目的研究18F-脱氧葡萄糖(18FFDG)PET/CT在乳腺癌诊断和术前分期评估 中的价值。方法 将我院收治的69例疑似 乳腺癌患者作为研究对象，均于术前进行 18F-FDG PET/CT检查，以术后病理结果为 标准，分析18F-FDG PET/CT对乳腺癌诊断 的信效度及与病理结果诊断一致性，并分 析18F-FDG PET/CT对术前分期信效度。结 果 69例患者术后病理结果显示为乳腺癌 者67例，18F-FDG PET/CT检查结果显示SUV 阳性者65例，阴性者4例，18F-FDG PET/ CT乳腺癌诊断诊断灵敏度为97.01%，特 异度为100.00%，准确度为97.10%，与病 理结果诊断一致性较好(κ=0.896)。18FFDG PET/CT检查确诊为乳腺癌的65例患者 中，术后病理结果显示I期者6例，II期 者35例，III期者8例，IV期者16例；18FFDG PET/CT检查显示I期者7例，II期者32 例，III期者11例，IV期者15例；18F-FDG PET/CT检查对乳腺癌术前分期诊断总准确 度为89.23%。结论 18F-FDG PET/CT乳腺 癌诊断准确性较高，能较好评估术前肿瘤 分期。

Objective To study the value of 18F-FDG PET/CT in diagnosis and preoperative staging assessment of breast cancer. Methods A total of 69 cases of patients with suspected breast cancer treated in our hospital were selected for the study and were given 18F-FDG PET/CT before operation. The postoperative pathological results were used as criteria to analyze the reliability and validity of 18F-FDG PET/CT and its consistency with pathological diagnosis in the diagnosis of breast cancer and to analyze the reliability and validity of 18F-FDG PET/CT to preoperative staging. Results The postoperative pathological results of 69 cases showed there were 67 cases of breast cancer, and the results of 18F-FDG PET/CT showed that 65 cases were SUV positive and 4 cases were negative, and the diagnostic sensitivity, specificity and accuracy of 18F-FDG PET/ CT were 97.01%, 100% and 97.10% in the diagnosis of breast cancer, and its consistency with pathological diagnosis was better (κ=0.896). Among 65 cases diagnosed with breast cancer by 18F-FDG PET/CT, postoperative pathological results showed that there were 6 cases in I stage, 35 cases in II stage, 8 in III stage and 16 in IV stage. 18F-FDG PET/ CT showed that there were 7 cases in I stage, 32 cases in II stage, 11 in III stage and 15 in IV stage. The total accuracy of 18F-FDG PET/CT was 89.23% in the diagnosis of preoperative staging of breast cancer. Conclusion 18F-FDG PET/CT has high diagnostic accuracy in breast cancer, and can better evaluate preoperative tumor staging.