After Shoulder Dystocia: Managing the Subsequent Pregnancy and Delivery

Abstract

Among risk factors for shoulder dystocia, a prior history of delivery complicated by shoulder dystocia is the single greatest risk factor for shoulder dystocia occurrence, with odds ratios 7 to 10 times that of the general population. Recurrence rates have been reported to be as high as 16%. Whereas prevention of shoulder dystocia in the general population is neither feasible nor cost-effective, intervention efforts directed at the particular subgroup of women with a prior history of shoulder dystocia can concentrate on potentially modifiable risk factors and individualized management strategies that can minimize recurrence and the associated significant morbidities and mortality.

Introduction

The occurrence of shoulder dystocia is largely considered unpredictable. Epidemiologically, although diabetics’ infants weighing more than four kilograms at birth are disproportionately at higher risk for shoulder dystocia compared to average weight newborns,^1–5 nearly half of shoulder dystocia events occur in non-macrosomic infants of otherwise healthy women.^2;6 Although both antepartum and intrapartum risk factors associated with occurrence of shoulder dystocia are known,^1;2 a substantial number of mother-infant pairs who actually experience shoulder dystocia during delivery has none of these risk factors while many more deliver uneventfully despite having several risk factors.⁷ However, among risk factors for shoulder dystocia, a history of shoulder dystocia in a prior delivery carries a recurrence risk of 10–16%.^8–10 Akin to the single greatest predictor of preterm delivery being a history of preterm delivery in a previous pregnancy, a prior history of delivery complicated by shoulder dystocia is indeed the single greatest risk factor for shoulder dystocia recurrence, with odds ratios 7 to 10 times that of the general population.^8–10 Table 1 provides a summary of studies describing recurrence risks.^8–11

Table 1:

Summary of the literature that investigates recurrence of shoulder dystocia among cohorts of women with a recorded history of shoulder dystocia in a prior delivery.

Study (Time Period)	Background rate of SD	Proportion having pregnancy after SD N (%)	Proportion with cesarean for history of SD N (%)	Recurrence rate of SD among all subjects with history of SD (RR)	SD recurrence rate / Cesarean rate for failed trial of labor
Smith et al8 (1980–85)	0.85%	51 / 203 (25.1%)	5/51 (9.8%)	9.8% (RR-17)	11.9% / 9.5%
Lewis et al9 (1983–92)	2.0%	101 / 747 (13.5%)	---	---(RR-7)	13.8% / unknown
Ginsberg et al10 (1993–99)	1.5%	73 / 602 (12.1%)	7/73 (9.6%)	15.1% (RR-10)	16.7% / 9.6%
Gurewitsch et al11 (1993–2004)	1.9%	78 / 385 (20.2%)	4/78 (5.1%)	11.5% (RR-6.3)	12.2% / 12.8%

SD = shoulder dystocia, RR = relative risk

Unfortunately, the pathogenesis of shoulder dystocia is multivariate, screening for “at-risk” individual pregnancies is poorly predictive (positively or negatively) of its actual occurrence, and thus effective interventions for prevention are few. However, although avoidance of shoulder dystocia in the general population is neither feasible nor cost-effective,^12–14 concentrating intervention efforts on the particular subgroup of women with a prior history of shoulder dystocia has distinct merit. First, the occurrence of shoulder dystocia in a previous pregnancy indicates the need for a targeted evaluation. Second, review of the details of a woman’s specific shoulder dystocia experience can elucidate potentially modifiable conditions and circumstances amenable to intervention in a subsequent pregnancy. Third, even if shoulder dystocia recurs, complications may be minimized by individualized management aimed at reducing or controlling those factors that predispose to significant morbidity and mortality arising from shoulder dystocia. In focusing on the women with a prior history of shoulder dystocia, this review will concentrate on those aspects of shoulder dystocia and injury pathophysiology that are knowable and hence modifiable.

It is axiomatic that birth remains a moderately hazardous process and that certain unintended and unavoidable outcomes will occur even when complications are anticipated and managed appropriately. Importantly, it is the untoward outcomes of the condition and not necessarily the condition itself that we wish to prevent. Indeed, avoidance of shoulder dystocia recurrence per se should neither be expected nor set as a goal. However, its outcome may be significantly improved if those factors that increase the risk for harm are prepared for and managed. As with any emergency, logical evidence-tested response to shoulder dystocia – either before it occurs or once in progress – is directly correlated with outcome, whether it occurs for the first time or is recurrent. Unfortunately, fully tested and evidence-validated approaches to management specifically of the subsequent pregnancy and delivery of a woman with a prior history of shoulder dystocia do not exist. Nevertheless, empiric modification of delivery method or of intrapartum management for women with a history of prior shoulder dystocia appear to be commonplace;^8;10;11 however, without targeted emphasis on modifiable risk factors, such strategies have had little impact on either recurrence or morbidity.^10;11 Our goal to propose a reasonable and logical approach to the management of a subsequent pregnancy and delivery after a shoulder dystocia based on interpretation of the available literature.

