Baseline serum bicarbonate levels independently predict short-term mortality in critically ill patients with ischaemic cardiogenic shock

Abstract

Background: Cardiogenic shock is a feared complication of acute myocardial infarction with high mortality rates. Data on the predictive role of acid base dysregulation in this clinical setting are sparse. We therefore embarked on investigating the predictive role of serum bicarbonate in critically ill intensive care unit (ICU) patients with cardiogenic shock. Methods: A total of 165 ischaemic cardiogenic shock patients (118 men, aged 68.4 years (interquartile range 59.0–77.4), APACHE II score 26.0 (interquartile range 21.0–29.0), after percutaneous coronary intervention were included in a single-centre analysis. Percutaneous coronary intervention-related data such as left ventricular ejection fraction and laboratory indices were recorded and routine clinical follow-up was obtained at hospital discharge and at one year. All-cause mortality was assessed and data were analysed using univariate and multivariate models. Results: All-cause mortality was highest (17%) during the first 48 hours following ICU admission (28-day mortality rate 43%). In a multiple regression model, age (hazard ratio (HR) 1.035, 95% confidence interval (CI) 1.011–1.059, P=0.004), APACHE II score (HR 1.036, 95% CI 1.002–1.072, P=0.037) and baseline serum bicarbonate levels (HR 0.93, 95% CI 0.866–0.998, P=0.046) independently predicted 28-day mortality (overall model fit χ2 22.9, P<0.0001). The HR for patients in the lowest baseline serum bicarbonate tertile for 365-day mortality was HR 2.06 (95% CI 1.20–3.53). Conclusions: In a large cohort of consecutive cardiogenic shock patients hospitalised in the ICU, low serum bicarbonate levels at admission independently predicted mortality. Given the widespread availability of blood gas analysers in ICUs, we propose baseline serum bicarbonate levels as an additional biomarker for identification and stratification of cardiogenic shock patients at risk.