**Resistance Training with Blood Flow Restriction Compared to Traditional Resistance Training on Strength and Muscle Mass in Non-Active Older Adults: A Systematic Review and Meta-Analysis**

**Darío Rodrigo-Mallorca 1 , Andrés Felipe Loaiza-Betancur 2 , Pablo Monteagudo 1,3 , Cristina Blasco-Lafarga 1, \* and Iván Chulvi-Medrano 1**


**Citation:** Rodrigo-Mallorca, D.; Loaiza-Betancur, A.F.; Monteagudo, P.; Blasco-Lafarga, C.; Chulvi-Medrano, I. Resistance Training with Blood Flow Restriction Compared to Traditional Resistance Training on Strength and Muscle Mass in Non-Active Older Adults: A Systematic Review and Meta-Analysis. *Int. J. Environ. Res. Public Health* **2021**, *18*, 11441. https://doi.org/10.3390/ ijerph182111441

Academic Editors: Amelia Guadalupe Grau and Paul B. Tchounwou

Received: 1 September 2021 Accepted: 26 October 2021 Published: 30 October 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

**Abstract:** Low-intensity training with blood flow restriction (LI-BFR) has been suggested as an alternative to high-intensity resistance training for the improvement of strength and muscle mass, becoming advisable for individuals who cannot assume such a load. The systematic review aimed to determine the effectiveness of the LI-BFR compared to dynamic high-intensity resistance training on strength and muscle mass in non-active older adults. A systematic review was conducted according to the Cochrane Handbook and reportedly followed the PRISMA statement. MEDLINE, EMBASE, Web of Science Core Collection, and Scopus databases were searched between September and October 2020. Two reviewers independently selected the studies, extracted data, assessed the risk of bias and the quality of evidence using the GRADE approach. Twelve studies were included in the qualitative synthesis. Meta-analysis pointed out significant differences in maximal voluntary contraction (MVC): SMD 0.61, 95% CI [0.10, 1.11], *p* = 0.02, I <sup>2</sup> 71% *p* < 0.0001; but not in the repetition maximum (RM): SMD 0.07, 95% CI [−0.25, 0.40], *p* = 0.66, I <sup>2</sup> 0% *p* < 0.53; neither in the muscle mass: SMD 0.62, 95% CI [−0.09, 1.34], *p* = 0.09, I <sup>2</sup> 59% *p* = 0.05. Despite important limitations such as scarce literature regarding LI-BFR in older adults, the small sample size in most studies, the still differences in methodology and poor quality in many of them, this systematic review and meta-analysis revealed a positive benefit in non-active older adults. LI- BFR may induce increased muscular strength and muscle mass, at least at a similar extent to that in the traditional high-intensity resistance training.

**Keywords:** hypertrophy; katsu; low-intensity training; occlusive exercise; sarcopenia

### **1. Introduction**

The number of people over 60 years of age is increasing rapidly worldwide due to the increase in life expectancy and the decrease in the fertility rate. According to World Health Organization (WHO) data [1], the world population in this group of age is expected to reach 2 billion by 2050, reflecting an increase of 900 million from the 1.1 billion dated in 2015, up to the 22% of the total population compared to the current 12%. Maintenance of quality of life and prevention from disability (i.e., larger health span more than just life span), is of outermost importance and a current public health challenge [2].

In this scenery, physical activity (PA) has widely been confirmed to counteract the deterioration associated with aging and the sedentary behaviors intrinsic to these last stages of life [3,4]. PA reduces the risk of mortality and chronic pathologies [5,6]. It also helps to prevent dynapenia (decreased muscle strength) and sarcopenia (loss of strength, a decline in muscle mass, and final severe functional capacity impairment in the older adults, in this order) [7]. More specifically, physical exercise, especially strength training [8], emerges as a non-pharmacological tool in the management of this impairment of muscle function and structure which frequently leads older adults to the frailty syndrome [9]. Sarcopenia is indeed an emergency and expensive comprehensive health issue related to many other non-communicable diseases, such as larger number of falls and fractures [5,10,11], osteoporosis [12], diabetes [13], overall disability [12], but also cognitive impairment [14], reduced daily living autonomy, frequent hospitalization, and, finally, larger comorbidity and risk of death (See Cruz-Jentoft et al. [15], for Sarcopenia: revised European consensus on definition and diagnosis).

Resistance training is widely accepted as the most common strategy in this nonpharmacological approach to sarcopenia treatment [8]. Notwithstanding, in the last decade, research has revealed alternative training proposals to traditional high-intensity strength training (>70% RM), such as training with blood flow restriction (BFR), which consists of applying partial peripheral vascular occlusion during low-load strength training (20%–30% of 1RM), causing a local hypoxia effect in the muscle. Recent systematic reviews have analyzed responses on athletic population profiles [16,17] and active adults across the age spectrum (20–80 years) [18–20] indicating that BFR is a similarly effective intervention to high-intensity training in stimulating strength and muscle mass gains. Despite increasing research, the literature on BFR in older adults remains sparse on these issues and the subject's functional status and moderating variables (pressure, cuff size, application volume), making further research necessary to strengthen the evidence on the efficacy of BFR in older adults.

Therefore, the present systematic review and meta-analysis aimed to determine the effectiveness of the low-intensity resistance training with blood flow restriction compared to dynamic high-intensity resistance training on strength and muscle mass in non-active older adults.

#### **2. Materials and Methods**

#### *2.1. Protocol and Registration*

This systematic review was conducted according to the Cochrane Handbook [21] and reported following the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) declaration [22]. It was registered in the International Prospective Register of Systematic Reviews (PROSPERO registration number: CRD42020214901).

#### *2.2. Information Sources and Search*

We conducted a systematic search according to Chapter 4 of the Cochrane Handbook [21]. MEDLINE, Web of Science Core Collection, Scopus, and EMBASE databases were searched between September and October 2020. The search strategy applied was the following: old OR eld OR sarcopenic OR frail AND blood flow restriction OR occlusive training OR vascular occlusion OR kaatsu OR ischemic training.

#### *2.3. Eligibility Criteria and Study Selection*

Selection criteria were built based on the participants, intervention, comparators, outcomes, study design (PICOs) approach acronym [21] as follows.

