Abstract
BACKGROUND: Recent data suggest that hyperchloremia induced by fluid resuscitation is associated with acute kidney injury (AKI) and mortality, particularly in sepsis. Experimental studies showed that hyperchloremia could affect organ functions. In patients with septic shock, we examined the relationship between serum chloride concentration and both renal function and survival. METHODS: Post hoc analysis of the "HYPER2S" trial database (NCT01722422) including 434 patients with septic shock randomly assigned for resuscitation with 0.9% or 3% saline. Metabolic parameters were recorded up to 72 h. Metabolic effects of hyperchloremia (> 110 mmol/L) were studied stratified for hyperlactatemia (> 2 mmol/L). Cox models were constructed to assess the association between chloride parameters, day-28 mortality and AKI. RESULTS: 413 patients were analysed. The presence of hyperlactatemia was significantly more frequent than hyperchloremia (62% versus 71% of patients, respectively, p = 0.006). Metabolic acidosis was significantly more frequent in patients with hyperchloremia, no matter the presence of hyperlactatemia, p < 0.001. Adjusted risk of AKI and mortality were not significantly associated with serum chloride, hyperchloremia, maximal chloremia and delta chloremia (maximal-H0 [Cl]). CONCLUSIONS: Despite more frequent metabolic acidosis, hyperchloremia was not associated with an increased risk for AKI or mortality. Trial registration ClinicalTrials.gov, identifier: NCT01722422, registered 2 November 2012.