Robert B. Gherman and Bernard Gonik
Table Of Contents
Robert B. Gherman, MD
Bernard Gonik, MD
Shoulder dystocia is an obstetric complication that has been recognized and discussed for at least two centuries. Because of its infrequent occurrence and unpredictability, however, obstetricians have had little opportunity to become familiar with all aspects of this obstetric emergency. Potentially devastating maternal and neonatal complications can occur without rapid resolution of the dystocia.
Shoulder dystocia occurs when further delivery of the fetal body is prevented by impaction of the fetal shoulder behind the maternal symphysis pubis. Most authors define this obstetric emergency to include delivery requiring maneuvers in addition to gentle downward traction on the fetal head to effect delivery.1 A recent study further clarified this by defining shoulder dystocia as delivery with longer than 60 seconds between delivery of the fetal head and body, delivery requiring the use of ancillary obstetric maneuvers, or both.2,3
Shoulder dystocia occurs infrequently, with an incidence ranging from 0.2% to 2.1% of all vaginal deliveries.4,5,6 This wide range has been attributed to the inherent subjectivity of the clinician's definition of shoulder dystocia, the degree of reporting, and differences in defining the study population. The actual incidence, therefore, may be considerably higher (approaching 10%).2 When shoulder dystocia is objectively defined by head-to-body times of more than 60 seconds, only 25% of cases are subjectively defined by the practitioner.3
Shoulder dystocia results from a size discrepancy between the fetal shoulders and the pelvic inlet. In normal labor, after internal rotation, the biparietal diameter rests in a transverse position with the bisacromial diameter in an oblique angle. Extension and restitution result in the occiput returning to the anteroposterior plane. It is speculated that a persistent anteroposterior location of the fetal shoulders at the pelvic brim occurs when there is increased resistance between the fetal skin and vaginal walls (e.g., with macrosomia), with a large fetal chest relative to the biparietal diameter, and when truncal rotation does not occur (e.g. precipitous labor).7 When this occurs, the anterior shoulder impacts behind the symphysis pubis. Shoulder dystocia also may occur from impaction of the posterior fetal shoulder on the maternal sacral promontory.8,9,10
Shoulder dystocia is strongly correlated with the triad of fetal macrosomia, maternal diabetes, and maternal obesity. Macrosomia is arbitrarily defined as a birth weight greater than 4000 g. Large for gestational age is defined by an estimated fetal weight greater than the 90th percentile for the gestational age.11 Repeated studies show that infants weighing in excess of 4000 g are statistically at an increased risk for shoulder dystocia.12,13,14 The percentages of births complicated by shoulder dystocia for unassisted births not complicated by diabetes were 5.2% for infants weighing 4000 to 4250 g, 9.1% for those weighing 4250 to 4500 g, 14.3% for those weighing 4500 to 4750 g, and 21.1% for those weighing 4750 to 5000 g.15
Fetal body configuration may be more important than macrosomia per se because of a larger trunk and chest circumference as well as an increased bisacromial diameter.16 Macrosomic infants of diabetic mothers also are characterized by larger shoulder and extremity circumferences, decreased head-to-shoulder ratio, significantly higher percentage of body fat, and thicker upper extremity skin folds compared with nondiabetic control infants of similar birth weight and length.17 These factors do not allow the fetal shoulders to rotate from the anteroposterior position to a more desirable oblique position. As clinical evidence of this, women with class A1 gestational diabetes experience up to a 3% rate of shoulder dystocia.18,19
A higher rate of macrosomia occurs in pregnancies complicated by diabetes mellitus. In those with birth weights more than 4000 g, 12.8% of diabetic patients showed evidence of shoulder dystocia.20 The risk of shoulder dystocia in unassisted births to diabetic mothers has been found to be 8.4%, 12.3%, 19.9%, and 23.5% when the birth weight is 4000 to 4250 g, 4250 to 4500 g, 4500 to 4750 g, or more than 4750 g, respectively. When infants of diabetic mothers are delivered by vacuum extraction or forceps, these rates climb to 12.2%, 16.7%, 27.3%, and 34.8%.15
Other previously described antepartum risk factors include postdate gestation, male gender of fetus, advanced maternal age, short maternal stature (less than 1.5 m [4 feet, 10 inches]), platypelloid pelvis, multiparity, excessive weight gain during pregnancy (more than 16 kg [35 lb]), previous history of a macrosomic infant, and prior shoulder dystocia. Intrapartum risk factors also include a protracted first stage of labor, prolonged deceleration phase (between 8 and 10 cm), and epidural anesthesia. There is a 28-fold increase in shoulder dystocia when a midpelvic operative vaginal delivery is performed after a prolonged second stage.21 Among 637 patients participating in a randomized trial of operative vaginal delivery, the rate of shoulder dystocia was 3.3%. Nearly 75% of cases of shoulder dystocia were associated with the use of the vacuum extractor.22
