A controlled clinical trial of leg thermal therapy in conjunction with standard cardiac rehabilitation in patients with decompensated heart failure

Abstract

[Purpose] Leg thermal therapy (LTT) using far-infrared-ray dry sauna stimulation improves hemodynamics in patients with heart failure (HF); however, its additive effect when combined with cardiac rehabilitation (CR) remains unknown. This study aimed to investigate whether incorporating LTT into acute-phase CR confers exercise capacity in hospitalized patients with HF. [Participants and Methods] Seventeen patients with decompensated HF (median age 75 years) admitted between June 2018 and November 2019 were prospectively assigned to the CR plus LTT (11 participants) or the CR group (6 participants). Both groups performed structured exercise sessions for 40 min, five times a week. The LTT group received additional leg heating at 45°C for 20 min. Cardiopulmonary function and blood tests were performed before and after the intervention. [Results] At one month, peak oxygen uptake was equal between groups. However, anaerobic threshold oxygen uptake increased in the CR plus LTT group, noradrenaline levels decreased, and no complications were observed. [Conclusion] The addition of LTT to CR did not improve overall exercise capacity in hospitalized patients with HF, but it contributes to specific physiological improvements. Hence, this could be a potential approach to optimize cardiac rehabilitation for patients with decompensated HF.

INTRODUCTION

Heart failure (HF) has been described as a life-threatening syndrome that gradually decreases tolerated exercise capacity¹⁾. Its prevalence in outpatients has rapidly increased as the population of older people has increased in Japan²⁾, and elderly patients are more susceptible to a cycle of frailty during cardiac decompensation³⁾. Cardiac rehabilitation (CR) in patients with HF is an essential treatment to improve exercise tolerance and prevent hospitalization^{4, 5)}. Early ambulation, leading to exercise therapy, should be initiated in conjunction with acute care for HF⁶⁾ because their skeletal muscle quickly and progressively loses strength and mass due to physical deconditioning⁷⁾. A multicenter retrospective cohort study demonstrated a substantial prognostic benefit of CR in the outpatient HF population in addition to being effective for hospitalized patients⁸⁾. However, there is a lack of consensus regarding the effectiveness of CR in elderly patients with HF. To improve the effectiveness of CR in this group, new practical and additional therapeutic interventions must be developed.

Blood flow through the skeletal muscle is 20 times greater during exercise than at rest⁹⁾. Endothelium-derived nitric oxide (NO) plays a significant role in blood flow redistribution during exercise. This means that endothelial dysfunction in patients with HF causes a decrease in exercise capacity due to attenuation in their hyperemic blood flow response induced by exercise. In other words, sensitivity to shear stress of skeletal muscle arterioles during adaptation of peripheral circulation in response to exercise plays an important role¹⁰⁾. In recent years, leg thermal therapy (LTT) using far-infrared radiation, which increases the core body temperature, has been shown to improve hemodynamics¹¹⁾ and sleep quality¹²⁾ in HF patients. LTT reduces oxidative stress and improves endothelial function in this group¹¹⁾. Notably, one study indicates that administering LTT prior to exercise may augment exercise tolerance in patients with cardiac disease¹³⁾. However, little is known about the clinical benefit of LTT on exercise capacity in HF patients. Therefore, we hypothesized that LTT, in combination with guideline-based therapy for HF¹⁾, would efficiently improve exercise tolerance in elderly patients. The present study aimed to investigate whether prescribing LTT in addition to CR confers benefits for exercise capacity as determined by cardiopulmonary exercise testing (CPET) in patients in the acute phase of HF. This work is expected to make an original contribution to the field of CR in hospitals and outpatients.

