Obstetric Anal Sphincter Lacerations
Victoria L. Handa, MD, Beate H. Danielsen, PhD, and William M. Gilbert, MD
OBJECTIVE: To estimate the frequency of obstetric anal
sphincter laceration and to identify characteristics associ-
ated with this complication, including modifiable risk fac-
tors.
METHODS: A population-based, retrospective study of over
2 million vaginal deliveries at California hospitals was
performed, using information from birth certificates and
discharge summaries for 1992 through 1997. We excluded
preterm births, stillbirths, breech deliveries, and multiple
gestations. The main outcome measure was obstetric anal
sphincter laceration (third and fourth degree).
RESULTS: The frequency of anal sphincter lacerations was
5.85% (95% confidence interval [CI] 5.82, 5.88), decreasing
significantly from 6.35% (95% CI 6.27, 6.43) in 1992 to
5.43% (95% CI 5.35, 5.51) in 1997 (P < .01). Using logistic
regression analysis, we identified primiparity as the domi-
nant risk factor (odds ratio [OR] for women with prior
vaginal birth 0.15; 95% CI 0.14, 0.15). Birth weight over
4000 g was also highly significant (OR 2.17; 95% CI 2.07,
2.27). Lacerations occurred more often among women of
certain racial and ethnic groups: Indian women (OR 2.5;
95% CI 2.23, 2.79) and Filipina women (OR 1.63; 95% CI
1.50, 1.77) were at highest risk. Episiotomy decreased the
likelihood of third-degree lacerations (OR 0.81; 95% CI
0.78, 0.85), but increased the risk of fourth-degree lacera-
tions (OR 1.12; 95% CI 1.05, 1.19). Operative delivery
increased the risk of sphincter laceration, with vacuum
delivery (OR 2.30; 95% CI 2.21, 2.40) presenting a greater
risk than forceps delivery (OR 1.45; 95% CI 1.37, 1.52).
CONCLUSION: Anal sphincter lacerations are strongly asso-
ciated with primiparity, macrosomia, and operative vagi-
nal delivery. Of the modifiable risk factors, operative vag-
inal delivery remains the dominant independent variable.
(Obstet Gynecol 2001;98:22530. 2001 by the Ameri-
can College of Obstetricians and Gynecologists.)
As maternal and fetal mortality have declined over the
past century, increased attention has been focused on the
morbidity of childbirth. A recent survey of female obste-
tricians in Britain1 found that 31% would prefer elective
cesarean delivery to minimize long-term maternal se-
From Department of Obstetrics and Gynecology, University of California Davis
School of Medicine, Sacramento, California; and Health Information Solutions,
Redwood City, California.
VOL. 98, NO. 2, AUGUST 2001
2001 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
quelae of vaginal delivery, such as urinary and anal
incontinence. Such concerns are creating new controver-
sies in womens health.
The prevalence of anal incontinence in women is
strongly associated with obstetric history and more spe-
cifically with obstetric lacerations of the anal sphincter.
Even with immediate diagnosis and repair of sphincter
lacerations, 40 50% of women report subsequent anal
incontinence.2 4 Because recognition and repair of lacer-
ations do not reliably prevent these sequelae, prevention
is paramount.
The epidemiology of obstetric anal sphincter lacera-
tion is incompletely understood. This complication has
been reported in 2.2% to 19% of vaginal births.5 8 Fac-
tors that have been associated with sphincter lacerations
include primiparity,5,7,9,10 birth weight,59,11 episioto-
my,511 and forceps delivery.5,6,8,10,12 Associations with
race6,7 have also been suggested. Prior studies have been
characterized by relatively small sample size and have
been performed primarily at teaching hospitals, poten-
tially limiting generalizablity.
The purpose of this study was to estimate the fre-
quency of anal sphincter laceration in a large, heteroge-
neous population and to identify risk factors for this
obstetric complication. Our goal was to identify modifi-
able risk factors and strategies for prevention.
MATERIALS AND METHODS
We employed a database of the California Office of State
Health Planning and Development, which links Califor-
nia birth certificates to maternal and newborn discharge
records since 1992. Ninety-eight percent of all California
deliveries are included in the database. Discharge data
include diagnostic codes (International Classification of
Diseases, 9th Revision) and procedural codes (Current
Procedural Terminology). This study was approved by
the institutional review boards of the California Office of
State Health Planning and Development and the Univer-
sity of California Davis.
