CT灌注预测急性缺血性卒中患者缺血半暗带和核心梗死区的最佳灌注参数阈值探讨

Thresholds of CT perfusion in predicting ischemic penumbra and infarct core in patients with acute ischemic stroke

目的：探讨CT灌注（CTP）预测缺血半暗带和核心梗死区的最佳灌注参数阈值.方法：回顾性分析2009年6月至2013年10月在浙江大学医学院附属第二医院神经内科接受静脉溶栓治疗并行基线CTP检查的39例前循环脑梗死患者的临床及影像学资料.根据24 h再灌注率界定无再灌注组（10例）和再灌注组（21例）,分别用于评价缺血半暗带与核心梗死区阈值.根据基线CTP和复查CTP分别计算各参数阈值所对应的低灌注体积、最终梗死体积.采用配对t检验、相关性分析、Bland-Altman图分析各灌注参数阈值所对应的低灌注体积与最终梗死体积的一致性,得出最佳阈值.结果：无再灌注组中延迟时间〉3 s与最终梗死体积一致性最高（偏差3.3 ml,95%一致性区间-41.7~48.3;r=0.933,P〈0.001）;再灌注组中相对脑血流量〈30%与最终梗死体积一致性最高（偏差-2.2 ml,95%一致性区间-25.6~ 21.2;r=0.923,P〈0.001）.结论：延迟时间〉3 s和相对脑血流量〈30%分别是预测半暗带、核心梗死区体积的最佳CTP参数阈值.

Objective： To determine the optimal parameters and their thresholds on CT perfusion （CTP） to predict the penumbra and core in patients with acute ischemic stroke. Methods： The data of 39 thrombolytic candidates with acute cerebral anterior- circulation ischemic stroke admitted in the Second Affiliated Hospital, Zhejiang University School of Medicine from June 2009 to October 2013 were retrospectively reviewed. Patients all underwent CTP at admission and CTP or magnetic resonance perfusion （MRP） 24 h after thrombolysis. Patients were classified as non-reperfusion group （ to define the threshold of penumbra, n = 10） and reperfusion group （ to define the threshold of infarct core, n = 21 ） by reperfusion status. According to the baseline CTP and 24 h imaging, the volumes of threshold-based hypoperfusion lesions and final infarction were calculated. Paired t test, correlation analysis and Bland-Ahman plot were performed to assess the optimal thresholds for predicting the penumbra and infarct core. Results： In non-reperfusion group, the best agreement was found between final infarct volume and delay time 〉3 s （bias 3.3 ml, 95% limits of agreement： -41.7 to 48.3 ml, r = 0.933, P 〈 0.001 ）, while in reperfusion group, the best agreement was noted between final infarct volume and rCBF 〈 30% （bias -2.2 ml, 95% limits of agreement ： -25.6 to 21.2 ml; r =0. 923, P 〈0. 001 ）. Conclusion： Delay time 〉3 s and rCBF 〈 30% are the optimal thresholds for predicting the penumbra and infarct core on CTP, respectively. These thresholds may be of help to estimate the mismatch status and select eligible patients for thrombolysis.