PARTICIPANTS AND METHODS

This study was an open-label, pseudorandomized, controlled trial. Patients were selected from 68 consecutive patients admitted to Sasayama Medical Center, Hyogo Medical University, for exacerbation of HF between June 2018 and November 2019. Of the 68 patients, 17 met the inclusion criteria and participated in this study. The inclusion criteria were as follows: 1) age ≥20 years; 2) hospitalization for HF with New York Heart Association (NYHA) class III to IV symptoms; 3) HF, defined as either left ventricular ejection fraction (LVEF) of 40% or less, peak oxygen uptake (VO_2peak) of 80% or less than the standard value, or plasma brain natriuretic peptide (BNP) of 80 pg/mL or above; and 4) written informed consent obtained from the patient or legal representative. The exclusion criteria were as follows: 1) acute inflammatory diseases such as bacterial infection or thrombophlebitis; 2) paresthesia of a leg; 3) uncontrollable bleeding or known bleeding diathesis with active bleeding; 4) cancer; 5) a condition or factor that contraindicated hyperthermia therapy; and 6) a contraindication to exercise testing or training. The patients who fulfilled the enrollment criteria entered an approximately one- or two-week run-in period. Treatment was based on oxygen administration for hypoxia, vasodilators adjunct to diuretics for fluid overload, inotropic support for depressed cardiac output, and short-term noninvasive positive-pressure ventilation for respiratory distress most appropriate to the initial severity of HF and adjusted according to the clinical response to achieve optimal benefits. Following the run-in period, participants were randomly assigned to either the CR plus LTT group (LTT group) or the CR alone group (control group). Randomization was performed based on even and odd medical record numbers, while the trial team provided unblinded treatment arm allocation information to participants and their physical therapists. It should be noted that the study did not implement blinding for assessors. However, all members of the nursing and pharmacy teams remained unaware of the treatment allocations. Although a 30-day intervention period was designated in this study, the necessary evaluations were carried out ahead of schedule in patients discharged before one month.

The trial protocol (No. 202102-081) was approved by the institutional review board of Hyogo Medical University on August 14, 2018 and registered at the University Hospital Medical Information Network Clinical Trials Registry (ID: UMIN000031444). We informed all patients about the research, and each signed a consent form. This investigation conformed to the principles outlined in the Declaration of Helsinki and all Consolidated Standards of Reporting Trials guidelines; the required information is reported accordingly¹⁴⁾.

Exercise testing was conducted by bicycle ergometry (Well Bike BE-260). During testing, the electrocardiogram was continuously recorded (MLX-1000, Fukuda Denshi, Tokyo, Japan), and blood pressure was measured every minute (Tango M2, Suntech, Carlsbad, CA, USA). Peripheral oxygen hemoglobin saturation was monitored by a pulse oximeter. The patients stood for five minutes at rest to stabilize their hemodynamic variables and then cycled at 50 to 60 rpm with no workload for three minutes. The workload increased by 10 watts every minute (ramp) after that until the symptom-limited maximal workload was reached. A gas analyzer (AE310-s, Minato Medical Science, Osaka, Japan) collected and analyzed expired respiratory gasses during exercise testing. Breath-by-breath gas exchange data were measured continuously during exercise and averaged every 15 s. VO_2peak was calculated as the highest oxygen uptake during the last 20–30 s of maximal effort during testing. Anaerobic threshold oxygen uptake (VO₂AT) was calculated by the V-slope method. Minute ventilation per carbon dioxide slope (VE/VCO₂ slope) was calculated as previously described¹⁵⁾.

An expert sonographer performed Doppler echocardiography in a blinded session. Standard views, including the parasternal long-axis, short-axis, and apical four- and two-chamber views, were recorded. Left atrial dimension (LAD), end-systolic left ventricular diameter (LVDs), and end-diastolic left ventricular diameter (LVDd) were measured with two-dimensional echocardiography (EPIQ 7G, Philips, Tokyo, Japan).

