Objectives: To synthesize the published literature on the efficacy, effectiveness, and toxicity of hydroxyurea (HU) when used for treatment of sickle cell disease (SCD); and to review the evidence regarding barriers to its use.
Data sources: Articles cited in MEDLlNE, EMBASE, TOXLine, and CINAHL through June 30, 2007.
Review methods: Paired reviewers reviewed each title, abstract, and article to assess eligibility. They abstracted data sequentially and then independently graded the evidence.
Results: In one small, randomized trial of HU in children with SCD; the yearly hospitalization rate was lower with HU than placebo (1.1 versus 2.8, p=0.002). The absolute increase in fetal hemoglobin (Hb F%) was 10.7 percent. Twenty observational studies of HU in children reported similar increases in Hb F%, while hemoglobin concentration increased by roughly 1 g/dl. One large randomized trial tested the efficacy of HU in adults with SCD and found that after 2 years of treatment, Hb F% increased by 3.2 percent and hemoglobin increased by 0.6 g/dl, The median number of painful crises was 44 percent (p<0.001) lower among patients treated with HU. The 12 observational studies of HU enrolling adults with SCD supported these findings. Panelists from the Center for the Evaluation of Risks to Human Reproduction reviewed the literature for potential toxicities of HU. They concluded that HU does not cause a growth delay in children 5-15 years old. There were no data on the effects on subsequent generations following exposure of developing germ cells to HU in utero. Some evidence supported impaired spermatogenesis with use of HU. Although we identified six patients taking HU who developed leukemia, the evidence did not support causality. Similarly, the evidence suggested no association between HU and leg ulcers in patients with SCD, although there was in patients with other illnesses. The literature supported neutropenia, skin rashes and nail changes associated with use of HU, but was sparse regarding skin neoplasms or other secondary malignancies in SCD. Only two studies investigated barriers to use of HU. Perceived efficacy and perceived safety of HU had the largest influence on patients' (or parents' ) choice to use HU. Providers reported barriers to be patient concerns about side effects; and their own concerns about HU in older patients, patient compliance, lack of contraception, side effects and carcinogenic potential, doubts about effectiveness, and concern about costs.
Conclusions: HU is efficacious in children and adults with SCD; with an increase in Hb F%, and reduction in hospitalizations and pain crises. However, few studies have measured the effectiveness of HU for SCD in usual practice. The paucity of long-term studies limits conclusions about toxicities and about mortality. Future studies of interventions to overcome the barriers to use of HU in patients with SCD are necessary.