The prevalence and correlates of hospital admission of patients with heart failure admitted to Jimma Medical Center, Ethiopia

¹Department of Internal Medicine, Jimma University, Jimma, Ethiopia, ²Department of Internal Medicine, Jimma University, Jimma, Ethiopia, ³College of Public Health and Medical Sciences, Jimma University, Jimma, Ethiopia, ⁴Department of Internal Medicine, College of Public Health and Medical Sciences, Ambo University, Ambo, Ethiopia

^&Corresponding author
Kedir Negesso Tukeni, Department of Internal Medicine, Jimma University, Jimma, Ethiopia

Introduction    

Heart failure is a complex clinical syndrome resulting from structural and functional impairments of ventricular filling or ejection of blood, which may arise due to abnormalities or disorders involving all aspects of cardiac structure and function [1]. The European Society of Cardiology (ESC) definition emphasizes typical symptoms and signs caused by the structural and/or functional cardiac abnormalities, resulting in a reduced cardiac output and/or elevated intracardiac pressures at rest or during stress [2]. Heart failure is a major cause of morbidity and mortality worldwide. The prevalence of heart failure increases significantly with age, occurring in 1-2% of the population aged 40-59 years and up to 12% of individuals >80 years old. According to the most recent statistics, hospitalizations decreased from 1,020,000 in 2006 to 809,000 in 2016. While prevalence of heart failure continues to rise, the incidence may be decreasing due to improved recognition and treatment of cardiovascular diseases and its comorbidities, as well as improved disease prevention strategies [3]. Overall, 5-year survival following the diagnosis of heart failure is about 50%. However, this depends on the severity of underlying heart impairment. For patients with severe heart failure, the 1-year mortality may be as high as 40%. The majority of these patients die of cardiovascular causes, most commonly progressive heart failure or sudden cardiac death [1,3]. Hospitalizations are common after the heart failure diagnosis. Following heart failure admission, mortality rates estimated to be as high as 14% at 30 days to 37% at 1 year to up to 75% at 5 years. Readmission with heart failure is also common, with about one-quarter of patients with heart failure readmitted at two months to nearly 50% at 6 months. With each subsequent admission, the risk of death rises [1]. This is linked to substantial burden for patients, careers and health care systems. Patients with chronic heart failure frequently experience recurrent hospitalizations as the disease progress to advanced stage.

Study done in Lebanon on readmission rate and factors associated with readmission among patients with heart failure showed that the majority of readmissions (73.61%) were due to heart failure exacerbations. Significant predictors of readmission were patient related comorbidities like history of diabetes mellitus, coronary artery disease, length of stay at the index admission. The other major risk identified risk for readmission was that the majority of the management of patients did not attain to the level of the expected evidence-based treatment guidelines [4]. Repeated hospital readmissions are frequent and increasing over time in patients with heart failure. The factors which contribute to readmission and frequent hospitalization after being discharged are non-compliance to treatment, poor adherence to medication due to financial issues, multiple co-morbidities associated with heart failure. According to the demographic data, elderly men with a mean age of 69 years accounts for the majority of patients who are readmitted with heart failure than young females. Anemia and iron deficiency are also among things which affect the quality of life of patients with heart failure, putting them at risk of rehospitalization and worsening underlying heart failure. Previous study has revealed that non-compliance, low hemoglobin and New York Heart Association (NYHA) Class IV of heart failure were the main factors associated with readmission [5]. When seen patient perspectives regarding reasons for heart failure readmission, there are many factors and cannot be easily categorized into mutually exclusive reasons. Distressing signs and symptoms, unavoidable progression of illness to advanced stage, influence of psychosocial factors, imperfect self-care adherence, and health system failures are the major ones [6].

