Reliability of postoperative photographs in assessment of facial nerve function after vestibular schwannoma resection.

Abstract

This study was undertaken to assess the reliability of observations of postoperative photographs in assigning House-Brackmann scores as outcome measures for patients following resection of vestibular schwannomas. Forty pictures of differing facial expressions typically elicited from patients for assigning House-Brackmann scores were individually evaluated by neurosurgery residents and faculty members at the University of Alabama at Birmingham; a score was assigned to each picture by the individual raters. The interrater reliability was measured using the Spearman correlation coefficient, Kendall coefficient of concordance, and kappa statistic; internal consistency was calculated using the Cronbach alpha reliability estimate. The Spearman correlation coefficients showed strong positive association among raters, with a range of values of 0.66 to 0.90. Internal consistency measured by the Cronbach alpha coefficient was excellent (α = 0.97). The Kendall coefficient of concordance for the ordinal grades suggested a substantial degree of agreement among the raters (w = 0.76, p < 0.001). Static postoperative photographs are a reliable outcome measure for determining facial nerve function after vestibular schwannoma resection and may serve as a surrogate for the dynamic patient interview.