Tachycardia and Pre-existing Chronic Kidney Disease Are Predictors of the Worse Clinical Outcomes in Patients Recently Hospitalized With Acute Heart Failure

Background: This study aimed to assess the factors contributing to the outcomes of recently hospitalized patients with heart failure (HF). Methods: A prospective data of 76 adults who were admitted due to acute HF between October 1, 2019 and June 30, 2020 at our center were analyzed. Endpoints included survival and rehospitalization within six months after discharge. Results: The mean age was 64.9 ± 13.8 years, with a male preponderance (68.4%). Approximately 60.5% of patients had the left ventricular ejection fraction (LVEF) <40%, whereas 26.3% of patients had LVEF ≥50%. Coronary artery disease (75%), arterial hypertension (72.4%), chronic kidney disease (46.1%), and diabetes mellitus (46.1%) were the most frequent comorbidities. Poor compliance (40.8%) and non-cardiac infection (21.1%) were the common precipitating factors for hospitalization. The majority of subjects had severe symptoms, indicated by the frequent need of intensive care unit (43%), high N-terminal prohormone brain natriuretic peptide levels [NT-proBNP; median, 4765 (1539.7-11782.2) pg/mL], and presence of either atrial fibrillation, severe mitral regurgitation, or significant pulmonary hypertension in approximately one-third of cases. Even though in-hospital mortality was relatively low (2.6%), the all-cause mortality and rehospitalization rates in the next six months after discharge were still high, reaching 22.54% and 19.72%, respectively. Further survival analysis showed that tachycardia on admission and pre-existing chronic kidney disease (CKD) resulted in low six-month survival rates among these patients. Conclusion: After hospital discharge, patients with HF were still exposed to higher risks of death and readmission albeit with the medication addressed. Tachycardia on admission and pre-existing CKD might predict worse outcomes.


Introduction
Heart failure (HF) is a major global problem and the leading cause of morbidity and mortality worldwide [1]. HF is the primary reason for hospitalization in the elderly population, and the number of readmissions is high in this population at risk [2]. These data showed that HF is a crucial issue and challenge for health services in the community. Acute HF (AHF) refers to the rapid onset or worsening of symptoms and/or signs of HF, which require prompt treatment and frequently related to unplanned hospital admission or emergency room visit [3]. Generally, AHF encompasses a complex syndrome, including various clinical presentations, underlying cardiac problems, trigger factors, comorbidities, and complications, leading to a relatively high mortality rate during hospitalization and after discharge [4]. Therefore, patient-based, rather than disease-based, data might provide essential information to evaluate miscellaneous aspects of hospitalized patients with AHF [5,6]. Consequently, this approach could improve the clinical assessment and management strategy to reduce in-hospital mortality and rehospitalization rates caused by AHF.
To date, there are numerous published data on the epidemiology, clinical management, and outcomes of AHF in Western countries. However, similar information is still lacking and less comprehensive in Southeast Asian countries. Multiple factors, including different cultures, less organized systems, limited access to healthcare, poor adherence of patients, limited diagnostic or therapeutic tools, and lack of trained investigators, affect the scarce data provided in most Asian countries, including Indonesia [6,7]. Moreover, this lack of data is a major barrier to participation in global collaboration of the relevant issue [8]. Notably, 1 2 2 2 2 3 1 1 1 1 several published registries of AHF in Asian countries showed that the demographic characteristics, comorbidities, treatment approaches, and outcomes of patients with AHF are slightly different from the Western data [9,10]. Thus, it is pivotal to identify any differences to elaborate on whether the Western data and guidelines regarding the management of patients with AHF can be practically applied to Asian patients, especially in Indonesia.
This study aimed to describe the clinical characteristics and analyze the essential factors contributing to outcomes of hospitalized patients with AHF. This study will be of value in adding databases of AHF during hospitalization and until six months after discharge, especially in the Indonesian population. Thus, the results might lead to a protocol to reduce mortality and rehospitalization rates related to AHF.

