Shoulder dystocia

1. What every clinician should know

Obstetricians have two major concerns when dealing with shoulder dystocia. The first is the technical aspects of attempting to predict who is at risk for shoulder dystocia, managing it when it occurs, and attempting to avoid the dreaded consequence of permanent brachial plexus injury to the neonate during its resolution. The second concern is the ever-present fear in the mind of every practicing obstetrician that if a baby is injured during a shoulder dystocia delivery, rightly or wrongly the obstetrician will be held to be at fault in the lawsuit that will almost certainly follow.

2. Diagnosis and differential diagnosis

Definition of shoulder dystocia

Shoulder dystocia occurs when there is an inability to deliver a baby’s shoulders after its head has emerged. In such cases, the baby’s shoulder has become impacted behind the mother’s pubic symphysis (Figure 1).

Figure 1.

Shoulder Dystocia

Continue Reading

The official American Congress of Obstetricians and Gynecologists (ACOG) definition of a shoulder dystocia delivery is one that requires additional obstetrical maneuvers following thefailure of gentle downward traction on the fetal head to effect delivery of the shoulders.

It is thought that when a neonate’s shoulders get stuck during a shoulder dystocia delivery, it is because instead of the shoulders descending into the wider oblique diameter of the maternal pelvic outlet, the shoulders instead descend in an anterior-posterior orientation. This diameter is narrower than the oblique diameter, increasing the chances that the baby’s anterior shoulder will get stuck behind the maternal symphysis. This prevents the normal sequence of prompt delivery of the shoulders following that of the head.

However, some authors have not found the ACOG definition sufficiently reproducible or quantifiable to be useful. Spong et al have defined shoulder dystocia as a prolonged head-to-body delivery time (e.g., more than 60 seconds) or the need for ancillary obstetrical maneuvers. In the Spong study, 60 seconds was found to be approximately two standard deviations above the mean value for head-to-body delivery time in uncomplicated shoulder dystocia deliveries.

Incidence of shoulder dystocia

The incidence of shoulder dystocia is 0.6% to 1.4% of all vaginal deliveries. This incidence has risen over the last decade, largely due to the almost epidemic national increase in maternal obesity, a major risk factor for fetal macrosomia. Yet, some of this observed increase is no doubt due to better reporting as awareness among obstetricians of the importance of proper documentation of shoulder dystocias has increased.

Risk factors for shoulder dystocia

While many factors have been cited as increasing the risk for shoulder dystocia, careful analysis shows that there are only four primary risk factors:

  • Macrosomia

  • Gestational diabetes

  • Previous shoulder dystocia

  • Instrumental vaginal delivery

All other supposed risk factors for shoulder dystocia turn out to merely be markers in one form or another of the above.


There is a direct linkage between birth weight and the risk of shoulder dystocia. This has been demonstrated repeatedly in innumerable studies over multiple decades. The data in Table I correlating birth weight with shoulder dystocia from a large university-affiliated obstetrical service are representative.

Table I.n

Relationship between birth weight and shoulder dystocia

Most other proposed risk factors for shoulder dystocia exert their influence because of their association with increased birth weight.

Despite this clear linkage between fetal macrosomia and shoulder dystocia, there are significant problems with clinicians attempting to use suspected fetal macrosomia to predict which patients will experience shoulder dystocia at delivery:

  • The ability to accurately predict fetal weight prior to delivery is notoriously poor. Best estimates are thought to be within 15% to 20% above or below the true fetal weight at delivery, and even wider differences are frequently noted. Some studies have shown that clinical estimates of fetal weight by palpation are more accurate than are ultrasound estimates.

  • Definitions of macrosomia are not consistent, either in clinical practice or in the obstetrical literature. For instance, the American College of Obstetricians and Gynecologists (ACOG) defines macrosomia as 4,500 g—but then goes on say that it is reasonable to “consider” cesarean section to avoid shoulder dystocia when the estimated fetal weight for a nondiabetic mother is ≥ 5,000 g, or for a diabetic mother an estimated fetal weight of ≥ 4,500 g. Moreover, many other textbooks and papers discuss 4,000 g and 4,500 g, respectively, as weights at which to “consider” recommending a cesarean section.