Recurrence Prevention Starts Immediately After the Initial Shoulder Dystocia

Preventive strategies for the next possible shoulder dystocia occurrence begin with the proper management of the aftermath of the index event. The incidence of shoulder dystocia is increasing.^1;15;16 This may be due to better diagnoses, better documentation or physical factors such as increasing birthweight and obesity. The same is true for shoulder dystocia-associated brachial plexus injury.¹⁷ Fortunately, the majority of shoulder dystocia events are inconsequential in terms of untoward outcome. Still, up to 27% of shoulder dystocia will be complicated by some maternal and/or fetal injury, with significant proportion (up to 10%) having attendant permanent sequelae.¹⁸ Whether the original shoulder dystocia event was mild, moderate or severe, the occurrence of shoulder dystocia at any delivery is worthy of documentation and evaluation.

Documentation

Regardless of outcome, details of the shoulder dystocia – even merely that it occurred – must be communicated to the mother and, by way of documentation in the medical record, to the next obstetrician who will care for her. Ideally, each participant in the delivery’s emergency response should author a note about the event itself and their role. However, there is no substitute for the primary delivering clinician’s giving a comprehensive accounting of each aspect of the management and its results. Items important to document are listed in Table 2.

Table 2:

Documentation suggested after a delivery complicated by a shoulder dystocia that may help with management of a subsequent pregnancy.

Second stage of labor details Duration Time spent pushing Time the head was on the perineum prior to initiating traction Total head to body delivery time Details of any operative interventions Type (forceps or vacuum) Indications Station initiated Asynclitism (if present) Rotation of the head(if performed) Anesthesia details Use of any prophylactic positioning or maneuvers Position of head and side of restitution (manual or spontaneous) Timing and type of episiotomy with extensions How and when shoulder dystocia was recognized (e.g., turtle sign, with first attempt at traction) Maneuvers (e.g., prophylactic McRoberts’) and traction prior to diagnosis Evidence for compound presentation Other procedures Nuchal cord management Suctioning of oropharynx Meconium management Personnel present Description of each maneuver performed Maneuver documentation Type and sequence of maneuvers Who performed which maneuvers Traction direction and orientation of head and thorax Effectiveness of each maneuver Fetal response to each maneuver Final successful maneuver Timing of management choices Clear descriptions of right versus left shoulder and anterior versus posterior shoulder (e.g., which was initially anterior, which was manipulated directly, which delivered first) Birth weight Apgar scores Arterial and venous cord gas Infant injury (if any) Maternal injury (if any)

Acker proposed a shoulder dystocia intervention form 15 years ago that was in keeping with the idea that this degree of detail and documentation is not only important, but should be standardized.¹⁹ The merits of documenting all of this information from risk management and medicolegal perspectives are reviewed elsewhere.^20–22 However, the relevance of such detail to assessment of recurrence risk should not be underestimated. Each aspect provides potential clues to the pathogenesis of that particular shoulder dystocia and, regardless of any retrospective assessment of the specific management’s propriety or prudence, thorough review of the precise sequence of events may identify potentially modifiable contributors to recurrence and outcome for that particular patient.

Relating delivery details to pathophysiology

Shoulder dystocia results from heterogeneous processes that produce the same clinical effect. Among impaired or dysfunctional births characterized by fetopelvic disproportion, there is a continuum of obstruction wherein delivery of the fetal trunk is impeded following completed (or near completed) delivery of the fetal head. While the most common definition of shoulder dystocia requires the presence of a bony impaction preventing spontaneous delivery of the fetal trunk, its pathogenesis derives from variable contributions of both size and positional incompatibilities between fetal shoulder and maternal pelvic dimensions. Some shoulder dystocia events are caused mainly by positional misalignments of only marginally incompatible shoulder and pelvic dimensions (usually of a normal-sized infant and normal-sized pelvis). Others arise from a significant shoulder-to-pelvis size incompatibility necessarily producing a “tight fit” between either a large-for-gestational age infant (most common) and a normal-to-large pelvic cavity or a normal-sized infant and a reduced-capacity pelvis (least common).

In a normal delivery, a winding, forward-progressing motion occurs as the head and body traverse the birth canal. It is this rotational-type motion that accomplishes delivery of the shoulders following emergence of the head through the introitus. This is effectively an anatomical requirement imposed by the typical arrangement of the bony boundaries of the pelvic inlet, midpelvis and pelvic outlet, which is akin to threads of a screw. Thus, just after delivery of the head, the shoulders often must come to lie in an oblique orientation relative to the maternal pelvis before they can to descend into the midpelvis and deliver through the pelvic outlet. Shoulder dystocia occurs when the shoulders retain the anteroposterior orientation they had assumed upon the head’s initial entry (usually in occiput transverse position) into the pelvic inlet and are unable to (or have yet to) rotate spontaneously to occupy the oblique diameter of the pelvis. Subsequently, the shoulders fail to deliver from behind the maternal pubic symphysis.

