The effects of whole-body muscle stimulation on body composition and strength parameters: A PRISMA systematic review and meta-analysis

Abstract

Background:

This systematic review and meta-analysis set out to determine the efficacy of whole-body muscle electrostimulation on body composition, strength, and muscle power in active and non-active adults (aged ≥18 years).

Method:

This review was reported in accordance with the Protocol Statement of Preferred Reporting Element Guidelines for Systematic Reviews and Meta-Analysis included controlled trials; whole-body electromyostimulation trials with at least 1 exercise and control group; participants >18 years old. Outcome measures were defined as standardized mean differences for muscle mass, body fat mass, strength, and power. Studies were searched in the following electronic databases: PubMed, Web of Science, Scopus, SPORTDiscus, and EMBASE for all articles published up to July 30, 2021. The risk of bias was assessed by 2 independent researchers using the Physiotherapy Evidence Database scale and Grading of Recommendations, Assessment, Development and Evaluations approach. Analyses were performed using the metafor package of the statistical software R (version 4.0.3; R Core Team, 2020). Random effects models, forest, and funnel plots to quantify the asymmetry associated with publication bias were fitted using the metafor library in R. Statistical heterogeneity was assessed using I² statistics.

Results:

In total, 26 studies representing 1183 participants were included (WB-electromyostimulation: n = 586 and control group: n = 597). The mean age of the participants ranged from a minimum of 20.4 to a maximum of 77.4 years old. Interventions lasted a minimum of 4 and a maximum of 54 weeks. Standardized mean difference was 0.36 (95% confidence interval [CI]: 0.16–0.57) for muscle mass, −0.38 (95% CI: −0.62–0.15) for body fat, 0.54 (95% CI: 0.35–0.72) for strength, and 0.36 (95% CI: 0.02–0.71) for power with significant differences between groups (all P < .04). I² revealed low heterogeneity of muscle mass (15%) and power (0%) between trials and medium heterogeneity of body fat (45%) and strength (55%).

Conclusion:

We concluded that WB-electromyostimulation has significant positive effects on muscle mass, body fat, strength, and power.

1. Introduction

It is common knowledge that physical exercise is essential for a healthy lifestyle. The American College of Sports and Medicine recommends regular cardiovascular physical activity of 150 minutes a week with 2 sessions of muscular resistance training of major muscle groups, for the improvement and maintenance of cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy individuals.^[1,2]

Body composition is one of the main indicators of physical health and well-being. In fact, changes in body composition throughout life are related to mortality risk.^[3] According to the American Council on Exercise,^[4] the healthy fat percentage for adults is up to 24% for men and 31% for women, with higher values considered as excess body fat, which is the main cause of obesity and other metabolic and cardiovascular diseases. In addition, the amount of muscle mass and strength play an important role, since as we age, muscle mass tends to decrease and its loss is directly related to a decrease in functional capacity, poorer quality of life, and dependence in older people.^[5] In recent years, new approaches have emerged with the premise of shorter and more efficient workouts. One example is high-intensity interval training programs, which have a positive impact on fat loss and muscle mass gain,^[6] and is nowadays one of the most widely used strategies to improve body composition, as this training method has shown similar results to those obtained after applying a traditional continuous training program of moderate intensity, with 40% less duration.^[7] Another recent approach using technology is whole-body electrostimulation, a time-saving training method used worldwide. Its use has increased in recent years among the population seeking faster results in less time.^[8] The training programs are variable according to the objectives and characteristics of its practitioners, increasing their physical condition and improving body composition with the most outstanding benefits, according to experts and manufacturers.^[9–11] Previous studies^[12–15] have proven the effectiveness of whole-body electromyostimulation (WB-EMS) and its use as an alternative sporting activity, both for those fleeing from conventional methodologies and for athletes who wish to improve their sporting performance through WB-EMS sessions. This training method has also demonstrated improvements in body composition and strength in older people^[16–20] and in active and healthy populations.^[21,22] The main users of this methodology are middle-aged women, who perform 2 workouts a week in order to lose weight, improve health, and gain muscle mass.^[23]

Although there is the possibility of using very diverse protocols, in the current literature the most common is the application of bipolar stimuli as more usual, with a period of stimulation and another of pause (intermittent) with a frequency of 50 to 80 Hz and depth of 300/400 µs. The average duration of this type of training is 20 minutes.^[24]

Given the number of randomized controlled trials in different types of populations seeking to determine the effectiveness of WB-EMS on body composition and strength parameters, in order to improve our knowledge of the use of WB-EMS and its effects we conducted a systematic review and meta-analysis of published studies associating these 2 variables. The primary objective of this study is to determine the efficacy of WB-EMS for composition improvement and secondly to evaluate the effects of this training on some strength parameters. Thus, our primary hypothesis was that WB-EMS enhances the positive effects on lean body mass and fat mass loss. Furthermore, our secondary hypothesis was that WB-EMS generates positive effects on strength and muscle power.

2. Materials and methods

2.1. Literature search and study selection

Studies were searched in the following electronic databases: PubMed, Web of Science, Scopus, SPORTDiscus, and EMBASE for all articles published until July 30, 2021, in the English language only, with no publication status limitations. All randomized clinical trials will be considered. Details of the Cochrane Library are presented in Table 1. It was registered in advance in INPLASY (INPLASY202120050). This review was reported in accordance with the Protocol Statement of Preferred Reporting Element Guidelines for Systematic Reviews and Meta-Analysis.^[25]

Table 1.

