OBJECTIVE: To estimate the differences in immediate ma- Data from the last 3 decades confirm that the rate of
ternal and neonatal effects of forceps and vacuum-assisted operative vaginal deliveries is decreasing in North Amer-
deliveries. ica.1– 4 The United States, Canada, Eastern Europe, and
METHODS: We conducted a medical record review of all South America prefer the use of forceps extractions. The
forceps and vacuum-assisted deliveries that occurred from vacuum is the instrument of choice in Western Europe,
January 1, 1998, to August 30, 1999, at Winthrop-Univer- Asia, Israel, and the Middle East.5
sity Hospital. Maternal demographics and delivery charac- It has been reported that 60% of residency programs in
teristics were recorded. Maternal outcomes, such as use of the United States perform less than 10% of their total
episiotomy and presence of lacerations, were studied. Neo- deliveries with the assistance of the vacuum or forceps.6 At
natal outcomes evaluated were Apgar scores, neonatal in- our institution, the use of vacuum has increased whereas
tensive care unit admissions, cephalohematomas, instru- forceps use has decreased during the past 10 years. Six
ment marks and bruising, and caput and molding. percent of the total deliveries are operative vaginal deliver-
RESULTS: Of 508 operative vaginal deliveries, 200 were ies. Review of the literature suggests differential maternal
forceps and 308 were vacuum assisted. Forceps were used and neonatal complications between forceps and vacuum
more often than vacuum for prolonged second stage of assisted deliveries.7–11 The purpose of this study was to
labor (P ⴝ .001). There was a higher rate of epidural (P ⴝ estimate the differences in immediate maternal and neona-
.02) and pudendal (P < .001) anesthesia, episiotomies (P ⴝ tal effects of forceps and vacuum-assisted deliveries in a
.01), maternal third- and fourth-degree perineal (P < .001) community-based teaching hospital with a residency pro-
and vaginal lacerations (P ⴝ .004) with the use of forceps, gram in obstetrics and gynecology.
whereas periurethral lacerations were more common in
vacuum-assisted (P ⴝ .026) deliveries. More instrument
marks and bruising (P < .001) were found in the neonates MATERIALS AND METHODS
delivered by forceps, whereas there was a greater incidence We performed a record review of all forceps and vacu-
of cephalohematomas (P ⴝ .03) and caput and molding um-assisted deliveries that occurred from January 1,
(P < .001) in the neonates delivered with vacuum. Multi- 1998, through August 30, 1999, at Winthrop-University
variable logistic regression analysis showed that forceps use Hospital. The study was approved by the institutional
was associated with an increase in major perineal and
review board. The deliveries were performed by a senior
vaginal tears (odds ratio [OR] 1.85; 95% confidence inter-
resident under the supervision of an attending physician.
val [CI] 1.27, 2.69; P ⴝ .001), an increase in instrument
marks and bruising (OR 4.63; 95% CI 2.90, 7.41; P < .001)
Maternal demographics recorded included age, parity,
and a decrease in cephalohematomas (OR 0.49; 95% CI and gestational age. Delivery characteristics recorded
0.29, 0.83; P ⴝ .007) compared with the vacuum. included indication for the use of an instrument, use of
oxytocin, fetal position and station, change in instru-
CONCLUSIONS: Maternal injuries are more common with
the use of forceps. Neonates delivered with forceps have
ment, delivery by cesarean, and type of anesthesia. Ma-
more facial injuries, whereas neonates delivered with vac- ternal outcomes of interest were the use of episiotomy,
uum have more cephalohematomas. (Obstet Gynecol lacerations sustained, and presence of vulvovaginal he-
2004;103:513– 8. © 2004 by The American College of matomas. The delivery information was entered in the
Obstetricians and Gynecologists.) medical record by the physician performing the delivery.
LEVEL OF EVIDENCE: II-3 Neonatal outcomes of interest were birth weight, Ap-
gar scores, neonatal intensive care unit (NICU) admis-
From the Department of Obstetrics and Gynecology, Winthrop-University Hospital, sions, cephalohematomas, instrument marks and bruis-
Mineola, New York, and the State University of New York at Stony Brook. ing, and caput and molding. The pediatricians who
.001) was independently associated with an increase in were seen in the vacuum-assisted group (2.4% versus 0;
major perineal and vaginal tears. In addition, a birth P ⫽ .04) than in the forceps group. More instrument
weight of 4,000 g or more (OR 2.09; 95% CI 1.06, 4.10; marks and bruising (37.6% versus 10.3%; P ⬍ .001) were
P ⫽ .03) and use of episiotomy (OR 2.22; 95% CI 1.22, found in the neonates delivered by forceps. However,
4.05; P ⫽ .01) were independently associated with major there was a greater incidence of cephalohematomas
tears (Table 4). When neonatal outcomes were evalu- (21.4% versus 13.8%; P ⬍ .04) and caput and molding
ated, the use of forceps was associated with an increase in (33.7% versus 16%; P ⬍ .001) in the neonates who were
instrument marks and bruising (OR 4.63; 95% CI 2.90, delivered by the assistance of the vacuum.
