2025 Volume 7 Issue 4 Pages 231-238
Background: Neuromuscular electrical stimulation (NMES) is an alternative therapy for patients unable to perform sufficient voluntary exercises. This randomised crossover study aimed to evaluate the safety and efficacy of home-based NMES as an adjunct to cardiac rehabilitation (CR) for improving physical function in frail older adult patients with chronic heart failure (CHF).
Methods and Results: 8 frail older adult patients with CHF underwent 8 weeks of CR supplemented with home-based NMES and 8 weeks of CR alone in random order, separated with a 4-week washout period. NMES at 50-Hz frequency was administered for 50 min/day, 5 times per week, with electrodes placed on the legs. Changes in the short physical performance battery (SPPB) score, leg strength, and the Barthel index were assessed between patients with CR with and without home-based NMES. No NMES-related adverse events were observed. CR with home-based NMES had a higher total SPPB score and 5-repetition sit-to-stand test time of 2.67 points and −10.67 s, respectively, than CR alone (95% confidence interval [CI] 0.3–5.0, P<0.05 and 95% CI −19.5 to −1.3, P<0.05, respectively). No significant leg strength or Barthel index changes were observed between CR with and without home-based NMES.
Conclusions: Home-based NMES safely improved physical function in frail older adult patients with CHF.
The prevalence of chronic heart failure (CHF) in older adult patients has steadily increased over the past 20 years in both Western and Asian countries, including Japan.1,2 Older adult patients with CHF have high rates of rehospitalization and death, leading to increasing healthcare costs.3,4 Because reduced exercise tolerance and muscle strength are factors associated with poor prognosis in patients with CHF,5,6 exercise-based cardiac rehabilitation (CR) is now widely recognized as an effective treatment to improve physical activity in patients with CHF and thus reduce rehospitalization and mortality rates.7,8 However, the implementation rate of outpatient CR for patients with CHF in Japan is reported to be low,9–12 related to aging, frailty, and reduced performance of activities of daily living (ADL).12 Guidelines recommend that patients with CHF undergo rehabilitation 3–5 times per week,13 but older adult patients with CHF often have poorer exercise adherence because of their significant comorbidities, and increased feelings of anxiety and depression.13,14 Thus there is a great need for home-based adjunct therapies that can be easily implemented for frail older adult patients with CHF undergoing CR.
Neuromuscular electrical stimulation (NMES) is an alternative therapy for patients with CHF who have difficulty performing sufficient voluntary exercise training.13,15 NMES therapy stimulates muscle contraction by the application of a transcutaneous electric current through nerves and skeletal muscles, requiring no voluntary effort from the patient, and not causing breathing difficulties.13,15 NMES therapy has a beneficial effect on functional capacity and status, vascular endothelial function, quality of life, and aerobic enzyme activity, as well as on patient adherence to rehabilitation programs.13,15 Short-term NMES therapy, prescribed as an adjunct to in-hospital rehabilitation, has been reported to improve lower extremity function in frail older adult patients (aged ≥75 years) hospitalized with acute HF.16
The aim of this study was to investigate the safety and efficacy of home-based NMES therapy as an adjunct to CR for improving physical function in frail older adult patients with CHF.
Home-dwelling patients who were receiving optimal medical therapy for CHF secondary to ischemic cardiomyopathy, dilated cardiomyopathy, arrhythmia, and/or valvular disease, and had undergone CR for >3 months were included in this study. The inclusion criteria were: patients aged ≥75 years who were frail, but hemodynamically stable (defined as having had no hospital admission for HF), and with no changes in their New York Heart Association (NYHA) functional class and subjective symptoms ≥1 month prior to study entry.17 A patient was defined as frail if they scored ≤8 points (for men) and ≤7 points (for women) on the short physical performance battery (SPPB),18 or ≥8 points on the Kihon checklist (KCL).19 Exclusion criteria were: patients with dementia; those who undertook moderate-to-vigorous physical exercise several times per week; those with cancer; or those with comorbid conditions that may have affected their performance or the evaluation or intervention within the past 3 months, including central and peripheral nervous system disease or severe joint disorders that would clearly affect improvement in motor function;20 and patients who could not provide consent. Dementia was assessed using the Mini-Cog, with dementia being defined as a Mini-Cog score ≤2 points.21
Study Design and ProtocolThis was a single-center, randomised, 2-period, controlled crossover trial. Study patients were randomly assigned to 1 of 2 sequences. In Sequence 1, patients underwent 8 weeks of CR+home-based NMES therapy during Phase 1, followed by a 4-week washout phase,22 and then 8 weeks of CR alone during Phase 2. Sequence 2 involved the same interventions, but in reverse order (Figure 1). All patients underwent baseline assessments of physical function, frailty, blood parameters, ADL, and physical activity 3 days prior to study entry. These assessments were also carried out within 3 days after the end of Phase 1, and 3 days before and after Phase 2. Blood samples were taken before and after phases 1 and 2 (Figure 1).