We studied all vaginal deliveries between 1 January
1992 and 31 December 1997, excluding breech deliver-
ies, preterm deliveries, multiple gestations, and still-
births. The outcome of interest was injury to the anal
0029-7844/01/$20.00 225
PII S0029-7844(01)01445-4
sphincter, including both third- and fourth-degree
sphincter lacerations. A third-degree tear is defined as a
perineal laceration involving the anal sphincter. A
fourth-degree tear is defined as a laceration that involves
the anal sphincter and the rectal mucosa. Because most
prior studies of laceration have not distinguished be-
tween third- and fourth-degree lacerations,59 we consid-
ered them as a single outcome.
A number of independent variables were examined
for their association with anal sphincter lacerations. De-
mographic variables of interest included maternal age,
race, education, and insurance type (private insurance,
Medicaid, health maintenance organization, self-insured,
and all other types of insurance). We also examined
parity, which was classified into four categories: women
delivering their first child (primiparous), women deliv-
ering their second child with a history of prior cesarean
delivery (prior cesarean only), women with a history
of one or more previous vaginal deliveries but no prior
cesarean deliveries (multiparous), and women with at
least two prior deliveries and a history of cesarean deliv-
ery (multiparous with prior cesarean).
Obstetric variables included macrosomia (birth
weight greater than 4000 g), maternal diabetes, postdates
pregnancy, abnormalities of the first stage of labor, pro-
longed second stage of labor, shoulder dystocia, prior
cesarean delivery, and fetal distress. We also examined
labor interventions (induction of labor, epidural anesthe-
sia, episiotomy, forceps delivery, and vacuum delivery).
The type of episiotomy (midline or mediolateral) was
not specified.
We determined laceration rates for each study year.
The Cochran-Armitage trend test13 was used to examine
changes over time in laceration rates and obstetrical
interventions. We first used bivariate descriptive statis-
tics to identify characteristics associated with lacerations.
For each apparent association, we calculated odds ratios
(OR) and two-tailed 95% confidence intervals (CI). We
then used logistic regression to identify statistically sig-
nificant associations while controlling for possible con-
founding variables. For each independent variable, we
obtained OR and two-tailed 95% CI. We performed
separate logistic regressions for each year to examine
changes in associations over time.
RESULTS
The study population included 2,101,843 births. The
overall frequency of third- and fourth-degree laceration
was 5.85% (123,009 of 2,101,843) (95% CI 5.82, 5.88).
The incidence of anal sphincter lacerations decreased
significantly over the 6 years of data collection, from
6.35% (95% CI 6.27, 6.43) in 1992 to 5.43% (95% CI
226 Handa et al Anal Sphincter Lacerations
Figure 1. Annual incidence of third- and fourth-degree
anal sphincter lacerations at California hospitals, 1992
1997. The decrease over time is significant in both (P
.01).
Handa. Anal Sphincter Lacerations. Obstet Gynecol 2001.
5.35, 5.51) in 1997 (P .01) (Cochran-Armitage trend
test for binomial proportions 2: 19.38; P .001; Figure
1).
Table 1 characterizes the study population and sum-
marizes the results of bivariate analyses. The most strik-
ing finding was the strong association between sphincter
laceration and first vaginal birth. Compared with nullip-
arous women, multiparas had one-sixth the risk of
sphincter laceration (OR 0.14; 95% CI 0.13, 0.14).
Women with abnormal labor or shoulder dystocia were
also at markedly increased risk of sphincter laceration.
Strong associations were observed for several obstetric
interventions, including labor induction, episiotomy,
and forceps or vacuum delivery. Epidural anesthesia was
recorded in only 3.3% of all deliveries (69,642 cases),
suggesting probable undercoding of this procedure at the
time of discharge. Because of potential undercoding, this
variable was excluded from further analysis.
Multivariate logistic regression analysis was stratified
by year of delivery to control for the potential effect of
changes over time. The results for the 1997 delivery
cohort are shown in Table 2. Again, the largest indepen-
dent protective effect was multiparity (OR 0.15; 95% CI
0.14, 0.15). We also observed large differentials with
respect to race-ethnicity, especially among women of
Indian descent (OR 2.50; 95% CI 2.23, 2.79). Strong
associations were noted for fetal macrosomia (OR 2.17;
95% CI 2.07, 2.27) and shoulder dystocia (OR 2.67; 95%
CI 2.47, 2.89). Both forceps and vacuum deliveries were
associated with an increased risk of sphincter laceration
(OR 1.45, 2.30; 95% CI 1.37, 1.52 and 2.21, 2.40).