Participants: Participants over 65 years, physically inactive, and characterized as healthy by the authors, defined as not achieving 150 min of moderate-to-vigorous-intensity physical activity per week or 75 min of vigorous-intensity physical activity per week or an equivalent combination of moderate and vigorous-intensity activity [23,24].

Intervention: Low-intensity blood flow restriction training (LI-BFR), based on the restriction of afferent and efferent blood flow during the performance of a low-intensity dynamic resistance exercise (20–50% of 1RM), causing a local hypoxia effect on the muscle using a pneumatic pressure cuff placed in the proximal region of the limb [25].

Comparators: Resistance training (RT) interventions were considered as any form of physical activity that is designed to improve muscular fitness by exercising a muscle or a muscle group against external resistance, performed systematically in terms of frequency, intensity, and duration, and is designed to maintain or enhance health-related outcomes. Resistance can come from fixed or free weights, elastic bands, body weight (against gravity), and water resistance. It may also involve static or isometric strength (holding a position or weight without moving against it). Often presented as a percentage of the participant's one-repetition maximum (1-RM), the maximum weight they can lift/move if they only must do it once [26].

Outcomes: Muscular strength (Kg and Nm) and muscle mass (cm<sup>2</sup> ).

Study design: Randomized controlled trial (RCT) where the intervention was RT with a follow-up period of at least 4 weeks. RCT is understood as a study in which many similar people are randomly assigned to 2 (or more) groups to test a specific drug, treatment, or other intervention. One group (the experimental group) has the intervention being tested, the other (the comparison or control group) has an alternative intervention, a sham dummy intervention (placebo), or no intervention at all. The groups are followed up to see how effective the experimental intervention was. Outcomes are measured at specific times and any difference in response between the groups is assessed statistically. This method is also used to reduce bias.

Eligibility criteria were applied independently by two blinded authors and disagreements were solved through consensus and active participation of a third author, likewise, the same authors inspected the reference lists from key journals and systematic reviews with a similar PICO to identified all promising or potential studies.

#### *2.4. Data Collection Process*

Two authors independently performed data extraction. Relevant data were extracted to a computer-based spreadsheet. The reviewers extracted the following information: authors' information, publication year, functional status, BRFT characteristics (cuff size and pressure) resistance training protocols (frequency, intensity, length, duration, and volume), and effect estimates (mean, standard deviation, standard error) (Table 1).



Abbreviations: yrs, years; RT, resistance training exercise group; LI-BFR, low-intensity blood flow restriction exercise group; d, days; wk, week; st, sets; RP, repetitions; LC, leg curl; LE, leg extension; LP, leg press; CSA, cross-sectional area; MRI, magnetic resonance imaging; MVC, maximal voluntary contraction.

#### *2.5. Risk of Bias of Individual Studies*

Two authors independently assessed the risk of bias. In the case of disagreement, the subject was discussed with another author. The risk of bias was assessed using the Cochrane risk-of-bias tool for randomized controlled trials (RoB 2.0) [39], which evaluates the risk of bias in five domains: the randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. A study is considered to be at a "low risk of bias" if all five domains have been judged to be at low risk of bias. A study is considered to have "some concerns" if it has been judged to raise some concerns in at least one domain. A study is considered to be at a "high risk of bias" overall if it is judged to be at a high risk of bias in at least one domain. The tool was applied to each outcome of interest.

#### *2.6. Summary Measures*

For continuous outcomes, the group size, the mean values, and the standard deviations (SDs) were recorded for each group compared in the included studies. Pooled effects were calculated using an inverse of variance model, and the data were pooled to generate a standardized mean difference (SMDs) with a corresponding 95% confidence interval (CIs). Most studies for each outcome reported data in the same units, so it was possible to pool all studies regardless of whether they reported changes in-between data at baseline and final data. Significance was set at *p* < 0.05. A random-effects model was used. We used Cohen's guidelines (no effect <0.2, small effect = 0.2 to 0.49, moderate effect = 0.5 to 0.79, large effect ≥ 0.80) [40] to report the magnitude of the effect and help with the interpretation of SMDs. All analyses were performed by a single reviewer using Review Manager (RevMan Version 5.4.1 The Cochrane Collaboration, 2020) and checked against the extracted data by the other author.

#### *2.7. Additional Analysis*

Subject to data availability, the subgroup analysis were performed considering the medium used to evaluate muscle strength and muscle mass on a specific muscle group or the evaluated kinetic chain.

#### *2.8. Certainty of the Evidence: GRADE Approach*

The reviewers decided *a posteriori* to evaluate the certainty of the evidence using the grading of recommendations, assessment, development, and evaluation (GRADE) approach to making the systematic review more usable for clinicians, trainers, decisionmakers, and developers of clinical practice guidelines. We followed the GRADE approach to assess the certainty (or quality) of evidence in three major outcomes. The GRADE approach considers the risk of bias and the body of evidence to rate the certainty of the evidence into one of four levels:

High certainty: We are very confident that the true effect lies close to that of the estimate of the effect.

Moderate certainty: We are moderately confident in the effect estimate—the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

Low certainty: Our confidence in the effect estimate is limited—the true effect may be substantially different from the estimate of the effect.

Very low certainty: We have very little confidence in the effect estimate—the true effect is likely to be substantially different from the estimate of effect.

#### **3. Results**

#### *3.1. Literature Search and Article Selection*

Initial database searches yielded a total of 1659 articles. After performing screening by title and abstract, and then removing duplicates, a total of 326 research papers were discarded, thus obtaining a total of 48 RCTs for full-text review. Subsequently, 36 RCTs were excluded for not assessing muscle mass and strength; apply BFR in aerobic exercise; results recorded on graphs only; apply BFR in pathological older adults. In total 12 studies were included in the Systematic Review [27–38] (Figure 1).

**Figure 1.** Preferred reporting items for systematic reviews and meta-analyses (PRISMA) flow-chart of the study selection.

*3.2. Risk of Bias Individual Studies*

The twelve studies present some methodological problems.

#### 3.2.1. Muscular Strength Outcome (RM Test)

Four (57%) studies were judged of low risk of bias in at least one domain. One of them (14%) related to the random sequence generation and deviations from intended interventions [32]; three (43%) related to the missing data outcome [28,32,36]; and the remaining one (14%) for the measurement of the outcome domain [28]. For further information on the risk of bias, see Figure 2.

**Figure 2.** Risk of bias summary: review authors' judgments about each risk of bias item for muscular strength outcome.

#### 3.2.2. Muscular Strength Outcome (MVC Test)

Four (57%) studies were judged of low risk of bias in at least one domain. Two of them (29%) were judged at low risk of bias in all domains [30,31]; four (57%) related to the missing data outcome [28–30,37], and the remaining three (43%) for the measurement of the outcome domain [28–30]. For further information on the risk of bias, see Figure 3.

**Figure 3.** Risk of bias summary: review authors' judgments about each risk of bias item for muscular strength outcome.

3.2.3. Muscle Mass Outcome (cm<sup>2</sup> )

Four (44%) studies were judged of low risk of bias in at least one domain. One of them (11%) related to the random sequence generation and deviations from intended interventions [31]; three (33%) related to the missing data outcome [29,36,37]; and the remaining one (11%) for the measurement of the outcome domain [28]. For further information on the risk of bias, see Figure 4.

**Figure 4.** Risk of bias summary: review authors' judgments about each risk of bias item for muscle mass outcome.

#### *3.3. Main Findings*

#### 3.3.1. Narrative Synthesis

Twelve studies investigated the effect of the LI-BFR on strength and muscle mass compared to RT [27–38]. All studies that measured strength gains by direct RM test (kg) indicated significant improvements in weight lifted (*p* < 0.05) [28,31–34,36]. However, in the case of the studies that measured strength employing the MVC (Nm) [27–30,35,37,39], the evidence is a bit more uncertain, as two of the seven studies that performed this test did not find significant improvements in strength (*p* > 0.05) [27,35]. Table 2 describes the articles not included in the meta-analysis.


#### **Table 2.**Data from studies not included in the meta-analysis.

Abbreviations: LI-BFR, Low-intensity blood flow restriction exercise group; RT, resistance training exercise group; CON, control group; LE, leg extension; LC, leg curl; LP, leg press; HG, handgrip; EF, elbow flexion; EE, elbow extension; QD, quadriceps; AD, adductors; HM, hamstrings; GM, gluteus maximus, CI, confidence interval; PI, percent increase; SD, standard deviation.

In the measurements concerning muscle mass of the included studies, those that measured changes in quadriceps thickness reported significant differences [27,28,31,33,34,36,38], however, for the lower limb, one study found no significant differences for adductors, hamstrings, and gluteus maximus [36]. In the case of upper extremities, one study reported significant differences in elbow flexor and extensor muscles [37].

#### 3.3.2. Quantitative Synthesis

The effects of BFR on muscular strength assessed in the RM-test, MVC-test, and muscle mass (cm<sup>2</sup> ) are shown in Figures 5–7, respectively.

#### LI-BFR vs. RT Alone on Muscular Strength via RM Test

As shown by Figure 5, when compared to resistance training alone, LI-BFR may have little to no effect in muscular strength measured by the RM test (SMD 0.07 (95% CI: −0.25 to 0.40) *p* = 0.66; I <sup>2</sup> = 0%, *p* = 0.53). However, this evidence is very uncertain. Likewise, this evidence is very uncertain when analyzing this comparison separately in leg press, knee extension, and knee flexion (ES 0.01, ES 0.08, and 0.12, respectively; see Table 3).

**Figure 5.** LI-BFR versus RT on muscular strength (RM test), standard means difference (SMD).


**Figure 6.** LI-BFR versus RT on muscular strength (MVC test), standard means difference (SMD).

#### LI-BFR vs. RT Alone on Muscular Strength via the MVC Testing

The LI-BFR effect on muscular strength measured using the MVC is larger than the one of RT alone (SMD 0.61, 95% CI [0.10 to 1.11], *p* = 0.02; I <sup>2</sup> = 71%, *p* < 0.0001), but again, the evidence of this benefit is very uncertain.

Subgroup analysis by movement patterns reveals that this benefit is mainly due to the knee extension pattern, which is also significant (*p* = 0.05) and has a similar larger effect (SMD 0.65, 95% IC [0.00, 1.29]). Benefits in knee flexion are smaller and non-significant (SMD 0.53, 95% IC [−0.55, 1.61]; *p* = 0.33). Equally, there is also very uncertain evidence about this comparison on the MVC, both in the knee extension (ES 0.65), and in the knee flexion (ES 0.53). Table 4 shows this information.


**Figure 7.** LI-BFR versus RT on muscle mass, standard means difference (SMD).

LI-BFR vs. RT Alone on Muscle Mass (cm<sup>2</sup> )

Our data point out that LI-BFR trend to increase muscle mass over resistance training alone with a moderate effect size (SMD 0.62, 95% CI [−0.09 to 1.34], *p* = 0.09; I <sup>2</sup> = 59%, *p* = 0.05), but the evidence is very uncertain (Figure 7). Likewise, the evidence is very uncertain about the effect of low-load BFR-RT when compared with RT alone on muscle mass in knee extensors (ES 0.26) and knee flexors (ES −0.20), and elbow flexors and extensors (ES 1.65), see Table 5.

#### **Table 3.**Summary of findings for the comparison: LI-BFR versus RT alone on muscular strength (RM test).

#### **Resistance Training with Blood Blow Restriction Versus Resistance Training Alone**

#### **Population:**Non-Active older adults

 **Intervention:** resistance training with blood flow restriction

#### **Comparison:** resistance training

**Setting:**laboratory


 The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RM: maximum repetitions; SMD: Standard mean difference. **\*** Effects size: 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect [29]. 1 Downgraded by two levels due to no randomization process, selection of the reported result, and measurement of the outcome. 2 Downgraded by two-level due to small sample size and wide confidence intervals (imprecision); 3 Downgraded by one level due to no randomization process, and selection of the reported result.

#### **Table 4.** Summary of findings for the comparison: LI-BFR versus RT alone on muscular strength (MVC test).

#### **Resistance Training with Blood Blow Restriction Versus Resistance Training**

**Population:**Non-active older adults

 **Intervention:** resistance training with blood flow restriction

#### **Comparison:** resistance training

#### **Setting:** laboratory


 The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; MVC: Maximum voluntary contraction; SMD: Standard mean difference. **\*** Effects size: 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect [29]. 1 Downgraded by one level due to inconsistency; 2 Downgraded by two-level due to small sample size and wide confidence intervals (imprecision).

#### **Table 5.** Summary of findings for the comparison: LI-BFR versus RT alone on muscle mass (cm<sup>2</sup>).

#### **Resistance Training with Blood Blow Restriction Versus Resistance Training**

**Population:**Non-active older adults

 **Intervention:** resistance training with blood flow restriction

#### **Comparison:** resistance training

**Setting:** laboratory


The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). cm2: Square centimeters; CI: Confidence interval; SMD: Standard mean difference. \* Effects size: 0.2 represents a small effect, 0.5 a moderate effect, and 0.8 a large effect [29]. 1 Downgraded by one level due to no randomization process; 2 Downgraded by one level due to inconsistency; 3 Downgraded by one level due to small sample size and wide confidence intervals (imprecision).

#### **4. Discussion**

#### *4.1. Summary of Main Results*

Our review aimed to determine the effectiveness of the low-intensity resistance training with blood flow restriction compared to dynamic high-intensity resistance training on strength and muscle mass in non-active older adults. We included 6 randomized controlled trials in the meta-analysis, revealing that low-intensity blood flow restriction led to larger significant improvements in muscular strength (MVC test) compared to traditional resistance training. This larger benefit was reduced to a trend when considering the effect on the muscle mass (cross-sectional area, in cm) and even disappeared when comparing differences in muscular strength improvements assessed utilizing the RM test. Particularly, all these outcomes shared a very low level of certainty due to poor quality study designs and disparities in the methodological approach.

Notably, subgroup analysis by movement patterns revealed that the above-mentioned benefit on muscular strength assessed utilizing the MVC was mainly due to the knee extension pattern.

#### *4.2. Certain of Evidence*

The included studies evaluated different resistance training programs with or without BFR. The protocols in these studies differed in terms of the number of sets and repetitions, exercises, and muscle groups involved, as well as in the level of occlusion cuff pressure. Their positions regarding the characteristics of the participants, more specifically on the functional status, were neither entirely clear, as they previously justify the use of BFR in older adults with sarcopenia, yet no information on specific diagnostic tests for sarcopenia was found [15]. Moreover, the functional status of the subjects was determined as inactive (more than 6 months without physical activity), but older adults are a highly heterogeneous population [41], and their exercise-response is also heterogeneous [42], which needs further details. Therefore, the articles included in this review lack clear and unified criteria in the process of sample selection.

Very low-quality evidence formed all the comparisons in this systematic review. Our certainty in the evidence was downgraded due to limitations in the risk of bias assessment, including lack of both randomization process, measurement of the outcome, and selection of the reported result. The absence of blinding of both participants and investigators can lead to an overestimation of the effect estimate, although in exercise interventions it is not easy to blind participants. Of outermost importance, this blinding process is even more difficult in protocols with blood flow restriction, since if familiarization with the device and prior measurement of arterial occlusion pressure with Doppler ultrasound (which all the included studies affirm) have been properly conducted, it is easy to know whether the cuff is exerting pressure on the involved limb. Blinding the intensity is a challenge. Furthermore, most of the studies had low numbers of participants, wide confidence intervals, and high heterogeneity in the effects across them. Importantly, undertaking a sensitivity analysis to explore these limitations was not appropriate due to the low number of studies, which could bias any effect estimate.

#### *4.3. Potential Biases in the Review Process*

The strength of this systematic review was the use of systematic methods to assess the certainty of the evidence. An important limitation in the review process has been, as mentioned above, the heterogeneity of the training and BFR protocols.

Regarding strength training protocols, the number of repetitions was very disparate among the included studies with a range between 6 and 30 repetitions, including one study on muscle failure [32]. This high heterogeneity makes a comparison between studies difficult because the influence of the exercise program on the BFR effect cannot be completely isolated.

Another example of the heterogeneity of the protocols is the occlusion pressure. The included studies used different pressure percentages within the range established by the current positionings [43], and the pressure was calculated in two different ways. Some studies used Doppler ultrasound to determine the maximum arterial occlusion pressure while others applied a pressure value 1.5 times the brachial systolic pressure. It also happens that some studies used variable occlusion protocols (no pressure exerted in the recovery periods between series) while the rest were based on a constant pressure during the entire intervention, making direct comparisons between the results of the studies difficult.

#### *4.4. Agreements and Disagreements with Other Studies or Reviews*

Our findings of low-quality evidence on the effects of BFR on strength and muscle mass align with those reported by two recent systematic reviews [20,44]. For instance, the increase in muscle strength was revealed with effect sizes ranging from 0.55 to 4.34 [44]. Aligned with it, Centner et al. [20] found a greater improvement in muscle strength with pooled effect sizes (ES) of 2.16 (95% CI 1.61 to 2.70). These authors also highlighted a very low level of evidence for the included studies due to the methodological diversity and the very small sample of participants. They included profiles of unhealthy subjects, and they also reported the variability of the profiles due to the high heterogeneity of the elderly. Similarly to us, these reviews revealed a favorable trend for LI-BFR in muscle mass gain, however, this effect did not reach statistical significance. Since the methodological diversity of the above-cited primary reviews [20,44] is similar to ours, we may conclude that the profile and heterogeneity of the physical condition of the participants, being in this systematic review and meta-analysis of older adults, may influence the results regarding LI-BFR resistance training. The agreement between their findings and ours could be also explained by several factors like the control of other important variables, such as the nutritional status [45].

#### *4.5. Implications for Practice and Further Research*

The findings of this systematic review highlight the need for more RCTs, but mostly with a more defined methodological approach in their interventions, since the disparity of the protocols is detrimental to the quality of the evidence, as determined by the grading of recommendations, assessment, development, and evaluation (GRADE). In addition, all the primary studies included, together with those found in other systematic reviews, analyze muscular strength gains through specific strength tests, but there is a lack of knowledge about the effect of LI-BFR on the functional status of the elderly. Of course, it has been previously proven that increasing strength and muscle mass benefits physical capacity in older adults [4,8], but future lines of research might include together strength testing some functional assessments, or even some specific motor tasks and challenges of daily living activities, to determine the impact of BFR on functional capacity and older individuals' autonomy.

