Alkalosis during emergency department evaluation of acute heart failure: Is there an association with mortality?

Abstract

Objective: To analyze the possible association between a finding of plasma alkalosis in patients diagnosed with acute heart failure (AHF) in the emergency department (ED) and in-hospital mortality.

Methods: Cases of AHF were identified in the registry for Epidemiology of Acute Heart Failure in Emergency Departments (EAHFE), which comprises entries for 24 248 episodes diagnosed in 53 Spanish EDs. Studied cases were those with registered plasma pH levels determined by analysis of acid-base equilibrium in the ED. Patients in the alkalosis group (AG) had a plasma pH greater than 7.45. Controls cases had a pH between 7.35 and 7.45. We gathered epidemiologic and comorbidity data, chronic medications, baseline status variables, and characteristics related to the decompensation episode and its severity. The outcome measure was in-hospital mortality from any cause. Logistic regression was used to analyze crude and adjusted associations between alkalosis and mortality, expressed as odds ratios (ORs) and 95% CIs.

Results: A total of 2522 AG cases and 8526 controls were studied. Betwee-group differences included age (AG patients were older), number of comorbidities and chronic medications (notably, more loop diuretics in AG patients), and multiple variables related to the AHF episode. In-hospital all-cause mortality was similar (AG, 7.5%; controls, 7.0%): crude and adjusted ORs, 1.062 (95% CI, 0.896-1.259) and 1.023 (95% CI, 0.854-1.225), respectively. In the AG, 676 patients had probable metabolic alkalosis (PCO2 > 40 mmHg) and 937 had probable respiratory alkalosis (PCO2 35 mmHg). Inhospital mortality was 9.3% in the probable metabolic alkalosis subgroup and 6.7% in the probable respiratory alkalosis subgroup: crude and adjusted ORs, 1.258 (95% CI, 0.942-1.681) and 0.919 (95% CI, 0.695-1.215), respectively.

Conclusions: This retrospective analysis of cases in the EAHFE registry found no association between alkalosis and higher in-hospital mortality after AHF. Nor were significant associations found when we analyzed mortality related to probable metabolic vs respiratory alkalosis.