The Journal of Maternal-Fetal Medicine 6:220–224 (1997)

Shoulder Dystocia and Operative Vaginal Delivery

James A. Bofill, MD,1* Orion A. Rust, MD,1 Meenakshi Devidas, PhD,2
William E. Roberts, MD,1 John C. Morrison, MD,1 and James N. Martin, Jr., MD1
1Departmentof Obstetrics and Gynecology, University of Mississippi Medical Center,
Jackson, Mississippi
2Department of Preventive Medicine, University of Mississippi Medical Center,

Jackson, Mississippi

Abstract Our objective was to determine the factors involved in the development of shoulder dystocia in
association with operative vaginal delivery. In this prospective study, patients who were candidates for operative
vaginal delivery were randomized either to forceps (N 5 315) or vacuum with M-cup (N 5 322) and timed from
initial placement of instrument to final delivery. Data were gathered prior to and after instrumental delivery.
Statistics used were Pearson chi square, Fisher’s exact, analysis of variance, and multiple logistic regression. There
were a total of 21 patients with shoulder dystocia in both groups (3.3% incidence). Discriminant factors that did
not meet significance included use of epidural analgesia (P 5 .12), station (P 5 .99), previous vaginal delivery
(P 5 .99), fetal gender (P 5 .54), indication for operative vaginal delivery (P 5 .63), .45 degree rotation
(P 5 .68), use of episiotomy (P 5 .62), maternal weight (P 5 .26), and maternal diabetes (P 5 .08). Nearly
attaining significance in univariate analysis was randomization to vacuum (P 5 .052). Significant factors
included gestational age (P 5 .03), time required to effect delivery (P 5 .007), and birthweight (P 5 .0001).
When these factors were subjected to stepwise multiple logistic regression, three factors remained as significant
associations with shoulder dystocia: randomization to vacuum (P 5 .04), time for delivery (P 5 .03), and
birthweight (P 5 .0001). In this operative vaginal delivery trial, shoulder dystocia was strongly associated with
large fetal size, longer time to delivery, and the use of vacuum for delivery. J. Matern.–Fetal Med. 6:220–224,
1997. r 1997 Wiley-Liss, Inc.

Key Words: forceps; vacuum; shoulder dystocia; time

INTRODUCTION Our group recently completed a large randomized efficacy

Shoulder girdle dystocia at vaginal delivery is an uncom-
mon event that may have catastrophic sequelae for the
infant. This complication of delivery is not predictable or
preventable [1,2]. Large descriptive series have noted rates
of shoulder dystocia from 0.15 to 2.1% [3–5]. These studies
have retrospectively noted some clinical findings to be
linked with the diagnosis of shoulder dystocia. These
associations include maternal diabetes, prolonged or ar-
rested labor, midpelvic instrumental delivery, and birth-
weight [3,6].
Little is known, however, about the relationship of
shoulder girdle dystocia and operative vaginal delivery. The
reports that have linked midpelvic instrumental delivery
with shoulder dystocia were written prior to the 1988
ACOG definitions of outlet, low, and midpelvic procedures
[7]. Likewise, these collections of shoulder dystocia cases are
retrospective in nature and are dependent for their data
collection on written delivery notes which may or may not
be entirely complete or accurate.
trial comparing obstetric forceps to the vacuum extractor
[8]. The results of the efficacy trial showed no difference in
the rate of successful delivery when obstetric forceps were
compared with the M-cup vacuum extractor (Neward
Enterprises, Rancho Cucamonga, CA [8]. The facts surround-
ing these deliveries were prospectively documented as
opposed to the erratic data collection that may obfuscate
the findings in retrospective chart reviews. In this report we
use these carefully collected data to compare the patients
who sustained shoulder dystocia with those patients who did
not suffer this complication at delivery. The purpose of our
study was to determine any factors associated with shoulder
dystocia in the setting of operative vaginal delivery.
*Correspondence to: James A. Bofill, M.D., c/o Donna White/Gail
Head, Ob-Gyn Publication Office, Department of Obstetrics and Gyne-
cology, University of Mississippi Medical Center, 2500 North State Street,
Jackson, MS 39216-4505.
Received 17 June 1996; Revised 21 November 1996; Accepted 22
January 1997

r 1997 Wiley-Liss, Inc.

