The trusted source for
healthcare information and
Professor, Department of Obstetrics and Gynecology, University of Colorado School of Medicine, Aurora
Dr. Hobbins reports no financial relationships relevant to this field of study.
SYNOPSIS: An individual participant data meta-analysis from Australia suggests that expectant management of patients with premature rupture of membranes between 34 and 36 weeks, compared with immediate intervention, results in comparable levels of composite neonatal adverse outcomes but in mixed maternal adverse outcomes that balance out in the final analysis.
SOURCE: Quist-Nelson J, de Ruigh AA, Seidler AL, et al. Immediate delivery compared with expectant management in late preterm prelabor rupture of membranes: An individual participant data meta-analysis. Obstet Gynecol 2018;131:269-279.
Preterm (before 37 weeks) premature (before labor begins) rupture of the membranes (PPROM) occurs in 3% of pregnancies. In these pregnancies, the clinician faces a conundrum regarding whether to move forward with the delivery or to allow further time for in utero maturity. Based on all the information at hand, in the absence of infection, the American College of Obstetricians and Gynecologists (ACOG) recommended in its recent practice bulletin that expectant management be employed when PPROM happens prior to 34 weeks.1 However, after that time the authors favored delivery. The thrust to deliver mostly has been to avoid neonatal sepsis, but many other potential morbidities fall into play with either option.
Investigators from Australia searched the literature for randomized, controlled trials (RCTs) that would provide individual participant data for meta-analysis. They only included those trials in which individual patient data could be mined from the published papers or obtained directly from the authors.
The primary outcome was a composite of immediate adverse neonatal outcomes, as well as neonatal sepsis, necrotizing enterocolitis, respiratory distress syndrome (RDS), or neonatal death. Secondary outcomes involved individual contributions to the composite. Maternal outcomes included need for cesarean delivery, antepartum hemorrhage, endometritis, and length of hospitalization. All patients in the meta-analysis had vaginal cultures for a variety of agents, including gram-positive beta Streptococcus (GBS).
Eight trials were eligible but five did not have individual patient data available, leaving three to yield adequate information on 2,563 patients with PPROM at 34 to 36 weeks. Of these, 1,289 patients were randomized to have immediate delivery and 1,281 had expectant management up until 37 weeks, after which they were delivered. Seventy-six percent of the “immediate” group and 78% of the “expectant” group received antibiotics.
Regarding the primary outcome, there were no significant differences in composite neonatal adverse outcome (9.6% vs. 8.3%; relative risk [RR], 1.20; 95% confidence interval [CI], 0.94-1.55). Also, no differences were noted in overall neonatal sepsis (2.6% vs. 3.5%; RR, 0.74; 95% CI, 0.47-1.15). However, immediate delivery resulted in significantly higher rates of RDS and hyperbilirubinemia. There were no differences in perinatal deaths, but immediate delivery resulted in higher rates of admission to the neonatal intensive care unit (NICU) and longer hospital stays. Mothers having immediate delivery were less likely to have antenatal hemorrhage and chorioamnionitis but had a higher cesarean delivery rate. Interestingly, in those with positive vaginal cultures at randomization, immediate delivery was less likely to be associated with neonatal sepsis (6.5% vs. 23%; RR, 0.35; 95% CI, 0.14-0.86). However, in the 15% and 17% of those who were culture-positive for GBS, there were no significant differences in neonatal sepsis between groups.
Who would have thought that PPROM was so complicated? We still ponder what causes it and wrestle with how to handle it. The obvious answer to the first question is that something weakens the membranes. However, there is no single “something” responsible. Studies certainly have shown that infection with a resulting increase in inflammatory cytokines, such as tumor necrosis factor and interleukins 1 and 6,2 is involved. Also, elevations of metalloproteinase,3 an enzyme found in amniotic fluid and fetal blood, imply that the fetus plays at least some role in PPROM. Finally, as far back as 2002, researchers found a failure of normal trophoblastic invasion of the myometrial portion of spiral arteries in the basal plate of placentas in PPROM, like that seen in preeclampsia.4 This suggests that a disruption of placentation occurs long before the membranes rupture. So, there is more to the problem than simply mechanically weak membranes.
On to the question of what to do about PPROM when it happens. Earlier studies showed that up until 34 weeks, watchful waiting in PPROM was associated with better outcomes than early intervention. However, between 34 and 36 weeks, opinions vary. The featured meta-analysis found no difference in the authors’ primary outcomes, including overall neonatal sepsis, although the outcomes of admissions to the NICU, hyperbilirubinemia, and RDS were seen more commonly with early intervention. In contrast, women in the expectant group had higher rates of chorioamnionitis and antepartum hemorrhage, but had a lower cesarean delivery rate. Overall, the outcome differences between the two methods seem to cancel out each other, resulting in a standoff. This implies that the move to intervene in PPROM between 34 and 36 weeks, as advocated by ACOG, now can be challenged by this meta-analysis, which points toward watchful waiting as a viable option. For brevity, this Alert will not deal with steroids, tocolytics, or antibiotics, but here are some suggestions for management of PPROM:
In some clinical scenarios, it is useful to know the approximate length of time between membrane rupture and when spontaneous labor tends to ensue. In general, the earlier in pregnancy PPROM occurs, the longer the latent period. However, a lesser-known study has shown that if the cervical length, assessed with transvaginal ultrasound, is < 2.0 cm, the average lag time to spontaneous labor is 59 hours, compared with 10 days if the cervical length is > 2.0 cm.5
Financial Disclosure: OB/GYN Clinical Alert’s Editor Jeffrey T. Jensen, MD, MPH, reports that he is a consultant for and receives grant/ research support from Bayer, Merck, ContraMed, and FHI360; he receives grant/research support from Abbvie, HRA Pharma, Medicines 360, and Conrad; and he is a consultant for the Population Council. Peer Reviewer Catherine Leclair, MD; Nurse Planners Marci Messerle Forbes, RN, FNP, and Andrea O’Donnell, FNP; Editorial Group Manager Terrey L. Hatcher; Executive Editor Leslie Coplin; and Editor Jonathan Springston report no financial relationships relevant to this field of study.