Shoulder dystocia: Risk factors and planning birth of high-risk pregnancies

INTRODUCTION — A vaginal birth is complicated by shoulder dystocia when, after expulsion of the head, additional obstetric maneuvers beyond gentle guidance are needed to enable extraction of the shoulders. Shoulder dystocia is an obstetric emergency because it may result in life-threatening newborn injuries, as well as less serious maternal injuries. It has been estimated that one newborn with hypoxic ischemic encephalopathy due to shoulder dystocia is born out of every 22,000 term vaginal births [1].

This topic will discuss antepartum issues related to shoulder dystocia: risk factors for the disorder and planning birth when these risk factors are identified before labor. Since many shoulder dystocias occur in pregnancies without risk factors, clinicians should always be prepared for the occurrence of the disorder. Intrapartum issues, including pathophysiology, diagnosis, management, and outcome of shoulder dystocia are reviewed separately. (See "Shoulder dystocia: Intrapartum diagnosis, management, and outcome".)

CAN SHOULDER DYSTOCIA BE PREDICTED? — There is expert consensus that the occurrence of shoulder dystocia cannot be accurately predicted by antenatal or intrapartum risk factors or imaging studies [2]. The predictive value of any one or combination of risk factors for shoulder dystocia is low (probably less than 10 percent), and at least 50 percent of pregnancies complicated by shoulder dystocia have no identifiable risk factors [3-5]. (See 'Risk factors' below and 'Lack of utility of pelvimetry and fetal biometry' below.)

Because shoulder dystocia cannot be accurately predicted, all obstetric care providers should be able to promptly recognize when gentle traction alone is inadequate for extraction of the shoulders and be able to proceed through an orderly sequence of maneuvers to deliver the fetus in a timely manner, with no or minimal maternal and fetal trauma. It should be noted, however, that permanent birth injury (eg, neonatal brachial plexus palsy), and even perinatal death, can occur in cases of shoulder dystocia that are appropriately identified and managed. Furthermore, neonatal brachial plexus palsy, which has been attributed to iatrogenic maneuvers that stretch the brachial plexus when the fetus with shoulder dystocia is born vaginally, can occur in infants from cesarean births and from vaginal births without shoulder dystocia [6,7]

RISK FACTORS — Clinicians should be aware of risk factors for shoulder dystocia, so they can discuss the possibility of scheduled cesarean birth with patients at highest risk of occurrence and appropriately manage a prolonged second stage of labor in less high-risk patients.

Antepartum

High birth weight — High birth weight is the major risk factor for shoulder dystocia. Although other risk factors have also been associated with shoulder dystocia (eg, maternal obesity, diabetes, older age at first birth), they are often related to high birth weight.

The incidence of shoulder dystocia increases progressively as birth weight increases over 4000 g (table 1) [3,8-11], and morbidity and mortality from shoulder dystocia increase substantially when birth weight is ≥4500 g [12,13]. Some risk factors for high birth weight are listed in the table (table 2).

Although birth weight is a risk factor for shoulder dystocia, it is not highly useful for predicting its occurrence because:

●The majority of extremely high birth weight newborns do not have shoulder dystocia. As an example, shoulder dystocia was reported in only 15.5 percent of 7859 newborns with birth weight ≥5000 g (11 lb) delivered vaginally in one series [14].

●Approximately 50 percent of shoulder dystocias occur in newborns with birth weight <4000 g [15].

●It is difficult for the clinician to predict birth weight prior to birth. Sonographic estimation of fetal weight does not accurately detect or exclude macrosomia. Clinical estimates of birth weight based upon abdominal palpation are also insensitive. (See "Fetal macrosomia", section on 'Sonography'.)

