目的

探讨¹⁸ F-脱氧葡萄糖(¹⁸ F-FDG)在诊断乳腺癌及腋窝淋巴结转移的价值。

方法

对34例乳腺癌患者的乳腺病灶(以健侧为对照)及腋窝淋巴结¹⁸ F-FDG PET/CT显像结果进行定性分析和半定量分析,结合病理结果,计算两种方法诊断的灵敏度、特异度、准确性。两种方法诊断与病理诊断行一致性检验。对乳腺病灶恶性组及对照组组间最大标准化摄取值(SUVmax)均数的比较采用两独立样本t检验。

结果

两组间SUVmax均数比较显示:恶性组¹⁸ F-FDG摄取高于对照组,差异有统计学意义(P=0.00)。1例为大汗腺导管内癌,系原位癌,对¹⁸F-FDG摄取不明显,PET/CT图像上未见明显结节影,目测显示0级,SUVmax为1.5。该例病灶微小,小于PET/CT 5～8 mm 的分辨率,造成定性和半定量分析均漏诊。18F-FDG PET/CT 定性分析诊断乳腺癌的灵敏度、特异度、准确性分别为97.7%、100.0%、97.9%;半定量分析以SUVmax 2.0时,诊断乳腺癌的灵敏度、特异度、准确性为佳,分别为93.2%、100.0%、93.6%。定性分析与病理诊断的一致性极好,Kappa值为0.85,半定量分析与病理诊断有较好一致性,Kappa值为0.64。在腋窝淋巴结转移的诊断上,定性分析的诊断灵敏度、特异度、准确性分别为55.9%、96.1%、90.1%,半定量分析(以文献SUVmax 1.8为标准)的诊断灵敏度、特异度、准确性分别为45.8%、97.0%、89.3%。两种分析诊断与病理诊均较好一致性,Kappa值分别为0.57(定性分析)和0.51(半定量分析)。

结论

¹⁸ F-FDG PET/CT诊断乳腺恶性病灶时,以SUVmax 2.0为临界值进行半定量分析诊断具有较好的诊断价值;在诊断腋窝淋巴结转移方面,定性分析和半定量分析与病理诊均有较好的一致性。

Objective

To investigate the value of 18F-fluorodeoxyglucose(FDG)PET/CT i maging in diagnosis of breast cancer and axillary ly mph node metastases.

Methods

According to the pathological diagnostic results 34 breast cancer patients were recruited for the study.Their ¹⁸F-FDG PET/CT i maging data were analyzed qualitatively by visual assess ment and semi-quantitatively by calculating maxi mu m standardized uptake values(SUVmax).The sensitivity,specificity and accuracy of the t wo methods were measured,Kappa test was used for the consistency of each of the t wo PET/CT diagnostic methods with pathological diagnosis,and the mean of SUVmax bet ween groups was compared using t wo independent samples t test.

Results

The comparison of the mean SUVmax bet ween the 2 groups showed the ¹⁸ F-FDGuptake was greater in the malignant breast lesion group than in the control group,with statistical significant difference bet ween the three groups(P=0.00).One case of pathological apocrine intraductal carcino ma,cancer in situ,was negatively uptaken by PET/CT.The qualitative analysis showed zero grade and semiquantitative analysis showed SUVmax as 1.5,because the lesion size was so s mall,＜5-8 c m,which was beyond the uptake range of PET/CT.In diagnosing breast malignancy,the diagnostic accuracy,sensitivity and specificity of ¹⁸ F-FDG PET/CT qualitative analysis was 97.9%,97.7%and 100.0%,respectively;for the semi-quantitative analysis,when the cut-off point of SUVmax was set at 2.0,the sensitivity,specificity,and accuracy was 93.2%,100.0% and 93.6%,respectively.The kappa value was 0.85 for qualitative analysis,with excellent agreement,and 0.64 for semi-quantitatively.In detecting axillary ly mph node metastases,the sensitivity,specificity,and accuracy of qualitative analysis were 55.9%,96.1%and 90.1%,respectively;for semi-quantitative analysis,the cut-off point of SUVmax was set at 1.8 (referenced to literature),the sensitivity,specificity,and accuracy were 45.8%,97.0%and 89.3%,respectively;the kappa value was 0.57 for the qualitative analysis and 0.51 for semi-quantitative analysis.

Conclusion

For detecting malignant breast lesions with ¹⁸ F-FDGPET/CT,when the cut-off point of SUVmax was set at 2.0,semi-quantitative analysis is opti mal.For detecting axillary ly mph node metastases,both qualitative analysis(Kappa value=0.57)and semi-quantitative analysis(Kappa value=0.51)have median agreement with pathological diagnosis.