The aim of the study was to compare the different approaches of pre-operative diffusion-tensor-imaging-based fibre tracking (FT) of the corticospinal tract (CST) focusing on the positioning of the seeding region of interest (seed ROI). Thirty-nine patients with brain lesions in the vicinity of the CST were evaluated pre-operatively. Imaging comprised a 3D T1-weighted sequence, a gradient echo echo-planar imaging sequence for functional magnetic resonance imaging (fMRI), and a diffusion-weighted sequence for diffusion tensor (DT) tractography. DT tractography was performed with two different procedures to track the corticospinal fibres: one downwards and one upwards. Downward FT was started with the seed ROI in the pre-central gyrus subjacent to the maximal fMRI activity while for the upward FT seed ROI was placed in the cerebral peduncle. In 16 patients, tracking results were individually compared with the unaffected contralateral hemisphere. Results were correlated with fractional anisotropy (FA) values and other factors potentially influencing fibre tracking results. On the side with the space-occupying lesion, downward FT yielded more positive tracking results (tracked fibres > 0) than the upward FT. On both the affected and the unaffected side, downward FT reconstructed fewer fibres than upward FT. For none of the two methods did the tracking results (number and volume of fibres) correlate with FA values or with other clinical data. FA values for tracts ipsilateral to the lesion correlated with age and lesion entity. We conclude that the sequence of ROI positioning influences significantly the tracking results. Upward FT may fail to track fibres, whereas the successful tracking results may be superior to downward FT. Hence, upward FT of the CST should be preferred in patients with space-occupying lesions. Downward FT should be performed if upward FT fails.