RESUMO
The estimation of tumor size is important for treatment strategies of breast cancer. The hyperechogenic zone around breast cancer is a recognized criterion for malignancy, but its impact on preoperative tumor size estimations has been poorly investigated. Data of prospectively maintained database of 513 patients with primary breast tumors were analyzed retrospectively. A total of 196 patients with complete datasets including preoperative ultrasound (US) were eligible for analysis. The median age of the patients was 58.5 years (range 33-87). With all of the 196 patients, US has been performed. In 170 of 196 (86.7 %) cases, an echogenic halo was detected. We use two ways to measure tumor size with US: without (US-0) and with (US-1) echogenic halo. Mammography (MG) was used as standard. Tumor size measured by US and MG was compared with the actual histopathological (HP) tumor size. Mean differences between the sizing obtained by US-0, US-1, and MG and the HP sizing were -6.5, -1.5, and -1.8 mm, respectively. All three methods tend to underestimate the tumor size. The US-1 measurement was the closest to the HP size in comparison to the MG and US-0 measurements and the match was higher in tumors <2 cm. The estimated Pearson correlation coefficients (r) were 0.72, 0.68, and 0.61 for US-1, US-0, and MG, respectively. Moreover, the predictive value of US-1 regarding tumor size was not influenced by histological type and grade of the tumor, receptor status, and presence of intraductal component. Estimation of tumor size by US should include the hyperechogenic zone around the tumor.