**Appendix A**

**Table A1.** Summary of the kinematic variables in the "Set" position. Data are the magnitude of the mean ± SD presented in the reviewed studies. Groups are male, female, and mixed (when authors joined data without discriminating by sex) sprinters. Studies are listed, in each variable, in reverse chronological order. Data, terms, conditions, and sprinters' performance levels are presented according to the original authors. Statistical differences between groups are marked with asterisks (\* *p* < 0.05; \*\*\* *p* < 0.001).










62






CM—center of mass; (a) internal angle between the thigh and trunk in flexion/extension plane; (b) relative angle between the pelvis and the thigh according to the Biomechanical Convention [53]; (c) relative angle between the thigh and the shank according to the Medical Convention [53]; (d) relative angle between the shank and the foot according to the Biomechanical Convention [53]; (e) rear leg hip angle measured as front-rear leg angle; (f) relative angle between the vector from hip to shoulder and the horizontal plane.

**Appendix B**

**Table A2.** Summary of the kinematic variables in the "Block Phase". Data are the magnitude of the mean ± SD presented in the reviewed studies. Groups are male, female, and mixed (when authors joined data without discriminating by sex) sprinters. Studies are listed, in each variable, in reverse-chronological order, followed by alphabetically for studies published in the same year. Data, terms, conditions, and sprinters' performance levels are presented according to the original authors. Statistical differences between groups are marked with asterisks (\* *p* < 0.05; \*\* *p*< 0.01; \*\*\* *p* < 0.001; # Cohen's d—large effect size (>0.8); § small effect size [0.2–0.6] of 90% confidence intervals; §§ moderate effect size [0.6–1.2] of 90% confidence intervals); Ϯ adults; ƥ different from descending phase—*p* < 0.05; ¥ significantly greater compared to the bunched start.


1








Anthropometric condition 3.50 ± 0.39

1

block phase).

Elongated start 2.89 ± 0.13 ¥












73


74

CM—center of mass; ROM—range of motion; (a) block time calculated from the difference between the average data of total block time and reaction time data; (b) probably an incorrect data from the original paper; (c) presumably the negative signal is a gap in the data reported in the original paper; (d) the take-off or push-off angle is the angle between the horizontal and the line passing through the most front part of the contact foot and the center of mass at block clearance; (e) center of mass projection angle is calculated as the resultant direction from the horizontal and vertical block exit velocities of the center of mass; (f) angular displacement during rear block contact only; (g) higher magnitude of dorsiflexion was correlated to a faster stretch velocity, which was related to increased force generation (maximal rate of force development, maximal resultant and horizontal push force, and also normalized average horizontal block power); (h) the angle, measured relative to the horizontal, between the line passing through the hip and shoulder (trunk segment) of the side of the body in which the athlete's front foot at the block take off instant; (i) relative angle between the pelvis and the thigh according to the Biomechanical Convention [53]; (j) relative angle between the thigh and the shank according to the Medical Convention [53]; (k) relative angle between the shank and the foot according to the Biomechanical Convention [53].


**Table A3.** Summary of the kinetic variables in the "Block Phase". Data are the magnitude of the mean ± SD presented in the reviewed studies. Groups are male and mixed (when authors joined data without discriminating by sex) sprinters. Studies are listed, in each variable, in reverse-chronological order, followed by alphabetically for studies published in the same year. Data, terms, conditions, and sprinters' performance levels are presented according to the original authors. Statistical differences between groups are marked with asterisks (\* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001; # Cohen's d—large effect size (>0.8); §§§ large effect size [1.2–1.6] of 90% confidence intervals; Ϯ adults; ƥ different from descending clearly associated with average horizontal power produced across the block phase—*p* < 0.05; Ϯ adults; ƥ different from descending Ϯ adults; ƥ different from descending moderately associated (moderate effect


1











1

compared with either leg in the block phase).


female data; (e) average horizontal external power is calculated as the product of anteroposterior force and horizontal velocity; (f) average horizontal external power was calculated based on the rate of change of mechanical energy in a horizontal direction (i.e., change in kinetic energy divided by time) [2]; (g) normalized average horizontal external power is the average horizontal external power normalized to the mass and the leg length of the sprinter [2]; (h) for normalization, the body height was used instead of the sprinter's leg length [25]; (i) joint data normalized to the mass and the leg length of the sprinter; (j) significantly larger (*p* < 0.05, Cohen's *d* = 2.02–11.09) than any other lower-limb and lumbosacral torques, although quantitative data for the remaining joint torques are not available.

**Appendix C**

**Table A4.** Summary of the kinematic variables in the "first two steps". Data are the magnitude of the mean ± SD presented in the reviewed studies. Groups are male, female, and mixed (when authors joined data without discriminating by sex) sprinters. Studies are listed, in each variable, in reverse-chronological order, followed by alphabetically for studies published in the same year. Data, terms, conditions, and sprinters' performance levels are presented according to the original authors. Statistical differences between groups are marked with asterisks (\* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001; # significant different from adults; § small effect size [0.2–0.6] of 90% confidence intervals; §§ moderate effect size [0.6–1.2] of 90% confidence intervals).








87









94

Medical Convention [53]; (i) relative angle between the shank and the foot according the Biomechanical Convention [53]; (j) data referring to the values recorded in the entire acceleration

phase (0–40 m) excluding the block phase.


**Table A5.** Summary of the kinetic variables in the "first two steps". Data are the magnitude of the mean ± SD presented in the reviewed studies. Groups are male and mixed (when authors joined data without discriminating by sex) sprinters. Studies are listed, in each variable, in reverse chronological order, followed by alphabetically for studies published in the same year. Data, terms, conditions, and sprinters' performance levels are presented according to the original authors. Statistical differences between groups are marked with asterisks (\* *p* < 0.05; \*\*\* *p* < 0.001; # significant different from adults; Ϯ adults; ƥ different from descending different from descending small effect size [0.2–0.6] of 90% confidence intervals; Ϯ significantly significantly greater compared with either leg in the block phase).







1




98

1






calculated during the increase in ankle joint moment; (d) ankle joint stiffness was calculated during the decrease in ankle joint moment; (e) only elite female data.