 ASSISTED DELIVERY AND SHOULDER DYSTOCIA
SUBJECTS AND METHODS
Over a 10-month period (October 1994 through July
1995), women who were candidates for operative vaginal
delivery were randomized to be delivered either by obstetric
forceps or by one of two vacuum extraction techniques
(continuous or intermittent suction). For this study, the
vacuum arms were combined into one group because there
were no differences is maternal or fetal outcome or in the
respective rates of shoulder dystocia [9]. In this trial all of
the data were collected and analyzed on an ‘‘intent to treat’’
basis, and patients remained in their randomized groups for
all statistical purposes. The M-cup (Neward Enterprises,
Rancho Cucamonga, CA) was the only vacuum cup used in
this study, but the physicians were free to choose the type of
forceps for the patients who randomized to that instrument.
Only women with pregnancies $34 weeks or with fetuses
expected to weigh $1,800, if gestational dating was uncer-
tain, were considered for this trial. These are the lower
limits that have traditionally been used for vacuum extrac-
tion at our institution. The delivery physicians were di-
rected to perform episiotomy selectively, not electively.
Patients gave consent for this trial only after the decision
had been made that they were candidates for operative
vaginal delivery. These patients signed an informed consent
document which was approved by the Institutional Review
Board of the University of Mississippi. After the consent
process, the next in a series of numbered, opaque envelopes
was opened which designated the patient to be delivered
either by forceps or by vacuum. Randomization envelopes
were prepared using a table of random numbers by an
uninvolved third party prior to the initiation of the study.
Randomization was usually accomplished while the patient
was being transported from the labor to the delivery room.
Data were gathered prior to, during, and after each
delivery. Maternal demographics, gestational age, the indi-
cation for operative delivery, and whether the parturient
had ever delivered vaginally were prospectively recorded.
Data gathered in the delivery room but prior to the
application of the delivery instrument included the station
and position of the fetal head. The type of analgesia and the
requirement for episiotomy were also recorded. Patients
underwent routine urinary bladder catheterization prior to
the application of the randomized delivery instrument.
Deliveries were timed with stopwatches from the insertion
of the instrument into the vagina (vacuum cup or first blade
of the forceps) until the complete delivery of the infant.
Instrumental rotation, if required, was likewise recorded.
These deliveries were accomplished by obstetric house
officers under the supervision of maternal-fetal medicine
fellows with extensive experience in both types of operative
vaginal delivery.
Neonatal data gathered after delivery included Apgar
scores and cord arterial blood gas specimens. For the
221
purposes of our study, shoulder dystocia was defined as the
requirement for ancillary maneuvers (e.g., McRobert’s posi-
tion, suprapubic pressure, delivery of posterior arm, etc.) for
the delivery of the fetal shoulders. Statistics used were the
Pearson chi square, Fisher’s Exact, analysis of variance
(ANOVA), and multiple stepwise logistic regression (SAS
Cary, NC).
RESULTS
This operative vaginal delivery trial included 637 women:
315 women randomized to obstetric forceps and 322 to
vacuum. Thirty-five of the women in the forceps group were
delivered by other means (12 spontaneously, 16 by vacuum,
and seven by cesarean). Twenty-one women who random-
ized to the vacuum group were delivered in another manner
(five spontaneously, 11 by forceps, and five by cesarean).
Overall, there was no difference in the efficacy rates of these
two instruments [8]. The indications for these operative
vaginal deliveries are described in another report [8].
There were 21 cases of shoulder dystocia (the study
group) recorded for an incidence of 3.3% (21/637). In 20
cases McRobert’s position and suprapubic pressure sufficed
to relieve the impaction of the anterior shoulder behind the
symphysis. In one case delivery of the posterior arm was
required to relieve the obstruction. There were nine cases in
which shoulder dystocia was diagnosed after an episiotomy
had been performed, and in one of these cases the episi-
otomy was extended to create space. In ten other cases an
episiotomy was not performed before or after the diagnosis
of shoulder dystocia. In two instances an episiotomy was
performed only after the diagnosis of shoulder dystocia was
established. For statistical purposes the two cases in which
episiotomy was performed after the diagnosis of shoulder
dystocia were considered to be in the ‘‘no episiotomy’’
group.
There were 15 cases of shoulder dystocia in the women
that randomized to the vacuum (incidence 4.7%), while
only six of the deliveries on the forceps arm had this
complication (incidence 1.9%) (P 5 .052). There was one
case of deep transverse arrest which randomized to the
forceps and was abandoned secondary to a poor forceps
application. The delivery physician completed a vacuum
delivery of a 4,300-g fetus which resulted in shoulder
dystocia and a clavicular fracture. For statistical purposes
this case remained in the forceps group. Table 1 demon-
strates that several factors were not significantly correlated
with shoulder dystocia in univariate analysis. These in-
cluded maternal demographic factors such as age, weight,
previous vaginal delivery, or maternal diabetes. Intrapartum
factors not associated with shoulder dystocia included
epidural analgesia, pelvic station, indication for operative
intervention, .45 degree rotational maneuvers, and use of
episiotomy.
222 BOFILL ET AL.

