The aim of our study was to test the possibility of using image subtraction in detecting enhancing lesions in brain MR scans with and without magnetization transfer (MT) in multiple sclerosis (MS). Ten MS patients underwent 1.5-T MR imaging of the brain with spin-echo T1-weighted sequences with and without MT, repeated after 0.1 mmol/kg of an usual two-compartment paramagnetic contrast agent (Gadoteridol, Gd-HP-DO3A). Precontrast images were subtracted from postcontrast. Enhancing lesions were counted on the postcontrast images only (post-Gd), comparing pre- and postcontrast images by direct visual control (pre/post-Gd), and on the subtracted images (SI) only. Without MT, 36 enhancing lesions were counted on post-Gd, 36 on pre/post-Gd, and 59 on SI; using MT, 69, 52, and 50, respectively. Significant differences were found for pre/post-Gd without MT vs SI without MT ( p=0.028) and vs pre/post-Gd with MT ( p=0.012) as well as for pre/post-Gd with MT vs post-Gd with MT ( p=0.028). With pre/post-Gd, MT allowed the detection of 1.6 enhancing lesions per patient more than without MT. Whereas the SI without MT allow the detection of an increased number of enhancing lesions, SI with MT do not. An off-site final assessment allowed calculation of sensitivity and positive predictive value as follows: without MT were 63 and 94% (post-Gd), 67 and 100% (pre/post-Gd), 96 and 88% (SI); and with MT were 93 and 73% (post-Gd), 96 and 100% (pre/post-Gd), 91 and 98% (SI), respectively. Thus, SI seem to increase the sensitivity without MT; moreover, they could be used to correct the pseudoenhancement that impair post-Gd images with MT.