Diabetes mellitus — Maternal diabetes mellitus increases the likelihood of shoulder dystocia severalfold over the nondiabetic population (table 1). This increase is due, in part, to the higher prevalence of high birth weight in patients with diabetes compared with nondiabetic patients [8,15]. In addition, the anthropomorphic measurements of newborns of diabetic mothers (IDMs) are different from those in offspring of mothers without diabetes. Specifically, the chest-to-head and shoulder-to-head ratios are increased in IDMs, thereby increasing the risk of shoulder dystocia independent of fetal weight and at weights <4000 g [16,17]. Of note, even a single abnormal glucose value in a 75 g two-hour glucose tolerance test is associated with adverse pregnancy outcome, including macrosomia and shoulder dystocia [18]. (See "Infants of mothers with diabetes (IMD)".)

Previous shoulder dystocia — Most studies report a recurrence risk of at least 10 percent, but recurrence risks ranging from 1 to 25 percent have been reported in retrospective studies (odds ratio range for recurrence: 6 to 28) [19-27]. This may be an underestimate of the true recurrence risk if vaginal birth is attempted because many patients and clinicians choose a planned cesarean birth in pregnancies subsequent to an episode of shoulder dystocia and thus avoid recurrence.

Recurrence is not surprising since high birth weight is a major risk factor for shoulder dystocia and tends to recur. In one study, one-third of nondiabetic patients with newborns ≥4500 g at their first birth also had newborns this large at their second birth, whereas only 0.3 percent of those with a normal weight newborn (3000 to 3500 g) at their first birth had a newborn weighing ≥4500 g in their second birth [28].

Recurrent shoulder dystocia is more likely when [21]:

●Birth weight is greater than in the previously affected pregnancy

●Prepregnancy weight is greater than in the previously affected pregnancy

●Gestational weight gain is greater than in the previously affected pregnancy

●Second stage of labor is longer than in the previously affected pregnancy

●Birth weight is >4000 g

The combination of a previous shoulder dystocia and high birth weight is particularly worrisome. In one small series, 11 of 21 patients (52 percent) with both risk factors experienced recurrent shoulder dystocia [29].

Of note, the absence of shoulder dystocia in a previous pregnancy does not preclude its occurrence in a subsequent pregnancy. In a retrospective cohort study including all patients in Norway with two consecutive singleton vaginal births during 1967 to 2005 (n = 537,316), 96 percent of cases of shoulder dystocia in the second birth were in patients without a previous history of the complication [19].

Postterm pregnancy — Postterm pregnancy is a risk factor for shoulder dystocia, presumably because of higher birth weights with advancing gestational age [10]. In a cohort study of term (n = 379,445) and postterm (n = 65,796) births from Norway, the relative risk of shoulder dystocia in postterm births was increased by 30 percent (0.73 versus 0.53 percent; relative risk [RR] 1.3, 95% CI 1.2-1.4) [30]. An increased prevalence of high birth weight was primarily responsible for the increased risk of shoulder dystocia because the difference was not apparent after stratification by birth weight. (See "Postterm pregnancy".)

Male fetal sex — Male sex is more common in pregnancies complicated by shoulder dystocia than in the overall birth population (55 to 68 percent versus 51 percent) [4,8,31]. Higher birth weights among males likely contribute, at least in part, to this finding. In one population-based series, 70 percent of newborns with birth weight >4500 g were male, whereas 51 percent of all newborns are male [13]. Anthropomorphic dimensions may also account for some of the increased risk of shoulder dystocia among males, similar to findings in IDMs [8].

Maternal obesity and excessive gestational weight gain — High maternal body mass index (BMI) and excessive gestational weight gain are risk factors for both high birth weight and diabetes, which are also risk factors for shoulder dystocia [32,33]. In a meta-analysis, the pooled relative risk of shoulder dystocia for patients with versus without obesity was 1.63 (95% CI 1.33-1.99) [33]. Excessive gestational weight gain was associated with a greater proportion of large for gestational age neonates than either BMI >25 kg/m² or gestational diabetes, in one study [32].

Maternal demographics — Advanced maternal age is a risk factor for shoulder dystocia; however, confounding variables such as an increased prevalence of diabetes and higher maternal weight probably account for this association [15]. Increased maternal age is also associated with increased parity, which may be another confounding factor since each successive newborn tends to be 200 g larger than the prior one up to the fifth pregnancy. Short maternal stature (<160 cm) is often reported as a risk factor for shoulder dystocia and cesarean birth [34,35]. In the United States, African American patients appear to be at higher risk of shoulder dystocia in offspring than White patients [36].

