The epidemiological transition is a process by which the change in the management of infants, children, and adults due to the immunization leads to the reduction of the infant mortality and the increase of people reaching adulthood and old age. It is responsible for the decline of transmissible diseases due to better access to primary health care and lower overall mortality [1] [2] . The passage of the rural population to the urbanization will cause changes in the environment and in the behaviour because of excessive consumption of fats, physical inactivity, smoking, high-calorie and high-sodium foods, sympathetic hyperactivity, psycho-social stress, and weak prevention [3] . These urban populations are exposed to high blood pressure, greater weight gain, increase in heart rate, stronger urinary excretion of sodium, hyperglycemia, and hypercholesterolemia [4] [5] . The long-term effect of all these modifiable cardiovascular risk factors associated with unmodifiable risk factors (age and heredity) results in heart failure (HF) [6] .

The heart failure is a frequent disease in the adult population in Africa. Diagnostic and management of this disease require specific heart investigations and treatments that are often inaccessible in the developing countries [7] [8] .

The non ischemic heart diseases, including hypertensive, valvular, rheumatic heart disease are the leading cause of HF in Africa [9] . Other HF causes, such as endomyocardial fibroses, are unique to some African regions [10] .

The purpose of this study was to determine the cardiovascular risk factors and the causes of HF and to evaluate the prognosis of the patients with HF at the Institute of Cardiology of Abidjan (ICA) and the National Police Hospital (NPH).

That was a prospective study conducted from January 2016 to December 2016 at the Institute of Cardiology of Abidjan and the Cardiology Department of the National Police Hospital in Côte d’Ivoire. This study included 989 patients who presented HF.

Ø Diagnostic and study parameters

The diagnostic of HF was established according to the Framingham criteria and to usual paraclinical signs, including the analysis of the electrocardiogram (ECG), chest x-ray, and echocardiography [11] [12] [13] . The dosage of the brain natriuretic peptide was tested for diagnosis and follow-up. Sex, age, clinical tolerance of the New York Heart Association (NYHA) score or grading were collected. Cardiovascular risk factors, such as smoking, type 2 diabetes, high blood pressure, and hypercholesterolemia were studied. Severe kidney impairment was retained when creatinine clearance was less than 30 ml/min. The virus of human immunodeficiency (HIV) testing (miniVidas 2 ELISA test) was carried out in all patients. The population of study underwent a resting ECG and a follow-up test (an exercise ECG). Echocardiography was done in all patients according to the recommendations for HF care [12] [14] . Left ventricular ejection fraction (LVEF) was evaluated by ultrasound at least once during the hospitalization and at least once per year over the follow-up.

Ø Etiologic research

The diagnosis of ischemic heart disease was made when the patient had either a medical background of typical angina or of myocardial infarction. Data from resting or exercise ECG and echocardiography with respect to the medical history of myocardial infarction or evolution of myocardial ischemia were considered for diagnosis of ischemic heart disease. The coronarography was performed in some patients. The diagnosis of hypertensive heart disease was made on the basis of a long-lasting medical history of high blood pressure with or without the use of long-term anti-hypertensive drugs associated with left ventricular hypertrophy [12] . Valvular heart disease was diagnosed with respect to echocardiography data regardless to rheumatic or degenerative causes. Echocardiography findings consisted of left ventricular dilatation (telediastolic diameter > 55 mm) or biventricular dilatation and systolic dysfunction without previously mentioned causes confirmed the diagnosis of primitive cardiomyopathy [9] [12] .

Ø Follow-up of patients

The patients were treated according to the current recommendations [12] . The evolution data were collected either by a consultation within 3 months after the first hospitalization and then regularly in cases where it was possible or by new hospitalizations. The patients or their family were contacted at the end of the data collection. The causes of death were reported.

Data were collected in an Excel database from Windows 8 (Microsoft Corporation, Redmond, WA, USA). Data were expressed as means with standard deviation. Discrete variables were expressed in percentages. The t-test and Chi-Square Test were used to compare quantitative variables and qualitative variables, respectively. A p-value ≥ 0.05 was not considered statistically significant (NS). Patients were compared according to NYHA score or grading, duration of QRS interval (> or < 120 ms), LVEF (LVEF thresholds: > or < 30% and > or < 45%), and the follow-up (death or survival).

During the 12-month study, 989 patients were hospitalized at the Department of Medicine of the ICA and the NPH. The mean age was 55 ± 12 years (extremes: 25 - 73 years). Patients under 50 years, between 50 and 70 years, and ≥ 70 years were 42%, 60%, and 18%, respectively. Cardiovascular risk factors were high blood pressure (60%), tobacco (19%), type 2 diabetes (11%), and hypercholesterolemia (8%). HIV positive patients, ignored upon admission, were 11% of cases. Severe kidney failure with 25% was observed in all hypertensive patients. Patients with atrial fibrillation were 20% compared to 80% of sinus rythm patients. The average duration of QRS was 102 ± 24 ms. Patients with complete left branch block were 20%. The mean of LVEF was 35.8% ± 13%. The LVEF was ≤ 45% in 57% of cases while 30% of patients had LVEF ≤ 30%. Ischemic heart disease (60%) is the most frequent cause of HF followed by hypertensive heart disease (20%) and rheumatic valvulopathy (12%). Among the ischemic heart disease patients, 80% had a myocardial infarction. 50% of patients with ischemic heart disease underwent a coronarography and a myocardial revascularization. Primitive cardiomyopathy patients were 15% with normal coronarography.

Over the follow-up (average: 12.6 ± 8 months), 18% of patients died, 14% of whom died within 1 year after the initial diagnosis and 5% occurred in 19 months after the first symptoms of HF (Table 1). Sudden death (5%), end stage HF (13%), and end stage kidney failure (2%) were the major causes of death of the study population. 28% of the patients were rehospitalized for HF while 5% of subjects presented an ischemic stroke over their follow-up. Regarding to the NYHA grading, 39% of patients were Class II; 43% were Class III; and 18% were Class IV at 1 month later after their admission to ICA.

No significant differences were observed between the deceased patients and the surviving individuals regarding to gender, age, risk factors or HF causes except the frequency of primitive dilated cardiomyopathy (p < 0.02). The change in LVEF was independant risk factor of mortality over the 12-month study. The NYHA score was strongly linked to the mortality with higher proportion of Classes II, III, and IV in deceased patients (p = 0.03).

60% of patients with LVEF > 30% suffered commonly from hypertensive heart disease (p < 0.05). Ischemic heart disease patients were younger (48 versus 56 years, p = 0.01) and were frequently diagnosed with diabetes (p < 0.021). They had a low average of LVEF which was not significantly different from that of