摘要目的 探讨钆塞酸二钠(Gd?EOB?DTPA)增强T1ρ和体素内不相干运动(IVIM)成像评价非酒精性脂肪肝炎(NASH)活动性评分(NAS)与炎症分级的诊断价值.方法 以高脂?高胆固醇饲料喂养4、8和12周建立NASH模型,对照组6只,实验组20只,行肝脏IVIM和Gd?EOB?DTPA增强T1ρ成像,采用组内相关系数(ICC)评价短期测量的一致性,以组织病理学为金标准,分析ADC、真性扩散系数(D)、假性扩散系数(D*)、灌注分数(f)、T1ρ及肝胆期T1ρ与NAS、炎症、纤维化的相关性,采用ROC分析T1ρ、IVIM各参数对NASH和进展期炎症的诊断价值.结果 实验组20只兔中,1只兔肝脏正常, 8只兔诊断为NASH,6只兔诊断为NASH可能,其余5只诊断为单纯性脂肪肝.f值与NAS评分呈负相关(r=-0.530,P<0.01),纤维化S1、2的f值较S0下降(P<0.01),D、D*、ADC在NASH评分、炎症、纤维化分期差异无统计学意义(P>0.05).T1ρ值和肝胆期T1ρ值与NAS评分均呈正相关,r值分别为0.658、0.750,P均<0.01;T1ρ和肝胆期T1ρ值与炎症评分呈良好正相关,r值分别为0.790、0.812,P均<0.01.T1ρ、肝胆期T1ρ、ADC、D、D*、f值诊断NASH的ROC下面积分别为0.849、0.949、0.728、0.596、0.522和0.871.肝胆期T1ρ结合f值诊断NASH、G2及3级炎症、S1及2级纤维化的ROC下面积分别为0.971、0.935和0.903,炎症(R2=0.746,P=0.002)与纤维化(R2=0.624,P=0.002)是肝胆期T1ρ主要独立影响因子.结论 Gd?EOB?DTPA增强T1ρ成像能反映NASH的严重性和炎症程度,IVIM参数不足以评价NASH炎症分级;肝胆期T1ρ结合f值是评价NASH活动性积分和炎症分级更好的无创性影像标记物.

abstractsObjective To explore the values of metrics on intravoxel incoherent motion (IVIM) and gadoxetic acid enhanced T1ρ imaging for staging of non?alcoholic fatty liver disease activity scores (NAS) and inflammation in nonalcoholic steatohepatitis (NASH) rabbits model. Methods NASH rabbits model was established by feeding with a varied duration (4, 8, 12 weeks) of high?fat, high?cholesterol diet. IVIM and gadolinium?ethoxybenzyl?diethylenetriamine pentaacetic acid (Gd?EOB?DTPA) enhanced T1ρ images were performed by a 3.0 T MR scanner. The inter?class correlation coefficients (ICC) and Bland?Altman analysis were applied to evaluate the reproducibility of the IVIM and Gd?EOB?DTPA enhanced T1ρ mapping measurers. Spearman correlation analysis were used to assess the correlation between MR metrics, including ADC, D, D*, f, T1ρ, T1ρ (hepatobiliary phase, HBP), and NAS score and inflammation grades respectively with reference to histopathology. ROC curve analysis was used to evaluate the diagnostic performance of T1ρ and IVIM parameters for NASH, inflammation grade, and hepatic fibrosis. Multiple linear regression equations were used to analyze the independent influence factors of T1ρ (HBP). Results The f value was negatively correlated with the NAS score (r=-0.530, P<0.01). The f value of the fibrosis S1?2 was significantly lower than that of the S0 (P<0.01). There was no statistical difference in D, D*, ADC among NASH score, inflammation, and fibrosis stage. T1ρ and T1ρ (HBP) values were positively correlated with NAS scores and inflammation grades. The area under curve (AUC) for the diagnosis of NASH for T1ρ, T1ρ(HBP), ADC, D, D*, and f values were 0.849, 0.949, 0.728, 0.596, 0.522, and 0.871, respectively. The AUCs of T1ρ (HBP)+f in the diagnosis of NASH, G2?3 inflammation, and F1?2 fibrosis were 0.971, 0.935, and 0.903, respectively. Fibrosis (R2=0.624, P=0.002) and inflammation (R2=0.746, P=0.002) were major independent factors of T1ρ (HBP). Conclusion Gd?EOB?DTPA enhanced T1ρ imaging can reflect the severity of NASH and degree of inflammation. IVIM measurements are not accurate enough to stage liver inflammatory activity of NASH. T1ρ (HBP)+f might be a superior noninvasive imaging biomarker than either non?enhanced T1ρ or IVIM for NASH activity and inflammation assessments.