Search terms used in literature search.

Category 1	Category 2	Category 3
Whole Body Electro muscle stimulation WB-EMS Whole-body-electro-myo-stimulation Whole-body Electromyostimulation Whole-Body Electromyostimulation training Neuromuscular Electrical Stimulation NMES	Body Composition Fat Mass OR/AND Muscle Mass Strength OR/AND Power	Randomized controlled trial Controlled trial Clinical trial

WB-EMS = whole-body electromyostimulation.

2.2. Eligibility criteria

2.2.1. Study types.

This study included randomized clinical trials investigating the effects of whole-body electrostimulation training on body composition and strength indicators.

2.2.2. Intervention types.

In the intervention group, all subjects must have performed the same exercise protocol with the full-body electrostimulation suit. In the control group (CG), participants must not have performed any training program.

2.2.3. Participant types.

All studies involving trained or untrained participants >18 years of age, with no previous experience with WB-EMS, will be considered.

2.2.4. Outcome measurements.

The primary outcomes of the study are fat-free mass or muscle mass and percentage fat mass or amount of fat measured by electrical bioimpedance, dual-energy X-ray absorptiometry, skinfolds, or anthropometric measurements. Secondary outcomes shall be the maximum strength and muscle power measured in different tests.

2.3. Study selection

Two authors independently screened study titles/abstracts and excluded unrelated studies. They then read the full articles of the remaining studies according to the eligibility criteria. The study selection process is shown in a flowchart according to the PRISMA guidelines (Fig. 1).

Figure 1.

Flow diagram of search process.

2.4. Data collection and management

Two authors independently collected data from all studies based on the data extraction form. This consisted of collecting the following information: title, authors, study design, participant characteristics, type of treatments and controls, outcomes, and other essential data elements. Any discrepancies were discussed with an experienced third author by discussion.

2.5. Missing data dealing with

Once we identified missing or unclear data, we contacted the original authors of the trial to request it. One author responded to our queries.^[26]

2.6. Risk of bias assessment

The risk of bias was assessed by 2 independent researchers using the Physiotherapy Evidence Database (PEDro) scale.^[27] The PEDro scale consists of 11 items, of which only 10 (0/1) are scored. The PEDro scale refers to randomization, allocation concealment, similarity at baseline, blinding of participants, staff and assessors, incomplete outcome data, intention-to-treat analysis, between-group comparison, and measure of variability. In addition, the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) system^[28] will be used to rank the quality of evidence and the strength of the recommendation. The GRADE approach to assessing the quality of evidence involves a 4-point scale including “high,” “moderate,” “low,” and “very low.” It started at the high level and was downgraded to lower levels of evidence when there was a risk of bias, inconsistency, indirectness, imprecision, or publication bias. In addition, the GRADE system^[28] will be used to rank the quality of evidence and strength of recommendation.

2.7. Data synthesis

Results from different studies have been organized in the same way, as effect sizes and corresponding standard errors. The changes in means and standard deviations, as well as sample size, for each group of different studies, have been used to calculate effect sizes as Hodges g using the library esc in the R statistical software (version 4.0.3; R Core Team, 2020; https://cran.r-project.org/web/packages/metafor/index.html).^[29] These effect sizes represent the changes between study groups over the intervention time.

For studies that reported their results as means and standard deviations in pre-post format, changes for both were computed using the following formula as suggested by Higgins (2011)^[30]:

$${\displaystyle \begin{array}{l}{\displaystyle Mea{n}_{diff}=Mea{n}_{post}-Mea{n}_{pre}}\end{array}}$$

$${\displaystyle \begin{array}{l}{\displaystyle S{D}_{diff}=\sqrt{S{D}_{pre}^2+S{D}_{post}^2-(2\ast 0,5\ast S{D}_{pre}\ast S{D}_{post}}}\end{array}}$$

Some studies reported its results as means and interval confidence of changes. For that scenario, using the same reference as previously, standard deviations changes were calculated as

$${\displaystyle \begin{array}{l}{\displaystyle S{D}_{diff}=\frac{\sqrt{N}\ast SDdi{f}_{UL}-SDdi{f}_{LL}}{3,92}}\end{array}}$$

2.8. Statistical analysis

The analyses were conducted using the metafor package of the statistical software R (version 4.0.3; R Core Team, 2020). Random effects models, forest, and funnel plots^[31] for quantifying asymmetry associated with publication bias were adjusted by means of metafor library in R.^[32] Statistical heterogeneity was assessed using I² statistics (low: 0–39%, moderate: 40–59%, substantial: >60%.^[33]

2.9. Ethics and dissemination

Ethical approval is not required as individual patient data will not be collected in this study. We will publish this study in a peer-reviewed journal.

3. Results

Full descriptive details of the included studies are shown in Table 2. Twenty-six studies were included in this systematic review and meta-analysis, with a total of 1183 subjects (randomized control trial: n = 586, CG: n = 597). The range of participants in each group varied from n = 8^[15] to n = 55^[34] in the WB-EMS group and in the CG. The mean age of the participants ranged from a minimum of 20.4 (electrostimulation exercise group [EEG] group) and 20.5 (CG)^[35] to 77.3 (EEG group) and 77.4 (CG).^[36] Interventions lasted a minimum of 4 weeks^[35] and a maximum of 54 weeks.^[37]

Table 2.

