CT能谱成像定量碘基物质图对肺栓塞的诊断价值

Spectral CT imaging in the diagnosis of pulmonary embolism using quantitative iodine-based material decomposition images

目的评价CT能谱成像定量碘基物质图对肺栓塞（PE）的诊断价值。方法53例怀疑PE的患者行CT常规平扫及能谱增强扫描，并进行数据后处理，同时获得单能量肺CTA及碘基物质图像。观察肺动脉内有无栓子，记录栓子的位置、数目及栓塞程度；分析碘基物质图表现，记录肺内灌注异常的位置及数目，测量碘含量。计数资料比较采用Х^2检验，不同碘含量比较分别采用两样本的秩和检验及符号秩和检验。结果33例患者的碘基物质图密度均匀，CTA未显示PE；19例患者共显示93个栓子，其中肺叶26个，肺段54个，亚段13个。51个栓子为闭塞性，42个栓子为非闭塞性。所有闭塞性栓塞均有灌注减低表现，非闭塞性栓塞中11个显示灌注降低。CTA栓子的栓塞程度与肺内有无灌注减低区差异有统计学意义（X。=39．94，P〈0．01）。正常肺实质区[（1．92±0．54）g／L]与灌注减低区[（0．30±0．20）g／L]碘基物质含量差异有统计学意义（Z=-5．63，P〈0．01）。PE低灌注区抗凝治疗前[（0．26±0．23）g／L]与治疗后[（0．94±0．50）g／L]的碘基物质含量差异有统计学意义（Z=-3．93，P〈0．01）。结论CT能谱成像可以为PE提供定性、定量分析，可以作为评价病变程度以及指导治疗的有效手段。

Objective To assess the diagnostic value of CT spectral imaging using quantitative iodine-based material decomposition images in the evaluation of pulmonary embolism. Methods Fifty-three patients underwent CT angiography with spectral imaging mode on a GE Discovery CT750HD scanner. Iodine distribution in the lung parenchyma using the iodine-based material decomposition images was quantitatively measured by post-processing. Monochromatic CT angiographic images were reconstructed from the same data sets and these images were reviewed for the identification and localization of pulmonary embolism as well as the degree （ partial or complete） of the embolic occlusion. The number and location of perfusion defects were recorded. The iodine content of perfusion defects and normal lung parenchyma on the iodine maps were measured by one reader using an ROI analysis. Comparative analyses were obtained using the Chi-square test for categorical data. Two independent samples rank test and 2 related samples signed-rank test were used to compare iodine densities between different groups. Results CT angiography showed no pulmonary embolism in 33 patients, and iodine distribution was homogeneous. A total of 93 clots with lobar （ n = 26）, segmental （n = 54） and sub-segmental （n = 13） distribution were detected in 19 patients; Fifty-one clots were occlusive and 42 clots were non-occlusive. The iodine-based material decomposition images of all occlusive clots showed lobar, segmental or sub-segmental iodine distribution defects; whereas eleven of 42 non- occlusive clots had evidence of iodine distribution defects. There was significant difference （ Х^2= 39. 94, P〈0.01） in the perfusion defects between occlusive and non-occlusive clots. There was a significant difference in iodine content between normal lung parenchyma [（ 1.92 ± 0. 54） g/L] and perfusion defects [ （0. 30 ± 0. 20）g/L ] （Z = -5.63, P 〈 0. 01 ）. There was a significant difference in the iodine content of perfusion defects before [ （0. 26 ± 0. 23 ）g/L ] and after anticoagulation [ （0. 94 ± 0. 50 ）g/L ]（Z = -3.93, P 〈0. 01 ）. Conclusion With the ability of iodine mapping, CT spectral imaging is areliable method in the evaluation of pulmonary embolism both qualitatively and quantitatively, and may be a useful tool in providing information regarding the severity of PE and monitoring therapeutic efficacy.