European Journal of Cardiovascular Medicine

Heart failure (HF) is a leading cause of cardiovascular morbidity and mortality in both developing as well as developed countries. Despite significant advancements in therapeutics targeted at preventing risk factors, early diagnosis and introduction of newer heart failure medications, the global burden of the disease still remains high and continues to rise. It is estimated that approximately 64.3 million people worldwide are currently living with heart failure(1). Hospitalization for heart failure (HHF) is often necessitated by the acute worsening of signs and symptoms of heart failure in patients with chronic HF. Recurrent hospitalizations for heart failure impose a substantial burden on healthcare services and also adds significant financial strain on patients and their families. The economic impact is particularly pronounced in low- and middle-income countries, where prolonged hospital stays and repeated admissions contribute to high expenditures. A subset of patients with chronic HF has tendency for frequent decompensation of HF more often than others despite strict adherence to guideline-directed medical therapy, lifestyle modifications, and dietary restrictions. In contrast other subset remain relatively symptom free with no recurrent hospitalizations. This variability indicates that multiple factors, including patient-specific characteristics, HF etiology, treatment adherence, and possibly environmental influences, contribute to the risk of recurrent HHF. Identifying these factors is crucial for developing targeted interventions to reduce hospital readmissions and improve patient outcomes. In this study, we aim to analyse and identify the key factors associated with recurrent HHF, which may help in risk stratification and individualized management strategies for heart failure patients.

The study was conducted at a tertiary care center in a Medical college and Hospital in Maharashtra. This was a prospective observational study in which a convenience sample of 150 patients with age more than 18 years were enrolled with either newly diagnosed heart failure or heart failure patients who were on regular guideline directed treatment and stable for at least preceding 30 days. Patients were followed regularly for one year with scheduled OPD visits every 3 months and during any hospitalizations for heart failure. Patients had HF due to various etiologies and were classified mainly as having HF due to Ischemic heart disease, Valvular heart Disease, Dilated Cardiomyopathy, Heart Failure with preserved ejection fraction due to various causes such as Hypertension, Chronic kidney disease etc. They were also classified as having heart failure with reduced ejection fraction (HFrEF) and Heart failure with preserved ejection fraction (HFpEF) as per established diagnostic criteria. Patient with congenital heart disease, HF with severe other comorbidities requiring repeated hospitalizations, Transient HF such as high-output states, Patients with end-stage renal disease on dialysis, Patients with advanced malignancies, Patients who lost to follow-up were excluded.

Baseline characteristics of all patients were recorded during enrollment which included Age, Sex, BMI, Dietary compliance, residence in rural or urban area, history of any addictions, Comorbidities like Diabetes Mellitus (DM), Hypertension (HTN), Chronic Kidney Disease, Chronic obstructive airway disease (COAD), Presence of Atrial Fibrillation (AF) etc. Routine blood investigations were also done for all patients which consisted of NTproBNP, CBC, HBA1C, Serum Creatinine, Serum Albumin etc. Detailed Echocardiography was done for all patients and parameters like Left ventricular ejection fraction (LVEF), LV end diastolic dimension (LVEDD), Presence of more than mild mitral regurgitation, Presence of Pulmonary hypertension was recorded.

Patient were prescribed standard guideline directed HF treatment and also medications for any coexisting comorbidities tailored to each individual. They were advised dietary recommendations to minimize salt intake, fluid restriction as needed. They were regularly examined during OPD visits and appropriate adjustments were made in medications as needed. Compliance of medications were checked in each visit. The primary outcome measure was the incidence of recurrent HHF defined as more than two hospitalizations for HF during study period of one year. Accordingly, patients were grouped into recurrent HHF group (2 or more HHF) and Non-recurrent HHF group (no or only 1 HHF). Probable trigger for decompensation of HF was also recorded for every HHF and was classified as Cardio-vascular (CV) related (e.g. Hypertension, Acute coronary syndrome, Arrhythmia) or Non-CV related such as Infections, worsening renal function, exacerbation of COAD etc.

STATISTICAL ANALYSIS

IBM SPSS software version 26 was used for statistical analysis. Continuous variables (e.g., age, BMI, LVEF, serum creatinine) were summarized using mean ± standard deviation (SD) or median (interquartile range, IQR) depending on normality. Categorical variables (e.g., sex, comorbidities, presence of atrial fibrillation) were presented as frequencies and percentages. Between the two groups continuous variables were compared using Independent t-test or Mann-Whitney U test. Categorical variables were compared using Chi-square test or Fisher’s exact test. Binary logistic regression was performed to identify independent predictors of recurrent HHF. Variables with a p-value <0.10 in univariate analysis were included in the multivariate model. Adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were reported. A p-value <0.05 was considered statistically significant.

