EP16.25: Does risk of emergent delivery with placenta accreta spectrum differ by placental location?

Sonographic findings may modify week-to-week risk of emergent delivery for placenta accreta spectrum (PAS). We aimed to assess whether risk differs by presence of previa. Cohort study of all singleton viable cases of PAS at a single centre from 2001-23. Outcomes were emergent delivery (maternal/fetal instability, unsuspected intraop diagnosis, or <2 hours from decision to deliver) and expedited delivery (moved up due to clinical circumstances). Scheduled deliveries were censored within each gestational week. We report probability of emergent delivery from lifetable analysis. We compare survival function between previa vs no previa with Kaplan Meier (log-rank test); and proportional hazard models. Of 213 eligible PAS cases, 76% had previa. In all PAS cases, the cumulative probability of emergent delivery was: 6.2% for 30-31 weeks, 7.6%, 32nd week, 8.2%, 33rd week, 9.4%, 34th week, 12.5%, 35th week, 18.5%, 36th week, and 21.2% after 37 weeks. Previa was associated with ≥2 Caesareans (78% vs 49% p < 0.001), vaginal bleeding during pregnancy (63% vs 33% p < 0.001), and undiagnosed PAS before delivery (3% vs 29%, p < 0.001). Groups did not differ in demographic characteristics, disease severity by grade, or other clinical characteristics. Emergent delivery (14% of all) and emergent or expedited delivery (37% of all) did not differ by previa (see figure 1). After adjusting for disease grade, prior preterm birth, and short cervix, the associations between previa the outcomes remained non-significant; HR 1.13 [95% CI, 0.42-3.05], for emergent delivery, and 1.78 [95% CI, 0.88-3.61] for emergent or expedited delivery. While meaningful differences cannot be excluded, sonographic presence of previa in patients with PAS did not significantly alter the week-to-week risk of emergent or expedited delivery. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.