Among infants weighing 3500 to 3999 g, Acker and colleagues noted a twofold increase in the incidence of shoulder dystocia among gravidas who experienced either protraction or arrest disorders of labor.12 The shoulder dystocia rate was increased another twofold if delivery ultimately occurred using low forceps. Two recent studies suggest that labor abnormalities may not be clinical predictors for subsequent development of shoulder dystocia.4,5
Preconceptive and prenatal risk factors do not allow the obstetrician to accurately and reliably predict shoulder dystocia. Only 25% of the cases of shoulder dystocia described by Lewis and colleagues23 had at least one significant risk factor. In this study, fundal height measurements had a positive predictive value of only 7.8%. Only 32% of patients were obese (more than 198 lb [90 kg]), 25% had excessive weight gain (more than 44 lb [20 kg]), 8% had short stature (less than 1.5 m [60 inches]), 6% were at more than 42 weeks' gestation, 3% were of advanced maternal age, and 2% had a history of diabetes.23
Although macrosomia clearly is a risk factor for shoulder dystocia, 50% to 60% of cases of shoulder dystocia occur in infants who weigh less than 4000 g.19 Among the 2924 infants identified in the macrosomic cohort of Kolderup and colleagues,24 only 48 (1.6%) injuries related to shoulder dystocia were noted. Among the 22 brachial plexus injuries with documented follow-up, only 5 (17%) were clinically evident at 6 months.24 In the series by Berard and colleagues of 87 macrosomic infants delivered vaginally, there were only 5 cases of Erb's palsy. By 3 months of age, all affected infants were without evidence of brachial plexus palsy.25 All of the 157 vaginally delivered infants with birth weights more than 4500 g described by Lipscomb and associates had no permanent sequelae by 2 months of age.26
Efforts to accurately diagnose fetal macrosomia during the antenatal or intrapartum period have met with poor results. When performed, ultrasound estimation of fetal weight is associated with a 10% to 15% margin of error27 and low positive predictive value for brachial plexus injury. Reported ultrasonographic predictors of shoulder dystocia also have been associated with low sensitivity and specificity. These have included a chest-to-head difference of 1.4 cm,28 shoulder-to-head difference of 4.8 cm,16 and abdominal-to-biparietal diameter difference of 2.6 cm among infants of diabetic mothers.29 Although an ultrasonographically derived fetal abdominal circumference of more than 35 cm identified 93% of macrosomic infants, only 13% of these actually had shoulder dystocia.30
Multiple studies demonstrate that induction of labor for suspected macrosomia (compared with expectant management) does not decrease the incidence of shoulder dystocia or brachial plexus injury and only results in an increased rate of cesarean delivery.31,32,33 A recent meta-analysis determined that 2345 and 443 cesarean sections in nondiabetic and diabetic gravidas, respectively, would be necessary to prevent one permanent brachial plexus injury.34 Finally, patients with a prior history of shoulder dystocia have recurrence risks of 9.8% to 13.8%.35,36
The first step in management is anticipation of shoulder dystocia, including recognition of the previously described antepartum and intrapartum risk factors. Every effort should be made to prepare a rapid, ingrained, and well-coordinated stepwise plan. When fetal macrosomia is suspected, the patient should be thoroughly counseled regarding the risks and benefits of a trial of vaginal delivery.37
At the time of delivery, if shoulder dystocia is a concern, some clinicians have empirically advocated immediately proceeding to delivery of the fetal shoulders to maintain the forward momentum of the fetus. Others support a short delay in delivery of the shoulders, arguing that the endogenous rotational mechanics of the second stage may spontaneously alleviate the obstruction.
Shoulder dystocia usually is heralded by the classic “turtle sign.” After the fetal head delivers, it retracts back onto the maternal perineum. As maternal efforts to expel the fetal body continue, the fetal shoulder becomes further impacted behind the symphysis pubis. If the physician is alone, immediate assistance should be recruited. Additional support from anesthesia, nursing, and pediatrics also should be recruited. The mother should be instructed to stop pushing while attempts are made to dislodge the impacted shoulder. Most cases of shoulder dystocia resolve within a few minutes, although it may seem longer.