  • Approximately half of all shoulder dystocias occur with newborns weighing less than 4,000 g, below anybody’s definition of macrosomia.

  • Even if: (a) fetal weight could be estimated accurately, and (b) a consensus could be reached on a weight level at which it made sense to recommend cesarean section in order to avoid a shoulder dystocia, the sensitivity and specificity of predicting shoulder dystocia are so poor as to make such a policy unreasonable. Estimating fetal weight in order to attempt to prevent shoulder dystocias by doing elective cesarean sections would result in the performance of thousands of cesarean sections—with all their inherent risks—to prevent a single permanent brachial plexus injury.

Gestational diabetes

In babies of diabetic mothers, the risk of shoulder dystocia in any weight group is 2.5- to 6-fold higher than if the baby’s mother does not have diabetes. Additionally, the incidence and severity of neonatal injury from shoulder dystocia is higher in babies born of diabetic mothers ( Table II).

Table II.n

Relationship between birth weight and risk of shoulder dystocia in babies of diabetic and nondiabetic mothers

The reason for the increased risk of shoulder dystocia in these babies has to do with their different growth morphology compared with babies of nondiabetic mothers. Given equal weights, babies born of diabetic mothers have larger shoulders, chest circumference, and abdominal circumference than those born of nondiabetic mothers.

Although tight maternal blood sugar control during pregnancy does not eliminate the risk disparity between babies of mothers with gestational diabetes and those without, tight control has been shown to reduce the risk of fetal macrosomia, shoulder dystocia, and neonatal injury.

Previous shoulder dystocia

The rate of recurrence of shoulder dystocia is approximately 10%. Moreover, there is a higher rate of injury and severity of injury in newborns whose delivery represents a recurrent shoulder dystocia.

Yet, while many clinicians recommend cesarean section for delivery of a pregnant patient who has had a previous shoulder dystocia delivery, the effectiveness of this policy in reducing risk is not clear. The ACOG Practice Bulletin on Shoulder Dystocia (2002) says, “because most subsequent deliveries will not be complicated by shoulder dystocia, the benefit of universal elective cesarean delivery is questionable in patients who have such a history of shoulder dystocia.”

Instrumental vaginal delivery

Deliveries requiring mid vacuum or forceps result in rates of shoulder dystocia that are 4.6 to 28 times higher than do spontaneous vaginal deliveries. This ratio is even higher if both vacuum and forceps are used sequentially. However, there is little information in the published literature on the contribution that low vacuum and low forceps deliveries—especially outlet interventions—make to the incidence of shoulder dystocia.

Other purported risk factors for shoulder dystocia ( Table III)

Table III.n

Other purported risk factors for shoulder dystocia

The bottom line is this:

  • There are a limited number of proven risk factors for shoulder dystocia. Most “risk factors” exert their influence because of an association with increased birth weight.

  • Most women who experience shoulder dystocia have no combination of risk factors that allows clinically useful prediction prior to delivery.

Suggested clinical approach to "risk factors"

Patients who have true risk factors for shoulder dystocia—suspected macrosomia, gestational diabetes, a history of a previous shoulder dystocia—must be counseled about their increased risk for shoulder dystocia, and this conversation must be documented in the medical record.

This discussion should include:

  • her risk of having a shoulder dystocia at delivery

  • the risk that her baby will suffer injury—temporary and permanent—from the dystocia

  • the option of having a cesarean section as a means of avoiding this risk

  • the risks cesarean section entails for this and future pregnancies

From a medical/legal point of view, the most important points are that:

  • You considered the possibility of shoulder dystocia.

  • You made a rational decision as to how to proceed based on the information available to you.

  • You discussed this risk and your recommendation with your patient.

  • You documented this discussion.


Although shoulder dystocia is, in most cases, unpredictable and unpreventable, there are certain precautions that can be taken which will enable you to be best prepared when it does occur:

(1) Make certain that all patients in your obstetrical practice are screened for shoulder dystocia risk. Perform a routine “shoulder dystocia review” at or around 36 weeks gestation looking for:

  • A history of a previous shoulder dystocia

  • Gestational diabetes

  • Estimated large fetal weight

  • Excessive maternal weight gain

  • Excessive fundal height measurement

  • Mother’s statement that “this baby feels very big.”

If any risk factors are present, counsel your patient about this risk and document this conversation in the medical record.

(2) If there is any concern about fetal size as term nears, obtain an ultrasound for estimated fetal weight.

(3) Train so as to always be prepared for the possibility of a shoulder dystocia. Simulation drills for shoulder dystocia—and other infrequent obstetrical emergencies—are vital. The only props needed are a pelvic model and a doll.

3. Management

There are seven aspects to management of shoulder dystocia emergencies, each of which is vitally important in increasing the chances of a safe, successful outcome:

1. Heightened awareness that a shoulder dystocia might occur in a particular case

When a clinician suspects that there is an increased risk of shoulder dystocia in a given patient, he or she should do the following:

  • Notify both the labor nurse and the supervising (charge) nurse of the increased risk of shoulder dystocia with this delivery

  • Arrange for anesthesia and pediatric backup to be immediately available

  • Arrange for an experienced obstetrician to be on hand for possible assistance at delivery

  • Consider prophylactic McRoberts positioning

  • Make certain that a step stool is available to allow an assistant to provide suprapubic pressure at the correct angle

2. Diagnosis

A shoulder dystocia presents with the inability of the anterior fetal shoulder to emerge from the vagina with maternal pushing and routine physician traction after delivery of the head. It is often preceded by the turtle sign. The clinician must promptly recognize this as a shoulder dystocia and immediately do the following:

  • Have the mother stop pushing and cease traction if one or two such efforts are not successful in delivering the shoulder.

  • Alert medical personnel in the room to the presence of a shoulder dystocia.Ask that extra nursing staff, a pediatrician, an anesthesiologist, and another obstetrician be called to assist.

  • Briefly explain to the patient that the baby’s shoulders are “temporarily stuck” and that you and your team will be working to get the baby out safely.

  • Begin the series of shoulder dystocia resolution maneuvers in a calm, deliberate, yet time-sensitive fashion.

3. Operational control/situational awareness

Make the management of this emergency as deliberate and efficient as possible. By default, you will immediately become the leader of what must instantly become a high-performance team. Provide clear and firm direction. Avoid an inappropriate sense of urgency while at the same time recognizing that you have somewhere between 6 to 10 minutes before central neurologic damage is likely to occur.

Assign one staff member to be a timekeeper and scribe. This person should call out loud each 30-second interval that passes so that the team is constantly aware of the duration of the shoulder dystocia. Such knowledge helps the obstetrician in charge to modulate the pace of shoulder dystocia resolution maneuvers and to know when an increase in the intensity of delivery efforts may be appropriate.

4. Maneuvers

McRoberts maneuver

The McRoberts maneuver (Figure 2) is the most commonly used shoulder dystocia resolution maneuver. It involves flexing the maternal thighs up against the mother’s chest. The McRoberts positioning works not by changing the actual dimensions of the maternal pelvis, but by straightening out the sacrum relative to the lumbar spine. This allows cephalic rotation of the synthesis pubis, enabling the fetal shoulder to slide under it. The McRoberts maneuver is almost always used in conjunction with suprapubic pressure. Approximately 50% of shoulder dystocias can be resolved solely with the use of the McRoberts maneuver.

Figure 2.

McRoberts Maneuver

Suprapubic pressure

Suprapubic pressure (Figure 3) is pressure applied just above the maternal symphysis pubis to the anterior shoulder of the fetus. The goal of this maneuver is to move the fetal shoulder away from its direct anterior-posterior orientation in the maternal pelvis into an oblique position. Used together, the McRoberts maneuver and suprapubic pressure have been shown to resolve 60% of shoulder dystocias.

Figure 3.

Suprapubic Pressure

Suprapubic pressure must be distinguished from fundal pressure, in which downward pressure is applied to the top of the uterus. Fundal pressure serves only to drive the impacted shoulder further into a nondeliverable position, and should never be employed in the context of a shoulder dystocia.


