Characteristics of Older Patients with Heart Failure Readmitted due to Acute Exacerbations within the Past Year

Abstract Objective: We aimed to examine the relationship between physical performance and readmission among older patients with heart failure (HF) over the past year. Methods: This retrospective cohort study included 325 patients with HF who were aged ≥65 years and were hospitalized for acute exacerbation between November 2017 and December 2021. We investigated age, sex, body mass index, length of hospital stay, initiation of rehabilitation, New York Heart Association (NYHA) class, Charlson comorbidity index (CCI) score, medications, cardiac/renal function, nutrition, maximal quadriceps isometric strength, grip strength, and Short Physical Performance Battery (SPPB) score. Data were analyzed using the χ2 test, Mann–Whitney U test, and logistic regression analysis. Results: Altogether, 108 patients met the inclusion criteria and were divided into the non-readmission (n = 76) and readmission (n = 32) groups. The readmission group exhibited longer hospital stay, more severe NYHA class, higher CCI score, higher brain natriuretic peptide (BNP) levels, lower muscle strength, and lower SPPB score compared to the non-readmission group. In the logistic regression model, BNP level and SPPB score were independent factors associated with readmission. Conclusion: BNP level and SPPB score were associated with readmission in patients with HF within the past year.

The number of older adults is increasing throughout the world.
In Japan, it has been reported that older adults account for 28.4% of the total population 1,2) . An increase in the incidence of heart failure (HF) due to aging has become a major problem. According to a previous study, the estimated prevalence of HF in the United States is 5.1 million, which is predicted to increase to 8 million by 2030 3) . A previous Japanese study estimated that the total number of patients with left ventricular dysfunction in Japan (both systolic and diastolic dysfunction) would increase from 979,000 in 2005 to 1,284,000 by 2030 4) . Patients with HF encounter problems such as increased readmission rates, increased medical costs, and multiple short-term hospitalizations 5,6) .
HF is a major cause of hospitalization and death among older adults worldwide 7) . Factors related to readmission of patients with HF include exacerbation of HF as well as various conditions such as respiratory diseases, kidney diseases, psychiatric diseases, bone and joint diseases, and diabetes 8,9) . Shiraishi et al. reported that 5.4% of the patients with HF were readmitted within 30 days, while 26.2% of them were readmitted within 1 year 10) . In addition, according to the Rationale and Design of Japanese Cardiac Registry of Heart Failure in Cardiology (JCARE-CARD), the readmission rate among patients with HF was as high as 35% within 1 year 11) , indicating a high risk of readmission among these patients within 1 year. Furthermore, patients with HF have to bear high medical costs, and it has been reported that patients with higher serum N-terminal pro-brain natriuretic peptide levels have to bear higher medical costs associated with hospitalization 12,13) . Thus, reducing the readmission rate of patients with HF would reduce the medical costs.
In recent years, many reports have shown that the Short Physical Performance Battery (SPPB) is useful for functional evaluation of older patients, and a meta-analysis revealed that an SPPB score of <10 points is an excellent predictor of all-cause mortality [14][15][16] . In addition, the SPPB score has also been reported to predict future decline in the activities of daily living (ADLs) among older adults 17) . The SPPB score at discharge may be useful in predicting future decline in the ADLs and readmission among patients with HF.
Therefore, we hypothesized that the SPPB score may be a strong predictor of readmission among patients with HF, and patients with HF having low SPPB scores would have been repeatedly hospitalized and discharged within the past year.

Participants
The present study was a single-center cross-sectional study. This retrospective cohort study included 325 patients with HF admitted to a community-based hospital in Kumamoto, Japan, for acute exacerbations from November 2017 to December 2021. An experienced cardiologist diagnosed acute decompensated HF based on the Framingham criteria and Japanese guidelines 18) . The inclusion criteria were age ≥65 years and ability to walk with assistance at discharge. Patients who did not consent to participation, patients who died during hospitalization, patients who were transferred to another hospital, and patients in a poor general condition that could have caused difficulties while conducting the measurements were excluded from this study. All participants in this study underwent cardiac rehabilitation.

Investigations
Patient characteristics and clinical parameters at discharge such as age, sex, body mass index (BMI), length of hospital stay, initiation of rehabilitation, New York Heart Association (NYHA) class, Charlson comorbidity index (CCI) score, comorbidities, medications, left ventricular ejection fraction (LVEF), brain natriuretic peptide (BNP), creatinine (Cr), blood urea nitrogen (BUN), estimated glomerular filtration rate (eGFR), hemoglobin (Hb), albumin (Alb), geriatric nutritional risk index (GNRI) score, maximal quadriceps isometric strength (QIS), grip strength, and SPPB score were investigated by reviewing the medical records. LVEF was not measured at discharge in many patients. Therefore, we used the data at admission. According to the Japanese guidelines, it has been shown that 60%-80% of the patients with HF having reduced ejection fraction and 60%-90% of the patients with HF having preserved ejection fraction have HF with unchanged LVEF 19) .

