Lumbar-Type Hybrid Assistive Limb Improves Short Physical Performance Battery for Aged Patients With Decompensated Heart Failure in Cardiac Rehabilitation

Abstract

Background: Unlike the outpatient cardiac rehabilitation (CR) program for chronic heart failure (HF), the acute-phase inpatient CR program is not well established. We aimed to examine whether CR using lumbar-type hybrid assistive limb (HAL; lumbar HAL-CR) therapy improves physical functions in hospitalized aged HF patients.

Methods and Results: Decompensated and hospitalized HF patients were recruited from 2018 to 2019. As soon as possible after the decompensated HF phase, during the same hospitalization, participants underwent 40-min daily sessions of sit-to-stand and squat exercises with the lumbar-type HAL (lumbar HAL-CR program). The outcome measure was the score on the Short Physical Performance Battery (SPPB) before and after lumbar HAL-CR therapy. Twenty-eight HF patients (median age 84.5 years; 18 males) were enrolled. The median number of lumbar HAL-CR therapy sessions was 5.0. The SPPB score (7.0 [5.3–8.8] to 9.0 [7.0–11.0] points; P<0.001) and quadriceps isometric strength (0.25 [0.22–0.38] to 0.30 [0.26–0.37] kgf/kg; P=0.040) were significantly improved. Patients with lower nutrition status had more improvement in SPPB score.

Conclusions: For the inpatient CR program, lumbar-type HAL therapy improved physical functions in elderly HF patients. Lumbar-type HAL therapy may improve physical functions in aged HF patients within the current short-term period of hospitalization for acute-phase HF.

Central Figure

Elderly hospitalized heart failure (HF) patients with frailty are a burden worldwide.^1–3 The number of those patients will continue to increase until 2040 in Japan.⁴ Cardiac rehabilitation (CR) is currently used to prevent a decrease in physical function during hospitalization for heart disease. The number of inpatients receiving CR for heart disease has increased 3.78 times within 10 years in Japan.⁵ However, unlike well-established outpatient CR programs, appropriate inpatient CR programs have yet to be defined.^3,6

Moreover, the mean HF hospitalization period is shorter than before.⁷ Patients often do not get enough CR sessions to improve their physical function during hospitalization. Previous reports indicate that older patients with heart disease need more than 10 days to improve their scores on the Short Physical Performance Battery (SPPB) or 6-min working distance (6MWD) test, which is a meaningful outcome parameter.^8–13 However, inpatient CR programs are needed to shorten the current clinical practice period. Thus, a short-term and effective inpatient CR program is desired.

Lumbar-type hybrid assistive limb (HAL) therapy is unique in that it detects bioelectric signals from the skin surface, which reflect voluntary motor intention, and assists muscle movement accordingly.^14,15 It facilitates the performance of stand-up exercise.¹⁶ In a previous report, 7.2 days of lumbar-type HAL and CR improved quadriceps isometric strength (QIS).¹⁷

We hypothesized that lumbar HAL and CR could augment the effects of exercise-based CR in elderly HF patients during the current short periods of clinical hospitalization. The present study aims to evaluate whether 5 sessions of lumbar HAL and CR therapy can improve physical function in elderly hospitalized HF patients.

Methods

Study Design

We included HF patients who were hospitalized in National Cerebral and Cardiovascular Center for acute decompensated or acute-on-chronic decompensated HF and underwent in-hospital CR using lumbar HAL (lumbar HAL-CR) between October 2018 and December 2019.

The rehabilitation was initiated in the ward as early as possible. Moreover, within a few days after admission, during the early recovery phase, patients who met the eligibility criteria for lumbar HAL-CR therapy were enrolled in this study.

The patients who were included in the present study met the following eligibility criteria: (1) certification of the need for care or support under the long-term care insurance system in Japan, specifically classified as support level or care level 1–3; (2) a sedentary lifestyle, defined as spending ≥6 h/day in sedentary behavior during waking hours before admission due to reduced physical activity; and (3) difficulty in standing up without assistance after admission.

Patients were excluded if they were: (1) bedridden before admission; (2) contraindicated to receive exercise therapy based on current guidelines;¹⁸ (3) on intravenous inotropic support at the start of CR; (4) scheduled for discharge within a week; (5) receiving hemodialysis; (6) deemed ineligible by the attending physician; or (7) denied admission to the CR program.

