In this study we assessed the safety, efficacy and cost-effectiveness of the use of triple dose gadolinium-DTPA (Gd) in serial monthly brain MRI of patients with multiple sclerosis, such as could be selected for clinical trials. The number of enhancing lesions, the number of new enhancing lesions and the number of active scans were used to evaluate the sensitivity of the contrast-enhanced MRI to disease activity. The dose of Gd, and the effect of introducing a delay between the contrast injection and the scan were both appraised. Every 4 weeks for 3 months, and in two separate sessions, scans were obtained from 40 patients with relapsing-remitting or secondary progressive multiple sclerosis, 5 min (early) and 20 min (delayed) after a standard dose (0.1 mmol/kg) or triple dose (0.3 mmol/kg) Gd injection. There were 435 enhancing lesions (242 of which were new) on the early standard dose scans, 479 (263 new) on the delayed standard dose, 772 (365 new) on the early triple dose and 827 (404 new) on the delayed triple dose. There were 109 scans revealing active disease on the early standard dose scans, 112 on the delayed standard dose, 119 on the early triple dose and 120 on the delayed triple dose. Statistical simulations indicated that the sample sizes needed for both cross-over and parallel-group trials with similar powers are lower if serial monthly triple dose MRI is used. No side-effects were reported and no significant changes in blood test parameters were found throughout the study. This study shows that the serial use of triple dose Gd is safe, and that it increases the sensitivity of serial monthly enhanced MRI in detecting multiple sclerosis activity significantly. Its use should enable preliminary trials of experimental therapies for multiple sclerosis to be conducted in small patient populations, over a short period of time.