Positional misalignments resulting in shoulder dystocia may also be produced by incomplete entry to the hollow of the sacrum by the posterior shoulder as the head delivers. This may occur when there is a compound presentation with the posterior arm beneath the posterior shoulder, from relative lordosis of the maternal lumbosacral spine or from a prominent sacral promontory impeding descent of the posterior shoulder. The anterior shoulder then “rides high” and impacts behind the pubic symphysis. Another cause is insufficient time allotted to permit spontaneous shoulder rotation to the normal oblique orientation relative to the pelvis. This can occur following precipitous delivery of the head produced either by increased compliance of pelvic tissues or else by instrumented deliveries. This can even occur simply by too-hasty application of traction to the head immediately upon its emergence from the birth canal (e.g., “to keep the momentum going”) without awaiting the next contraction. It is also noteworthy that shoulder dystocia is disproportionately more common among anomalous or stillborn fetuses, suggesting that physiologic negotiation of the pelvic inlet and outlet is indeed a dynamic process, facilitated by normal fetal responsiveness.¹⁶ It is likely that any of these antecedents to a given shoulder dystocia can lead to an “accidental” or dynamically-determined misalignment of otherwise compatible fetal shoulder and maternal pelvic dimensions. This latter situation is probably associated with a lower recurrence risk compared to shoulder dystocias from other causes. For these women, a recurrence may best be prevented by avoiding instrumented deliveries in the future or by patiently allowing the fetal head to spontaneously rotate to the oblique diameter, before traction is applied to the head and neck.

More severe shoulder dystocia occurs when the anterior shoulder becomes tightly impacted behind the pubic symphysis. In the most severe cases, the posterior shoulder is simultaneously impacted on the sacral promontory. The most common cause for this type of shoulder dystocia is a large-for-gestational age or macrosomic fetus, whose shoulder width is typically 14 cm or more.²³ Less commonly, the maternal pelvis may be contracted or have a greater transverse dimension than the obstetric conjugate at the pelvic outlet, as in the platypelloid pelvis. Acquired deformities of the pelvis, such as coccygeal fracture from either trauma or a previous delivery, may also cause such difficulty in delivery of the shoulders. It is these types of shoulder dystocia events – those encountered in significantly size-discrepant mother-infant dyads – that are associated more often with untoward outcome.²⁴ It is also these types of shoulder dystocia that would be more likely to recur in subsequent deliveries. The approach to such women in the next pregnancy would focus on careful monitoring of fetal growth antenatally and timely delivery for some; others with narrow or reduced pelvic capacity may be appropriately managed with planned cesarean birth. We return to these strategies later in this review.

The apparent simplicity of the above distinction between positionally-determined and size-determined pathogenesis of the index shoulder dystocia is misleading. The contributions of positional misalignment and true fetopelvic disproportion to the occurrence of shoulder dystocia at a given delivery in a given woman are variable and are by no means mutually exclusive. A grand multipara who experiences shoulder dystocia after a precipitous second stage labor and delivery of an infant comparably sized to those she has delivered several times before without incident may have been caused by an acquired pelvic deformity that prevented normal occupancy of the hollow of the sacrum by the infant’s posterior shoulder. Thus, an apparent dynamically-determined positional misalignment was also impacted upon by narrowed pelvic dimensions. Similarly, an unanticipated finding of a randomized controlled trial of forceps versus vacuum delivery was that shoulder dystocia followed the vacuum deliveries more often than it followed forceps deliveries.²⁵ One explanation for this outcome was that the vacuum may be more apt than forceps to be successfully applied to the head of a macrosomic infant already maximally occupying the pelvic cavity since, unlike with forceps, the vacuum instrument itself does not have to be accommodated between the head and the pelvic sidewalls. Thus, shoulder dystocias seemingly attributable to insufficient time for shoulder rotation following instrumented assistance with delivery of the head often also involve true size discrepancy between a large-for-gestational-age fetus and normal maternal pelvic capacity.

Retrospective evaluation of pelvic capacity and infant biometry

The contour of the pelvis undergoes modification and expansion during pregnancy and delivery by a progressive laxity that develops in the cartilage between the pubic rami and within the sacroiliac joints. Thus, the prospective de novo assessment of the likelihood of shoulder dystocia by traditional clinical or even medical image-derived pelvimetry performed in either a nullipara or someone who previously has had uneventful deliveries is bound to have a low yield. It is also probable that the simple assessment of pelvic diameters as done in traditional pelvimetry may be too crude an assessment of pelvic capacity where propensity toward shoulder dystocia is concerned. It is unknown whether volumetric (three-dimensional) rather than geometric (two-dimensional) assessment or surface-rendered (topographical) analysis of bony contours could better differentiate women with otherwise “normal” or “adequate” pelves who nonetheless may have a predilection for shoulder dystocia from similar sized pelves of women who are unlikely to have shoulder dystocia.

Similarly, not all fetuses of the same weight, even when large-for-gestational age, are equally predisposed to shoulder dystocia. The notion of macrosomia refers not simply to an arbitrary weight cut-off, but rather a disproportionately large fetal trunk relative to the head. Thus anthropomorphic measurements of an infant whose delivery was complicated by shoulder dystocia may also yield some insight into pathogenesis.