Randomised allocation and data collection timeline. CR, cardiac rehabilitation; NMES, neuromuscular electrical stimulation.
This study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Tokoha University (Kenshizu18-19). All patients were informed of the study and provided written informed consent. The study followed the CONSORT guidelines.23
Home-Based NMES TherapyFor each patient, 8 adhesive electrodes (50×90 mm) were placed on the skin over the upper lateral and lower medial aspects of the quadriceps muscle of both legs and over the upper and lower portions of the gastrocnemius muscle of both legs. An electrostimulator (ESPERGE; ITO Co. Ltd., Saitama, Japan) was used for NMES. Stimulation parameters were set as follows: biphasic current of 50 Hz, 250-µs pulse duration, mode on-time of 5 s, and mode off-time of 5 s. The stimulation intensity was gradually increased according to the patient’s tolerance until muscular contraction was observed. A physical therapist supervised the home-based NMES therapy on three occasions prior to commencing the study. The patients then underwent home-based NMES therapy for 50 min, 5 times per week for 8 weeks, as instructed. We based the NMES program in this study on that of 2 previous studies, which reported that NMES therapy for ≥30 h was necessary to improve exercise tolerance.22,23 In addition, a preliminary trial confirmed that 50-min NMES sessions were feasible for older adult patients with HF.
CRAll patients participated in a supervised, standardized, CR program either at home or in an outpatient setting, once or twice each week. The program followed Japanese Circulation Society and Japanese Association of Cardiac Rehabilitation guidelines with a combination of endurance and resistance training, including ergometer cycling or walking, and strength training using body weight or a resistance training machine.13 For endurance and resistance training, the prescribed exercise was moderate in intensity. If necessary, patients were prescribed ADL and balance training.
Study OutcomesThe primary outcome was defined as changes in the total SPPB score.24 The secondary outcomes were changes in quadriceps isometric strength (QIS), ADL, life-space mobility, and safety parameters.
SPPB ScoreOverall physical function was assessed using the SPPB score, which is a standardized, reproducible measure of global physical function that has been validated in frail older adults to predict a wide range of clinical outcomes.25 The SPPB evaluates the following 3 physical domains: (1) standing balance with a progressively narrow base of support, (2) usual gait speed over a 4 m distance, and (3) time to complete the 5-repetition sit-to-stand (5-STS) test to assess functional and muscle power of the lower limb. Each domain is scored based on patient performance, with scores ranging from 0 to 4 points, with a total SPPB score ranging from 0 to 12.
Leg Strength, ADL, and Life-Space MobilityQIS was measured twice for each leg using a handheld dynamometer (μTas F-1, Anima, Tokyo, Japan) as a parameter of leg strength. Patients were seated on a bench with their hips and knees flexed at 90°.17
The Barthel index (BI) was used to assess basic ADL and consists of 10 items: feeding, transfers, grooming, toilet use, bathing, ambulation, stair climbing, dressing, and bowel and bladder care. Scores range from 0 to 100, with 0 indicating complete dependence and 100 indicating complete independence.26
A life-space assessment (LSA) was used to measure life-space mobility levels27 because it has been shown to have a positive correlation with higher step counts and physical activity time.28
Safety ParametersThe safety of the intervention was checked weekly by either a physical therapist or a nurse for all patients throughout the study period. As a safety check, we assessed skin redness, rash, delayed-onset myalgia, and itching in the electrode attachment area of the legs. We also assessed cardiovascular adverse events such as exacerbation of HF and serious arrhythmia during the study period. Blood tests were performed to evaluate B-type natriuretic peptide (BNP), creatine kinase (CK), and high-sensitivity C-reactive protein (hs-CRP) levels as safety indicators.
Other ParametersPatients’ characteristics at baseline, including age, sex, body mass index, etiology of CHF, NYHA functional classification, current medications, and comorbidities, were obtained from medical records on the first study day. Left ventricular ejection fraction (LVEF) was measured using echocardiography within 1 month prior to study entry.
