Introduction Parasites are under consideration as therapy for Crohn's disease and hookworm treatment is marketed for this purpose, despite a lack of controlled evidence. We therefore conducted a study of the effects of the hookworm Necator americanus on patients with mild to moderately active Crohn's.

Methods Fifty-six patients with a Crohn's disease activity index (CDAI) of 220–450 were randomised to receive 10 hookworm larvae or placebo (histamine) under a plaster patch on the forearm. Clinical activity and immunological changes were monitored over 12 weeks before hookworm was eradicated by treatment with mebendazole, and there was a final visit at 24 weeks. The primary analysis of effect was of the change from baseline at 12 weeks in the CDAI assessed by a generalised linear model (analysis of covariance) with change from baseline as response with terms for treatment arm and baseline CDAI as explanatory variables. Assuming a standard deviation of the change in the Crohn's disease activity index of 110 this study of 56 patients had 80% power to detect a difference of 85 in the CDAI with hookworm treatment compared to control (α=0.05).

Results Clean file was declared and the treatment code broken at 5 p, on 30 November 2009. Groups were well matched for age (mean 38.2±11.7 (SD) years for hookworm treatment, 42.2±13.2 for placebo), sex (50% vs 54% female), smoking, drug use and disease location. Preliminary analysis of the primary endpoint appeared to show a strong trend against any evidence of therapeutic benefit with hookworm. Patients treated with hookworm had a 64 point increase in CDAI (95% CIs ranging from a 15-point benefit to a 143-point deterioration). There was no obvious difference in advers events in each group.

Conclusion Our preliminary analysis appears to exclude any significant overall benefit from treatment with 10 Necator americanus larvae in active Crohn's Disease and raises the possibility that such treatment may even be harmful. Further analysis will be needed to determine whether apparent deteriorations in CDAI may reflect gastrointestinal symptoms induced by hookworm infection and whether there might be benefit in subgroups. Further work could explore whether different dosing regimens would be more successful.