TABLE 1. Factors Without Associationa

Shoulder dystocia
No Yes p
Epidural
No 340 8 .122
Yes 276 13
Station
Mid 30 1 .999
Low 295 10
Outlet 291 10
Previous vaginal
No 470 16 1.000
Yes 146 5
Indication
Maternal 176 7 .638
Fetal distress 120 5
Other 126 3 Fig. 1. Percentage of operative vaginal deliveries with shoulder dysto-
Elective 119 2 cia by birthweight.
Prolonged second
stage 75 4
Rotation .45
No 566 19 .686
Yes 50 2
Episiotomy
No 319 12 .629
Yes 297 9
Maternal diabetes
No 569 17 .08
Yes 47 4
Maternal age (years) 20.86 (65.02) 20.00 (64.7) .439
Maternal weight (lbs) 176.6 (638.7) 186.2 (642.0) .264
aData are presented as ‘‘N’’ or as mean 6 SD.

TABLE 2. Maternal/Fetal Factorsa

Shoulder dystocia
No Yes p Fig. 2. Percentage of operative vaginal deliveries with shoulder dysto-
cia by the time required to complete the delivery.
Total time (second) 200 (6191) 323 (6264) .007
Birthweight (g) 3098 (6464) 3760 (6584) .0001
Gestational age (weeks) 38.95 (61.8) 39.84 (61.4) .026
92% of vacuum deliveries, an insignificant difference [8].
Apgar at 1 minute 8.65 (6.89) 8.09 (61.26) .006
Apgar at 5 minutes 8.92 (6.36) 8.90 (6.30) .828
Time as a variable was available for 19 of the 21 cases of
Cord arterial pH 7.26 (6.07) 7.24 (6.07) .325 shoulder dystocia. Overall, time was recorded in 90% (576
of 637) of these deliveries [8]. Cases of shoulder dystocia
aData are presented as mean 6 SD.
were most common when the operative delivery time
exceeded 6 minutes (Fig. 2).
Table 2 shows the maternal factors that were associated In the shoulder dystocia group two infants sustained
with shoulder dystocia in univariate analysis. The gesta- transient Erb’s palsy and one infant was noted to have a
tional age of the pregnancy, the birthweight of the fetus, and fractured clavicle. One infant who was delivered without
the time required to achieve delivery were significantly the diagnosis of shoulder dystocia also sustained a clavicular
different in the mother-infant pairs who sustained shoulder fracture. For the study infants the Apgar score at 1 minute
dystocia compared with those who did not. Only 38% was significantly lower, but by 5 minutes their Apgar score
(8/21) of the infants in shoulder dystocia cases weighed was equivalent to that of their control counterparts (Table
.4,000 g. Figure 1 demonstrates the frequency of shoulder 2). There were no apparent differences in cord arterial pH
dystocia by birthweight categories. The time required for values in the study and the control infants. The lowest cord
delivery was recorded in 89% of forceps deliveries and in arterial pH in the shoulder dystocia group was 7.09.
 ASSISTED DELIVERY AND SHOULDER DYSTOCIA
The variables that appeared to be associated with shoul-
der dystocia were then subjected to multiple logistic regres-
sion analysis. Interestingly, vacuum delivery achieved signifi-
cance in the logistic regression (P 5 .045), whereas it was
only of borderline importance in simple univariate analysis.
The variable of gestational age was forced out of the model
by birthweight (P 5 .0001), whereas the time required to
achieve an operative vaginal delivery retained its signifi-
cance (P 5 .027).
DISCUSSION
The diagnosis of shoulder dystocia is subject to the
discretion of the delivery physician, and it is possible that
this diagnosis is underreported secondary to medicolegal
concerns. Indeed, Gonik et al. noted that only 29% of
infants who were delivered with injuries usually associated
