目的探讨18F-FDG 符合线路 SPECT/CT 显像半定量分析法预测乳腺癌腋窝淋巴结转移的最佳阈值.资料与方法以58例行腋窝淋巴结清扫术的乳腺癌患者为研究对象,18F-FDG 符合线路 SPECT/CT 测量腋窝淋巴结(L)与对侧相应部位腋窝本底(B)最大放射性计数比值(L/B 值).以病理为“金标准”,通过 ROC曲线建立预测乳腺癌腋窝淋巴结转移的最佳阈值,将最佳阈值下的判定结果与病理结果行一致性检验.结果58例患者共检测到119枚与手术区匹配的淋巴结(直径≥8 mm).L/B 值的 ROC 曲线下面积为0.938,最佳阈值为2.22.采用 L/B 值≥2.22预测腋窝淋巴结转移的灵敏度、特异度、阳性预测值、阴性预测值分别为88.9%、97.9%、98.5%、85.2%,与病理诊断的一致性较好(Kappa=0.862, P<0.05).结论采用 L/B 值≥2.22作为预测直径≥8 mm 的腋窝淋巴结转移的标准具有较高的准确性,可用于腋窝淋巴结转移的评估.%Purpose To investigate the optimal threshold of 18F-FDG coincidence SPECT/CT imaging using semiquantitative analysis method in the prediction of axillary lymph node metastasis in breast cancer. Materials and Methods Fifty-eight patients with breast cancer confirmed by axillary lymph node dissection were enrolled. Maximum radioactive count ratio (L/B value) of axillary lymph node (L) and the corresponding areas of contralateral axillary background (B) were measured by the 18F-FDG coincidence SPECT/CT. ROC curve was established to predict the optimal threshold of axillary lymph node metastasis in breast cancer, and Kappa analysis was used to do consistency test of judgment results under optimal threshold and pathologic findings (gold standard). Results 119 lymph nodes in 58 patients matched with operation area ( ≥ 8 mm diameter) were detected. The area under the ROC curve of L/B value was 0.938, the optimum threshold value was 2.22. Sensitivity, specificity, positive predictive value, and negative predictive value, using L/B ≥ 2.22 to predict axillary lymph node metastases as judge of malignant and pathological diagnosis criteria (Kappa=0.862, P<0.05), were 88.9%, 97.9%, 98.5%and 85.2%, respectively. Conclusion It has higher accuracy for the threshold of the L/B ≥ 2.22 as a predictor of axillary lymph node metastasis ( ≥ 8 mm diameter), and it can be used for the assessment of axillary lymph node metastasis.
展开▼