#### **References**


## *Article* **Effect of Physical Training on Body Composition in Brazilian Military**

**Luis Alberto Gobbo <sup>1</sup> , Raquel David Langer <sup>2</sup> , Elisabetta Marini 3,\* , Roberto Buffa <sup>3</sup> , Juliano Henrique Borges <sup>2</sup> , Mauro A. Pascoa <sup>2</sup> , Vagner X. Cirolini <sup>2</sup> , Gil Guerra-Júnior <sup>2</sup> and Ezequiel Moreira Gonçalves <sup>2</sup>**


**Abstract:** The military are selected on the basis of physical standards and are regularly involved in strong physical activities, also related to particular sports training. The aims of the study were to analyze the effect of a 7-month military training program on body composition variables and the suitability of specific 'bioelectrical impedance vector analysis' (spBIVA), compared to DXA, to detect the changes in body composition. A sample of 270 male Brazilian cadets (19.1 ± 1.1 years), composed of a group practicing military physical training routine only (MT = 155) and a group involved in a specific sport training (SMT = 115), were measured by body composition assessments (evaluated by means of DXA and spBIVA) at the beginning and the end of the military routine year. The effect of training on body composition was similar in SMT and MT groups, with an increase in LST. DXA and spBIVA were correlated, with specific resistance (Rsp) and reactance (Xcsp) positively related to fat mass (FM), FM%, LST, and lean soft tissue index (LSTI), and phase angle positively related to LST and LSTI. Body composition variations due to physical training were recognized by spBIVA: the increase in muscle mass was indicated by the phase angle and Xcsp increase, and the stability of FM% was consistent with the unchanged values of Rsp. Military training produced an increase in muscle mass, but no change in FM%, independently of the sample characteristics at baseline and the practice of additional sports. SpBIVA is a suitable technique for the assessment of body composition in military people.

**Keywords:** bioelectrical impedance; vector analysis; lean soft tissue; fat mass; muscle mass; phase angle

#### **1. Introduction**

The military paradigm is associated with healthy appearance, athletic bearing, and high-level physical performance. Indeed, the military are selected based on physical standards and are regularly involved in strong physical activities, also related to sports training, which requires monitoring for variations in body composition variants [1].

There are various methods usable to evaluate body composition, including anthropometry; bioimpedance; and more accurate techniques, such as potassium 40 counting, water isotope dilution, underwater weighing, imaging techniques, and dual energy X-ray absorptiometry (DXA) [2].

Due to the high suitability and low cost, the anthropometric techniques are the most used in many fields of application, including the routine military practice [3–5]. These methods, however, are not very accurate in detecting the main body compartments. For

**Citation:** Gobbo, L.A.; Langer, R.D.; Marini, E.; Buffa, R.; Borges, J.H.; Pascoa, M.A.; Cirolini, V.X.; Guerra-Júnior, G.; Gonçalves, E.M. Effect of Physical Training on Body Composition in Brazilian Military. *Int. J. Environ. Res. Public Health* **2022**, *19*, 1732. https://doi.org/10.3390/ ijerph19031732

Academic Editor: José Carmelo Adsuar Sala

Received: 15 December 2021 Accepted: 29 January 2022 Published: 2 February 2022

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

**Copyright:** © 2022 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/).

example, body mass index does not distinguish lean mass from fat mass [6–8] and so is incapable of evaluating muscle mass gain concomitant to fat weight loss (as generally occurs with intense military training) [9]. Accordingly, Pierce et al. [10] have recently demonstrated that BMI is not associated with performance on military relevant tasks in U.S. Army soldiers. Further, waist circumferences, largely used among the military [3–5] due to the associations with intra-abdominal fat and the related morbidity outcomes [11], are subject to intra- and inter-observer errors of measurement [12] and the need for a strong standardization because of the different possible measurement sites. Research results in military members are discordant, showing both a good and a poor agreement between the circumference measurement body composition method and dual-energy X-ray absorptiometry (DXA) [5,13].

Bioelectrical impedance analysis (BIA) is a non-invasive, low cost, and easy to operate technique, which needs a very short time compared to the more sophisticated body composition methods [14]. BIA has been rarely applied to research in the military showing a good agreement with DXA results [13,15,16]. The traditional two-component approach of BIA uses predictive equations, including bioelectrical values (generally resistance), and considers other variables (age, sex, and height) for the evaluation of fat mass and fat-free mass [17]. However, the application of predictive equations in samples differing from those where they have been calibrated can introduce a source of error. Otherwise, the use of population/group-specific equations reduces the comparability of results.

Alternative approaches, that have been proposed to avoid the use of equations and possible related errors, are based on the analysis of raw bioelectrical data of resistance (R, ohm) and reactance (Xc, ohm). The phase angle (arctan Xc/R 180/π, degrees) is an indicator of nutritional status related to body cell mass and cell membrane integrity, that is largely used in clinical practice [18,19]. Phase angle has also been analyzed in relation to resistance training, and an increasing trend of its values has been registered [20]. However, as shown by Mereu et al. [21], the analysis of body composition based on the phase angle only can be inaccurate and is significantly improved if the information given by the vector length (R<sup>2</sup> + Xc<sup>2</sup> ) 0.5 is also considered.

Such a vectorial approach has been proposed by Piccoli et al. [22], who conceived the bioelectrical impedance vector analysis (BIVA). The classic BIVA procedure analyzes the bioelectrical values of resistance and reactance, standardized for body height (a proxy of conductor length). A BIVA variant defined as 'specific bioelectrical impedance vector analysis' (spBIVA) implies the standardization of resistance and reactance by length and by cross-sections of the body as well [22–24]. SpBIVA has been shown to be effective in the evaluation of fat mass percentage [23–25] and skeletal muscle mass [9,23,26,27]. Specific reference values have been proposed for 50 different populations, such as Italian-Spanish, U.S. young adults, and Italian healthy elderly [22,27,28]. The classic BIVA approach has been sporadically used in relation to sport and exercise [29], and specific BIVA even less [29–32]. Neither classic nor specific BIVA has been applied to the military samples.

The aims of the present study were two-fold: (1) to analyze the effect of a 7-month military training program on body composition variables, and (2) to analyze the correlation between the changes in body composition measured by spBIVA and DXA in a Brazilian military sample.

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

In accordance with the Helsinki Declaration, a written informed consent was obtained from all participants. The research was approved by the Ethics Committee of the School of Medical Sciences, University of Campinas. All procedures followed Resolution No. 466 of 2012 of the National Health Council of the Ministry of Health of Brazil.

