Bell's Palsy

To the Editor: In his Clinical Practice article on Bell's palsy, Gilden (Sept. 23 issue) 1 recommends treatment of the patient in the vignette with oral prednisone. The Cochrane Collaboration 2 has pointed out that the statement of the American Academy of Neurology — that early treatment with corticosteroids is “probably effective” 3 — is most likely invalid because their evidence synthesis (i.e., their summary of the best available evidence) included a trial with a high rate of loss to follow-up 4 and a nonrandomized trial 5 — trials that both had serious threats to validity. Without these studies, the pooled results are no longer . . .