Assessment of risk of bias for included studies (n = 27) according to PEDro scale.

References	Eligibility criteria	Random allocation	Allocation concealment	Inter group homogeneity	Blinding subjects	Blinding personnel	Blinding assessors	Participation ≥85% allocation	Intention to treat analysis	Between group comparison	Measure of variability	Total score PEDro
Alvaro Pano-Rodrigues et al (2020)	Y	1	1	1	0	1	1	1	1	1	1	9
Alvaro Pano-Rodrigues et al (2020b)	Y	1	1	1	1	0	1	1	1	1	1	9
Andre Filipovic et al (2019)	Y	1	1	1	0	0	0	1	1	1	1	7
Anja Weissenfels (2018)	Y	1	1	1	0	0	1	1	1	1	1	8
Anja Weissenfels (2019)	Y	1	1	1	0	0	1	1	1	1	1	8
Berger, J (2020)	Y	1	1	1	1	1	0	1	1	1	1	9
Carina Zink-Ruckel et al (2021)	Y	1	1	1	1	0	0	1	1	1	1	8
Evangelista A, et al (2019)	Y	1	1	1	0	0	1	1	1	1	1	8
Evangelista A, et al (2021)	Y	1	1	1	0	0	0	1	1	1	1	7
Florian Mick (2018)	Y	1	1	1	0	0	0	0	1	1	1	6
Francisco J. Amaro-Gahete et al (2018)	Y	1	1	1	0	0	0	0	1	1	1	6
Francisco J. Amaro-Gahete et al (2018b)	Y	1	1	1	0	0	1	0	1	1	1	7
Hyeng-Kyu Park et al (2020)	Y	1	1	1	1	1	1	1	1	1	1	10
Jiyoun Kim et al (2020)	Y	1	0	1	1	0	0	1	1	1	1	7
Katharina Wittmann et al (2016)	Y	1	1	1	0	0	1	1	1	1	1	8
Sebastian Willert et al (2019)	Y	1	1	1	0	0	1	1	1	1	1	7
Simon von Stengel (2015)	Y	1	1	1	1	0	0	1	1	1	1	8
Stefano D’ottavio (2019)	Y	1	1	1	0	0	0	1	1	1	1	7
Sunhee Park et al (2021)	Y	1	1	1	1	0	0	1	1	1	1	8
L. Jurado-Fasioli et al (2019)	Y	1	1	1	0	0	1	1	1	1	1	8
Ulrike Dörmann (2019)	Y	1	1	1	1	0	0	0	1	1	1	7
Wolfgang Kemmler et al (2010)	Y	1	0	0	0	0	1	1	1	1	1	6
Wolfgang Kemmler et al (2013)	Y	1	1	1	1	0	1	0	1	1	1	8
Wolfgang Kemmler et al (2016)	Y	1	1	1	0	0	1	1	1	1	1	8
Wolfgang Kemmler et al (2018a)	Y	1	1	1	0	0	1	0	1	1	1	7
Wolfgang Kemmler et al (2018b)	Y	1	1	1	0	0	1	1	1	1	1	8
Yong-Seok Jee (2019)	Y	1	0	1	1	0	0	1	0	1	1	6

PEDro = Physiotherapy Evidence Database.

3.1. Characteristics of the studies and participants

In relation to the training volume of the intervention groups, there was a variation of 1 training per week and a maximum of 3^[38–42] with a duration between 8^[13] and 40 minutes.^[39] Among the 26 studies included, 11 have been conducted with women only^{[16,20,35–39,42–45]} and 7 were only with men.^{[13,26,41,46–50]} Two interventions have been performed in patients with chronic low back pain,^[34,51] 4 in people with sarcopenia and obese,^{[20,39,46,52]} 4 in pre- and post-menopausal women^[43–45,53] and 1 in older people with osteoarthritis.^[38] Two studies have worked with athletes.^[13,15] All other studies have been conducted in healthy, untrained, or inactive people.

3.2. Type of interventions and groups

Most studies have isolated the application of WB-EMS in the intervention group, while 2 studies added protein supplementation^[26,36] and another study restricted energy intake.^[45] In addition, in 2 other interventions, the authors included groups with different stimulus frequencies (Hz).^[54,55] A single article worked with groups with different stimulus intensities.^[41]

The training comparison groups were performed with the same exercise program as the WB-EMS group, with the exception of 1 study where the WB-EMS was compared with a CG without any exercise program ^[51] and 4 other articles in which the comparison group performed other training programs parallel to the WB-EMS group.^{[21,37,45,56]}

3.3. WB-EMS protocol used

The WB-EMS protocols (i.e., the pulse parameters) were fairly homogeneous across the studies. All studies applied low-frequency bipolar protocols of 80 to 85 Hz with a rectangular pulse waveform, with the exception of 4 studies that applied 20 Hz, 55 Hz, and 50 Hz.^{[43,44,54,55]} The pulse was specified between 200 and 400 ms. All studies combined WB-EMS with dynamic voluntary movements, with the exception of 3 studies that applied isometric exercises.^[38,41,54]

The intensity of the application pulse has been predominantly prescribed according to Borg rating of perceived exertion,^[57] ranging from consistently strong^[5] to very strong^[7] on the 10-point scale and 15 to 19 on the 20-point scale. However, 5 working groups used a maximum impulse tolerance approach.^{[35,38,39,48]} and a single article worked the stimulus intensity between 80 and 100 mA.^[21] It remains important to note that none of the studies reported negative side effects of WB-EMS applications Table 3. Study and intervention characteristics of the included articles (n = 26).