Materials And Methods
This prospective study included all patients aged ≥18 years who were admitted to our hospital with the primary indication of AHF between October 1, 2019 and June 30, 2020, at Siloam Hospital Kebon Jeruk, Jakarta. AHF was defined based on the criteria delineated in 2016 European Society of Cardiology guidelines for HF [3]. Patients were further categorized to acute de novo HF (admitted for the first time without a history of HF) and acute decompensated HF (ADHF; sudden worsening of HF in previously diagnosed or hospitalized patients). Patients who were discharged or transferred by their own intention before fulfilling treatment or admitted by other primary reasons than AHF were excluded.
Data regarding the patients' demographic, clinical characteristics, comorbidities, history of cardiovascular procedures, HF etiology, precipitating factors, laboratory investigation, electrocardiography, echocardiography, and management during hospitalization were directly collected from hospital medical records. The outcomes were in-hospital mortality, defined as all-cause death during hospitalization for AHF; six-month rehospitalization, defined as rehospitalization for HF within six months after hospital discharge; and six-month mortality, defined as all-cause mortality within six months after hospital discharge. The outcome data regarding in-hospital mortality were collected from the medical records, and data on six months of rehospitalization and mortality were collected by either telephone contact or clinical visit within six months after hospital discharge.
Descriptive statistics were used to summarize the data. Categorical variables were presented as frequencies and percentages. For continuous variables, Shapiro-Wilk or Kolmogorov-Smirnov normality test was performed as appropriate. Normally distributed data were summarized as mean and standard deviation. Otherwise, median and interquartile range (25th and 75th percentiles) were used. Chi-square test or Fischer's exact test were performed to compare and determine the independent variables according to sixmonth mortality that will be included in the logistic regression model. Independent variables included in the Cox regression model were variables with chi-square or Fisher's exact ρ-value < 0.25. Survival estimates for six-month mortality were calculated using the Kaplan and Meier method. Statistical analysis was conducted using SPSS version 22.0.

Results
This study included a total of 76 patients admitted with the primary complaint of AHF during a nine-month registry period. The characteristics, comorbidities, and history of cardiovascular procedures are presented in  Approximately 43% of patients required intensive care during hospitalization. Meanwhile, the duration of hospitalization in the cardiac ward and intensive care unit was 5 (4-9) days and 4 (2.5-7) days, respectively. The common comorbidities observed among our patients were coronary artery disease (CAD) (75%), arterial hypertension (72.4%), chronic kidney disease (CKD) without hemodialysis (46.1%), and non-insulindependent diabetes mellitus (46.1%). Some patients had a history of undergoing previous cardiovascular procedures including percutaneous coronary intervention (18.4%), coronary artery bypass grafting (14.5%), surgical valve repair/replacement (5.3%), and cardiac resynchronization therapy (1.3%).
Most patients (65.8%) had systolic blood pressure between 90 and 139 mmHg at the time of admission, whereas the median heart rate was 88 beats per minute (bpm) ( Table 2). Underlying CAD was the most common etiology for AHF, encompassing chronic coronary syndrome (50% cases), prior myocardial infarction (17.1%), and acute coronary syndrome (15.8%). Based on the time interval of the event, only 13.2% of patients had de novo HF, whereas the remaining patients had ADHF. Meanwhile, with respect to the clinical profiles, the majority of patients (88.2%) presented with ADHF, and only 10.5% and 1.3% of them had acute lung edema with respiratory distress and cardiogenic shock, respectively. In patients with ADHF, the common precipitating factors were poor compliance (40.8%), non-cardiac infection (21.1%), acute coronary syndrome (9.2%), hypertensive emergency (3.9%), and acute mechanical cause (2.6%).

