Eighteen randomised controlled trials (RCTs) were included in the review (n=2,786 participants); eight trials assessed corticosteroids, seven trials assessed antiviral agents, and three trials assessed both interventions. The mean sample size was 155 (range 40 to 829). Median follow-up was six months (range 10 weeks to 12 months). Six trials failed to describe randomisation sequence generation. Seven trials did not adequately report allocation concealment. Ten trials used blinded outcome adjudicators. Six trials used an intention-to-treat analysis, or a variant of this. Four trials reported loss to follow-up of more than 20%. The quality of the evidence was considered to be high for the impact of corticosteroids on outcomes of unsatisfactory recovery and synkinesis and autonomic dysfunction, and moderate for other outcomes.
Corticosteroid monotherapy: There was a statistically significantly lower risk of unsatisfactory recovery with corticosteroid treatment alone (relative risk 0.69, 95% confidence interval (CI): 0.55 to 0.87; number-need-to-treat 11, 95% CI: 8 to 25; 10 RCTs).
Antiviral monotherapy: There was no statistically significant difference between groups treated with antiviral agents alone and control groups (two RCTs).
Combined corticosteroid and antiviral therapy: Antiviral agents plus corticosteroids were significantly more effective than antivirals alone (relative risk 0.48, 95% CI: 0.29 to 0.79; number-need-to-treat 6, 95% CI: 4 to 14; two RCTs). There was a trend towards greater efficacy for antiviral agents plus corticosteroids to be more effective than corticosteroids alone (relative risk 0.75, 95% CI: 0.56 to 1.00; number-need-to-treat 20, 95% CI: 11 to ∞; eight RCTs).
There was no evidence of significant statistical heterogeneity in any of the analyses, with the exception of antivirals alone (I2=47%).
There was a statistically significant effect of dose with corticosteroid treatment, with higher doses (over 450 mg) associated with a larger treatment effect (p=0.02). No significant interactions were found with other variables.
Corticosteroids, but not antivirals were associated with a significant benefit in short term recovery and reduction in synkinesis and autonomic dysfunction. There was no evidence for significant adverse events associated with either treatment (nine RCTs).
There was no evidence of publication bias.