Williamson IG, Whelan TR.
The clinical problem of Bell's palsy:
is treatment with steroids effective?
Br J Gen Pract. 1996 Dec;46:743-7.

Objective
To review disease characteristics and to determine the effectiveness of steroid therapy for facial recovery in patients with Bell's palsy.

Data sources
Studies were identified by hand searching the literature on facial paralysis using the Index Medicus (1970 to 1995) and by searching MEDLINE and BIDS (1985 to 1995) using the keywords facial paralysis, Bell's palsy, steroids, drug therapy, prednisolone, adrenocorticotrophic hormone, cortisone, and steroid treatment. Additional studies were identified by scanning the bibliography of a relevant review; searching the Science Citation Index; contacting 6 steroid manufacturers; and contacting 5 ear, nose, and throat specialists.

Study selection
Studies were selected if they were randomised controlled trials (RCTs) that compared steroid therapy with either a placebo or an untreated control arm. Studies were excluded if the outcomes were difficult to categorize or if serious flaws in methods existed.

Data extraction
Data were extracted on recovery rates, study design, drug regimen, patient numbers and characteristics, and severity of palsy.

Main results
4 studies were included in the analysis (392 patients, age range 5 to 70 y). All trials included complete and incomplete palsy. Recovery rates for nontreated Bell's palsy ranged from 57% to 85% for hospitalized patients. Full recovery was likely for patients with partial paralysis. 1 trial compared cortisone with placebo, 2 trials compared prednisolone with placebo, and 1 trial compared prednisolone with an untreated control arm. Follow-up was from 5 months to 1 year. Pooled data showed an improvement in the complete facial recovery rate in patients who received steroid treatment compared with those who received control treatment (77% vs 68%, {P = 0.03}*). {The weighted absolute benefit increase of 9% means that 11 patients would need to be treated with steroid therapy for 1 additional patient to completely recover, 95% CI 6 to 117; the relative benefit increase was 13%, CI 1% to 27%.}* None of the trials alone showed a difference in complete facial recovery rates between patients who received steroid therapy and patients in the control groups. 1 trial showed that autonomic synkinesis was reduced in patients who received steroid therapy compared with those who were left untreated.

Conclusion
Compared with placebo or no intervention, steroid therapy improves the complete facial recovery rate in patients with Bell's palsy.

Source of funding: South West Regional Health Authority.+

For article reprint: Dr. I.G. Williamson, Primary Medical Care Group, School of Medicine, Southampton University, Southampton S016 5ST, England, UK. FAX 44-1703-701-125. E-mail I.G.Williamson@soton.ac.uk.

*Numbers calculated from data in article.

+Information provided by author.

---

Commentary

Existing negative RCTs of corticosteroid therapy in idiopathic facial nerve paresis (Bell's palsy) have inadequate power to detect a treatment effect because of the small numbers of enrolled patients . The review by Williamson and Whelan combines 4 RCTs that used similar entry and treatment approaches. Their approach seems well justified.

The data show that full recovery is highly likely for patients who never develop complete facial paresis. This finding suggests an approach of watchful waiting with such patients, with re-examination at about 10 to 14 days from onset for those seen early in the course to detect progression to complete paresis. Pain in patients with incomplete paresis can be treated by nonsteroidal anti-inflammatory drugs or analgesics, and great attention should be paid to the protection of the eyes from desiccation.

In patients who present with complete paresis or who progress to complete paresis after re-examination, the data from this review would support a course of oral corticosteroids, and I would concur with the recommendation of 1 mg/kg of body weight per day (up to 80 mg/d) for 7 to 10 days without tapering. The presence of corticosteroid contraindications should modify this plan because the benefit of treatment is not great and the likelihood of complications from corticosteroid ther-apy may tip the risk-benefit ratio away from treatment in these patients.

The cause of Bell's palsy is uncertain: A viral neuritis caused by such agents as herpes simplex or herpes zoster is a possible mechanism, even in patients who do not have a skin eruption. This possibility has led to the widespread empiric use of such antiviral agents as acyclovir or fam-ciclovir (usually in addition to corticosteroid treatment) in an attempt to improve Bell's palsy recovery rates. The cost of these agents and the residual uncertainty about the efficacy of corticosteroids makes it important that an RCT with enough power to evaluate both cortico-steroid and antiviral therapy in patients with Bell's palsy be done in the near future.

PRINTER FRIENDLY VERSION OF THIS PAGE