Objective To investigate the etiology and clinical manifestations of liver failure in pregnancy and the value of TPL predictive model based on total bilirubin( TBil),prothrombin activity( PTA),and lactic acid( LACT) in evaluating the prognosis of liver failure in pregnancy. Methods A total number of 69 pregnant patients who were diagnosed with liver failure in The Third Affiliated Hospital of Guangzhou Medical University from January 1,2009 to December 31,2019 were enrolled,and according to prognosis,they were divided into death group with 22 patients and survival group with 47 patients. The two groups were compared in terms of etiology,clinical manifestation,laboratory markers,and prognosis. A multivariate logistic regression analysis was used to investigate the independent risk factors for death in patients with liver failure in pregnancy,and a TPL predictive model was established. The t-test was used for comparison of normally distributed continuous data between two groups,and the Wilcoxon signed-rank test was used for comparison of non-normally distributed continuous data between two groups; the chi-square test was used for comparison of categorical data between two groups. The receiver operating characteristic( ROC) curve was plotted,and the area under the ROC curve( AUC) was used to analyze the value of TPL model in predicting the prognosis of patients with liver failure in pregnancy. Results Of all 69 patients,22 died and 47 survived,with a mortality rate of31. 9%. Acute fatty liver of pregnancy( AFLP) was the most common cause of liver failure in pregnancy( 37. 7%),followed by viral hepatitis( 27. 5%). There was no significant difference in mortality rate between the patients with different etiologies( χ2= 4. 013,P > 0. 05).Jaundice was the most common clinical manifestation of liver failure in pregnancy( 79. 7%),followed by poor appetite( 63. 8%) and edema of both lower limbs( 52. 2%). There were no significant differences in clinical manifestations between the survival group and the death group( P > 0. 05). Compared with the survival group,the death group had significantly higher TBil,LACT,and international normalized ratio and significantly lower PTA and platelet count( Z =-2. 691,Z =-1. 998,Z =-2. 640,t =-2. 545,Z =-2. 222,all P < 0. 05). The multivariate logistic regression analysis was used to include TBil,PTA,and LACT into an equation and establish the TPL model( all P<0. 05),and the TPL model had a sensitivity of 90. 9%,a specificity of 68. 1%,a positive predictive value of 57. 1%,and a negative predictive value of 94. 1%. The TPL model had an AUC of 0. 833( 95% confidence interval [CI]: 0. 771-0. 965,P < 0. 05),and the TPL model had a significantly higher AUC than the TBil model( AUC = 0. 702,95% CI: 0. 594-0. 805,P < 0. 05),PTA model( AUC =0. 673,95% CI: 0. 550-0. 796,P < 0. 05),and LACT model( AUC = 0. 650,95% CI: 0. 494-0. 772,P < 0. 05). According to the cut-off value of the ROC curve,patients' mortality rate increased with the increase in the score of the TPL model( χ2= 20. 312,P <0. 05). Conclusion AFLP and viral hepatitis are common causes of liver failure in pregnancy,and jaundice,poor appetite,and edema of both lower limbs are common clinical manifestations of liver failure in pregnancy. The TPL predictive model is more accurate than the single index in predicting the prognosis of liver failure in pregnancy and has a better clinical guiding value.
DownLoad: