The term “Acute Heart Failure” encompasses a wide variety of clinical presentations, from acute pulmonary oedema (APO) to chronic decompensated heart failure including isolated right heart failure and cardiogenic shock [1] . In developed countries, heart failure prevalence is approximately 1% - 2% of the adult population, reaching a percentage above 10% among the over-70s [2] . This is the leading cause of hospitalization in cardiology departments in Africa, with a prevalence respectively estimated at 25.6% and 37.7% in Lomé and Dakar [3] [4] . Therapeutic management of acute heart failure focuses on symptom reduction and hemodynamic stabilization, in conjunction with specific treatment of a precipitating factor [1] . Its prognosis is severe, with 10% as the hospital mortality; this rate is close to 25% among patients over 70 years [5] [6] . Readmission rate at 60 - 90 days reaches 10% in Europe [1] . In Burkina, a study conducted in 2014 found a hospital mortality estimated at 12.8% [7] . The absence of universal health insurance in Burkina Faso and poverty represents specific challenges for the management of acute heart failure. In Africa, morbidity is estimated at 15% and mortality between 10% and 20%.

Previous studies conducted in Burkina have not addressed the out-of-hospital evolution and the impact of poor prognostic factors on mortality. Hence, our work aims to study the epidemiological, clinical, paraclinical, therapeutic and evolutionary aspects of acute heart failure.

This was a retrospective observational, descriptive study conducted over a period of three (03) months, from February 1, 2016 to April 31, 2016 within the cardiology department of Yalgado Ouedraogo Teaching Hospital (CHU-YO) in Ouagadougou, BURKINA FASO. Patients admitted in the cardiology department of CHUYO and diagnosed with acute heart failure were included in the study. All patients were included in our study:

Data were entered on the software Epi Data version 3.1 and analyzed through STATA 13.1. Multivariate regression was used to analyze the effect of poor prognostic variables on death occurrence. The test was statistically significant if p < 0.05.

We encountered a number of difficulties in the course of our work. The main difficulties related to: the failure of some patients to comply with the schedule of visits, which prevented us from determining the actual rate of post-hospitalisation service; not all the echocardiographic parameters, such as the kinetics of the walls of the left ventricle and pulmonary pressures, were systematically recorded; and the census of patients (some of whom could not be registered during the study period), due to the multi-site nature of our study. Also, the failure to carry out the prescribed (routine) paraclinical investigations (blood tests, X-rays) constituted a bias in the analysis of certain parameters. Selection bias may have been introduced because all our recruited heart failure patients had at least a cardiac ultrasound and an ECG. The hospital setting of our study does not allow us to generalise our results to the whole of Burkina Faso.

The prevalence of AHF in our department during the study period was 45.25%. Our results are close to those of Ouédraogo in Ouagadougou, who found 27.6% [7] . However, they were higher than those of Pio et al. in Lomé who found 25.6% [4] . Heart failure is getting more and more frequent, in parallel with the increase in cardiovascular risk factors.

Our study noted a male predominance with a sex ratio at 1.58. EFICA study has also noted a male predominance [8] .

The average age was 58.24 ± 19.04 years, with extremes of 15 and 88 years. The age range [55 - 75 years] was the most represented group. Our results were lower than those observed in American and European series, whose average age was respectively 74 (ADHERE) and 73 years (EFICA) [8] . In underdeveloped countries, the high life expectancy and the later onset of ischemic and hypertensive heart diseases associated with improved management of pathologies such as coronary artery disease and hypertension in industrialized countries justify these discrepancies.

Congestive heart failure was present in 88.71% of cases among the clinical pictures of AHF upon admission within the department. Our results are above those found in other African series. Thiam et al. [3] in Dakar and Pio et al. [4] in Lomé have respectively found 67.6% and 67%. Dyspnea was the primary symptom observed by Coulibaly in Ségou, who noted 96.9% of patients admitted with dyspnea [9] . Indeed, dyspnea is the first sign of heart failure.

Ten patients (16.13%) had atrial fibrillation on ECG. This result was higher than that of Yassine Ragbaoui, who found 10.6% in Morocco [10] . This difference could be explained by the small number and young average age of our population.

In our study, LVEF was impaired in 54.10% of cases and preserved in 26.23% (diastolic heart failure). It was preserved in 11.40% of cases in the study by Pio et al. in Lomé [4] .

In our series, 32.26% of cases had high creatinine levels. This figure is lower than that noted in the EFICA study, which was estimated at 53% [8] . This could be explained by the small size of our population and the young average age in our study compared with the one in EFICA study. Indeed, renal function deterioration increases with age.

In our study, hypertensive heart disease was the most frequent etiology (33.87%). Our results can be compared to African data on cardiovascular diseases, where hypertension is the most frequent etiology, followed by valvulopathy [11] [12] .

Loop diuretics (furosemide) were the main medication used (90.32%), followed by ACE inhibitors. The latter were the most prescribed in Galinier’ study [13] . Mohamed Hassane noted a use of diuretics in 96% of cases. This could be explained by the fact that congestive heart failure was the most displayed mode in our study. Indeed, loop diuretics are the cornerstone in the treatment for acute congestive heart failure [14] . Nowadays, the combination of SGLT2 inhibitors provides more satisfactory results [15] [16] .

As for the average length of hospital stay, it was lower than that found in Jamal Kheyi’s study who found 12.1 ± 6.6 days [17] . This could be due to the difference in the organization, the operation of the health care system and the management methods.

Hospital mortality was 11.30%, a rate lower than that found in Ikama’ study in Congo and N’djessan in Abidjan, which were respectively 20.2% and 25.9% and is justified by the high number of comorbidities in their populations [11] . Our rate is lower than the rate found in EFICA study [8] in France (28%), and below the one of Thiam et al. in Dakar, which was 25.9% [3] . This difference could be related to the small size of our population. The rehospitalization rate was 1.82%. This low rate could be explained by the fact that 15 patients, corresponding to 27.27%, were lost to follow-up.

In our study, there was a statistically significant association between the rate of death and the low systolic blood pressure on admission; which corroborates literature data. Indeed, the mode of presentation remains the main determinant of early lethality [8] . Acute heart failure syndromes with shock are logically more serious, as shown by the EFICA results: the initial intrahospital lethality was 58% for patients with shock compared to 16% for patients without shock [8] .

Acute heart failure is the leading cause of hospitalization within the cardiology department of Yalgado Ouedraogo Teaching Hospital. It is a serious syndrome with a long duration of hospitalization and a non-negligible intra-hospital lethality, all the more important when it is associated with collapse or cardiovascular shock upon admission. The implementation of a national program to combat arterial hypertension, the main etiology of heart failure in our context, could reduce the incidence of this syndrome. For better emergency management, the development of appropriate algorithms at each level of our health pyramid should help in reducing morbidity and mortality.