Blood samples were taken to determine the levels of BNP, noradrenaline (NA), tumor necrosis factor-alpha (TNF-α), plasminogen activator inhibitor-1 (PAI-1), and insulin-like growth factor-1 (IGF-1) in the supine position after 30 min of quiet rest, according to the manufacturer’s procedures. BNP was measured by chemiluminescent enzyme immunoassay, NA was measured by high-performance liquid chromatography, TNF-α was measured by enzyme-linked immunosorbent assay, PAI-1 was measured by latex photometric immunoassay, and IGF-1 was measured by immunoradiometric assay.

Both groups underwent supervised exercise for 40 min five times a week, based on the standard CR program for HF¹⁶⁾. Each session included a warm-up period (5–10 min), resistance training (10 min), aerobic exercise (10–20 min), and a cool-down period (5–10 min). Aerobic exercise was performed by bicycle ergometry or walking on the ground at the heart rate (HR) of the anaerobic threshold or at a perceived exertion of 11–13 on the Borg scale. Resistance training was performed as 2–3 sets of ten repetitions of closed-chain exercises (squat and heel raise) or leg press (workload corresponding to 40–60% of the one-repetition maximum) at 11–13 on the Borg scale. In contrast, stretching was mainly performed in patients with NYHA class IV, and aerobic exercise with assistance was performed using the ergometry function of the electric cycling machine (PBE-100, Meisei, Tokyo, Japan). Exercise intensity and duration were adjusted based on medical considerations and clinical responses.

The LTT protocol was performed as previously reported¹²⁾. Patients received leg heating with far infrared radiation (Leghot^®, Fujika Co., Ltd., Tokyo, Japan) at 45°C for 20 min and then remained in bed at rest with a blanket to keep them warm for 30 min (Fig. 1). The LTT was performed either before or after CR, depending on the patient’s decision. Additionally, the core body temperature under the tongue was measured using an electronic thermometer (C405, TERUMO, Tokyo, Japan) both before and after the completion of the LTT¹⁷⁾.

Fig. 1.

A: leg thermal therapy (LTT) using far-infrared radiation, B: keeping warm with a blanket after LTT.

The primary endpoint was the change in VO_2peak from baseline to follow-up before discharge. The secondary outcomes were the changes in VO₂AT, VE/VCO2 slope, LAD, LVDs, LVDd, plasma BNP level, NA level, TNF-α level, PAI-1 level, and IGF-1 level from baseline to follow-up before discharge. We reported any adverse events associated with LTT, such as local burns.

The required sample size for the study was calculated using G*Power (version 3.1.9.7) (Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany), a statistical software. It is crucial to ascertain that the difference in VO_2peak, which is the primary outcome defined in the study, is roughly 1.0 mL/kg/min between the LTT and control groups. Moreover, prior research has indicated that the standard deviation of VO_2peak measurements is approximately 2.0 mL/kg/min¹⁸⁾. Consequently, with an alpha level (α) of 0.05, a statistical power (1-β) of 0.70, and an effect size of 0.5, the requisite sample size was calculated to be approximately 100 participants.

The baseline characteristics of the patients were compared between groups using the Mann–Whitney U test, χ² test and Fisher’s exact test. Comparisons of medications and outcome measures before and after the intervention in each group were performed using the Wilcoxon’s signed-rank tests, χ² test and Fisher’s exact test. Intergroup comparisons of medications and outcome measures were performed using the Mann–Whitney U test, χ² test and Fisher’s exact test. A univariate regression analysis employing Spearman’s correlation coefficient was conducted to assess the relationship between the duration of the intervention and the resulting changes in VO_2peak, as well as the correlation between the duration of the intervention and the subsequent changes in VO₂AT. The effect size was also calculated as r. Cohen’s guideline for small (r=0.10), medium (r=0.30), and large effects (r=0.50) was used to evaluate the magnitude of this effect size. All statistical analyses were performed using SPSS Statistics for Windows (version 22; IBM, Armonk, NY, USA). The α level was set to 0.05 (two-sided).