Assessment of the NYHA class, maximal QIS, and grip strength
We evaluated the maximal QIS, grip strength, and NYHA class at discharge. The NYHA class was determined by an experienced cardiologist and a physiotherapist specializing in cardiac rehabilitation.
The maximal QIS was measured using a handheld dynamometer (Mobie MT-100; SAKAI, Tokyo, Japan). Patients sat on a chair with their arms folded and knees flexed at 90°. The pad position was established at 30 mm above the lateral malleolus in all patients. Knee extension strength was measured twice on both the right and the left sides with maximum force. We used the maximum value multiplied by the leg length and divided by the body weight (Nm/kg).
Grip strength was measured to evaluate the upper limb muscle strength. A digital grip force meter (Grip D T.K.K. 5401; Takei Scientific Instruments, Niigata, Japan) was used to measure grip strength. The measurements were performed according to the method recommended by the Ministry of Education, Culture, Sports, Science, and Technology 20) . However, since many patients were unstable in the standing posture, measurements were performed in the sitting position. The maximum grip strength was measured twice on each side and the highest value was used.

Assessment of the SPPB score
The SPPB score was assessed based on standing balance, 4-m walking velocity, and time to rise five times from the seated position 14) . Participants were asked to hold each of the three standing positions, namely closed-leg standing, semi-tandem stance (heel of one foot touching the base of the toe of the other foot), and full tandem stance (heel of one foot directly touching the other toe) for 10 s during the assessment for balance. Walking velocity was measured using a 4-m walk performed at the patient's usual pace. Participants were allowed to use canes or walkers. For measurement of time to rise from the seated position, participants sat on a chair with their arms folded across their chest and stood up five times consecutively as quickly as possible. The time taken to complete this maneuver was measured.
These tasks were assigned scores from 0 to 4 according to the established methods. The scores from the three tasks were added to obtain the total SPPB score (range: 0-12). A higher score indicated better physical performance.

Statistical analysis
Continuous variables such as patient characteristics and clinical parameters were expressed as mean ± standard deviation. Categorical variables were expressed as numbers (percentages). Normality of distribution for continuous variables was evaluated using the Shapiro-Wilk test. Patients with HF were divided into two groups based on readmission within the past year. Baseline characteristics and physical function were compared using the Mann-Whitney U test and χ 2 test. Logistic regression analysis was performed to evaluate whether patient characteristics, clinical parameters, and physical function at discharge were associated with readmission within the past year. A model was constructed using a stepwise forward selection procedure based on Akaike's information criterion. The explanatory variables used in this model were those that showed statistical significance in the univariate analysis (P <0.05).
All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R (R Core Team, 2022). Specifically, it is a modified version of R commander designed to add statistical functions frequently used in biostatistics 21) .

Ethical consideration
This study was approved by the Ethics Committee of the Medical Corporation Tanakakai Musashigaoka Hospital (no. H30-3) and complied with the principles of the Declaration of Helsinki.

Patient characteristics and clinical parameters
A flowchart of patient selection is shown in Figure 1. Among 325 patients hospitalized with HF, 174 met the inclusion criteria. Sixty-six patients were excluded for not providing consent (n = 1), death during hospitalization (n = 21), transfer to another hospital (n = 13), and poor general condition that might have caused difficulties in measurements (n = 31). Finally, 108 patients were included and divided into the non-readmission group (n = 76) and readmission group (n = 32). Table 1 shows the clinical characteristics of all patients and the results of the comparison between the groups.
The mean age of the patients was 85.5 ± 7.2 years, and 50% were female. Altogether, 34 patients were over 90 years of age. There were no significant differences in age, sex, BMI, and time from admission to the start of rehabilitation between the readmission and non-readmission groups. Patients in the readmission group had a significantly longer hospital stay compared to those in the non-readmission group. All patients enrolled in this study were classified into NYHA classes II and III. The readmission group exhibited significantly higher severity of NYHA class than the nonreadmission group.
Significant differences were observed in the CCI scores between the groups. However, the incidence of comorbidities such as hypertension, diabetes, atrial fibrillation, valvular disease, and ischemic heart disease was not significantly different between the groups. No significant differences were observed in the use of beta-blockers, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, or diuretics between the readmission and non-readmission groups.
No significant differences were observed in the LVEF, Cr level, BUN level, or eGFR. However, BNP was significantly higher in the readmission group than in the non-readmission group. No significant differences were observed in the Hb level, Alb level, or GNRI between the groups.
The maximal QIS, grip strength, and SPPB were significantly lower in the readmission group than in the nonreadmission group. Table 2 shows the results of the logistic regression analysis. The NYHA class, BNP level, maximal QIS, grip strength, and SPPB score were selected as explanatory variables and analyzed using stepwise selection. BNP level and SPPB score were identified as independent factors associated with readmission within the past year.