Methods     

Study design and setting: a quantitative prospective observational hospital-based study was conducted at Jimma Medical Center, Ethiopia, from February 3^rd to March 3^rd, 2023. Jimma Medical Center is a tertiary teaching hospital located in Jimma Town, Oromia region of Ethiopia, serving a population of over 20 million population catchment areas. The medical center has over 1,500 healthcare providers, serving about 160,000 outpatients, 11,000 emergency cases and more than 4700 deliveries per year. The hospital provides coronary care services, including at outpatient follow-up clinics, emergency care services and inpatient services with electrocardiogram, echocardiography and stress testing services for patients with heart failure, and other major facilities among which diabetes care and follow-up, dialysis and higher imaging services including computed tomography (CT) and magnetic resonance imaging (MRI) are listed. The set-up also has installed catheterization machine, which is a brand new of its kind and to start on both diagnostic and interventional angiography very soon.

Data collection: a structured data collection format, which was tested for consistency prior to the data collection, was used including the socio-demographic characteristics (age, sex, chart number, marital status, level of education, monthly income, and area of residence) and clinical profiles of the patients with heart failure. The charts of the patients were collected and reviewed for additional data regarding the previous patient care, type and kind of medication the patient's taking, previous history of admission and the possible reason and management provided. The data collectors were used personal protective equipment like alcohol-based hand sanitizer, and face mask to reduce any risk of transmitting infections from the patients to the data collectors and vice versa.

Data analysis: data were analyzed using Epidata manager version 3.1 and then transported to SPSS version 26 for analysis. Demographic characteristics of the study subjects were analyzed descriptively. Categorical data was analyzed using frequencies (n) and percentages (%) and represented in graphs and pie charts. Continuous data were analyzed using mean (SD) for normally distributed data or median (IQR) for skewed data. Chi-square, Fischer's exact, students were used in bi-variate analysis while binary logistic was used to conduct multivariate analysis. Fischer's exact was used when cell frequency was less than five (n<5) while Chi-square test was used for cell frequency (n>5). Significance level required was <0.05. Odds ratio was used to show direction of association between the independent and dependent variables.

Study limitation: since this study is a cross-sectional study, it is impossible to analyze causal relationship between variables of the study. It also included only admitted patients to the medical ward. In addition to the small sample size, this makes it difficult to find factors associated with admitted cases with respect to non-admitted patients. Also, this is a finding from a single institution.

Ethical consideration: ethical clearance was obtained from the Institutional Review Board (IRB) of the Institute of Health, Jimma University. Written informed consent was obtained from the study participants before starting the data collection process. All patients were granted the right to withdraw from the research at any time. The confidentiality and privacy of patients were assured throughout the study period by removing the identifiers from the data collection tools using different codes. Neither the data records nor the extracted data were used for any other purpose.

Results     

Socio-demographic characteristics: all 69 patients who were admitted to Internal Medicine Wards of Jimma Medical Center, Ethiopia, with heart failure during study period were interviewed and enrolled into the study, with a 100% response rate. Most of the study participants were male (53.6%, Ratio of Male to female being 1:1.16) and most of them were within the age from 45 to 60 years. The majority of patients (52.2%) have attended elementary school while significant number of them (44.9%) didn´t have formal education. The approximate yearly income of most of the study participants (43.5%) was in the range of $464 - 928.56, while about the third of them had yearly income in the range of $928.56 - 1857.12. Most of the patients (89.9%) were married and living with their family. More than two-thirds of the study population need to travel a minimum of 10 kilometers to come to Jimma Medical Center for better care (Table 1).