RESULTS

A total of 68 patients were assessed for eligibility. Fifty-one patients were excluded for not meeting the inclusion criteria. Thus, 17 patients were enrolled in this study and were assigned to either the LTT group (n=11) or the control group (n=6). One patient was excluded from the LTT group during the intervention due to active infection. No intervention-related complications occurred during the study period in either group. Figure 2 shows the study’s flow diagram of patient recruitment and retention. The baseline characteristics following the run-in period of the patients who completed the intervention are shown in Table 1. Three patients improved to NYHA class II as a result of the initial treatment (LTT group: 2; Control group: 1). There were no significant differences in any variables at baseline between groups. Each group’s mean age was over 75 because the location of our rural hospital is in an aging community.

Fig. 2.

The flow chart of the study procedure.

Table 1. Baseline characteristics.

	LTT group (n=10)	Control group (n=6)
Age, years	75.0 [72.0–91.3]	73.5 [69.8–83.3]
Sex, male:female, n	7:3	4:2
BMI, kg/m2	18.9 [17.5–20.3]	20.6 [18.4–21.3]
NYHA class, Ⅰ:Ⅱ:Ⅲ:Ⅳ, n	0:2:7:1	0:1:4:1
LVEF, %	40.9 [31.3–51.8]	26.2 [17.5–35.4]
LVEF <50%, n (%)	4 (40.0)	3 (50.0)
Duration of admission, days	45.0 [32.5–57.3]	44.0 [27.8–57.3]
Duration of intervention, days	30.5 [23.8–38.8]	25.0 [20.0–27.8]
Underlying heart disease, n (%)
DCM	3 (30.0)	2 (33.3)
HCM	1 (10.0)	1 (16.7)
IHD	3 (30.0)	1 (16.7)
HHD	1 (10.0)	1 (16.7)
Others	2 (20.0)	1 (16.7)
Comorbidities, past or present, n (%)
Hypertension	7 (70.0)	4 (66.7)
Diabetes mellitus	4 (40.0)	4 (66.7)
Hyperlipidemia	4 (40.0)	5 (83.3)
Myocardial infarction	2 (20.0)	1 (16.7)
COPD	4 (40.0)	3 (50.0)
Cerebral infarction	2 (20.0)	1 (16.7)

Data are median [interquartile range] or number of patients (%). There were no significant differences in any variables at baseline between groups. LTT: leg thermal therapy; BMI: body mass index; NYHA: New York Heart Association; LVEF: left ventricular ejection fraction; DCM: dilated cardiomyopathy; HCM: hypertrophic cardiomyopathy; IHD: ischemic heart disease; HHD: hypertensive heart disease; COPD: chronic obstructive pulmonary disease.

At the baseline, all patients received β-blockers in the LTT group, and none received sodium-glucose cotransporter-2 inhibitors or digoxin in the control group. In the LTT group, one patient no longer needed an inotropic drug, pimobendan, and no patients received it after the intervention. No one received digoxin in the control group during the study period. There were no significant differences in any other patterns of medication use between the LTT group and the control group before or after the intervention in either the within- or the between-group analyses (Table 2). Core body temperature showed a significant increase pre- and post-intervention in the LTT group, recorded as follows: Pre 36.5 (IQR 36.2–36.6) and Post 36.7 (IQR 36.5–36.8)°C (p=0.04, r=0.64).

Table 2. Medications before and after intervention.

