Abstract
BACKGROUND: Hyperchloremia is associated with increased mortality in critically ill patients. The objective of this study was to investigate the association between increased chloride levels and mortality outcomes in intracerebral hemorrhage (ICH) patients admitted to the intensive care unit (ICU). METHODS: We performed a retrospective study of all patients diagnosed with ICH and included in the Medical Information Mart for Intensive Care (MIMIC-Ⅲ) from 2001 to 2012. Inclusion criteria were the first diagnosis of ICH, ICU length of stay (LOS) over 72 h, and not receiving hypertonic saline treatment. Serum chloride perturbation within 72 h of admission was evaluated as a predictor of outcomes. The increase in chloride from baseline was dichotomized based on an increase in chloride in 72 h (≤5 mmol/L or >5 mmol/L). The primary outcome was 90-day mortality. RESULTS: A total of 376 patients (54.5% male, median age 70 years, interquartile range:58-79 years) were included. The overall 90-day mortality was 32.2% (n=121), in-hospital mortality was 25.8% (n=97), and Day 2 acute kidney injury (AKI) occurred in 29.0% (n=109) of patients. The prevalence of hyperchloremia on admission, during the first 72 h, and an increase in chloride (>5 mmol/L) were 8.8%, 39.4%, and 42.8%, respectively. After adjusting for confounders, the hazard ratio of increase in chloride (>5 mmol/L) was 1.66 (95% confidence interval:1.05-2.64, P=0.031). An increase in chloride (>5 mmol/L) was associated with a higher odds ratio for 90-day mortality in both the AKI and non-AKI groups. CONCLUSIONS: An increase in chloride from baseline is common in adult patients with ICH admitted to ICU. The increase is significantly associated with elevated mortality. These results support the significance of diligently monitoring chloride levels in these patients.