Table 3.

Study and intervention characteristics of the included articles (n = 26).

References	Sample size (n); population; AEG (mean ± SD)	Groups	Intervention	Mains outcomes
Alvaro Pano-Rodrigues et al (2020)	Postmenopausal women N = 32 61,38 + −3,95 yr old	WB-EMS = 16 e.g. = 16	10 wk. 2 sessions × 20 min per wk. WB-EMS group performed 20 repetitions of 3 exercises (squat, deadlift, and bench press). Stimulus of 55 Hz, 200–400 μs with RPE 15 (Borg Scale-20). e.g. performed the same exercise protocol as the WB-EMS group (without the EMS).	Body composition, power, and speed
Alvaro Pano-Rodrigues et al (2020b)	Postmenopausal women N = 34 61,4 + −4,0 yr old	WB-EMS = 17 e.g. = 17	10 wk. 2 sessions × 20 min per wk. WB-EMS group performed 20 repetitions of 3 exercises (squat, deadlift, and bench press). Stimulus of 55 Hz, 200–400 μs with RPE 15 (Borg Scale-20). e.g. performed the same exercise protocol as the WB-EMS group (without the EMS).	Balance, strength, flexibility, and agility
Andre Filipovic et al (2019)	Soccer players N = 28 WB-EMS = 24,4 + −4,2 yr old TG = 21,1 + −1,9 yr old CG = 23,6 + −3,9 yr old	WB-EMS = 10 e.g. = 10 CG = 8	7 wk. 2 sessions × 8 min per wk. WB-EMS group performed 3 × 10 jump squats. Stimulus of 85 Hz, 350 μs with RPE 16–19 (Borg Scale-20). e.g. performed the same exercise protocol as the WB-EMS group (without the EMS). CG did not exercise.	Body composition, muscle biopsy, and strength
Anja Weissenfels (2018)	Low back pain patients N = 30 WB-EMS = 54,6 + −5,7 CG = 59,4 + −7,7	WB-EMS = 15 CG = 15	12 wk. 1 session × 20 min per wk. WB-EMS group performed light movements to relieve back pain. Stimulus of 85 Hz, 350 μs with RPE 5–7 (Borg10 Scale). Control group did not exercise.	Low back pain intensity, isometric maximum force of extension, and flexion of the trunk
Anja Weissenfels (2019)	Low back pain patients N = 110 WB-EMS = 57,4 + −7,6 CG = 54,4 + −7,4	WB-EMS = 55 e.g. = 55	12 wk. 1 session × 12–20 min per wk. WB-EMS group performed 6 specific exercises for the trunk muscles. Stimulus of 85 Hz, 350 μs with RPE 5–7 (Borg Scale10). e.g. performed the same exercise protocol as the WB-EMS group (without the EMS)	Intensity of lumbar pain, maximum isometric strength of extension, and flexion of the trunk
Berger, J (2020)	N = 51 24,9 + −3,9	WB-EMS = 19 WB-EMS20 = 14 CG = 14	10 wk. 1.5 sessions × 20 min per wk. WB-EMS group performed 10–12 repetitions of 10 selected exercises. Stimulus of 20–85 Hz, 350 μs with RPE 6–7 (Borg Scale 10). e.g. performed the same exercise protocol as the WB-EMS group (without the EMS). Control group did not exercise.	Jump, speed, and strength
Carina Zink-Ruckel et al (2021)	Amateur golfers N = 54 WB-EMS = 42,7 + −16,6 yr old CG = 43,0 + −13,4 yr old	WB-EMS = 27 e.g. = 27	16 wk. 1 session of 20 min per wk. WB-EMS group performed 6–8 repetitions of light movements. Stimulus of 80 Hz, 350 μs with RPE 6–7 (Borg Scale-10). e.g. performed the same exercise protocol as the WB-EMS group (without the EMS)	Intermuscular adipose tissue and volume and interfacial muscle tissue
Evangelista A, et al (2019)	N = 58 WB-EMS = 25,5 + −6,1 e.g. = 25,1 + −3,2 CG = 27,1 + −4,1	WB-EMS = 21 e.g. = 21 GC = 16	8 wk. 2 sessions × 20 min per wk. WB-EMS group performed 3 sets of 8 to 12 repetitions maximum of 3 exercises. Stimulus of 80–85 Hz, 350 μs with RPE 5–8 (Brog Scale-10). e.g. performed the same exercise protocol as the WB-EMS group (without the EMS). Control group did not exercise.	Maximum strength and muscle thickness
Evangelista A, et al (2021)	Inactive older men N = 20 75,1 + −6,58 yr old	WB-EMS = 10 e.g. = 10	6 wk. 2 sessions × 20 min per wk. WB-EMS group performed 8 bodyweight exercises in 2 sets × 8 repetitions. Stimulus with 85 Hz, 350 μs with RPE 7–8 (Borg-10 Scale). e.g. performed the same exercise protocol as the WB-EMS group (without the EMS).	Body composition and physical condition
Florian Mick (2018)	Sport students N = 18 WB-EMS = 22,8 + −3,0 CG = 22,8 + −2,5	WB-EMS = 8 e.g. = 8	8 wk. 2 sessions × 20 min per wk. WB-EMS group performed 5 exercises (2 strength + 3 dynamic jump exercises). Stimulus with 85 Hz, 350 μs, and 70% of the maximum tolerance capacity. e.g. performed the same exercise protocol as the WB-EMS group (without the EMS).	Strength, power, jump, and speed
