Agreement Between House-Brackmann and Sunnybrook Facial Nerve Grading Systems in Bell's Palsy in Children: Secondary Analysis of a Randomized, Placebo-Controlled Multicenter Trial

Background There is limited evidence on the use of facial nerve function grading scales in acute facial nerve paralysis in children. Objective To investigate the agreement between and the usability of the House-Brackmann and Sunnybrook scales in children with idiopathic facial paralysis (Bell's palsy) and to compare their ease of administration. Methods Data from a randomized controlled trial in children aged 6 months to <18 years with Bell's palsy was used. Children were recruited within 72 hours of symptom onset and assessed using the House-Brackmann and the Sunnybrook scales at baseline and at 1, 3, and 6 months until recovered. Agreement between the scales was assessed using the intraclass correlation coefficient (ICC) at each time point and using a Bland-Altman plot. Ease of administration was assessed using an 11-point Likert scale. Results Comparative data were available for 169 of the 187 children randomized. The ICC between the 2 scales across all time points was 0.92 (95% confidence interval [CI] 0.91-0.93), at baseline 0.37 (95% 0.25, 0.51), at 1 month 0.91 (95% CI 0.89-0.94), at 3 months 0.85 (95% CI 0.80-0.89), and at 6 months 0.96 (95% CI 0.95-0.97). The median score for the ease of administration for the House-Brackmann and Sunnybrook scales was 3 (interquartile range [IQR]: 1-5) and 7 (IQR: 4-8) respectively (P < .001, Wilcoxon signed-rank test). Conclusions There was excellent agreement between House-Brackmann and Sunnybrook scales, with poorer agreement at baseline. Clinicians found the House-Brackmann scale easier to administer. These findings suggest that both scales can be applied in children.