AIMS: Investigate the efficacy of risedronate sodium (Procter and Gamble, Cincinnati, USA) for treating reduced lumbar spine (LS) bone mineral density (BMD) in non-ambulatory patients. METHODS: Nine (10-39 years, mean age 23.0 years, 7 males) in the risedronate arm and 10 (10-35 years, mean age 21.4 years, 8 males) in the placebo arm completed 24 months of therapy at baseline, 6, 12, 18, and 24 months. The primary outcome was change in LS BMD assessed by dual energy x-ray absorptiometry (DXA). Secondary outcomes included changes in serum bone markers, bone specific alkaline phosphatase, osteocalcin, and N-telopeptides. Mixed models examined group, time, and the group by time interaction for the 4 post-baseline time points. RESULTS: The change in LS BMD score from baseline to 24 months was 0.069 (95% CI 0.014 to 0.124) in risedronate participants compared to -0.015 (95% CI -0.073 to 0.042) (t Value = -2.40, P > t=0.03) in the controls. When controlling for baseline scores, the difference was consistent across four post-baseline time points tested (F=5.67, Pr > F=0.03). No differences in serum bone markers were observed. CONCLUSIONS: Risedronate increases LS BMD in non-ambulatory patients with minimal side effects.