The mean age of the cohort was 61.91 ± 10.8 years, with 65.3% (n=98) males and 34.7% (n=52) females. The majority of patients (47.33%, n=71) had heart failure due to ischemic heart disease (IHD), followed by dilated cardiomyopathy (22.7%, n=34), HFpEF due to hypertension, chronic kidney disease other causes (18%, n=27) and valvular heart disease (12%, n=18). HFrEF was present in 70% (n=105) of patients, while HFpEF was observed in 30% (n=45).

Table 1: Baseline parameters of a cohort

The mean left ventricular ejection fraction (LVEF) was 36.2% ± 8.6%. Among comorbidities, Hypertension (HTN) was present in 68.7% (n=103), Diabetes Mellitus (DM) in 46.7% (n=70), Chronic Kidney Disease (CKD) in 20% (n=30), and Atrial Fibrillation (AF) in 19.3% (n=29). A total of 58.7% (n=88) of patients resided in rural areas, and 36% (n=54) had a history of smoking or alcohol consumption. Mean serum creatinine was 1.3 ± 0.5 mg/dL, and NTproBNP levels were elevated in the majority of patients (median: 2480 pg/mL, IQR: 1800-3600 pg/mL).

Table 2: Baseline Categorical Patient Characteristics

The mean BMI was 27.3 ± 4.5 kg/m². Mean hemoglobin levels were 12.9 ± 2.1 g/dL, and HbA1c levels were 7.2 ± 1.5%. More than mild mitral regurgitation (MR) was present in 22% (n=33), and pulmonary hypertension was identified in 35.3% (n=53). Regarding medical therapy, ARNI was prescribed in 22% (n=33) of patients, ACEi/ARB in 56% (n=84), Beta-blockers in 81.3% (n=122), SGLT2i in 25.33% (n=38), MRA in 50.7% (n=76), and Diuretics in 65.3% (n=98).

Table 3: Comparison of various parameters in study groups

A total of 150 patients were considered for analysis after excluding patients who lost to follow-up. During study period of one year from the day of enrollment, a total of 9 patients died during hospitalization for heart failure. The earliest time of death from enrollment was 2 months and 6 days. Two patients died before first scheduled OPD follow-up and had severely reduced LVEF and high baseline NTproBNP. One patient died before 6 months of enrollment and rest of 6 patients died during last 3 months of the study. All deaths were due to direct result of decompensated heart failure and associated complications, and most of them were hospitalized multiple times during study period. 46 patients had hospitalization for HF two or more times during study period and were included in recurrent HHF group, while 104 patients were included in Non-recurrent HHF group. Out of total patients died 8 patients had recurrent HHF suggesting higher mortality in this group.

Trigges for Decompensation Leading to HF Hospitalization

Number of total hospitalizations for HF recorded were 169. Among the CV-related triggers, Hypertension accounted for 28% (n=47), Acute Coronary Syndrome (ACS) for 21.9% (n=37), and Arrhythmias for 11.2% (n=19). Non-CV-related triggers included Infections in 20.1% (n=34), Worsening renal function in 12.4% (n=21), and Exacerbation of COPD or other respiratory diseases in 6.5% (n=11).

Comparison Between Recurrent and Non-Recurrent HHF Groups

Patients in the recurrent HHF group were significantly older (mean age: 65.1 ± 9.4 years vs. 60.5 ± 11.2 years, p=0.02) and had lower LVEF (mean: 32.8% ± 7.4% vs. 38.5% ± 8.2%, p=0.001). Male sex was associated with a higher number of HF hospitalizations, accounting for 67.4% (n=31) of the recurrent HF cases, compared to females (p=0.04). Comorbidities were more prevalent in the recurrent HHF group, particularly DM (67.2% vs. 44.6%, p=0.005), CKD (41.4% vs. 17.4%, p=0.002), and AF (29.3% vs. 12.0%, p=0.009). The mean serum creatinine level was 1.5 ± 0.6 mg/dL in the recurrent HHF group and 1.2 ± 0.4 mg/dL in the non-recurrent HHF group, with a statistically significant difference (p=0.03). NTproBNP levels were also significantly higher in patients with recurrent HHF (median: 3150 pg/mL, IQR: 2450-4100 pg/mL vs. 2080 pg/mL, IQR: 1500-2900 pg/mL, p=0.004).

Patients in the recurrent HHF group had a higher prevalence of moderate-to-severe mitral regurgitation (32.8% vs. 15.2%, p=0.01) and pulmonary hypertension (48.3% vs. 29.3%, p=0.03). Poor medication compliance was more frequently noted in the recurrent HHF group (37.9% vs. 18.5%, p=0.01).

