Liang Jing, Liu Fang, Wang Fengmei, Han Tao, Jing Li, Ma Zhe, Gao Yingtang
Department of Gastroenterology and Hepatology, Tianjin Third Central Hospital, Tianjin 300170, China; Tianjin Key Laboratory of Artificial Cell, Tianjin 300170, China; Artificial Cell Engineering Technology Research Center of Public Health Ministry, Tianjin 300170, China.
Tianjin Key Laboratory of Artificial Cell, Tianjin 300170, China; Artificial Cell Engineering Technology Research Center of Public Health Ministry, Tianjin 300170, China; Molecular Biology Laboratory, Tianjin Third Central Hospital, Tianjin 300170, China.
Biomed Res Int. 2017;2017:8793278. doi: 10.1155/2017/8793278. Epub 2017 Mar 2.
. To develop a noninvasive score model to predict NASH in patients with combined CHB and NAFLD. . 65 CHB patients with NAFLD were divided into NASH group (34 patients) and non-NASH group (31 patients) according to the NAS score. Biochemical indexes, liver stiffness, and Controlled Attenuation Parameter (CAP) were determined. Data in the two groups were compared and subjected to multivariate analysis, to establish a score model for the prediction of NASH. . In the NASH group, ALT, TG, fasting blood glucose (FBG), M30 CK-18, CAP, and HBeAg positive ratio were significantly higher than in the non-NASH group ( < 0.05). Multivariate analysis showed that CK-18 M30, CAP, FBG, and HBVDNA level were independent predictors of NASH. Therefore, a new model combining CK18 M30, CAP, FBG, and HBVDNA level was established using logistic regression. The AUROC curve predicting NASH was 0.961 (95% CI: 0.920-1.00, cutoff value is 0.218), with a sensitivity of 100% and specificity of 80.6%. . A noninvasive score model might be considered for the prediction of NASH in patients with CHB combined with NAFLD.
开发一种无创评分模型以预测合并慢性乙型肝炎（CHB）和非酒精性脂肪性肝病（NAFLD）患者的非酒精性脂肪性肝炎（NASH）。65例合并CHB和NAFLD的患者根据NAS评分分为NASH组（34例）和非NASH组（31例）。测定生化指标、肝脏硬度和受控衰减参数（CAP）。比较两组数据并进行多因素分析，以建立预测NASH的评分模型。在NASH组中，谷丙转氨酶（ALT）、甘油三酯（TG）、空腹血糖（FBG）、细胞角蛋白18（CK-18）M30、CAP和乙肝e抗原（HBeAg）阳性率显著高于非NASH组（P<0.05）。多因素分析显示，CK-18 M30、CAP、FBG和乙肝病毒脱氧核糖核酸（HBVDNA）水平是NASH的独立预测因素。因此，采用逻辑回归建立了一个结合CK18 M30、CAP、FBG和HBVDNA水平的新模型。预测NASH的受试者工作特征曲线下面积（AUROC）为0.961（95%可信区间：0.920~1.00，截断值为0.218），敏感性为100%，特异性为80.6%。对于合并CHB和NAFLD的患者，可考虑使用无创评分模型预测NASH。
