The term shoulder dystocia describes a series of difficulties encountered with release of the fetal shoulders in cephalic deliveries and more objectively the need to use additional manoeuvres when axial traction on the fetal head has failed. An overall incidence between 0.58 and 0.70 % of vaginal deliveries is reported in the largest observational studies.

The main mechanism behind the occurrence of shoulder dystocia is the retention of the anterior shoulder behind the pubic symphysis, while the posterior shoulder is usually located in the maternal pelvis (Fig. 3.1). In rare situations, both shoulders are retained above the pelvic brim.

The main risk factors for shoulder dystocia are listed in Table 3.1. Previous shoulder dystocia stands out as a major risk factor for recurrence, and it is reported to be 10-times higher than in the general population, for an overall incidence of 1–25 %. The anatomical characteristics of the maternal pelvis that predispose to shoulder dystocia and may cause it to be recurrent in nature are poorly understood. When additional risk factors are present, such as maternal diabetes or suspected fetal macrosomia or when previous fetal injury occurred in association with shoulder dystocia, serious consideration should be given to elective caesarean delivery, and this option should be discussed with the mother. In the remaining situations, management remains controversial.

Another major risk factor is fetal macrosomia, and when coexistent with poorly controlled maternal diabetes, an additional 2–4-fold risk is present, posed by the increased diameter of the fetal shoulders.

With isolated macrosomia, there is recent evidence from a randomised controlled trial that fetuses whose estimated weight is above the 95th percentile close to term benefit from induction of labour at 37–38 weeks, as it reduces the risk of shoulder dystocia by about 70 % and also slightly increases the number of vaginal deliveries. An estimated fetal weight above 5000 g is considered in many guidelines to be an indication for elective caesarean section, and others include this recommendation for estimated weights above 4500 g.

With maternal diabetes, labour induction at 38–39 weeks has been shown to decrease the incidence of shoulder dystocia and is recommended in several guidelines. Some institutions recommend elective caesarean section when estimated fetal weight is above 4250 g or above 4500 g.

The above recommendations have been criticised for the limited strength of the evidence on which they are based. Weight estimation by ultrasound has a well-demonstrated inherent error, particularly in macrosomic fetuses. One should also not forget that caesarean section for fetal weights above 4500 g is associated with a 2.5 % risk of trauma. Nevertheless, these recommendations remain important sources of guidance, while stronger evidence is awaited.

The most frequent complication of shoulder dystocia is brachial plexus injury, of which Erb’s palsy is the usual presentation. The latter manifests by a characteristic position of the affected arm that hangs by the side of the body and is rotated medially. The forearm is usually extended and pronate (Fig. 3.2). It affects about 0.15 % of all births, and in some countries, the incidence appears to be decreasing. Older studies report brachial plexus injury to occur in 2–16 % of shoulder dystocias, but recent data from centres performing regular staff training refer that this can be reduced to about 1.3 %. Brachial plexus injury appears to be related mainly to the traction force applied on the fetal head. Improved awareness of the fact and simulation-based training of the force that can be safely applied to the fetal head may be responsible for the decreasing incidence of this complication.

The underlying cause for injury is believed to be exaggerated mechanical distention of the brachial nerve, but the mechanism is incompletely understood, as it may also affect the posterior arm, and cases have been reported to occur in elective caesarean section and vaginal delivery without shoulder dystocia. In 24–53 % of brachial plexus injuries, no documentation of shoulder dystocia was found, and 4–12 % occurred after caesarean section, so it is possible that they are also caused by abnormal intrauterine fetal positions.

Of all brachial plexus injuries diagnosed at birth, the majority disappear after treatment, and only 10–23 % remain after 12 months. In the majority of cases of residual paralysis, some degree of recovery is achievable after surgery.

Shoulder dystocia is also a cause of perinatal mortality, although the incidence appears to have decreased in the last decades. Confidential enquiries carried out in the United Kingdom indicate that it may be responsible for about 8 % of intrapartum fetal deaths. The main cause of perinatal death is acute fetal hypoxia/acidosis (see Chap. 2).

There is uncertainty about how many minutes may elapse before the fetus is at risk of injury from acute hypoxia/acidosis. The phenomenon is probably faster when there are nuchal cords and when cord clamping takes place before the shoulders are released. Five such cases with resolution taking 3–7 min were reported in association with cerebral palsy, but it is uncertain whether intrapartum hypoxia/acidosis was already present before delivery of the head. Sustained cord occlusion is likely to occur in these situations, and therefore umbilical blood gas values may not translate the severity of hypoxia/acidosis. Hypoxic-ischaemic encephalopathy has been documented in cases with only moderate acidaemia on cord gas analysis.

Compression of fetal neck vessels may also play an important part in the pathophysiological mechanism, and it may be the main cause of cerebral injury when no nuchal cords are present. Again, umbilical blood gas values may not translate the severity of hypoxia/acidosis occurring in the brain, and hypoxic-ischaemic encephalopathy has been documented in cases with only moderate acidemia on cord gas analysis. In a small but well-documented observational study, no cases of hypoxic-ischaemic encephalopathy were found when resolution took less than 5 min, and only mild cases of hypoxic-ischaemic encephalopathy were reported when it lasted 5–9 min. Serious complications of hypoxia/acidosis were only described in one case where more than 12 min elapsed.

Although there are no certainties as to the time that may elapse before the fetus is at risk of injury from hypoxia/acidosis, it seems wise not to clamp nuchal cords after the head is delivered unless there is no other alternative and to attempt resolution preferably within 5 min. When 12 min have elapsed, fetal prognosis is likely to be poor. Different timings must be considered when fetal oxygenation is already compromised before the occurrence of shoulder dystocia or when there is fetal growth restriction.

Other criteria have been proposed for the definition of shoulder dystocia, based on the time interval between delivery of the fetal head and release of the shoulders, and they consider cut-offs of 1 or 2 min. However, accurate measurement of these intervals may not be practicable in many hospitals.