论文部分内容阅读
目的 探讨~(18)F-FDG PET/CT术前诊断结直肠癌区域淋巴结转移的价值及其最佳诊断阈值。资料与方法 结直肠癌初诊患者76例,均在PET/CT检查后1周内行结直肠癌根治术,将所有与PET/CT相匹配的淋巴结依据其与原发肿瘤的位置关系分为近端和远端淋巴结,同时以病理结果为“金标准”,生成淋巴结短径与最大标准化摄取值(SUVmax)的受试者工作特性(ROC)曲线,分析其诊断效能。结果 近端与远端淋巴结ROC曲线示淋巴结短径、SUVmax的最佳阈值为6.5 mm、1.9和5.5 mm、1.81,淋巴结短径最佳阈值时诊断近端淋巴结转移的敏感度、特异度、准确度分别为84.85%、73.02%、77.52%,诊断远端淋巴结转移的敏感度、特异度、准确度分别为97.62%、65.45%、79.38%;SUVmax最佳阈值时诊断近端淋巴结转移的敏感度、特异度、准确度分别为84.85%、95.81%、91.64%,诊断远端淋巴结转移的敏感度、特异度、准确度分别为92.86%、94.55%、93.81%。SUVmax最佳阈值时诊断特异度、准确度均高于淋巴结短径最佳阈值(P<0.01),且SUVmax最佳阈值时与病理结果一致性较高(Kappa=0.813、0.874,P<0.01)。结论 最佳阈值法能提高18F-FDG PET/CT对结直肠癌区域淋巴结转移的诊断效能,且SUVmax标准优于淋巴结短径标准。
Objective To investigate the value of 18F-FDG PET/CT in the diagnosis of regional lymph node metastasis of colorectal cancer and its optimal diagnostic threshold. Materials and Methods 76 newly diagnosed patients with colorectal cancer underwent colorectal cancer radical resection within 1 week after PET/CT examination. All lymph nodes matched with PET/CT were classified as proximal according to their positional relationship with the primary tumor. With the distal lymph nodes, the ROC curves of the short-term lymph nodes and the maximum normalized uptake (SUVmax) were generated with pathological results as the “gold standard”, and the diagnostic efficacy was analyzed. Results The ROC curves of the proximal and distal lymph nodes showed the short-term lymph node diameter and the optimal threshold of SUVmax were 6.5 mm, 1.9 and 5.5 mm, and 1.81. The sensitivity, specificity, and accuracy of the diagnosis of proximal lymph node metastasis when the lymph node short-thickness threshold was optimal. The degrees of sensitivity, specificity, and accuracy of diagnosis of distal lymph node metastasis were respectively 84.85%, 73.02%, and 77.52%, which were 97.62%, 65.45%, and 79.38%. The sensitivity of proximal lymph node metastasis was diagnosed when the optimal threshold of SUVmax was reached. The specificity and accuracy were 84.85%, 95.81% and 91.64%, respectively. The sensitivity, specificity, and accuracy of diagnosis of distal lymph node metastasis were 92.86%, 94.55%, and 93.81%, respectively. The specificity and accuracy of the SUVmax optimal threshold were higher than the shortest diameter of the lymph node (P<0.01), and the SUVmax optimal threshold was consistent with the pathological results (Kappa=0.813, 0.874, P<0.01). . Conclusion The optimal threshold method can improve the diagnostic efficacy of 18F-FDG PET/CT in regional lymph node metastasis of colorectal cancer, and the SUVmax criterion is superior to the standard of lymph node shortness.