Intrahepatic cholestasis of pregnancy: dilemma in diagnosis and management.

Abstract

Objective: To determine the obstetrical outcomes of women delivered for the diagnosis of intrahepatic cholestasis of pregnancy (ICP).

Methods: Retrospective study of singleton pregnancies diagnosed with ICP between 1 May 2014 and 31 December 2017. Population was analyzed based on bile acids: normal (<10 µmol/L), mild (10 to 40 µmol/L), moderate-severe (>40 µmol/L), and not obtained. Receiver operating characteristic curves established critical values for aspartate aminotransferase (AST) and alanine aminotransferase (ALT) to predict elevated bile acids. Statistical analyses included χ² for categorical variables and ANOVA for continuous variables. All tests used a 2-sided α level of significance of .05.

Results: Bile acids were normal in 39 (45.9%) women, 30 (35.3%) had mild cholestasis, 10 (11.8%) had moderate-severe cholestasis and not obtained for six (7%) women. Gestational diabetes was more common in mild cholestasis (p = .03). There were no differences in demographics, clinical presentation, obstetric interventions and neonatal outcomes. Bile acids took 5-6 days to result. Rate of labor inductions was high in all groups. Postpartum complications occurred in four women in the normal group and in one woman in the mild cholestasis group. Five (12.8%) neonates in the normal group, six (20%) in the mild group, and one (10%) in the severe group were admitted to the NICU. There was no fetal asphyxia, no 5-minute Apgar score <7, and no perinatal deaths. An AST of 27.5 IU/L (p = .002) with sensitivity of 81% and specificity of 76%, and an ALT of 26.7 IU/L (p = .004) with sensitivity of 78% and specificity of 68% predicted elevated bile acids. Improving the sensitivity of AST and ALT to 95%, the ROC curve identified an AST of 62 IU/L with a specificity, positive and negative predictive values of 32, 58 and 86%, respectively; and an ALT of 106 IU/L with a specificity, positive and negative predictive values of 27, 57 and 83%, respectively.

Conclusions: ICP should not be presumed in patients with pruritus. This practice may lead to early term delivery and associated complications.