The present study followed the Declaration of Helsinki and Ethical Guidelines for Medical and Biological Research Involving Human Subjects in Japan. We conducted a single-center, retrospective observational study. Informed consent was obtained using an opt-out method on National Cerebral and Cardiovascular Center’s website or a notice on a bulletin board in the stroke rehabilitation room at the National Cerebral and Cardiovascular Center (approval no. M30-090-5).

Lumbar-Type HAL System

HAL is an exoskeletal, cyborg-type robot suit that fuses human capabilities with robotic and information system capabilities.¹⁹ This system is an exoskeletal cyborg that supports sit-to-stand movements, assisting patients in standing up from a seated position, and effectively supporting lower-limb motor function. The components of a lumbar-type HAL (HAL-FB02-SSSJP; Cyberdyne Inc., Tsukuba, Japan) are as follows: an exoskeletal frame, power units, a battery, lumbar and thigh support belts, and electrodes with bioelectrical signal sensors (Figure 1).²⁰ An assistive intensity can be set according to one of 5 levels based on the patient’s condition (i.e., the higher the level, the greater the assistive intensity).

Figure 1.

Components of a lumbar-type hybrid assistive limb.

Our Lumbar HAL-CR Program

The CR program using the lumbar-type HAL was conducted for 40 min daily, 5 times a week. Before initiating the program, the level of assistive intensity was selected to ensure that the patient could exercise easily. All patients started with 3 rounds of 10 sets of squat exercises with the lumbar-type HAL while holding horizontal support bars, with 1 min of rest between each round (Figure 2A). Subsequently, following the squat exercises, 3 rounds of 10 sets of sit-to-stand exercises with the lumbar-type HAL were conducted in a similar way (Figure 2B). Patients removed the device after completing the squat and sit-to-stand exercises. They then walked 50–100 m on the track at a comfortable speed, taking intermittent breaks as needed, for 15–20 min under the supervision of a physical therapist.

Figure 2.

Cardiac rehabilitation program using a lumbar-type hybrid assistive limb. In the squat exercise (A), patients start in a standing position, lower their body without touching the seat, and then return to the standing position with assistive support. In the sit-to-stand exercise (B), patients begin in the seated position, and the device assists them as they stand up.

Baseline Characteristics

At the start of the CR program, we collected baseline characteristics from medical charts, including sex, age, body mass index (BMI), days from admission to the start of CR, left ventricular ejection fraction (LVEF) as evaluated by echocardiography, New York Heart Association (NYHA) classification on admission, and laboratory data such as B-type natriuretic peptide (BNP) and creatinine levels. We also reviewed the medication profiles at the start of the CR. The nutritional status on admission was assessed using the Controlling Nutritional Status (CONUT) score, which ranges from 0 to 12, with higher scores indicating worse nutritional status.²¹

Outcome Measures

The SPPB was measured, and included a balance test (side-by-side, semi-tandem, and tandem), usual walking speed over a 4-m distance, and a 5-chair-stands test (total score ranging from 0 to 12, with higher scores indicating better physical function).²²

QIS was measured using a handheld dynamometer (μTas F-1; ANIMA, Tokyo, Japan). For examination of the QIS, patients sat upright on a bench with knee joints fixed at 90° flexion, and strength was measured twice on each side using a dynamometer. The average value (kgf/kg) was recorded.²³

These outcome measures were examined in the CR room at the beginning of the early recovery phase (baseline) and after the completion of CR at discharge. The changes in SPPB score and QIS were calculated between the baseline and discharge assessments (∆SPPB score and %∆QIS).

Furthermore, we collected exercise-related adverse events from medical charts, which included skin troubles associated with wearing the lumbar-type HAL, falls, deterioration of NYHA classification, and the occurrence of cardiovascular (CV) events until the date of discharge.

Statistical Analysis

The χ² test and Wilcoxon rank-sum test were used as appropriate to determine differences in patient characteristics between the 2 groups. The Wilcoxon signed-rank test was used to compare the variables in each group between baseline and discharge. We also performed receiver operating characteristic (ROC) curve analysis to determine the optimal cut-off values for predicting significant improvement of the SPPB score. The area under the curve (AUC) with 95% confidence intervals (CI) was calculated, and sensitivity and specificity were assessed. The optimal cut-off values were determined using the Youden index. Additionally, we used multiple linear regression models to determine the factors leading to improved SPPB score with the use of lumbar HAL-CR therapy.

