 2 1 50 ml second stage of labor, instrumental delivery, episi-
on post-partum day 4. otomy, and epidural analgesia. Some of these in-
itial findings were compatible with previous similar
studies. Nulliparity was found to be associated
Discussion
with more post-partum urinary retention (1, 2).
Cesarean section patients were excluded from this Nulliparous women also had a significantly in-
study because the aim of the present study was to creased post-partum maximum cystometric ca-
examine the relationship between labor and va- pacity and urine volume at first desire to void in
ginal delivery and the lower genito-urinary tract urodynamic studies ( 3 ) . In Andolf et a1.k study,
and subsequently the effect on the post-partum forceps delivery and vacuum extraction were
post-void residual bladder volume. shown to be associated with a higher rate of
In previous studies, bladder volume or function asymptomatic urinary retention on day 3 post-par-

Table 111 Number of patients with covert urinary retention detected by serial ultrasound examinations. Student's t test comparing the post-void residual bladder
volumes on consecutive days (post-partum days 1, 2 and 3) showed no statisticai significant difference
~

Number of patients with PVR

volume 2150 rnl (covert Percentage in study Post-void residual bladder Range of post-void residual
retention) population n=691 volume (ml) rnean5s.d. bladder volume (ml)

Day 1 post-partum 67 9.70 359.72223.2 150.2-1067.9

Day 2 post-partum 12 1.74 260.2+101.5 152.8-448.2
Day 3 post-parturn 2 0.29 179.9?13.0 170.7-189.1
Day 4 post-partum 0 0 - -
 Urinury retention in the post-purtum period 671
tum (2). Epidural analgesia’s effect on the urinary It is unclear whether the two groups of retention
bladder and urinary retention has been well docu- (overt and covert) patients represent two genuinely
mented (1, 2, 8, 9, 10). Our study patients who had distinct groups with separate pathophysiology, or
had episiotomy or long labor also had a higher they constitute a spectrum of the same disease.
rate of urinary retention, which has not been re- Further investigations using urodynamics or elec-
ported in other studies. In our study, long labor is trophysiological techniques like electromyography
defined as either the first stage of labor and was is necessary to clarify this matter.
more than 12 hours or the second stage of labor, We have demonstrated that covert urinary reten-
which was more than 2 hours. tion is a self-limiting phenomenon. The number of
The major contributory obstetric parameters to patients with significant urinary retention de-
post-partum urinary retention were the duration of creased rapidly and progressively and none had a
the first and the second stages of labor. We believe significant post-void residual bladder volume 4
this is the first study that has used multiple re- days after delivery (Table 111). No specific treat-
gression to determine which obstetric parameters ment was required in these women as long as they
are significant variables in post-partum urinary re- remained asymptomatic. Some of these women
tention. The length of the first or second stages of had post-partum post-void residual bladder vol-
labor were directly related to the post-partum post- umes as large as 1000 ml.
void residual bladder volume. The regression analy- Detection of patients with covert urinary reten-
sis was performed with epidural analgesia con- tion in the post-partum period did not lead to
trolled for several reasons. Epidural analgesia is al- short-term benefit. The long term effect of this
ready a well documented associated factor with short term and self-limiting post-partum covert
post-partum urinary retention (1,2,8,9,10) and the urinary retention is unclear. Andolf et al. followed
intention of this study is to investigate other poss- up a group of such patients with post-partum uri-
ible obstetric parameters associated with post-part- nary retention 4 years after delivery and they
um urinary retention. The number of patients in this found one third had voiding difficulties (2). How-
study who had epidural analgesia was small (n= 19), ever, the follow-up was incomplete and the num-
and the majority of patients who had epidural anal- bers were small. A proper randomized, controlled
gesia also had urinary retention (10/19=52.6%). In- trial with systematic long-term follow-up of
clusion of such strong urinary retention-associated women who have post-partum urinary retention is
covariate into the regression model will invariably necessary to detirmine whether catheterization
mask the rest of the covariates. prevents urinary continence problems. The relief
One hypothesis for the association between post- of the retention in the women with asymptomatic
partum urinary retention and the duration of the retention may be counterproductive if it leads to
first and second stage of labor is that during pro- more intervention-related problems than it pre-
tracted labor the presenting part of the fetus may vents in the long term.
exert pressure on the pelvic floor and the pelvic
soft tissues, which include the pelvic nerve plexus-
es. This may subsequently lead to either urinary References
outflow obstruction by tissue edema, or detrusor 1. Saultz JW, Toffler WL, Shackles JY. Postpartum urinary
dysfunction due to neuropraxia. These pressure-re- retention. J Am Board Fam Pract 1991; 4(5): 3414.
2. Andolf E, Iosif CS, Jorgensen C, Rydstrom H . Insidious
lated injuries to the components of the voiding urinary retention after vaginal delivery: Prevalence and
mechanisms may lead to urinary retention. symptoms at follow-up in a population-based study. Gyn-
It is logical to speculate that the longer the lab- ecol Obstet Invest 1994; 38: 51-3.
or, the longer the pelvic structures would be ex- 3. Kerr-Wilson RHJ, Thompson SW, Orr JW, Davis RO,
posed to unusual pressure. It has been shown that Cloud GA. Effect of labor on the postpartum bladder. Ob-
stet Gynecol 1984; 64: 115-7.
to expel the fetus the uterus needs to generate sub- 4. Jeffery TJ, Thyer B, Tsokos N, Taylor JD. Chronic urinary
stantial pressure during the first and second stages retention postpartum. Aust NZ J Obstet Gynaecol 1990;
of labor (1 1, 12). This pressure is transmitted onto 30(4): 364-6.
the pelvic floor structures in the process of birth 5. Bump RC. Prevention and management of complications
following continence surgery. In: Ostergard DR, Bent AE,
leading to injury and perhaps urinary retention. eds. Urogynecology and urodynamics - theory and prac-
The duration of labor may be considered as pro- tice. 3rd edition. Baltimore: Williams and Williams, 1985:
portional to the total amount of force that the pel- 480.
vic structures have absorbed (12) and therefore it 6.Griffiths CJ, Murray A, Ramsden PD. Accuracy and re-
is not surprising that it is related to urinary reten- peatability of bladder volume measurement using ultra-
sonic imaging. J Urol 1986; 136: 808-12.
tion. Likewise, stretching of the perineal muscles I. Granberg S, Wikland M. Comparison between endovaginal
during labor has been shown to lead to neuroprax- and transabdominal transducers for measuring ovarian vol-
ia and long term urinary problems ( I 3). ume. J Ultrasound Med 1987; 6: 649-53.