#### **5. Conclusions**

The findings of this systematic review point out that strength training with blood flow restriction may induce increased muscular strength and muscle mass in non-active older adults, at least at a similar extent to that in the traditional high-intensity resistance training. However, caution should be when considering these findings, since the evidence is very uncertain about the effect of low-load BFR-RT when compared with RT alone on our outcomes. Further randomized controlled trials with a more defined and standardized methodological protocol are still required and more research is needed to reach a more certain conclusion.

**Author Contributions:** Conceptualization, D.R.-M., A.F.L.-B. and I.C.-M.; methodology, A.F.L.- B.; software, A.F.L.-B. formal analysis, A.F.L.-B.; data curation, A.F.L.-B.; writing—original draft preparation, D.R.-M., A.F.L.-B., P.M., C.B.-L. and I.C.-M.; writing—review and editing, D.R.-M.; A.F.L.-B., P.M., C.B.-L. and I.C.-M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


### *Article* **The Anaerobic Power Assessment in CrossFit ® Athletes: An Agreement Study**

**Tomás Ponce-García 1, \* , Javier Benítez-Porres 1 , Jerónimo Carmelo García-Romero 1 , Alejandro Castillo-Domínguez <sup>2</sup> and José Ramón Alvero-Cruz 1, \***


**\*** Correspondence: tomas\_ponce@uma.es (T.P.-G.); alvero@uma.es (J.R.A.-C.)

**Abstract:** Anaerobic power and capacity are considered determinants of performance and are usually assessed in athletes as a part of their physical capacities' evaluation along the season. For that purpose, many field tests have been created. The main objective of this study was to analyze the agreement between four field tests and a laboratory test. Nineteen CrossFit ® (CF) athletes were recruited for this study (28.63 ± 6.62 years) who had been practicing CF for at least one year. Tests performed were: (1) Anaerobic Squat Test at 60% of bodyweight (AST60); (2) Anaerobic Squat Test at 70% of bodyweight (AST70); (3) Repeated Jump Test (RJT); (4) Assault Bike Test (ABT); and (5) Wingate Anaerobic Test on a cycle ergometer (WG). All tests consisted of 30 s of max effort. The differences among methods were tested using a repeated-measures analysis of variance (ANOVA) and effect size. Agreement between methods was performed using Bland–Altman analysis. Analysis of agreement showed systematic bias in all field test PP values, which varied between −110.05 (AST60PP—WGPP) and 463.58 (ABTPP—WGPP), and a significant proportional error in ABTPP by rank correlation (*p* < 0.001). Repeated-measures ANOVA showed significant differences among PP values (*F*(1.76,31.59) = 130.61, *p* =< 0.001). In conclusion, since to our knowledge, this is the first study to analyze the agreement between various methods to estimate anaerobic power in CF athletes. Apart from ABT, all tests showed good agreement and can be used interchangeably in CF athletes. Our results suggest that AST and RJT are good alternatives for measuring the anaerobic power in CF athletes when access to a laboratory is not possible.

**Keywords:** anaerobic power; peak power; HIFT, high-intensity functional training; crossfit; athletes; field test

#### **1. Introduction**

Anaerobic capacity has been defined as the total amount of ATP re-synthesized, by the whole body, during a maximal intensity and short duration effort by means of the anaerobic metabolic pathways [1]. The time interval to best measure the anaerobic capacity is 30 s [2] since up to 80% of the energy consumed in 30 s of maximal effort comes from anaerobic sources [3,4]. In addition, in a longer test, individuals tend not to apply the maximum intensity [5]. There are several laboratory tests to assess the anaerobic performance [6]. However, most are expensive and difficult to perform due to the specific equipment they require. For that reason, one of the most widely used laboratory tests to assess this ability is the Wingate test, which consists of pedaling with arms or legs at maximum effort for 30 s against a resistance determined by the participant's body weight. WG has shown to be a reliable test, having a test-retest correlation in many populations ranging from 0.89 to 0.98 [7]. Two main variables are determined from this test, peak power (PP) and mean power (XP). PP is also known as "anaerobic power" and is determined by the peak mechanical power recorded during the test, normally occurring in the first 5 to 10 s. In addition, XP is considered by many authors as the "anaerobic capacity" and represents

**Citation:** Ponce-García, T.; Benítez-Porres, J.; García-Romero, J.C.; Castillo-Domínguez, A.; Alvero-Cruz, J.R. The Anaerobic Power Assessment in CrossFit ® Athletes: An Agreement Study. *Int. J. Environ. Res. Public Health* **2021**, *18*, 8878. https://doi.org/10.3390/ ijerph18168878

Academic Editor: Ewan Thomas

Received: 1 July 2021 Accepted: 18 August 2021 Published: 23 August 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

the average mechanical power maintained during the 30 s, taken at 1, 3 or 5 s periods [7]. Some authors have shown PP and XP to be associate with performance in some team and individual sports, especially those performed at high intensity or a combination of low-moderate intensities with higher intensity peaks such as CF [8], surfing [9], alpine ski [10], soccer [11], track and field athletes [12] and many others.

In order to assess this ability out of the laboratory, numerous field tests, consisting of different exercises or tasks, have been created. Some of them based on different modalities of jumps [5,12–16]; running [14,17,18]; squat exercise [14,19,20]; and other exercises such as skipping [21]. All those tests have been studied in active individuals [17,18,21,22] as well as athletes of different sports such as soccer [14], volleyball [5,15], track and field [7,12,20,23], and cyclists [24,25]. They have shown to be valid tools to assess these parameters in athletes [5,12,18,19].

In the last decade, Functional Fitness Training has become one of the top fitness trends around the world [26,27]. One of these functional fitness programs, which has developed into a competitive sport, was branded as CrossFit®. CF is a multimodal highintensity functional training program that combines weightlifting, gymnastics and athletics, among other movements in just one training or competition bout and develops all physical domains such as endurance, strength, stamina, etc. [28]. The multimodality characteristic of this sport, combined with the fact that the tests carried out in competition are not previously announced or standardized, means that CF athletes must be prepared for the unknown and therefore have an optimal development of all physical capacities such as maximum strength, stamina, power, speed, cardiorespiratory fitness, etc. [8,29–35]. Additionally, its intensity component indicates that CF competitors must exhibit a great deal of anaerobic performance to excel in this sport [29].