All statistical analyses were performed using JMP® version 14.0 (SAS Institute Inc., Cary, NC, USA). Measurements are presented as median [interquartile range] or number (percentage), and a P value of <0.05 was considered statistically significant.

Based on a previous study,¹³ we estimated the required sample size using the G*Power 3.1 program (Heinrich-Heine-Universität, Düsseldorf, Germany). The calculation was performed with the following parameters: effect size=0.56; α error probability=0.05; and power (1−β error probability)=0.8. As a result, a sample size of 28 participants was deemed necessary.

Results

Baseline Characteristics and SPPB and QIS Outcomes

Of 288 consecutive HF patients, 28 were included in the present study. The median age was 84.5 years, and 64.3% were male. The median SPPB score was 7.0 [5.3–8.8]. The lumbar HAL-CR was performed in a median of 5.0 [3.0–6.8] sessions. After lumbar HAL-CR, the SPPB score and QIS improved significantly (SPPB score: 7.0 [5.3–8.8] to 9.0 [7.0–11.0]; P<0.001; QIS: 0.25 [0.22–0.38] to 0.30 [0.26–0.37]; P=0.040; Figure 3).

Figure 3.

Comparison of SPPB score and QIS at baseline and discharge following HAL-CR. Box plots illustrate the changes in SPPB score (A) and QIS (B) from baseline to discharge following CR using the lumbar-type HAL (HAL-CR). Both SPPB score and QIS significantly improved after HAL-CR (SPPB score: 7.0 [5.3–8.8] to 9.0 [7.0–11.0]; P<0.001; QIS: 0.25 [0.22–0.38] to 0.30 [0.26–0.37]; P=0.040). CR, cardiac rehabilitation; HAL, hybrid assistive limb; QIS, quadriceps isometric strength; SPPB, Short Physical Performance Battery.

All patients completed the CR program without experiencing any exercise-related adverse events.

Factors of SPPB Score Improvement (∆SPPB Score ≥2)

The median ∆SPPB score in this study was 2 points, and the patients were divided into 2 groups: Improved group, defined as those with a ∆SPPB score ≥2 (n=14); and Unchanged group, defined as those with a ∆SPPB score <2 (n=14).

Patients in the Improved group were younger (age 82.5 [79.5–85.0] vs. 87.0 [83.0–89.0] years; P=0.034) and had worse nutrition status (CONUT score [5.5 (4.8–7.3) vs. 3.5 (3.0–4.3)]; P=0.005; Table 1).

Table 1.

Clinical Characteristics of the Study Patients

	Overall (n=28)	Improved group (n=14)	Unchanged group (n=14)	P value
Age (years)	84.5 [81.3–88.0]	82.5 [79.5–85.0]	87.0 [83.0–89.0]	0.034
Male	18 (64.3)	11 (78.6)	7 (50.0)	0.115
BMI (kg/m2)	20.5 [18.9–22.8]	20.2 [19.0–22.5]	20.9 [18.2–23.4]	0.696
Admission to initiation of CR (days)	5.5 [4.0–12.3]	6.0 [4.0–16.5]	5.0 [3.0–8.0]	0.139
CR (no. sessions)	9.5 [5.3–14.8]	12.0 [5.0–16.3]	9.0 [5.8–13.3]	0.610
CR using lumbar HAL (no. sessions)	5.0 [3.0–6.8]	5.0 [3.0–11.0]	5.0 [3.0–5.3]	0.361
LVEF (%)	47.5 [30.0–58.5]	50.0 [28.8–57.5]	44.0 [30.0–60.0]	0.628
NYHA classification ≥III	12 (42.9)	7 (50.0)	5 (35.7)	0.445
Etiology of HF
CAD	12 (42.9)	5 (35.7)	7 (50.0)	0.445
VHD	9 (32.1)	4 (28.6)	5 (35.7)	0.686
DCM/HCM	1 (3.6)	0 (0)	1 (7.1)	0.309
Other cardiac disease	6 (21.4)	5 (35.7)	1 (7.1)	0.065
BNP (pg/mL)	288.4 [189.3–421.3]	253.0 [107.4–404.9]	365.8 [235.4–709.4]	0.085
Creatinine (mg/dL)	1.3 [1.0–1.6]	1.3 [1.0–1.7]	1.3 [1.0–1.7]	0.783
CONUT score	4.5 [3.0–6.0]	5.5 [4.8–7.3]	3.5 [3.0–4.3]	0.005
Medications
β-blocker	19 (67.9)	11 (78.6)	8 (57.1)	0.225
ACE inhibitor/ARB	21 (75.0)	11 (78.6)	10 (71.4)	0.663
MRA	13 (46.4)	6 (42.9)	7 (50.0)	0.705
SGLT2I	1 (3.6)	1 (7.1)	0 (0)	0.309
SPPB score (points)
Baseline	7.0 [5.3–8.8]	6.0 [3.8–7.3]	8.0 [7.0–10.3]	0.007
Discharge	9.0 [7.0–11.0]**	9.5 [6.8–10.3]**	9.0 [6.5–11.0]	0.982
ΔSPPB score	1.5 [1.0–2.8]	2.5 [2.0–4.3]	1.0 [0.0–1.0]	<0.001
QIS
Baseline (kgf/kg)	0.25 [0.22–0.38]	0.27 [0.20–0.37]	0.25 [0.23–0.38]	0.927
Discharge (kgf/kg)	0.30 [0.26–0.37]*	0.29 [0.24–0.39]	0.31 [0.26–0.35]	1.000
%ΔQIS (%)	7.60 [−4.19, 30.38]	8.07 [−8.44, 32.15]	7.11 [0.58–21.97]	0.982

