415 Risk of Obstructive Sleep Apnea after Treatment of Head and Neck Squamous Cell Carcinoma

Abstract

Abstract Introduction Head and neck cancers (HNC) or their treatment may be associated with an increased risk of obstructive sleep apnea (OSA). Small studies that examined OSA risk factors in adults with HNC reported conflicting results. This study examined associations between tumor characteristics and risk of OSA among patients at least one year free of head and neck squamous cell carcinoma (HNSCC). Methods For this cross-sectional study of HNSCC patients at a large academic medical center, inclusion criteria were age ≥18 years, with absence of tracheostomy or mental impairment. The STOP-BANG questionnaire, with a threshold ≥3, was used to identify high risk for OSA. Descriptive statistics were used to compare demographic and health characteristics between OSA risk groups. Logistic and linear regression models adjusted for age and gender were used to examine associations between demographics, anthropometric measures, and OSA risk. Results Among 67 participants, 57 (85%) were male, mean age was 62.0±8.0 (s.d.) years, mean body mass index (BMI) was 28.7±4.6 Kg/m2, and mean neck circumference (NC) was 16.3±1.2 inches. A total of 50 (75%) participants received chemoradiation only. High OSA risk was observed in 40 (60%) of the participants. Tumor location, tumor stage, and type of cancer treatment were not different between OSA risk groups. Body mass index and NC were greater in the high OSA risk group (BMI 29.6±4.5 Kg/m2 vs. 27.3±4.1 Kg/m2, p=0.03; NC 16.5±1.3 inches vs. 15.8±0.5 inches, p=0.01). In age and gender-adjusted logistic regression models, BMI (OR=1.2, 95% CI 1.0, 1.4) and NC (OR=2.9, 95% CI 1.1, 7.3) were associated with high OSA risk. Adjusted linear regression models showed that BMI (β=0.10, 95%CI 0.04, 0.17) and NC (β=0.64, 95%CI 0.32, 0.96) were associated with STOP-BANG scores. Conclusion High OSA risk was quite common after HNSCC treatment. However, measured HNSCC characteristics were not different between high and low OSA risk groups. Instead, OSA risk factors included BMI and NC, as often reported in non-HNSCC patients as well. Prospective studies before and after cancer treatment will be needed to further elucidate potential roles of HNSCC and its treatment in subsequent OSA incidence. Support (if any) Dr. Gavidia’s work was supported by an NIH/NINDS T32-NS007222 grant.