^(18)F-FDG PET-CT预测非小细胞肺癌胸腔镜肺叶切除中转开胸手术的效果

The predictive value of ^(18)F-FDG PET-CT in the conversion from thoracoscopic lobectomy to thoracotomy for non-small cell lung cancer

目的探讨18F-FDG PET-CT检查在非小细胞肺癌胸腔镜肺叶切除中转开胸手术中的预测价值。方法回顾性分析2014年1月至2021年6月于中日友好医院就诊的193例原发性非小细胞肺癌胸腔镜手术患者的临床资料及CT、PET-CT图像资料。193例患者中,中转开胸手术45例(中转组),未中转开胸手术148例(非中转组)。对两组临床病理特征和影像参数进行单因素分析,采用多因素logistic回归分析影响中转开胸手术的因素。以最终是否中转开胸手术为金标准,通过受试者工作特征(ROC)曲线分析筛选出的变量预测中转开胸的效果,采用Delong检验比较各变量预测的曲线下面积(AUC)。结果中转组胸部CT的支气管周围或血管周围淋巴结(PLN)、支气管周围软组织袖带(PCS)、胸膜钙化、肺结节钙化、PLN或PCS钙化或密度增高者的比例均高于非中转组(均P<0.05),PET最大标准摄取值(SUV_(max))较非中转组高(P<0.001)。多因素logistic回归分析示,年龄(OR=4.663,95%CI 2.191~9.923,P<0.001)及胸部CT扫描的PLN或PCS密度(OR=2.824,95%CI 1.791~7.303,P<0.001)均为胸腔镜肺叶切除中转开胸手术的独立影响因素。经ROC曲线分析,18F-FDG PET和胸部CT组合预测中转开胸手术的效能[AUC=0.891(95%CI 0.831~0.951);SUV_(max)最佳临界值为3.45、CT最佳临界值为70 Hu:灵敏度84.4%,特异度83.8%]优于单纯胸部CT[AUC=0.678(95%CI 0.591~0.766);CT最佳临界值为70 Hu:灵敏度62.2%,特异度62.8%;P<0.001]及年龄[AUC=0.625(95%CI 0.532~0.719);最佳临界值为65.5岁:灵敏度75.6%,特异度60.1%;P<0.001]。结论胸部CT扫描的PLN或PCS密度及患者年龄对非小细胞肺癌胸腔镜肺叶切除中转开胸手术有预测价值,CT与PET联合预测胸腔镜肺叶切除术是否中转开胸手术具有附加作用。

Objective To investigate the predictive value of 18F-FDG PET-CT in the conversion from thoracoscopic lobectomy to thoracotomy for non-small cell lung cancer(NSCLC).Methods The clinical data,CT and PET-CT images of 193 patients with primary NSCLC undergoing thoracoscopic surgery from January 2014 to June 2021 at China-Japan Friendship Hospital were retrospectively analyzed.All patients were divided into 45 cases who were converted to thoracotomy(the conversional group)and 148 cases who were not converted to thoracotomy(the non-conversional group).Univariate analysis was performed on clinicopathological characteristics and image parameters.Multivariate logistic regression was used to analyze the factors affecting the conversion to thoracotomy.Taking the final conversion to thoracotomy or not as the gold standard,the predictive effect of variables in the conversion to thoracotomy was analyzed through the receiver operating characteristic(ROC)curve.Delong test was used to compare the area under the curve(AUC)predicted by all variables.Results In the conversional group,the proportion of peribronchial lymph node(PLN),peribronchial cuffs of soft(PCS),pleural calcification,pulmonary nodule calcification,PLN or PCS calcification or increased density on chest CT was higher than that in the non-conversional group(all P<0.05);and the maximum standardized uptake value(SUV_(max))of PET in the conversional group was higher compared with that in the non-conversional group(P<0.001).Univariate logistic regression analysis suggested age(OR=4.663,95%CI 2.191-9.923,P<0.001)and PLN or PCS density of chest CT scan(OR=2.824,95%CI 1.791-7.303,P<0.001)were independent influencing factors of the conversion from thoracoscopic lobectomy to thoracotomy.ROC analysis showed that the effect of the conversion to thoracotomy predicted by the combination of 18F-FDG PET and chest CT[AUC=0.891(95%CI 0.831-0.951);the optimal cut-off value of SUV_(max) and CT was 3.45,70 Hu:the sensitivity was 84.4%,the specificity was 83.8%]was better than that by chest CT alone[AUC=0.678(95%CI 0.591-0.766);the optimal cut-off value of CT was 70 Hu:the sensitivity was 62.2%,the specificity was 62.8%;P<0.001]and by age[AUC=0.625(95%CI 0.532-0.719);the optimal cut-off value was 65.5 years:the sensitivity was 75.6%,the specificity was 60.1%;P<0.001].Conclusions PLN or PCS density on chest scan and age are valuable in predicting the conversion from thoracoscopic lobectomy to thoracotomy for NSCLC patients.The combination of PET and CT has an additional role in predicting the conversion to thoracotomy during thoracoscopic lobectomy.