Do critically ill patients with AKI benefit from furosemide? Further real-word evidence from a large multi-center database

Table 1 Association between furosemide use and clinical outcomes in critically ill patients with acute kidney injury

Outcomes	Non-diuretics	Furosemide	p value	HR (95% CI)
Pre-matched cohorts	48,495	11,256
Primary
In-hospital mortality, n (%)^a	5048 (10.4)	1327 (11.8)	0.18	0.96 (0.90–1.02)
In-hospital mortality, n (%)^b	5048 (10.4)	1327 (11.8)	0.98	1.00 (0.93–1.07)
Secondary
Recovery of renal function, n (%)^b	30,031 (61.9)	7105 (63.1)	< 0.001	1.10 (1.05–1.14)
Length of ICU stay, mean (SD)^c	4.1 (5.0)	4.7 (5.2)	< 0.001	1.53 (1.35–1.73)
Length of hospital stay, mean (SD)^c	10.3 (9.0)	11.1 (8.7)	< 0.001	1.46 (1.18–1.80)
Post-matched cohorts	11,256	11,256
Primary
In-hospital mortality, n (%)^a	1415 (12.6)	1327 (11.8)	0.03	0.92 (0.85–0.99)
In-hospital mortality, n (%)^b	1415 (12.6)	1327 (11.8)	0.49	0.97 (0.89–1.06)
Secondary
Recovery of renal function, n (%)^b	6791 (60.3)	7105 (63.1)	< 0.001	1.12 (1.06–1.19)
Length of ICU stay, mean (SD)^c	4.3 (4.8)	4.7 (5.2)	< 0.001	1.41 (1.20–1.65)
Length of hospital stay, mean (SD)^c	10.7 (8.9)	11.1 (8.7)	< 0.001	1.55 (1.19–2.00)

^aCox proportional hazard regression was used to adjust the confounding variables
^bLogistic regression was used to adjust the confounding variables
^cLinear regression was used to adjust the confounding variables
The confounding variables include age, gender, ethnicity, admission type, comorbidities, mechanical ventilation use, vasopressor use, fluid balance first day, serum creatinine, eGFR, and APCHE IV predicted hospital mortality. Sensitivity analysis was conducted by selecting a part of all variables by clinical experts, and the p value never changed from < 0.05 to > 0.05 or from > 0.05 to < 0.05, which presented robust
Abbreviations: HR hazard ratio, CI confidence interval, ICU intensive care unit, SD standard deviation

ISSN: 1364-8535