When a field test is developed to assess any ability of the athletes throughout the season, experts attempt to simulate the specific sporting gestures of the discipline for which it is created (running in soccer, for example). In the case of CF, as a multimodal sport made up of many elements of different kinds (squatting, jumping, running, lifting, etc.), it might seem challenging to succeed in choosing a specific exercise that encompasses all the skills and abilities necessary for this activity and evaluate any capacity accurately. Nevertheless, taking into account the specific characteristics of these athletes, it may be assumed that any field test might be a valid and interchangeable tool to assess any of the physical capacities. Hence, they might show a good performance in any test with jumping, running, cycling, squatting, etc.

In the current work, to assess the anaerobic performance by different exercises and determine their validity and level of agreement, four tests were chosen: a continuous jump test used in previous work by Dal Pupo et al. [5] (RJT), as well as three other tests that, to our knowledge, have not been used previously: two weighted deep squat tests (AST60 and AST70) at different percentages of the athlete's bodyweight (60% and 70%) and a test performed with a particular machine used in CF where upper and lower limbs are used simultaneously called Assault Bike® (ABT).

In CF athletes, some authors have evaluated the physiological determinants of performance in [8,30–35]. Most of them using laboratory tests to assess both the aerobic or anaerobic capacities and comparing the results with those obtained in standardized CF workouts. However, no study of agreement between field methods has been found. Therefore, the main purpose of this study is to analyze the agreement between four different modalities of field test measuring anaerobic performance (AST60, AST70, RJT and ABT) against the gold standard, Wingate test, in CF athletes.

#### **2. Materials and Methods**

#### *2.1. Participants*

Nineteen CF participants volunteered to participate in this study, approved by Málaga University Ethics Committee (CEUMA: 43-2018-H). They were experienced athletes who followed the same competitors' training program and had competed in some national or in-

ternational competition. Data collection was carried out over four weeks off-season. Except for the rest periods established before each test, the athletes followed their regular training regimen throughout those weeks. They were asked to stop taking any supplementation or performance-enhancing products one week prior to data collection. The participants were recruited and tested in a local CF center. All participants provided written informed consent. As inclusion criteria, a minimum of one year of CF practice was established. Any participants with the presence or suspicion of any cardiac pathology, suffering or having suffered recently any musculoskeletal injury or any other condition that prevented exercising properly were excluded. Descriptive data are shown in Table 1.

**Table 1.** Descriptive data of the sample (*n* = 19).


#### *2.2. Study Design*

A cross-sectional study was conducted over four weeks. Despite the fact that all participants were familiar with the exercises in all tests, a familiarization session was also scheduled during the first two weeks. All trials were separated by at least 48 h and performed at the same daytime to avoid the effects of circadian rhythms [36]. Participants were also advised to refrain from any strenuous physical activity in the previous 24 h of each trial. Tests performed were: (1) Anaerobic Squat Test at 60% of bodyweight (AST60); (2) Anaerobic Squat Test at 70% of bodyweight (AST70); (3) Repeated Jump Test (RJT); (4) Assault Bike Test (ABT); and (5) Wingate Anaerobic Test on a cycle ergometer (WG). Tests order execution was randomly assigned. The chronology of the tests is shown in Figure 1.

**Figure 1.** The chronology of the tests.

#### *2.3. Procedures*

#### 2.3.1. Anthropometry, Body Composition and Other Physiological Variables

On the first day, to detect any possible cardiac pathology, all participants underwent an electrocardiogram assessed by a qualified physician. Furthermore, some anthropometric data were taken; height, by a wall-mounted stadiometer (SECA® 206; SECA, Hamburg, Germany) with a precision of 1 mm and body mass, by a scale with a precision of 100 gr (SECA® 803; SECA, Hamburg, Germany). Additionally, body composition was measured by a Medisystem Multifrequency Impedanciometer (Sanocare Human System SL, Madrid, Spain). Participants were asked to go fasting or without consuming any drink or food for at least 4 h, not having consumed alcohol in the last 48 h nor diuretics in the last 7 days or having performed strenuous physical activity in the previous 12 h [37]. Before the measure, they remained supine for 5 min with the upper limbs positioned about 30 degrees apart from the trunk and the lower limbs about 45 degrees apart [38]. Fat mass in kg was estimated according to Segal's formula [39], Lean body mass in kg was calculated by subtracting fat mass from total body mass and muscle mass in kg according to Janssen's formula [40]. Body composition variables were also calculated as a percentage (Table 1).

#### 2.3.2. All-Out Anaerobic Tests

#### Anaerobic Squat Test (AST60 and AST70)

The AST consisted of 30 s at the maximum effort of deep squats with a percentage of the participant bodyweight. The maximum number of squats had to be performed within that interval. Deep squat was established as a squat in which the iliac crest is below the highest part of the knee in its lowest position, and the leg, thigh and trunk segments are fully aligned at the highest position (Figure 2).

**Figure 2.** Full squat movement requirements. (**A**): start position; (**B**): lowest position; (**C**): final position.

The equipment used was a standard olympic lifting set composed of a 20 kg barbell, plates between 5 and 15 kg, with increases of 5 kg, and fractional discs from 0.5 and 2.5 kg, with 0.5 kg increments, from Xenios Usa® (Xenios Usa LLC, New York, NY, USA). The power of each repetition was registered by Beast® accelerometry sensor (Beast technologies) attached to the participant's wrist through a bracelet "ad hoc" (see Figure 3) and data processed by its smartphone application. Beast® sensor has shown to be a valid and reliable tool to measure full-squat values [41]. Two trials with different loads were executed, 60% (AST60) and 70% (AST70) of participant bodyweight. Participants were weighed before each trial to determine the barbell load, rounded to the closest 0.5 kg. As a warm-up, they started with five minutes easy run, followed by one set of ten repetitions with an empty barbell, two more sets of ten repetitions with the assigned percentage and finished with 5 min easy run. Afterwards, a 5 min interval for recovery was established and used to set the accelerometry sensor. At the count of 3, 2, 1 . . . "Go!" the participant began to work at maximum effort, trying to execute as many squats as possible, being verbally motivated by the examiner throughout the test. To cool down, they were asked to easy walk for 5 min.