Data are presented as median [interquartile range] or n (%). The improved group of ΔSPPB score was defined as ≥2 (n=14), and the unchanged group of ΔSPPB was <2 (n=14). ACE, angiotensin-converting enzyme; ARB, angiotensin II receptor blocker; BMI, body mass index; BNP, B-type natriuretic peptide; CAD, coronary artery disease; CONUT, Controlling Nutritional Status; CR, cardiac rehabilitation; DCM/HCM, dilated cardiomyopathy/hypertrophic cardiomyopathy; HAL, hybrid assistive limb; HF, heart failure; LVEF, left ventricular ejection fraction; MRA, mineralocorticoid receptor antagonist; NYHA, New York Heart Association; QIS, quadriceps isometric strength; SGLT2I, sodium glucose co-transporter-2 inhibitor; SPPB, Short Physical Performance Battery; VHD, valvular heart disease. *P<0.05 vs. baseline. **P<0.001 vs. baseline.

Using ROC curve analysis, the optimal cut-off value to predict improvement in SPPB (∆SPPB ≥2) was identified as age 85 years (AUC of 0.737; 95% CI [−0.405, 0.015]; P=0.029). However, in comparing the 2 groups divided by this cut-off value of 85 years, no significant differences were observed in ∆SPPB score or %∆QIS.

Similarly, ROC curve analysis for nutritional status revealed that a CONUT score cut-off of 5.0 significantly predicted improvement in SPPB (∆SPPB ≥2; AUC=0.808; 95% CI [0.016, 1.151]; P=0.010; sensitivity 78.6%, specificity 57.1%).

Compared with the low CONUT group (CONUT score <5), the high CONUT group (CONUT score ≥5) demonstrated significantly greater ∆SPPB score (2.5 [1.8–4.3] vs. 1.0 [0.0–1.3]; P=0.002) and significantly greater %∆QIS (23.7 [5.2–35.2] vs. 2.73 [−11.7, 9.7]; P=0.023; Figure 4).

Figure 4.

Comparison of ∆SPPB score and %∆QIS between low and high Controlling Nutritional Status (CONUT) groups. Box plots of ∆SPPB score (A) and %∆QIS (B) in the low and high CONUT groups. ∆SPPB score was greater in the high CONUT group than in the low CONUT group (2.5 [1.8–4.3] vs. 1.0 [0.0–1.3]; P=0.002). Percentage of ∆QIS was also greater in the high CONUT group than in the low CONUT group (23.7 [5.2–35.2] vs. 2.73 [−11.7, 9.7]; P=0.023).

