Serum Anion Gap is Associated with Risk of All-Cause Mortality in Critically Ill Patients with Acute Myocardial Infarction

PURPOSE: Anion gap (AG) is a valuable and easily obtained clinical tool for differentially diagnosis of acid-base disorders. Current understanding of the prognostic impact of AG on mortality after acute myocardial infarction (AMI) is limited. We aimed to investigate whether AG is a predictor of short-term and long-term all-cause mortality after AMI. PATIENTS AND METHODS: We examined 1806 patients diagnosed with AMI in intensive care unit from the Medical Information Mart for Intensive Care III (MIMIC-III) database. We analyzed the association of AG with 30-day, 180-day and 1-year all-cause mortality on a continuous scale and in categories, using multivariable Cox regression. We utilized restricted cubic splines to evaluate the linearity between hazard ratio (HR) and AG concentrations. RESULTS: AG was associated with a higher risk of 30-day, 180-day and 1-year all-cause mortality, with adjusted HRs of 1.083 (95% CI 1.051 to 1.117), 1.077 (95% CI 1.049 to 1.105), and 1.074 (95% CI 1.047 to 1.101), respectively. The results were consistent in subgroup analyses. The association between AG and all-cause mortality was linear for 180-day and 1-year mortality, and near linear for 30-day mortality, as higher concentrations were associated with high all-cause mortality. When stratified according to quartiles, AG was associated with 30-day mortality (HR[95% CI]: second quartile, 2.243[1.273, 3.955]; third quartile, 3.026[1.763, 5.194]; top quartile, 4.402[2.573, 7.531]), 180-day mortality (HR[95% CI]: second quartile, 1.719[1.118, 2.645]; third quartile, 2.362[1.575, 3.542]; top quartile, 3.116[2.077, 4.676]), and 1-year mortality (HR[95% CI]: second quartile, 1.700[1.143, 2.528]; third quartile, 2.239[1.536, 3.264]; top quartile, 2.876[1.969, 4.201]) using bottom quartile as reference. CONCLUSION: We firstly demonstrated that higher AG was significantly associated with increased 30-day, 180-day and 1-year all-cause mortality in AMI patients. AG as an easily obtained marker is of strong and reliable predictive value for AMI mortality during follow-up.