The finding of the current study indicated that more than half of the patients admitted with AHF were males. This finding is lower than the finding of the EHFS II (2) but is similar to the findings of Esubalew and colleagues' (13), ADHERE (16), THESUS-HF(17), KorAHF(22), and OPTIMIZE-HF(23) studies.
The mean age of patients enrolled in this study was 48.33 ± 18.9 which is comparable to the findings of Esubalew et al (13) and THESUS-HF (17) but is lower than the ADHERE, KorAHF, and OPTIMIZE-HF studies (16, 22–23).
This study found that 51.1% of patients enrolled had ADHF syndrome and 70.2% had been admitted for this diagnosis within the previous year, which differs from the findings of EHFS II and ADHERE, which found that 63% of patients had ADHF and 37% had denovo HF, and 75 percent of patients had ADHF and 25% had denovo HF, respectively (2, 16). This may be due to the prevalence of poor drug adherence and literally high illiteracy in which 81% of study participants attended primary school and less; which could contribute for recurrent admission among patients with ADHF.
In this study, on admission to hospital, patients with NYHA IV, III, and II, were 76%, 20.1%, and 3.3%, respectively. This finding is comparable to the study done at TASH by Tirfe and his colleagues, among AHF patients presenting with NYHA IV, III, and II functional classes, which showed 69.2%, 24.3%, and 6.5%, respectively, and the ADHERE registry, 32%, 44%, and 20%, respectively(14, 16). When compared to patients in the ADHERE registry, there was a larger proportion of patients with NYHA IV functional class who came to JUMC. This could indicate that, as compared to the participants in these investigations, more patients were admitted with significant decompensation.
The most common signs and symptoms in our patients were paroxysmal nocturnal dyspnea (99.5%), cardiomegaly (91.5%), and neck vein distension (88%). In addition, Unlike the ALARM-HF registry, which found that the most prevalent presenting symptoms were rales (61%), orthopnea (73%), and elevated jugular venous pressure (40%). This could mean that more patients with substantial functional impairment were admitted to JUMC compared to the patients in the ALARM-HF registry (24).
In this study the mean (± SD) systolic blood pressure and pulse rate were 116.17(± 24.5) mmHg and 104.15 (± 21.5) bpm, respectively. This contrasts to the findings of THESUS survey which found 130.4(± 33.5) mmHg systolic blood pressure and 103.7 (± 21.6) bpm pulse rate; OPTIMIZE-HF which found 142.6 (± 33.2) mmHg systolic blood pressure and 86.6 (± 21.5) bpm pulse rate; and CHINA-HF registry which found 128 (± 26) mmHg systolic blood pressure and 82 (± 25) bpm pulse rate ( 17, 23, 25). This may be due to a difference in genetic background, racial differences, and diet, as well as a difference in study sample size, which could have an effect on the mean blood pressure measurement.
We found sinus tachycardia in 33.2%, and AF in 23.9%, of the patients in contrast to Tirfe et al's findings, which reported AF was the most common (54.1%), followed by sinus tachycardia (33.9%). Similarly, the KorAHF registry was AF (34.9%), followed by pathologic Q wave in 23.9% (14, 22).
In our study patients, the primary underlying causes of heart failure was ischemic heart disease. This finding is similar to the findings of EHFS II (2)and KorAHF registry (2, 22). But, our finding contrasts to the findings of THESUS survey (17), TaHeF study and two studies from Nigeria (Abuja Heart Study on urban Nigerians and Contemporary profile AHF in southern Nigerians) in which hypertension was commonest cause of acute heart failure (26, 27, 28).
In this study, 52.7% of patients had a depressed left ventricular ejection fraction (≤ 40%). In ADHERE registry, KorAHF registry, and the Oman Acute Heart Failure Registry, patients with an ejection fraction of less than forty percent were 56%, 60.5%, and 63%, respectively (16, 22, 29). This could be because HFrEF is commonly associated with ischemic heart disease, as discovered in this study, which found a link between IHD and low ejection fraction (p value = 0.000, 95% CI [3.13, 12.62], OR = 6.280).