Episiotomy was protective for sphincter laceration, with
OR 0.89 (95% CI 0.86, 0.92). However, in a separate
analysis (results not shown), we found that episiotomy
increased the likelihood of fourth-degree laceration (OR
OBSTETRICS & GYNECOLOGY
Table 1. Characteristics of Women With and Without Obstetric Anal Sphincter Lacerations: Bivariate Analysis
Deliveries without Deliveries with
anal sphincter anal sphincter Odds ratio Continuity
lacerations lacerations (95% confidence adjusted
Characteristic (n 1,978,834) (n 123,009) interval) P
Demographic characteristics
Parity
Primiparous 710,254 (35.8%) 92,532 (75.2%) Reference Reference
Multiparous without prior cesarean 1,241,467 (62.7%) 22,694 (18.4%) 0.14 (0.13, 0.14) .01
Prior cesarean only 48,298 (2.4%) 7733 (6.3%) 1.23 (1.20, 1.26) .01
Multiparous with prior cesarean 50,789 (2.6%) 1806 (1.5%) 0.27 (0.26, 0.29) .01
Maternal age (y)
18 99,465 (5.0%) 7546 (6.1%) 1.22 (1.19, 1.25) .01
1835 1,170,106 (86.4%) 106,296 (86.4%) Reference Reference
35 169,042 (8.5%) 9155 (7.4%) 0.87 (0.85, 0.89) .01
Maternal race
White 692,687 (35.0%) 51,119 (38.7%) Reference Reference
Hispanic 936,361 (47.3%) 47,257 (38.4%) 0.68 (0.67, 0.69) .01
Black 130,741 (6.6%) 5197 (0.2%) 0.54 (0.52, 0.55) .01
Filipina 45,323 (2.3%) 4994 (4.1%) 1.49 (1.45, 1.54) .01
Other Pacific Islander 10,221 (0.5%) 493 (0.4%) 0.65 (0.60, 0.72) .01
Indian 12,955 (0.7%) 2153 (1.8%) 2.25 (2.15, 2.40) .01
Other Asian 129,220 (6.5%) 10,432 (8.5%) 1.09 (1.07, 1.11) .01
Native American 9364 (0.5%) 376 (0.3%) 0.54 (0.49, 0.60) .01
Did not finish high school 700,497 (35.4%) 32,985 (26.8%) 0.67 (0.66, 0.68) .01
Medical insurance
Medicaid 944,476 (47.7%) 45,213 (36.8%) Reference Reference
HMO 573,713 (29.0%) 43,623 (35.5%) 1.59 (1.57, 1.61) .01
Private 372,098 (18.8%) 29,226 (23.8%) 1.64 (1.62, 1.67) .01
Self-insured 56,906 (2.9%) 2919 (2.4%) 1.07 (1.03, 1.11) .01
Other 26,409 (1.3%) 1780 (1.4%) 1.41 (1.34, 1.48) .01
Obstetric characteristics
Fetal macrosomia (birth weight 4000 g) 206,233 (10.4%) 21,010 (17.1%) 1.77 (1.74, 1.80) .01
Prior cesarean delivery 100,484 (5.1%) 9681 (7.9%) 1.60 (1.56, 1.63) .01
Maternal diabetes 49,916 (2.5%) 3815 (3.1%) 1.24 (1.20, 1.28) .01
Postdates pregnancy 89,816 (4.5%) 7085 (6.0%) 1.34 (1.31, 1.38) .01
Abnormal first stage of labor 87,742 (4.4%) 13,050 (10.6%) 2.55 (2.51, 2.61) .01
Prolonged second stage 17,120 (0.9%) 5254 (4.3%) 5.11 (4.95, 5.28) .01
Fetal distress 131,674 (6.7%) 15,656 (12.7%) 2.05 (2.01, 2.08) .01
Shoulder dystocia 33,811 (1.7%) 6467 (5.3%) 3.19 (3.10, 3.28) .01
Obstetric interventions
Epidural anesthesia 62,819 (3.2%) 6823 (5.5%) 1.79 (1.74, 1.84) .01
Labor induction 227,997 (11.5%) 16,339 (13.3%) 1.18 (1.16, 1.20) .01
Episiotomy 753,045 (38.1%) 70,929 (57.7%) 2.21 (2.19, 2.24) .01
Forceps delivery 128,837 (6.5%) 24,609 (20.0%) 3.59 (3.54, 3.64) .01
Vacuum delivery 234,759 (11.9%) 42,943 (34.9%) 3.98 (3.94, 4.03) .01
HMO health maintenance organization.