#### *2.1. The Sample*

A sample of 270 young men (19.1 ± 1.1 years) from all the regions of Brazil (South, Southeast, Midwest, North, and Northeast) enrolled in the Preparatory School of Army

Cadets (EsPCEx) of the city of Campinas, SP, Brazil, was selected. Data were collected over two years (2013 and 2014), in two periods: at the beginning (March/April) and the end (October/November) of the military routine year.

The sample was divided into two groups: (1) the cadets who were involved in the military physical training routine only (MT, n = 155); (2) the cadets who were involved, by their own choice, in the military physical training routine plus a specific sport training for military competition (SMT, n = 115): track and field (n = 25), basketball (n = 16), fencing (n = 3), soccer (n = 18), judo (n = 3), swimming (n = 13), trekking (n = 4), shooting (n = 2), triathlon (n = 11), volleyball (n = 15), or chess (n = 5).

All cadets were included in the sample, except those who did not sign the consent form, who did not attend the day of the evaluations (even if only the second ones), who had a history of musculoskeletal injury at the time of the assessments, or were disconnected from the school.

#### *2.2. Military Physical Training*

Military physical training was performed 5 days/week during 90 min/day for 34 weeks, according to the academy military physical training manual, where the cadets were supposed to undergo a physical training that consisted of: (a) 2–3 sessions/week of continuous or interval running, with a weekly increased load; (b) 1 session/week of calisthenics exercises (7–15 repetitions of push-up, push-up/stand-up, sit-up, squat with hands on hip, lunge with hands on hip, and jumping jacks); (c) 1–2 sessions/week of circuit resistance training (2 sets of bench press, sit-up and its variations, half squat, barbell curl, pull-up, stair jumps, jump rope, and wrist roller, with 30 s of each exercise and 30 s of rest interval); (d) 1 session/week of swimming; and (e) 2 sessions/week of sports training. Before each session, all participants went through ~8 stretching exercises, ~7 neuromuscular warm-up, and ~7 general warm-up exercises. For sports training, each participant performed specific training for each modality [32,33].

#### *2.3. Measurements*

All subjects underwent anthropometric, BIA and DXA assessments, in the same sequence, in the morning.

Anthropometric measurements were performed following standard procedures [34], by an accredited International Society for the Advancement of Kinathropometry (ISAK) technician. Body weight (kg) and height (cm) were measured using a digital scale with precision of 0.1 kg (Filizola, São Paulo, Brazil) and a Harpenden stadiometer with precision of 1 mm (Holtain Limited, Crosswell, UK), respectively. Relaxed upper arm, waist and calf girths were measured using an anthropometric tape (precision of 1 mm). Body mass index in kg·m−<sup>2</sup> was calculated by the ratio between body weight, in kilograms, and height squared, in meters (BMI).

A fan beam equipment model iDXA (GE Healthcare Lunar, Madison, WI, USA), enCoretm 2011 software (version 13.6), was used to determine body composition. Total body composition was measured with the subject lying in the supine position, with the scanning time for the full length of approximately seven minutes. Total fat mass (FM, kg), fat percent (%FAT), lean soft tissue (LST, kg), and bone mineral content (BMC, kg) were measured. LSTI (kg·m−<sup>2</sup> ) was calculated as LST/height squared in meters. To determine the reproducibility of the variables estimated by the equipment, coefficient of variation (CV%) and the technical error of measurement (TEM) were determined, based on the testing and retesting conducted with 23 subjects, and retested within 24 h. The values of CV% were 0.74%, 0.28%, and 0.26% for FM, BMC, and LST, respectively, and TEM were 0.25 kg (FM), 0.02 kg (BMC), and 0.25 kg (LST).

Bioelectrical measurements (resistance (R), ohm; reactance (Xc) ohm; at 50 kHz and 425 µA) were taken following the standard procedure [14]. With a Bioelectrical Body Composition Analyzer, tetrapolar device, single frequency (50 kHz), and model Quantum II (RJL Systems, Detroit, MI, USA). Specific bioelectrical impedance vector analysis was

applied [24]. Specific bioelectrical values (resistivity (Rsp) ohm cm; reactivity (Xcsp) ohm cm) were obtained by multiplying resistance and reactance by a correction factor (A/L), where area (A, cm<sup>2</sup> ) and length (L, cm) were estimated as follows: A = (0.45 upper arm area + 0.10 waist area + 0.45 calf area) and L = 1.1 stature (in cm). The segment areas were calculated as C2/4π, where C (cm) is the girth of the upper arm, waist, or calf. The phase angle (degrees) was calculated as arctan (Xc/R 180/π) and the impedivity vector (Zsp, ohm cm) as (Rsp<sup>2</sup> + Xcsp<sup>2</sup> ) 0.5 .

All participants should have fasted for at least 4 h, not ingest caffeinated foods or alcoholic beverages 24 h prior to the test, not perform strenuous physical activity less than 12 h before the test, not use any diuretics for at least 7 days before the test, urinate about 30 min before the test, and remove all metals (bracelets, watch, chains, etc.). During the assessment, the volunteers remained in the supine position, on a stretcher isolated from electrical conductors, in the supine position, with the legs abducted at an angle of approximately 45 degrees [35]. The values of CV% were 0.35% and 0.33% for R and Xc, respectively, and TEM were 3.54 Ω and 0.49 Ω, respectively, for R and Xc, for the same 23 subjects retested for DXA. Gonzalez et al. [36] validated the same equipment used in this study in a Brazilian sample, also using DXA as a reference method.

The within-sample variability was investigated by considering the distribution of bioelectrical values in the tolerance ellipses, representing the bivariate percentiles of the reference population. At this purpose, the tolerance ellipses for the Italo-Spanish adults (18–30 years) [28] have been used. The major axis of the ellipses refers to variations of FM% (higher values towards the upper pole) and the minor axis to variations of skeletal muscle mass and ECW/ICW (lower values on the left side).

#### *2.4. Statistical Analyses*

Bioelectrical values of MT and SMT groups were compared by mean of tolerance and confidence ellipses, using a two-sample Hotelling's T<sup>2</sup> test.

