Peak serum chloride and hyperchloremia in patients undergoing cardiac surgery is not explained by chloride-rich IV fluid alone: A post-hoc analysis of the LICRA trial

Abstract Objectives With the exception of 0.9% saline little is known about factors that may contribute to increased serum chloride concentration (SCl−) in patients undergoing cardiac surgery. We sought to characterize the association between administered chloride load from intravenous fluid and other perioperative variables with peak perioperative SCl−. Design Secondary analysis of data from a previously published controlled clinical trial in which patients were assigned to a chloride-rich or chloride-limited perioperative fluid strategy (NCT02020538) Setting Academic medical center Participants 1056 adult patients with normal preoperative SCl− undergoing cardiac surgery Interventions Nil Measurements & Main Results Peak perioperative SCl− and hyperchloremia, defined as peak SCl− >110 mmol/l, were selected as co-primary endpoints. Regression modeling identified factors independently associated with these endpoints. Mean (sd) peak perioperative SCl− was 114 (5) mmol/l; hyperchloremia occurred in 824 (78.0%) of the cohort. In addition to administered volume of 0.9% saline multivariable linear and logistic regression modeling consistently associated preoperative SCl− (regression coefficient 0.5; 95% CI 0.4-0.6 mmol/l and OR 1.60; 95% CI 1.41-1.82 per 1 mmol/l increase) and cardiopulmonary bypass duration (regression coefficient 0.1; 95% CI 0.1-0.2 mmol/l and OR 1.12; 95% CI 1.06-1.19 per 10 minutes) with both co-primary outcomes. Multivariable modeling only explained approximately 50% of variability in peak SCl−. Conclusions Our data identified an association for both 0.9% saline administration and other non-fluid variables with peak perioperative SCl− and hyperchloremia. Stand-alone strategies to limit administration of chloride-rich IV fluid may have limited ability to prevent hyperchloremia in this setting.