Our study has shown that the median latency to delivery is 3 weeks for pre-viable PPROM and around 2 weeks for cases of PPROM between 24 and 32 weeks. Additionally, we observed that all cases of placental abruption and cord prolapse occurred in women that had membrane rupture before 28 weeks of gestation, and that neonatal survival at discharge increased by more than 33% when PPROM occurred after 24 weeks of gestation compared to earlier gestational ages. The occurrence of placental abruption was significantly associated with the latency to delivery (median latency 32 days in women with abruption vs 12 days in women without abruption, p = 0.032).
Closer examination of latency at different gestational ages at PPROM in our study group revealed that the median latency tends to be longer in cases of previable PPROM compared to PPROM occurring later, although the difference was not statistically significant. Many previous studies have suggested an increased latency with earlier PPROM [4, 5, 16–18]. In previable PPROM, we observed a median latency of 22 days, which is consistent with the findings by Manuck et al. [19]. On the other hand, in their review of 15 studies on previable PPROM, Sim et al. reported that the median latency ranged between 7 and 49 days, which reflects the heterogeneity between studies [3].
The median latency that we observed among PPROM after 24 weeks (13.5 days for all PPROM occurring between 24 and 31+ 6 weeks) is comparable to that reported in the study by Baser et al. [18], who observed a mean latency of 15 days in the whole study cohort, although they included cases of PPROM up to 35 weeks of gestation. When comparing different groups of gestational age, the median latency observed in the 24–28 weeks PPROM group in our study is similar to the mean latency reported by Baser et al [18] in the same GA interval. However, in the 24–28 weeks group, they observed a much longer mean latency of 24 days (compared to a median of 11 days in our cohort). Nonetheless, the authors reported the mean value, along with a large standard deviation [18]. Given the non-parametric distribution of the data on latency, we opted for median and IQR, limiting the ability to compare with the results of that study.
Regarding neonatal outcomes, the rate of neonatal survival at discharge in cases of previable PPROM, where the mother delivered at a mean GA of 25 weeks, was 58%, which is higher than that reported by Sim et al. (44.9%) [3] and closer to the data by Manuck et al. [20]. This improvement likely reflects advancement in neonatal intensive care unit techniques. In fact, our data collection covers the last 5 years, while the studies reviewed by Sim et al. included women with PPROM between 1988 and 2014.
In cases of PPROM occurring after 24 weeks, neonatal survival at discharge in our cohort exceeded 90%, and it reached 100% for PPROM after 28 weeks. Such survival rate is higher than that reported in older studies [21] and likely reflects advances in obstetric and neonatal care.
In terms of adverse obstetric outcomes, the overall incidence of placental abruption in our cohort was 8%, which is higher than that reported by Major et al.(5%) and Ananth et al. (2%) in women with PPROM [10, 22]. This difference can be attributed to the fact that those studies included all cases of membrane rupture before 37 weeks of gestation. Since our analysis focused on early-, very early-, and previable PPROM, we anticipated a higher incidence of such outcome. Indeed, all cases of placental abruption (7/86) and cord prolapse (6/86) in our cohort occurred in women who experienced PPROM before 28 weeks.
The high incidence of cord prolapse and cesarean section in PPROM that occurred before 28 weeks may be partly attributed to fetal malpresentation and fetal heart rate abnormalities, respectively, which are more common at earlier gestational ages compared to later ones. Although the difference in the incidence of these outcomes between PPROM occurring before and after 28 weeks was not statistically significant, this may be attributed to the limited number of cases.
The overall rate of clinical chorioamnionitis in our cohort, at 20%, is similar to that observed in the study by Baser et al. [18] and higher than that reported in the study by Ramsey et al [23]. This difference may be explained by the lower mean gestational age at PPROM in our cohort (26.1 weeks) compared to that study (32.4 weeks). It has been previously observed that the incidence of chorioamnionitis increases with decreasing gestational age at PPROM [23]. Differently from other studies, we also reported the rate of chorionamnionitis upon histologic examination, which was more than 2 times higher than the rate of chorionamnionitis suspected clinically. Among cases of PPROM before 28 weeks, the rate of histologic chorioamnionitis was notably high, detected in almost half of the cases in our study.
The strengths of the present study include the relatively large sample size, considering the low incidence of PPROM < 32 weeks (less than 1% of all deliveries in our hospital during the study period), and the fact that all cases were managed in a single tertiary center, with a uniform management based on the hospital’s protocol. The stratification of outcomes based on the timing of PPROM is another strength, as it provides specific information allowing for more personalized counselling for patients regarding maternal and neonatal outcomes.
This study also has some limitations, including its retrospective design, the absence of data on cases of PPROM that opted for elective termination of pregnancy before viability, which represents a common inherent selection bias in studies on PPROM, and the lack of neonatal long-term follow-up data.