磁共振ADC值联合血液生化学指标诊断肝纤维化的临床初探

Combination of MRI ADC Value and Biochemical Indicators for the Assessment of Fibrosis in Chronic Hepatitis:Preliminary Results

目的探讨磁共振表观扩散系数(ADC)值联合血液生化学指标诊断肝纤维化的临床价值。资料与方法利用扩散成像技术检测41例慢性肝病患者磁共振ADC值(b=600s/mm2),同时测量透明质酸(HA)并计算肝纤维化诊断预测模型Forns和APRI指数。以肝纤维化病理分期作为金标准,应用受试者工作特征(ROC)曲线分析以上指标诊断肝纤维化的价值。用Logistic回归综合两指标信息,根据ROC曲线确定诊断界值,分析其诊断肝纤维化的价值。结果在判断S≥2肝纤维化时,ADC值、HA、Forns和APRI的曲线下面积(Az)分别为0.88、0.80、0.72和0.76;在判断S≥3肝纤维化时,Az分别为0.93、0.84、0.77和0.74。而在判断S≥2肝纤维化时ADC+HA、ADC+Forns和ADC+APRI两两联合诊断时的Az分别为0.93、0.93和0.92;在判断S≥3肝纤维化时,Az分别为0.96、0.95和0.94,且Logistic回归方法的诊断敏感性和特异性比单个指标诊断均有所提高。结论ADC值与血液生化学多指标联合建立Logistic回归模型,和单独检测相比诊断准确性提高,可供临床参考。

Objective To evaluate the apparent diffusion coefficient （ADC） combined with biochemical indicators for the diagnosis of hepatic fibrosis in patients with chronic hepatitis. Materials and Methods 41 patients with chronic hepatitis who received liver biopsy were included in this study. All of them underwent DWI on 1.5T MRI device,and ADC values of liver were measured when b value were set as 600s/mm2. Meanwhile blood samples were obtained in all patients to observe blood biochemiology of hepatic fibrosis including hyaluronic acid（HA）,and Forns index,APRI index. The values of these methods for assentment of fibrosis in chronic hepatitis were evaluated by receiving operating characteristic （ROC） curve. The logistic regression model was used to combine the results from multiple tests to increase the accuracy. Results In discriminating patients staged S2-S4,the areas under the receiving operating characteristic curves （Az） were 0.88 for ADC,0.80 for HA,0.72 for Forns index,and 0.76 for APRI index. In discriminating patients staged S3-S4,the Az were 0.93 for ADC,0.84 for HA,0.77 for Forns index,and 0.74 for APRI index. Three combinations were ADC＋HA,ADC＋Forns and ADC＋APRI,whose Az reached 0.93,0.93 and 0.92 respectively when in discriminating patients staged S2-S4. In discriminating patients staged S3-S4,their Az reached 0.96,0.95 and 0.94 respectively. Furthermore,combination tests achieved a higher specificity and sensitivity. Conclusion Combining multiple test results by a logistic model may increase the diagnostic accuracy of hepatic fibrosis than single indicator.