Variables n (%)
Clinical presentation  ACS, acute coronary syndrome; ADHF, acute decompensated heart failure; AHF, acute heart failure; bpm, beats per minute; DBP, diastolic blood pressure; DCM, dilated cardiomyopathy; HF, heart failure; HR, heart rate; IQR, interquartile range; JVP, jugular venous pressure; SBP, systolic blood pressure; SD, standard deviation; VHD, valvular heart disease. Table 3, the mean hemoglobin level at admission was 12.51 ± 2.3 mg/dL. Most patients had renal impairment with a median creatinine level of 1.7 (1.2-2.6) mg/dL, indicating a median estimated glomerular filtration rate of 38 (22.9-59.6) mL/min. Unfortunately, only a few patients (31.6%) underwent N-terminal prohormone brain natriuretic peptide (NT-proBNP) tests, and the median level of this cardiac biomarker at the time of admission was 4765 (1539.7-11782.2) pg/mL. Atrial fibrillation was identified in 29% of initial electrocardiography (ECG) of our patients. Widened QRS complex ≥120 ms on ECG was found in 13.3% of subjects. The premature ventricular complex was common in the ECG result, which was noted in 13.3% of patients.

Variable Value
Laboratory results, median (IQR), unless specified otherwise  By echocardiography study, the left ventricular ejection fraction (LVEF) < 40% was found in 60.5% of patients, whereas LVEF ≥50% in 26.3% of subjects. The LVEF was measured using the standard of Simpson's procedure from the apical four-chamber view. Since most patients had diastolic dysfunction, restrictive diastolic dysfunction was noted in 41.7% of patients. Nearly three-quarters of patients had mitral regurgitation, and severe regurgitation was identified in 27.1% of cases. More than one-third of subjects had estimated pulmonary artery systolic pressures >50 mmHg, indicating a high likelihood of pulmonary hypertension. The right ventricular dysfunction as simply indicated by reduced tricuspid annular plane systolic excursion was found in 40.4% of subjects.

Complication during admission
Respiratory distress and respiratory failure -  Regarding the management of patients, loop diuretics and nitrates were the most common intravenous drugs used, which contributed to 92.1% and 43.4% of patients, respectively. Additionally, inotropes were needed in approximately one-third of the subjects, with dobutamine and norepinephrine as the most preferred agents. Only 21.1% of patients received parenteral anticoagulation for venous thromboembolism treatment or prophylaxis.

Variables
Further analysis of these parameters in Kaplan Meier revealed that there were poorer six months survival rates in patients with tachycardia on admission (log-rank test 0.06; Figures 1 and 2) and underlying CKD (0.018; Figures 3 and 4). Patients with initial tachycardia and previous renal dysfunction had a lower mean survival time compared to those without these conditions (133 days vs 161 days, and 136 vs 169 days, respectively). It is important to note that none of these covariates reached the median survival at six months follow-up.

FIGURE 4: Kaplan-Meier estimate of mortality between patients with
AHF with CKD and without CKD (log-rank test 0.018).