Francisco J. Amaro-Gahete et al (2018)	Middle-aged adult N = 65 3,5 + −4,9 yr old	WB-EMS = 19 e.g.=16 PAR = 16 GC = 14	12 wk. 2 sessions × 20–30 min per wk. WB-EMS group performed 2 different sessions: HIIT with long intervals and HIIT with short intervals. Stimulus with 15–75 Hz, 200–400 μs, and 80–100 mA of intensity. e.g. performed the same exercise protocol as the WB-EMS group (without the EMS). PAR performed the WHO-recommended exercise protocol 3 × per wk. Control group did not exercise.	Body composition, physical activity assessment, and dietary intake assessment
Francisco J. Amaro-Gahete et al (2018b)	Amateur runners N = 12 27,0 + −6,8 yr old	WB-EMS = 6 e.g. = 6	6 wk. 1 session × 16–20 min per wk. WB-EMS group performed wave periodization model The training sessions were divided into 4 parts: warm-up (phase A), strength training part (phase B), high-intensity interval power training part (phase C), and high-intensity training part. (Phase D). Stimulus of 12–90 Hz, 200–400 μs, and RPE 10–17 (Borg-10 scale). e.g. kept up his race training routine	VO2 max, running economy, muscle power, and body composition
Hyeng-Kyu Park et al (2020)	Young women N = 23 WB-EMS = 23,5 + −4,2 CG = 25,2 + −5,7	WB-EMS = 11 e.g. = 12	6 wk. 3 sessions × 20 min per wk. WB-EMS group performed low-intensity strength exercises. 80 Hz stimulus. e.g. performed the same exercise protocol as the WB-EMS group (without the EMS).	Body composition; laboratory lipid profile; magnetic resonance imaging; assessment of isokinetic muscle function; evaluation of balance function and cardiopulmonary function test
Jiyoun Kim et al (2020)	Obese older women N = 25 EC = 71,75 + −2783 CG = 70,38 + −2,93	WB-EMS = 13 e.g. = 12	8 wk. 3 sessions × 40 min per wk. WB-EMS group performed anaerobic and aerobic exercises with dance. Stimulus of 85 Hz, 350 μs, and intensity of 60–80% of the maximum tolerance capacity. e.g. performed the same exercise protocol as the WB-EMS group (without the EMS).	Body composition, biomarkers, and caloric intake/expenditure
Katharina Wittmann et al (2016)	Sarcopenic obese older women N = 67 EC = 77,3 + −4,9 yr old ECP = 76,4 + −2,9 yr old CG = 77,4 + −4,9 yr old	WB-EMS = 24 WB-EMSP = 21 CG = 22	26 wk. 1 session × 20 min per wk. WB-EMS group performed a guided video program in a sitting position with light movements for the lower and upper body. Stimulus with 85 Hz, 350 μs, and RPE 5–6 (Borg-10 Scale). e.g. performed the same exercise protocol as the WB-EMS + protein supplementation group. Control group did not exercise.	MetS Z-score, waist circumference; mean arterial pressure, triglycerides, fasting plasma glucose, and high-density lipoprotein cholesterol
Sebastian Willert et al (2019)	Overweight premenopausal women N = 90 EC = 38,4 + −8,0 yr old PA = 34,4 + −8,3 yr old CG = 35,3 + −7,4 yr old	WB-EMS = 30 PA = 30 CG = 30	16 wk. 1.5 sessions × 20 min per wk. WB-EMS group performed 2 sets of 6–8 repetitions of light exercises. Stimulus with 85 Hz, 350 μs, and RPE 5–7 (Brog-10 Scale). PA group increases their normal daily activity by 250 kcal/d and reduces energy intake by 250 kcal/d. CG reduced energy intake by 500 kcal/d.	Lean body mass and body fat mass
Simon von Stengel (2015)	Osteopenia women N = 60 WB-EMS = 74,7 + −3,7 CG = 74,7 + −4,4	WB-EMS = 32 e.g. = 28	54 wk. 1.5 sessions × 18–19 min per wk. WB-EMS group performed 10–14 exercises 1–2 sets of 8 repetitions. Stimulus with 85 Hz, 350 μs, and RPE 14–16 (Borg Scale-20). e.g. performed the same exercise protocol as the WB-EMS group (without the EMS).	Bone mineral density, lean body mass, and grip strength
Stefano D’ottavio (2019)	N = 22 26,7 + −3,15	WB-EMS = 6 WB-EMS50 = 8 e.g. = 8	6 wk. 2 sessions × 20 min per wk. WB-EMS group performed 9 isometric exercises. Stimulus with 50–85 Hz, 350 μs, and RPE 14–16 (Borg-20 Scale). WB-EMS50 performed the same exercise protocol with a 50 HZ stimulus. e.g. performed the same exercise protocol as the WB-EMS and WB-EMS50 groups (without the EMS).	Strength and power