Pneumonia, AF, medication discontinuation, and uncontrolled hypertension, as well as acute coronary syndrome, were the most common precipitating causes of AHF in this study, which makes this study comparable to the Get with the Guidelines-HF (GWTG-HF) database. OPTIMIZE-HF, a registry which involved 48612 patients from 259 US hospitals, showed the most common precipitating factors were pneumonia, acute coronary syndrome, arrhythmia, uncontrolled hypertension, non-adherence to medication, and worsening renal function (23). Pneumonia/respiratory processes, arrhythmia, medication noncompliance, increasing renal failure, and uncontrolled hypertension were the most common precipitating causes of heart failure between January 2005 and September 2013 (30). This suggests that early detection of pneumonia and educating and counseling patients on cardiac drug compliance need due attention to decrease the acute heart failure associated with hospital admission.
In this study, hypertension, pulmonary hypertension, chronic renal disease, chronic obstructive pulmonary disease, diabetes mellitus, TB, hyperthyroidism, asthma, and HIV/AIDS were all recognized as comorbidities. However, hypertension, coronary artery disease, and diabetes were the most common comorbid conditions in the ADHERE registry. Similarly, in KorAHF registry hypertension, ischemic heart disease, diabetes, AF, cerebrovascular illness, chronic renal failure, and chronic lung disease were the major comorbidities. Yet, in the China HF registry, hypertension, coronary heart disease, AF, and diabetes mellitus were the most common comorbidities (16, 22, 25).
The median hospital stay in this study was 9 days, which is comparable to the findings of the EHFS II(2), KorAHF registry(22), and CHINA-HF registry(25), but higher than the findings of the ADHERE (4.3 days)(16), OPTIMIZE-HF (4 days)(23), and THESUS-HF (7) days (2, 16, 17, 22, 23, 25).
In this study, poor treatment outcomes occurred in 52 patients, of whom 37 (20.1%) had in hospital mortality. Mortality was higher in patients with ADHF (51.35%) than in denovo HF (48.6%). Over all, in hospital mortality in this study is almost comparable to the prospective study done at TASH in which in hospital mortality was 17.20%, the data from a retrospective study done by Esubalew and his colleagues showed in hospital mortality of 24.4%, and the data from the prospective Tanzanian acute heart failure (TaHeF) study in which mortality was 22.4 per 100-year observation (13, 14, 26).
In contrast to these, hospital mortality rates for patients evaluated in OPTIMIZE-HF, KorAHF, ADHERE registry, and EHFS II ranged from 4–6.7% (2, 16, 22, 23). A prospective cohort study of international congestive heart failure (INTER-CHF) patients revealed regional variability in mortality among AHF patients, with mortality being highest in African patients, intermediate in South East Asia, and lowest in China, South America, and the Middle East (15).So, the difference in mortality rate may be related to health care infrastructure and variability in patients' characteristics at admission.
In this study, independent predictors of in hospital mortality identified were the presence of acute coronary syndrome as a precipitating factor, smoking, chronic kidney disease as comorbidity, and elevated BUN (≥ 43 mg/dl). In the CHINA-HF registry, acute coronary syndrome, BUN, infection, left bundle branch block on ECG, low systolic blood pressure, and elevated bilirubin level were all independent predictors of in-hospital mortality (25), while BUN, high serum creatinine, and low systolic blood pressure were all independent predictors in the ADHERE registry (16). In the absence of any other prognostic indicators, OPTIMIZE-HF found that risk-adjusted post-discharge mortality was considerably higher in patients whose hospitalization was prompted by ischemia/ACSs or worsening renal function (23). The majority of these studies show that acute renal function detriment or the presence of CKD and acute coronary syndrome were major predictors of mortality as in this this study.
Despite its efforts, the current study has some limitations. The laboratory tests for serum uric acid level, NT pro-BNP, BNP, and CRP cardiac troponin determination was not consistently available for all of the enrolled patients (troponin was only checked for 120 patients), making it impossible to draw a link between poor outcome and these variables. This is one of the limitations. It was also difficult for us to estimate the amount of salt consumption behavior among the patients and the cost of therapy since they are subjectively influenced. Yet, in contrast to retrospective researches, the strength of this study lies in its prospective study design, which yielded more or less useful information.