Cardiogenic shock in general intensive care unit: a nationwide prospective analysis of epidemiology and outcome

Abstract

Abstract Funding Acknowledgements None. Background Cardiogenic shock (CS) is a heterogenous, life-threatening disease burdened by a mortality up to 50%. An increasing incidence of the CS related to other etiological causes, including acute decompensated heart failure, has been reported. However, most of the epidemiological data are derived from patients admitted to dedicated acute cardiac care units. Epidemiology and management of CS patients admitted to general intensive care unit (ICU) in Italy is yet unexplored. Purpose To describe the epidemiology, patients feature, management and outcome of CS patients admitted to Italian general ICUs. Methods Prospective multicentric epidemiological study. Over 600 thousand patients admitted to 316 hundred Italian general intensive care units from 2011 to 2018 were screened trough PROSAFE software. The Margherita PROSAFE project is an observational project for the continuous electronic collection of data on patients admitted to ICUs; it is developed by the GiViTI (Gruppo italiano per la valutazione degli interventi in terapia intensiva - Italian Group for the Evaluation of Interventions in Intensive Care Medicine). The primary outcome of the analysis was hospital mortality. Secondary outcomes included description of patients features and management. Results 11661 patients (72.1±12.1 years old.; 39.5% women) with a diagnosis of CS were included in the analysis. ICU and in-hospital mortality were respectively 45.5 % and 53.9%. Highest mortality was observed for pulmonary embolism (58.4%), right heart failure (56.3%) and acute decompensated heart failure (ADHF) (44.2%). 46.9% of patients presented with more than 3 organ failure (panel A). CS commonest aetiologies were ADHF (63,3%), acute myocardial infarction (19.8%) and right heart failure (11%). Invasive mechanical ventilation was used in the 88.6% of cases. Intra-aortic balloon pump was placed in 15.2%, with higher frequency in ischemic aetiologies (43.9%) and ADHF (13.7%); VA-ECMO was most frequently placed in the right heart failure (3.7%) and acute myocardial infarction (2.6%) groups. The factors independently associated with increased hospital mortality were increasing age (p <0.001), lower Glasgow coma scale (p <0.001), hypoxemia (PaO2 <100, OR: 1.91; 95% Cl: 1.58 – 2.3; p <0.001), bradycardia (OR: 1.84; 95% Cl: 1.49 – 2.26; p <0.001), hypotension (OR: 1.27; 95% Cl: 1.12 – 1.43; p <0.001), bilirubin above 12 mg/dl (OR: 7.90; 95% Cl: 2.56 – 24.3), increased creatinine (p <0.001) and reduced platelets (<50000 U/mL, OR: 2.43; 95% Cl: 1.78 – 3.31; p <0.001) (panel B). Curiously, in the predictive model a BMI above normal values was protective as well as tachycardia compared to severe bradycardia (panel C). Conclusions The etiology of CS in general intensive care units is heterogeneous with acute heart failure being the commonest. The mortality rate remains unacceptably high with right heart failure and acute heart failure being burden by the worst outcome.