Multiple Regression Analysis for ∆SPPB Score

Although there were no differences in age, sex, BMI, or LVEF between the 2 groups, the high CONUT group had longer durations from admission to the initiation of CR (9.0 [5.0–16.5] vs. 4.0 [3.0–6.3] days; P=0.008), lower BNP levels (230.6 [107.4–378.1] vs. 384.1 [261.3–709.4] pg/mL; P=0.037), lower SPPB score at baseline (6.0 [3.8–7.3] vs. 8.5 [7.0–10.3]; P=0.002), and lower QIS at baseline (0.24 [0.20–0.28] vs. 0.32 [0.24–0.43] kgf/kg; P=0.026) than those in the low CONUT group, respectively (Table 2).

Table 2.

Low and High CONUT Groups

	Low CONUT group (n=14)	High CONUT group (n=14)	P value
Age (years)	84.5 [82.3–89.0]	84.5 [80.8–87.3]	0.565
Male	7 (50.0)	11 (78.6)	0.115
BMI (kg/m2)	20.7 [19.0–23.5]	20.4 [17.9–22.5]	0.370
Admission to initiation of CR (days)	4.0 [3.0–6.3]	9.0 [5.0–16.5]	0.008
CR (no. sessions)	8.0 [5.0–13.3]	12.0 [5.8–16.3]	0.296
CR using lumbar HAL (no. sessions)	5.0 [3.0–5.3]	5.0 [3.0–11.0]	0.639
LVEF (%)	46.5 [30.0–59.3]	47.5 [30.8–57.8]	0.982
NYHA classification ≥III	5 (38.5)	6 (42.9)	0.816
Etiology of HF
CAD	6 (42.9)	7 (50.0)	0.705
VHD	6 (42.9)	3 (21.4)	0.225
DCM/HCM	1 (7.1)	0 (0)	0.309
Other cardiac diseases	1 (7.1)	4 (28.6)	0.139
BNP (pg/mL)	384.1 [261.3–709.4]	230.6 [107.4–378.1]	0.037
Creatinine (mg/dL)	1.1 [0.9–1.6]	1.4 [1.1–1.8]	0.232
CONUT score	3.0 [3.0–4.0]	6.0 [5.0–8.0]	<0.001
Medications
β-blocker	9 (64.3)	10 (71.4)	0.686
ACE inhibitor/ARB	10 (71.4)	11 (78.6)	0.663
MRA	7 (50.0)	6 (42.9)	0.705
SGLT2I	0 (0)	1 (7.1)	0.309
Baseline SPPB score (points)	8.5 [7.0–10.3]	6.0 [3.8–7.3]	0.002
Baseline QIS (kgf/kg)	0.32 [0.24–0.43]	0.24 [0.20–0.28]	0.026

Data are represented as median [interquartile range] or n (%). Abbreviations as in Table 1.

In the multiple linear regression analysis, Model 1 (admission to the initiation of CR and BNP) and Model 2 (baseline SPPB score) showed that a high CONUT score (CONUT score ≥5) was independent of SPPB improvement (∆SPPB score ≥2; Table 3).

Table 3.

Results of the Multiple Linear Regression Analysis

Model / Variable	Adjusted B	95% CI	P value
Model 1
High CONUT group	0.44	0.074, 1.582	0.033
Admission day to the initiation of the CR	0.03	−0.096, 0.113	0.865
BNP	−0.22	−0.003, 0.001	0.225
Model 2
High CONUT group	0.47	0.107, 1.647	0.027
Baseline SPPB score	−0.12	−0.362, 0.198	0.553

CI, confidence interval. Other abbreviations as in Table 1.

Discussion

We reported that the 5-session lumbar HAL-CR therapy improved SPPB (SPPB score: 7.0 [5.3–8.8] to 9.0 [7.0–11.0]) and QIS in elderly patients with acute phase HF. SPPB was more improved in HF patients with low nutrition status.

Our patients were aged 84.5 years and hospitalized in the acute phase of HF. Their physical function status was low (SPPB score7.0 [5.3, 8.8]). Kitzman reported that older frail patients exhibited improved SPPB after 3 months of outpatient CR (age 70 years, SPPB score 6.0, 36 sessions).⁸ Regarding the inpatient CR program, 2 weeks of CR improved 6MWD in patients (aged 80 years) with transcatheter aortic valve replacement.⁹ In the Early Rehabilitation in Cardiology (ERIC) study, 12 exercise sessions improved 6MWD in decompensated HF patients.¹⁰ Oliveira et al. also reported that 10 days of inpatient CR improved 6MWD.¹¹ The mean ages of these patients ranged from 58 to 68 years. In the PEARL study (a randomized controlled trial on early cardiac rehabilitation for acute decompensated heart failure), older HF patients improved their SPPB (age 80.1 years; SPPB score 5.9–9.6) and needed 10.7–11.7 CR days.¹² Rinaldo et al. reported that older heart disease patients improved their SPPB (age 81.1 years; SPPB score 6.9–8.4) and needed an average of 9.9 days.¹³ In general, it takes older heart disease patients more than 10 days to improve their SPPB score or 6MWD.