Discussion
The patients in this study were younger (mean age, 64.9 years) than populations in Western studies [11], and were predominantly male and overweight or obese. Specifically, approximately one-third of patients were a very elderly group with age >75 years. CAD and hypertension were prevalent comorbidities in patients included in this registry. Interestingly, the majority of HF patients enrolled were overweight and obese. It seems that this overnutrition condition should be considered as an important comorbidity in our population. Additionally, CAD, with various clinical profiles, was the most common etiology of HF in our center and this fact is consistent with the data worldwide [9,11]. Other common causes identified in our population were hypertensive and valvular heart diseases.
At our center, acute decompensation of chronic HF was much more prevalent than the case of de novo counterpart. Importantly, poor compliance of the patients was the commonest precipitating factor of the worsening of HF symptoms, as it was found in nearly half of the admitted patients. This finding delineates the crucial role of patient education and sustainable services after hospital discharge to avoid rehospitalization or mortality. Patients with HF often need to undergo long-term and complex management. Thus, low health literacy may affect patient compliance to HF medication. It is understood that several patients did not understand their current situation and medication, as well as management plan of the disease. This lacking of understanding would easily result in abrupt cessation of the medication soon after the symptoms subsided [12]. Additionally, elderly patients tend to discontinue the medication when adverse drug effects appear [13]. It is important to provide adequate information regarding the common side effects of drugs to the patients. Routine evaluation of drug interaction and polypharmacy could be an effective solution to this problem.
Other frequent precipitants of ADHF in our subjects were non-cardiac infection, and pneumonia was the most common type of infection that we encountered. As generally known, patients with HF had an increased risk of developing pneumonia. Alveolar and interstitial congestion could disrupt physiological mechanisms of alveolar lining fluid as the cellular barriers. This damage would impair the microbial clearance and increase the susceptibility of the respiratory tract to infection [14]. Meanwhile, pneumonia and other systemic infections would increase cardiac workload, which is a detrimental situation for patients with chronic HF. The most common bacterial causative agents reported in patients with HF are Streptococcus pneumoniae and Staphylococcus aureus [15]. Besides, various types of viruses, such as influenza, parainfluenza virus, coronavirus, and human metapneumovirus, are also common causes of communityacquired pneumonia in this population. Nevertheless, co-infection by bacteria and viruses often occur [16,17]. Regarding this issue, the guideline recommends that patients with HF should receive pneumococcal and yearly influenza vaccination to reduce worsening of symptoms and hospitalization [3]. Besides lung infection, other non-cardiac infections, such as sepsis, urinary tract infection, and even soft tissue infection, can lead to worsening of HF symptoms and hospitalization [18].
Patients with low LVEF (<40%) dominated in this study (60.5% subjects), and this finding is similar to other Asian registries [9,11]. The higher proportion of HFrEF in our center might be correlated to CAD as the most common etiology and comorbidity encountered here. It is important to note that approximately one-third of patients with HF in this study had either atrial fibrillation, severe functional mitral regurgitation, or significant pulmonary hypertension. The high median NT-proBNP levels (4765 pg/mL) might indicate the relative severe HF symptoms in our population.
Intravenous diuretic, especially furosemide, was the most commonly administered drug during hospitalization. This agent is effective in a majority of cases of acute HF to relieve the volume overload symptoms, thus gaining negative water balance before discharge [9]. Although diuretic resistance might prohibit decongestion strategy, this problem could be solved by combining some diuretic agents [19]. Intravenous nitrates were also commonly administered to optimize symptom relief at the initial period, as long as there was no hypotension.
The in-hospital mortality rate at our center (2.6%) was considerably lower compared to the previously reported data from Indonesia, which were 6.7% and 3% [7,20]. Despite this lower death rate during hospitalization, the six-month mortality and rehospitalization rates significantly increased to 22.54% and 19.72%, respectively. Nevertheless, this six-month death rate was still lower than those of the previous reported Asian studies, which were 26.3% and 45.8% [21,22]. The relatively high mortality and hospital readmission rates within the next six months after discharge emphasized that HF is a serious disease with a rapidly progressive condition, albeit proper management during hospitalization. Thus, sustainable optimization of treatment after discharge is of paramount importance to reduce adverse events in the future. Delivering education and improving patients' compliance might offer an effective way to obtain better long-term outcomes; particularly, poor compliance was the most prevalent trigger of rehospitalization in our center. In contrast, clinician inertia might lead to suboptimal management of patients with HF. Since the Asian population has lower body weight and higher sensitivity to drugs than the Western population, underdosing and underprescription of HF-modifying drugs were common [23]. As generally known, suboptimal doses of ACE inhibitors, ARBs, beta-blockers, and aldosterone antagonists could subsequently increase the mortality and rehospitalization rate in patients with HF, particularly HFrEF.
