A new prognostic model for pregnancy-specific liver disease

Objective To investigate the prognostic factors for pregnancy-specific liver disease ( PSLD) , to establish a simple effective model ( NEW mode) using the statistic method, and to investigate the value of Model for End-Stage Liver Disease ( MELD) and NEW model in the prognostic evaluation of PSLD patients. Methods A total of 84 pregnant patients who were admitted to The Third Affiliated Hospital of Guangzhou Medical University due to jaundice and hepatic insufficiency from January 1, 2015 to December 31, 2017 and were diagnosed with acute fatty liver of pregnancy, preeclampsia, or hemolysis, elevated liver enzymes, and low platelets ( HELLP) syndrome were enrolled. According to their prognosis, they were divided into death group with 14 patients and survival group with 70 patients. The MELD model and the NEW model were used for scoring, the sensitivity, specificity, positive predictive value, and negative predictive value of the two models were calculated, and the receiver operating characteristic ( ROC) curve was used to assess the value of these two models in predicting the prognosis of PSLD. The t-test was used for comparison of continuous data between two groups, the chi-square test was used for comparison of categorical data between two groups, and the logistic regression model was used to analyze the prognostic models of PSLD.Results The univariate analysis showed that there were significant differences between the two groups in total bilirubin, direct bilirubin, serum creatinine, international normalized ratio ( INR) , alkaline phosphatase, serum total protein, and albumin ( t =-6. 33, -5. 38, -3. 38, -11. 56, -4. 63, 3. 00, and 2. 19, all P < 0. 05) , among which INR showed the most significant difference and was thus used to establish the NEW model ( R = 5. 812 × INR-11. 3) . Both MELD model and NEW model demonstrated a good predictive value, with an area under the ROC curve of 0. 952 ( 95% confidence interval [CI]: 0. 906-0. 997, P < 0. 05) and 0. 982 ( 95% CI: 0. 958-1. 000, P < 0. 05) , respectively, a sensitivity of 92. 9% and 100%, respectively, and a specificity of 87. 3% and 94. 4%, respectively. Higher scores of MELD model and NEW model were associated with higher fatality rates; the fatality rate was 0 when MELD score was lower than 25, but it increased to 81. 8% when MELD score was above 40; the fatality rate was 0 when R score was lower than-3, but it increased to 84. 6%when R score was above 0. Conclusion Both MELD model and NEW model can predict the prognosis of PSLD patients. The NEW model with reference to INR alone achieves a good value in evaluating prognosis and thus has a good clinical value.