Abstract
OBJECTIVE: Sodium intake is associated with cardiovascular outcomes. However, no study has specifically reported an association between cardiovascular mortality and urinary sodium concentration (U(Na)). We examined the association of U(Na) with mortality in a cohort of type 2 diabetes (T2D) patients. METHODS: Patients were followed for all-cause death and cardiovascular death. Baseline U(Na) was measured from second morning spot urinary sample. We used Cox proportional hazard models to identify independent predictors of mortality. Improvement in prediction of mortality by the addition of U(Na) to a model including known risk factors was assessed by the relative integrated discrimination improvement (rIDI) index. RESULTS: Participants (n = 1,439) were followed for a median of 5.7 years, during which 254 cardiovascular deaths and 429 all-cause deaths were recorded. U(Na) independently predicted all-cause and cardiovascular mortality. An increase of one standard deviation of U(Na) was associated with a decrease of 21% of all-cause mortality and 22% of cardiovascular mortality. U(Na) improved all-cause and cardiovascular mortality prediction beyond identified risk factors (rIDI = 2.8%, P = 0.04 and rIDI = 4.6%, P = 0.02, resp.). CONCLUSIONS: In T2D, U(Na) was an independent predictor of mortality (low concentration is associated with increased risk) and improved modestly its prediction in addition to traditional risk factors.