The consistency of the results obtained with specific BIVA and DXA was evaluated by means of Pearson's correlation between bioelectrical and DXA variables at baseline and comparing the trend of longitudinal body composition variations described by the two techniques. The effect of training (pre- vs. post-training) in the two sub-samples of SMT and MT was analyzed using two-way ANOVA (anthropometric, DXA output and bioelectrical values) and paired one-sample Hotelling's T<sup>2</sup> (confidence ellipses).

Statistical analyses were performed using IBM SPSS Statistics 19 (IBM SPSS Statistics for Windows, Version 19.0. Armonk, NY: IBM Corp) and the specific BIVA software (freely available at the website: http://specificbiva.unica.it/ (accessed on 17 August 2018).

#### **3. Results**

On average, at baseline, the sample of military people was in the normal weight BMI category and had a percent fat mass within the normal range for men (Table 1).




*Int. J. Environ. Res. Public Health* **2022**, *19*, x FOR PEER REVIEW 5 of 11

Waist crf, cm 76.2 4.8 75.7–76.8

FM, kg 12.2 3.7 11.7–12.6 LST, kg 55.2 6.4 54.4–55.9

**Table 1.** *Cont.*

Legend: SD = standard deviation; BMI = body mass index; LSTI = lean soft tissue index; FM = fat mass; LST = lean soft tissue; BMC = bone mineral content; FM% = fat mass percent; Rsp = specific resistance; Xcsp = specific reactance. mass; LST = lean soft tissue; BMC = bone mineral content; FM% = fat mass percent; Rsp = specific resistance; Xcsp = specific reactance.

Bioelectrical values were quite totally within the reference tolerance ellipses, but slightly shifted toward the lower pole, indicative of low FM% (Figure 1). Bioelectrical values were quite totally within the reference tolerance ellipses, but slightly shifted toward the lower pole, indicative of low FM% (Figure 1).

**Figure 1.** Distribution of bioelectrical values of Brazilian Military onto tolerance ellipses representing Italian-Spanish young adults, at the beginning of the routine year. **Figure 1.** Distribution of bioelectrical values of Brazilian Military onto tolerance ellipses representing Italian-Spanish young adults, at the beginning of the routine year.

DXA and specific BIVA were correlated (Table 2). In fact, Rsp and Xcsp were positively related to FM, FM%, LST, and LSTI, while phase angle was positively related to LST and LSTI. DXA and specific BIVA were correlated (Table 2). In fact, Rsp and Xcsp were positively related to FM, FM%, LST, and LSTI, while phase angle was positively related to LST and LSTI.

**Table 2.** Matrix of correlation between bioelectric and DXA variables (N = 270) at baseline.


LST, kg 0.229 0.000 0.300 0.000 0.189 0.002 LSTI, kg·m−2 0.292 0.000 0.497 0.000 0.400 0.000 Legend: r = Pearson correlation coefficient; *p* = *p* value; FM = fat mass; FM% = fat mass percent; LST = lean soft tissue; LSTI = lean soft tissue index; Rsp = specific resistance; Xcsp = specific reactance; PA = phase angle.

Military people practicing sport activities (SMT) showed body composition differences with respect to those practicing military training only (MT). In fact, SMT group had higher values of LST and BMC, and lower values of FM and FM% (Table 3). The bioelectrical values of specific reactance and phase angle were significantly higher in SMT than in MT, indicating higher muscle mass (Table 3, Figure 2). ences with respect to those practicing military training only (MT). In fact, SMT group had higher values of LST and BMC, and lower values of FM and FM% (Table 3). The bioelectrical values of specific reactance and phase angle were significantly higher in SMT than in MT, indicating higher muscle mass (Table 3, Figure 2). **Table 3.** Descriptive and comparative statistics.

Legend: r = Pearson correlation coefficient; *p* = *p* value; FM = fat mass; FM% = fat mass percent; LST = lean soft tissue; LSTI = lean soft tissue index; Rsp = specific resistance; Xcsp = specific reactance;

Military people practicing sport activities (SMT) showed body composition differ-


*Int. J. Environ. Res. Public Health* **2022**, *19*, x FOR PEER REVIEW 6 of 11

**Table 3.** Descriptive and comparative statistics. **SMT (N = 115) MT (N = 155)**

Legend: SMT: Sports and Military Training; MT: Military Training only; F, F test of two-way ANOVA for group (Fg), training (Ft), and group-training interaction (Fgxt); BMI = body mass index; FM = fat mass; FM% = fat mass percent; LST = lean soft tissue; LSTI = lean soft tissue index; Rsp = specific resistance; Xcsp = specific reactance; PA = phase angle. Legend: SMT: Sports and Military Training; MT: Military Training only; F, F test of two-way ANOVA for group (Fg), training (Ft), and group-training interaction (Fgxt); BMI = body mass index; FM = fat mass; FM% = fat mass percent; LST = lean soft tissue; LSTI = lean soft tissue index; Rsp = specific resistance; Xcsp = specific reactance; PA = phase angle.

**Figure 2.** Confidence ellipses with T<sup>2</sup> Hotelling's test in the two groups before and after training. Legend: SMT: Sports and Military Training; MT: Military Training only. Comparisons between SMT and MT were performed using two-sample Hotelling's T2 tests, while those between pre- and posttraining groups were performed with paired one-sample Hotelling's T<sup>2</sup> tests.

#### **4. Discussion**

In this sample of military personnel, DXA and specific BIVA showed a consistent scenario of body composition variations related to physical training. In fact, specific bioelectrical variables were correlated with DXA (FM, FM% and LST, and LSTI), and both the techniques showed: (a) different body composition in the military practicing physical training routine only (MT) or a specific sport training as well (SMT); (b) an increase in fat-free mass and a steady percentage of fat mass in relation to training, in both SMT and MT groups.

The sample, especially the SMT group, showed body composition characteristics adequate to the military standard, as suggested by the BMI indicative of normal weight [37], and the percentage of fat mass, which was lower than the body fat limits of approximately 20%, desirable for the U.S. army men [4]. The values of fat-free mass were higher in SMT than MT. The period of over ~7 months of military training induced, in both MT and SMT groups, a gain of lean soft tissue that contributed to the higher value of weight and BMI. However, the absolute and relative quantity of body fat did not change.