Intrapartum

Abnormal progress of labor — Labor abnormalities alone are not useful predictors of shoulder dystocia due to the high frequency of labor abnormalities in the general obstetric population and the relatively low frequency of shoulder dystocia [37,38]. A relationship between shoulder dystocia and abnormal labor progression, including both precipitous and prolonged second stage, has been reported at both average and high birth weights, but data are inconsistent [39-44].

The combination of labor abnormalities, suspected high birth weight (>4000 g), and assisted vaginal birth appears to be more predictive of shoulder dystocia than labor abnormalities alone. In a classic study, the combination of prolonged second stage, birth weight >4000 g, and midpelvic assisted vaginal birth was associated with a 21 percent incidence of shoulder dystocia [45]. In comparison, when only prolonged second stage and midpelvic assisted vaginal birth were present, the risk fell to 4.6 percent and was 0.16 percent in the absence of these risk factors.

Assisted vaginal birth — Forceps- or vacuum-assisted vaginal birth is associated with shoulder dystocia [46], but it is unclear whether instrument-assisted fetal descent increases the risk of shoulder impaction or whether shoulder malposition inhibits descent, resulting in an increased frequency of assisted vaginal birth. A meta-analysis of observational studies reported a similar risk of shoulder dystocia for vacuum and forceps deliveries (odds ratio 1.11, 95% CI 0.25-4.89; four studies) [46]. Widely inconsistent results among the studies may be explained by the fact that the choice of vacuum versus forceps depends in part on the clinical setting, and some clinical settings are associated with a higher risk of shoulder dystocia than others. Operator expertise with each instrument may be another factor.

The only randomized trial of obstetric forceps versus the M-cup vacuum extractor reported a higher risk of shoulder dystocia with the use of a vacuum [47]. It is hypothesized that this finding, which is supported by some observational data [48-50], may be explained by different traction force vectors on the head with vacuum versus forceps, such that the vacuum instrument is more likely to pull the anterior shoulder into the symphysis pubis and leave the posterior shoulder above the sacral promontory [48].

High birth weight is an important confounder. As discussed above, in a classic study, the combination of birth weight >4000 g, prolonged second stage, and midpelvic assisted vaginal birth was associated with a 21 percent incidence of shoulder dystocia compared with 4.6 percent when only prolonged second stage and midpelvic assisted vaginal birth were present [45].

RISK REDUCTION STRATEGIES — Although most risk factors for shoulder dystocia are not modifiable, individuals have some control over their prepregnancy weight, gestational weight gain, and glucose level (in patients with diabetes).

●Dietary and lifestyle interventions that reduce prepregnancy obesity and gestational weight gain may reduce the frequency of large for gestational age infants and, in turn, shoulder dystocia. Dietary and physical activity-based interventions also have other health benefits and are not harmful.

A meta-analysis of randomized trials reported that such interventions reduced the overall risk of shoulder dystocia by 60 percent (relative risk [RR] 0.39, 95% CI 0.22-0.70; four trials, n = 2317) compared with usual care, but the difference was not statistically significant when patients with diabetes were excluded from the analysis [51]. (See "Obesity in pregnancy: Complications and maternal management" and "Gestational weight gain".)

●The US Preventive Services Task Force and the National Institutes of Health concluded that treatment of gestational diabetes can reduce the risk of macrosomia (RR 0.53, 95% CI 0.41-0.68) and, in turn, shoulder dystocia (RR 0.42, 95% CI 0.23-0.77) [52,53].

Although the level of glycemia in patients with pregestational diabetes is a factor in macrosomia risk, additional metabolic factors appear to be involved such that good but not stringent glycemic control may not substantially reduce the frequency of macrosomia and shoulder dystocia in this population [54]. In one study, the mean blood glucose level had to be at least one to two standard deviations below the mean blood glucose levels of pregnant people without diabetes to achieve comparable macrosomia rates [55]. (See "Pregestational (preexisting) diabetes: Preconception counseling, evaluation, and management" and "Pregestational (preexisting) diabetes mellitus: Antenatal glycemic control".)

