钆塞酸二钠增强MRI在小儿亲体肝移植供体一站式评估中的可行性研究

Feasibility study of gadoxetic acid disodium enhanced MRI in preoperative one-stop-shop evaluation for parent liver donors

目的探讨钆塞酸二钠（Gd-EOB-DTPA）增强MRI在小儿亲体肝移植供体一站式评估中的可行性及临床应用价值。方法前瞻性收集拟行小儿亲体肝移植的40例供体，均行一站式上腹部MRI平扫及Gd-EOB-DTPA增强扫描。由2名放射科医师分别记录肝动脉、门静脉、肝静脉、胆管的解剖分型及满足胆管分型需要的最短延迟时间，进行图像质量评分，并测量移植肝体积。由1名移植科医师术中记录肝血管及胆管分型，并测量移植肝体积。分析术前影像和术中解剖分型结果的符合率。采用组内相关系数（ICC）评价2名放射科医师术前在影像上测量移植肝体积的一致性，采用Pearson相关分析评价术前影像测量移植肝体积与术中测量体积间的相关性，采用Kappa检验评价2名放射科医师间图像质量评分的一致性。结果32例行肝移植手术，影像评估肝动脉、门静脉、肝左静脉与肝中静脉共干、胆管解剖与术中结果完全一致，符合率为100％（32／32）。术前2名放射科医师测量的移植肝体积分别为（249．89±39．93）、（242．76±39．04）cm^3，一致性高（ICC=0．983）。术中测量的移植肝体积为（219．95±46．11）cm^3，和术前测量结果具有高度相关性（r值分别为0．820、0．850，P均〈0．01）。2名医师对肝动脉和胆管的图像质量评分均为1～3分，门静脉及肝静脉图像质量评分均为2、3分，均可满足临床诊断要求，对肝动脉、门静脉、肝静脉评分的一致性好（Kappa值分别为0．841、1．000、1．000），对胆管评分的一致性一般（Kappa值为0．658）。40例肝胆期延迟15min图像均能满足胆管分型需求（100％，40／40）。结论Gd-EOB—DTPA增强MRI能够有效地对小儿亲体肝移植供体进行一站式术前评估，且用于胆管分型的延迟时间可缩短至15min。

Objective To explore the feasibility and clinical application of gadoxetie-acid-disodium （Gd-EOB-DTPA）-enhanced MRI in preoperative one-stop-shop evaluation for parent liver donors. Methods Forty parent liver donors were included in this prospective study. They all underwent Gd-EOB-DTPA-enhanced MRI for preoperative one-stop-shop evaluation, Two radiologists evaluated the anatomical variations of hepatic vessels, bile ducts, the shortest timing for bile ducts, image quality and graft liver volume. One transplantation surgeon recorded the intraoperative anatomy of hepatic vessels and bile ducts, and measured the transplanted liver volume. The consistency between imaging results and intraoperative findings was calculated. Interohserver agreements measuring the graft volume was assessed by intra-class correlation coefficient （ICC）. Pearson correlation coefficients were measured between MRI-derived graft volume and actual graft volume. Interobserver agreements regarding the image quality were determined by calculating Kappa values. Results Thirty-two parent donors proceeded to liver donation. All anatomical findings of hepatic vessels and bile ducts accurately confirmed by intraoperative findings （32/32, 100%）. MRI-derived graft volume measured by the two radiologists were （249.89 ±39.93）cm^3, （242.76 ± 39.04） cm^3, respectively and there was a high interobserver consistency （ICC=0.983）. The intraoperative graft volume was （219.95±46.1l） cm^3 and correlated with actual graft volume moderately in both reviewers （r=0.820, 0.850, P〈0.O1）. The image quality score were from 1 to 3 for hepatic arteries and bile duets, 2 or 3 for portal and hepatic veins, which all met the clinical diagnostic requirements. The interobserver agreement was good for evaluating hepatic vessels （Kappa values were 0.841, 1.000, 1.000） and moderate for bile ducts （Kappa value was 0.658）. Fifteen minutes delay in hepatoboliary phase can meet the demand of bile duets classification in all donors. Conelttsion One-stop-shop Gd-EOB-DTPA-enhaneed MRI can be used in preoperative evaluation of parent donors effectively, and 15 minutes delay scanning is enough for eholangiography.