The observed differences of body composition are consistent with the effects of physical training described in the general population, and in the military [38,39]. Aerobic, stretching and resistance training are among the main interventions that can affect fat mass, fat free mass, and skeletal muscle mass, especially in young adults. These effects can be achieved in adults in a period of 3 to 12 months, depending on the characteristics of the sample and the volume of training, as well as other influent factors, such as daily habits and, particularly, alimentary style [40].

Research focused on military training has shown in general an increase in fat-free mass [15,16,34,41–43], but not in Margolis et al. [44], while the results on fat mass are less consistent among the studies. Mikkola et al. [16], in Finnish military performing regular, rather high-intensity, physical activity, over a period from 6 to 12 months, observed an increase in fat mass (in normal weight individuals), but a decrease in visceral fat. Indeed, intense physical activity promotes a greater reduction of visceral than subcutaneous adipose tissue, even if weight increases [45].

As previously presented, despite the inability of indicators such as BMI and waist circumference to validly identify lean and fat mass in physically active soldiers [6–12], there is still a gap in studies with DXA, for example, to test the agreement with the bioimpedance technique [5,13].

From a methodological point of view, similarly to the present research, previous studies realized in U.S. adults [9,23] and elderly Italians [24] detected a high correlation between DXA and specific BIVA variables. In particular, Rsp and Xcsp showed a positive correlation with FM% (especially Rsp; [9,24]) and with FFMI (especially Xcsp; [9]), while phase angle was positively related to FFMI only [9]. It is noteworthy that such convergent results have been obtained in samples characterized by different geographical provenience (Brazil, present study; US and Italy) [9,23,24], age class (Young adults, adults, and elders), and lifestyle (military, general population, and retirees). Indeed, the observed relationships are expected. In fact, resistance is negatively related to total body water and electrolytes, and hence, in normal-hydrated individuals, increases with the proportion of low conductive tissues, such as fat mass [9,24]. On the other side, the capacitive component (reactance) and phase angle are associated with the polarization produced by cell membranes and tissue interfaces and are positively related with body cell mass [22]. In this study, the correlation between DXA and spBIVA is shown by the trajectory of vector migration in relation to training, that, in both MT and SMT groups, is associated with increased values of reactance and phase angle (increasing muscle mass), but quite unchanged Rsp values (stable FM%). Such results can be comparable to those of Campa et al. [46], who analyzed three different sports modalities (volleyball, soccer, and rugby) and observed higher PhA values in athletes with a high mesomorphic component, which means, higher skeletal muscle component.

However, inconsistent results between DXA/spBIVA and waist circumference have been observed. In fact, SMT group showed higher values of waist circumferences, increasing in time, with respect to MT, but lower FM% levels, which remained stable after the military training. A similar disagreement between DXA and the circumference methods has already been described in the military [5]. In our research, we have also observed that sp-BIVA results, similarly to DXA, were not consistent with the pattern of waist circumference differences. However, compared to DXA, spBIVA did not recognize a lower percentage of fat mass in SMT with respect to MT. The observed gaps between abdominal circumference and DXA or spBIVA are noteworthy, considering the particular emphasis given to circumference measurement to calculate body fat percentage among the military [42]. The inconsistencies can be likely related to the different distribution of body components in the central and peripheral regions of the body and maybe to the greater effect of training on visceral than on subcutaneous fat, discussed above.

Despite the fact that the present study analyzed cadets who practiced 11 different sports, approximately 85% of the total SMT group practiced either teams' sports or individual sports, such as cyclical sports (swimming, athletics, or triathlon), therefore, the physiological characteristics were not so different when comparing practitioners of sports modalities, such as basketball, with practitioners of judo or fencing. In this way, the variations in the subjects' body composition, especially in the SMT group, are partly explained by training in some sports that total almost 90% of all the modalities practiced.

Considering, for example, the practice of team sports, such as soccer and volleyball, as was verified in different studies [47,48] that the longer the training time, the greater the phase angle and the lower the resistance values, indicating higher lean body mass values and, consequently, higher musculoskeletal mass. Micheli et al. [47] demonstrated that elite Italian professional football players (Series A and B) trained 9 weeks more throughout the year, with three more training sessions per week and one more game per week than amateur players, and consequently presented 7% higher and 8% lower phase angle and resistance values, respectively, indicating better body compositing status. In contrast, in our study, the differences between the phase angle (MT × SMT) were due to the greater values of reactance for those who practice sports activities. Such a situation can be explained by the amount of military training of both groups, which naturally provides good physical fitness, with lower resistance values. The SMT group, with specific sports training, presented higher values of phase angle, explained by higher reactance values, influenced mainly by greater amount of cell mass.

The main strengths of the present research are related to the application of a standardized protocol with cross-sectional and longitudinal measures, the use of reference techniques for body composition assessment in association with specific BIVA, and a wellcontrolled sample. However, some limitations are also present and are related to the poor representation of cases in the different disciplines, characterized by different training protocols, which made it impossible to recognize possible differences in body composition changes and in the underlying physiological mechanisms. Further, there was some disagreement between methods (anthropometry, DXA, and specific BIVA), likely related to regional differences of body components, which should be better analyzed by means of localized body composition analysis.

#### **5. Conclusions**

In conclusion, this research showed that spBIVA is a suitable technique for the assessment of body composition in the population studied. The effect of training on body composition was independent of sample characteristics or type of physical exercise: muscle mass increased, while the percentage of fat mass remained unchanged.

**Author Contributions:** Conceptualization, L.A.G., E.M. and R.B.; methodology, R.D.L. and E.M.G.; software, E.M.; formal analysis, E.M. and L.A.G.; investigation, R.D.L., E.M.G., J.H.B., M.A.P., V.X.C. and G.G.-J.; resources, V.X.C.; data curation, R.D.L., J.H.B. and E.M.G.; writing—original draft preparation, L.A.G., E.M. and R.B.; writing—review and editing, L.A.G., E.M. and R.B.; visualization, E.M.; project administration, G.G.-J. and E.M.G. 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 the Ethics Committee of the School of Medical Sciences, University of Campinas. All procedures followed Resolution No. 466 of 2012 of the National Health Council of the Ministry of Health of Brazil.

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

**Acknowledgments:** The authors are grateful to the officers and the cadets of the "Preparatory School Cadets Army" (EsPCEx) of Campinas-SP for their authorization and collaboration in this study.

