Correspondence

Prednisolone or Acyclovir in Bell's Palsy

N Engl J Med 2008; 358:306-307January 17, 2008

Article

To the Editor:

Sullivan et al. (Oct. 18 issue)1 report that early treatment with prednisolone significantly improves the chances of complete recovery in patients with Bell's palsy, whereas no evidence of a benefit of acyclovir could be found. Idiopathic peripheral-facial-nerve paralysis of sudden onset is difficult to diagnose because other causes that are associated with a significantly higher risk of defective healing2 must be ruled out. Since zoster sine herpete — which can mimic the signs and symptoms of Bell's palsy — caused by varicella–zoster virus has an incidence rate of up to 34% among patients with “Bell's palsy,”3 information about the virologic backgrounds of all participating patients, obtained by means of serologic and polymerase-chain-reaction examinations, is needed. However, the authors have not provided virologic data to ensure the exclusion of patients with zoster sine herpete from this study.

Dirk Beutner, M.D.
University of Cologne, 50924 Cologne, Germany
dirk.beutner@uni-koeln.de

3 References

1
Sullivan FM, Swan IRC, Donnan PT, et al. Early treatment with prednisolone or acyclovir in Bell's palsy. N Engl J Med 2007;357:1598-1607
Full Text | Web of Science | Medline

2
Kawaguchi K, Inamura H, Abe Y, et al. Reactivation of herpes simplex virus type 1 and varicella-zoster virus and therapeutic effects of combination therapy with prednisolone and valacyclovir in patients with Bell's palsy. Laryngoscope 2007;117:147-156
CrossRef | Web of Science | Medline

3
Grosheva M, Guntinas-Lichius O. Significance of electromyography to predict and evaluate facial function outcome after acute peripheral facial palsy. Eur Arch Otorhinolaryngol 2007;264:1491-1495
CrossRef | Web of Science | Medline

To the Editor:

Sullivan and colleagues found no benefit of acyclovir (at a dose of 400 mg five times per day for 10 days) for the treatment of Bell's palsy, either alone or in combination with prednisolone. They excluded patients with herpes zoster. However, occult varicella–zoster virus reactivation (zoster sine herpete) may cause up to 29% of cases of Bell's palsy.1 The optimal dose of acyclovir for the treatment of varicella–zoster virus infection (according to the package insert) is 800 mg five times per day. Thus, a significant subgroup of patients with Bell's palsy might still benefit from higher-dose antiviral treatment than that used in the study by Sullivan et al.

Steven Leiner, N.P.
Mission Neighborhood Health Center, San Francisco, CA 94110
stevenleiner@hotmail.com

1 References

1
Furuta Y, Ohtani F, Kawabata H, Fukuda S, Bergstrom T. High prevalence of varicella-zoster virus reactivation in herpes simplex virus-seronegative patients with acute peripheral facial palsy. Clin Infect Dis 2000;30:529-533
CrossRef | Web of Science | Medline

To the Editor:

The methodologic superiority of the Scottish pivotal trial of early treatment of Bell's palsy, reported by Sullivan et al., as compared with the Japanese trial recently reported by Hato et al.,1 is obvious and clearly highlighted in Gilden and Tyler's editorial accompanying the article by Sullivan et al.2 However, one clinically relevant difference between the two trials that favors the Japanese study seems to have been disregarded. Hato et al. classified their patients according to the baseline severity of the facial palsy and presented their efficacy data for the different subgroups. Sullivan et al. mention adjustment of the analysis for baseline characteristics, including the score on the House–Brackmann scale, without clearly presenting results for this. It would be helpful for daily practice and would strengthen their findings if Sullivan and colleagues could confirm that their efficacy findings hold for subgroups of patients with lower and higher baseline House–Brackmann scores. If not, we agree with Gilden and Tyler that the early use of (val)acyclovir (in combination with glucocorticoids) could still be considered in patients with severe or complete facial palsy.

Esther S. Korf, M.D., Ph.D.
Joep Killestein, M.D., Ph.D.
VU Medical Center, 1081 HV Amsterdam, the Netherlands
j.killestein@vumc.nl

2 References

1
Hato N, Yamada H, Kohno H, et al. Valacyclovir and prednisolone treatment for Bell's palsy: a multicenter, randomized, placebo-controlled study. Otol Neurotol 2007;28:408-413
CrossRef | Web of Science | Medline

2
Gilden DH, Tyler KL. Bell's palsy -- is glucocorticoid treatment enough? N Engl J Med 2007;357:1653-1655
Full Text | Web of Science | Medline

Author/Editor Response

We agree with Beutner and Leiner that a definitive diagnosis of idiopathic facial paralysis within the 48-to-72-hour window for early intervention is difficult. Baseline virologic studies would potentially have been valuable in determining whether the subgroup of patients with zoster sine herpete had a recovery rate that was different from the rate for those who did not have zoster sine herpete. However, we estimate that 800 patients would have been needed to complete follow-up, in order to provide equivalent power for our study. The dose of acyclovir that we used (2 g daily) rather than the higher dose suggested by Leiner was chosen because that was the dose used in two of the three studies included in the Cochrane review that was available at that time (one study used 2.4 g daily).1

In response to Korf and Killestein's point, we present here the subgroup analysis of response according to initial disease severity. For 450 of 496 patients who completed the study and for whom we had relevant data, we characterized those with a grade of 2 to 4 at onset as having moderate Bell's palsy and those with a grade of 5 to 6 as having severe Bell's palsy. The recovery rates at 9 months for patients treated with prednisolone and with acyclovir are shown in Table 1Table 1Rate of Recovery According to the Assigned Treatment and the Severity of Bell's Palsy at Onset..

Overall, the recovery rate in the prednisolone group was 93.8%, as compared with 80.4% in the no-prednisolone group (odds ratio, 3.66; 95% confidence interval [CI], 1.92 to 7.28; P<0.001; interaction between prednisolone and baseline severity was nonsignificant, P=0.18). The recovery rate in the acyclovir group was 84.6%, as compared with 89.6% in the no-acyclovir group (odds ratio, 0.64; 95% CI, 0.36 to 1.12; P=0.12; interaction was nonsignificant, P=0.35). The combined recovery rate in the moderately affected group was 90.7%, as compared with 78.3% in the severely affected group (odds ratio, 2.69; 95% CI, 1.49 to 4.82; P<0.001). As with the point about the dose of acyclovir, we cannot say whether valacyclovir may be effective because it was not an intervention in our study. Absence of evidence is not evidence of absence.2

Frank Sullivan, Ph.D.
University of Dundee, Dundee DD1 4HN, United Kingdom
f.m.sullivan@chs.dundee.ac.uk

Iain Swan, M.D.
University of Glasgow, Glasgow G12 8QQ, United Kingdom

Fergus Daly, Ph.D.
University of Dundee, Dundee DD1 4HN, United Kingdom

2 References