	Pre-intervention		Post-intervention
	LTT group	Control group	LTT group	Control group
β-blocker	10 (100.0)	5 (83.3)	10 (100.0)	6 (100.0)
ACE inhibitors/ARB	7 (70.0)	5 (83.3)	6 (60.0)	5 (83.3)
MRA	6 (60.0)	2 (33.3)	7 (70.0)	1 (16.7)
SGLT2 inhibitors	2 (20.0)	1 (16.7)	2 (20.0)	1 (16.7)
Diuretics	8 (80.0)	5 (83.3)	7 (70.0)	5 (83.3)
Pimobendan	1 (10.0)	1 (16.7)	0 (0.0)	1 (16.7)
CCB	1 (10.0)	1 (16.7)	1 (10.0)	1 (16.7)
Statins	4 (40.0)	5 (83.3)	4 (40.0)	5 (83.3)
Digoxin	1 (10.0)	0 (0.0)	1 (10.0)	0 (0.0)
Antiplatelet	3 (30.0)	1 (16.7)	3 (30.0)	1 (16.7)

Number of patients (%). There were no significant differences in any other patterns of medication use between the LTT group and the control group before or after the intervention in either the within- or the between-group analyses. LTT: leg thermal therapy; ACE: angiotensin-converting enzyme; ARB: angiotensin-receptor blocker; MRA: mineralocorticoid receptor antagonist; SGLT2: sodium glucose cotransporter 2; CCB: calcium-channel blocker.

Table 3 shows the outcome measures before and after the intervention in both groups. Regarding the primary outcome measure, there were no significant differences in VO_2peak between the groups. In addition, VO_2peak showed no significant differences before and after the intervention in either group. In the between-group analysis for secondary outcome measures, no significant differences were apparent in any parameters. In the within-group analysis for secondary outcome measures, there was a significant increase in VO₂AT (r=0.63) and a decrease in NA (r=0.85) before and after the intervention in the LTT group (Fig. 3). In the control group, IGF-1 levels significantly increased (p=0.03, r=0.61) during the study. BNP decreased in both LTT and control groups during the study period (p=0.08, r=0.56 and p=0.06, r=0.81, respectively). In this study, the median peak respiratory exchange ratio (RER), which serves as an index to reflect the quality of CPET¹⁹⁾, went from 1.07 [interquartile range (IQR) 1.01–1.22] at baseline to 1.08 [IQR 1.06–1.17] after intervention in the LTT group and 1.13 [IQR 0.99–1.29] to 1.11 [IQR 1.02–1.21] in the control group. Statistical analyses indicated no significant differences in peak RER comparisons, with a baseline p-value of 0.79 and post-intervention results showing a p-value of 0.91. Regarding the duration of the intervention, no significant correlation was found between the length of the intervention and the changes in VO_2peak (p=0.20, r=0.34). Additionally, no correlation was observed between the duration of the intervention and the changes in VO₂AT (p=0.34, r=0.26).

Table 3. Effects of intervention on outcome measures.

	Pre-intervention		Post-intervention
	LTT group	Control group	LTT group	Control group
Primary outcome
VO2peak, mL/kg/min	10.3 [8.1–12.1]	10.6 [9.8–10.9]	10.7 [10.3–11.6]	10.9 [9.9–11.3]
Secondary outcome
VO2AT, mL/kg/min	7.1 [6.4–7.5]	6.9 [6.5–7.3]	7.7 [6.8–8.1]*	7.2 [6.9–7.3]
VE/VCO2 slope	42.4 [38.2–47.0]	40.7 [35.2–55.6]	42.6 [34.3–46.2]	46.0 [42.3–57.8]
LAD, mm	41.9 [36.7–45.7]	42.5 [39.7–44.5]	43.5 [32.4–46.7]	40.2 [39.5–40.7]
LVDd, mm	50.2 [46.4–53.6]	56.6 [51.7–60.2]	49.7 [44.9–52.7]	50.6 [49.3–54.9]
LVDs, mm	40.0 [33.8–46.7]	44.6 [42.0–53.3]	42.6 [29.3–44.4]	36.7 [35.8–42.4]
BNP, pg/mL	191.5 [123.0–610.3]	671.0 [628.5–808.0]	181.5 [125.3–358.8]	386.5 [266.3–654.5]
NA, pg/mL	598.5 [375.3–962.0]	491.5 [390.5–903.8]	500.5 [264.3–754.3]*	545.5 [299.5–809.5]
TNF-α, pg/mL	2.0 [1.4–2.4]	2.0 [1.5–2.6]	1.9 [1.4–2.3]	2.2 [1.3–3.3]
PAI-1, ng/mL	15.5 [12.3–16.8]	11.5 [10.3–16.5]	14.5 [10.5–17.5]	15.5 [13.5–17.5]
IGF-1, ng/mL	83.5 [61.8–130.3]	78.0 [67.8–87.5]	91.5 [83.3–109.0]	100.5 [81.5–117.3]*