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

#### **References**


## *Article* **The Relationship between VO<sup>2</sup> and Muscle Deoxygenation Kinetics and Upper Body Repeated Sprint Performance in Trained Judokas and Healthy Individuals**

**André Antunes <sup>1</sup> , Christophe Domingos <sup>2</sup> , Luísa Diniz <sup>1</sup> , Cristina P. Monteiro 1,3 , Mário C. Espada 2,4 , Francisco B. Alves 1,3 and Joana F. Reis 1,3,\***


**Abstract:** The present study sought to investigate if faster upper body oxygen uptake (VO<sup>2</sup> ) and hemoglobin/myoglobin deoxygenation ([HHb]) kinetics during heavy intensity exercise were associated with a greater upper body repeated-sprint ability (RSA) performance in a group of judokas and in a group of individuals of heterogenous fitness level. Eight judokas (JT) and seven untrained healthy participants (UT) completed an incremental step test, two heavy intensity square-wave transitions and an upper body RSA test consisting of four 15 s sprints, with 45 s rest, from which the experimental data were obtained. In the JT group, VO<sup>2</sup> kinetics, [HHb] kinetics and the parameters determined in the incremental test were not associated with RSA. However, when the two groups were combined, the amplitude of the primary phase VO<sup>2</sup> and [HHb] were positively associated with the accumulated work in the four sprints (ΣWork). Additionally, maximal aerobic power (MAP), peak VO<sup>2</sup> and the first ventilatory threshold (VT<sup>1</sup> ) showed a positive correlation with ΣWork and an inverse correlation with the decrease in peak power output (Dec-PPO) between the first and fourth sprints. Faster VO<sup>2</sup> and [HHb] kinetics do not seem to be associated with an increased upper body RSA in JT. However, other variables of aerobic fitness seem to be associated with an increased upper body RSA performance in a group of individuals with heterogeneous fitness level.

**Keywords:** VO<sup>2</sup> kinetics; muscle oxygenation; judo; upper body; arm crank; near-infrared spectroscopy; repeated sprint ability

#### **1. Introduction**

Judo is a technically and tactically demanding sport, involving several intermittent efforts of high-intensity activity, interceded by short rest periods [1]. An official, seniorlevel match may last up to 4 min, and judokas may perform up to seven matches during a tournament, including preliminary rounds, main rounds and finals, all in the same day. This sport is reported to rely heavily on upper body strength and power [2], and it has also been suggested that the high-intensity efforts that occur throughout a match are mainly supported by anaerobic energy systems, while the oxidative energy system may be crucial to the recovery process in between high-intensity efforts and between matches [2].

It has been shown that at the elite level, judo contest winners have a higher activity profile over the course of a match, performing more offensive actions per match (56 offensive actions/match in gold medalists vs. 49 offensive actions/match in silver medalists) [3].

**Citation:** Antunes, A.; Domingos, C.; Diniz, L.; Monteiro, C.P.; Espada, M.C.; Alves, F.B.; Reis, J.F. The Relationship between VO<sup>2</sup> and Muscle Deoxygenation Kinetics and Upper Body Repeated Sprint Performance in Trained Judokas and Healthy Individuals. *Int. J. Environ. Res. Public Health* **2022**, *19*, 861. https://doi.org/10.3390/ ijerph19020861

Academic Editor: Paul B. Tchounwou

Received: 5 November 2021 Accepted: 11 January 2022 Published: 13 January 2022

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

**Copyright:** © 2022 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/).

These results highlight that for high-level judo athletes, the ability to perform multiple high-intensity actions over time may be a crucial aspect in determining a contest winner. Therefore, the study of the factors underlying the ability to perform more high-intensity actions over the course of a match in this group of athletes seems to be of relevance.

Many sport activities rely on the ability to repeat several efforts over time. The ability to perform these activities seems to be dependent on the ability to quickly restore phosphocreatine (PCr) stores, ensuring that high rates of muscular work can be sustained over the course of several high-intensity bouts [4]. The ability to restore PCr stores back to near-resting level seems to be dependent on muscular oxidative capacity [5]. Moreover, as these short-duration efforts are repeated over time, the contribution of the oxidative energy system seems to increase [6]. The ability to quickly attain a high rate of ATP resynthesis from oxidative phosphorylation, as expressed by the rate at which oxygen uptake (VO2) rises, seems to be an important aspect in delaying the onset of fatigue [7]. As this quick rise is associated with a reduction in oxygen deficit [8], it can contribute to an increased high-intensity exercise tolerance.

The speed at which VO<sup>2</sup> rises to attain a given value necessary to support the exercise workload can be characterized by the time constant of the primary component of VO<sup>2</sup> kinetics (τphase II), which represents the time necessary for 63% of the final oxygen uptake response to be complete [9–11]. In response to moderate-intensity workloads, the VO<sup>2</sup> response is characterized by three phases [12,13]: Phase I is the cardiodynamic phase, and corresponds to the phase during which VO<sup>2</sup> rises as a consequence of increased pulmonary blood flow. Phase II corresponds to the primary component, where VO<sup>2</sup> rises in an exponential manner, until a steady-state VO<sup>2</sup> is achieved, corresponding to phase III. The profile of VO<sup>2</sup> response during phase II seems to closely reflect muscle VO<sup>2</sup> profile [13–15]. During heavy-intensity constant load exercise, the attainment of a steady-state VO<sup>2</sup> is delayed due to the rise in VO2, which exceeds the expected values of VO<sup>2</sup> based on the VO2–exercise intensity relationship established during submaximal (moderate intensity domain) workloads, coinciding with the emergence of a VO<sup>2</sup> slow component (VO2SC).

Faster VO<sup>2</sup> kinetics, characterized by shorter values for τphase II, have been observed in trained individuals [16–20] and have been associated with a smaller decrease in speed over a repeated-sprint ability test (RSA) in a group of soccer players [21]. Moreover, a shorter τ has been associated with longer high-speed running distances in a group of young high-level soccer athletes [22]. Trained individuals have also been shown to have a smaller VO2SC at a given workload compared to untrained individuals [23,24], which has been associated with an increase in the ability to sustain high-intensity exercise workloads over time at a given absolute workload [25]. Some studies have also shown that upper body trained individuals have faster VO<sup>2</sup> kinetics compared to untrained subjects [18,19]. To our knowledge, no study to date has sought to understand the characteristics of the response of VO<sup>2</sup> kinetics during heavy-intensity upper body exercise, nor attempted to establish a relationship between VO<sup>2</sup> kinetics variables and upper body high-intensity exercise performance in a group of judo athletes.