The patients in the present study were older compared with previous studies and showed improvements in SPPB scores after a relatively short course of inpatient CR. Lumbar HAL-CR may improve SPPB in older HF patients within 6 days. Moreover, we did not observe CV events during the lumbar HAL-CR program.

Many HF patients have low physical function and are hospitalized in Japan due to this age-related illness.^3,4 Over time, hospitalization periods for HF in Japan have been getting shorter, decreasing from 26 days in 2007 to 16 days in 2015,⁷ and 11.7 (7.8–15.6) days in 2023.²⁴ Generally, within the 11–12 day hospitalization period, it takes 5 days to treat oxygen demand and do cardiac catheterization or other examinations, and the remaining 3–5 days following a weekend holiday is the only time left for CR and improving the patient’s physical status (Figure 5). It has also been reported that Waon-CR therapy improved the 6MWD within 5 days in older HF patients.²⁵ The lumbar HAL-CR program, which improves SPPB score in a short period of time, is desirable in the present short treatment timeframe for HF and also for preventing hospital-associated disability.

Figure 5.

Conceptual schematic of current short-term and acute-phase heart failure (HF) hospitalization.

Furthermore, Kato et al. reported that HAL and CR improved SPPB score (from 6.8 to 8.6) and QIS (from 0.29 to 0.35) in patients with chronic HF.¹⁷ Although our patients were in the acute phase of HF, the improvement of SPPB score and QIS seemed similar to the improvements in previous HAL-CR reports.

The reason why the SPPB score improves more in HF patients with low nutritional status is still debatable. Older adults with poor nutritional status often exhibit reduced peripheral nerve conduction velocity due to vitamin B12 deficiency and related metabolic impairments.²⁶ The greater improvements in physical function observed in the high CONUT group may be partly attributed to enhanced neuromuscular adaptation facilitated by the lumbar-type HAL support. In malnourished older patients with HF, HAL-assisted exercise may have been particularly effective in reactivating neuromuscular pathways.

Furthermore, the greater functional improvements in patients with higher CONUT scores imply that poor nutritional status may represent a potential target population for HAL-assisted CR. These findings suggest that combining nutritional support with robot-assisted exercise therapy could yield synergistic effects, warranting further investigation in future studies.

Study Limitations

The present study has several limitations. First, we started with 20-min daily rehabilitation sessions in the yard for several days. We then provided lumbar HAL-CR therapy in the CR room for a median of 5.5 sessions, measuring SPPB scores before and after the intervention. Conducting rehabilitation in a yard may affect the SPPB values at baseline. However, since SPPB measurement is taken just before lumbar HAL-CR, this impact is low. Second, this was a single-center, retrospective pilot study with a small sample size, which may limit the generalizability of the results. Third, we did not include a control group without HAL intervention, so the independent effect of lumbar HAL-CR cannot be fully determined.

Conclusions

In this pilot study, CR using the lumbar-type HAL significantly improved physical function in older patients hospitalized for HF. Notably, patients with poor nutritional status showed greater improvement. These findings suggest that this robot-assisted rehabilitation approach may be particularly effective in enhancing physical performance in elderly patients with HF within the current limited clinical hospitalization timeframe.

Acknowledgment

We sincerely thank Dr. Yoichi Goto for his advice during our study.

Sources of Funding

This research was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture of Japan (JSPS KAKENHI grant no. 20K11250) and Cyberdyne Inc. Japan (#C677).

