Incident sleep disordered breathing in old age

Abstract

Background: Little is known about progression of and risk factors for sleep disordered breathing (SDB) in old age. We prospectively examined elderly volunteers to understand how changes in body weight are related to SDB for a period of 20-30 years.

Methods: Participants were 30 surviving members of a community-based cohort (mean entry age = 57.8) studied over a median follow-up of 23.4 years. SDB was quantified as the apnea-hypopnea index (AHI) via in-lab polysomnography from 215 nights, representing 733.3 person-years of follow-up. Weights were recorded in kilograms. We used linear regression to derive individual trajectories of AHI and weight regressed on time.

Results: Individuals had relatively low AHI (X = 2.3 [SD = 3.5]) and body mass index (kg/m(2); X = 24.6 [SD = 4.6]) at entry. Rates of change in AHI were characterized by positive slopes and linear increases by least squares regression. Mean rate of change was +0.43 events per hour per year, a 3.3% yearly increase relative to the maximum AHI observed for each case. Within individuals, curve fitting indicated statistically significant AHI increases associated not only with increases, but also decreases, in weight.

Conclusions: Rates of increase in AHI were larger than for aging reported for other organ systems (eg, autonomic, musculoskeletal, and respiratory), possibly reflecting complex mechanistic determination of SDB in old age. Association between decreased weight and increased SDB with advancing years represents an important "proof of concept," perhaps compatible with failure to maintain airway patency during sleep as a component of generalized muscle weakness in old age.

Figure 1.

Individual participants’ graphs (first 15 cases) showing time-dependent changes in log apnea–hypopnea index (AHI; blue) and body weight (in kg; red) over time. Both log AHI and weight are plotted as standardized scores on the y-axis. The x-axis represents age in months. Straight lines represent least squares regression fits.

Figure 2.

Individual participants’ graphs (second 15 cases) showing time-dependent changes in log apnea–hypopnea index (AHI; blue) and body weight (in kg; red) over time. Both AHI and weight are plotted as standardized scores on the y-axis. The x-axis represents age in months. Straight lines represent least squares regression fits.