with the diagnosis of shoulder dystocia had a maternal chart
with that designation [5]. Neonatal brachial plexus impair-
ment is the condition most often considered to be related to
shoulder dystocia, and it is second only to asphyxia with
regard to long-term morbidity. However, cases of neonatal
brachial plexus impairment have been documented in
which there was no evidence of shoulder dystocia suggesting
that some cases may not result from events at delivery [10].
Indeed, cases of neonatal brachial plexus impairment have
been reported in which the infants were delivered by
cesarean section without instrumental or any other type of
traction in an attempt to effect vaginal delivery [1,10]. It is
likely that brachial plexus impairment may occur without
shoulder dystocia, without extreme lateral traction imposed
on the fetal head, and with the best of intrapartum care [10].
It is clear that the diagnosis of shoulder dystocia should be
standardized and made less subjective. Spong and colleagues
have demonstrated that only 16 of 27 patients in whom
ancillary maneuvers were used for delivery had shoulder
dystocia listed as a clinical diagnosis [11]. Going even
further, this group measured the head-to-body delivery time
intervals and noted that the upper limit of normal (mean 6 2
standard deviations) was 60 seconds [11]. In their study, if
shoulder dystocia was a combined diagnosis of ancillary
maneuvers and/or prolonged head-to-body delivery time
then a group of infants was identified that had higher
birthweights and lower 1-minute Apgar scores than control
infants [11]. Our data seem to confirm the findings of Spong
et al [11]. First, although we measured time from the
placement of the instrument into the vagina until the
delivery was completed, it is clear that those operative
vaginal deliveries with shoulder dystocia took significantly
longer to complete. We cannot be certain that the extra
time was entirely due to a prolonged head-to-body interval.
Second, the infants with shoulder dystocia in our study had
significantly lower 1-minute Apgar scores and higher weights
than the controls.
223
Our prospective study does have limitations. Even though
this is the largest randomized forceps versus vacuum trial
reported thus far, the number of patients included cannot
compete with the larger populations described in the
retrospective analyses. A small sample size is a hindrance
when attempting to study rare events such as shoulder
dystocia. The finding of a trend with regard to maternal
diabetes and shoulder dystocia (P 5 .08) may well be a Type
II statistical error. There was no evidence that the pelvic
station from which the operative vaginal delivery was
initiated was a significant risk factor for shoulder dystocia.
Only one delivery initiated from the midpelvis resulted in a
shoulder dystocia. This probably represents a careful selec-
tion process for attempted midpelvic deliveries. When fetal
macrosomia is a concern and there is second-stage midpel-
vic arrest the most likely recourse is cesarean section
without the benefit of a trial of operative vaginal delivery.
Antenatal prediction of birthweight, whether deter-
mined clinically or by ultrasonography, is notoriously impre-
cise. Importantly, the time required for delivery is unlikely
to be measured in most centers when forceps are utilized. It
should be stressed that of the three factors we identified as
being associated with shoulder dystocia only the use of