Data are median [interquartile range]. *Within-group comparison (pre vs. post), p<0.05. There were no significant differences in any parameters between groups. LTT: leg thermal therapy; VO_2peak: peak oxygen uptake; VO₂AT: anaerobic threshold oxygen uptake; VE/VCO₂ slope: regression slope relating minute ventilation to carbon dioxide output; LAD: left atrial diameter; LVDd: left ventricular diastolic dimension; LVDs: left ventricular systolic dimension; BNP: brain natriuretic peptide; NA: noradrenaline; TNF-α: tumor necrosis factor-α; PAI-1: plasminogen activator inhibitor-1; IGF-1: insulin-like growth factor-1.

Fig. 3.

Comparisons of anaerobic threshold oxygen uptake and noradrenaline before and after the intervention in both groups. VO₂AT: anaerobic threshold oxygen uptake; LTT: leg thermal therapy.

DISCUSSION

To the best of our knowledge, this study is the first controlled clinical trial to investigate the efficacy of LTT plus CR for HF patients. The primary outcome measures showed no significant differences between the two groups, whereas patients in the LTT group exhibited a significant increase in VO₂AT compared to baseline. Additionally, their plasma NA level decreased substantially during the study period. The indices of VO_2peak and VO₂AT have been shown to be associated with prognosis in HF patients²⁰⁾. It could be of clinical importance that LTT, in conjunction with CR, improved VO₂AT in elderly HF patients, including those with severe HF, and even in patients early after an acute episode of cardiac decompensation, for whom the effect of CR alone has not been demonstrated. Our findings might suggest an additional treatment option for HF in older patients.

A meta-analysis concluded that the most profound effects of CR in patients with mild to moderate HF were found in measures that reflect exercise capacity, such as maximum HR, maximum cardiac output, VO_2peak, VO₂AT, 6-min walking distance, and health-related quality of life²¹⁾. The most impressive increase was found in VO₂AT, although there are some problems associated with detecting the VO₂AT in patients with HF²²⁾. This finding suggests that sub-maximal exercise testing may detect changes more sensitively than maximal exercise training. In performing symptom-limited CPET, the level of peak workload is crucial for detecting changes in VO_2peak after a period of training. Peak RER >1.1 is an objective criterion of maximal effort²³⁾. Although we found no statistical significance in peak RER comparisons, we can speculate that the increased VO₂AT after the intervention may indicate an additive effect of LTT because the peak RER in CPET in the LTT group, whose median was ≤1.1, would not be enough to detect a VO_2peak leading to variations in values.

A repeated sauna therapy called Waon therapy (WT) improved the exercise tolerance and prognosis of HF patients^24,25,26). Alternative LTT raises the core body temperature by only approximately 0.3°C, whereas WT can raise it by nearly 1.0–1.2°C²⁷⁾. Nonetheless, LTT appears to be comparable to WT because the rate of improvement in endothelial function is almost the same¹¹⁾, possibly related to the modest recovered exercise tolerance shown in our study. Our participants suffering from severe HF might have had deteriorated physical conditions that were too poor to achieve the same level of improvement as in other studies. However, knee extensor strength improved significantly in the LTT group, probably indicating induced rapid adaptations in skeletal muscle function promoted by CR with LTT (Supplementary Fig. 1). Indeed, knee extensor strength has been correlated with VO_2peak and VO₂AT²⁸⁾ and negatively correlated with VE/VCO₂ slope²⁹⁾. Intriguingly, our results show that combined treatment with CR and LTT reduces NA levels. Additionally, we did not observe any effect on the reduction of NA in the control group. Exercise training has been shown to reduce resting plasma catecholamine levels³⁰⁾. NA reduction might be implicated in restoring autonomic balance and could be one of the mechanisms by which exercise tolerance appeared to be improved in the LTT group. In addition, LTT, which uses a handy far-infrared radiant heater for warming legs, has the advantage that it can be performed quickly on-site and continued at home in contrast to WT. LTT might be useful for enhancing the effects of home-based CR.

