3．0TMR标准化表观扩散系数值评价肝纤维化程度的研究

Diffusion weighted imaging for quantification of liver fibrosis in patients using normalized apparent diffusion coefficient values at 3.0 T MR

目的探讨标准化ADC值对肝纤维化程度定量分析的能力。方法采用3．0TMRDWI检查回顾性分析10名健康志愿者（对照组）和43例经肝脏穿刺活检病理证实为肝功能代偿期的肝病患者（慢性肝病组）资料，慢性肝病组患者进行肝纤维化分期（S分期）。测量不同肝纤维化分期下肝脏ADC值、脾脏ADC值、肾皮质ADC值、脾脏标准化ADC值（S-ADC）及肾脏标准化ADC值（R—ADC）。采用非参数Spearman检验分析S分期与ADC值之间的相关性；采用单因素方差分析比较各期ADC值间的差异，采用Logistic回归分析评价ADC值预测纤维化程度的能力，选择约登指数最大作为截断点，计算曲线下面积（AUC）、敏感度及特异度。结果10名对照组志愿者均为S0期。慢性肝病组S0、S1、S2、S3、S4期分别有2、5、9、12和15例。S0-S4期肝脏ADC值分别为（1．37±0．13）×10^-3、（1．33±0．16）×10^-3、（1．17±0．16）×10^-3、（1．23±0．14）×10^-3和（1．12±0．11）×10^-3mm^2／s，S—ADC分别为1．86±0．18、1．68±0．12、1．3±0．14、1．48±0．15和1．34±0．10，R—ADC分别为0．71±0．08、0．68±0．12、0．61±0．09、0．64±0．11和0．60±0．08，差异均有统计学意义（F值分别为6．48、18．70和3．04，P〈0．05）。S-ADC、肝脏ADC值和R—ADC均与肝纤维化分期呈负相关性（r值分别为-0．71、-0．51和-0．41，P值均〈0．01），但S-ADC与肝纤维化分期的相关性高于肝脏ADC值和R-ADC。对于预测≥S2期、≥S3期及S4期纤维化，S-ADC的效果均优于肝脏ADC值、R—ADC，约登指数分别为0．91、0．58和0．59。结论S-ADC在评价肝纤维化方面优于肝脏ADC值及R—ADC，具有良好的诊断准确性。

Objective The purpose of this study is to discuss the diagnostic accuracy of normalized liver ADC using the spleen and renal cortex as reference organs for the diagnosis of liver fibrosis. Methods Forty three patients with liver disease （chronic liver disease group） at compensated stage and 10 healthy volunteers （control group） were retrospectively assessed with diffusion-weighted imaging at a 3.0 T MR unit. Liver ADC, spleen ADC, renal ADC and normalized ADC （ defined as the ratio of liver ADC to spleen ADC or renal cortex ADC, S-ADC and R-ADC for short） were measured in patients stratified by fibrosis stage. Spearman analysis was used to see the correlation between fibrosis stages and ADC, one-way ANOVA was used to compare the ADCs in different fibrosis stages. Logistic regression analysis was used to determine the performance of ADC for prediction of liver fibrosis, and show the area under the curve （AUC） , sensitivity and specificity choosing the optimal cutoff value that maximized the Youden index. Results Ten volunteers belonged to SO stage. From SO to $4 stage, there were 2, 5, 9, 12 and 15 patients, correspondingly, liver ADC were （1.37±0.13） ×10-3, （1.33±0.16） ×10-3, （1.17±0.16） ×10-3, （1.23±0.14） ×10-3 and （1. 12 ±0. 11） × 10 -3mm2/s, S-ADC were 1. 86 ±0. 18, 1. 68 ±0. 12, 1. 34 ±0. 14, 1.48 ±0. 15 and 1.34±0. 10, R-ADC were 0.71 ±0.08, 0.68 ±0.12, 0.61 ±0.09, 0.64 ± 0. 11 and 0.60 ±0.08 respectively, and the differences among them were significant （F = 6.48, 18.70 and 3.04 ,P 〈 0. 05 ）. Thecorrelation between fibrosis stage and S-ADC was stronger than between fibrosis stage and liver ADC, R-ADC （r = -0. 71, -0. 51, -0. 41 ;P 〈 0. 01 ） . S-ADC was superior to liver ADC and R-ADC for detection of S2, S3 and S4 fibrosis stage （Youden index： 0. 91, 0. 58, and 0. 59）. Conclusion Spleen normalized liver ADC improves diagnostic accuracy for detection of liver fibrosis than liver ADC and renal normalized liver ADC.