vacuum is known prior to the delivery. This certainly adds
credence to the assertion that shoulder dystocia, even when
associated with operative vaginal delivery, is unpredictable
and unpreventable.
This is not the first study to suggest that use of the
vacuum extractor is more likely to cause shoulder dystocia.
In a 1978 study, Benedetti and Gabbe [6] noted a trend
toward a higher incidence of shoulder dystocia with vacuum
extraction as opposed to Kielland forceps. They reasoned
that the vacuum extractor may be able to deliver a larger
infant than the Kielland forceps. Obstetric forceps are
instruments that laterally cradle the head of the infant and
thereby add to the volume of material that must negotiate
the pelvis and introitus. It is certainly possible that if a
particular infant is only marginally suited to the size of the
pelvis the extra space taken up by the forceps blade may not
allow for delivery of the fetal head. The vacuum extractor,
which takes up no room in the pelvis, may allow for the
delivery of the infant who is only marginally suited for the
size and architecture of the maternal pelvis. We have not
restricted the use of the vacuum extractor in our labor and
delivery suite with regard to midpelvic applications, possible
fetal macrosomia, maternal diabetes, or in the situation of a
prolonged second stage of labor. We continue to adhere to a
strict time limit of 20 minutes for vacuum-assisted vaginal
delivery although we are now aware that those deliveries
that last longer than 6 minutes more often result in shoulder
dystocia. We do, however, prescribe common sense in the
application of any instrument for delivery when one or more
of the above clinical parameters is present.
224 BOFILL ET AL.
ACKNOWLEDGMENTS
This study was supported in part by the Vicksburg
Hospital Medical Foundation.
REFERENCES
1. Morrison JC, Sanders JR, Magann EF, Wiser WL: The diagnosis and
management of dystocia of the shoulder. Surg Gynecol Obstet
175:515–522, 1992.
2. Gross TL, Sokol RJ, Williams T, Thompson K: Shoulder dystocia: a
fetal-physician risk. Am J Obstet Gynecol 156:1408–1418, 1987.
3. Acker DB, Sachs BP, Friedman EA: Risk factors for shoulder dystocia.
Obstet Gynecol 66:762–768, 1985.
4. Gross SJ, Shime J, Farine D: Shoulder dystocia: predictor and outcome.
Am J Obstet Gynecol 156:334–336, 1987.
5. Gonik B, Hollyer VL, Allen R: Shoulder dystocia recognition:
differences in neonatal risks for injury. Am J Perinatol 8:31–34, 1991.
6. Benedetti TJ, Gabbe SG: Shoulder dystocia: a complication of fetal
macrosomia and prolonged second stage of labor with midpelvic
delivery. Obstet Gynecol 52:526–529, 1978.
7. American College of Obstetricians and Gynecologists: Operative
vaginal delivery. Washington, DC: ACOG Technical Bulletin, No.
196, 1994.
8. Bofill JA, Rust OA, Schorr SJ, Brown RC, Martin RW, Martin JN,
Morrison JC: A randomized prospective trial of the obstetric forceps
versus the M-cup. Am J Obstet Gynecol 175:1325–1330, 1996.
9. Bofill JA, Rust OA, Schorr SJ, Brown RC, Roberts WE, Morrison JC:
A randomized prospective trial of vacuum extraction technique.
Obstet Gynecol (in press).
10. Jennett RJ, Tarby TJ, Kreinick CJ: Brachial plexus palsy: an old
problem revisited. Am J Obstet Gynecol 166:1673–1677, 1992.
11. Spong CY, Beall M, Rodrigues D, Ross MG: An objective definition of
shoulder dystocia: prolonged head-to-body delivery intervals and/or
use of ancillary obstetric maneuvers. Obstet Gynecol 86:433–436,
1995.