Objective: To determine the reasons for cesarean section (C/S) for the second twin following vaginal delivery of twin A, and causes of neonatal death (NND).
Study design: Five hundred forty-one twin deliveries were studied retrospectively from 1987 to 1995. Deliveries were fractionated by fetal presentation for twin A and twin B (i.e., vertex A/vertex B, vertex A/breech B, vertex A/transverse lie B, breech A/vertex B, ... transverse lie A/transverse lie B, yielding nine groups). The variation in C/S rate was noted with respect to fetal presentation for twin A and twin B. The indications for C/S of twin B following successful delivery of twin A were noted, as were sources of NND.
Results: The C/S rate rose as the presentation of twin A changed from vertex to breech to transverse lie (13.8%, 67.4% and 100%, respectively, for twin B vertex). A similar trend was found for twin B as breech or transverse lie. The overall C/S rate was 34.6%, and of these, 27 were solely to deliver twin B, for 5.0% of all twins and 14.4% of all C/S. The risk for C/S for the second twin was increased 7.6x if twin A was vertex rather than breech. The prime reasons for C/S of twin B varied with the presentation of twin B. Cord prolapse of twin B was most common for vertex A/vertex B (7/8), whereas inability to turn and extract twin B was most common for vertex A/transverse lie B (back down) (9/14). The other two reasons for C/S were fetal distress of twin B (two) and abruption (two). Two cases of neonatal demise resulted from asphyxia: one due to cord prolapse (vertex A/vertex B), and one due to traumatic internal version and extraction (vertex A/transverse lie B).
Conclusion: C/S delivery for the second twin is most common in vertex twin A pairs since it is these that are generally allowed to be delivered vaginally until an untoward event complicates the delivery of twin B. Hence, 96% of these C/S deliveries occur when twin A is vertex. The "safest" configuration (vertex A/vertex B) results in 26% of cases delivered by C/S for cord prolapse of twin B, while 52% of C/S deliveries are for change in presentation of twin B, with inability to perform internal podalic version and extraction. These two indications accounted for 81.5% of C/S and all the neonatal deaths.