The aim of this study was to evaluate the effort, accuracy, histological diagnoses and value of MR-guided vacuum biopsy (10-gauge, 9-gauge) as the diagnostic procedure for suspicious breast lesions visible on MRI alone. 389 MR-guided vacuum biopsies of suspicious MRM findings were performed in 365 patients either with a Vacora system (10G, Bard Company) or an ATEC system (9G, Suros Company). The retrospective study included the number of specimens, the table time, the complication rate, and the histopathological results for open biopsy and the findings after follow-up. The study included 341 unilateral unilocular, 12 unilateral bilocular and 12 bilateral unilocular MR-guided vacuum biopsies. In 27 patients (3.9%) the planned intervention was canceled because the lesion could not be reproduced. The average number of specimens was 15.1 (range 4 - 75) with the 10G technique and 14.6 (range 4 - 38) with the 9G technique. The table time was 43.2 min (range 17 - 95 min). Histology revealed concordant benign lesions in 231 cases (59.4%), borderline lesions in 50 cases (12.8%), malignant tumors in 106 cases (DCIS 30 [7.7%], invasive carcinoma 76 [19.5%]), and discordant findings in 2 cases (0.5%). The complication rate was less than 1%. MR-guided vacuum biopsy of the breast is an effective method for the minimally invasive percutaneous evaluation of suspicious breast lesions seen on MRI alone. As a consequence, primary open biopsy can be avoided and the rate of unnecessary surgical interventions reduced. There were no major differences between 10G and 9G vacuum biopsy systems.