LAUSR

ABSTRACT Background: It is unknown whether early renal replacement therapy (RRT) initiation strategy in intensive care unit (ICU) patients with both acute respiratory distress syndrome (ARDS) and sepsis with or without renal failure is clinically beneficial. Patients and methods: A total of 818 patients with both ARDS and sepsis admitted to the ICU of Tianjin Medical University General Hospital were included in the analysis. Early RRT was defined as initiating the RRT strategy within 24 h of admission. The relationship between early RRT and clinical outcomes, including primary (30-day mortality) and secondary (90-day mortality, serum creatinine, Pao2/Fio2, duration of invasive mechanical ventilation, cumulative fluid output, and cumulative fluid balance) outcomes, was compared using propensity score matching (PSM). Results: A total of 277 patients (33.9% of the total population) underwent an early RRT initiation strategy before PSM. After PSM, a cohort of 147 patients with early RRT and 147 patients without early RRT with matched baseline characteristics (including serum creatinine at admission) were constructed. Early RRT was not significantly associated with 30- (hazard ratio [HR], 1.25; 95% confidence interval [CI], 0.85-1.85; P = 0.258) or 90-day mortality (HR, 1.30; 95% CI, 0.91–1.87, P = 0.150). At each time point within 72 h after admission, there was no significant difference in serum creatinine, Pao2/Fio2 and duration of mechanical ventilation between the early and the no early RRT groups. Early RRT significantly increased total output at all time points within 72 h of admission and reached a statistically significant negative fluid balance at 48 h. Conclusions: Early RRT initiation strategies had no statistically significant survival benefit in ICU patients with both ARDS and sepsis, with or without renal failure, nor did they significantly improve serum creatinine and oxygenation or shorten the duration of mechanical ventilation. The use and timing of RRT in such patients should be thoroughly investigated.