Incidence, risk factors and maternal outcomes of unsuspected placenta accreta spectrum disorders: a retrospective cohort study

Abstract

Background: To identify incidence and underlying risk factors for unsuspected placenta accreta spectrum (PAS) and compare the maternal outcomes between suspected and unsuspected cases in three large academic referral centers. Methods: A retrospective cohort study was conducted in three university-based tertiary referral centers from Jan 1st, 2013, to Dec 31st, 2022. All cases of PAS confirmed by pathology were included in the study. Unsuspected PAS cases were diagnosed at the time of delivery, while suspected cases served as the control group. Potential risk factors were compared between the two groups. Multivariable regression model was also performed to identify risk factors. Maternal outcomes were also evaluated. Results: A total of 339 pathology-confirmed PAS cases were included in the study out of 415,470 deliveries, of which 35.4% (n = 120) were unsuspected cases. Unsuspected PAS cases were 7.9 times more likely to have a history of intrauterine adhesions (adjusted odds ratio [aOR] 7.93; 95% confidence interval [CI] 2.35–26.81), 7.0 times more likely to have a history of clinically confirmed PAS (aOR, 6.99; 95% CI 2.85–17.18), 6.3 times more likely to have a posterior placenta (aOR, 6.30; 95% CI 3.48–11.40), and 3.4 times more likely to have a history of placenta previa (aOR, 3.41; 95% CI 1.18–9.82). On the other hand, cases with gravidity > 3, placenta previa, and/or a history of previous cesarean delivery were more likely to be diagnosed antenatally (aOR 0.40, 0.19, 0.36; 95% CI 0.22–0.74, 0.09–0.40, 0.19–0.70). Although the suspected PAS group had a higher proportion of invasive cases and abdominal and pelvic organ injuries (74.4% vs. 25.8%, p < 0.001; 6.8% vs. 1.7%, p = 0.037), the maternal outcomes were more favorable in the sPAS group, with a lower median volume of 24-hour blood loss and blood product transfusion (estimated blood loss in 24 h, 1000 [800–2000] vs. 2000 [1400–2400], p < 0.001; RBC unit transfusion, 0 [0-800] vs. 800 [600–1000], p < 0.001; fresh-frozen plasma transfusion, 0 [0-450] vs. 600 [400–800], p < 0.001). Conclusions: Our findings indicate that 35% of patients with PAS were unsuspected prior to delivery. Factors associated with PAS being unsuspected prior to delivery include a history of intrauterine adhesions, a history of clinically confirmed PAS, a posterior placenta, and a history of placenta previa. Additionally, gravidity > 3, a history of previous cesarean delivery, and placenta previa increase the likelihood of antenatal diagnosis.