Clinical characteristics, dietary habit, type and average doses of medications for heart failure: majority of patients (56.5%) had previous admission with heart failure and were having follow-up at the Jimma Medical Center, while the rest, 43.5% of them were newly admitted, either they were having follow-up or came from other health facilities with no previous history of admission to the medical center. About 60% of these patients were admitted to cardiac unit while the rest of them were admitted to General medical wards (39.1%) and to Medical ICU (2.9%) (Table 2). Out of 39 readmitted patients, majority (n=17, 43.6%) were readmitted after one year of their previous admission. Twelve patients were readmitted within the next 7 to 11 months of their prior admission, while 6 patients were readmitted within 3 months of their most recent admission and the rest 4 patients were readmitted within the next 4 to 6 months of their most recent admission (Figure 1). Among the readmitted patients, majority (n=16) were stayed as inpatient for 6 to 10 days in wards during their prior admission. Fourteen patients stayed for 11 to 15 days while 5 of them stayed as inpatient for 16 to 20 days, and 3 patients stayed even longer for 21 to 30 days in the hospital. Only one patient was discharged from ward within 5 days of their previous admission (Figure 2). Among the total 39 patients readmitted with heart failure, 14 of them (20.3%) were those diagnosed with ischemic heart disease and registered for follow-up. Valvular heart diseases account for 18.8% (n=13) among which chronic rheumatic valvular heart disease is the majority. The rest of patients had idiopathic dilated cardiomyopathy and hypertensive heart disease (Figure 3).

While majority of patients (82%) used to eat processed, boxed-canned and restaurant prepared foods few days in a month, about 10.4% of them didn´t use it at all during the last month. Two patients out of total 39 readmitted ones had taken processed or restaurant prepared food most days in a week; while only one patient used restaurant prepared food few days in a week during the last month of current admission. Majority of patients (53.8%) were adherent to their medications and having their usual follow-up, while 18 (46.2%) of the readmitted ones had discontinued their follow-up or had stopped taking their usual medications that was prescribed while on follow-up at the clinic. Among patients who discontinued their follow-up or medications, 38.8% of them were presented within the first 2 weeks of medication discontinuation, while about a half of them (50.1%) have discontinued their medication and follow-up for more than one month before current readmission. The majority of readmitted patients (51.3%) use approximate one teaspoon table salt per day. About 25.6% of them use two teaspoons, while 12.8% of them use three teaspoon of table salt per day. The average (mean) of teaspoon of table salt used per day is 1.87 in this study population (Table 3). Enalapril, long-acting Metoprolol, Losartan and Spironolactone are the most commonly prescribed anti-remodeling medications for heart failure in this study participants, of which the long-acting Metoprolol is the most prescribed one (n=30) which is followed by Enalapril (n=27) as anti-remodeling drug among readmitted patients. Out of 39 readmitted patients, 27 were taking Enalapril (15 patients were on 10 mg per day while the rest, 12 patients, were taking 5 mg per day). Only two patients were taking Losartan 50 mg per day. Regarding Long-acting metoprolol, 30 patients, of which 20 of them were taking 25 mg per day, 8 of them were taking 50mg per day and 2 patients were taking 100mg per day. Among 39 readmitted patients, 18 of them were taking Spironolactone 25 mg per day Table 4).

Causes of underlying heart failure, phenotypic classification, precipitating factor and comorbidities: pneumonia is found to be the leading causes of precipitating factors for heart failure, accounting for 56.6%, followed by drug discontinuation or nonadherence to management. About 10% of patients had atrial fibrillation as a precipitating factor. In some patients (7.2%), both chest infection and drug discontinuation have accounted for worsening of their heart failure. Hypertension, anemia and thyrotoxicosis are among the least common precipitating factors of current heart failure admission. Only one patient is documented, as there was no identifiable precipitating factor that it was heart failure. Regarding the etiologies for heart failure, about a third (33.3%) of them had ischemic heart disease, While valvular heart disease is the next common cause of heart failure. Hypertensive heart disease and idiopathic dilated cardiomyopathy accounts 15.9%, 14.5% respectively. Cor pulmonale and congenital heart disease are the least common cause of heart failure among the total admitted patients during the study periods. Though about half of patients didn´t have documented comorbidity, hypertension (29%) was the most common comorbidity associated with heart failure. Chronic lung diseases was about 10.1% while diabetes accounted only 7.2% of comorbidities among total admitted patients (Table 5). Nearly half of patients (48%) had heart failure with preserved ejection fraction, while 45% of them had heart failure with ejection fraction less than or equal to 40%. Only few patients (7%) had mildly reduced ejection eraction (Figure 4).