1.12; 95% CI 1.05, 1.19), while decreasing the likelihood
of third-degree laceration (OR 0.81; 95% CI 0.78, 0.85).
There were no significant changes over time in the OR
shown in Table 2, with the exception of the OR associ-
ated with operative deliveries. Figure 2 illustrates
changes over time in the association between sphincter
lacerations and operative deliveries. From 1992 to 1997,
there was a significant increase in the OR for vacuum
delivery and a decrease in the OR for forceps delivery.
There was a simultaneous decline in the incidence of
forceps delivery and increase in the incidence of vacuum
delivery (Figure 3). The incidence of episiotomy signifi-
cantly decreased over time, from 41.6% (95% CI 41.4,
41.7) in 1992 to 35.3% (95% CI 35.1, 35.4) in 1997
(Cochran-Armitage trend test P 0.01).
DISCUSSION
The strength of this study is the size and heterogeneity of
the population examined. Prior studies5 8 have included
a relatively small number of cases (276 to 1124 sphincter
lacerations), usually from a single institution. The large

VOL. 98, NO. 2, AUGUST 2001 Handa et al Anal Sphincter Lacerations 227
Table 2. Factors Associated With Obstetric Anal Sphincter
LacerationsMultiple Logistic Regression Analysis
Odds ratio
(95% confidence
Characteristic interval)
Demographic characteristics
Parity
Primiparous Reference
Multiparous 0.15 (0.14, 0.15)
Prior cesarean only 1.12 (0.71, 1.77)
Multiparous with prior cesarean 0.28 (0.17, 0.44)
Maternal age (y)
18 0.81 (0.75, 0.87)
1835 Reference
35 1.09 (1.03, 1.16)
Maternal race Figure 2. Odds ratio (and 95% confidence interval) of
White Reference sphincter laceration for forceps delivery and vacuum deliv-
Hispanic 0.94 (0.90, 0.98) ery, 19921997.
Black 0.69 (0.63, 0.74)
Filipina 1.63 (1.50, 1.77) Handa. Anal Sphincter Lacerations. Obstet Gynecol 2001.
Other Pacific Islander 0.74 (0.57, 0.95)
Indian 2.50 (2.23, 2.79) Our data revealed some surprising findings regarding
Other Asian 1.37 (1.29, 1.45)
Native American 0.63 (0.48, 0.85) modifiable risk factors for obstetric sphincter lacerations.
Did not finish high school 1.04 (0.99, 1.08) The effect of episiotomy was the most unanticipated
Medical insurance finding, in that this procedure was associated with a 10%
Medicaid Reference decrease in sphincter laceration. Since the early 1980s,
HMO 1.32 (1.27, 1.38) several retrospective case-control studies suggested that
Private 1.31 (1.25, 1.38)
Self-insured 1.02 (0.91, 1.13) episiotomy independently increases sphincter lacera-
Other 0.98 (0.82, 1.16) tions, with OR as high as 8.9.7 We observed a significant
Obstetric characteristics decrease in the use of episiotomy over time and speculate
Fetal macrosomia (birth weight 2.17 (2.07, 2.27) that obstetricians and others are avoiding episiotomy in
4000 g) women perceived to be at increased risk of complex
Prior cesarean 0.94 (0.51, 1.48)
Maternal diabetes 1.24 (1.14, 1.36) lacerations. This may be responsible for the apparent
Postdates pregnancy 1.13 (1.05, 1.20) protective effect of episiotomy. However, we cannot
Abnormal first stage of labor 1.20 (1.12, 1.29) exclude the possibility that episiotomy is less consistently
Prolonged second stage of labor 1.49 (1.35, 1.66) documented (or coded) in the setting of a laceration. If
Fetal distress 1.31 (1.24, 1.37) the physician does not document an episiotomy in this
Shoulder dystocia 2.67 (2.47, 2.89)
Obstetric interventions setting (or if the episiotomy is not coded at hospital
Labor induction 1.01 (0.96, 1.06)
Episiotomy 0.89 (0.86, 0.92)
Forceps delivery 1.45 (1.37, 1.52)
Vacuum delivery 2.30 (2.21, 2.40)
Abbreviation as in Table 1.