PLANNING BIRTH IN HIGH-RISK PREGNANCIES

Lack of utility of pelvimetry and fetal biometry

●Pelvimetry – Almost all shoulder dystocias occur in patients with normal pelvic dimensions; therefore, imaging studies (radiograph, computed tomography, ultrasound) and clinical pelvimetry are not useful for identifying patients at increased risk, except in rare cases of a severe fetal or pelvic abnormality [8,56].

●Fetal biometry – The clinical relevance of markedly abnormal fetal biometry should be interpreted on a case-by-case basis, whereas mildly abnormal biometry is not very predictive of birth outcome. Several fetal biometric parameters (eg, difference between the abdominal and biparietal diameters, chest circumference, humerospinous distance, cheek-to-cheek diameter, shoulder width) have been used to predict shoulder dystocia, primarily in fetuses of patients with diabetes [57-61]. The value of this approach has not been tested in large, prospective studies. Likewise, head circumference/abdominal circumference and femur length/abdominal circumference ratios may be routinely included in obstetric ultrasound reports in late gestation, but neither normal nor most abnormal ratios are useful for predicting or excluding shoulder dystocia. In a systematic review of studies of various fetal biometric parameters for predicting shoulder dystocia, abdominal diameter minus biparietal diameter ≥2.6 cm, abdominal circumference >90^th centile, and estimated fetal weight >4000 g had positive predictive values of only 7 to 11 percent [62].

Shared decision making — The planned route of birth is a shared decision made by the clinician and patient, ideally after nondirective counseling about the risks of labor and vaginal birth, including a discussion about shoulder dystocia. If planned cesarean birth is being considered, the risks of this procedure should also be discussed.

There is general consensus among experts that planned cesarean birth is appropriate in those pregnancies most likely to result in shoulder dystocia with long-term complications in order to reduce this potential outcome (see 'Patients with prior shoulder dystocia or estimated weight greater than 4500 or 5000 grams' below). Although this will prevent some shoulder dystocias and their associated complications, most cases of shoulder dystocia, including those that result in neonatal injury, cannot be predicted before birth, thus they cannot be prevented, as discussed above [41]. (See 'Can shoulder dystocia be predicted?' above.)

Similarly, there is general consensus favoring a trial of labor when the risk of shoulder dystocia is very low, such as for a multiparous patient without a history of difficult vaginal birth, who is in spontaneous labor at 39 weeks of gestation, and has an estimated fetal weight under 4000 g. However, most clinical settings fall into gray areas since at least 50 percent of pregnancies complicated by shoulder dystocia have no identifiable risk factors and most risk factors are weakly predictive of morbidity from shoulder dystocia. Our approach is summarized in the algorithm (algorithm 1) and discussed below.

Patients with prior shoulder dystocia or estimated weight greater than 4500 or 5000 grams — We consider the following clinical situations as particularly high risk for shoulder dystocia and brachial plexus injury, in alignment with the American College of Obstetricians and Gynecologists (ACOG) [2,63,64]. Scheduled cesarean birth in these scenarios is reasonable as it should reduce the occurrence of shoulder dystocia and associated morbidity. Ideally, the estimated fetal weight is based on sonographically determined weight performed within one week of birth.

●Prior shoulder dystocia, especially with a severe neonatal injury – There are no data from prospective studies to guide birth planning in pregnancies at increased risk of recurrent shoulder dystocia. The clinician should counsel the patient about the potential risk of recurrence (probably at least 10 percent), including factors in the current pregnancy that may impact this risk (eg, estimated fetal weight, presence or absence of gestational diabetes), and elicit her participation in making a decision about route of birth. (See 'Previous shoulder dystocia' above.)

●Estimated fetal weight >5000 g in patients without diabetes (estimated risk of shoulder dystocia >20 percent).

●Estimated fetal weight >4500 g in patients with diabetes (estimated risk of shoulder dystocia approximately 15 percent).