Near infrared spectroscopy (NIRS) has been used to examine the relative matching of O<sup>2</sup> delivery with tissue oxygen utilization during constant-workload exercise transitions [26]. The hemoglobin/myoglobin deoxygenation ([HHb]) signal derived from NIRS measurements is reported to reflect the balance between O<sup>2</sup> delivery and O<sup>2</sup> utilization and has been used as a non-invasive index of O<sup>2</sup> extraction from muscle capillaries during exercise [27,28].

Given that several variables of aerobic fitness, such as maximal aerobic speed [29] and τphase II [21], have been associated with increased RSA, we hypothesized that participants with faster upper body VO<sup>2</sup> and [HHb] kinetics would achieve a higher upper body RSA performance. Therefore, the present study sought to understand if faster VO<sup>2</sup> and [HHb] kinetics were associated with a higher performance, expressed as a lower decrease in peak power output (PPO) and mean power output (MPO), as well as a higher accumulated work (ΣW), over the course of an upper body RSA test in a group of judo athletes and in a group of healthy individuals of heterogenous fitness level.

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

Eight male judo athletes (JT) (age 21.1 ± 3.0 years, height 172.3 ± 4.5 cm, body mass 71.5 ± 7.1 kg, triceps skinfold thickness 4.5 ± 0.7 mm) and seven male untrained healthy participants (UT) (age 22.6 ± 1.0 years, height 172.7 ± 4.5 cm, body mass 64.3 ± 5.8 kg, triceps skinfold thickness 6.6 ± 1.6 mm) volunteered to participate in the study. The JT were all black belts, of national (placed in 1st–7th place in the national championships) and international (placed 3rd–9th place in European and World cups) level and had been training (13.1 ± 2.8 year) and competing regularly (6.0 ± 1.3 competitions in the previous year) for at least three years; the UT were not involved in any upper body exercise modalities, although they were all healthy, active (at least 150 min. of physical activity/wk) individuals [30].

None of the participants were suffering from any upper body injuries at the time of testing or recovering from any major upper body injury that had occurred in the past 12 months, nor taking any medicine. None of the individuals of JT were cutting weight nor preparing for a major competition at the time that the testing sessions were undertaken.

In order to determine the sample size for the present study, a priori statistical power analysis was performed with G-Power [31] based on the studies of Dupont [21] and McNarry [32], aiming for a power of 85% (alpha = 0.05, two-tailed). The sample size suggested was of 10 individuals for correlations and 7 for each group for the comparisons. Given the strenuous nature of the tests that were undertaken, and that participants could drop out of the study at any time, additional participants were recruited for a total sample of 15 individuals.

All the participants were fully informed of any risks before giving their written informed consent to participate in the study, in accordance with the requirements outlined by the Ethics Committee of the Faculty of Human Kinetics of the University of Lisbon (approval code 42/2021) and in accordance with the Declaration of Helsinki [33].

The participants were required to report to the laboratory on three occasions. To avoid circadian rhythm effects, testing occurred at the same time of day, with each session separated by at least 48 h and all testing sessions were completed within 2 weeks. All subjects were required to present themselves in the laboratory with comfortable clothes, in a rested and hydrated state, to refrain from drinking any sort of alcoholic beverages at least 24 h prior to each testing session, and from eating or taking caffeine 3 h prior to each test.

All tests were performed on an electronically braked arm crank ergometer (Lode Angio, Groningen, Netherlands). In their first session, participants performed an incremental step test to determine maximal aerobic power (MAP), peak oxygen uptake (peak VO2), first ventilatory threshold (VT1) and its respective workload (W\_VT1). In the second session, participants performed two heavy-intensity square-wave exercise transitions to determine VO<sup>2</sup> kinetics and [HHb] kinetics of triceps brachii. Each transition began with 3 min of baseline cranking at 0 W, following which the transition workload was imposed. Each square-wave transition was separated by 1 h of passive recovery. In the third session, participants completed a standardized warm-up, followed by an RSA test, consisting of four 15 s upper body all-out sprints, each interceded by 45 s of passive recovery in between.

#### *2.1. Gas Exchange and Muscle Deoxygenation Measurements*

Gas exchange variables were collected breath-by-breath with a gas analyzer (Meta-Max 3B, Cortex Biophysik, Leipzig, Germany), after calibration according to the manufacturer's instructions.

The changes in the [HHb] signal in the local circulation of the long head of the triceps brachii were monitored using a continuous-wave tissue oximeter (NIMO, Nirox, Brescia, Italy) using the NIRS technique. In order to reliably collect the NIRS signal, the local skin of each participant's upper arm was initially shaved and cleaned. A probe consisting of a

photon emitter and a photon receptor, emitting and detecting near-IR beams with three different wavelengths (685 nm, 850 nm and 980 nm), was attached to the skin surface, and secured with tape and then covered with an optically dense elastic bandage in order to minimize movement, prevent loss of near-IR signal and stray light interference, and also to constrain the signal emission-reception site. The signal was sampled at a frequency of 40 Hz. To account for the effects of adipose tissue thickness on the NIRS signal, the skinfold thickness at the site where NIRS probes were placed was measured with a caliper (Slim Guide Caliper, Creative Health, Ann Arbor, MI, USA) and a correction factor was used in the analysis software (Nimo Data Analysis Peak). All NIRS measurements were conducted on the right limb, and [HHb] was monitored during the second and third testing sessions (square-wave transitions and RSA test, respectively). The validity and limitations associated with the measurements obtained via this oximeter have been reviewed by Rovati and associates [34].