**Figure 3.** Beast sensor placement on the athlete's at right wrist.

Peak power (PP), mean power (XP) and minimal power (MP) were determined. Fatigue index (FI), understood as the loss of power during the 30 s interval, was calculated by the following formula FI (%) = (PP − PM/PP) × 100 [7].

#### Repeated Jump Test (RJT)

As previously described by Dal Pupo et al. [5], this test consisted of the maximum number of countermovement jumps in 30 s at the maximum height. Before the trial, participants warmed up with 5 min easy run, 3 sets of 10 forward jumps, 3 sets of 5 vertical jumps and 5 additional minutes easy run. Afterwards, a 5 min interval was established to rest and set the sensors. At the count of 3, 2, 1 . . . "Go!" the participant started to jump as high and fast as possible. In order to keep the maximum intensity, the participant was encouraged by the researchers during the whole interval. Right after the test, they were asked to easy walk for 5 min to calm down. Jumping variables were registered by a Polar® V800 with Running Bluetooth® Smart. This sensor has been shown to be valid and reliable to determine jumping variables [42]. PP, XP, MP and FI were determined.

#### Assault Bike Test (ABT)

This test was performed with an Assault Bike® Classic model (Assault Fitness Products; Carlsbad, CA, USA). The Assault Bike® is an air-resisted bike with the peculiarity of using both upper and lower extremities simultaneously (Figure 4). This machine has gained its popularity by being used by most CF centers and official competitions worldwide. The test consisted of 30 s at maximal effort. It began with a 15 min warm-up of cycling at 50 rpm (approximately 176 watts). Next, a 5 min recovery interval was established. The test was carried out from a static position without any inertia. To facilitate the initial start, the crank of the dominant leg was previously set to 45 degrees.

Wingate Anaerobic Test (WG)

The wingate anaerobic test is considered the gold standard when measuring the anaerobic capacity and consists of 30 s at maximum speed on a cycle ergometer with a constant resistance of 0.075 kp per kg bodyweight [7]. The test was executed with a Monark 828E cycle ergometer (Monark Exercise AB, Vansbro, Sweden) calibrated before each trial. Since trials were completed in morning-time (between 9:00 am and 2:00 pm), the warm-up was extended from 5 to 15 min, as proposed by Souissi et al. [36]. To warm up, all participants were asked to ride at 50–70 rpm at 1 kp (50–70 watts) for 15 min. Afterwards, they took a 5 min recovery interval. Straightaway, at the count of "3, 2, 1 . . . Go!" the participant started to ride as fast as possible. The researcher motivated them verbally during the whole time. A 5 min recovery ride at a warm-up pace was set to calm down. Every 5 s, power values were registered. PP, XP, MP and FI were determined. ∑

**Figure 4.** Assault Bike® Classic.

#### *2.4. Statistical Analysis*

≤

The Statistical Package for the Social Sciences (SPSS 21, IBM Corp., Armonk, NY, USA) and MedCalc Statistical Software (MedCalc 18.6, MedCalc Software Ltd., Ostend, Belgium) were used to carry out statistical analyses. The level of significance was set at *p* ≤ 0.05. Data were checked for normality by the Shapiro–Wilks analysis, and the agreement for the PP of the four methods was performed by using Bland–Altman analysis [43]. In order to evaluate the proportional error, Tau Kendall's rank correlation of the difference and mean of every method paired with WG was carried out. Previously, variables of difference and mean were computed for each pair. Furthermore, the differences among PP, XP, MP and IF of the five methods were tested for statistical significance (*p* < 0.05) using a repeatedmeasures analysis of variance (ANOVA). When a significant difference was found, post hoc 2-tailed paired *t*-tests to determine which values were significantly different were used. The Bonferroni adjustment was applied to keep the overall significance level at 0.05. The

assumption of sphericity was tested using Mauchly's test. Additionally, the pairwise effect size was calculated by Cohen's d using G\*Power 3.1.9.6 software.

#### **3. Results**

All variables showed a normal distribution in Shapiro–Wilks analysis (*p* => 0.05), except for RJTXP (*p* = 0.001) and RJTMP (*p* = 0.022). Since the sphericity was violated, Greenhouse–Geisser corrected results are reported (ε = 0.44). The repeated-measures ANOVA showed significant differences among PP values, (*F*(1.76,31.59) = 130.61, *p* =< 0.001). Pairwise effect sizes are shown in Table 2. Additionally, absolute PP, XP and MP values of the AST60 and AST70 tests were slightly lower than the reference test. AST60 PP, XP and MP underestimated WG values by −110.05 (−14.12%), −101.07 (−15.20%) and −94.11 (−17.37%) watts, respectively. AST70 also underestimated WG values by −75.11 (−9.64%), −68.38 (−10.29%) and −56.16 (−10.37%) watts. In addition to the minor underestimation, the differences between AST70 and WG remained quite regular among all power values, around 10%, which was the only test that showed not statically significant differences by ANOVA test and showed the smallest effect size in all variables (Table 2).

**Table 2.** Absolute values of peak, mean, minimal power and fatigue index of the tests.


AST60, anaerobic squat test at 60% of body weight; AST70, anaerobic squat test at 70% of body weight; RJT, repeated jump test, ABT, assault bike test; PP, peak power; XP, mean power; MP, minimal power; FI, fatigue index; SD, standard deviation; *p*, ANOVA *p*-values; *d*, pairwise effect sizes.

> In contrast, the homologous absolute values RJT and ABT were notably higher. With an overestimation of RJT values of 343.22 (44.06%), 393.24 (59.16%) and 380.21 (70.18%), and ABT values of 463.58 (59.52%), 286.05 (43.04%) and 262.10 (48.38%) (Table 2).