The weight-based approach takes into account the difficulty in accurately estimating fetal weight (especially when >4500 g), the relatively low incidence of shoulder dystocia-related trauma (even in high birth weight newborns), and the fact that prophylactic cesarean birth for patients with suspected high birth weight will likely result in many unnecessary cesarean births of both normal weight and high birth weight neonates [65-73].

In patients with diabetes, a decision-analysis model estimated that 443 cesarean births would need to be performed at the >4500 g weight threshold to prevent one permanent brachial plexus injury; other adverse outcomes of shoulder dystocia were not considered in the analysis [67]. The model found the 4500 g threshold was much less beneficial in patients without diabetes, as 3695 cesarean births would need to be performed to prevent one permanent brachial plexus injury. In patients without diabetes, a 5000 g threshold for cesarean birth is reasonable as it is estimated to prevent at least 20 percent of shoulder dystocias with minimal effect on the total cesarean birth rate, which is already 50 to 60 percent for newborns at this birth weight [74,75]. This approach might also prevent those shoulder dystocias with the highest risk of severe morbidity since the risk of permanent brachial plexus injury in shoulder dystocia appears to be related to birth weight [76,77].

Patients with a prolonged second stage — Patients with estimated fetal weights >4000 g and <4500 g (diabetics) or <5000 g (nondiabetics) undergoing a trial of labor become high risk for shoulder dystocia intrapartum if they have a prolonged second stage and need an operative vaginal birth (see 'Intrapartum' above and 'High birth weight' above). The principle of cesarean birth to prevent morbidity from shoulder dystocia also applies in this setting.

●We suggest intrapartum cesarean birth rather than a low pelvic or outlet assisted vaginal birth in pregnancies with a prolonged second stage and estimated fetal weight >4500 g.

●We suggest intrapartum cesarean birth rather than a midpelvic assisted vaginal birth when the estimated weight is >4000 g.

●In other scenarios, assisted vaginal birth can be considered case by case, based on factors such as fetal size and position, past obstetric history, maternal habitus (eg, short versus normal stature, obesity versus normal body mass index), and operator expertise. (See "Labor: Diagnosis and management of a prolonged second stage", section on 'Timing of operative delivery'.)

ACOG has opined that clinicians should exercise good judgment when deciding whether to perform an assisted vaginal birth in patients with a prolonged second stage and other risk factors for shoulder dystocia [2]. They state that assisted vaginal birth can be reasonable in selected patients, depending on their combination of specific risk factors.

Patients with imminent macrosomia — Induction of labor has been proposed when macrosomia appears imminent, with the goal of preventing the potential adverse consequences associated with continued fetal growth. Although logical, concerns have been raised about the limitations of ultrasound estimated fetal weight, the number of inductions that would need to be performed to prevent one adverse outcome, the maternal and neonatal consequences of induction, and the lack of evidence establishing the effectiveness of this approach.

Our approach is described below.

Patients without diabetes — We believe that it is reasonable to offer induction of labor to patients at 39 weeks of gestation without diabetes and with estimated fetal weight between 4000 and 5000 g, but expectant management is also reasonable, given limited data indicating benefit. The potential reduction in shoulder dystocia and its sequelae with induction at 39 weeks is likely to be less than with induction at 37 or 38 weeks, but common neonatal morbidities, such as hyperbilirubinemia and respiratory problems, are higher with induction at 37 or 38 weeks [78]. This approach is consistent with ACOG, which discourages induction of labor before 39+0 weeks for suspected macrosomia because of insufficient evidence that the benefits of reducing the risk of shoulder dystocia override the harms of early birth [64]. Further trials of early-term induction for suspected fetal macrosomia are needed to refine the optimum gestational age and estimated fetal weight for induction.

