Relationship of Admission Serum Anion Gap and Prognosis of Critically Ill Patients: A Large Multicenter Cohort Study

Table 2

The association between serum AG and mortalities in all patients.

OR (95% CI)		OR (95% CI)		OR (95% CI)
ICU mortality
Model 1	2.281 (1.003-5.186)	0.049	1(ref)		2.539 (2.197-2.934)	<0.001
Model 2	2.270 (0.965-5.339)	0.060	1(ref)		1.851 (1.592-2.151)	<0.001
Model 3	2.308 (0.977-5.452)	0.056	1(ref)		1.834 (1.571-2.141)	<0.001
Model 4	2.493 (1.034-6.014)	0.042	1(ref)		1.533 (1.308-1.797)	<0.001
Model 5	2.504 (1.039-6.035)	0.041	1(ref)		1.530 (1.305-1.794)	<0.001
Hospital mortality
Model 1	1.641 (0.755-3.570)	0.211	1(ref)		2.531 (2.243-2.855)	<0.001
Model 2	1.593 (0.706-3.594)	0.262	1(ref)		1.890 (1.665-2.145)	<0.001
Model 3	1.587 (0.700-3.595)	0.269	1(ref)		1.866 (1.638-2.125)	<0.001
Model 4	1.691 (0.737-3.881)	0.215	1(ref)		1.611 (1.409-1.841)	<0.001
Model 5	1.700 (0.742-3.895)	0.210	1(ref)		1.618 (1.415-1.849)	<0.001

Model 1: unadjusted; model 2: adjusted for age, gender, ethnicity, and APACHE IV score; model 3: adjusted for age, gender, ethnicity, APACHE IV score, heart failure, respiratory failure, renal failure, liver diseases, coagulopathy, sepsis, trauma, shock, hypertension, and diabetes mellitus; model 4: adjusted for age, gender, ethnicity, APACHE IV score, heart failure, respiratory failure, renal failure, liver diseases, coagulopathy, sepsis, trauma, shock, hypertension, diabetes mellitus, and serum lactate; model 5: adjusted for age, gender, ethnicity, APACHE IV score, heart failure, respiratory failure, renal failure, liver diseases, coagulopathy, sepsis, trauma, shock, hypertension, and diabetes mellitus, serum lactate, renal replacement therapy, and mechanical ventilation. AG, anion gap; APACHE, acute physiology and chronic health evaluation; CI, confidence interval; ICU, intensive care unit; OR, odds ratio.