number of cases in this report (123,009 sphincter lacera-

tions) allows us to simultaneously examine a large num-
ber of risk factors. Prior studies found anal sphincter
lacerations in 2.2% to 19% of deliveries.5 8 In our pop-
ulation, 5.85% (95% CI 5.82, 5.88) of women experi-
enced a third- or fourth-degree laceration at the time of
vaginal delivery. As in prior reports,5,7,11 our results Figure 3. Annual incidence of forceps delivery and vacuum
suggest that the dominant risk factor for anal sphincter delivery in California hospitals, 19921997. The decrease
degree laceration is primiparity. Compared to women in forceps delivery and increase in vacuum delivery are
with a prior vaginal delivery, primiparas had more than both significant (P .01).
six times the risk of a sphincter laceration. Handa. Anal Sphincter Lacerations. Obstet Gynecol 2001.

228 Handa et al Anal Sphincter Lacerations OBSTETRICS & GYNECOLOGY

discharge), the result would be an apparent weakening of
the association between episiotomy and laceration. This
is a potential weakness of this study design.
Our findings indicate that episiotomy may decrease
third-degree but increase fourth-degree lacerations. We
speculate that episiotomy may cause a third-degree lac-
eration to extend to the rectal mucosa. Because some
aspects of continence may be more severely compro-
mised by fourth-degree lacerations,3 our data do not lead
us to recommend a more liberal use of episiotomy.
We found an increased risk of laceration with both
forceps and vacuum deliveries. This effect was relatively
modest, however. We observed an OR of 1.4 for forceps
delivery, in contrast to published reports estimating that
sphincter lacerations increase seven to eight times with
forceps.5,6 We speculate that prior reports have affected
obstetric practice in California, with obstetricians and
others avoiding forceps delivery among women per-
ceived to be at increased risk of sphincter lacerations.
Over 5 years, the incidence of forceps delivery de-
creased, whereas the incidence of vacuum delivery in-
creased, consistent with national trends.14 As vacuum
deliveries have apparently replaced some forceps deliv-
eries in California, there has been an observed increase
in the association between sphincter lacerations and vac-
uum delivery. We speculate that vacuum deliveries are
being substituted for forceps among women believed to
be at high risk for perineal trauma.
Our results agree with prior reports suggesting a
strong association between sphincter laceration and in-
creased birth weight.5,7,8 This association is presumably
due to the mechanical stress of delivering a large baby.
Other factors that may reflect disproportion between the
size of the baby and the maternal pelvis include shoulder
dystocia, labor abnormalities, and prolonged second
stage of labor. Our findings suggest that these obstetric
diagnoses are all strongly associated with sphincter lac-
eration.
Racial differences in the incidence of obstetrical lacer-
ations have been previously reported.7 The increased
risk of laceration among Asian women, Filipinas, and
Indian women was striking in our study. Possible expla-
nations for our observations include differences in body
type7 and variations in perineal anatomy.15 We were not
able to control for maternal height and weight in this
model and these may confound apparent racial differ-
ences.
Although this study is based on a large, heterogeneous
population, the use of coded discharge data can be
criticized. Less critical diagnoses may be omitted in
complex cases. Documentation and data collection meth-
ods are not standardized, and coding practices may vary.
One example in our data set is the apparent undercoding
VOL. 98, NO. 2, AUGUST 2001
of epidural anesthesia. Prior hospital-based studies have
suggested that epidural anesthesia is not independently
associated with sphincter laceration.12 Therefore, we do
not feel that the lack of data on this procedure is a
significant source of bias in our results. However, it does
raise the question of whether other misclassifications
could affect apparent associations. To assess the validity
of the coding of sphincter lacerations, data were obtained
from an independent, blinded review of a stratified sam-
ple of cases in this database (personal communication,
2001, P. S. Romano, University of California Davis,
Sacramento, CA). The sensitivity and positive predictive
value of coding for lacerations were 90.3% and 91.5%.
These findings are reassuring. Nevertheless, accurate
coding is dependent on accurate documentation in the
medical record.
Another limitation of this study is the possibility that
important risk factors may be missing from the model.
For example, we were unable to control for the experi-
ence and training of the birth attendant, measures of
maternal body type, and other obstetrical interventions
(such as maternal position at the time of delivery and the
use of perineal massage or fundal pressure). These inter-
ventions may influence the risk of lacerations.7,11,16 18
Our data suggest that anal sphincter lacerations are
decreasing but remain a concern, especially for women
undergoing a first vaginal birth. Observed trends in the
use of episiotomy and forceps delivery suggest that the
obstetric community may be responding to concerns
about these procedures. The majority of risk factors for
sphincter lacerations, such as parity and birth weight, are
not modifiable. Operative delivery remains the domi-
nant modifiable independent variable.
