Clinical and demographic predictors of buccal fat pad volume in thyroid eye disease

Abstract

ABSTRACT Purpose To understand how thyroid eye disease (TED) impacts buccal fat pad (BFP) volume. Methods In this cohort study, computed tomography (CT) scans and charts of adult patients with and without TED were obtained from an institutional database. The primary outcome was BFP volume in cubic centimeters. Three independent, blinded observers analyzed scans using Horos, a free, open-source medical image viewing software. Bivariate and multivariable analyses were performed. We determined sample size using an effect size based on published reports of the minimum amount of fat excision needed to notice a clinical difference. Equivalence testing against upper and lower bounds set by the same effect size was conducted to assess practical significance of the results. Results Our sample was sufficient to detect a difference as large as 1.5cc with 95% power. 72 scans were included in our study, 24 TED patients and 48 controls. Mean BFP volume was not statistically different between TED patients and controls (3.96 cc vs 4.06 cc, p = .778). Analysis of covariance adjusting for relevant patient factors (age, sex, and BMI) also failed to find a significant difference between groups. Equivalence testing was significant (p < .001) and revealed the observed difference between groups was less than any clinically meaningful difference. For an effect size of 1.5cc, the data suggests there is a 5% risk of a false negative. Conclusions TED was not associated with a significant difference in BFP volume, suggesting that the BFP is spared from TED-related soft-tissue expansion.