Abstract
Morbidly adherent placenta (MAP) is a growing concern currently and is still a diagnostic challenge for obstetricians. As emergency hysterectomy due to unscheduled delivery in MAP carries significant risks, we aimed to evaluate whether first and second trimester serum analytes may be used in the prediction of MAP requiring hysterectomy. A retrospective chart review of all identified cases of placenta previa totalis with and without MAP was performed. A total of 316 pregnant women diagnosed as placenta previa totalis were identified and included in the analysis. Cases were examined in three groups (Group 1: 204 nonadherent placenta previa patients; Group 2: 61 MAP patients managed with endouterine hemostatic square sutures and/or Bakri balloon tamponade; and Group 3: 51 patients with MAP requiring hysterectomy). Among all first and second trimester screening analytes only maternal serum alphafetoprotein (MS-AFP) levels were significantly higher in patients with MAP requiring hysterectomy (p < 0.001). According to the Receiver Operating Characteristic (ROC) analysis performed for the predictive value of MS-AFP levels, the area under the curve (AUC) was 0.742 [95% confidence interval (CI): 0.505-0.979]. The best MS-AFP cut-off value was 1.25 multiple of the median (MoM) with 85.94% sensitivity and 71.43% specificity (p = 0.036). The best predictors which affect the increased risk of hysterectomy, was further evaluated by multivariate logistic regression analyses. Only elevated maternal serum alphafetoprotein (MS-AFP) was found to be an independent predictor of MAP requiring hysterectomy [odds ratio (OR) = 25.329, 95% confidence interval (CI):1.487-43.143, p = 0.025]. In conclusion, increased second trimester MS-AFP levels independently predict morbidly adherent placenta requiring hysterectomy among women with placenta previa totalis.