> In addition, Bland–Altman's analysis of agreement showed systematic bias in all field test PP values (*p* > 0.05). The smallest difference between all PP values and WGPP was observed for the AST70 with an underestimation of −75.11 watts (95% CI, −124.80, −25.41). AST60 also underestimated PP by −110.05 watts (95% CI, −157.74, −62.36). Nevertheless, the other two tests, RJT and ABT, overestimated PP by 343.22 watts (95% CI, 312.63, 373.80) and 463.58 watts (95% CI, 380.18, 546.98), respectively.

> Furthermore, only a significant proportional error was found in ABTPP by Tau Kendall's rank correlation (Table 3 and Figure 5d).



AST60, anaerobic squat test at 60% of body weight; AST70, anaerobic squat test at 70% of body weight; RJT, repeated jump test, ABT, assault bike test; CI, confidence Interval.

**Figure 5.** Bland–Altman's plots representing differences (Y axes) and mean (X axes) of measurements between: (**a**) AST60 and WG; (**b**) AST70 and WG; (**c**) RJT and WG; (**d**) ABT and WG.

#### **4. Discussion**

− − − The main purpose of the present study was to evaluate the agreement between the five methods to assess anaerobic power in CF athletes. Since to our knowledge, this is the first study to analyze the agreement between various methods to estimate anaerobic power in CF athletes. Bland–Altman's analysis revealed a systematic bias with a mean difference that can vary between −110.05 watts (AST60PP−WGPP) and 463.58 Watts (ABTPP−WGPP). Despite the systematic bias shown by all the field tests compared with the laboratory test, the results showed good agreement between all methods (*p* > 0.05) since more than 80% of the dots on the graph were within the limits of agreement. In contrast, Tau Kendall's rank correlation analysis showed a proportional error in ABTPP (*p* < 0.001), where the differences were small for low PP values in the range of measurements and become higher as the true value increases. Additionally, the lowest within-subject variability in all the variables studied in the present work suggests that the AST70 is a valid field test to assess the power and anaerobic capacity in CF athletes.

Some of the field tests practiced in this study, such as AST and ABT, have not been previously used. AST is a test based on the squat exercise tested with two different percentages of the participants' body mass (60 and 70). The underestimation of PP absolute values, supported by the findings of Luebbers et al. [20], suggests that it might be interesting to replicate the study using higher percentages (75 and 80) to achieve more accurate agreement. In addition, some studies have shown underestimation of absolute values in a running test assessed in armed forces operators [21] and cycling athletes [25], as well as

a kicking test studied in taekwondo athletes [44]. On the other hand, the overestimation of the RJT PP value is consistent with the findings of Sands et al. [16], where absolute power values of the Bosco test were higher than WG. In our study, overestimation was also found in ABT, and it might be due to the simultaneous use of lower and upper limbs to generate power instead of only the lower limbs as in WG. We have not found any previous study carried out with this machine that can provide data in this regard. However, the simultaneous use of the muscles of the lower limbs involved in pedaling and those of the upper limbs involved in pulling and pushing may suggest a more significant muscle mass implication and thus a greater capacity to generate power.

The results abovementioned are consistent with the WGPP differences reported by Gacesa et al. [45] in a comparison testing of maximum anaerobic performance on different elite athletes. Their findings suggest that the ability to generate power may be dependent on the activity since the highest values were found in anaerobic predominant sports such as volleyball, basketball, hockey, boxing, and wrestling, and lower values in soccer, rowing, and long-distance running athletes, which are predominantly aerobic types of sports. Further, some authors have found differences in power values between participants of different positions in basketball [46] and elite runners of different distances [23]. Consequently, it might be thought anaerobic power to be related to specific disciplines or attributed to some degree of specificity of the athletes tested. However, the results shown in the present study, due to the need of CF athletes to face multiple physical demands with a high level of intensity, may indicate that these athletes are able to exhibit outstanding anaerobic performance in tests of different nature (jumping, squatting, cycling, etc.).

Many comparisons or validity studies where authors studied the level of agreement between only one field test and WG were found. However, a lack of agreement works between more than one field method and the laboratory test in the literature makes it difficult to compare our results with any other. Moreover, as mentioned above, most of their results show some level of under or overestimation of field-test values which may be attributable to the biomechanical, technical or any other difference in the sporting gesture used for each test together with the intrinsic characteristic of the athlete tested. Future studies analyzing the agreement between different task tests may be of interest to find the cause of that variability and the most suitable field test for each discipline, especially in a multimodal sport as it is CF.

One limitation of the present work was not considering any other variables, such as kinematics, that could reflect the different biomechanical or lifting strategies related to performance in AST or any other test. Future research should aim to record these variables mentioned above and evaluate the interaction in the outcomes, replicating this work with other tests composed by other CF-specific exercises or in athletes of different experience/fitness levels.

In practice, the use of AST70 or RJT as a method to assess the anaerobic power in CF athletes could provide an alternative for coaches interested in assessing or monitoring their athletes at any point of the season without the need of taking them to a sports medicine laboratory.

#### **5. Conclusions**

Since to our knowledge, this is the first study to analyze the agreement between various methods to estimate anaerobic power in CF athletes. In conclusion, our results show a good level of agreement between all four methods and WG, being greater in AST70, which suggests that they may be used interchangeably with the exception of ABT. The proportional error found in ABT might make its use doubtful. Moreover, the results of the present study suggest that the magnitude of peak power values seems to be dependent on the type of exercise and athlete characteristics.

**Author Contributions:** Conceptualization, T.P.-G. and J.R.A.-C.; methodology, T.P.-G. and J.R.A.-C.; software, T.P.-G. and J.R.A.-C.; validation, J.C.G.-R., J.R.A.-C., A.C.-D. and J.B.-P.; formal analysis, T.P.-G. and J.R.A.-C.; investigation, T.P.-G.; data curation, T.P.-G. and J.R.A.-C.; writing—original draft preparation, T.P.-G.; writing—review and editing, T.P.-G., J.C.G.-R., J.R.A.-C., A.C.-D. and J.B.-P.; supervision, J.C.G.-R. and J.R.A.-C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki and approved by Ethics Committee of Universidad de Málaga (Code: CEUMA 43-2018-H).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

**Acknowledgments:** The authors wish to thank CrossFit Teatinos and all participants for their collaboration and the support from the University of Málaga (Campus of International Excellence Andalucía Tech).

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