Recently, there have been many reports on the effect of CR on exercise tolerance in elderly HF patients, but most have targeted stable, chronic HF. The outcome or duration of intervention varies between studies, and older patients, usually with various complications, are heterogeneous populations. Thus, it is not easy to provide CR programs supported by solid evidence for elderly HF participants. One randomized controlled trial unveiled that elderly HF patients undergoing CR for four weeks early after an acute episode of cardiac decompensation improved their VO_2peak and VO₂AT³¹⁾. However, we enrolled participants who were much older and included a higher percentage of women than in their study. Exercise training in patients with HF improves exercise capacity more in men than in women³²⁾. In addition, we enrolled a challenging set of patients, those with NYHA class IV, at the start of the intervention, while previous research targeted NYHA class II and III participants. It is well known that HF patients participating in CR who do not show improvements or exhibit less than a 6% increase in VO_2peak within six months, called nonresponders, should be considered at higher risk for worsening HF³³⁾. Most of our participants had several factors involved in nonresponse to CR, such as higher age and lower baseline VO_2peak³⁴⁾. In fact, the median VO_2peak showed a modest increase of 2.8% in the control group and 3.9% in the LTT group, while the median VO₂AT improved by 4.3% in the control group compared to an 8.4% increase in the LTT group, indicating a potentially favorable response to CR among participants in the LTT cohort. Furthermore, we had to exclude more candidates from the study than expected. Some patients have back or knee problems that limit their physical activity, while others require nursing care and suffer from cognitive decline. Indeed, a prospective multicenter cohort study has shown that physical frailty, social frailty, and cognitive disabilities are common in hospitalized elderly patients with HF, even those who are able to walk alone without the help of another person and overlap considerably with each other³⁵⁾.

Both groups were able to perform the exercises safely without any adverse events during the study period. The safety of LTT in chronic HF patients^{11, 12)} and of CR in elderly patients with decompensated HF²⁶⁾ has been reported, and our results support those findings. Furthermore, it is important that even though the patients in this study were older than those in earlier studies and some had severe diseases, the study was conducted safely.

This was a non-blinded, single-center study with a small number of participants, so its results cannot be generalized. In fact, because of the coronavirus disease 2019 outbreak, it was challenging to recruit a newly qualified inpatient participant, which brought a premature end to our study. Future multicenter studies recruiting a large number of participants are needed. The assessors involved in this study were not blinded to participant allocation, which may have introduced potential assessment bias and influenced outcome evaluations. It is recommended that future studies implement blinded assessments to minimize this risk. Additionally, there was variation in the distribution of diseases and severities among the participants. Although an accurate conclusion about efficacy can only be drawn after fully considering the available CPET-derived variables, very few participants were eligible for CPET in our study because of their frailty. Further studies should address this issue, although there are few alternative methods to assess exercise tolerance in patients with HF.

In conclusion, among HF patients soon after an acute cardiac decompensation episode, exercise capacity as determined by CPET was not higher among those who used LTT during CR than among those who underwent standard CR. LTT was well tolerated, and no intervention-related complications occurred during the study period.

Funding

There were no sources of funding for the present study.

Conflicts of interest

The authors declare that they have no conflicts of interest.