PURPOSE: To compare the accuracy of preoperative breast tumor size measurements obtained on three imaging modalities (mammography [MM], sonography [US], and MRI) with those obtained on final pathologic examination for different breast densities and various tumor types. METHODS: Records from patients who underwent breast cancer lumpectomy between 2008 and 2012 and in whom tumor was seen on all three imaging modalities were retrospectively reviewed for maximum tumor size measurements. Patients with positive tumor margins and those who had undergone neoadjuvant chemotherapy were excluded. Tumor size measurements obtained on the three imaging modalities were compared for accuracy with those obtained during the final pathologic examination. Differences were analyzed for the whole group and for subgroups according to breast density and tumor type. RESULTS: In total, 57 patients were included, in whom wire-localization lumpectomy was performed without neoadjuvant chemotherapy; negative surgical margins for tumor were obtained, and tumor was preoperatively visualized on all three imaging modalities. The mean (± SEM) tumor size measured on MRI was significantly greater than that measured on pathology (p < 0.001), whereas the sizes measured on US and MM were not statistically significantly different from that measured on pathology (p = 0.62 and p = 0.57). Tumor size measured on MRI was greater than that measured on both US and MM (p = 0.003 and p < 0.001). Compared with the measurements obtained on pathology, that obtained on US showed moderate agreement (Lin concordance correlation coefficient [CCC], 0.71; 95% confidence interval [CI], 0.56-0.82); poorer agreement was found for the sizes obtained on MM (CCC, 0.58; 95% CI, 0.38-0.72) and MRI (CCC, 0.50; 95% CI, 0.31-0.65). No difference in comparative accuracy of size measurement was noted between dense and nondense breast tissue. MRI overestimated tumor size in ductal cancers (p < 0.001) and slightly underestimated it in lobular cancers. CONCLUSIONS: Preoperative MRI significantly overestimated tumor size. Measurements obtained on US and MM were more accurate irrespective of breast density, with US measurements being slightly more accurate than MM measurements.
PURPOSE: To compare the accuracy of preoperative breast tumor size measurements obtained on three imaging modalities (mammography [MM], sonography [US], and MRI) with those obtained on final pathologic examination for different breast densities and various tumor types. METHODS: Records from patients who underwent breast cancer lumpectomy between 2008 and 2012 and in whom tumor was seen on all three imaging modalities were retrospectively reviewed for maximum tumor size measurements. Patients with positive tumor margins and those who had undergone neoadjuvant chemotherapy were excluded. Tumor size measurements obtained on the three imaging modalities were compared for accuracy with those obtained during the final pathologic examination. Differences were analyzed for the whole group and for subgroups according to breast density and tumor type. RESULTS: In total, 57 patients were included, in whom wire-localization lumpectomy was performed without neoadjuvant chemotherapy; negative surgical margins for tumor were obtained, and tumor was preoperatively visualized on all three imaging modalities. The mean (± SEM) tumor size measured on MRI was significantly greater than that measured on pathology (p < 0.001), whereas the sizes measured on US and MM were not statistically significantly different from that measured on pathology (p = 0.62 and p = 0.57). Tumor size measured on MRI was greater than that measured on both US and MM (p = 0.003 and p < 0.001). Compared with the measurements obtained on pathology, that obtained on US showed moderate agreement (Lin concordance correlation coefficient [CCC], 0.71; 95% confidence interval [CI], 0.56-0.82); poorer agreement was found for the sizes obtained on MM (CCC, 0.58; 95% CI, 0.38-0.72) and MRI (CCC, 0.50; 95% CI, 0.31-0.65). No difference in comparative accuracy of size measurement was noted between dense and nondense breast tissue. MRI overestimated tumor size in ductal cancers (p < 0.001) and slightly underestimated it in lobular cancers. CONCLUSIONS: Preoperative MRI significantly overestimated tumor size. Measurements obtained on US and MM were more accurate irrespective of breast density, with US measurements being slightly more accurate than MM measurements.
Authors: Kathrin Ogris; Andreas Petrovic; Sylvia Scheicher; Hanna Sprenger; Martin Urschler; Eva Maria Hassler; Kathrin Yen; Eva Scheurer Journal: Forensic Sci Med Pathol Date: 2017-03-01 Impact factor: 2.007
Authors: Susanne Wienbeck; Johannes Uhlig; Uwe Fischer; Martin Hellriegel; Eva von Fintel; Dietrich Kulenkampff; Alexey Surov; Joachim Lotz; Christina Perske Journal: Medicine (Baltimore) Date: 2019-09 Impact factor: 1.817