Table 3 depicts the results of traditional and novel pelvimetry and corresponding infant biometry performed within 24 hours of delivery on three mother-infant pairs who experienced shoulder dystocia.²⁶ Based on clinical history alone without benefit of postpartum pelvimetry, it would be reasonable to surmise that Patient 1 must have a “borderline” pelvis where large-for-gestational age fetuses are concerned. She was unable to deliver a 10-lb baby at all and a baby weighing just less than 9 lbs experienced shoulder dystocia during delivery. Patient 2, on the other hand, by clinical judgment and without benefit of retrospective pelvic and fetal biometry, could have been judged to be at reduced risk for developing shoulder dystocia based on her documented compliance with blood sugar management and ultrasound estimated fetal weight that was in a reasonable range.²⁷ Yet, of the three shoulder dystocia cases, hers was the most severe. Patient 3 was only modestly compliant with management of her blood sugar, as demonstrated by sonographic evidence of accelerated fetal growth by the time she reached term. However, given the extensive history of her “tested pelvis” for presumably similarly “macrosomic” (trunk-to-head disproportionate) infants in the past and the relatively mild nature of the shoulder dystocia she experienced in the context of her precipitous delivery, it might be reasonable to suspect that positional misalignment resulting from precipitous delivery was a greater factor in her shoulder dystocia pathogenesis than true fetopelvic disproportion.

Table 3:

Clinical and Computerized Tomographic Pelvimetric Data from Mother-Infant Dyads Examined Within 24 Hours of Delivery Complicated by Shoulder Dystocia

	Case 1	Case 2	Case 3
Clinical Presentation	VBAC, prior c-sec for CPD Prior infant 4,318g BMI 46.9, non-diabetic, 28-lb wt gain, EFW 3800g, 39-wk NSVD after 27-min second stage	Primigravida, insulin-requiring GDM, euglycemic during pregnancy, BMI 34.5, 33-lb wt gain, EFW 3900g, induction at 39 wk, NSVD after 70-min second stage	Grand grand multipara (10th delivery), recurrent insulin requiring GDM, largest prior infant 3,973g, no prior history shoulder dystocia, 3,476g-sono EFW @ 37 wk, precipitous delivery on arrival to hospital at 38 wks
Shoulder Dystocia Severity	Head-to-body interval 70 seconds Required McRoberts’, posterior Rubins’ and Modified Woods’ Screw	Head-to-body interval 2 minutes, Required McRoberts’, suprapubic pressure, and Woods’ Screw	Head-to-body interval 45 seconds, Required McRoberts’ and suprapubic pressure
Immediate Neonatal Outcome	Apgars 9/9, cord pH 7.3, facial bruising, no fractures or neurapraxia	Apgars 7/9, no neonatal complications	Apgars 8/9, Temporary Erb’s palsy
Birth Weight (g)	3,933	3,613	4,488
Bisacromial Width (cm)	14.33	14.83	14.67
Largest Transverse Diameter (normal > 12 cm)	15.76	12.09	11.72
Sacral Promontory to Top of Pubic Symphysis (normal > 11 cm)	11.63	10.41	11.21
Interspinous Distance (normal >10 cm)	12.47	10.57	11.18
Intertuberous Distance	14.71	13.36	13.53
Symphyseal Separation (cm)	1.82	0.62	0.64
Right / Left Sacroiliac Joint Space (cm)	0.49 / 0.42	0.34 / 0.48	0.58 / 0.62
Volume (cm3)	1,205	714	1,463
Surface Area (cm2)	2,141	1,650	2,690

Novel pelvimetry was derived using 3D Doctor^™ and reconstructing CT images in Matlab^™ via intensity-based segmentation, stacking and surface-fitting of a mesh to the images. Results were validated by comparing 3D Doctor^™ measurements against those derived by the radiologist interpreting the CT scans.

It is noteworthy that all three infants had similar bisacromial widths. Babies 1 and 3 had proportionate upper and lower body anthropomorphic dimensions, whereas Baby 2 had markedly asymmetric shoulder girth despite normal overall body weight and ponderal index, the latter of which is consistent with well-controlled gestational diabetes.²⁸ Retrospectively, the traditionally-derived sonographic estimation of fetal weight of Baby 2 obtained prior to delivery overestimated the actual birth weight, yet did not fully detect this shoulder width asymmetry that perhaps might have been detectable by additional non-standardized means.^29–32

Anecdotally, Patient 1 had a subsequent pregnancy that was scheduled for a repeat cesarean section at 39 weeks in order to avoid recurrent shoulder dystocia. However, she presented only a few hours before her scheduled operation already in rapidly progressive labor. She successfully delivered vaginally without shoulder dystocia a 3,945g infant, similar to the previous child who developed shoulder dystocia. Interestingly, the pelvimetric assessment following her initial shoulder dystocia delivery was consistent with a low recurrence risk since it suggested that a true fetopelvic disproportion was not evident. In light of the retrospective evaluation of pelvic capacity and infant biometry following this patient’s shoulder dystocia delivery, it is more probable that dynamic positional misalignment was more contributory to the pathogenesis of shoulder dystocia and her 27-minute second stage was more significant in producing this than was any presumed discrepancy in fetal shoulder and maternal pelvic dimensions based on her prior cesarean delivery indication.

