The purpose of this study was to evaluate the safety and efficacy of hydroxyurea (HU) in spinal muscular atrophy (SMA) in a randomized, double-blind, placebo-controlled trial. Twenty-eight patients with type 2 SMA and 29 patients with type 3 SMA were randomly assigned (2:1) to receive HU or matching placebo for 18 months. HU was initiated at 10 mg/kg/day with an 8-week titration to 20 mg/kg/day. Subjects were assessed at baseline (T0) and monthly for the first 2 months (T1-T2) and then every 2 months throughout treatment (T3-T10) and posttreatment periods (T11-T13). The primary outcome measures were the Gross Motor Function Measure (GMFM), Manual Muscle Test (MMT), and serum full-length survivor motor neuron (flSMN) mRNA. The secondary outcome measures were Modified Hammersmith Functional Motor Scale and forced vital capacity (FVC). Fifty-five patients completed this trial, which lasted from March 2007 to June 2009. Except for neutropenia, we found no differences in adverse events between the 2 groups. Compared with the placebo group, the HU group had -1.88 for GMFM (p = 0.11), -0.55 for MMT (p = 0.49), and 2.17 for flSMN mRNA (p = 0.13). Similarly, we found no difference in mean improvement of the secondary endpoints. Both groups had a trend toward a decline in FVC with little change in strength and motor function. Under the current regimen and schedule, HU brought about no improvement in patients with type 2 and 3 SMA, and its main side effect was neutropenia. This trial provides Class I evidence that HU 20 mg/kg/day does not effectively treat SMA.