Patients with recurrent HHF had significantly lower hemoglobin levels (11.9 ± 2.0 g/dL vs. 13.4 ± 2.2 g/dL, p=0.01) and significantly higher left ventricular internal diastolic diameter (LVIDD) (56.8 ± 6.2 mm vs. 52.3 ± 5.8 mm, p=0.002), indicating an association between these parameters and recurrent heart failure hospitalizations. In contrast, BMI (27.5 ± 4.6 kg/m² vs. 27.2 ± 4.4 kg/m², p=0.72), HbA1c levels (7.3 ± 1.4% vs. 7.1 ± 1.5%, p=0.58), serum albumin levels (3.7 ± 0.5 g/dL vs. 3.8 ± 0.5 g/dL, p=0.45), residence status (urban vs. rural, p=0.45), history of addiction (p=0.38), and dietary compliance (good vs. poor, p=0.41) were not significantly different between the recurrent and non-recurrent HHF groups.

Predictors of Recurrent HHF

On univariate analysis, age, LVEF, DM, CKD, AF, NTproBNP levels, moderate-to-severe mitral regurgitation, and pulmonary hypertension were significantly associated with recurrent HHF. On multivariate logistic regression analysis, independent predictors of recurrent HHF were, Lower LVEF (OR: 1.12 per 5% decrease, 95% CI: 1.05-1.20, p=0.002), Chronic Kidney Disease (OR: 2.45, 95% CI: 1.30-4.61, p=0.005), Atrial Fibrillation (OR: 2.17, 95% CI: 1.08-4.36, p=0.03), NTproBNP ≥ 3000 pg/mL (OR: 1.89, 95% CI: 1.12-3.18, p=0.02), Poor medication compliance (OR: 2.78, 95% CI: 1.43-5.41, p=0.003). Additionally, those who were not on diuretics (OR: 2.31, 95% CI: 1.24-4.32, p=0.008) and not on beta-blockers (OR: 1.96, 95% CI: 1.12-3.42, p=0.02) had a higher risk of recurrent HF hospitalizations. At the end of one year, overall mortality was 6% (n=9), with significantly higher mortality in the recurrent HHF group (17.4% vs. 1.0%, p=0.001).

Figure 1: Multivariate Logistic Regression Analysis of Recurrent HHF Predictors

Heart failure is a chronic and often progressive disease with risk of decompensation due to various triggers. Heart failure (HF) decompensation is a complex process driven by multiple interrelated pathophysiological mechanisms. The primary triggers for decompensation often involve hemodynamic stressors, neurohormonal activation, and inflammatory responses. Increased preload due to volume overload, reduced cardiac output leading to organ hypoperfusion, and heightened sympathetic nervous system activity contribute significantly to worsening HF symptoms. Additionally, activation of the renin-angiotensin-aldosterone system (RAAS) and excessive vasopressin release lead to sodium and water retention, exacerbating congestion and increasing ventricular wall stress.(2) These processes are further amplified by systemic inflammation, oxidative stress, and endothelial dysfunction, all of which can precipitate recurrent HF hospitalizations.(3) Behind these complex pathophysiological processes there are multiple patient related , disease related, treatment related and other epidemiological factors which can be measured and monitored by treating physicians or patients themselves so as to prevent ultimate occurrence of HF decompensation. Understanding these mechanisms is crucial for identifying at-risk patients and implementing targeted interventions. Some of these factors have definite relationship with decompensation of HF and are well studied in prior studies, but many of other important factors also need attention.

In this study, we found that 30.7% of patients (n=46) experienced recurrent hospitalizations for HF, consistent with prior reports that highlight the high burden of rehospitalization among HF patients(4). Although most of previous studies have reported re-hospitalization rate over short period of time such as 30 to 60 days and considered only the first hospitalization, our study on the other hand has focused on recurrent re-hospitalizations over longer period of one year. So the rate of recurrent HHF is quite high despite regular 3 monthly OPD follow-ups, meaning that this subset needs even more frequent follow-ups. Patients in the recurrent HF hospitalization (HHF) group were significantly older and had lower left ventricular ejection fraction (LVEF), a finding in line with existing literature that suggests old age and reduced LVEF as a strong predictor of adverse HF outcomes(5). Elderly patients often have multiple co-morbidities and high prevalence of arrhythmias like atrial fibrillation which make them predisposed for HF decompensation. Patients with severe LV dysfunction and low LVEF are very sensitive even for minor degree of changes in volume status and sympathetic drive, Hence the decompensation occurs secondary to different triggers which demand high cardiac output and increase in heart rate. Consequently, this subset needs frequent follow-up and monitoring of various clinical parameters. They need to follow strict dietary adherence & restrict their fluid intake and also take adequate dose of diuretics.