Sunhee Park et al (2021)	Elderly women with knee osteoarthritis N = 75 EC = 65,68 + −3,24 ISO = 66,88 + −4,61 CON = 68,04 + −4,16	EC = 25 ISO = 25 GC = 25	8 wk. 3 sessions × 20 min per wk. WB-EMS group performed isometric exercises with the duration of the device contraction (6 s). Stimulus with 85 Hz, 350 μs, and 60–70% of maximum tolerance capacity. ISO e.g. performed the same exercise protocol as the WB-EMS group (without the EMS). CG performed meditation and gentle stretching.	Body composition, biomarkers, strength, and degree of knee injury and osteoarthritis
L. Jurado-Fasioli et al (2019)	Sedentary adults N = 67 53,3 + −5,0	WB-EMS = 17 HIIT = 18 PAR = 17 GC = 15	12 wk. 2 sessions × 20–30 min per wk. WB-EMS group performed 2 different types of training: HIIT with long breaks and HIIT with short breaks. Stimulus of 15–75 Hz, 200–400 μs, and 80–100 mA of intensity. HIIT group performed the same exercise protocol as the WB-EMS group (without EMS). The PAR group performed the exercise protocol recommended by the WHO. GC did not exercise.	Body composition and caloric intake
Ulrike Dörmann (2019)	Active women N = 22 20,5 + −2,3 yr old	EC = 11 e.g. = 11	4 wk. 2 sessions × 20 min per wk. WB-EMS group trained with different exercises. Session a) strength b) speed and jump exercises. Stimulus with 85 Hz, 350 μs, and 70% of the maximum tolerance capacity. e.g. performed the same exercise protocol as the WB-EMS group (without the EMS).	Jump and speed test Strength and power
Wolfgang Kemmler et al (2010)	Postmenopausal women N = 30 64,5+ −5,5 yr old	WB-EMS = 15 e.g. = 15	14 wk. 1.5 sessions × 20 min per wk. WB-EMS group performed 1–2 sets of 10 exercises with 6–8 repetitions and conventional group training. Stimulus with 85 Hz, 350 μs, and RPE 6–7 (Borg-10 Scale). e.g. only did the conventional workouts.	Body composition, strength, and metabolic rest rate
Wolfgang Kemmler et al (2013)	Older women N = 60 EC = 74,4 + −3,7 yr old CG = 74,7 + −4,4 yr old	WB-EMS = 32 WB-EMS = 28	54 wk. 1.5 sessions × 18 min per wk. WB-EMS group performed exercises of low intensity and amplitude. Stimulus with 85 Hz, 350 μs, and RPE 14–16 (Borg-20 Scale). WB-EMS performed 1 session × 60 min for 10 wk interspersed with 10 wk off during the 54 wk.	Body composition and strength
Wolfgang Kemmler et al (2016)	Untrained men N = 42 EC = 41,9 + −6,4 HIIT = 43,7 + −6,1	EC = 22 HIIT = 20	16 wk. 1.5 sessions × 20 min per wk. WB-EMS group performed 1–2 series of 12 exercises and 6–8 repetitions. Stimulus with 85 Hz, 350 μs, and RPE 6–7 (Borg-10 Scale). HIIT did 2–3 sessions a wk, 10–13 exercises. Two sets of 15 repetitions in high-intensity interspersed training	Body composition and strength
Wolfgang Kemmler et al (2018a)	Older men with sarcopenic obesity N = 100 EC = 77,1 + −4,3 PC = 78,1 + −5,1 CG = 76,9 + −5,1	WB-EMS = 33 WB-EMSP = 33 GC = 34	16 wks. 1.5 sessions × 20 min per week. WB-EMS group performed 1–2 series of 12 exercises and 6–8 repetitions. Stimulus with 85 Hz, 350 μs, and RPE 6–7 (Borg-10 Scale). e.g. performed the same exercise protocol as the WB-EMS group (without the EMS). PG was protein supplemented without exercising; GC did not exercise.	Body composition, sarcopenia Z-score, and grip strength
Wolfgang Kemmler et al (2018b)	Older men with sarcopenic obesity N = 67 77,0 + −5,0 yr old	WB-EMS = 33 e.g. = 34	16 weeks. 1.5 sessions × 20 min per week. WB-EMS group performed 1–2 series of 12 exercises and 6–8 repetitions. Stimulus with 85 Hz, 350 μs, and RPE 6–7 (Eschar Borg-10). e.g. performed the same exercise protocol as the WB-EMS group (without the EMS).	Body composition, strength, MetS, and kidney function
Yong-Seok Jee (2019)	Healthy college men N = 54 WB-EMS80hz = 25,00 + −0,82 WB-EMS60Hz = 23,25 + −2,22 WB-EMSHz50 = 24,00 + −3,00 e.g.=28,33 + −2,52	WB-EMS80Hz = 13 WB-EMS60Hz = 14 WB-EMS50Hz = 12 e.g. = 13	6 wk. 3 sessions × 20 min per wk. WB-EMS group performed isometric exercises, 12 repetitions during the impulse phase (6 s). Stimulus with 85 Hz, 350 μs, and 50%, 60%, and 80% of the maximum tolerance capacity. WB-EMS80 performed the exercises with an intensity of 80%; WB-EMS60 performed the exercises with an intensity of 60%; WB-EMS50 performed the exercises with an intensity of 50%. e.g. performed the same exercise protocol as the WB-EMS group (without the EMS).	Body composition, strength, and thigh circumference

