Treating neurocysticercosis medically: a systematic review of randomized, controlled trials

OBJECTIVE To summarize the evidence from randomized controlled trials on th e effects of cysticidal therapy used for treating human cysticercosis. METHODS Published and unpublished studies in any language identified throug h MEDLINE (1966 - June 1999) specialized databases, abstracts, proceedings and contact with experts were analysed. Those which compared, using randomi zed or quasi-randomized methods, any cysticidal drug with placebo or sympto matic therapy were entered in the study. Data were extracted independently by two reviewers and trial quality assessed. Meta-analysis using fixed effe cts models calculated provided there was no significant heterogeneity, expr essed as relative risk. RESULTS Four trials met the inclusion criteria, treating intraparenchymatou s neurocysticercosis with either albendazole or praziquantel compared to pl acebo or no treatment. In the two trials reporting clinical outcomes, treat ment was not associated with a reduction in the risk of seizures, although numbers were small (RR 0.95, 95% CI 0.59-1.51). Four trials reported radiol ogical outcomes, and cysticidal treatment was associated with a lower risk of cyst persistence of scans taken within six months of start of treatment (RR 0.83, 95% CI 0.70-0.99). Subsidiary analysis assuming different outcome s in patients lost to follow-up did not alter the findings of the main anal ysis. CONCLUSIONS There is insufficient evidence to determine whether cysticidal therapy is of any clinical benefit to patients with neurocysticercosis. The review does not exclude the possibility that more patients remain seizure- free when treated with cysticidal drugs. Further testing through placebo-co ntrolled trials is required.