Episiotomies are only useful if there is insufficient room in the vagina for the clinician to put his or her hand inside to perform necessary maneuvers. Shoulder dystocia results from a misfit between the fetal shoulder and the boney pelvis; it is not a soft-tissue dystocia. Episiotomies in and of themselves do not aid in the resolution of shoulder dystocia.

Woods corkscrew and Rubin’s maneuvers

These maneuvers are variations of ways of rotating the fetal shoulders in order to change their orientation in the maternal pelvis.

In the Wood’s corkscrew maneuver (Figure 4), the anterior surface of either uppr or lower fetal shoulder is pushed by the deliverer’s hand in an effort to “torque” the baby out of the vagina.

Figure 4.

Rotational Maneuver

In Rubin’s maneuver (Figure 4), the posterior aspect of either shoulder is pushed. The Rubin’s approach has the added benefit of “flexing” the shoulders, bringing them closer together and thus decreasing the biacromial diameter.

It is not particularly important whether it is the anterior or posterior shoulder or its front or back surface that is pushed during rotational maneuvers. The clinician should grasp whichever shoulder is most easily reached and push it in whichever direction the fetus turns most easily. High rates of success have been reported with the use of rotational maneuvers to resolve shoulder dystocias. Figure 4.

Delivery of the posterior arm

In this maneuver, the clinician places his or her hand in the posterior aspect of the >mother’s vagina and reaches up to locate one of the baby’s arms or hands. The hand or arm is then swept across the fetal chest and delivered. Once the entire arm and shoulder are exteriorized, it is easy to rotate the baby so as to free up the stuck anterior shoulder. The remainder of the baby delivers without difficulty.

Other maneuvers

While other maneuvers to resolve shoulder dystocia are described, they are rarely employed, either because of their high rate of complications or the difficulty of performing them (Table IV).

Table IV.n

Other maneuvers to resolve shoulder dystocia

5. Do not cut the umbilical cord prematurely

While cutting the nuchal cord after delivery of the head is often routine, doing so in the context of a shoulder dystocia may prove fatal. Umbilical cord transection may prevent even minimal blood flow to the baby throughout the duration of the dystocia. Before cutting an umbilical cord, the deliverer must be very certain that the entire baby will emerge within seconds thereafter.

6. Measure cord blood gases

Being able to document whether there was in fact real asphyxia at birth and whether there was a base deficit—indicative of longer-term asphyxia—can provide important information both as to the baby’s status and as a refutation to later claims of inappropriate care.

7. Have a discussion with the family about the events that transpired—and document this conversation

Family and friends observing the delivery see a relatively calm labor room erupt into a frenzy of activity with voices becoming tense and multiple medical practitioners coming and going. At delivery, the baby may temporarily appear blue and not cry. Inexperienced observers often draw a host of erroneous impressions as to what has happened in such a situation. It is vital, therefore, for the delivering clinician to make time to sit and talk to the family and friends after the delivery and to explain what really transpired and why.

4. Complications

Neonatal injury

Damage to the brachial plexus

The brachial plexus (Figure 5) is a series of nerves whose roots lie between C-5 and T-1. They control sensation and muscular function of the shoulder, upper arm, lower arm, wrist, and hand. If any of these nerves are permanently injured, paralysis develops in the muscles innervated by the affected nerve.

Figure 5.

Brachial Plexus Schema

Although the quoted rate of such injuries following shoulder dystocias varies widely, it is generally accepted that approximately 10% of all shoulder dystocia deliveries will be accompanied by some degree of brachial plexus palsy. Of these, 10% will be permanent. Thus, if shoulder dystocias occur in roughly one in 100 vaginal deliveries, the rate of permanent brachial plexus injury is one in 10,000 vaginal deliveries. Given that there are approximately 4 million babies born in the United States each year—roughly 3 million by vaginal delivery–there will be approximately 300 babies each year in the United States born with birth-related permanent brachial plexus injuries. Not only is this a tremendous tragedy for each child and family involved, but a very large percentage of these cases wind up in litigation.