Investigations, laboratories and echocardiography and their results: from laboratory results, values of serum Creatinine, Sodium, Potassium, white blood cell count and hemoglobin were assessed in almost all of them, and were within the locally standardized normal range in the majority of the patients. Serum Creatinine level was higher in 13 patients, while 19 patients had low serum sodium (likely hypervolemic hyponatremia). Potassium was high in three of the study population, while again in three of them it was below normal laboratory level. Twelve patients had leukocytosis, 6 of them were having leukopenia while 25 of the patients were having varies degree of anemia as per locally acceptable laboratory standard (Figure 5). Serum level of troponin was high in two patients, nearly two-thirds (62.3%) of patients had serum Troponin within laboratory limit, while in the rest it was not determined may be there was no indication. Among patients for which echocardiography is done, about one-third of patients (31.8%) have echocardiographic evidence of myocardial infarction and also a similar number of patients (30.45%) had evidence of valve lesion. Evidence of cardiomyopathy other than myocardial infarction was seen in 23.2% of patients. While congenital heart disease and Cor pulmonale is seen in two patients each, 6 of the patients didn´t have echocardiographic evaluation during the study periods (Table 6).

Factors associated with hospital admission; precipitating factors and correlates: in this study, variables with significant association at P<0.05 in the Chi-square categorical test and P<0.25 in the binary logistic were entered into multivariate analysis to determine associated factors of heart failure admission and among these factors those with P<0.05 were considered to be statistically significant (Table 7). Accordingly, age has significant association with heart failure admission (P-value=0.020). Similarly, there is statistically significant association between precipitating factors and heart failure admission with P-value of 0.011. In binary logistic regression age, distance from hospital, precipitating factors, phenotype of heart failure (HFrEF), functional class of heart failure, lower serum creatinine, lower serum potassium and echocardiographic finding had P-value of <0.25. However, upon further analysis only precipitating factors had significant association with Heart Failure admission with a P-value of 0.029, AOR 4.006 CI 95% (1.156-13.882).

Discussion     

Heart failure is a major cause of mortality and morbidity worldwide. While prevalence of heart failure continues to rise, new incidences are decreasing due to improved healthcare systems and early recognition as well as treatment that changes the natural history of heart failure. Generally, a 5-year survival of heart failure patients is about 50%. Furthermore, for patients with severe heart failure, one year mortality could be as high as 40%. Majority of patients die from progressive nature of underlying heart failure than associated comorbidities [1,3]. Hospitalizations are quite common after heart failure diagnosis. Mortality and morbidity rate of heart failure increases with each hospitalization. Following heart failure admission, mortality rate is about 14% at one month, which increase to 37% and 75% at one year and 5 years respectively. In addition to increasing mortality and morbidity, it also increases burden for caregivers, cost for medications or transportation and with recurrent hospitalizations, severity of the heart failure or associated comorbidities also increase [1].

The overall proportion of patients who are readmitted to medical wards of Jimma Medical center is 56.5%. Almost all of them are having follow up at cardiac clinic of the Jimma Medical Center. The rest proportion of heart failure cases are newly admitted, either they are diagnosed as outpatient or referred from other healthcare facilities with no previous history of admission with heart failure. The readmission rate in this study is high when compared to similar studies done elsewhere. For example, the study done in Tertiary Hospital in Tanzania shows readmission rate of patients with heart failure to be around 38.5% [7], whereas in the USA the studies show the readmission rates ranging from 14% to 30% [7,8]. The possible reason for this gap could be due to difference in health care system, optimization of guideline directed medical theraphy among others. This study shows that 43.6% of patients with heart failure were readmitted after one year of their previous admission, which is followed by readmission after 7 to 11 months and within 3 months respectively. Study in Lebanon on readmission rate found that readmission rates were 15%, 22.2%, and 27.8% at one month, 2 months and 3 months respectively. In Tanzanian tertiary hospital study, the readmission rate within one month of discharge from hospital was 25%. According to medicare analysis [9], heart failure accounted for 28% of all hospital readmissions in 6 to 9 months following the initial heart failure hospitalization. This is similar to the retrospective cohort study in Dutch which included 22,476 patients to see association of heart failure admission and drugs at discharge which shows median time up to readmission was 29.3 months [10]. In this study, the hospital readmission rate of patients within 6 months of their previous admission is lower than readmission after one year of index initial admission.

