Background: The annual incidence of Bell's palsy (BP) is 15 to 20 per 100,000 with 40,000 new cases each year, and the lifetime risk is 1 in 60. For decades, clinicians have searched the prognostic tests of sufficient accuracy for acute facial paralysis. Objective: The present study was designed to verify in BP which clinical or electrophysiological parameters could be considered as predictive of the degree of recovery of normal facial muscle function. Methods: Sixty-three patients with BP were initially assessed according to the House and Brackmann facial function scoring system "HB system". All patients were followed for 3 months, the functional recovery then reassessed according to HB system. Nerve conduction studies were measured on the affected side via a bipolar surface stimulator placed over the stylomastoid foramen. Results: We could not find statistically significant differences between BP with good and poor prognosis as regard age, sex, onset, diabetes, hypertension, dyslipidemia, or the initial HB Score. Compound motor action potential amplitude (CMAP) detected during the initial electroneurography (ENoG) was statistically significant between BP with good and poor prognosis. Conclusions: The initial ENoG is more predictive of recovery of Bell's palsy than the initial clinical grading using the HB system. Age, sex, hypertension, diabetes, and dyslipidemia do not seem to correlate with the degree of recovery in Bell's palsy.