The Cox regression model of six-month mortality was presented in Table 8. From this study, the hazard ratios of tachycardia during admission and CKD were 1.938 and 2.165, respectively. Tachycardia on admission and CKD increase the risk of mortality at the six-month follow-up even though it is not statistically significant. It can as the effect of a smaller number of respondents compared to other studies. Assessment for tachycardia and CKD is needed in the management of a patient with increasing survival as the finding in this study showed shorter time survival in patients with tachycardia and CKD.
Tachycardia at admission and pre-existing CKD could be predictors for worse clinical outcomes in the next six months after discharge. Although these two variables were not statistically significant, which might be related to the insufficient number of respondents, the confidence interval indicated a tendency of higher death rate, as shown in the survival rates on the Kaplan Meier estimate. Higher heart rate during the acute event of HF results from neurohormonal disturbances. This neurohormonal activation acts as a mechanism to compensate for low cardiac output, which is beneficial in the short term [24]. However, in the long term, tachycardia might increase myocardial oxygen demand, induce additional ischemia, and cause arterial stiffness [25]. Most patients in this population had acute decompensation episodes from chronic HF; therefore, they might be previously treated with heart-rate-lowering agents, such as beta-blockers. In this situation, tachycardia at admission may reflect poor heart rate control in chronic HF. Eventually, worsened clinical outcomes in patients with AHF with tachycardia may result from the interaction between neurohormonal alteration, poor heart rate control, suboptimal previous treatment, and severity of the syndrome itself [25][26][27][28].
Several established studies have shown that CKD is associated with poor prognosis in both acute and chronic HF. The higher the blood urea and creatinine levels, the worse outcomes in patients admitted with AHF [29][30][31][32][33]. Renal worsening in patients with AHF also contributed to the need for ICU, respiratory failure requiring mechanical ventilation, and prolonged length of stay in the hospital [32]. Notably, CKD presents in >40% of patients with HF [34]. CKD affects HF and vice versa, namely, cardiorenal syndrome to describe the relationship of these two [33]. The risk of cardiac impairment increases gradually with the decline in kidney function. In the long term, CKD causes fluid retention that further increases preload. Resistant hypertension, which is commonly encountered in advanced CKD, could increase the cardiac workload as a consequence of the elevated afterload [30]. In contrast, low cardiac output and congestion as the results of HF could cause hypoperfusion and renal vein congestion, leading to the deterioration of renal function [34]. Even though RAAS acts as a compensatory mechanism in HF, its benefits decline over time, even contributing to further cardiac remodeling which amplifies the clinical progression of HF. Long-term cardiac effects of RAAS are cardiac hypertrophy and fibrosis. Whereas the detrimental effects in the kidney are reduced medullary blood flow, tubular fibrosis, and efferent and afferent arterial vasoconstriction [30,32]. The result of this complex interaction may explain the worse outcome reported in patients with AHF with CKD compared to patients with AHF without pre-existing renal dysfunction.
This study has several limitations. First, the number of subjects was rather small. This unfavorable situation was related to a significant decrease in the number of patients visiting hospitals as the COVID-19 pandemic began in March 2020 in our region. Besides, in the same year, there was a cooperation withdrawal of the national insurance with our institution. This situation further limited the patient's preference to seek medical service to our center. Second, this evaluation came from a single-center experience and might not represent the characteristics and outcomes of HF in Indonesia's general population. Third, some parameters from the echocardiography examination were incompletely recorded. Fourth, natriuretic peptide and iron studies from the laboratory were only performed in a small number of patients, precluding further elaboration of these data. Lastly, our six-month follow-up did not include the type of medication with their doses. This situation contributed to the difficulty in evaluating whether the patients had received optimal medication after discharge. Hopefully, an upcoming larger representative nationwide registry could provide additional and valuable information regarding the clinical outcomes of the patients recently hospitalized with AHF.

Conclusions
This single-center and prospective study analyzed the six-month clinical outcomes following hospital discharge in patients with AHF who had predominantly HFrEF, were overweight or obese, had CAD and hypertension, and presented with ADHF. Poor patient compliance was considered the most common precipitating factor for rehospitalization. Although in-hospital mortality was relatively low, the six-month death and readmission rates were still high and worrisome. Tachycardia on admission and pre-existing CKD had been shown as potential predictors for the worse long-term adverse events in this population.
Human subjects: Consent was obtained or waived by all participants in this study. Siloam Hospital Ethics Commitee issued approval 051/SHKJ/S/V/2021. We, the undersigned, herewith certify that: The institution has been informed regarding the observational research activities in hospitalized patients with heart failure at the Cardiology Department from October 2019 until January 2021 under the work of dr. Leonardo Paskah Suciadi and team. The research mechanisms have been directed to follow the standard operating protocols of our institution. The researcher team has been given permission and approval upon data collected and analysed at the cardiac intensive care unit, wards, and outpatient clinics. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work.
Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.