Brachial plexus injuries can affect any or all of the C-5 to T-1 nerve roots. However, they generally fall into three categories:

  • Injury to C7 results in loss of innervation of the forearm, wrist, and finger extensors. This causes the wrist to flex and the fingers to curl up into the “waiters tip” position. The injury to C-5 to C-7 in Erb’s palsy results in paralysis or weakness of the shoulder muscles, elbow flexors, and forearm supinators. The affected arm hangs down and is internally rotated, extended, and pronated. Sometimes Erb’s palsy is accompanied by injury to C-4, which results in phrenic nerve damage and diaphragmatic paralysis.

  • Klumpke’s palsy involves damage of the C-8 and T-1 nerve roots. It is characterized by weakness of the triceps, forearm, pronators, and wrist flexors. This leads to a clawlike, paralyzed hand with—if it is in isolation—good elbow and shoulder function. Klumpke’s palsy is more likely to be permanent than Erb’s palsy; only 40% of cases of Klumpke’s palsy resolve by one year, whereas 90% of cases of Erb’s palsies do. The proximity of some portions of the cervical sympathetic nerve chain to the C-8 and T-1 nerve roots sometimes leads to sympathetic nerve damage when there is a severe brachial plexus injury. This results in Horner’s syndrome: sensory deficits on the affected side, contraction of the pupil, and ptosis of the eyelids.

  • When the entire chain of brachial plexus nerves are injured—a pan-brachial plexus injury (C5-T1)—all of the muscles of the arm are affected. This can result in a “flail” arm, which is an arm that hangs uselessly by the side and because of a lack of sensation is subject to various traumatic injuries (Table V). There are three categories of injury to the nerves: (1) stretching of some of the component nerve fibers; (2) total rupture of the nerve; and (3) avulsion of the nerve (i.e., the tearing of the nerve off of its attachment to the spinal cord).

Table V.n

Types of brachial plexus injuries

A nerve that is partially torn can often regenerate itself with restoration of function; ruptured or avulsed nerves do not regenerate and cannot be repaired. In these cases, permanent injury results.

What causes brachial plexus injuries?

While it is universally acknowledged that inappropriate traction on a baby’s head during an attempt to resolve a shoulder dystocia can cause a brachial plexus injury, the consensus view – as expressed in the ACOG Practice Bulletin on shoulder dystocia and in the major obstetrical textbooks – is that there are multiple potential etiologies for brachial plexus injury (see Figure 5):

  • Stretching of the brachial plexus nerves by inappropriate physician traction

  • Damage to the brachial plexus nerves despite appropriate traction and delivery maneuvers due to the intrinsic variation in the strength of nerve fibers between individual neonates (i.e., normal biologic variation)

  • Stretching of the brachial nerves by endogenous forces—maternal pushing and uterine contractions. This occurs when the long axis of the baby’s body is pushed downward while its anterior shoulder is trapped behind the mother’s pubic bone. Empirical evidence of this is seen in the turtle sign.

  • Direct crushing pressure of the brachial plexus nerves against the maternal pubic bone caused by the endogenous forces of uterine contractions and maternal expulsive efforts.

Other shoulder dystocia-related neonatal injuries

Fractured clavicle and humerus

Fracture of the clavicle is seen in approximately 3% to 9% of shoulder dystocia deliveries. These almost always heal without complication. The statistics for fracture of the humerus are similar.


Once the fetal head is delivered, the umbilical cord is usually compressed between the fetal chest and the maternal vaginal wall. This results in either total or partial cessation of blood flow to the fetus. Should this condition persist too long, cerebral hypoxia and resulting brain damage can occur. The exact duration of shoulder dystocia necessary to cause central neurologic damage is uncertain but is thought to be between 6 and 10 minutes.

Fetal death

Prolonged hypoxia due to (1) a shoulder dystocia that cannot be resolved or (2) trauma from overly vigorous efforts to resolve a shoulder dystocia can in extremely rare cases result in fetal demise.

Maternal injuries

Fortunately, injuries to the mother following shoulder dystocia deliveries are relatively uncommon and are of a lesser magnitude than those sustained by neonates. There is approximately a 10% rate of postpartum hemorrhage and an 8% percent rate of fourth degree perineal lacerations following shoulder dystocias. Other injuries sometimes seen are:

  • Postpartum atrophy of the bladder due to prolonged compression.

