Haptoglobin phenotype, sleep-disordered breathing, and the prevalence of cardiovascular disease: the Sleep Heart Health Study.

Background: Diabetes is an independent risk factor for cardiovascular disease, and there is growing evidence that sleep-disordered breathing also may increase the risk of cardiovascular disease. The mechanism responsible for increased susceptibility of people with diabetes to cardiovascular disease is thought to share several features with sleep-disordered breathing, notably increased oxidative stress. We recently demonstrated that a particular haptoglobin phenotype that is associated with differential antioxidant activity is an independent risk factor for cardiovascular disease in individuals with diabetes. We therefore sought to determine whether sleep-disordered breathing and cardiovascular disease are more strongly associated among people with the unfavorable haptoglobin phenotype.

Methods: We tested this hypothesis in 2612 middle-aged and older participants from the Sleep Heart Health Study. Haptoglobin phenotyping was performed by gel electrophoresis. Respiratory disturbance index was assessed by standard methods. Logistic regression analysis was performed to estimate the association between haptoglobin phenotype and cardiovascular disease, adjusting for known cardiovascular risk factors (age, sex, diabetes, smoking, lipid levels, and hypertension). Possible modification by haptoglobin phenotype of the association of sleep-disordered breathing with cardiovascular disease prevalence was explored by examining interaction terms.

Results: We found no significant association between haptoglobin phenotype and prevalent cardiovascular disease in this cohort, nor were significant interactions found between haptoglobin phenotype and sleep-disordered breathing on the prevalence of cardiovascular disease.

Conclusions: Sleep-disordered breathing did not appear to interact with haptoglobin phenotype in modifying the association with prevalent cardiovascular disease in the Sleep Heart Health Study. These findings could be due to the absence of association or to survivor bias in these cross-sectional analyses.