0569 Soft Palate Fat Between Lean Adults with Obstructive Sleep Apnea and Healthy Control

Abstract

Abstract Introduction Previous studies have shown that obese patients with obstructive sleep apnea (OSA) have a significantly greater percentage of fat tissue in soft palate than normal subjects. However, the influence of soft palate fat is not clear in non-obese adults with OSA. This study compared the volume of fat in the soft palate between lean adults with OSA and lean controls. Methods We examined soft palate fat in 21 lean OSA cases and 16 lean controls with body mass index (BMI) <25 kg/m2. All subjects underwent a magnetic resonance imaging (MRI) with three-point Dixon scan. We used volumetric reconstruction algorithms to quantify the amount of soft palate fat, which was compared between apnecis and controls. Analysis reproducibility was quantified using intraclass correlation coefficients (ICC) from repeated analyses of 20 randomly-chosen MRIs. Results Analysis of soft palate fat was highly reproducible, with an ICC (95% confidence interval) of 0.968 (0.923, 0.987). Lean apneics were younger than lean controls (45.3±13.0 vs. 62.1±10.4 years; p<0.0001). No significant differences between apneics and controls were observed in the average BMI (23.4±2.2 vs. 23.5 ± 2.6 kg/m2; p=0.824), the fat pads volume (4198±1728 vs. 3880±1544 mm3; p=0.646), and the proportion of males (61.9% vs. 68.8%; p=0.666). In unadjusted analyses, the lean OSA group showed significantly higher soft palate fat volume than lean controls (7605±2109 vs. 5327±1783 mm3; p=0.003). When adjusting for age, gender and BMI, no differences was observed between groups in soft palate fat volume (p=0.122) and fat pads volume (p=0.702). Conclusion Analysis of soft palate fat volume from Dixon MRI is highly reproducible. Our results indicate no significant difference in deposition of fat at soft palate between lean patients with OSA and lean controls when accounting for age, gender and BMI. Support This study is supported by National Institutes of Health Grant: 2P01HL094307-06A1. LX is supported by Young Elite Scientists Sponsorship Program of China Association for Science and Technology.