Statistical AnalysisBased on 3 registered patients, the mean±standard deviation (SD) SPPB was 2.33±1.18. When the correlation was set to 0.3, and the power and type 1 error in a 2-sided test fulfilled the conditions >0.9 and <0.05, respectively, using SAS version 9.4 software (SAS Institute Inc., Cary, NC, USA), the minimum sample sizes for comparison between 2 groups were calculated as 4 and 7 patients, respectively. Accordingly, the sample size in this study was determined to be 9 patients, considering potential dropouts.
We sought to determine the study objective with regard to the primary and secondary endpoints. In this randomised crossover study, the same patients performed 2 different interventions in 2 periods. Continuous variables were analyzed using mixed-effects models with compound symmetry covariances for repeated measures in each patient. We included the following prespecified baseline factors as fixed effects: intervention allocation, period, and intervention-by-period interactions. Patient identification was treated as a random effect. The least-squares mean (LSM) was calculated from a model adjusted for period and carryover effects. An intention-to-treat analysis was performed. Categorical and continuous data are presented as the number of patients (percentage) and mean±SD, respectively. Analyses were performed using R version 4.3.1 software (R Foundation, Vienna, Austria) and the mixed-effects model (nlme) library.
In total, 35 patients met the study inclusion criteria. We excluded patients with severe dementia (n=8), those who underwent moderate-to-vigorous physical exercise several times per week (n=1), those with cancer or comorbid conditions that may have affected the performance of the evaluation or intervention within the past 3 months (n=9), and those unable to provide consent (n=8) (Figure 2). One patient was also excluded after enrolment because of hospitalization. Finally, the 8 eligible patients who agreed to participate in the study were then randomised into Sequence 1 or 2, and assessed for baseline status (Table 1). The median patient age was 85 years, 38% were men, and 75% had preserved LVEF. Atrial fibrillation, orthopedic disorders, and chronic kidney disease rates in the study population were high. The study patients had moderate impairment of physical function and were deemed to be frail.
Study flowchart. CR, cardiac rehabilitation; NMES, neuromuscular electrical stimulation.
Patients’ Baseline Characteristics
| Factor | Overall (n=8) |
|---|---|
| Age, years | 85.5 [84.0, 88.0] |
| Male, n (%) | 3 (37.5) |
| BMI, kg/m2 | 22.2 [20.6, 23.1] |
| Etiology of heart failure, n (%) | |
| ICM/DCM/HCM/valvular/other | 3 (37.5)/0 (0)/2 (25.0)/2 (25.0)/1 (12.5) |
| NYHA functional class II/III, n (%) | 4 (50)/ 4 (50) |
| Comorbidities, n (%) | |
| Atrial fibrillation | 6 (75.0) |
| Orthopedic disease | 8 (100.0) |
| Diabetes mellitus | 5 (62.5) |
| Chronic kidney disease | 6 (75.0) |
| Laboratory data | |
| CPK, U/L | 66.0 [52.5, 74] |
| hs-CRP, mg/dL | 0.14 [0.07, 0.21] |
| eGFR, mL/min/1.73 m2 | 33.8 [22.8, 48.4] |
| BNP, pg/mL | 232.1 [166.0, 364.0] |
| Echocardiographic parameters | |
| LVEF, % | 53 [49, 54] |
| E/e′ | 17 [12.4, 20.6] |
| Medications, n (%) | |
| Diuretic | 8 (100.0) |
| ACE or ARB | 4 (50.0) |
| SGLT2 inhibitor | 3 (37.5) |
| β-blocker | 6 (75.0) |
| Physical function | |
| SPPB, points | 7.4±2.7 |
| Balance score, points | 2.1±1.3 |
| Gait speed score, points | 2.8±1.3 |
| 5-STS score, points | 2.0±0.9 |
| Usual gait speed, m/s | 0.69±0.3 |
| Grip strength, kgf | 19.0±4.4 |
| QIS, kgf | 14.9±7.2 |
| Frailty and comprehensive geriatric assessment | |
| J-CHS, points | 2.0 [2, 4] |
| KCL, points | 11 [4, 12] |
| BI, points | 90 [89, 91] |
| LSA, points | 49.0 [42.8, 60.1] |
Data are shown as n (%), mean±SD, or median [interquartile range]. 5-STS, 5-repetition sit-to-stand test; ACE, angiotensin-converting enzyme; ARB, angiotensin-receptor blocker; BI, Barthel index; BMI, body mass index; BNP, B-type natriuretic peptide; CPK, creatine phosphokinase; DCM, dilated cardiomyopathy; eGFR, estimated glomerular filtration rate; HCM, hypertrophic cardiomyopathy; HF, heart failure; hs-CRP, high-sensitivity C-reactive protein; ICM, ischemic cardiomyopathy; J-CHS, Japanese version for Cardiovascular Health Study Criteria; KCL, Kihon Checklist; LSA, life-space assessment; LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; SPPB, short physical performance battery.
