全肝CT灌注技术在人正常肝脏及肝隐匿性转移瘤中的初步研究

Application of total liver perfusion CTin human normal liver and occult hepatic metastases

目的:应用全肝CT灌注技术探讨正常肝脏组织各肝段血流灌注情况,以及通过血流灌注的变化推断存在肝隐匿性转移的肝段。方法:采用全肝CT灌注技术对14例健康志愿者及10例存在肝隐匿性转移的病例进行扫描,扫描图像应用最大斜率法数学模型进行后处理,可以获得肝动脉灌注量(HAP)、肝动脉灌注指数(HPI)等灌注参数。分析对照组正常肝脏各肝段灌注参数的差异,以及分析存在隐匿性转移肝段与正常肝段灌注参数的差异。结果:对照组各肝段之间HAP、HPI无显著统计学差异,对照组中肝段HAP均值为(15.81±0.49)ml·min-1·100g-1,HPI均值为(19.60±0.57)%。存在隐匿性转移肝段HAP、HPI分别高于对照组,有显著性差异(P<0.05)。采用ROC曲线分析,HAP对诊断隐匿性转移肝段最佳阈值为23.97ml·min-1·100g-1,灵敏度为73%,特异度为89%。HPI对诊断隐匿性转移肝段最佳阈值为29.18%,灵敏度为67%,特异度为89%。结论:全肝CT灌注可以反映正常肝脏组织各肝段灌注情况,并且提示肝脏各段血流灌注情况相近。全肝CT灌注可以通过肝段灌注参数值变化反映存在隐匿性转移肝段的血流灌注变化,从而可能对隐匿性转移肝段做出早期诊断。

Objective：To investigate the perfusion in each liver segment of normal liver tissue, by application of the total liver perfusion CT, as well as through changes in perfusion to infer the presence of occult metastases liver seg- ment. Methods： From February 2010 to February 20ll, 10 patients diagnosed with ＂ occult hepatic metastases＂ （OHM） and 14 healthy volunteers had accepted the total liver CT perfusion scanning. The data was analyzed with the software of Functional Analysis. Maximum slope method for the mathematical model was used in the processing, and hepatic artery perfusion（HAP） and hepatic perfusion index（HPI） of the perfusion pseudo - color pictures were ob- tained. First, the values of HAP, HPI in S1 - $8 had been calculated, and we determined whether there were differ- ences in the control group. Then, we tried to find the differences by statistical analysis of HAP, HPI of hepatic seg- ments between the control group and patients group. Finally,we tried to determine the best threshold of HAP, HPI for detecting OHM and the corresponding sensitivity and specificity. Results： Among hepatic segments of the control group, there was no significant statistical difference hepatic in HAP, so as HPI. The mean of HAP was （ 15.81 -＋ 0.49） ml · min-1 . 100g-I ,and the mean of HPI was （ 19.60 ± 0.57 ） %. The HAP, HPI of hepatic segments with OHM were higher than the control group, and there was significant difference （ P 〈 0. 05 ）. By analysis of ROC curve, the best threshold of HAP to detect OHM was 23.97ml· min-X . 100g-1 with the sensitivity of 73% and specificity 89%. The best threshold of HPI was 29.18% with the sensitivity of 67% and specificity 89%. Conclusion：The total fiver perfusion CT can accurately reflect the normal liver tissue perfusion of the hepatic segments, and prompts the pa- rameters among each segment are similar. Perfusion parameters can reflect the changes in blood flow of liver segment with OHM, and thus early diagnosis of OHM can be make.