多普勒超声及Fibroscan评分系统在慢性乙型肝炎肝纤维化诊断中的应用价值

目的:探讨基于多普勒超声及Fibroscan构建的肝纤维化无创评分系统对慢性乙型肝炎患者的肝纤维化诊断价值。方法:选取在医院行超声引导下肝穿刺活检的300例乙型肝炎患者,根据肝穿刺活检结果将其分为肝纤维化组(172例)和非肝纤维化组(128例)。比较两组患者的性别、年龄、病程、彩色多普勒超声检查结果与Fibroscan硬度值(LSM值),将差异有统计学意义的变量纳入多因素Logistic回归分析中确定肝纤维化影响因素。根据逐步Logistic回归分析筛选出影响因素的标准化回归系数β得到相应的评分赋值,构成肝纤维化无创评分系统。绘制受试者工作特征(ROC)曲线,分析该评分系统诊断肝纤维化的诊断效能,确定最佳临界值以及敏感性和特异性。为进一步验证该评分系统对肝纤维化的诊断价值,另选行肝穿刺活检的100例乙型肝炎患者进行评分。结果:肝纤维化组中男性、肝被膜欠光滑、肝实质回声不均匀、肝内光点增粗、肝静脉清晰度较差的占比均明显高于非肝纤维化组,差异有统计学意义(χ²=4.089,χ²=12.105,χ²=7.792,χ²=9.383,χ²=7.053;P<0.05);脾长径、脾厚度、脾脏面积、胆囊壁厚度、肝固有动脉峰值血流速度(HAV_max)、肝固有动脉/门静脉峰值流速的比值(A/P)、肝脏硬度值(LSM)均明显高于非肝纤维化组,门静脉峰值血流速度(PVV_max)、脾静脉峰值血流速度(SVV_max)、肝脏循环指数(HCI)明显低于非肝纤维化组,差异有统计学意义(t=2.296,t=2.228,t=2.364,t=5.970,t=2.864,t=6.083,t=10.268,t=2.925,t=5.749,t=2.209;P<0.05)。多因素Logistic回归分析证实,患者性别、肝被膜光滑程度、肝实质回声、肝内光点增粗、肝静脉清晰度、胆囊壁厚度、HCI、A/P以及LSM值是肝纤维化的影响因素。根据筛选出的影响因素的标准化回归系数β得到对应评分赋值,最终该评分系统的总分为0~12分,其中0~7分为低危,8~9分为中危,10~12分为高危。以患者总评分为自变量、肝纤维化情况为因变量绘制的ROC曲线下面积为0.948(95%CI:0.835~0.989),最佳临界值为10分,此时灵敏度和特异度分别为93.46%和81.05%。100例乙型肝炎患者基于评分系统的诊断结果与肝穿刺结果有高度一致性(Kappa值=0.762)。结论:基于彩色多普勒超声和FibroScan所构建的肝纤维化无创评分系统可为肝纤维化的诊断及抗纤维化提供有力依据。

Application value of Doppler ultrasound and Fibroscan scoring system in diagnosis of chronic hepatitis B fibrosis

Objective:To explore the diagnostic value of liver fibrosis non-invasive scoring system which was constructed by Doppler ultrasound and Fibroscan for patients with chronic hepatitis B.Methods:300 patients with hepatitis B who underwent ultrasound-guided aspiration biopsy on liver were selected and were divided into liver fibrosis group(172 cases)and non-liver fibrosis group(128 cases)according to the results of aspiration biopsy on liver.The differences of gender,age,course of hepatitis B,color Doppler ultrasound results and Fibroscan hardness value(LSM value)between the two groups were compared.Statistically significant variables were included in multivariate logistic regression analysis to determine the influencing factors of liver fibrosis.According to the stepwise Logistic regression analysis,the standardized regression coefficientβof the influencing factors was selected to obtain the corresponding score assignment,which constituted a non-invasive scoring system for liver fibrosis.The receiver operating characteristics(ROC)curve was drawn to analyze the diagnostic efficacy of the scoring system in diagnosing liver fibrosis,and to determine the optimal cut-off value as well as sensitivity and specificity.For further verification of the diagnostic value of this scoring system for liver fibrosis,other 100 patients with hepatitis B who underwent aspiration biopsy on liver were selected for scoring.Results:The proportions of men with obvious fibrosis,the liver capsule is not smooth,uneven echo of liver parenchyma,thickening of intrahepatic light spot,poor clarity of hepatic vein of liver fibrosis group were significantly higher than those of non-liver fibrosis group(χ²=4.089,χ²=12.105,χ²=7.792,χ²=9.383,χ²=7.053,P<0.05).And the splenic diameter,splenic thickness,spleen area,Gallbladder wall thickness,the blood flow velocity of peak value of arteria hepatica propria(HAVmax),the ratio of flow velocity of peak value of arteria hepatica propria to portal vein(A/P)and LSM value of liver fibrosis group were significantly higher than those of non-liver fibrosis group,while the blood velocity of portal vein peak(PVVmax),the blood velocity of splenic vein peak(SVVmax),hepatic cycle index(HCI)of liver fibrosis group were significantly lower than those of non-liver fibrosis group(t=2.296,t=2.228,t=2.364,t=5.970,t=2.864,t=6.083,t=10.268,t=2.925,t=5.749,t=2.209,P<0.05).Multivariate logistic regression analysis confirmed that gender,smoothness of liver capsule,liver parenchymal echo,thickness of intrahepatic spot lightening,hepatic vein clarity,gallbladder wall thickness,HCI,A/P and LSM value were the influencing factor of liver fibrosis.According to the standardized regression coefficientβof the selected influencing factors,the corresponding score assignments were obtained.And finally the total score of the scoring system was 0 point to 12 points.Among of them,0 point to 7 points were classified as low risk,and 8 points to 9 points were classified as medium risk,and 10 points to 12 points were classified as high risk.The area under the receiver operating characteristics(ROC)curve drawn with the patient's total score as the independent variable and with liver fibrosis as the dependent variable was 0.948(95%CI:0.835-0.989),and the optimal cut-off value was 10 points.The sensitivity and specificity were respectively 93.46%and 81.05%at this time.The results of aspiration biopsy on liver of 100 patients with hepatitis B were highly consistent with the results based on the scoring system(Kappa value=0.762).Conclusion:The liver fibrosis non-invasive scoring system based on color Doppler ultrasound and FibroScan can provide a strong basis for diagnosis and anti-fibrosis of liver fibrosis.