CG = control group, EC = exercise group, EMS = electromyostimulation, HIIT = high intensive interval training group, PAR = physical activity recommendation group, RPE = rated perceived exertion, WB-EMS = whole-body electromyostimulation group.

3.4. Risk of bias assessment

Table 2 shows the risk of bias for included studies according to the PEDro scale. According to the PEDro scale, most of the studies have a high methodological quality (22 of 26 included articles). Four studies were rated as studies of moderate methodological quality.^{[15,16,41,58]}

The quality of evidence according to the GRADE approach was “low,” meaning that “Further research is very likely to have a major impact on our confidence in the estimate of effect and is likely to change the estimate” or “Our confidence in the estimate effect is limited: The true effect may be substantially different from the estimate of the effect.”^[59,60]

3.5. Effects of WB-EMS on muscle mass

Fourteen studies with the same number of WB-EMS groups evaluated the effect of WB-EMS on muscle mass (Fig. 2). In summary, the WB-EMS intervention produced significant effects (P = .002). The standardized mean difference (SMD) between all groups was = 0.36; 95% confidence interval (CI): 0.16–0.57 with a low level of heterogeneity between trials (I² = 15%).

Figure 2.

Forest plot of the results of the meta-analysis on muscle mass. Data are shown as pooled standardized mean differences (SMD) with 95% CI for changes in the WB-EMS and non-EMS training groups. CI = confidence interval, WB-EMS = whole-body electromyostimulation.

In summary, the Funnel plot (Fig. 3) provided evidence of small study bias.^[61] Egger regression test^[31] for funnel plot asymmetry did not indicate significant asymmetry (P = .7).

Figure 3.

Funnel plot of WB-EMS studies addressing muscle mass. WB-EMS = whole-body electromyostimulation.

3.6. Effects of WB-EMS on body fat

Seventeen studies with the same study groups determined the effect of EEG on total body fat mass (Fig. 4). In summary, EEG significantly (P = .003) affected total body fat mass. The SMD was −0.38 95% CI: −0.62–0.15. A moderate level of heterogeneity was observed between trials (I² = 45%, P = .02).

Figure 4.

Forest plot of the results of the meta-analysis on body fat. Data are shown as pooled standardized mean differences (SMD) with 95% CI for changes in the WB-EMS and non-EMS training groups. CI = confidence interval, WB-EMS = whole-body electromyostimulation.

Figure 5 shows the Funnel plot of the WB-EMS on the effects of total body fat that provided no evidence of significant bias.^[61] Egger regression test^[31] for funnel plot asymmetry did not indicate significant asymmetry (P = .7).

Figure 5.

Funnel plot of WB-EMS studies addressing body fat. WB-EMS = whole-body electromyostimulation.

3.7. WB-EMS effects on maximal strength

Twenty-eight maximal strength variables (knee extension, handgrip, trunk flexion, and extension strength) from 19 studies determined the effect of whole body electromyostimulation on maximal strength (Fig. 6). In summary, WB-EMS very significantly (P < .0001) affected strength. The pooled estimate from random effects analysis was SMD 0.54 95% CI: 0.35–0.72. A moderate level of heterogeneity was observed between trials (I² = 55%, P = .02).

Figure 6.

Forest plot of the results of the meta-analysis on strength. Data are shown as pooled standard mean differences (SMD) with 95% CI for changes in the WB-EMS and non-EMS training groups. CI = confidence interval, WB-EMS = whole-body electromyostimulation.

Figure 7 shows the Funnel plot of the WB-EMS on the effects of total body fat that provided no evidence of significant bias.^[61] Egger regression test^[31] for funnel plot asymmetry did not indicate significant asymmetry (P = .3).

Figure 7.

Funnel plot of WB-EMS studies on maximal strength. WB-EMS = whole-body electromyostimulation.

3.8. Effects of EEG on muscle power

For muscular power, a total of 7 works determined the effect of the WB-EMS (Fig. 8). Whole body electromyostimulation affected muscle power significantly (P = .04). The standard mean difference was SMD 0.36 95% CI: 0.02–0.71. A very low level of heterogeneity was observed between trials (I² = 0%, P = .52).

Figure 8.

Forest plot of the results of the meta-analysis on muscle power. Data are shown as pooled standardized mean differences (SMD) with 95% CI for changes in the WB-EMS and control groups. CI = confidence interval, WB-EMS = whole-body electromyostimulation.

Figure 9 shows the Funnel plot of the WB-EMS on the effects of total body fat that provided no evidence of significant bias.^[61] Egger regression test^[31] for funnel plot asymmetry did not indicate significant asymmetry (P = .4).

Figure 9.

Funnel plot of WB-EMS studies addressing muscle power. WB-EMS = whole-body electromyostimulation.

4. Discussion

The existing literature is not unanimous on the effects of WB-EMS training on body composition. Mainly on body fat reduction, there are studies reporting significant differences for the WB-EMS group^[16,52,56] and others in which its effects were not significant.^[42,62]

This heterogeneity, to a large extent, can be explained by 4 main factors: application of different types of intensity (from low to high intensity); different training volumes (from 20 to 90 minutes per week, from 6 weeks to 12 months of intervention); little or no control over participants’ caloric intake in most studies; and different population types (WB-EMS, trained participants, healthy, obese, cancer patients, and others).

