Inflammation and oedema of the facial nerve are implicated in causing Bell's palsy. Corticosteroids have a potent anti-inflammatory action which should minimise nerve damage and thereby improve the outcome of patients suffering from this condition. The objective of this review was to assess the effect of steroid therapy in the recovery of patients with Bell's palsy. We searched the Cochrane Neuromuscular Disease Group register for randomised trials, as well as MEDLINE, EMBASE and LILACS (to December 2000). We contacted known experts in the field to identify additional published or unpublished trials. Randomised trials comparing different routes of administration and dosage schemes of corticosteroid or adrenocorticotrophic hormone therapy versus a control group where no therapy considered effective for this condition was administered, unless it was also given in a similar way to the experimental group. Two reviewers independently assessed eligibility, trial quality, and extracted the data. Three trials with a total of 117 patients were included. One trial compared cortisone acetate with placebo; one compared prednisone plus vitamins, with vitamins alone; and one, not-placebo controlled, tested the efficacy of methylprednisolone. Allocation concealment was appropriate in two trials, and the data reported allowed an intention-to-treat analysis. Overall 13/59 (22%) of the patients allocated to steroid therapy had incomplete recovery of facial motor function six months after randomisation, compared with 15/58 (26%) in the control group. This reduction was not significant (relative risk 0.86, 95% confidence interval 0.47 to 1.59). The reduction in the proportion of patients with cosmetically disabling sequelae six months after randomisation was also not significant (relative risk 0.86, 95% confidence interval 0.38 to 1.98). The available evidence from randomised controlled trials does not show significant benefit from treating Bell's palsy with corticosteroids. More randomised controlled trials with a greater number of patients are needed to determine reliably whether there is real benefit (or harm) from the use of steroid therapy in patients with Bell's palsy.