Source: Lockhart P, Daly F, Pitkethly M, Comerford N, Sullivan F. Antiviral treatment for Bell's palsy (idiopathic facial paralysis). Cochrane Database Syst Rev. 2009 Oct 7;(4):CD001869. Review. PubMed PMID: 19821283.
Efficacy Endpoints: Incomplete recovery of facial function, motor synkinesis or crocodile tears syndrome
Harm Endpoints: Adverse drug effects, unspecified
Narrative: Bell's palsy is a one-sided facial nerve paralysis affecting roughly 11-40 people per 100,000, and it is thought to potentially be caused by a herpes virus infection of cranial nerve VII, the 'facial nerve'. Because full recovery of nerve function takes weeks to months and only occurs in about 85% of people, antiviral treatment has been proposed as a way to speed or improve the chances of full recovery of the nerve. This review summarizes the data on antiviral treatment. The Cochrane review updates an earlier Cochrane review that concluded insufficient evidence existed. Since the earlier review, a number of large, high-quality studies have been published, leading to the current re-analysis.
The review includes 1987 patients, with 1787 more than in the earlier review due to the addition of 5 new studies. There was no identifiable benefit to taking antivirals in the treatment of Bell's palsy. When compared to placebo, the risk ratio was 1.14 for full recovery with a confidence interval, or margin of error, including 1.0 (confidence interval = 0.80-1.62), indicating no identifiable difference between antivirals and placebo. When evaluating trials that included patients receiving both antivirals and corticosteroids (see Caveats below), the risk ratio was 0.88, and again the confidence interval included 1.0 (0.65-1.18). Only 1 study with <100 subjects evaluated motor synkinesis (involuntary facial contractions or eye tearing) and found no statistically identifiable benefit to antiviral drugs. The authors conclude that there is no benefit and that the medical community should perhaps reconsider the theory implicating the herpes virus as a causal agent in Bell's palsy.
Caveats: Due to another theory that corticosteroids may have an impact on Bell's palsy (see this NNT review for more information), many of these studies had multiple treatment arms leading to the possibility of overlapping effects: antiviral + corticosteroids vs corticosteroids vs. antivirals vs. placebo, for example. This makes it more difficult to assess antivirals separately, however the authors of this review address this in their analysis and find that treatment with corticosteroids led to a better outcome than treatment with antivirals.
It is difficult to assess "medication adverse effects," given the above patient arms that included corticosteroids. Common side effects of antivirals include nausea, vomiting and diarrhea, stomach upset, and headache. Obviously some patients will experience these effects however, as is common in studies powered and designed to detect a possible benefit, these studies seem to have been much more diligent in their tracking and detection of beneficial effects than in their tracking of harms. Therefore in this meta-analysis including over 1500 patients there was no identifiable difference between the treatment and placebo arms for medication adverse effects.
Author: Graham Walker, MD
Published/Updated: September 1, 2010
The title bar is color-coded with our overall recommendation.
If you have suggestions, requests, or questions about a particular NNT review, please send us a message and we’ll try to address it as soon as possible.