Crude mean changes in total SPPB scores were 1.37 points for the NMES+CR group and −0.12 points for the CR group (Supplementary Table). The adjusted LSM was 2.00 points for the NMES+CR group and −0.67 points for the CR group. The NMES+CR group had an SPPB score 2.67 points greater than that of the CR group (95% CI 0.3–5.0, P=0.046) (Figure 3). Crude mean changes in the 5-STS test scores were −8.07 s for the NMES+CR group and −0.55 s for the CR group. The adjusted LSM was −10.38 for the NMES+CR group and −0.29 s for the CR group. The NMES+CR group had a 5-STS test score that was −10.67 s higher than that of the CR group (95% CI −19.5 to −1.3, P=0.045) (Figure 3).
Changes in total SPPB and SPPB subcategory scores. 5-STS, 5-repetition sit-to-stand test; CI, confidence interval; CR, cardiac rehabilitation; LS Mean, least-squares mean; NMES, neuromuscular electrical stimulation; SE, standard error; SPPB, short physical performance battery.
No significant differences in the adjusted LSM of the QIS, BI, and LSA were observed between the NMES+CR and CR groups (Table 2).
Changes in Muscle Strength, ADL, Life-Space Mobility, and Safety Parameters
| NMES + CR (n=8) | CR (n=8) | Δ | 95% CI | P value | |
|---|---|---|---|---|---|
| LS mean±SE | |||||
| QIS, ADL, LSA | |||||
| ΔQIS, kgf | 0.63±1.3 | −0.37±1.3 | 1.0 | −2.6 to 3.8 | 0.71 |
| ΔBI, points | 3.00±2.1 | 0.00±1.6 | 3.0 | −1.1 to 7.1 | 0.17 |
| ΔLSA, points | 3.3±12.1 | −1.50±9.6 | 4.8 | −20.4 to 27.0 | 0.79 |
| Safety parameters | |||||
| ΔBNP, pg/mL | 41.99±61.5 | −29.33±48.6 | 70.3 | −78.5 to 162.5 | 0.51 |
| Δhs-CRP, mg/dL | −0.04±0.1 | 0.10±0.1 | −0.14 | −0.3 to 0.2 | 0.72 |
| ΔCK, IU/L | 24.87±17.3 | −9.67±13.7 | 34.5 | −9.1 to 58.9 | 0.17 |
Values are expressed as least square means±SE. ADL, activities of daily living; CI, confidence interval; CK, creatine kinase; CR, cardiac rehabilitation; LS mean, least-squares mean; NMES, neuromuscular electrical stimulation; QIS, quadriceps isometric strength. Other abbreviations as in Table 1.
No cardiovascular adverse events occurred in any patient throughout the study period; 2 patients experienced temporary redness and 3 experienced itching at the electrode application site on the legs, but none discontinued NMES. The implementation rate of NMES in this study was investigated using patient self-reports, and was very good at 100%.
In this study, as an adjunct to CR, home-based NMES therapy increased SPPB scores compared with CR alone in frail older adult patients with CHF. No cardiovascular events or changes in BNP levels were observed during NMES therapy, suggesting that NMES therapy could be a viable additional treatment option for improving physical function in this cohort of patients.
