Heart failure prognostic model

Heart failure (HF) is a common, costly, disabling and deadly syndrome. Heart failure is a progressive disease characterized by high prevalence in society, significantly reducing physical and mental health, frequent hospitalization and high mortality (50% of the patients survive up to 4 years after the diagnosis, the annual mortality varying from 5% to 75%). The purpose of this study is to develop a prognostic model with easily obtainable variables for patients with heart failure. Methods and Results. Our lot included 101 non–consecutive hospitalized patients with heart failure diagnosis. It included 49,5% women having the average age of 71.23 years (starting from 40 up to 91 years old) and the roughly estimated period for monitoring was 35.1 months (5–65 months). Survival data were available for all patients and the median survival duration was of 44.0 months. A large number of variables (demographic, etiologic, co morbidity, clinical, echocardiograph, ECG, laboratory and medication) were evaluated. We performed a complex statistical analysis, studying: survival curve, cumulative hazard, hazard function, lifetime distribution and density function, meaning residual life time, Ln S (t) vs. t and Ln(H) t vs. Ln (t). The Cox multiple regression model was used in order to determine the major factors that allow the forecasting survival and their regression coefficients: age (0.0369), systolic blood pressure (–0.0219), potassium (0.0570), sex (–0.3124) and the acute myocardial infarction (0.2662). Discussion. Our model easily incorporates obtainable variables that may be available in any hospital, accurately predicting survival of the heart failure patients and enables risk stratification in a few hours after the patients' presentation. Our model is derived from a sample of patients hospitalized in an emergency department of cardiology, some with major life–altering co morbidities. The benefit of being aware of the prognosis of these patients with high risk is extremely beneficial. The use of this model may ease the estimation of the vital prognosis, to improve the compliance and increase in the use of life–saving medical or surgical therapy (pacemakers, implantable defibrillators or transplantation).