Association of muscular fitness with rehospitalization for heart failure with reduced ejection fraction

Limited information is available regarding the prognostic potential of muscular fitness parameters in heart failure (HF) with reduced ejection fraction (HFrEF).


| INTRODUCTION
Heart failure (HF) is a chronic debilitating disease associated with a 50% mortality rate within 5 years of diagnosis. 1 Despite advancements in treatment strategies, HF still poses a significant threat to patient outcomes. 2 Patients with HF often complain of fatigue and shortness of breath even with low-intensity physical activity, which causes detrimental effects to their quality of life. 3,4 Exercise intolerance is a consequence of HF and a major determinant of its prognosis. 5,6 Measurement of peak oxygen uptake (peak VO 2 ) during exercise is an objective method for assessing functional capacity and is an important predictor of long-term prognosis in patients with HF. 7 Peak VO 2 is used for determining the timing of heart transplantation. 8 The loss of muscle mass and strength occurs progressively with aging. 9 Irrespective of aging, chronic diseases accelerate the atrophy of muscle fibers or lower the efficiency of energy production in the muscles, leading to low levels of muscular fitness, which is also associated with poor prognosis. 10,11 In patients with HF, muscular strength predict long-term survival. 6 However, the effect of muscle fitness on rehospitalization has not been fully investigated. Rehospitalization is an important outcome for patients with HFrEF, as it relates to quality of life and high financial burdens for the community. 12 Muscular strength is one of the most common indicators of muscular fitness.
However, muscle performance can also be measured as muscle power (MP), which indicates the ability of the muscle to perform forceful and high-velocity movements. Furthermore, as the measurement of MP is technically simple and relatively easy to assess, it could be incorporated into a standard prognostic assessment for patients with HFrEF.
Nevertheless, indices of muscular fitness are rarely used in real-world clinical practice, unlike peak VO 2 .
Therefore, we aimed to do the following: (a) investigate the predictive potential of knee extensor muscle strength and power on rehospitalization; and (b) evaluate the correlation between exercise capacity and muscular fitness in patients newly diagnosed with HFrEF.

| Study participants
We recruited consecutively a total of 99 patients, who were hospitalized with a new diagnosis of HF from January 2013 to November 2015, received subsequent inpatient treatment, and were discharged.
The inclusion criteria were as follows: left ventricular ejection fraction <40% and sufficient clinical stability to undergo exercise testing.
Patients who underwent surgical procedures such as coronary artery bypass, valve replacement, heart transplantation, and those on renal replacement therapy were excluded. Aerobic exercise capacity and muscular fitness were examined immediately before discharge. We investigated the medications prescribed at discharge and obtained blood chemistry data at the time of the 1-2 week follow-up visit. Definition of heart failure rehospitalization was a hospitalization caused by worsening heart failure symptoms and signs requiring the augmentation of previous medications. 13 We identified cases of heart failure rehospitalization by chart reviewing. All patients provided written informed consent at enrollment, and the Ethics Committee of Sever-

| Assessment of muscular fitness
The assessments of two muscular fitness parameters were performed using the Primus RS, version 11 (Baltimore Therapeutic Equipment

| Statistical analysis
Data were expressed as mean ± standard deviation (SD) frequency (%), or median (interquartile range [IQR]). Patients were grouped based on whether they required HF rehospitalization. For group comparison of continuous variables, the Student's t-test, one-way analysis of variance, the Mann-Whitney U test, or the Kruskal-Wallis test was used. Categorical variables were evaluated using chi-squared test or Fisher's exact test. Correlation analysis between aerobic exercise capacity and muscular fitness was performed using Pearson's correlation coefficient. The effect of aerobic exercise capacity or muscular fitness on HF rehospitalization was analyzed using Kaplan-Meier curves. Receiver operating characteristic (ROC) analysis was performed to identify the best cut-off value of peak VO 2 , MVIC, and MP for HF rehospitalization. The association of the tertile of aerobic exercise capacity or muscular fitness with HF rehospitalization was evaluated using a multivariable Cox proportional hazard model with adjustment for age, sex, body mass index, diabetes mellitus, and left ventricular ejection fraction (LVEF), and N-terminal pro-B-type natriuretic peptide (NT-proBNP). Statistical analyses were conducted using R software, version 3.5.3 (R Foundation for Statistical Computing, Vienna, Austria), assuming a threshold of significance at P < 0.05. had lower LVEF and a higher rate of diabetes than those without HF rehospitalization. Regarding HF guideline-directed medications at baseline, the use of beta-blockers was significantly lower in patients with HF rehospitalization than in those without HF rehospitalization.

| Baseline characteristics of patients
The use of other medications was not significantly different between the two groups. Table 2 shows the results of CPET and muscular fitness tests.
Patients with rehospitalization had significantly lower heart rates at rest and maximal exercise in CPET. The systolic blood pressure did not differ between the two groups at rest; however, it was lower during maximal exercise in patients with rehospitalization. Moreover, the exercise time and peak VO 2 were significantly lower in patients with