Among comorbidities, diabetes mellitus (DM), chronic kidney disease (CKD), and atrial fibrillation (AF) were more prevalent in the recurrent HHF group. Diabetes is a significant and established risk factor for recurrent hospitalizations in HF patients due to its association with worsening cardiac function, endothelial dysfunction, and fluid retention(6). Poor glycemic control contributes to increased inflammation, myocardial fibrosis, and renal impairment which explains frequent decompensations in this subset(7). CKD has been consistently associated with worse HF prognosis due to altered volume regulation (reduced GFR), neurohormonal activation (RAAS) and endothelial dysfunction (8). There is pathophysiologic interplay between the cardiac and renal axes which is disturbed in CKD. Similarly, AF leads to loss of atrial contraction, irregular and often fast ventricular response, and increased thromboembolic risk, further exacerbating HF progression(9). Elevated NTproBNP levels were significantly associated with recurrent HHF, reflecting higher cardiac wall stress and worsening myocardial dysfunction, as noted in previous studies (10). Elevated NT-proBNP levels indicate increased cardiac wall stress and volume overload, which are key contributors to heart failure progression and recurrent hospitalizations. Persistently high levels suggest ongoing myocardial dysfunction, poor response to treatment, and heightened neurohormonal activation, all of which lead to frequent decompensations(11).

Our findings emphasize the critical and obvious role of medication adherence in preventing HF decompensation. Poor compliance was significantly higher in the recurrent HHF group as expected and emerged as an independent predictor of recurrent hospitalization and this was despite regular three monthly follow-ups. This aligns with studies indicating that non-adherence to guideline-directed medical therapy (GDMT) results in increased morbidity and mortality among HF patients(12). Strategies to enhance medication adherence, such as patient education, maintaining medication diary, frequent follow-ups, and simplified dosing regimens, may help reduce hospital readmissions.  Additionally, our analysis revealed that moderate-to-severe mitral regurgitation and pulmonary hypertension, which are interrelated were significantly associated with recurrent HHF(13). Mitral regurgitation leads to increased left atrial pressure and volume overload especially during increased heart rate secondary to various triggers and in any increased afterload situation, contributing to pulmonary congestion and recurrent HF exacerbations (14).

Anemia was significantly more prevalent in the recurrent HHF group, with lower mean hemoglobin levels, and was associated with an increased risk of hospital readmissions. Low hemoglobin results in reduced oxygen delivery, increased cardiac workload, and worsening myocardial function. Previous studies have shown that anemia in heart failure patients contributes to worse clinical outcomes, including higher hospitalization rates and increased mortality(15,16). There was no statistically significant difference in recurrent HHF based on residence (rural vs. urban), addiction history (smoking or alcohol consumption), or diet adherence, suggesting that these factors may have had a lesser impact on hospitalization rates compared to other clinical variables. The probable reason for their non-significance could be that heart failure progression and hospitalizations are more strongly driven by intrinsic cardiac dysfunction, comorbid conditions, and adherence to pharmacological therapy rather than these lifestyle factors. As far as heart failure medications are concerned the use of diuretics was associated with lower rates of rehospitalization as they help in symptomatic relief and volume management, thereby preventing acute decompensation episodes(17). Similarly, beta-blocker therapy was significantly linked to lower rehospitalization rates, as these drugs provide mortality benefits and reduce HF exacerbations by improving cardiac function, preventing arrhythmic events and fast ventricular response in patients with atrial fibrillation(18).

Regarding the triggers for decompensation, hypertension, acute coronary syndrome (ACS), and arrhythmias were the most common cardiovascular causes, while infections, worsening renal function, and chronic obstructive airway disease (COAD) exacerbation were the predominant non-cardiovascular triggers. These findings align with prior studies that highlight the multifactorial nature of HF decompensation and underscore the need for comprehensive management addressing both cardiovascular and non-cardiovascular factors(19,20).

LIMITATIONS

This study was conducted at a single tertiary care center, with relatively smaller sample size limiting the generalizability of our findings to a broader population. Patients in our study were closely followed with 3 monthly OPD visits which may not reflect real world scenario as patients tend to avoid hospital visits once they start feeling better, hence without such follow-ups re-hospitalization rate can be even higher. Some parameters like dietary compliance, history of addictions and even compliance to medications were reported by patients themselves and could have some bias.