By contrast, the shoulder dystocia deliveries of Patients 2 and 3 were more likely to have resulted from size discrepancy than from positional misalignment of otherwise compatible bony dimensions between mother and fetus. In the case of Patient 2, postpartum CT pelvimetry obtained after her index shoulder dystocia event revealed a reduced pelvic capacity, suggesting that she and her infant were size incompatible. Given these findings, it would seem reasonable to propose that recurrence of shoulder dystocia would be likely in this patient, and she would be a good candidate for a scheduled elective cesarean delivery in her next pregnancy

Patient 3’s postpartum CT evaluation (Figure 1) shows a rather capacious pelvic cavity by standard measurements. Given the clinical history alone, it would appear as though the shoulder dystocia was more likely “sporadic” and attributable to positional misalignment in this patient. Even though her infant was large-for-gestational age it was not “macrosomic” as defined by trunk-to-head disproportion (or even by criteria for diabetic gravida given in the 2002 ACOG Practice Bulletin on Shoulder Dystocia). Indeed, her antenatal estimated fetal weight was at the 72^nd percentile. Based on this, it might have been reasonable to surmise that recurrence of shoulder dystocia based on fetopelvic size discrepancy would be relatively low in this patient, and potentially could be modified further with improved diabetic management to avoid excess neonatal adiposity.

Figure 1: Postpartum CT Pelvimetry Following Shoulder Dystocia.

CT images were obtained from a grand grand multiparous (para 9) woman immediately following delivery, which was complicated by shoulder dystocia. a) largest transverse diameter; b) anteroposterior diameter (sacral promontory to top of pubic symphysis); image shows marked angulation of sacrococcygeal joint consistent with healed fracture; c) interspinous diameter; d) intertuberous diameter. Acquired pelvic deformity, likely sustained during earlier delivery, probably contributed to “high-riding” anterior shoulder resulting in shoulder dystocia.

However, a surprise finding of an acquired pelvic deformity from a healed coccygeal fracture (likely sustained at the delivery that immediately preceded the index shoulder dystocia delivery) suggests a different pathogenesis for shoulder dystocia. The sharp angulation of the sacrococcygeal joint (Figure 1b) likely impeded normal occupancy of the hollow of the sacrum by the posterior shoulder in the index delivery. This, in turn, would have caused the anterior shoulder to be displaced forward and cephalad and become lodged behind the pubic symphysis in an anteroposterior orientation. Such a sacrococcygeal deformity could potentially reduce the efficacy of first-line shoulder dystocia maneuvers that still maintain the shoulders in the anteroposterior orientation.³³ Additionally, the borderline small transverse diameter of her pelvic inlet, the anteroposterior orientation is perhaps more naturally predisposed to being the “path of least resistance” for fetal descent in this patient. Thus, the combination of her acquired deformity and her anthropoid-type pelvic architecture make her likely to experience recurrent shoulder dystocia in future deliveries. A planned cesarean delivery for future pregnancy should be considered. If a vaginal delivery is planned, then antenatal management should focus on better antenatal control of fetal growth and, with awareness of the potential interference by the pelvic deformity, the delivering clinician can opt to prioritize rotational type maneuvers as initial management for any recurrent shoulder dystocia, which could reduce the risk of injury.³⁴

Patient Counseling

It is often counterintuitive or seemingly premature to attempt to counsel a patient regarding the possible pathogeneses of a just-experienced shoulder dystocia or to prognosticate either about its possible recurrence or about the long-term outcome of any associated injuries. Nevertheless, it is important to explain, in a candid and honest manner, what transpired, what are the possible causes and what might be anticipated in the future and to begin to involve the patient actively in her own and her child’s future care. Perhaps precisely because there are so many unknowns about exact pathogeneses, risk factors, prognostic indicators and future management plans, it is critically important to educate the woman whose delivery was complicated by shoulder dystocia about the often unpredictable and unpreventable nature of the complication, ensure her understanding that it is a significant issue as far as future childbearing is concerned, and elicit the patient’s preferences and priorities when planning future deliveries.

Until more information is amassed through epidemiological and clinical studies, it is reasonable to encourage the patient’s cooperation in investigating possible causes and commitment to compliance with proposed management schema that would aim to ameliorate potentially modifiable contributors to risk. However, the inability to guarantee a particular outcome or to entirely avoid risk in any chosen mode of delivery should always be explained clearly. Depending on a particular patient’s degree of risk aversion, she should be encouraged to consider how willing she would be to maintain a flexible, dynamic and evolving plan for delivery throughout the next pregnancy and, perhaps even during her next trial of labor. Knowledge of the occurrence of shoulder dystocia, even if otherwise uneventful in terms of difficulty of management or associated morbidity is important for the empowerment of both the patient and the obstetrician who will deliver her next child.

