Age-related ahi cut-offs associated with cardiovascular and cerebrovascular disorders: clinical implications

Abstract Introduction Several studies have shown that the association of obstructive sleep apnea (OSA) with clinical outcomes, such as hypertension, weakens in older adults. An apnea hypopnea index (AHI) ≥5 is recommended as the cut-off for initiating treatment without any consideration of possible age differences. We aimed to examine at which cut-off point of AHI, OSA is associated with cardiovascular and/or cerebrovascular diseases (CBVD) in middle-aged adults and in older adults. Methods We studied 1,741 adults from the Penn State Adult Cohort (age 20-88 years, 52.3% female, 12.4% racial/ethnic minority), who underwent a 1-night sleep laboratory evaluation, clinical history and physical examination. Hypertension was defined as a diastolic blood pressure ≥90 mmHg or a systolic blood pressure ≥140 mmHg or the use of anti-hypertensive medication. CBVD was defined based on a self-report of a physician diagnosis of heart disease and/or stroke. Logistic regression models examined the odds of having hypertension or CBVD in a step-wise manner starting from an AHI ≥5 up to AHI ≥30 by increments of 5 events, while simultaneously adjusting for sex, race and BMI and stratifying by age. Results The odds of hypertension were significantly increased in adults aged <60y (43.2±9.1y) with an AHI≥5 (OR=1.56, 95% CI 1.01-2.41, p=0.045), while the odds of hypertension were not significantly increased in adults aged ≥60y (68.4±6.3y) for any given AHI cut-off. The odds of CBVD were significantly increased in adults aged ≥60y with an AHI≥15 (OR=2.50, 95% 1.24-5.04, p=0.011), while the odds of CBVD were not significantly increased in adults aged <60y for any given AHI cut-off. Conclusion These data suggest that AHI cut-offs warranting treatment of OSA should be adjusted based on age and prevalent clinical disorders, consistent with the concept of personalized medicine. These findings are also consistent with the notion that OSA in older adults is a distinctly different phenotype than in young and middle-aged adults. Support (If Any)