A Cochrane review included 22 RCTs with a total of 1938 patients with neurocysticercosis. Six trials with viable lesions included 322 patients. Nine trials with non-viable lesions included 941 patients. Seven trials with mixed viable and non-viable lesions included 676 patients. Albendazole was used in all but two trials. All trials were based in Central and South America and South-East Asia.
• Viable lesions: In adults fewer participants with albendazole had lesions at follow up (RR 0.56, 95% CI 0.45 to 0.70; 2 trials, n=192) but no difference was detected for albendazole compared with no treatment for recurrence of seizures (one trial, n= 116). In a small trial fewer patients receiving albendazole had no resolution of symptoms at 3 months than in the control group (RR 0.25, 95% CI 0.07 to 0.93; n=15). There were no trials for viable lesions in children. No trials reported on deaths and hospital admissions.
• Non-viable lesions: In children seizures recurrence was less common with albendazole compared with no treatment (RR 0.49, 95% CI 0.32 to 0.75; 4 trials, n= 329). There was no difference detected in the persistence of lesions at follow up (6 trials, n= 570). There were no trials for non-viable lesions in adults. No trials reported on resolution of symptoms, deaths or hospital admissions.
• Trials including viable, non-viable or mixed lesions (in both children and adults): headaches were more common with albendazole alone (RR 9.49, 95% CI 1.40 to 64.45; 106 participants, 2 trials), but no difference was detected when giving albendazole with corticosteroids (one trial, n=116). There was no significant difference between albendazole and treatment groups in the number of deaths (2 trials, n= 470). For resolution of symptoms there was no difference between the albendazole and control groups up to 4 years (one trial, n=298).
Comment: The quality of evidence is downgraded by study quality (unclear allocation concealment) and indirectness (differences in studied patients and outcomes).