Heart failure (HF) is a major public health problem with increasing prevalence worldwide [1]. Once established, worsening heart failure is frequent and associated with significantly diminished quality of life, recurrent hospital admissions and direct impacts in healthcare costs [2]. The estimated prevalence of HF is 1 to 2% of the adult population in developed countries [3]. In Brazil, the HF prevalence is 2 million patients and its incidence is 240,000 new cases per year [4].

Despite the fact that most data on outcomes in patients with HF come from North America and Europe [5]; recently, several studies reported the risk factor prevalence and mortality predictors variation among races [6‐8]. Brazil has the largest universal health system in the world; in addition, it is characterized by intense racial admixture, social inequalities and cultural traditions that may impact the natural history of HF. Finally, comprehensive epidemiological, clinical and therapeutic data on chronic HF are still lacking and making the definition of population strategies for disease treatment and prevention at the least difficult to forecast.

In this scenario, we have conducted the GENIUS-HF (Genetic and ElectroNic medIcal records to predict oUtcomeS in Heart Failure patients) study, a Brazilian cohort that aims to contribute with the characterization of risk predictors and the impact of multimorbidity related to chronic HF [9].

The purpose of this study was to describe baseline characteristics, one-year outcomes and predictors of mortality and hospitalization of chronic heart failure patients.

We observed an overall mortality of 6.8% and a composite outcome of death and hospitalization of 17.7% in 1-year of follow-up. It is known that the HF mortality increases with the follow-up time and can reach a median of 40% in 2.5 years [13]. Our results are in agreement with previous reports. For instance, Maggioni et al. observed all-cause mortality rate at 1 year of 7.2% in chronic stable HF in a pilot study [14]. In addition, the continuation of this study showed all-cause 1-year mortality rate of 6.4% and combined endpoint of mortality or HF hospitalization within 1 year of 14.5% [15].

In our cohort, the variables associated with all-cause mortality at 1 year were elevated BUN or log BNP and lower SBP. On the other hand, considering the composite endpoint death and hospitalization, the predictors were age, high sensitive troponin, BNP and BUN. Previous studies have shown a variety of risk predictors [13, 16, 17] and between then, it is common to find creatinine as representative of renal function and systolic blood pressure. However, although BNP is markedly related to prognosis [18, 19] it is not present in the best-known risk models [13, 16]. Due to the fact that studies in chronic heart failure are scarce in the Brazilian population, there are no known critical variables in our population, except in acute patients as in the BREATHE Registry [20].

The majority of included individuals were in NYHA class I/II (81%) denoting the stable-outpatient character population. Besides, the use of beta-blocker (96.8%) and ACEi or ARB (91%) was high. These facts refer to the good clinical care of the cohort, certainly influencing outcomes.

There was a predominance of hypertensive (26.0%) and ischemic (21.9%) etiologies in our sample, on the other hand, the number of chagasic patients was also important (17.0%). Most studies describe ischemic as the main etiology found in the HF population, however, it depends on study design and ascertainment approaches [5]. Nonetheless, in our population, Chagas disease is still a major concern. In Brazil, there are endemic areas of Chagas’ disease such as the Midwest region [21] and migration movements can explain this relative high prevalence. Recently, Nadruz et cols evaluated the population attributable risk (PAR) of Chagas cardiomyopathy for 2-year mortality among patients with HF enrolled at years 2002–2004 (era 1) and 2012–2014 (era 2). The era 2 population is part of our cohort and the results found were that although the absolute death rates decreased over time in the Chagas cardiomyopathy and non-Chagas cardiomyopathies groups, the PAR of Chagas cardiomyopathy for mortality increased among patients with HF. Therefore, the current knowledge indicate that of all etiologies, Chagasic HF has the worst prognosis [22]. In addition, Bernardez-Pereira et al. analyzed the association between genetic ancestry, self-declared race and hemodynamic parameters in the GENIUS-HF cohort and observed a predominance of European ancestry in the entire study population [23].Taken together, these facts make our cohort sui generis and suggest care in the use of risk prediction models from other populations.

The SEATTLE heart failure model was an example of a web-based calculator that estimates 1, 2 and 3-year survival using clinical and pharmacological data easily obtained, but could need calibration in different ethnic populations [16]. Regardless, many publications on HF risk scores in Europe, North America [13, 16, 17] and nowadays-in Asian populations [6‐8], there is a shortage of studies in the Brazilian population.

We found that patients with ischemic etiology have worse survival free of death and hospitalization in 12 months compared to others etiologies. This is a common observation to many reports and also supports prior community-based epidemiological studies that reported greater risks of coronary heart disease-related deaths [24].

High levels of BUN and BNP and low systolic blood pressure were independent predictors of one-year overall mortality in our population. Considering the composite endpoint death and hospitalization, independent predictors were age, high sensitive troponin, BNP and BUN.