IntroductionMagnetic resonance imaging (MRI) is a robust neuroimaging technique and is the preferred method for stereotactic radiosurgery (SRS) planning. However, MRI data always contain distortions caused by hardware and patient factors. Research questionCan these distortions potentially compromise the effectiveness and safety of SRS treatments? Material and methodsTwenty-six MR datasets with multiple metastatic brain tumors (METs) used for Gamma Knife radiosurgery (GKRS) were retrospectively evaluated. A commercially available software was used for distortion correction. Geometrical agreement between corrected and uncorrected tumor volumes was evaluated using MacDonald criteria, Euclidian distance, and Dice similarity coefficient (DSC). SRS plans were generated using uncorrected tumor volumes, which were assessed to determine their coverage of the corrected tumor volumes. ResultsThe median target volume was 0.38 cm3 (range,0.01–12.38 cm3). A maximum displacement of METs of up to 2.87 mm and a median displacement of 0.55 mm (range,0.1–2.87 mm) were noted. The median DSC between uncorrected and corrected MRI was 0.92, and the most concerning case had a DSC of 0.46. Although all plans met the optimization criterion of at least 98% of the uncorrected tumor volume (median 99.55%, range 98.1–100%) receiving at least 100% of the prescription dose, the percent of the corrected tumor volume receiving the total prescription dose was a median of 95.45% (range,23.1–99.5%). Discussion and conclusionMRI distortion, though visually subtle, has significant implications for SRS planning. Regular utilization of corrected MRI is recommended for SRS planning as distortion is sometimes enough to cause a volumetric miss of SRS targets.