| Muscular fitness as a predictor for rehospitalization
Supplemental Table 1 shows the incidence of HF rehospitalization during the follow-up period according to the peak VO 2 , MVIC, and MP tertiles. The incidence of HF rehospitalization was significantly different according to the peak VO 2 tertile. Patients were divided into three groups (low, middle, and high) according to the tertiles of peak VO 2 , MVIC, and MP. Patients with lower aerobic exercise capacity had more HF rehospitalizations during follow-up ( Figure 2). Patients with lower MVIC tended to present with HF rehospitalization more frequently; however, the difference was not statistically significant (Supplemental Table 1), as was the case with the Kaplan-Meier survival curve (Figure 2). Nevertheless, patients with low MP had significantly more HF rehospitalizations than the other groups (Supplemental Table 1). The Kaplan-Meier curve showed that patients with low MP were rehospitalized for HF at early periods of follow-up ( Figure 2). ROC curves showed that the best cut-off values of peak VO 2 , MVIC, and MP for HF rehospitalization was 20.1 ml/kg, 320 N, and 87 Watt, respectively (Supplemental Figure 2). The best cut-off value of peak VO 2 was similar to the high tertile value, while the best cut-off values of MVIC and MP were similar to the low tertile values.
In the multivariable Cox proportional hazard model (

| DISCUSSION
In this study, we found that both MP and MVIC were lower in patients with HF rehospitalization than in those without rehospitalization. Both these parameters also exhibited a significant correlation with the peak VO 2 . However, low MP (and not MVIC) was significantly associated with the risk of rehospitalization for HF in patients with HFrEF.
HF rehospitalization incurs high medical costs, putting a great burden on patients, the health care system, and the social economy.
In addition, it may contribute to the long-term progression of HF and LV dysfunction. 14 In this study, DM and LVEF are the most significant factors for predicting HF rehospitalization. It is known that HF patients with DM have a higher risk of HF rehospitalization and a worse prognosis than those without DM. 15 Also, low LVEF is closely associated with cardiovascular outcomes in HF patients. 16 Despite adjustment of important prognostic factors such as diabetes and LVEF, our study showed that low aerobic exercise capacity and low muscular fitness are significant prognostic factors related to HF rehospitalization.
Peak VO 2 is the most objective indicator of physical fitness that represents the use of oxygen in the cardiac, circulatory, and respiratory systems and muscles. 17 CPET is recommended in the 2016 European HF Guidelines for identifying the cause of unexplained dyspnea or for determining the treatment policies. 2 CPET can provide a more global assessment of patients with HF. CPET parameters are valid prognostic factors for HF, especially peak VO 2 is closely related to the long-term prognosis of HF. 7 In our analysis, among the CPET parameters, peak VO 2 was most closely associated with to HF rehospitalization (Supplemental Table 2). Our study showed that muscular fitness is also a major predictor of HF prognosis. The isokinetic strength test of the knee flexor muscle showed that the strength index was significantly associated with mortality, although peak VO 2 was adjusted in multivariable analysis for patients with advanced HF. 6 Despite the clear association between muscular fitness and long-term outcomes of HF, muscular fitness is used less frequently than peak VO 2 for assessing the patients' condition in the clinical settings.
Consistent poor exercise tolerance was observed in the patients with HF and hypothesized to be due to the pathophysiological changes in the skeletal muscle. 18 Histologic studies on the skeletal In our study, muscular fitness and peak VO 2 were significantly correlated, and patients with low MP, probably those with low peak VO 2 , had poor prognosis for HF. These findings suggested that the association of MP with the prognosis of HF is comparable to that of peak VO 2 . In addition, muscular fitness variables in combination with peak VO 2 can more reliably predict hospitalization for heart failure. The prognostic ability of the combination of peak VO 2 and muscular fitness variable was analyzed using the best cut-off values obtained using ROC analysis. When the peak VO 2 was low and the MP or MVIC was low, the risk of re-hospitalization due to HF was 9-20 times higher than that of the high peak VO 2 and high MP or MVIC (data not shown). However, since the number of subjects is small, further research is needed.
As it is generally recommended to measure peak VO 2 rather than muscular fitness in evaluating the physical fitness of patients with HF, exercise training is primarily focused on aerobic endurance exercise. 2 Resistance training is a form of exercise that contracts the muscles against opposing forces that create resistance, overloading the musculoskeletal system to prevent muscle loss and improve the muscle strength. 23 Although resistance training has survival benefits in patients with HF, increased afterload during the lifting phase in resistance training may adversely affect the LV function and cause negative remodeling. 24  Abbreviations: BMI, body mass index; CI, confidence interval; DM, diabetes mellitus; HF, heart failure; HR, hazard ratio; LVEF, left ventricular ejection fraction; MP, muscle power; MVIC, maximum voluntary isomeric contraction; NT-proBNP. N-terminal pro b-type natriuretic peptide; VO 2 , oxygen uptake. muscular fitness. Therefore, there was a high possibility that patients with severe muscle loss were not included, and it might be difficult to generalize the results of this study to all patients with severe advanced HF. Nevertheless, measuring MP will not be difficult in patients with disability because its feasibility has been demonstrated in older populations with sarcopenia. Second, we did not demonstrate that MP was related to the prognosis of HF independently of peak VO 2 . This is because peak VO 2 and MP were highly correlated and subjects with low peak VO 2 or low muscular fitness shared clinical features such as old age, female, and high NT-proBNP levels (Supplemental Table 3-7). However, the fact that MP was significantly associated with the prognosis of HF as much as peak VO 2 suggests that measuring MP in patients is an alternative to measuring physical fitness. Third, the medications used after discharge were not reflected in the research results. In addition, lower prescription of beta-blockers at discharge in patients with low MP may have had an impact on long-term prognosis. Fourth, the number of subjects who participated in the study was relatively small.
In conclusion, aerobic exercise capacity and muscle fitness were associated with the prognosis of patients with HFrEF. Compared with MVIC, which is the traditional method of measuring knee extensor fitness, measurement of MP was found to be a better predictor for HF rehospitalization in these patients.