RESUMO
Shoulder dystocia is defined as vaginal cephalic delivery that requires additional obstetric maneuvers to deliver the fetus after the head has been delivered and gentle traction has failed. A bigger difference between the transverse abdominal diameter (TAD) (abdominal circumference [AC]/π) and biparietal diameter (BPD) (TAD-BPD) has been reported as a risk factor for shoulder dystocia in different countries; however, it remains unclear if this relationship is relevant in Japan. This study aimed to clarify the association between TAD-BPD and shoulder dystocia after adjusting for potential confounding factors in a Japanese cohort. We retrospectively examined 1,866 Japanese women who delivered vaginally between 37+0 and 41+6 weeks of gestation at the University of Yamanashi Hospital between June 2012 and November 2018. The cutoff value of TAD-BPD associated with shoulder dystocia and the association between TAD-BPD and shoulder dystocia were evaluated. The mean maternal age was 32.5±5.3 years; the patients included 1,053 nulliparous women (57.5%), 915 male infants (49.0%), 154 women with gestational diabetes mellitus (GDM) (8.3%), and 5 infants with macrosomia (0.3%). The mean TAD-BPD was 9.03±4.7 mm. The overall incidence of shoulder dystocia was 2.4% (44/1866). The cutoff value to predict shoulder dystocia was 12.0 mm (sensitivity, 61.4%; specificity, 73.8%; likelihood ratio, 2.34; positive predictive value, 5.4%; negative predictive value, 98.8%). We then used a multivariable logistic regression analysis to examine the association between TAD-BPD and shoulder dystocia while controlling for the potential confounding factors. In multivariate analyses, TAD-BPD ≥12.0 mm (adjusted odds ratio [OR], 4.39; 95% confidence interval [CI], 2.35-8.18) and GDM (adjusted OR, 3.59; 95% CI, 1.71-7.52) were associated with shoulder dystocia. Although TAD-BPD appears to be a relevant risk factor for shoulder dystocia, sonographic fetal anthropometric measures do not appear to be useful in screening for shoulder dystocia due to a low positive predictive value.