The McRoberts maneuver is recommended as the initial technique for disimpaction of the anterior shoulder. In a retrospective review of 236 cases of shoulder dystocia occurring between 1991 and 1994 at Los Angeles County-University of Southern California Medical Center, this maneuver alone alleviated 42% of cases.38 When combined with suprapubic pressure, the McRoberts maneuver results in resolution of 58% of cases.39 Trends toward lower rates of maternal and neonatal morbidity have been associated with the McRoberts maneuver.38 Objective testing also has shown that the McRoberts maneuver may reduce fetal shoulder extraction forces and brachial plexus stretching.40
The McRoberts maneuver involves flexion of the mother's legs onto her abdomen. An x-ray study of the McRoberts maneuver confirms that it is associated with an increase in the mean angle of inclination between the symphysis pubis and the sacral promontory. There also was a 24% decrease in the angle created by drawing a line bisecting the symphysis pubis relative to the horizontal. With the McRoberts maneuver, the angle created by a line bisecting the longitudinal axis of the fifth lumbar vertebra and the longitudinal axis of the upper sacrum also increased.41
The McRoberts maneuver does not change the actual dimensions of the maternal pelvis; instead, it presumably works by straightening the maternal sacrum relative to the lumbar spine, with consequent cephalic rotation of the symphysis pubis. This position is thought to also enhance passage of the posterior fetal shoulder over the sacrum and through the pelvic inlet, positioning the plane of the pelvic inlet to its maximal dimension perpendicular to the maximum maternal expulsive force. Limitations of the technique include the need for two assistants and the extra time required to elevate and flex the patient's legs. Potential difficulty also can be encountered in moving the very obese patient or the patient with a dense epidural motor blockade.
The need for additional maneuvers after the McRoberts maneuver has been associated with larger fetal birth weights, longer active phases, and longer second stages of labor.38 Simultaneous suprapubic pressure, applied either posteriorly or laterally, can displace the impacted shoulder into the oblique diameter and effect delivery. Because shoulder dystocia is considered to be a “bony dystocia,” cutting a proctoepisiotomy may not be necessary. If the McRoberts maneuver fails, however, enlarging the episiotomy may allow more room for the next proposed maneuver.
In the Woods (corkscrew) maneuver (Fig. 1), the practitioner pushes the posterior shoulder through a clockwise 180-degree arc by applying pressure on the anterior surface of the posterior shoulder. In the reverse Woods maneuver, or Rubin maneuver (Fig. 2), pressure is applied to the posterior surface of the anterior shoulder to effect counterclockwise rotation of the posterior shoulder.
Should these maneuvers fail, the physician's hand can be passed into the vagina following the posterior arm to the elbow. After pressure is applied at the antecubital fossa to flex the fetal forearm, the arm is swept out over the infant's chest and delivered over the perineum (Fig. 3). Rotation of the fetal trunk to bring the posterior arm anteriorly may be required.
Zavanelli Maneuver, Symphysiotomy, and Hysterotomy
Before attempting the Zavanelli maneuver or a symphysiotomy, all of the previously described maneuvers it would be reasonable to reattempt with the patient under anesthesia. In the Zavanelli maneuver, the fetal head is rotated back to a direct occiput anterior position and then flexed. Constant firm pressure is used to push the head back into the vagina; a cesarean section subsequently is performed. Tocolytic agents or uterine-relaxing general anesthesia may be administered in preparation for and during the maneuver. Among 92 reported cases of partially born fetuses in vertex presentation, the Zavanelli maneuver successfully returned 78 fetuses into the vagina.42 Severe maternal complications of the Zavanelli maneuver include uterine infection requiring hysterectomy, vaginal “rupture,” laceration of the lower uterine segment, and uterine rupture.42
To perform a symphysiotomy, the patient should be placed in an exaggerated lithotomy position and have a Foley catheter placed to identify the urethra. With the physician's index and middle finger displacing the urethra laterally, the cephalad portion of the symphysis is incised with a scalpel blade. Because of the significant maternal morbidity associated with this procedure, it represents a last attempt to preserve fetal life and should be initiated within 5 to 6 minutes of delivery of the fetal head. Among three recently described cases, two patients had significant lower urinary tract complications and also required blood transfusion.43
For catastrophic cases unresponsive to the traditional maneuvers, performance of hysterotomy may be used to either primarily resolve the shoulder dystocia or assist with vaginal techniques.44 O'Shaughnessy describes delivery of the posterior fetal arm through a transverse uterine incision with subsequent passage of the hand through the vagina to an assistant. Posterior arm delivery then was completed vaginally while the abdominal surgeon applied pressure on the anterior fetal shoulder to allow rotation to the oblique pelvic diameter.45
Initially described in 1976 by Gaskin, the “all-fours” maneuver consists of placing the gravid patient onto her hands and knees.46 In their analysis of 82 consecutive cases of shoulder dystocia, Bruner and coworkers noted that 68 patients (83%) had successful delivery of the fetus with this maneuver alone.47 Rates of maternal and neonatal morbidity were 1.2% and 4.9%, respectively, with a single case of maternal postpartum hemorrhage, one infant with a fractured humerus, and three neonates with low Apgar scores. The average time needed to assume the all-fours position and complete delivery was 2 to 3 minutes. The downward force of gravity or a favorable change in pelvic diameter produced by this maneuver may be the mechanisms that allow for disimpaction of the fetal shoulder.47
Fracture of the clavicle may be attempted by applying direct pressure away from the fetal lung. In reality, however, this is difficult to accomplish in the setting of shoulder dystocia.
