Predictors of mortality in patients with stress-induced cardiomyopathy developed during critical care

Abstract

PURPOSE: The aims of this study were to define predictors of in-hospital mortality and to explore the implication of Acute Physiology and Chronic Health Evaluation (APACHE) II score in patients with stress-induced cardiomyopathy (SCM) developed during critical care. MATERIALS AND METHODS: All patients admitted to intensive care unit and underwent transthoracic echocardiography (TTE) were consecutively enrolled from January 2008 to May 2011. Clinical, demographic and laboratory data, APACHE II score, and transthoracic echocardiography finding were analyzed using a logistic regression model to investigate predictors of in-hospital mortality. RESULTS: A total of 71 patients (60 ± 18 years, 37% male) were included in the final analysis. In univariate and multivariate logistic regression analyses, underlying malignancies, male sex, age less than 65 years, and APACHE II score higher than 15 remained independent risk factors for in-hospital mortality of SCM. The area under the receiver operating characteristic curve for APACHE II was 0.745 (95% confidence interval, 0.630-0.861; P = .001), and an APACHE II score of 15 (sensitivity 73%, specificity 68%) was the optimal cutoff value in predicting in-hospital mortality of SCM during critical care. CONCLUSION: The in-hospital mortality in patients with SCM that developed during critical care was associated with underlying malignancy, male sex, old age, and APACHE II score.