The Interval Period: Before the Next Pregnancy

Follow-Up of Infant Status

If the infant had sustained an injury at the index shoulder dystocia delivery, whether skeletal or neurologic, the importance of follow-up of the infant’s condition must be stressed to the patient (in order to optimize long-term outcome¹⁸), as well as to the delivering clinician. Not only is ongoing sensitive and caring communication with the family significant from a risk management perspective,^35;36 the information is also invaluable to management of the subsequent pregnancy and delivery. Parental input is extremely important in this process.

For the delivering clinician, knowledge of whether an infant that was discharged from the newborn nursery with an as-yet unresolved brachial plexus palsy eventually recovers completely or requires surgical intervention and remains with a permanent deficit is critical to retrospective assessment of the management of the index shoulder dystocia. Whereas many brachial plexus injuries of a temporary or even mild permanent nature (e.g., restricted to the upper plexus with only mild functional deficit in the shoulder’s active range of motion) may have myriad etiologies (including malpositioning in utero),³⁷ shoulder dystocia-associated brachial plexus injuries involving all nerve roots from C5 to C8/T1 and/or avulsion of any nerve root from the spinal cord must have involved some degree of externally-applied lateral traction of sufficient magnitude and rate to produce an injury of such an extent.^38–41 Setting aside whether the degree of traction applied was within the standard of care, the contribution of externally-applied traction to the eventual outcome cannot – and should not – be denied. Rather, it behooves the clinician to learn from the experience, and familiarize him/herself with shoulder dystocia management techniques that could reduce necessary traction at any delivery.^34;42;43

Preconceptual Management of Maternal Risk Factors

Among women with a history of shoulder dystocia in a previous pregnancy, predictors of recurrence seem to vary. Significance differences, as well as lack thereof between women with and without recurrent shoulder dystocia have been found for such variables as parity at index delivery (e.g. prior successful delivery without shoulder dystocia), maternal weight or diabetic status, operative delivery and length of second stage.^8–10 However, a consistent correlation has been found with high birth weight, as well as comparative birth weight with the index pregnancy, especially in diabetic pregnancies.⁴⁴

The most recognized antepartum risk factors for shoulder dystocia are also the same risk factors for fetal macrosomia. Besides a history of shoulder dystocia in a prior pregnancy, other risk factors include maternal obesity, weight gain during pregnancy of more than 35 pounds, gestational diabetes and pregestational diabetes without vascular complications, and postdatism.²¹ As shown in Table 4, severe compared to mild shoulder dystocia, is more likely to be associated with obese gravidas weighing more than 200 pounds or who gain excessive weight during pregnancy. If a clinician is fortunate enough to have a motivated patient with a history of shoulder dystocia who wishes to reduce the risk of fetal macrosomia prior to her next pregnancy, recommendation of weight reduction may be the single intervention with the greatest impact.

Table 4:

Comparison of Risk Factors Between Mild And Severe Shoulder Dystocia⁵⁶

	Mild		Severe
	Shoulder Dystocia		Shoulder Dystocia		P-value
	N	Value	N	Value
Maternal height (cm)	266	162.9 ± 6.7	143	162.7 ± 6.8	0.82
Maternal weight (kg)	274	87.4 ± 18.4	148	93.7 ± 22.0	0.002
Weight Gain (≥15.9 kg)	250	100 (40.0)	138	81 (58.7)	0.0004
Second stage abnormality	279	125 (44.8)	154	61 (39.6)	0.30
Prolonged second stage	279	43 (15.4)	154	23 (14.9)
Precipitous second stage	279	82 (29.3)	154	38 (24.6)
Operative vaginal delivery	290	93 (32.1)	170	62 (36.4)	0.34
Forceps	290	36 (12.4)	170	13 (7.6)
Vacuum	290	47 (16.2)	170	38 (22.3)
Combined	290	10 (3.4)	170	11 (6.5)

Severe shoulder dystocia defined as have met at least one of the following three criteria: 1) head-to-body interval ≥90 seconds; 2) use of either deliberate proctoepisiotomy or direct manipulation of the fetus (e.g., rotational maneuvers, delivery of the posterior arm); and 3) evidence of neonatal depression at birth (5-minute Apgar <7 and/or arterial cord pH <7.1).

Screening for occult diabetes is also reasonable. Some women with a history of a false-positive glucose challenge test may well be predisposed to macrosomia, suggesting some degree of impaired glucose tolerance not yet manifested as overt diabetes.⁴⁵ In most instances, these women likely have some degree of a metabolic syndrome or excess adipose tissue that contributes to hyperinsulinemia, a precursor to overt diabetes. Proper nutrition and exercise are well established as effective modifiers of this effect. A commitment to sustained lifestyle changes may improve future pregnancy outcome, as well as the woman’s long-term health status overall.

Antenatal Management of the Subsequent Pregnancy

Eliciting the History and Assessing Patient-Specific Recurrence Risk

Just as it is important for the delivering clinician who managed a given patient’s shoulder dystocia to carefully document and communicate to the patient the details of the event, so too should any practitioner meeting an obstetric patient for the first time specifically attempt to elicit a history of shoulder dystocia in a prior pregnancy. The major antepartum (listed above) and intrapartum risk factors for shoulder dystocia should be assessed. Intrapartum risk factors include operative vaginal delivery and abnormal second-stage length, either prolonged or precipitous. Even if the is no history of gestational diabetes, the patient should be asked to recall whether she had had a false-positive glucose challenge test⁴⁵ or perhaps had not been screened. If the infant whose delivery was complicated by shoulder dystocia had weighed more than four kilograms at birth, this is reason enough to perform an early glucose screen at registration. A specific discussion of nutrition and monitored weight gain is warranted, especially if the previous infant was large-for-gestational age. Vigilance for impaired glucose tolerance leading to accelerated fetal growth, with dietary modification and flexible medical therapy as needed,^46;47 should be a mainstay of antenatal management throughout the next pregnancy of any such patient, especially one with a history of prior shoulder dystocia.

For those women whose early glucose screen is negative, it is important to rescreen at the appropriate gestational age (26–28 weeks). For those with a false positive one-hour 50g glucose screen (normal three-hour glucose tolerance test) early in gestation, repeat “screening” in the early third trimester should be by the diagnostic three-hour test rather than the one-hour screening test. This is based on the principle that a false-positive screening test may not be sensitive enough, when repeated in a given patient, to consistently detect the condition being screened for.

Patients who have a normal glucose screening test at 26–28 weeks who later develop either clinical or sonographic evidence of accelerated fetal growth should be considered for retesting again at 30–34 weeks since there may be delayed detection of 15% of gestational diabetics. For those women with a history of shoulder dystocia who, in the subsequent pregnancy, manifest one abnormal value on the three-hour glucose tolerance test obtained after an abnormal glucose screen at 26–28 weeks, retesting should probably occur at 30–34 weeks, regardless of growth parameters by that stage. If there is any maternal or fetal suggestion of glucose intolerance, aggressive management of diet and blood glucose is warranted.

For those women with a history of shoulder dystocia who are diagnosed with gestational diabetes in a subsequent pregnancy, it is especially important to monitor blood glucose levels at several time points each day. A low threshold for initiating oral hypoglycemic or insulin therapy may be appropriate in these gravidas as well, especially if there is evidence of accelerated fetal growth while on diet despite report of normal blood glucose levels.⁴⁷

Assessment of Fetal Growth

It is well established that the margin for error for ultrasound estimation of fetal weight is too large to be relied upon heavily for predicting delivery complications.⁴⁸ However, when there has been a previous shoulder dystocia (similar to cases with gestational diabetes), serial ultrasound measurements of the trajectory of fetal growth are often more informative than a single growth estimation for assessing the need for intervention. Particular attention should be given to evidence of asymmetrical growth using ponderal indices.

The most common way to assess for accelerated truncal growth is the head circumference-to-abdominal circumference ratio.⁴⁹ If below 0.9 near term, the risk for shoulder dystocia may be increased, even when the overall estimated fetal weight has not surpassed the 90^th percentile.

For the gestational diabetic apparently well managed on diet alone, there is some evidence to suggest that empiric initiation of insulin therapy once the abdominal circumference exceeds the 75^th percentile may curtail further accelerated fetal growth.⁵⁰ Although not studied prospectively, this may be reasonable in a gestational diabetic with a history of a previous shoulder dystocia who desires to have a subsequent vaginal delivery.

Planning the Delivery: Mode and Timing

The practice of inducing labor for “impending macrosomia” has no proven benefit. Such practices increase the rate of cesarean delivery performed for failed induction without impacting the incidence of shoulder dystocia.^51–53 This lack of evidentiary support for early “elective” inductions and the persistent inability to shrink the margin of error in ultrasound estimation of fetal weight,¹⁴ led the American College of Obstetricians and Gynecologists in 2002 to revise their Practice Bulletin entitled “Shoulder Dystocia,” raising the cutoffs for estimated fetal weight above which primary cesarean section without trial of labor may be offered – to 4500 g for the diabetic gravida and to 5000 g for the non-diabetic gravida.

However, these recommendations may not apply for some women. There is still some evidence that lower estimated fetal weight thresholds for cesarean birth may be appropriate, particularly in women with “pre-existing” predilection for shoulder dystocia.^54;55 Though this evidence is not specific to the woman with a history of shoulder dystocia, these women may benefit the most from lower thresholds. Continued research is needed to more appropriately individualize intrapartum management for such women.

