Impact of pre-ICU balanced crystalloids versus saline on mortality in sepsis: a propensity score-matched multicenter cohort study.

Abstract

The optimal crystalloids for initial fluid resuscitation in sepsis remain uncertain. Most previous studies have focused on intensive care unit (ICU), leaving the pre-ICU admission impact of crystalloid type unclear. Is there an association between pre-ICU admission crystalloid type administration and clinical outcomes in patients with sepsis? This nationwide multicenter retrospective cohort study analyzed adult patients with sepsis between September 2019 and December 2023. Patients receiving initial fluid resuscitation before ICU admission were categorized into balanced crystalloid or 0.9% sodium chloride (saline) groups. The primary outcome was 30-day mortality. Among 3,688 patients (mean [standard deviation] age 71.9 [13.8] years; 58% male), 1,381 (37.4%) received balanced crystalloids and 2,307 (62.6%) received saline before ICU admission. After propensity score matching, 1,335 patients remained in each group. Balanced crystalloid use was associated with lower 30-day mortality, resulting in an adjusted odds ratio (aOR) of 0.69 (95% confidence interval [CI] 0.56-0.83). The finding was consistent after inverse probability of treatment weighting (aOR, 0.78; 95% CI 0.69-0.88). The association between balanced crystalloids and mortality was consistent across age, sex, disease severity, liberal fluid management (pre-ICU fluid volume >30 mL/kg), and pre-ICU vasopressor use. Notably, mortality risk was lower with balanced crystalloids in patients without chronic kidney disease, received fluids in the emergency department or ward (vs. operating room or other ICU), with baseline chloride ≤110 mmol/L, and baseline pH <7.35. Further, balanced crystalloid use was associated with reduced need for continuous renal replacement therapy during ICU stay, reduced renal replacement therapy at ICU discharge, and increased 28-day ventilator-free days. Pre-ICU administered balanced crystalloid use was associated with lower 30-day mortality in patients with sepsis than saline.