Treatment of Bell Palsy: Title and subTitle BreakTranslating Uncertainty Into Practice

Bell palsy is an inflammatory condition of the facial nerve that is associated with herpes simplex or varicella virus infections and can result in temporary or permanent facial disfigurement and pain. With an annual incidence of 13 to 43 per 100 000 population, more than 60 000 cases are diagnosed each year in the United States.^{1 - 2} Approximately 15% of patients with Bell palsy will have persistent facial nerve dysfunction and related impairments in quality of life.³

Evidence about the efficacy of medications for improving the resolution of Bell palsy has accumulated rapidly in recent years. Both corticosteroids and antiviral agents such as acyclovir and valacyclovir have been extensively studied in randomized controlled trials. In 2004, a systematic review of 4 trials suggested that corticosteroids were not superior to placebo.⁴ However, 2 large randomized controlled trials published in 2007 and 2008 showed that short-term corticosteroid use significantly improved recovery of facial nerve function.^{3 ,5}

In this issue of JAMA, the meta-analysis by de Almeida et al⁶ further confirms the benefit of corticosteroids for treatment of Bell palsy. The findings of this meta-analysis of 10 trials suggest that treatment of 11 patients with corticosteroids leads to satisfactory resolution in 1 patient who otherwise would not experience resolution (ie, number needed to treat of 11). This evidence is not likely to be contradicted by additional clinical trials and thus has important implications for clinical practice.⁷

Evidence about the benefits of antiviral medications for Bell palsy has accumulated at a comparable rate. However, it has not led to a similar degree of clinical confidence about the efficacy of antiviral therapy. A systematic review in 2004 concluded that antivirals provided no benefit for resolution of Bell palsy.⁸ Two large trials using factorial designs that tested corticosteroids and antivirals separately and in combination found no benefit from antiviral therapy.^{3 ,5} The systematic review by de Almeida and colleagues⁶ confirms that antiviral medications are ineffective as single agents. However, in contrast to the primary trials, the meta-analysis suggests that antivirals may be beneficial when combined with corticosteroids, although the relative risk of 0.75 did not reach statistical significance (P = .05). Additional large clinical trials will be required to resolve this issue because the marginal benefit of antiviral medications in patients also receiving corticosteroids appears to be relatively small. Nevertheless, the availability of generic, low-cost antiviral agents such as acyclovir and the absence of major adverse effects suggest that such trials are warranted.⁶

In the world of clinical research, knowledge is cumulative but certainty can remain elusive. A new randomized controlled trial can either reinforce or challenge existing evidence about the efficacy of a treatment because of differences in study design, enrollment criteria, or even random variation. Because any trial may be difficult to assess in isolation, systematic reviews or meta-analyses allow assessment of the entire body of evidence. Such reviews may identify benefits or harms of treatment that are not evident from individual trials, define questions that should be addressed in subsequent studies, and help inform clinical practice. Readers must remember that these reviews summarize the clinical evidence at a given time point, and that subsequent primary trials can contradict the findings of the aggregated literature.⁹

Translating research into improved clinical practice can be slow even when the findings are consistent and treatment efficacy is high.¹⁰ Translating the findings by de Almeida et al⁶ about the role of antiviral medications in Bell palsy is even more difficult because of residual clinical uncertainty. When the scientific evidence is not definitive, translational messages can take 3 forms. The first, chosen by de Almeida and colleagues, is to qualify their conclusions and call for more primary trials. Most researchers will be comfortable with this strategy. However, clinicians and patients want clear evidence that can inform their daily treatment decisions. Given the possibility of marginal benefit and the absence of major harm with antiviral therapy, clinicians may convert uncertain knowledge into definitive action by adding an antiviral medication to a corticosteroid for their next patient with Bell palsy.

Researchers also may adopt a second approach to translating evidence into practice by overstating the conclusions of their research. If a researcher contends that a study is conclusive when in fact the findings are incremental, subsequent, contradictory studies can make a fundamentally healthy process of knowledge accumulation and self-correction appear like a simple disagreement between experts.

A third approach is to transfer the burden of decision making to the clinician and the patient. In essence, the message is: here are the facts according to the available evidence, now the clinician and the patient must decide. This approach has substantial appeal because it encourages communication and partnership between clinicians and patients. In addition, it allows clinicians to share uncertainty with their patients by acknowledging that some clinical decisions are not clear-cut, and that the scientific basis of those decisions continues to change.¹¹ De Almeida et al⁶ adopt this approach when they note that considerations of treatment costs and patient values should enter into decisions about the use of antivirals.

By finding patterns and ambiguities that individual studies may not reveal, systematic reviews and meta-analyses help synthesize findings and highlight the residual uncertainty in a body of research, heeding Shakespeare's adage that “modest doubt is call’d the beacon of the wise.”¹² The systematic review by de Almeida et al⁶ of medications for treatment of Bell palsy helps resolve lingering doubt about the benefits of corticosteroids, but raises questions about the adjunctive role of antiviral medications. Until the next generation of clinical trials is completed, clinicians and patients will have to deal with substantial uncertainty in deciding whether to add antiviral drugs to corticosteroids for Bell palsy. By assessing how clinicians alter their prescribing patterns and how treatment guidelines are revised in response to this new evidence, it will be possible to learn more about how clinical uncertainty is translated into practice.