Preclinical studies suggest that fluoxetine may have neuroprotective properties. In this pilot study forty-two patients with secondary or primary progressive MS were randomized to receive fluoxetine 20 mg twice daily or placebo for 2 years. Every 3 months the Expanded Disability Status Scale (EDSS), 9-hole peg test (9-HPT) and ambulation index (AI) were assessed. Brain MRI scans, Multiple Sclerosis Functional Composite, Fatigue Impact Scale, Guy's neurological disability Scale and SF-36 were performed at baseline, year 1 and year 2. Seven out of 20 (35%) patients in the fluoxetine group and 7 out of 22 (32%) patients in the placebo group had sustained progression on the EDSS, 9-HPT, or AI at 2 years. No differences were identified between the 2 treatment groups with respect to secondary clinical outcomes and T2 lesion load, grey matter volume and white matter volume. An unanticipated low rate of disability progression in the placebo group decreased the statistical power. At least 200 patients would have been needed to detect a 50% treatment effect. This trial shows that fluoxetine was generally well tolerated, but no assumptions can be made about a possible treatment effect. An adequately powered controlled trial of fluoxetine in progressive MS is still warranted. This trial is registered with Current Controlled Trials ISRCTN38456328.