In a meta-analysis of labor induction versus expectant management in pregnancies with suspected macrosomia at 37 to 40 weeks of gestation (four randomized trials, n = 1190 primarily nondiabetic patients [some patients treated with nutritional therapy alone were included]), induction [79]:

●Reduced the risk of neonatal fractures (4 versus 20 per 1000, relative risk [RR] 0.20, 95% CI 0.05-0.79)

●Reduced the risk of shoulder dystocia (41 versus 68 per 1000, RR 0.60, 95% CI 0.37-0.98)

The number of inductions needed per fracture prevented was 60. The only two brachial plexus injuries were in the expectant management group, but the number of patients in the trials was too small and power was insufficient to determine whether this difference was clinically significant. The authors also noted that a policy of labor induction for suspected macrosomia at or near term did not significantly affect the rate of cesarean birth (RR 0.91, 95% CI 0.76-1.09) or assisted vaginal birth (RR 0.86, 95% CI 0.65-1.13), but did suggest a trend toward higher risk of third- and fourth-degree anal sphincter tears (RR 3.03, 95% CI 0.62-14.92). Although the trials could not be blinded to either staff or patients, risk of bias was deemed to be low.

Available data have several limitations that preclude a strong recommendation in favor of induction. For example, the definition of suspected macrosomia varied among the trials: in two trials pregnancies were included when fetal weight, estimated by ultrasound examination, was between 4000 g and 4500 g or between 4000 g and 4750 g, the third trial included pregnancies with estimated fetal weight more than the 97^th percentile at the time of inclusion, and the fourth trial included pregnancies with estimated fetal weight more than the 95^th centile. In addition, induction was performed at 37 to 40 weeks; therefore, it is unknown whether induction at 39 weeks (the preferred minimum age for induction to minimize risk of respiratory problems) would still be effective.

Patients with diabetes — In patients with pregestational diabetes, timing of birth is based primarily on the maternal and perinatal risks associated with the disease rather than concerns about shoulder dystocia alone. Timing the birth of these pregnancies is reviewed separately. (See "Pregestational (preexisting) diabetes mellitus: Obstetric issues and management", section on 'Delivery'.)

In patients with gestational diabetes and estimated fetal weight between 4000 and 4500 g at 39 weeks of gestation, we discuss the risks and benefits of induction. In such cases, the author believes that the maternal risks of induction and neonatal risks of respiratory distress and neonatal intensive care unit admission are small and balanced by the benefit of a reduction in shoulder dystocia, based on results of the two trials described below [80,81]. Specific patient factors to be considered during shared decision making include whether the estimated fetal weight lies within the 4000 to 4500 g range; glycemic control throughout pregnancy (poorer control is associated with the habitus of a newborn of a diabetic mother); prior labor, birth weight, and vaginal birth history (eg, has the patient had previous uncomplicated vaginal births of large infants?); and maternal habitus, including height, weight, and adequacy of pelvis (eg, height <160 cm, obesity, and/or a clinically small pelvis increase the likelihood of dystocia).

In the only randomized trial of induction versus expectant management of pregnancies with gestational diabetes mellitus (n = 200), induction at approximately 39 weeks resulted in fewer infants with birth weight ≥4000 g (15/100 versus 27/100) and birth weight ≥4500 g (0/100 versus 2/100), and fewer shoulder dystocias (0/100 versus 3/100 [all mild with no traumatic injuries]) [80]. A systematic review including this trial and four observational studies found that active intervention (induction, scheduled cesarean birth) was associated with reduced rates of macrosomia and related complications compared with expectant management, but the heterogeneity of study designs prevented any clear conclusions about selection of pregnancies for induction versus cesarean birth versus expectant management and the optimum gestational age for the intervention [81].

Patients with postterm pregnancy — Induction of labor at 41 weeks of gestation is common obstetric intervention, with the primary goal of reducing perinatal mortality. (See "Postterm pregnancy".)

A meta-analysis of randomized trials of induction of labor at 41 weeks of gestation compared with expectant management found that induction reduced the birth of newborns >4000 g [82]. The effect on shoulder dystocia was not reported, but one of the included trials reported that induction resulted in a lower rate of shoulder dystocia (0.3 versus 2.3 percent).