Discussion
In the present study, we examined the factors associated with readmission of patients with HF within the past year in Japan, which is becoming a super-aging society. Reportedly, ADLs are associated with readmission in Japan, and the presence or absence of sarcopenia affects the prognosis [22][23][24][25] . The readmission group in the present study exhibited higher NYHA class, more comorbidities, higher BNP levels, and lower physical function and physical performance than the non-readmission group. Furthermore, BNP level and SPPB score were identified as independent risk factors for readmission in the multivariate analysis. The mean age of patients in the present study was 85.5 ± 7.2 years, which was higher than that reported in previous studies [22][23][24] . In Japan, no definite view has been obtained about the effect of cardiac rehabilitation on older patients with HF, and the influence of physical performance of older patients with HF on the risk of readmission within the past year is unknown. The main finding of the present study was that physical performance affected the risk of readmission within the past year in older patients with HF having an average age of 85 years or above, suggesting the importance of cardiac rehabilitation in these patients.
BNP is a cardiac hormone synthesized in the ventricles and a biomarker that sensitively reflects the load on the Values are presented as mean ± standard deviation or number (percentage).
BMI, body mass index; NYHA, New York Heart Association functional classification; CCI, Charlson comorbidity index; HT, high blood pressure; DM, diabetes mellitus; Af, atrial fibrillation; ACE-I, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; LVEF, left ventricular ejection fraction; BNP, brain natriuretic peptide; Cr, creatinine; BUN, blood urea nitrogen; eGFR, estimated glomerular filtration rate; Hb, hemoglobin; Alb, albumin; GNRI, Geriatric Nutritional Risk Index; QIS, quadriceps isometric strength; SPPB, Short Physical Performance Battery ventricles [26][27][28][29] . Previous studies have reported that a high BNP level at discharge is associated with higher readmission rates and mortality 30) . BNP is secreted by the myocardium in response to wall stress, and it is affected by various factors. For example, obesity lowers the BNP levels, whereas aging, atrial fibrillation, and worsening renal function increase the BNP levels [31][32][33][34] . There was no significant difference in BMI and renal function between the readmission non-readmission groups in the present study. In addition, BNP level was found to be an independent risk factor for readmission in the multivariate analysis. However, the odds ratio was 1.000 (95% confidence interval: 1.000-1.000), and further research is needed to determine how BNP affects readmission in patients with HF. Patients with HF having low physical performance are more likely to be readmitted due to the increased cardiac load required for daily activities after discharge. Additionally, patients with HF may have decreased physical performance after discharge and may be readmitted due to events such as cardiovascular diseases, respiratory diseases, exacerbation of orthopedic diseases, and falls 22) . Previous studies have reported a relationship between readmission and ADLs 22) , which is influenced by physical performance (such as that determined using the SPPB score) and may result in a decline in ADLs due to inactivity caused by decreased physical capacity. A previous study reported that the SPPB score is a useful predictor of future decline in ADLs among older adults 17) . Therefore, the SPPB score at discharge is considered an important factor for predicting the risk of readmission. Long-term inactivity during hospitalization makes daily life after discharge difficult due to deterioration of physical function including muscle strength and exercise tolerance and concomitant deterioration of ADLs. Therefore, it is important to improve physical function, physical performance, and ADLs by introducing cardiac rehabilitation during acute treatment of patients to reduce the risk of readmission.
In Japan, people aged ≥90 years are defined as "superelderly" 35) . The present study included 34 patients aged ≥90 years, accounting for more than 30% of all patients. Our data may be important for preventing readmission among older adults with HF. Reports of early readmission (90-180 days) after discharge have increased over the past few years 22,24,36) , but very few studies have followed patients for longer periods. It is clear that Japanese patients with HF are at a high risk of readmission within 1 year 13) . The readmission rates in previous studies were 5.4% within 30 days, approximately 20% within 90 days, approximately 24% within 180 days, and 26%-35% within 1 year; and the risk of readmission increased with increasing observation period 10,11,22,36) . Therefore, continued follow-up after preventing early readmission is important. Moreover, results of the present study suggest the importance of cardiac rehabilitation to maintain physical performance. However, in Japan, there is insufficient evidence on the effectiveness of cardiac rehabilitation in older patients with HF, and very few studies have investigated the relationship between physical performance and readmission. Thus, further research is required on this topic.

Limitations
This study has some limitations. It was a single-center study with a small sample size. To avoid sampling bias, we randomly selected participants from both the groups. Based on the inclusion and exclusion criteria, only 33.2% of the patients with HF who were hospitalized for acute exacerbations were included in this study. Therefore, the results of this study may not be generalizable to other patients with severe HF. A detailed evaluation of frailty and body composition could not be conducted. Frailty and sarcopenia have been shown to affect the prognosis of patients with HF 37,38) . Therefore, detailed assessment of frailty and sarcopenia may be necessary. The present study did not evaluate left ventricular diastolic function. The effects of cardiac rehabilitation interventions have not been investigated. Determining the effects of cardiac rehabilitation is an important issue in Japan, where the average age of the population is expected to increase in the future. Studies that consider the interventional effect of cardiac rehabilitation are required to prevent readmission.

Conclusion
Patients who were readmitted due to exacerbation of HF within the past year exhibited longer hospital stays, multiple comorbidities, higher NYHA class, higher BNP levels, and lower physical performance than those from the nonreadmission group. In addition, the multivariate analysis showed that BNP level and SPPB score were associated with readmission for HF.