  • Maternal ligament and tendon strain or rupture due to vigorous application of shoulder dystocia resolution maneuvers.


Careful and complete documentation of shoulder dystocia deliveries and the response to them is vital. Below are some of the features that any such documentation record should include:

  • Predelivery estimate of fetal weight and pelvic capacity

  • The timing of the active phase and the second stage of labor

  • The form of anesthesia that was in place, if any

  • When and how the shoulder dystocia was diagnosed

  • Whether assistance of other personnel was called for

  • Whether or not an episiotomy was made

  • A description of the various maneuvers used and how long each was attempted

  • A description of the estimation of force applied at various stages of resolution attempts

Use terms that convey an accurate sense of the magnitude of the force used: “Similar to that of a standard vaginal delivery,” “minimal traction,” “pull requiring great force,” etc. Also note the outcomes of each maneuver used and of each instance of clinician traction.

  • The total head-to-body delivery time

  • Evaluation of the baby’s status after delivery

  • Documentation of the conversation with parents following delivery

5. Prognosis and outcome

Controversies regarding shoulder dystocia

Is shoulder dystocia predictable?

No. As the most recent ACOG bulletin on shoulder dystocia states, “Most cases of shoulder dystocia cannot be accurately predicted or prevented.” The major problem with trying to predict shoulder dystocias is both the poor predictive value and the high false positive rates of any of the predictive protocols thus far described. Specifically:

  • Despite macrosomia being the most important risk factor for shoulder dystocia, just under 50% of shoulder dystocias occur in babies weighing <4,000 g.

  • Macrosomia itself cannot be accurately predicted prior to delivery. Ultrasound examination has proven no better at fetal weight estimation than clinical examination. To have 90% confidence based on ultrasound examination that a baby weighs more than 4,000 g, the ultrasound estimate must exceed 4,600 g.

  • Fetal weight predictions are even less accurate at higher birth weights

Can shoulder dystocia be avoided?

What this question really asks is whether all women suspected of carrying macrosomic babies—or who have had previous shoulder dystocia deliveries–should be delivered via elective cesarean section. Again, the majority of those who have studied shoulder dystocia would answer “No.”

Rouse and Owen showed that the policy of prophylactic cesarean delivery for suspected macrosomia would require several thousand cesarean deliveries and millions of dollars to avert a single permanent brachial plexus injury. Even for higher risk patients whose mothers have gestational diabetes and suspected macrosomia, 443 unnecessary cesarean sections would have to be performed to prevent one permanent brachial plexus injury.

Moreover, cesarean sections in this frequently obese, often diabetic patient population are not necessarily benign procedures. And to the surgical risks must be added: (1) the cost of the increased number of cesarean sections, (2) the longer recovery periods, and (3) the fact that doing an initial cesarean section likely condemns a woman to future cesarean sections—with increased risk of placenta accreta and the complications this entails.

Can early induction of labor decrease the incidence of macrosomia and thus decrease the incidence of shoulder dystocia and permanent brachial plexus injury?

While such a policy has intuitive plausibility, it has been refuted by direct observation. Multiple studies have shown that induction “early” to avoid extra fetal weight gain does not improve maternal or fetal outcomes. Rather, it was found that in nondiabetic patients, labor induction doubled the risk of cesarean section without reducing the rate of shoulder dystocia or newborn morbidity.


Shoulder dystocia remains—and is likely to continue to remain—an unpredictable and unpreventable obstetrical emergency. Even if tools existed that allowed obstetricians to precisely determine fetal weight prior to delivery, the biologic variability of fetal shape, maternal pelvic dimensions, and the direction of forces in labor would make the predictability of shoulder dystocia extremely unreliable.

Obstetricians and society will have to make a decision as to the number of unnecessary cesarean sections it is willing to have performed in order to prevent a single case of permanent brachial plexus injury.

For now, best practice—and standard of care—is the following:

  • Be aware of each patient’s risk factors and evaluate each patient throughout their pregnancy to identify those at highest risk for shoulder dystocia.

