In Summary, for those who took the steroids:

Benefits in NNT

  • 90.7% saw no benefit
  • 9.3% were helped by being saved from incomplete recovery of their facial paralysis
  • 8.3% were helped by avoiding facial motor nerve after effects
  • 1 in 11 were helped (complete recovery)
  • 1 in 12 were helped (no 'motor synkinesis')

Harms in NNT

  • 0% were harmed by adverse drug effects
  • None were harmed (medication adverse effects)

View As:

Source: Salinas RA, Alvarez G, Daly F, Ferreira J. Corticosteroids for Bell's palsy (idiopathic facial paralysis). Cochrane Database Syst Rev. 2010 Mar 17;3:CD001942. Review. PubMed PMID: 20238317.

Efficacy Endpoints: Incomplete recovery of facial function after 6 months, cosmetic after effects of paralysis, motor synkinesis or autonomic dysfunction

Harm Endpoints: Adverse drug effects

Narrative: Bell's palsy, or unilateral facial nerve paralysis affects 11-40 people per 100,000. One theory is that Bell's palsy is an inflammatory reaction. For this reason corticosteroids, which are anti-inflammatory agents, have been studied to improve the outcome of patients with Bell's. This review summarizes the data on corticosteroid treatment.



The review includes 1,507 subjects from 7 trials, 4 of which occurred since 2007. There was a benefit found to taking steroids. On the secondary outcome of cosmetically disabling persistent after effects (variably defined), no benefit was found. This outcome included 5 trials with 668 subjects, with a risk ratio of 0.97 and wide confidence intervals (0.44-2.15) that include 1.0, indicating no identifiable effect. Steroids did, however, improve facial motor after effects. The authors also note a small number of adverse drug effects (3 insomnia, 1 stable heart arrhythmia), and 3 deaths in the combined trials, though all were among the placebo group, and none were thought to be related to the study or the Bell's palsy.



The authors of the review suggest that the effect of corticosteroids on resolution of Bell's palsy symptoms is significant enough that future trials should not be conducted using a placebo arm.

Caveats: The authors point out the "marginally significant" heterogeneity (i.e. variation in results) of these trials (P = 0.04, I2 = 55%), but note that two trials showed larger improvements than others, while all trials showed a benefit There was also a smaller benefit seen in trials with more subjects that started the steroids 48 hours or more after symptom onset.



Also, in other literature on steroids harms have been identified. While they are not identified in this review they certainly exist, though they are likely minor compared to the benefits of treatment. Finally, an optimal dosage of steroids that may provide the largest benefit while exposing patients to the lowest risk has not been identified.

Author: Graham Walker, MD

Published/Updated: September 1, 2010

  1. The Title Bar

    The title bar is color-coded with our overall recommendation.

    • Green: Benefits outweigh risks.
    • Yellow: Unclear risk/benefit profile.
    • Red: Benefits do not outweigh risks.
    • Black: Obvious harms, no clear benefits.
  2. Tip content...