0 Actu Ohstct Gjnwof Scum! 76 (1997)

612 S.-K. Yiy et d
8. hsantila R, Eklund P, Rosenberg PH. Epidural analgesia 12. Fairlie FM, Phillips GF, Andrews BJ, Calder AA. An
with 4 nig of morphine following Cesarean section: Effect analysis of uterine activity in spontaneous labour using ;I
of injected volumc. Acta Anaesthesiol Scand 1993; 37(8): microcomputer. Br J Obstet Gynaecol 1988; 95: 57-64.
7647. 13. Snooks SJ, Badenoch DF, Tiptaft RC, Swash M. Perinenl
9. Ramsay IN. Torbet TE. Incidence of abnormal voiding par- nerve damage in genuine stress incontinence - an electro-
arnctcrs in the immediate post-partum period. Neurourol physiological study. Rr J LJrol 1985; 57: 422-6.
Urodyn 1993; 1 3 2 ) : 179-83.
10. Weil A, Reyes 1-1, Rottenberg RD, Beguin F, Hcrrman WL.
Effect of lumbar epidural analgesia on lower urinary tract
function in the immediate post-partum period. Br J Obslet Addrtss ji)r correspondcwt,:
Gynaecol 1983; 90: 428-32. Dr. Shing-Kai Yip, M.R.C.O.G.
11. Arulkumaran S, Gibb D M E Ratnani SS, Kun KC, Heng Department of Obstetrics and Gymecology
SH. Total uterinc activity in induced labour an index of
~ Chinese University of Hong Kong
cervical and pelvic tissue resistance. Br J Obstet Gynaecol Shatin, NT
1985; 92: 693-7. Hong Kong

0 Artu Obstet Gvrrecol S c u d 76 (1997)