Multifetal pregnancies are often complicated by obstetrical conditions that lead to preterm delivery, placing the neonates at risk of prolonged hospitalization, serious morbidity, and mortality. Typically, all of the fetuses of a multiple gestation are delivered within a short interval of one another. However, in selected cases, the preterm birth of one sibling may not require delivery of the other fetus(es), who may remain in utero for an extended period, thereby improving their chance of survival and decreasing morbidity among the survivors. These are referred to as delayed-interval deliveries. Delayed-interval deliveries have been the subject of numerous case reports and reviews [1-10].
A variety of techniques have been proposed to select and manage these unique and challenging pregnancies. No strategy has been subjected to a randomized controlled trial. The clinician's intent to provide the best possible outcome in each circumstance precludes achieving high order scientific evidence on which to base any recommendations to families.
Several reviews on the technique of delayed-interval delivery have been published [11-14]; many of the same case reports are included in all of the manuscripts. There is a large ascertainment bias in the publication of these cases; successful outcomes often are reported while many unsuccessful cases go unreported. Three representative studies are illustrated below:
- An analysis of the 1995 to 1998 United States Matched Multiple Birth File database identified 200 twin pregnancies in which the first twin was live-born between 17 and 29 weeks of gestation and the second twin was delivered two or more days later [15,16]. The median duration of delay was six days (range 2 to 107 days). These delayed interval deliveries were matched to nondelayed twins presenting at the same gestational ages. One-year survival rates in the delayed delivery group were 56 percent compared to 24 percent for the nondelayed twins. The authors concluded that a strategy of delayed interval delivery resulted in improved survival and less neonatal morbidity than would otherwise be anticipated with the standard approach .
- A subsequent large series reported on 17 years of experience with delayed-interval delivery from the same center utilizing a uniform protocol . In this study, 38 twin and 12 triplet pregnancies underwent asynchronous delivery. In contrast to our own practice protocol (described below), the authors did not perform cerclage or routine amniocentesis for the retained sibling, and monochorionic twins and triplets were considered appropriate candidates for delayed interval delivery. Anti-D immune globulin was given to Rh(D)-negative mothers after delivery of the first twin/triplet, as indicated. Delayed interval delivery was useful in some settings: perinatal outcomes of second twins were improved when the first sibling was delivered between 20 and 29 weeks.
- Another series from one institution also reported significantly increased survival in the retained sibling(s) . Maternal morbidity was high (32 percent), similar to that in our own and other reports. These authors did not use a uniform approach to management and noted that amniocentesis was not always successful in excluding maternal morbidity.
Most authors recommend placement of a cerclage [19-25]. An important reason to consider cerclage is that silent cervical dilatation may have played a role in the early birth of the first sibling. If this is a possibility, then cerclage may forestall the birth of the remaining sibling(s). If a different mechanism of the early birth is apparent (eg, placental abruption, labor unrelated to cervical insufficiency, labor related to fetal demise), cerclage is unlikely to be useful. There are numerous reports of successful delayed-interval deliveries without surgical intervention [17,26-28]. Of note, women with a cerclage placed prior to consideration of delayed-interval delivery have poorer outcomes; these pregnancies are often complicated by intraamniotic infection ultimately necessitating delivery.