Liver fibrosis index-based nomograms for identifying esophageal varices in patients with chronic hepatitis B related cirrhosis
Shi-Hao Xu,Fang Wu,Le-Hang Guo,Wei-Bing Zhang,Hui-Xiong Xu
DOI: https://doi.org/10.3748/wjg.v26.i45.7204
IF: 5.374
2020-12-07
World Journal of Gastroenterology
Abstract:Esophageal varices (EV) are the most fatal complication of chronic hepatitis B (CHB) related cirrhosis. The prognosis is poor, especially after the first upper gastrointestinal hemorrhage.To construct nomograms to predict the risk and severity of EV in patients with CHB related cirrhosis.Between 2016 and 2018, the patients with CHB related cirrhosis were recruited and divided into a training or validation cohort at The First Affiliated Hospital of Wenzhou Medical University. Clinical and ultrasonic parameters that were closely related to EV risk and severity were screened out by univariate and multivariate logistic regression analyses, and integrated into two nomograms, respectively. Both nomograms were internally and externally validated by calibration, concordance index (C-index), receiver operating characteristic curve, and decision curve analyses (DCA).A total of 307 patients with CHB related cirrhosis were recruited. The independent risk factors for EV included Child-Pugh class [odds ratio (OR) = 7.705, 95% confidence interval (CI) = 2.169-27.370, = 0.002], platelet count (OR = 0.992, 95%CI = 0.984-1.000, = 0.044), splenic portal index (SPI) (OR = 3.895, 95%CI = 1.630-9.308, = 0.002), and liver fibrosis index (LFI) (OR = 3.603, 95%CI = 1.336-9.719, = 0.011); those of EV severity included Child-Pugh class (OR = 5.436, 95%CI = 2.112-13.990, < 0.001), mean portal vein velocity (OR = 1.479, 95%CI = 1.043-2.098, = 0.028), portal vein diameter (OR = 1.397, 95%CI = 1.021-1.912, = 0.037), SPI (OR = 1.463, 95%CI = 1.030-2.079, = 0.034), and LFI (OR = 3.089, 95%CI = 1.442-6.617, = 0.004). Two nomograms (predicting EV risk and severity, respectively) were well-calibrated and had a favorable discriminative ability, with C-indexes of 0.916 and 0.846 in the training cohort, respectively, higher than those of other predictive indexes, like LFI (C-indexes = 0.781 and 0.738), SPI (C-indexes = 0.805 and 0.714), ratio of platelet count to spleen diameter (PSR) (C-indexes = 0.822 and 0.726), King's score (C-indexes = 0.694 and 0.609), and Lok index (C-indexes = 0.788 and 0.700). The areas under the curves (AUCs) of the two nomograms were 0.916 and 0.846 in the training cohort, respectively, higher than those of LFI (AUCs = 0.781 and 0.738), SPI (AUCs = 0.805 and 0.714), PSR (AUCs = 0.822 and 0.726), King's score (AUCs = 0.694 and 0.609), and Lok index (AUCs = 0.788 and 0.700). Better net benefits were shown in the DCA. The results were validated in the validation cohort.Nomograms incorporating clinical and ultrasonic variables are efficient in noninvasively predicting the risk and severity of EV.
gastroenterology & hepatology