Regardless of the specific estimated fetal weight cut-off used to offer a primary cesarean section, only a small percentage of parturients will ever meet these criteria. What then can be said about the strategy of elective induction of labor in an attempt to ensure a lower birth weight than would result from expectant management and, thereby, potentially avoid shoulder dystocia? While ineffective as a strategy when applied to the general population, this may have merit in the gravida with a history of prior shoulder dystocia. First, since she is already parous, her risk of failed induction is significantly lower compared to her nulliparous counterpart. Second, postdatism is among the potentially modifiable antepartum risk factors for shoulder dystocia that would be eliminated by elective induction at term. Third, the risk of recurrence of shoulder dystocia correlates with similar or greater birth weight at subsequent delivery compared to the index shoulder dystocia delivery.¹⁰ Finally, and most importantly, it is the severe shoulder dystocia, which is more likely to occur in obese gravida with macrosomic infants,⁵⁶ that is most predictive of subsequent injury,⁵⁷ and that we would most wish to prevent or mitigate. However, these severe shoulder dystocias occur in the same type of patient in whom complications of cesarean delivery are also more likely to occur. Therefore, a strategy of early induction of labor at term, prior to development of substantial trunk-to-head size discrepancy, may present a balanced alternative to elective primary cesarean delivery in these women.

Anticipating Shoulder Dystocia Recurrence

Simulation-Based Training and Rehearsal

Since the exact threshold for permanent injury in in vivo shoulder dystocia is not known, the goal in shoulder dystocia management should be to reduce uterine force and clinician traction as much as possible. Maternal pushing and uterine forces can only be controlled in a limited way; these also have limited contribution to injury.⁵⁸ Therefore, training and on-going research must focus on how clinician-applied traction might be reduced.⁴³ This is especially important for shoulder dystocia management because the very natural, unconscious response when attempting to deliver the fetus once first attempts have failed is to increase traction on subsequent attempts.^33;43;59 This can be accomplished through simulation-based training and with drills.

Already addressed in other fields of medicine, obstacles to comparative research and provider training within the clinical setting are best overcome by use of medical simulation.^60;61 For surgically-oriented skills including vaginal delivery techniques, there is no substitute for mechanical simulation to allow haptic feedback and biofidelic learning for the student obstetric provider.^62;63 Real time measurement of actual clinician-applied forces during simulated shoulder dystocia deliveries – where associated mechanical fetal response can be measured prospectively in a fetal model – may enable clinicians to self-assess more accurately the magnitude, direction and rate of traction they apply during delivery.⁶⁴

Simulation-based experimentation has also proven the hypothesis that rotational maneuvers to resolve shoulder dystocia require with less applied force than McRoberts’ positioning. Indeed, brachial plexus stretch is also reduced by a factor of three with use of Rubin’s maneuver as an initial approach to shoulder dystocia management.³⁴ The comparative advantage of fetal maneuvers for reducing brachial plexus strain was also predicted computationally.⁶⁵ Clinicians are enjoined not to fear fetal manipulation, but to familiarize themselves with the techniques by availing themselves of simulation-based training and even to practice routine assessment of fetal shoulder position by direct palpation at every delivery.⁶⁶

The likelihood of favorable outcome of shoulder dystocia is maximized by preparation and coordination of a well-rehearsed response by all members of the health care team. Such resource utilization and management is best ensured by systematic rehearsal.⁶⁷ Communication and continuous feedback to team members regarding progress or lack thereof is essential to this effort. Otherwise, cross-purposeful actions of two or more team members can, albeit unintentionally, increase the risk for injury. Important components of shoulder dystocia management include a continuous assessment of the success or failure of each maneuver, employing an alternative maneuver within approximately 30 seconds of a previous failed maneuver,⁶⁸ and maintaining a calm and unhurried approach by the primary clinician. Each of these elements, as well as how and when to communicate with the family during and after a shoulder dystocia, can be rehearsed during drills until they progress smoothly.

Conclusion

Considered one of the greatest fears of obstetric providers, associated with considerable risk for injury to both mother and fetus and fraught with potential liability for the clinician, severe shoulder dystocia is an emergency that no one would care to relive. Thus, after a woman has experienced the complication, managing the risk of its recurrence in a subsequent delivery is desirable and prudent. Indeed, unlike other sporadically occurring and unpredictable complications of pregnancy, recurrence of shoulder dystocia is not infrequent. Yet shoulder dystocia’s definitive prevention, namely cesarean delivery, while expedient and facile is also costly and potentially risky^54;69 especially in the obese and diabetic gravida. Since many women choose to attempt a vaginal birth after a shoulder dystocia, it is important to use what information there is in the literature to form a rational and reasonable management plan for risk modification and for management of a shoulder dystocia event should it recur. At least some of the risk of shoulder dystocia recurrence indeed may be quantifiable and potentially modifiable during subsequent pregnancies of individual women with a history of shoulder dystocia. Importantly, avoidance of shoulder dystocia recurrence per se should neither be expected nor used a measure of success in management. As with other medical and obstetric conditions at higher risk for morbidity and mortality, it is the untoward outcomes of shoulder dystocia and not necessarily shoulder dystocia itself that we wish to prevent. Thus, circumstances and conditions that increase the risk of injury from shoulder dystocia should be targeted and if there is recurrence, the goal becomes the atraumatic resolution of the shoulder dystocia.⁷⁰