Long-chain n-3 fatty acids and mortality in elderly patients

Background: Long-chain n-3 fatty acids may favorably modulate many diseases. The evidence is firm for coronary heart disease, less certain for stroke, and only speculative for other diseases. The impact of these fatty acids on mortality among acutely sick elderly patients is unknown. Objective: The objective was to investigate the relation between long-chain n-3 fatty acids and overall mortality in acutely sick elderly patients. Design: Frail, elderly patients (n = 254) acutely admitted to St Olavs Hospital in central Norway were examined. The plasma phospholipid concentration of eicosapentaenoic acid (EPA) was used as a surrogate marker for dietary intake of marine fatty acids. Mortality rates were evaluated after 3 y of follow-up. Cox proportional hazard analysis was used to calculate hazard ratios adjusted for important biochemical and clinical covariates. Results: The hazard ratio of overall mortality was significantly higher in patients with EPA concentrations in the lowest quartile than in patients in the upper 3 quartiles (adjusted hazard ratio: 0.52; 95% CI: 0.35, 0.77). The upper 3 quartiles were not significantly different from one another (P = 0.94). Conclusions: Overall mortality in frail, elderly, acutely sick patients was inversely and nonlinearly associated with EPA concentrations. Approximately 25% of the population had EPA concentrations below the indicated threshold for maximal protection, suggesting that only this part of the population might have benefited from additional EPA intake.