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Shoulder dystocia and macrosomia".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Shoulder dystocia (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Diagnosis – A vaginal birth is complicated by shoulder dystocia when, after expulsion of the fetal head, additional obstetric maneuvers beyond gentle guidance are needed to enable extraction of the fetal shoulders. Shoulder dystocia is an obstetric emergency because it may result in life-threatening infant injuries, as well as less serious maternal injuries. (See 'Introduction' above.)

●Pathogenesis – Newborns of mothers with diabetes are at an increased risk of shoulder dystocia because they have higher birth weights and a higher chest-to-head and shoulder-to-head ratio than newborns of mothers without diabetes. (See 'Diabetes mellitus' above.)

●Risk factors/prediction

•The occurrence of shoulder dystocia cannot be predicted accurately by antenatal risk factors, labor abnormalities, or pelvimetry/fetal biometry. Since at least 50 percent of pregnancies complicated by shoulder dystocia have no identifiable risk factors, the predictive value of any one or combination of risk factors for shoulder dystocia is low (less than 10 percent). (See 'Can shoulder dystocia be predicted?' above.)

•High birth weight is a major risk factor for shoulder dystocia and difficult to diagnose accurately antepartum. The incidence of shoulder dystocia increases as birth weight increases (table 1). However, 50 percent of cases occur in newborns <4000 g. (See 'High birth weight' above.)

•The combination of suspected high birth weight, labor abnormalities, and assisted vaginal birth has a particularly high risk for shoulder dystocia: 21 percent in one report. (See 'Assisted vaginal birth' above.)

•The incidence of recurrent shoulder dystocia ranges from 1 to 25 percent, compared with <1 percent incidence in the general obstetric population. (See 'Previous shoulder dystocia' above.)

●Candidates for cesarean birth to reduce the risk of shoulder dystocia

•We suggest scheduled cesarean birth in the following clinical settings:

-Estimated fetal weight >5000 g in patients without diabetes or >4500 g in patients with diabetes. (See 'Patients with prior shoulder dystocia or estimated weight greater than 4500 or 5000 grams' above.)

-Prior shoulder dystocia, especially with a severe neonatal injury. (See 'Patients with prior shoulder dystocia or estimated weight greater than 4500 or 5000 grams' above.)

•We suggest intrapartum cesarean birth rather than a low pelvic assisted vaginal birth in pregnancies with a prolonged second stage and estimated fetal weight >4500 g.

•We suggest intrapartum cesarean birth rather than a midpelvic assisted vaginal birth when the estimated weight >4000 g.

•In other scenarios, assisted vaginal birth can be considered case by case, based on factors such as fetal size and position, past obstetric history, maternal habitus, and operator expertise. (See 'Patients with a prolonged second stage' above.)

●Candidates for induction

•No diabetes – In patients at 39 weeks of gestation without diabetes and with estimated fetal weight between 4000 and 5000 g, we suggest induction (Grade 2C) but expectant management is also reasonable. The potential reduction in shoulder dystocia and its sequelae with induction at 39 weeks is likely to be less than with induction at 37 or 38 weeks, but common neonatal morbidities such as hyperbilirubinemia and respiratory problems are likely to be higher with induction at 37 or 38 weeks. (See 'Patients without diabetes' above.)

•Gestational diabetes – In patients at 39 weeks of gestation with gestational diabetes and estimated fetal weight between 4000 and 4500 g, we discuss the risks and benefits of induction. We believe that the maternal risks of labor induction and neonatal risks of respiratory distress and neonatal intensive care unit admission are small and balanced by the benefit of a reduction in shoulder dystocia. Specific patient factors to be considered during shared decision making include whether the estimated fetal weight lies within the 4000 to 4500 g range; glycemic control throughout pregnancy; prior labor, birth weight, and vaginal birth history; and maternal habitus, including height, weight, and adequacy of pelvis. (See 'Patients with diabetes' above.)

•Pregestational diabetes – In patients with pregestational diabetes, timing of birth is based primarily on the maternal and perinatal risks associated with the disease rather than concerns about shoulder dystocia alone. (See "Pregestational (preexisting) diabetes mellitus: Obstetric issues and management", section on 'Delivery'.)