Materials and Methods
We conducted a retrospective cohort study of all women who experienced a clinically diagnosed shoulder dystocia requiring obstetric maneuvers at term (≥37 weeks) at a single tertiary care hospital from 2005 through 2008. We excluded women with a known major fetal anomaly, intrauterine death, multiple gestation, and preterm gestational age. This study was approved by the Washington University Human Research Protection Office.
The comparison groups were defined by exposure to a particular maneuver including McRoberts/suprapubic pressure only, Rubin maneuver, Wood’s screw maneuver, or delivery of the posterior arm. Based on the ACOG recommendation that McRoberts and suprapubic pressure be the initial maneuvers attempted in a shoulder dystocia, we chose our reference group in this study to include all of those patients whose dystocia was resolved by McRoberts/suprapubic pressure alone. With the exception of the reference group, some women were exposed to >1 additional maneuver.
The primary outcome was a composite morbidity of neonatal injury and neonatal depression. We defined neonatal injury as the occurrence of clavicular or humeral fracture, or brachial plexus injury. Neonatal depression included any of the following: Apgar <7 at 5 minutes, arterial cord pH <7.1, continuous positive airway pressure use, intubation, or respiratory distress. Respiratory distress was defined as the requirement of oxygen support after 6 hours of life or any need for mechanical ventilation. Beyond those neonates delivered by McRoberts/suprapubic pressure only, neonatal morbidity was included under each maneuver to which the neonate was exposed.
Extensive data were extracted from the maternal and neonatal medical records to obtain sociodemographic information, medical and antenatal history, and neonatal outcomes. Data on the shoulder dystocia including duration of the shoulder dystocia defined as time from delivery of the fetal head to delivery of the fetal shoulders in seconds, and obstetric maneuvers performed were collected from a standardized delivery record. Baseline characteristics were compared between the reference group of McRoberts/suprapubic pressure only and women who required the use of an additional maneuver. Categorical variables were compared using the χ 2 or Fisher exact test, as appropriate. Continuous variables were assessed for normality using Kolmogorov-Smirnov test. Variables that were not normally distributed were compared with the Mann-Whitney U test and normally distributed variables were compared using the Student t test. The relationship between number of advanced maneuvers utilized and duration of shoulder dystocia was secondarily explored.
Odds ratios (ORs) were calculated for the composite outcome comparing each of the exposure groups (delivery of posterior arm, Rubin maneuver, or Wood’s screw maneuver) to the reference group (McRoberts/suprapubic pressure only). Logistic regression was used to control for confounding factors. A backward stepwise approach utilizing the likelihood ratio test to assess the impact of each covariate on the model was used. Duration of shoulder dystocia was included as a continuous variable in the model because there was a linear relationship between duration and morbidity as demonstrated by plotting the predicted probability of the composite outcome against time. The final model adjusted for nulliparity and duration of shoulder dystocia.
The relationship between duration of shoulder dystocia and risk of the composite morbidity was further assessed and presented visually by Kaplan-Meier curve. Differences in the duration of shoulder dystocia in pregnancies with and without the composite morbidity were compared using the log rank test.
All analyses were performed using Stata Special Edition 12.1 (StataCorp, College Station, TX).
Results
Of 8390 deliveries in the study period, 231 met the inclusion criteria. Of those, 135 were delivered by McRoberts/suprapubic pressure alone (57.9%). In all, 83 women were exposed to Rubin maneuver, 53 were exposed to Wood’s screw maneuver, and 36 were exposed to delivery of the posterior arm.
Women who were delivered by McRoberts/suprapubic pressure alone did not differ by parity, gestational age, prevalence of diabetes (either pregestational or gestational), rate of operative vaginal delivery, or infant weight >4000 g as compared to women delivered using additional maneuvers ( Table 1 ). However, women delivered by McRoberts/suprapubic pressure alone did have a significantly shorter duration of shoulder dystocia than women delivered by one of the other maneuvers with a median duration of 29 seconds (interquartile range, 29–30 seconds) as compared to 60 seconds (interquartile range, 40–90 seconds) ( Table 1 ).
| Characteristic | McRoberts and/or suprapubic only, n = 135 | Advanced maneuvers, n = 96 | P value |
|---|---|---|---|
| Median gestational age, wk | 40 (39–40) | 39 (38–40) | .11 |
| Nulliparity | 42 (31.1) | 28 (29.2) | .75 |
| Advanced maternal age | 11 (8.1) | 9 (9.4) | .74 |
| Race | .63 | ||
| Black | 89 (65.9) | 62 (64.6) | |
| White | 33 (24.4) | 21 (21.9) | |
| Other | 13 (9.6) | 13 (13.5) | |
| Body mass index ≥30 kg/m 2 | 91 (69.5) | 64 (69.6) | .99 |
| Diabetes | 12 (8.9) | 10 (10.4) | .70 |
| Regional anesthesia | 115 (85.1) | 83 (86.5) | .79 |
| Operative vaginal delivery | 24 (17.8) | 18 (18.9) | .82 |
| Infant weight >4000 g | 37 (27.4) | 26 (27.1) | .96 |
| Median duration of shoulder dystocia, s | 29 (29–30) | 60 (40–90) | < .01 |
Secondarily, we explored the relationship between number of advanced maneuvers utilized and duration of dystocia. We found that the median duration of dystocia increased with the utilization of an increasing number of maneuvers ( P < .01) ( Table 2 ).
| Variable | n (%) | Median duration, s | Interquartile range, s |
|---|---|---|---|
| McRoberts and/or suprapubic only | 135 (58.4) | 29 | 29–30 |
| +1 additional maneuver | 43 (18.6) | 60 | 29–60 |
| +2 additional maneuvers | 30 (13.0) | 60 | 42.5–80 |
| +3 additional maneuvers | 23 (10.0) | 90 | 60–157.5 |
The neonatal morbidity composite occurred in 65 of the 231 patients (28.1%). The rate of neonatal morbidity was higher with additional maneuvers compared with McRoberts/suprapubic pressure alone (22.2% vs 50.0% for delivery of the posterior arm, 36.1% for Rubin maneuver, and 39.6% for Wood’s screw maneuver). In unadjusted analysis, each type of additional maneuver appeared to be associated with an increased risk of the composite outcome relative to the reference group delivered by McRoberts/suprapubic pressure alone: relative risk (RR), 2.25 (95% confidence interval [CI], 1.42–3.54) for delivery of the posterior arm; RR, 1.63 (95% CI, 1.06–2.49) for Rubin maneuver, and RR 1.78 (95% CI, 1.13–3.03) for Wood’s screw maneuver). However, after controlling for nulliparity and duration of shoulder dystocia, these differences were no longer statistically significant: adjusted OR (aOR), 1.77 (95% CI, 0.54–5.79) for delivery of the posterior arm; aOR, 1.36 (95% CI, 0.63–2.93) for Rubin maneuver; and aOR, 1.17 (95% CI, 0.45–3.03) for Wood’s screw maneuver ( Table 3 ).