REFERENCES
1. Al-Mufti R, McCarthy A, Fisk NM. Survey of obstetri-
cians personal preference and discretionary practice. Eur J
Obstet Gynecol Reprod Biol 1997;73:1 4.
2. Sorensen SM, Bondesen H, Istre O, Vilmann P. Perineal
rupture following vaginal delivery. Long-term conse-
quences. Acta Obstet Gynecol Scand 1988;67:315 8.
3. Haadem K, Dahlstrom JA, Ling L, Ohrlander S. Anal
sphincter function after delivery rupture. Obstet Gynecol
1987;70:53 6.
4. Kammerer-Doak DN, Wesol AB, Rogers RG, Dominguez
CE, Dorin MH. A prospective cohort study of women
after primary repair of obstetric anal sphincter laceration.
Am J Obstet Gynecol 1999;181:131723.
5. Angioli R, Gomez-Marin O, Cantuaria G, OSullivan MJ.
Severe perineal lacerations during vaginal delivery: The
University of Miami experience. Am J Obstet Gynecol
2000;182:10835.
Handa et al Anal Sphincter Lacerations 229
 6. Robinson JN, Norwitz ER, Cohen AP, McElrath TF,
Lieberman ES. Episiotomy, operative vaginal delivery,
and significant perineal trauma in nulliparous women.
Am J Obstet Gynecol 1999;181:1180 4.
7. Green JR, Soohoo SL. Factors associated with rectal injury
in spontaneous deliveries. Obstet Gynecol 1989;73:732 8.
8. Peleg D, Kennedy CM, Merrill D, Zlatnik FJ. Risk of
repetition of a severe perineal laceration. Obstet Gynecol
1999;93:1021 4.
9. Buchhave P, Flatow L, Rydhstroem H, Thorbert G. Risk
factors for rupture of the anal sphincter. Eur J Obstet
Gynecol Reprod Biol 1999;87:129 32.
10. Wood J, Amos L, Rieger N. Third degree anal sphincter
tears: Risk factors and outcomes. Aust N Z J Obstet
Gynaecol 1998;38:414 7.
11. Zetterstrom J, Lopez A, Anzen B, Norman M, Holmstrom
B, Mellgren A. Anal sphincter tears at vaginal delivery:
Risk factors and clinical outcome of primary repair. Obstet
Gynecol 1999;94:21 8.
12. Robinson JN, Norwitz ER, Cohen AP, McElrath TF,
Lieberman ES. Epidural analgesia and third- or fourth-
degree lacerations in nulliparas. Obstet Gynecol 1999;94:
259 62.
13. Margolin BA. Test for trend in proportions. In: Kotz S,
Johnson NL, eds. Encyclopedia of statistical sciences. Vol-
ume 9. New York: John Wiley & Sons, 1988:334 6.
230 Handa et al Anal Sphincter Lacerations
14. Curtin SC, Park MM. Trends in the attendant, place, and
timing of births, and the use of obstetric interventions:
United States, 1989 97. National Vital Statistics Reports
1999;47:112.
15. Zacharin RF. A Chinese anatomythe pelvic supporting
tissues of the Chinese and Occidental female compared
and contrasted. Aust N Z J Obstet Gynecol 1977;17:111.
16. Shipman MK, Boniface DR, Tefft ME, McCloghry F.
Antenatal perineal massage and subsequent perineal out-
comes: A randomised controlled trial. Br J Obstet Gynae-
col 1997;104:78791.
17. Labrecque M, Eason E, Marcoux S, Lemieux F, Pinault J,
Feldman P, et al. Randomized controlled trial of preven-
tion of perineal trauma by perineal massage during preg-
nancy. Obstet Gynecol 1999;180:593 600.
18. Sampselle CM, Hines S. Spontaneous pushing during
birth. J Nurse Midwifery 1999;44:36 9.
Reprints are not available. Address correspondence to: Victo-
ria L. Handa, MD, Department of Obstetrics and Gynecology,
Johns Hopkins University School of Medicine, 600 North
Wolfe Street, Harvey 319, Baltimore, MD 21287.
Received November 30, 2000. Received in revised form April 9, 2001.
Accepted April 12, 2001.
OBSTETRICS & GYNECOLOGY