7.41; P ⬍ .001; Table 5) and a decrease in cephalohema- There were 151 nulliparas in the forceps group and
tomas (OR 0.49; 95% CI 0.29, 0.83; P ⫽ .007; Table 6). 211 nulliparas in the vacuum group. Forceps were used
One hundred eighty-one women in the forceps group more often than vacuum for prolonged second stage of
and 252 women in the vacuum group had an episiotomy labor (15.9% versus 6.6%; P ⫽ .008). There was a higher
performed. This subgroup of women was analyzed for rate of epidural (95.4% versus 86.7%; P ⫽ .01) and
injuries. We found a greater incidence of fourth-degree pudendal (12.6% versus 1.4%; P ⬍ .001) anesthesia in
perineal lacerations (12.2% versus 4.8%; P ⫽ .005) and the forceps group than in the vacuum-assisted group.
vaginal lacerations (18.8% versus 9.5%; P ⫽ .005) with The difference in episiotomy use (92.7% versus 89.6%;
the use of forceps. There were no significant differences P ⫽ .31) and in the number of women without injury
in the incidence of third-degree perineal lacerations be- (42.5% versus 37.9%; P ⫽ .28) was not significant be-
tween the 2 groups (33.7% forceps versus 27.4% vac- tween the forceps and vacuum groups. We found a
uum; P ⫽ .16). However, more periurethral lacerations greater incidence of fourth-degree perineal lacerations
VOL. 103, NO. 3, MARCH 2004 Johnson et al Forceps and Vacuum Deliveries 515
Table 4. The Effect of Forceps Use on Major Perineal and Table 6. The Effect of Forceps Use on Neonatal Cephalo-
Vaginal Lacerations After Adjusting for Other Vari- hematomas After Adjusting for Other Variables on
ables on Multivariable Logistic Regression Analysis Multivariable Logistic Regression Analysis
Odds 95% confidence Odds 95% confidence
ratio interval P ratio interval P
Parity 0.80 0.52, 1.24 .32 Parity 0.59 0.33, 1.06 .08
Birth weight ⱖ 4,000 g 2.09 1.06, 4.10 .03 Birth weight ⱖ 4,000 g 1.26 0.55, 2.89 .58
Episiotomy 2.22 1.22, 4.05 .01 Episiotomy 1.53 0.71, 3.30 .28
Forceps 1.85 1.27, 2.70 .001 Forceps 0.49 0.29, 0.83 .007
Prolonged 2nd stage 1.80 0.94, 3.47 .08 Prolonged 2nd stage 1.44 0.67, 3.13 .35
Epidural 1.69 0.95, 3.01 .08 Epidural 0.89 0.45, 1.74 .73
VOL. 103, NO. 3, MARCH 2004 Johnson et al Forceps and Vacuum Deliveries 517
Maternal and neonatal morbidity in instrumental deliver- 14. Gardella C, Taylor M, Benedetti T, Hitti J, Critchlow C.
ies with the Kobayashi vacuum extractor and low forceps. The effect of sequential use of vacuum and forceps for
Acta Obstet Gynecol Scand 1987;66:643–7. assisted vaginal delivery on neonatal and maternal out-
11. Johanson RB, Rice C, Doyle M, Arthur J, Anyanwu L, comes. Am J Obstet Gynecol 2001;185:896 –902.
Ibrahim J, et al. A randomised prospective study compar- 15. Learman L. Regional differences in operative obstetrics: a
ing the new vacuum extractor policy with forceps delivery. look to the south. Obstet Gynecol 1998;92:514 –9.
Br J Obstet Gyneaecol 1993;100:524 –30.
12. Ecker JL, Tan WM, Bansal RK, Bishop JT, Kilpatrick SJ. Address reprint requests to: Dr. Reinaldo Figueroa, Depart-
Is there a benefit to episiotomy at operative vaginal deliv- ment of Obstetrics and Gynecology, Winthrop-University
ery? Observations over ten years in a stable population. Hospital, 259 First Street, Mineola, NY 11501; e-mail:
Am J Obstet Gynecol 1997;176:411– 4. rfigueroa@winthrop.org.
13. Leighton BL, Halpern SH. Epidural analgesia: effects on
labor progress and maternal and neonatal outcome Received November 24, 2003. Received in revised form November 25,
[review]. Semin Perinatol 2002;26:122–35. 2003. Accepted December 4, 2003.