Among patients who were readmitted, 35.9% have Echocardiographic evidence of ischemic heart disease and 33.3% of them has evidence of primary valvular lesions, commonly chronic rheumatic valvular heart diseases. This is different from a retrospective descriptive registry review done on 1140 patients on follow-up clinic at tertiary hospital in Mogadishu, Somalia, which shows 39.8% of patients have evidence of hypertensive heart disease rather than Ischemic heart diseases while Valvular heart disease is nearly similar to a second causes of cardiac lesions, which accounts for 34.6% by echocardiographic examination [11]. Nearly half (46.2%) of patients who are readmitted are found to be either nonadherent to medications or have stopped their follow-up activities. About half of them have discontinued their medications for more than one month. This is much higher when compared to a cross-sectional study done in Gondar Comprehensive Specialized hospital on 245 adult patients with heart failure, which shows 23.7% of them were nonadherent to medications [12]. Despite the increased number, non-adherence to the medication is among common risk factors for readmission for heart failure. This is also supported from a prospective cohort study sample of 280 patients with heart failure to identify predictors of hospitalization, poor medication adherence was among important risk factors identified for hospital readmission [13].

Enalapril, long-acting metoprolol, Losartan and Spironolactone are among the most common heart failure medications prescribed in this study. From readmitted patients, 27 are taking Enalapril; 15 of them have been taking 10 mg of daily dose and 12 have been taking 5 mg daily dose. Majority of patients who have been taking Metoprolol also took 25 mg daily dose. Only 2 patients among readmitted took the level of 100 mg of Metoprolol per day. Heart failure treatment landmark trials show that treatment with medications at maximum tolerable dose decreases hospitalizations and all-cause mortality and morbidity. In European prospective study which aimed to investigate characteristics and treatment indication bias among patients with heart failure with reduced ejection fraction that did not take recommended doses of Angiotensin-converting Enzyme inhibitors, Angiotensin receptor blockers and beta-blockers, there is greater risk of death and heart failure hospitalization among patients taking less than 50% of recommended doses of these drug compared to those patients reaching greater than or equal to 100% [14]. Almost the majority of patients who were readmitted in this study took suboptimal dose of cardiac anti-remodeling agents. This study shows pneumonia and drug discontinuation are the most common and statistically significant precipitating factors, (P-0.029, AOR 4.006 CI 95% {1.156-13.882}), accounting 56.6% and 14.5% respectively for heart failure admission and/or readmission. This is somehow similar to a prospective observational study done in Tikur Anbessa hospital in which 169 patients included, pneumonia and Atrial fibrillation are the commonest precipitators for heart failure admission [15]. Furthermore, drug discontinuation was the second-commonest precipitating factor, while it was the third commonest in that study. Atrial fibrillation was observed in 10.1% of patients in this study. In OPTIMIZE-HF study 61% patients had 1 or more precipitating factors with pneumonia the commonest, similar with our findings, followed by Ischemia and arrhythmia [15]. These precipitating factors are associated with in hospital mortality and post discharge death or re-hospitalization [16].

