Associations of total and high-molecular-weight adiponectin with all-cause and cardiovascular mortality in older persons: the Cardiovascular Health Study.

BACKGROUND

Adiponectin shows opposite associations with adverse outcomes in healthy middle-aged populations (lower risk) and cohorts with prevalent cardiovascular disease, heart failure, or advanced age (higher risk).

METHODS AND RESULTS

In a population-based study of older adults, we examined the relationships of total and high-molecular-weight adiponectin with mortality among subgroups defined by baseline cardiovascular status: No cardiovascular disease, heart failure, or atrial fibrillation (group 1); cardiovascular disease but no heart failure/atrial fibrillation (group 2); and heart failure/atrial fibrillation (group 3). We found significant differences in the associations with all-cause mortality across the groups. The association in group 1 was U-shaped; increasing levels of total adiponectin up to 12.4 mg/L were associated with lower mortality after adjustment for confounders (hazard ratio=0.81 per 1 SD [95% confidence interval, 0.65-0.95]), but above this cut point, higher levels conferred greater risk (hazard ratio=1.19 [95% confidence interval, 1.12-1.27]). Further adjustment for diabetes mellitus or insulin resistance, protection against which has been proposed to mediate the beneficial relationships of adiponectin with outcome, attenuated the association in the lower range. There was no significant association in group 2, but in group 3, total adiponectin showed a direct adjusted association. Additional adjustment for putative metabolic/inflammatory intermediates suggested a direct association for group 2, and magnified the one for group 3 (hazard ratio=1.31 [1.15-1.50]). Results were similar for high-molecular-weight adiponectin and for cardiovascular mortality.

CONCLUSIONS

Adiponectin exhibits distinct associations with mortality in elders, which shift from U-shaped to flat to direct with greater baseline cardiovascular dysfunction but become more consistently adverse after accounting for metabolic/inflammatory factors presumed to be favorably regulated by the adipokine. These findings advance understanding of the adiponectin paradox as it relates to older adults.