Recurrent HF hospitalizations remain a significant challenge in HF management. In this study advanced age, lower LVEF, comorbidities (DM, CKD, AF), elevated NTproBNP, and poor medication adherence have emerged as key predictors for recurrent hospitalizations. Addressing these subsets through optimized medical therapy, patient education, frequent physical or telephonic follow-ups and preventing triggers responsible for decompensation can help reduce hospitalizations and improve patient outcomes.

1. Lou ZC, Lou ZH, Zhang QP. Traumatic tympanic membrane perforations: a study of etiology and factors affecting outcome. Am J Otolaryngol. 2012;33(5):549–55.
2. Bishnoi T, Marlapudi SK, Sahu PK. Factors Influencing the Outcome of Spontaneous Healing of Traumatic Tympanic Membrane Perforation: A Clinical Prospective Observational Study. Indian J Otolaryngol Head Neck Surg. 2023 Sep;75(3):1774–1781.
3. Kim AS, Betz JF, Reed NS, Ward BK, Nieman CL. Prevalence of tympanic membrane perforations among adolescents, adults, and older adults in the United States. Otolaryngology- Head and Neck Surgery. 2022 Aug;167(2):356–8.
4. Sogebi OA, Oyewole EA, Mabifah TO. Traumatic tympanic membrane perforations: characteristics and factors affecting outcome. Ghana Med J. 2018 Mar;52(1):34–40.
5. World Health Organization. International classification of impairments, disabilities, and handicaps: a manual of classification relating to the consequences of disease, published in accordance with resolution WHA29.35 of the Twenty-ninth World Health Assembly, May 1976 [Internet]. Geneva (CH): World Health Organization; 1980. [Jun 30; 2024 ]. https://iris.who.int/handle/10665/41003 Available from: [Google Scholar]
6. Rollin M, Rogers P, Robinson P. Natural history of pediatric tympanic membrane perforation. Otol Neurotol. 2011 Feb;32(2):246–51.
7. Sagiv D, Migirov L, Glikson E, Mansour J, Yousovich R, Wolf M, et al. Traumatic Perforation of the Tympanic Membrane: A Review of 80 Cases. J Emerg Med. 2018 Feb;54(2):186–90.
8. da Lilly-Tariah OB, Somefun AO. Traumatic perforation of the tympanic membrane in University of Port Harcourt Teaching Hospital, Port Harcourt. Nigeria. Niger Postgrad Med J. 2007 Jun;14(2):121–4.
9. Orji FT. Non-explosive blast injury of the tympanic membrane in Umuahia, Nigeria. Niger J Med. 2009 Oct-Dec;18(4):365–9.
10. Hempel JM, Becker A, Müller J, Krause E, Berghaus A, Braun T. Traumatic tympanic membrane perforations: clinical and audiometric findings in 198 patients. Otol Neurotol. 2012 Oct;33(8):1357–62.
11. Jellinge ME, Kristensen S, Larsen K. Spontaneous closure of traumatic tympanic membrane perforations: observational study. J Laryngol Otol. 2015 Oct;129(10):950–4.
12. Lin deman P, Edström S, Granström G, Jacobsson S, von Sydow C, Westin T, et al. Acute traumatic tympanic membrane perforations. Cover or observe? Arch Otolaryngol Head Neck Surg. 1987 Dec;113(12):1285–7.
13. Orji FT, Agu CC. Patterns of hearing loss in tympanic membrane perforation resulting from physical blow to the ear: a prospective controlled cohort study. Clin Otolaryngol. 2009 Dec;34(6):526–32.
14. Adegbiji WA, Olajide GT, Olajuyin OA, Olatoke F, Alabi SA, Nwawolo CC. Traumatic Tympanic Membrane Perforation in Tertiary Health Institution. The Journal of Middle East and North Africa Sciences. 2018;4(1):1–7.
15. Sailesh AV, Arumugham V, Pakanati SSR, Potnuru S. Traumatic tympanic membrane perforation: an overview in a tertiary care centre-Khammam. International Journal of Otorhinolaryngology and Head and Neck Surgery. 2021;7(8):1275–8.
16. Ritenour AE, Wickley A, Ritenour JS, et al. Tympanic membrane perforation and hearing loss from blast overpressure in Operation Enduring Freedom and Operation Iraqi Freedom wounded. J Trauma. 2008;64(2 Suppl):S174–8.
17. Armstrong B. Traumatic perforations of the tympanic membrane: observe or repair? Laryngoscope. 1972;82:1822–30.
18. Orji FT, Agu CC. Determinants of spontaneous healing in traumatic perforations of the tympanic membrane. Clin Otolaryngol. 2008;33:420–6. doi: 10.1111/j.1749-4486.2008.01764.x.