This heterogeneity in the literature is again evidenced by our review work with meta-analysis, where our findings indicated that WB-EMS significantly affected participants’ decrease in body fat, which differs from that reported in another review conducted in 2021 on the effects of WB-EMS on body composition and strength in untrained.^[24] In this case, this difference can possibly be explained by the fact our study included a larger number of studies because it is more current. Nor have we differentiated between trained and untrained populations.

On the other hand, there is more consensus in the literature regarding alterations in muscle mass. Several studies report a positive effect with significant differences for the increase in muscle mass for the WB-EMS training group.^{[26,37,39,46]} In our review, in which we have included 14 articles, the WB-EMS had significant effects (P = 000.2) with the SMD (0.36; 95% CI: 0.16–0.57). In the 2021 revision by Wolfgang Kemmler,^[24] the results are even more significant (1.23; 95% CI: 0.71–1.76) for the intervention group in a total of 13 groups analyzed. In addition, WB-EMS appears to be very effective in maintaining muscle mass in the process of weight loss (calorie restriction), as demonstrated by these 2 studies.^[45,63]

There is some evidence of a dose/response effect of pulse intensity that affects body fat more than muscle mass.^[41] Few studies^[45,62] have taken into account different energy intakes (restriction from 250 to 500 kcal), while others have supplemented participants with different doses of whey protein,^{[26,46,64,65]} nutritional conditions decisive for weight loss (negative energy balance) and for muscle mass gain, respectively (minimum adequate protein intake).^[40,66] Regarding the type of sessions, most of the studies conducted and included in the reviews have used the same programs and types of exercises, being used for both objectives. In other words, there was no specific approach for each training objective. These factors may go some way to explain the less obvious effects on body fat loss, which contrasts with the greater evidence for effects on muscle mass. In summary, more studies of high methodological quality are still needed to determine the scientific evidence for the use of WB-EMS for body fat loss. Although the factors that determine strength production are several, we know that the amount of muscle mass (cross-sectional area) is one of the main ones.^[67] Thus, the effectiveness of WB-EMS in increasing muscle mass also seems to be accompanied by an increase in maximal strength. In the review by Wolgang Kemmler (2021), significant effects on maximal extension and trunk extension strength were observed, with a total of 10 papers included in the meta-analysis. This result is in agreement with our systematic review and meta-analysis, where we found statistically significant differences for the intervention group (P < .0001), with a SMD of 0.54; 95% CI: 0.35–0.92, in a total of 19 included papers and 28 analyzed groups with different muscle groups evaluated.

In contrast, we also analyzed muscle power, where we also found statistically significant differences in favor of the intervention group in a total of 7 articles, although with a lower degree of significance (P = .036) and smaller effect (0.36; CI 95%:0.01–0.70). These results are, in part, inconsistent with a mini-review conducted with 5 studies and 112 participants, where the authors found statistically significant differences for maximal strength, but not for power.^[68] It should be noted that this mini-review was conducted with moderately trained young adults, unlike ours, where we included all types of populations.

4.1. Limitations

The findings of the study should be considered in the context of the following limitations:

• •The criteria used to judge the level of evidence have not yet been standardized. Different authors of systematic reviews use different criteria, and the same author may use different criteria in different studies.^[69]
• •The use of different criteria is related to the decision to include only randomized clinical trials or to also consider studies of low methodological quality, where the measurement scales may also vary.^[70]

The best method for assessing the risk of bias has not been determined. The search strategy only looked for articles in English, which implied a risk of bias, as publishing significant results is easier than publishing non-significant results, and the latter is more likely to appear in national journals written in languages other than English.^[71]

5. Conclusion

This systematic review and meta-analysis provided further evidence of the positive effects of WB-EMS on body composition and strength parameters. The hypotheses we defined for this work have been confirmed. WB-EMS showed a significant effect on muscle mass and on the reduction of body fat. The effects on the secondary hypotheses were also confirmed with a significant effect on maximal strength and muscular power.

Being a sample with these characteristics, extrapolation of the results should only be made to cohorts in terms of age (middle age or older) and level of physical activity (untrained or less trained). Furthermore, it should be considered that the present results can only be attributed to WB-EMS protocols that focus on moderate to high impulse intensity and low to moderate voluntary workload (functional bodyweight exercises).

Further systematic reviews and meta-analyses with more studies in young and healthy populations, will be necessary for further evidence and generalization of the results obtained with this work.

Author contributions

Conceptualization: Luiz Rodrigues-Santana, Louro Hugo, Miguel A. Hernández-Mocholí, José C. Adsuar.

Data curation: Miguel A. Hernández-Mocholí, Nicolás Contreras-Barraza.

Funding acquisition: José C. Adsuar.

Methodology: Nicolás Contreras-Barraza.

Software: Miguel A. Hernández-Mocholí.

Writing – original draft: Luiz Rodrigues-Santana, Jorge Pérez-Gómez, Miguel A. Hernández-Mocholí, Jorge Carlos-Vivas, Pilar Saldaña-Cortés, Nicolás Contreras-Barraza, José C. Adsuar.

Writing – review & editing: Luiz Rodrigues-Santana, Jorge Pérez-Gómez, Miguel A. Hernández-Mocholí, Jorge Carlos-Vivas, Pilar Saldaña-Cortés, Nicolás Contreras-Barraza, José C. Adsuar.