探讨慢性乙型肝炎病毒(HBV)携带者肝纤维化的影响因素并构建其相关危险因素模型,对符合诊断的肝功能正常的慢性HBV携带者应用FibroScan瞬时弹性扫描仪检测肝硬度(LSM)值,筛选出发生显著肝纤维化者(LSM值≥7.5 kPa),并通过Logistic回归分析统计方法对性别、年龄、HBV DNA载量、乙肝表面抗原(HBsAg)、乙肝e抗原(HBeAg)等相关因素进行分析,从而构建出Logistic回归模型.结果显示:纳入研究的239例慢性HBV携带者中,未发生显著肝纤维化的占45.19%(108例),发生显著肝纤维化的患者占54.81%(131例),其中有16.32%(39例)的患者进展为肝硬化.经t-检验发现,HBV DNA载量、年龄、性别为男性、HBsAg高水平和HBeAg阴性均为慢性HBV携带者发生显著肝纤维化的危险因素,性别为女性及HBeAg阳性是HBV携带者发生显著肝纤维化的保护因素.影响因素Logistic回归模型为P=eS/(1+eS),S=-5.258+1.807X1+0.749X2-1.154X3+0.219X4-0.124X5.Hosmer-Lemeshow拟合优度检验p=0.123>0.1,表明该回归方程的拟合度好.通过受试者操作特征(ROC)曲线评价并确定诊断临界点:ROC曲线下面积为0.842,标准误为0.037,p=0.008<0.05,95%置信区间为0.528～0.872,表明该模型对于慢性HBV携带者转归的预测价值中等.综上,HBV DNA载量、年龄、性别、HBsAg和HBeAg是本组研究患者的主要相关因素,该模型或可应用于临床上对肝纤维化进行治疗指导.
In this study,we explored factors related to liver fibrosis in chronic hepatitis B virus(HBV)carriers and constructed a related risk factor model.The liver stiffness(LSM)value determined with FibroScan was used to screen out those with significant liver fibrosis from diagnosed chronic HBV carriers(LSM value≥7.5 kPa),and then their gender,age,HBV DNA load,hepatitis B surface antigen(HBsAg),hepatitis B e antigen(HBeAg)and other related factors were statistically analyzed to construct a Logistic regression model.Among the 239 chronic HBV carriers included in the study,45.19%(108)showed no significant liver fibrosis,54.81%(131)had significant liver fibrosis,and 16.32%(39)had liver cirrhosis.According to t-test,we found that HBV DNA load,age,male,high HBsAg level and negative HBeAg were risk factors related to significant liver fibrosis in chronic HBV carriers; while female and positive HBeAg were protective factors against liver fibrosis in these carriers.The Logistic regression model of related factors was:P=eS/(1+eS),S=-5.258+1.807X1+0.749X2-1.154X3+0.219X4-0.124X5.According to the goodness test of Hosmer-Lemeshow fitting,p=0.123>0.1,indicating that the regression equation had a good fitting.In addition,a receiver-operating characteristic(ROC)curve was used to evaluate and determine the diagnostic critical points,with the area under the ROC curve of 0.842 and the standard error of 0.037(p=0.008<0.05,95% confidence interval 0.528-0.872),indicating moderate predictive power of this model in prognosis of chronic HBV carriers.In summary,HBV DNA load,age,gender,HBsAg and HBeAg were the major factors related to liver fibrosis.This model may serve to guide the clinical practice against liver fibrosis.