The use of fundal pressure to alleviate the shoulder dystocia should be avoided because it only further impacts the anterior shoulder behind the symphysis pubis. The case-control study of Phelan and associates compared 59 infants with documented Erb's palsy whose births were complicated by shoulder dystocia with 59 cases of shoulder dystocia in which the infants had no evidence of brachial plexus impairment. The incidence of fundal pressure was significantly higher for cases than for control subjects (32% versus 2%; odds ratio = 27.5).48
Maternal complications of shoulder dystocia include postpartum hemorrhage, cervicovaginal lacerations, fourth-degree lacerations of the rectum, bladder atony, and uterine rupture. Symphyseal separation and maternal femoral neuropathy have been associated with overly aggressive hyperflexion of the maternal legs.49,50 A recent large retrospective study evaluating 285 cases of shoulder dystocia reported an overall fetal injury rate (nerve or bone injury) of 25%.51
Approximately 80% of brachial plexus injuries involve the nerve roots of C5–6 (Erb-Duchenne palsy). Over 90% of these injuries resolve by 1 year of neonatal life, with only an approximate 5% to 8% rate of persistent nerve injury. Although occurring less often, only 40% of the injuries involving the C8-T1 nerve roots (Klumpke's palsy) can be expected to resolve by 1 year of age. Brachial plexus injury occurs in approximately 11.8% to 16.8% of shoulder dystocia cases; its occurrence appears to have no relation to the number and type of maneuvers used to disimpact the shoulder.8,51,52,53 Even with use of the McRoberts' maneuver alone, 11.6% of infants sustained brachial plexus injury.38 Brachial plexus injury is more common in diabetic gravidas (10.5/1000 deliveries) than in nondiabetic gravidas (0.56/1000 deliveries).54 Compressive forces resulting from the application of fundal pressure have been associated with thoracic spinal cord injury, manifesting as lower extremity motor dysfunction, overflow urinary incontinence, and rectal incontinence in the newborn.55
Clavicular and humeral fracture may also occur, although these usually resolve without any long-term sequelae. Although the fetal pH declines at a rate of 0.04 U/min between delivery of the head and trunk, death due to shoulder dystocia is exceedingly rare.54
Risks to the physician mainly involve litigation; brachial plexus injury accounts for a large proportion of shoulder dystocia-related lawsuits. For this reason, it is important to clearly document the following facts57:
When and how dystocia was diagnosed
Shoulder dystocia is an infrequent obstetric emergency that often occurs without warning and can lead to serious neonatal and maternal morbidity. When the risk factors of fetal macrosomia, maternal diabetes, postdatism, maternal obesity, previous history of macrosomia and shoulder dystocia, second-stage abnormalities, and operative vaginal delivery are present, an even higher index of suspicion must be present. Proper management of this disorder should be aimed toward minimizing fetal and maternal morbidity through an organized and expeditious clinical approach. All obstetricians should frequently review the management of shoulder dystocia.
2. Spong CY, Beall M, Rodrigues D, Ross MG: An objective definition of shoulder dystocia: Prolonged head-to-body delivery intervals and/or the use of ancillary obstetric maneuvers. Obstet Gynecol 86: 433, 1995