Lymphatic filariasis and intestinal helminth infections are important disorders in tropical areas. Periodic treatment with albendazole is now used in many school-based intestinal helminth-control programmes. However, few such programmes exist for lymphatic filariasis, despite evidence that single-dose treatment with ivermectin can greatly reduce the concentration of Wuchereria bancrofti microfilariae in the blood for months to years. We aimed to assess the potential for school-based control of lymphatic filariasis by investigating the efficacy and tolerability or combined ivermectin and albendazole in Haitian schoolchildren. In January, 1996, we collected 832 20 microL capillary blood samples for inclusion in a randomised controlled study from children aged 5-11 years, and examined them by microscopy for W bancrofti microfilariae. Infected children were randomly assigned treatment with placebo (n = 29), a single 200-400 micrograms/kg dose of ivermectin (mean, 273 micrograms/kg, n = 28), 400 mg albendazole (n = 29), or a combination of 200-400 micrograms/kg ivermectin and 400 mg albendazole (n = 24). Children with high concentrations of microfilariae in the blood were admitted to hospital and adverse reactions were monitored for 3-5 days, otherwise children were examined at school or during a visit to their home. 4 months after treatment, we examined blood samples again for microfilariae. 113 microfilaraemic children were enrolled (mean age 7.8 years). 4 months after treatment, the proportion of children who remained positive for microfilariae was significantly lower in the ivermectin plus albendazole group (four [17%]), but there were no significant changes in the other three groups (20 [69%] placebo, 22 [76%] albendazole alone, 17 [61%] ivermectin alone remained positive; p = 0.004). Geometric mean microfilarial concentration decreased from 9.3 to 5.3 per 20 microL blood among children who received placebo; from 15.5 to 1.5 per 20 microL blood among those who received ivermectin only (p = 0.032); from 14.1 to 5.1 per 20 microL blood among those who received albendazole alone; and from 13.7 to 0.3 per 20 microL blood among those who received both ivermectin and albendazole (p = 0.0001). Systemic adverse reactions did not differ significantly between children who received ivermectin alone and those who were treated with ivermectin and albendazole [corrected]. For children with W bancrofti microfilaraemia, combined treatment with ivermectin and albendazole was more effective than treatment with ivermectin only, with no measurable increase in severity of adverse reactions.