  • Counsel such patients as to their risk of having a shoulder dystocia at delivery, and let them know their options and the pros and cons of these options.

  • Always be prepared during any vaginal delivery for the possibility of a shoulder dystocia. Know how to: (a) efficiently organize the medical personnel at hand, and (b) how to assume your role as team captain of what must instantly become a high-performance team.

  • Know and practice your hospital’s shoulder dystocia protocol. Keep your shoulder dystocia knowledge and skills sharp with simulation drills, on-line and in-person refresher courses, and memory tools, such as checklists and documentation forms.

6. What is the evidence for specific management and treatment recommendations

Acker, DB, Gregory, KD, Sachs, BP, Friedman, EA. “Risk factors for Erb-Duchenne Palsy”. Obstet Gynecol. vol. 71. 1988. pp. 389-92. (This seminal article was one of the first to evaluate from actual medical records (not discharge summaries or birth certificates) risk factors for shoulder dystocia. It is still a key source for data on the percentage of shoulder dystocia deliveries in (1) various neonatal weight groups and (2) between babies born of diabetic and nondiabetic mothers. Caveat: All of the percentages listed for shoulder dystocia in this study are about 50% higher than in most subsequent papers.)

Allen, R, Sorb, J, Gonik, B. “Risk factors for shoulder dystocia: an engineering study of clinician-applied forces”. Obstet Gynecol. vol. 77. 1991. pp. 352-55. (This article is famous, or rather notorious, because it purports to show exactly how much force on the part of a deliverer it takes to injure a neonate's brachial plexus. Unfortunately its data has been abused in the medical-legal world as setting a ceiling for force used—when, in fact, the paper only examined forces in two babies who had a shoulder dystocia and only one with a brachial plexus injury—and that one only temporary! Nevertheless, if you are sued for a brachial plexus injury, this is the article the plaintiff attorney will use to try to demonstrate the unfounded conclusions that: (1) all brachial plexus injuries are caused by "excess" physician traction, and that (2) this article says exactly what that "excessive" amount is. In fact, the article does no such thing.)

“American College of Obstetricians and Gynecologists”. Practice bulletin number 40: Shoulder dystocia. 2002. (While there is a disclaimer on the first page stating, "These guidelines should not be construed as dictating an exclusive course of treatment or procedure," in reality, these guidelines are considered de facto national standards of care. This Practice Bulletin does a good job of summarizing issues of the predictability of shoulder dystocia and of physician management of it; the bulletin does not go into specific maneuvers or protocols.)

Gherman, RB, Goodwin, TM, Ouzounian, JG. “Brachial plexus palsy: An in utero injury?”. Am J Obstet Gynecol. vol. 180. 1999. pp. 1303-7. (This is one of the first articles looking critically at the claim that all brachial plexus injury is caused by "excessive" physician traction. It outlines much evidence against this oft-claimed but unproven hypothesis.)

Gonik, B, Walker, A, Grimm, M. “Mathematic modeling of forces associated with shoulder dystocia: A comparison of endogenous and exogenous forces”. Am J Obstet Gynecol. vol. 182. 2000. pp. 689-91. (Gonik's mathematical "proof" that endogenous forces—uterine contractions and maternal pushing—can be 4 to 9 times greater than exogenous forces—physician traction—added additional support to those formulating alternative hypotheses regarding the causation of brachial plexus injuries. This theoretical framework squares with the many reports of brachial plexus injuries from cesarean section deliveries, when there was no shoulder dystocia, and with the observation of the turtle sign.)

Lerner, HM, Salamon, E. “Permanent brachial plexus injury following vaginal delivery without physician traction or shoulder dystocia”. Am J Obstet Gynecol. vol. 198. 2008. pp. e7-8. (This article was the first to document the delivery of a baby in which: (1) there was no shoulder dystocia (the baby was delivered in its entirety in a single push), and (2) hence, no physician traction that resulted in a permanent brachial plexus injury. Allen had published a similar article in 2005 (Obstet Gynecol 2005:105(S Pt 2):1210-2) involving a baby that had a temporary brachial plexus injury.)