The leading causes of underlying cardiac disease identified in this study are IHD, valvular heart disease, hypertensive heart disease, idiopathic DCM, Cor pSulmonale and Congenital Heart disease, in decreasing order. The leading cause of primary valvular disease is chronic Rheumatic heart disease followed by degenerative Valvular Heart Disease. This is different from a cross-sectional study conducted on Patterns of Cardiac disease at Jimma University Medical center about ten years back, in which 781 patients are included Rheumatic Heart Disease was the leading followed by hypertensive heart disease [17]. Prevention of Rheumatic Heart Disease that increasingly implemented since then might be the cause for decreasing prevalence, while the increment of coronary artery disease risk factors including diabetes mellitus, hypertension and others might lead to the increased risk for ischemic heart disease as a cause of hospital admission.

Conclusion     

Coronary artery disease is the leading cause of hospitalization with heart failure in this study. More than half of these patients were readmitted to hospital within the first year of their hospital discharge. Chest infections (community-acquired pneumonia) and medication discontinuation are the leading precipitating factors identified in the patients admitted and readmitted with the heart failure. Hypertension is the commonest co-comorbidities identified in the patients admitted with heart failure during the study periods. This may imply that there might be epidemiological changes of etiology of heart failure from infectious causes, mainly rheumatic origin, to the coronary artery diseases, which may be the result of increasing novel risk factors for the atherosclerosis, and still there may be the need to work on infection prevention and hygiene specially for patients with comorbidities to reduce risk of recurrent hospitalization with heart failure.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Kedir Negesso Tukeni, Megersa Negesa Geleta and Robsan Jaleta Regassa: contributed to the conceptualization, design, investigation, analysis and write-up of the first draft. Kedir Negesso Tukeni, Megersa Negesa Geleta, Robsan Jaleta Regassa and Eyob Girma Abera: participated in the design, data curation drafting, interpretation and edition of the data and supervision. Kedir Negesso Tukeni, Megersa Negesa Geleta and Eyob Girma Abera: were involved in the design, supervision and edition of the manuscript. All authors reviewed, edited the manuscript and approved the final version of the manuscript.

Acknowledgments     

We are grateful to Jimma Medical Center Medical ward staff and other healthcare professionals working in the ward for their kind cooperation throughout the study period.

Tables and figures     

Table 1: socio-demographic characteristics of patients with heart failure admitted to Jimma Medical Center, from February 3^rd to March 3^rd, 2023, Jimma Ethiopia

Table 2: admission characteristics of patients with heart failure admitted to Jimma Medical Center, from February 3^rd to March 3^rd, 2023, Jimma Ethiopia

Table 3: frequency of taking processed and restaurant foods, follow up status and approximate table salt usage among patients with heart failure readmitted to Jimma Medical Center, Ethiopia from February 3^rd to March 3^rd, 2023 (N=39)

Table 4: types and doses of cardiac anti-remodeling medications among patients with heart failure readmitted to Jimma Medical Center, Ethiopia from February 3^rd to March 3^rd, 2023 (N=39)

Table 5: precipitating factors, comorbidities and causes of underlying heart failure among patients with heart failure admitted to Jimma Medical Center, Ethiopia from February 3^rd to March 3^rd, 2023 (N=69)

Table 6: serum troponin level and echocardiographic findings among patients with heart failure admitted to Jimma Medical Center, Ethiopia from February 3^rd to March 3^rd, 2023 (N=69)

Table 7: factors associated with admission among patients admitted with heart failure to Jimma Medical Center, Ethiopia from February 3^rd to March 3^rd, 2023 (N=69)

Figure 1: duration of gaps between current admission and the most recent admission of patients with heart failure readmitted to Jimma Medical Center, Ethiopia

Figure 2: duration of hospital stays during previous most recent admission among patients with heart failure patients readmitted to Jimma Medical Center, Ethiopia

Figure 3: types of echocardiography-confirmed cardiac lesions during follow-up among patients with heart failure readmitted to Jimma Medical Center, Ethiopia

Figure 4: phenotypic classification of heart failure depending on ejection eraction of patients with heart failure admitted to Jimma Medical Center, Ethiopia

Figure 5: selected laboratory test results among patients with heart failure admitted to Jimma Medical Center, Ethiopia

References