Association between serum chloride and mortality in patients on maintenance hemodialysis: the Osaka dialysis complication study (ODCS).

BACKGROUND

Recent studies have shown that a lower level of serum chloride is a predictor of all-cause mortality (ACM) in cohorts of hypertension, heart failure, and chronic kidney disease, although it is unknown whether such an association is present in those on maintenance hemodialysis (MHD).

METHODS

This was a post hoc analysis of our prospective cohort study in 1653 patients on MHD. The exposure of interest was serum chloride, and the key outcome was ACM during 5-year follow-up.

RESULTS

The median (interquartile range) age, dialysis duration, and chloride were 68 (60, 75) years, 64 (29, 133) months, and 102 (99, 105) mEq/L, respectively. We identified 467 cases with ACM. Cox proportional hazards model showed that a lower chloride level was an independent predictor of ACM with adjusted hazard ratio [95% confidence interval] of 0.82 [0.75, 0.90] per 1 standard deviation of chloride. The inverse association was consistent across various subgroups and remained significant after additional adjustment for serum sodium, potassium, use of renin-angiotensin system inhibitors, and use of loop diuretics. Although deaths from cardiovascular disease (CVD) and from infection were also predicted by chloride, chloride was not independently associated with occurrence of CVD or hospitalization for infection. Instead, chloride was independently associated with deaths following CVD events and hospitalization for infection.

CONCLUSIONS

An inverse association between chloride and ACM was also present in MHD patients. Serum chloride was more closely related to survival after CVD and infection events rather than the occurrence of these events in this population.