588: Placenta accreta spectrum FIGO clinical classification in association with histopathological findings

Abstract

Placenta accreta spectrum (PAS) covers a wide spectrum of placental adherence/ invasion with varied clinical significance. Histopathological examination is considered the confirmatory gold standard but is only obtained sometime after definitive treatment. The International Federation of Gynecology and Obstetrics (FIGO) has recently published a new clinical classification that can be assigned at delivery, and we aimed to investigate the association between this new FIGO classification and histopathology. We studied a retrospective cohort of 185 subjects with histopathology proven PAS managed at our referral center between September 2012 and January 2019. Two experienced surgeons retrospectively reviewed charts and assigned the FIGO grading based on findings reported at delivery. A third experienced reviewer adjudicated to determine classification used for final analysis. Subjects were further subcategorized into accreta/increta vs. percreta and FIGO grade 2-3 vs. grade 4-6 since FIGO grading above 4 represents invasion through the uterine serosa correlating with placenta percreta. Among 185 subjects, there were 41 (22%) placenta accrete, 44 (24%) increta, and 100 (54%) percreta on histopathology. FIGO grades included 72 (39%) grade 2-3, and 113 (61%) grade 4-6. The interrater reliability was found to be substantial with Kappa = 0.661 (p < 0.001), 95% CI (0.449, 0.872). There was a significant association between all histopathology groupings and the FIGO clinical classification, P< 0.001, in particular, percreta was associated with FIGO clinical classification 4-6 (P< 0.001), as 95 (84%) of subjects with FIGO clinical classification 4 and above had placenta percreta. The new FIGO clinical classification is strongly associated with histopathologic findings. We have demonstrated that with robust documentation and detailed record review, and bring mindful of the inherent limitations, retrospective assignment of FIGO class may be feasible. Reporting of FIGO Classification at the time of delivery may be important in reducing bias in future clinical studies.