P5379
Predictive value of omega-3 polyunsaturated fatty acids (PUFAs) for cardiovascular- and all-cause mortality in chronic haemodialysis patients

Extract

Background: Decreased omega-3 (n-3) polyunsaturated fatty acids (PUFAs) levels are closely associated with incidence of cardiovascular disease (CVD). Although haemodialysis (HD) patients are widely recognized to be at highest risk of CVD, the association of the PUFAs with CVD prognosis remains unclear. We investigated the predictive value of n-3 PUFAs for CVD- and all-cause mortality in chronic HD patients.

Methods: A total of 360 outpatients stably undergoing maintenance HD therapy (male 68%, age 66±11years, diabetes 41%) were enrolled. They underwent measurement of eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA) and arachidonic acid (AA), and were followed-up for 3 years.

Results: During follow-up period (32 months), 84 patients died (23.3%) including 48 CVD cause (13.3%). On Cox multivariate analysis, DHA [hazard ratio (HR) 0.89, 95% confidence interval (CI) 0.81–0.97, p=0.013], EPA (HR 0.86, 95% CI 0.78–0.95, p=0.0014) and EPA/AA ratio (HR 0.77, 95% CI 0.65–0.89, p=0.0002) were identified as independent predictors of CVD mortality. Similar results were also obtained for all-cause mortality. However, the addition of DHA to a prediction model based on traditional risk factors had no effect on model discrimination as measured by C-index. On the other hand, only the addition of EPA/AA ratio could improve C-index for both CVD- and all-cause mortality, moreover, even if compared to the prediction model with DHA (Table). Kaplan-Meier survival rate at 3-year was lower in patients with EPA/AA<0.56 as a value of third quartile, compared to EPA/AA≥0.56 for CVD mortality (82.6% vs. 95.3%, p=0.0035), and all-cause mortality (72.9% vs. 87.7%, p=0.0042), respectively.