Disclosures

Y.S. is a University Professor at the University of Tsukuba, and the founder, shareholder, and CEO of CYBERDYNE Inc., a university-affiliated venture company based in Ibaraki, Japan. He holds equity in CYBERDYNE Inc., and the University of Tsukuba receives patent royalties from the company, a portion of which is distributed to Y.S. following the university’s conflict of interest management rules. The conflicts of interest have been fully disclosed and are reviewed by both the University of Tsukuba and CYBERDYNE Inc. in compliance with relevant institutional and national regulations. CYBERDYNE Inc. had no role in the study design, data collection, analysis, data interpretation, manuscript preparation, funding, or the decision to submit this article for publication. All relevant disclosures are documented in the participant consent forms. The other authors declare no conflicts of interest.

IRB Information

This study was approved by the Institutional Research and Ethics Committee (M30-090-5) of the National Cerebral and Cardiovascular Center.

Data Availability

The deidentified participant data will be shared upon reasonable request to the corresponding author. Data requests will be accepted for up to 36 months after the publication of this article. For any purpose, the data will be shared in Excel files via email.

References

• 1.   Mollar A, Bonanad C, Diez-Villanueva P, Segarra D, Civera J, Sastre C, et al. Frailty and hospitalization burden in patients with chronic heart failure. Am J Cardiol 2022; 183: 48–54, doi:10.1016/j.amjcard.2022.08.013.
• 2.   Lai HY, Huang ST, Anker SD, von Haehling S, Akishita M, Arai H, et al. The burden of frailty in heart failure: Prevalence, impacts on clinical outcomes and the role of heart failure medications. J Cachexia Sarcopenia Muscle 2024; 15: 660–670, doi:10.1002/jcsm.13412.
• 3.   Kitai T, Kohsaka S, Kato T, Kato E, Sato K, Teramoto K, et al. JCS/JHFS 2025 Guideline on diagnosis and treatment of heart failure. Circ J 2025; 89: 1278–1444, doi:10.1253/circj.CJ-25-0002.
• 4.   Fujimoto W, Toh R, Takegami M, Hayashi T, Kuroda K, Hatani Y, et al. Estimating incidence of acute heart failure syndromes in Japan: An analysis from the KUNIUMI Registry. Circ J 2021; 85: 1860–1868, doi:10.1253/circj.CJ-20-1154.
• 5.   Kanaoka K, Iwanaga Y, Nakai M, Sumita Y, Saito Y, Miyamoto Y. Temporal trends and regional variations in cardiovascular care in Japan, 2010–2019. Int Heart J 2023; 64: 53–59, doi:10.1536/ihj.22-445.
• 6.   Makita S, Yasu T, Akashi YJ, Adachi H, Izawa H, Ishihara S, et al. JCS/JACR 2021 Guideline on rehabilitation inpatients with cardiovascular disease. Circ J 2022; 87: 155–235, doi:10.1253/circj.CJ-22-0234.
• 7.   Shiraishi Y, Kohsaka S, Sato N, Takano T, Kitai T, Yoshikawa T, et al. 9-Year trend in the management of acute heart failure in Japan: A report from the national consortium of acute heart failure registries. J Am Heart Assoc 2018; 7: e008687, doi:10.1161/JAHA.118.008687.
• 8.   Kitzman DW, Whellan DJ, Duncan P, Pastva AM, Mentz RJ, Reeves GR, et al. Physical rehabilitation for older patients hospitalized for heart failure. N Engl J Med 2021; 385: 203–216, doi:10.1056/NEJMoa2026141.
• 9.   Kleczynski P, Trebacz J, Stapor M, Sobczynski R, Konstanty-Kalandyk J, Kapelak B, et al. Inpatient cardiac rehabilitation after transcatheter aortic valve replacement is associated with improved clinical performance and quality of life. J Clin Med 2021; 10: 2125, doi:10.3390/jcm10102125.
• 10.   Delgado B, Novo A, Lopes I, Rebelo C, Almeida C, Pestana S, et al. The effects of early rehabilitation on functional exercise tolerance in decompensated heart failure patients: Results of a multicenter randomized controlled trial (ERIC-HF study). Clin Rehabil 2022; 36: 813–821, doi:10.1177/02692155221088684.
