ABSTRACT
PURPOSE: To develop a model based on non-invasive clinical and ultrasonographic parameters for predicting the likelihood of subsequent histologic chorioamnionitis in women with preterm premature rupture of membranes (PPROM) and to determine whether the inclusion of invasive test results improves the predictive value of the model. MATERIALS AND METHODS: This retrospective cohort study included 146 consecutive women presenting with PPROM (20-33 weeks). Transvaginal ultrasonographic assessment of cervical length was performed. Maternal serum C-reactive protein (CRP) levels and white blood cell (WBC) counts were measured after amniocentesis. Amniotic fluid (AF) obtained by amniocentesis was cultured, and interleukin-6 (IL-6) levels and WBC counts were determined. The primary outcome measure was histologic chorioamnionitis. RESULTS: Risk scores based on serum CRP concentrations and gestational age (model 1) were calculated for each patient. The model was shown to have adequate goodness of fit and an area under the receiver operating characteristic curve (AUC) of 0.742. When including AF test results (e.g., AF IL-6 levels) in model 1, serum CRP concentrations were found to be insignificant, and thus, were excluded from model 2, comprising AF IL-6 levels and gestational age. No significant difference in AUC was found between models 1 and 2. CONCLUSION: For women with PPROM, the newly developed model incorporating non-invasive parameters (serum CRP and gestational age) was moderately predictive of histologic chorioamnionitis. The inclusion of invasive test results added no predictive information to the model in this setting.