Previous studies investigating the effects of NMES in patients with CHF have used exercise tolerance measurements such as peak V̇O2 and the 6-min walk distance as outcomes.29–31 However, it is often challenging to assess exercise tolerance in frail older adult patients with CHF owing to their extremely poor physical function. The SPPB has recently been suggested as a useful tool for assessing daily functioning in clinical trials involving frail older adults.32 It has also been utilized as an indicator to predict clinical outcomes in older adult patients with CHF.25 In this study, 8 weeks of home-based NMES plus CR increased SPPB scores by an average +2.67 points compared with CR alone in frail older adult patients with CHF. Tanaka et al. investigated the clinical efficacy of add-on NMES therapy to exercise-based early CR during hospitalization in frail older adult patients with acute HF.28 Add-on NMES therapy increased the mean SPPB score by +2.3 (range, 0.5–4.1) points compared with a control group receiving early CR alone.33 The settings and NMES intervention period in Tanaka et al.’s study differed from those in our study, but both studies consistently showed the efficacy of NMES therapy in improving motor function in frail older adult patients with CHF. Furthermore, NMES therapy resulted in sufficient improvement in motor function in this study, as the minimal clinically important difference (MCID) in the SPPB score has been reported to be +1.0 points.34
Based on previous studies,35,36 we anticipated an increase in maximal isometric knee extension strength after NMES therapy, but no significant changes were observed. One study investigating the effects of home-based NMES therapy reported that lower extremity muscle strength did not improve in patients with severe CHF and NYHA III and IV.37 Chronic skeletal muscle disorders or wasting is often observed in patients with advanced HF, which may explain this observation.37 The chronic impairment of muscle tissue owing to HF affects muscle and skin nerve receptors, thereby influencing the contractility of the weakened muscle.38 The current study patients were older adult patients with CHF and low motor function who may have had chronic muscle tissue disorders. It is possible that the NMES prescription used in this study had a lower current intensity or shorter treatment duration for patients with decreased muscle contractility. In contrast, the 5-STS time was significantly reduced to a mean of −10.67 s using NMES therapy. The 5-STS is an index of lower limb muscle power expressed as the product of muscle strength and velocity.39 Compared with maximum leg muscle strength, leg muscle power more strongly correlates with climbing stairs, getting up and down, and walking.40 NMES therapy for 30 min/day for 10 weeks has been reported to significantly increase the conductivity of nerves in the lower limbs of patients with diabetes mellitus.41 Therefore, a decrease in 5-STS time and muscle power might be due to an improvement in nerve conduction velocity. We consider that NMES therapy resulted in sufficient improvement in muscle power in our study, as the MCID in 5-STS time has been reported to range from −1.7 to −6.3 s.42,43
The effect of NMES therapy on changes in ADL was limited. Our study patients had a median BI score of 90 points at baseline, indicating a high level of ADL independence. The lack of improvement in ADL may be explained by the ceiling effect of the BI. Many patients reported difficulty with climbing stairs and bathing, which are among the most challenging ADL. To translate improvements in NMES motor function into ADL, we consider it necessary to provide supervised ADL training in addition to NMES.
No cardiovascular events, including worsening HF or changes in BNP levels, were observed during NMES therapy. NMES mainly affects peripheral skeletal muscle, with only an indirect effect on central cardiac function and hemodynamics, suggesting that it may not have a significant effect on BNP.44 To assess potential NMES-related skeletal muscle damage and systemic inflammation in these frail older adult patients with CHF, we analyzed serum CK and hs-CRP levels, which did not significantly increase using NMES therapy, consistent with findings reported previously.45–47 Our findings indicated that home-based NMES therapy is safe for frail older adult patients with CHF.
Study LimitationsThe sample size determination was based on statistical estimates using data from 3 patients, which has statistical limitations. In future studies, we expect to achieve more reliable results by calculating the sample size based on data from a larger number of patients. The sample size calculation predicted a difference of 2.67 points in the adjusted SPPB score and set the statistical power (β) at 0.9. However, considering the baseline score of 7.4 and the ceiling effect, this predicted difference may have been clinically too large. The lack of significant improvement in QIS may be due to a β error resulting from the small sample size. However, the absence of statistical significance does not imply the absence of clinical significance. In future studies, we plan to revalidate these findings with a larger sample size. NMES implementation was evaluated using patient self-reported surveys, which may have introduced uncertainty regarding the actual implementation of NMES. Automatic log storage devices should be used in future studies to determine the exact implementation status. This study targeted older adult patients with HF; however, careful attention should be paid to external validity given the diversity among older adults. One recent study of >10,000 older adult patients with HF (mean age, 83 years; males, 50%) reported that the BI score of older adult patients with HF who were stable at home was 88±19 points.48 In addition, the median KCL score in that study to assess daily living functions was 11 points.48 The values of these indices in that study are similar to those in our study (median BI and KCL in this study: 90 and 11 points, respectively); therefore, there appear to be no major differences in ADL or living conditions between patients in that study and the older adult patients with HF who were living at home in our study. The statistical analyses were performed by an independent statistician, but no other blinding was performed. As it was difficult to set up a sham treatment with the NMES device used in this study, a placebo group could not be included in the study design. To enhance reliability and validity, it is necessary to blind the examiner and/or use sham electrical stimulation in the control group in future studies.
Home-based NMES therapy as an adjunct to CR improved physical function and was safe, suggesting it is a viable additional treatment option for improving physical function in frail older adult patients with CHF.
We thank the staff of the Omote-Jyunkankika Cardiovascular Clinic for facilitating this study.
The authors declare they have no conflicts of interest.
This research was supported by JSPS KAKENHI (grant no. 21K17501).
This study was approved by the Ethics Committee of Tokoha University (Kenshizu 18-19).
Please find supplementary file(s);
https://doi.org/10.1253/circrep.CR-24-0091