• 11.   Oliveira MF, Santos RC, Artz SA, Mendez VMF, Lobo DML, Correia EB, et al. Safety and efficacy of aerobic exercise training associated to non-invasive ventilation in patients with acute heart failure. Arq Bras Cardiol 2018; 110: 467–475, doi:10.5935/abc.20180039.
• 12.   Nakaya Y, Akamatsu M, Ogimoto A, Kitaoka H. Early cardiac rehabilitation for acute decompensated heart failure safely improves physical function (PEARL study): A randomized controlled trial. Eur J Phys Rehabil Med 2021; 57: 985–993, doi:10.23736/S1973-9087.21.06727-7.
• 13.   Rinaldo L, Caligari M, Acquati C, Nicolazzi S, Paracchini G, Sardano D, et al. Functional capacity assessment and minimal clinically important difference in post-acute cardiac patients: The role of Short Physical Performance Battery. Eur J Prev Cardiol 2022; 29: 1008–1014, doi:10.1093/eurjpc/zwab044.
• 14.   Suzuki K, Mito G, Kawamoto H, Hasegawa Y, Sankai Y. Intention-based walking support for paraplegia patients with Robot Suit HAL. Advanced Robotics 2007; 21: 1441–1469.
• 15.   Lee S, Sankai Y. Virtual impedance adjustment in unconstrained motion for an exoskeletal robot assisting the lower limb. Adv Robot 2005; 19: 773–795.
• 16.   Watanabe H, Koike A, Pak YJ, Wu L, Kubota H, Konno H, et al. Effects of a lumbar-type hybrid assistive limb on cardiopulmonary burden during squat exercise in healthy subjects. J Clin Neurosci 2019; 66: 226–230, doi:10.1016/j.jocn.2019.05.026.
• 17.   Kato H, Watanabe H, Koike A, Wu L, Hayashi K, Konno H, et al. Effects of cardiac rehabilitation with lumbar-type hybrid assistive limb on muscle strength in patients with chronic heart failure: A randomized controlled trial. Circ J 2021; 86: 60–67, doi:10.1253/circj.CJ-21-0381.
• 18.   Izawa H, Yoshida T, Ikegame T, Izawa KP, Ito Y, Okamura H, et al. Standard cardiac rehabilitation program for heart failure. Circ J 2019; 83: 2394–2398, doi:10.1253/circj.CJ-19-0670.
• 19.   Sankai Y, Sakurai T. Exoskeletal cyborg-type robot. Sci Robot 2018; 3: eaat3912, doi:10.1126/scirobotics.aat3912.
• 20.   Watanabe H, Koike A, Wu L, Kato H, Konno H, Sato A, et al. Efficacy of cardiac rehabilitation with assistance from hybrid assistive limb in patients with chronic heart failure: Protocol for a randomized controlled Study. Cardiology 2019; 142: 213–219, doi:10.1159/000500263.
• 21.   Ignacio de Ulibarri J, Gonzalez-Madrono A, de Villar NG, Gonzalez P, Gonzalez B, Mancha A, et al. CONUT: A tool for controlling nutritional status. First validation in a hospital population. Nutr Hosp 2005; 20: 38–45, https://www.ncbi.nlm.nih.gov/pubmed/15762418.
• 22.   Guralnik JM, Simonsick EM, Ferrucci L, Glynn RJ, Berkman LF, Blazer DG, et al. A short physical performance battery assessing lower extremity function: Association with self-reported disability and prediction of mortality and nursing home admission. J Gerontol 1994; 49: M85–M94, doi:10.1093/geronj/49.2.m85.
• 23.   Yokota C, Kamada M, Nakatsuka K, Takegami M, Miura H, Murata M, et al. Effect of outpatient cardiac rehabilitation on motor function and health-related quality of life in stroke survivors. J Clin Neurosci 2024; 123: 1–6, doi:10.1016/j.jocn.2024.03.013.
• 24.   National Cerebral and Cardiovascular Center. Hospital Quality Indicators, Reiwa 4. https://www.ncvc.go.jp/hospital/about/quality/template_R05.html#qi03 (accessed April 2, 2025).
• 25.   Yamamoto K, Noda T, Ito K, Miura H, Murata M, Yokota C. Effect of inpatient cardiac rehabilitation combined with waon therapy on exercise capacity in elderly patients with heart failure: A pilot study. Physical Therapy Research 2025, doi:10.1298/ptr.E10322.
• 26.   Leishear K, Boudreau RM, Studenski SA, Ferrucci L, Rosano C, de Rekeneire N, et al. Relationship between vitamin B12 and sensory and motor peripheral nerve function in older adults. J Am Geriatr Soc 2012; 60: 1057–1063, doi:10.1111/j.1532-5415.2012.03998.x.