锥光束CT与MRI术前评估乳腺癌病灶大小准确性的对比研究

Comparative Accuracy of Preoperative Tumor Size Assessment on Breast Cancer Based on Cone Beam Breast CT and MRI

目的对比锥光束乳腺CT(CBBCT)与乳腺MRI测量原发性乳腺癌病变大小及T分期的准确性,并分析可能影响测量准确性的因素。方法回顾性分析行术前CBBCT、MRI检查并有术后病理结果证实的157例原发性乳腺癌患者。以术后新鲜大体病理最大径为标准,以两种影像学测量乳腺癌病灶最大径与乳腺癌术后病理切除标本最大径差值的绝对值进行分组,差值绝对值≤0.5 cm者为一致,>0.5 cm者为非一致。并且利用两种影像方法以及大体病理所测肿瘤大小获得影像T分期和病理T分期。通过对两种影像学测量方法与大体病理所测得病灶大小的一致率以及肿瘤T分期的一致率行χ^(2)检验,比较CBBCT与MRI在测量乳腺癌病灶大小和肿瘤T分期准确性的差异。结果CBBCT与MRI测量乳腺癌病灶大小与手术病理的一致率差异无统计学意义(66.2%vs.61.1%,χ^(2)=0.881,P=0.348)。CBBCT高估率小于MRI(10.2%vs.18.5%,χ^(2)=4.384,P=0.036),但两者低估率差异没有统计学意义(23.6%vs.20.4%,χ^(2)=4.64,P=0.496)。CBBCT与术后病理测量病灶大小差值的绝对值小于MRI(0.52±0.54 vs.0.63±0.67;t=2.008,P=0.046),且差异主要发生在未绝经组(0.54±0.54 vs.0.77±0.84;t=2.564,P=0.012)。CBBCT与MRI对肿瘤T分期与病理T分期的一致率差异无统计学意义(71.6%vs.71.6%,χ^(2)=0.000,P=1.000),且两者对肿瘤T分期的高估率和低估率差异均无统计学意义(11.3%,16/141 vs.18.4%,26/141,χ^(2)=2.798,P=0.094;17.0%,24/141 vs.9.9%,14/141,χ^(2)=3.041,P=0.081)。结论CBBCT可以作为MRI的补充方法对乳腺癌病灶大小和分期进行术前评估,而且对于未绝经期患者的病灶测量具有更高准确性。

Objective To compare the accuracy of cone beam breast CT(CBBCT)and magnetic resonance imaging(MRI)in measuring breast lesion size and staging of breast cancer,and analyze which factors affect the accuracy.Methods In this retrospective study,the study cohort included 157 patients acquired who underwent preoperative CBBCT and MRI examinations with histopathologic confirmed primary breast cancer.Tumor size measurements obtained on the two imaging modalities were compared for accuracy with those obtained during the final pathologic examination.Referring to tumor size measurement on pathologic examination,all the lesions were divided into two groups.Concordance was defined as the discrepancy in diameter smaller or equal to 0.5 cm,and the discordance was more than 0.5 cm.Two imaging methods and tumor size measured by pathologic examination were used to obtain clinical T stage and pathological T stage.Concordance rates between MRI,CBBCT and pathology were compared using the Chi square test,and the discrepancy between CBBCT and MRI in measuring the size of breast cancer and the accuracy of staging were compared.Results The concordance rate with final pathologic examination was no significance difference between CBBCT and MRI in measuring the size of breast cancer lesions(66.2%vs.61.1%,χ^(2)=0.881,P=0.348).The overestimation rate of CBBCT was lower than that of MRI(10.2%vs.18.5%,χ^(2)=4.384,P=0.036),but the underestimation rate of both was no significance difference(23.6%vs.20.4%,χ^(2)=4.64,P=0.496).The absolute value of the difference between CBBCT and pathologic examination was less than that of MRI(0.52±0.54 vs.0.63±0.67;t=2.008,P=0.046),and the difference mainly occurred in the premenopausal group(0.54±0.54 vs.0.77±0.84;t=2.564,P=0.012).The concordance rate was no significance difference between CBBCT and MRI in staging of breast cancer lesions(71.6%vs.71.6%,χ^(2)=0.000,P=1.000),and the overestimation rate and underestimation rate were also no significance difference(11.3%,16/141 vs.18.4%,26/141,χ^(2)=2.798,P=0.094;17.0%,24/141 vs.9.9%,14/141,χ^(2)=3.041,P=0.081).Conclusion CBBCT can be used as a supplementary method of MRI to evaluate the lesion size of breast cancer and staging of breast cancer before operation,and it has higher accuracy for lesion measurement of non climacteric patients.