Establishment of a nomogram model for predicting risk of intrahepatic cholestasis in pregnant women during pregnancy

  Objective  To establish a nomogram model for predicting the risk of intrahepatic cholestasis of pregnancy (ICP), and verify its efficiency.

  Methods  A total of 575 pregnant women were enrolled and were divided into modeling group (n=460) and verification group (n=115) according to simple randomization method. Logistic regression analysis was performed to determine the risk factors for ICP. A nomogram model was established using R software, and its predictive performance was evaluated.

  Results  Hepatitis B virus infection, gestational hypertension, low selenium intake, family history of ICP, fetal protection treatment with progesterone and hyperuricemia were risk factors of ICP (P < 0.05). The area under the curve (AUC) value and C index of the nomogram for predicting the risk of ICP were both 0.840 (95%CI, 0.764 to 0.916). The calibration curve showed that predictive results of the nomogram were highly consistent with the results of actual diagnosis. In the verification group, the AUC value and C index of the nomogram for predicting the risk of ICP both were 0.801(95%CI, 0.711 to 0.892). The calibration curve showed that predictive results of the nomogram were highly consistent with the results of actual diagnosis.

  Conclusion  Establishing anomogram can help to predict the risk of ICP, which is conducive to screening high-risk groups of ICP and formulating appropriate intervention measures.