Association of serum sodium levels with all-cause and cardiovascular mortality in chronic kidney disease: Results from a prospective observational study

Aims: The prevalence of hyponatraemia in the outpatient setting has not been thoroughly explored, and little is known about the prognostic implication of dysnatraemia in chronic kidney disease (CKD) patients, in particular accommodating the effect of concurrent medications. Methods: This is a prospective observational study of non-dialysis-dependent CKD patients managed in a nephrology clinic. Patients enrolled between 2002 and 2012 in the Chronic Renal Insufficiency Standards Implementation Study were assessed. Survival analyses were performed using baseline sodium and 12-month time-averaged sodium, with adjustment for co-morbid diseases, laboratory findings and concurrent medications. Results: At baseline (n = 2093), mean estimated glomerular filtration rate was 32.8 +/- 15.9ml/min per 1.73 m(2), median age was 67 (interquartile range 56-75) years and median serum sodium concentration was 140 (138-142) mmol/l. After a followup of 41 (18-67) months, therewere 684 deaths, 174 from cardiovascular causes; 1925 time-averaged sodium values were analysed. In the Cox multivariate adjusted regression, baseline hyponatraemia, but not hypernatraemia, was independently associated with all-cause mortality (hazard ratio (HR) 1.35, 95% confidence interval (CI) 1.02-1.78, P = 0.04, and HR 1.15, 95% CI 0.84-1.57, P = 0.39, respectively). This was similarly the case for time-averaged hyponatraemia and hypernatraemia (HR 2.15, 95% CI 1.59-2.91, P<0.01, and HR 1.47, 95% CI 0.93-2.38, P = 0.10, respectively). However, the association of baseline and time-averaged hyponatraemia with cardiovascular mortality was not significant. Conclusion: Hyponatraemia in the ambulatory setting is associated with all-cause but not cardiovascular mortality in CKD, independent of concomitant medications and co-morbidities.