Respiratory disturbance index: an independent predictor of mortality in coronary artery disease

Abstract

Cardiovascular mortality was prospectively investigated in consecutive coronary artery disease (CAD) patients with versus without obstructive sleep apnea (OSA) during a follow-up period of 5 yr. An overnight sleep/ventilatory study was performed in patients requiring intensive care (n = 62, mean age 67.6 +/- 10.4 yr, range 44 to 86) during a stable condition (New York Heart Association [NYHA] functional class I-II) 4 to 21 mo after discharge from the hospital. OSA, defined as a respiratory disturbance index (RDI) of 10/h or more was found in 19 patients (mean RDI 17.5 +/- 8.3). Three OSA subjects who were successfully treated with continuous positive airway pressure (CPAP) during the observation period were excluded from the final analysis. There was no statistically significant difference (Fisher two-tailed exact test) between the OSA and non-OSA patient groups in terms of number of elderly subjects (age >/= 65 yr), gender, obesity (body mass index [BMI] >/= 30 kg/m(2)), smoking history, presence of hypertension, diabetes mellitus, hypercholesterolemia, or history of myocardial infarction at the study start. During the follow-up period, cardiovascular death occurred in six of 16 OSA patients (37.5%) compared with 4 (9.3%) in the non-OSA group (p = 0.018). The univariate predictors of cardiovascular mortality were RDI (p = 0.007), OSA (p = 0.014), age at baseline (p = 0.028), hypertension at baseline (p = 0.036), history of never-smoking (p = 0.031), and digoxin treatment during the follow-up period (p = 0.013). In a Cox multiple conditional regression model, RDI remained as an independent predictor of cardiovascular mortality (exp beta = 1.13, 95% confidence interval [CI] 1.05 to 1.21, two-sided p < 0.001). We conclude that untreated OSA is associated with an increased risk of cardiovascular mortality in patients with CAD. Furthermore, it appears appropriate that RDI is taken into consideration when evaluating secondary prevention models in CAD.