#### *2.2. Incremental Step Test*

Participants performed an incremental exercise test for determination of MAP, peak VO2, VT<sup>1</sup> and W\_VT1. Participants performed a 3 min step of baseline cranking at 0 W, following which the power was increased 15 W each min (step) until participants reached voluntary exhaustion. The participants were instructed to crank the wheel at the rate of 70 rotations per minute (rpm), grabbing the handles of the ergometer in a standard position, in which they stood upright with their feet shoulder width apart, flat on the floor, and with their shoulder joint levelled with the pedal crank axle. Handle height and ergometer configuration were recorded and reproduced in subsequent tests. The present incremental test protocol's characteristics were based on the protocols of Koppo and associates [20] and Schneider and associates [35], which also studied upper body VO<sup>2</sup> kinetics of a group of heterogenous fitness level. Breath-by-breath pulmonary gas-exchange data were collected continuously during the incremental step test. The peak VO<sup>2</sup> was taken as the highest 30 s average value attained before the participants reached volitional exhaustion. The MAP was defined as the minimal workload which elicited peak VO2.

The VT<sup>1</sup> was estimated by monitoring the ventilatory equivalents for oxygen (VE/VO2) and carbon dioxide (VE/VCO2), determined by inspection to define the point at which an increase in VE/VO<sup>2</sup> was observed, with no concomitant increase in VE/VCO<sup>2</sup> [36]. The workload over which these responses were observed was defined as the W\_VT1. Throughout the test, heart rate (HR) was monitored continuously (ONRHYTHM 500, Kalenji, France) and the highest HR value observed in the last stage of exercise was registered as peak HR. The workload associated with VT<sup>1</sup> was used to determine the intensity for the square-wave transitions, which was set at 20%∆, calculated as W\_VT<sup>1</sup> plus 20% of the difference between the W\_VT<sup>1</sup> and the MAP.

#### *2.3. Square-Wave Transitions*

The participants performed two square-wave constant workload transitions for the determination of VO<sup>2</sup> and [HHb] kinetics, with a workload of 20%∆, corresponding to a heavy-intensity workload. After a 3 min period of baseline cranking at 0 W, the target workload was imposed. Each square-wave transition lasted 6 min and the transitions were separated by 1 h of passive rest. Given the lower exercise tolerance associated with upper body exercise, a 20%∆ workload was chosen, to ensure that the subjects were working in the heavy-intensity exercise domain without incurring excessive fatigue, which would compromise performance in the subsequent square-wave transition, for both groups, and therefore confound the underlying physiological response.

The VO<sup>2</sup> data were collected breath-by-breath from each transition and were examined to exclude errant breaths and values lying more than 4 standard deviations from the local mean (based on 5 breaths), and subsequently linearly interpolated to provide 1 s values. The data from the two transitions were then time aligned to the start of exercise and averaged to reduce signal noise and enhance the underlying physiological response characteristics [37].

VO<sup>2</sup> kinetics parameters were calculated by an iterative procedure, minimizing the sum of the residuals, according to the following bi-exponential model:

$$\text{VO}\_2\text{ (t)} = \text{VO}\_{2\text{baseline}} + \text{A}\left[1 - \text{e}^{-(t-\text{TDp})/\text{TP}}\right] + \text{Asc}\left[1 - \text{e}^{-(t-\text{TDsc})/\text{TP}}\right]$$

where VO<sup>2</sup> (*t*) represents the absolute VO<sup>2</sup> at a given time *t*, VO2baseline represents the mean VO<sup>2</sup> under unloaded conditions 30 s prior to the work transition; A, TDp, and τ represent the amplitude, time delay, and time constant, of the phase II of the increase in VO<sup>2</sup> after the onset of exercise, and Asc, TDsc, and τsc represent the amplitude of the slow component, time delay before the onset of, and time constant of the slow component phase of VO<sup>2</sup> kinetics, respectively [38].

The end-exercise VO<sup>2</sup> was defined as the mean VO<sup>2</sup> value obtained in the last 30 s of the 6 min constant workload transitions. The first 20 s of VO<sup>2</sup> data were excluded from the analysis to remove the influence of the cardiodynamic phase on the subsequent response [39]. Because the asymptotic value of the second function is not necessarily reached at the end of the exercise, the amplitude of the slow component was defined as

$$\mathbf{A}'\_{\mathbf{SC}} = \mathbf{Asc} \left[ 1 - \mathbf{e}^{-(\mathrm{te} - \mathrm{TDsc}/\tau \mathrm{sc})} \right],$$

where *t*e was the time at the end of the exercise bout [20].

Throughout each square-wave transition, the [HHb] signal was monitored in order to provide a non-invasive surrogate of the changes in O<sup>2</sup> saturation of the hemoglobin/ myoglobin in the local circulation of the long head of the triceps brachii.

The [HHb] data were normalized to resting values, considering the average of the 3 min rest before the unloaded pedaling, and the [HHb] response was characterized according to a monoexponential model, with a timed-delay (TD) at the onset of exercise, followed by an exponential increase [27] until the end of the exercise period:

$$\left[\text{HCHO}\right](t) = \left[\text{HCHO}\right]\_{\text{baseline}} + \text{AH} \text{H} \text{b} \left[1 - \text{e}^{-(t - \text{TDHHb})/\tau \text{HHb}}\right]$$

where [HHb] (*t*) represents the [HHb] at a given time *t*, [HHb]baseline represents the 60 s average [HHb] prior to the participant gripping the handles, and A [HHb] and τ [HHb] correspond to the amplitude and time constant of the exponential phase of [HHb] kinetics, respectively. The TD was defined as the time between the onset of exercise and the time at which a first increase in the [HHb] signal was observed [40], which was determined by visual inspection. [HHb] data were fit from the time of initial increase in [HHb] to 180 s. The exponential-like phase of the [HHb] kinetics was also characterized by an "effective" time constant (τ 0 ), which corresponded to the sum of TD and τ [40].

#### *2.4. Repeated Sprint Exercise*

The upper body RSA test consisted of four 15 s all-out sprints, each separated by 45 s of passive rest. Participants performed a 6 min warm-up at 30 W with a cadence of 70 rpm, with three brief sprints (<5 s duration) during the last 3 min of the warm-up. Participants were then given 2 min of rest before commencing the upper RSA test. Thirty seconds before the start of the test, the participants were asked to grip the ergometer handles. Throughout the whole test, the participants were verbally encouraged to give their maximum effort.

The exercise workload was set at 5% of the body mass of each individual [41]. The peak power output (PPO) and mean power output (MPO) attained during each sprint were monitored, and the total work performed (Work) during each sprint was derived as the integral of power output over the 15 s period.

The 15 s work period was chosen by considering the data reported by Soriano and associates [42] based on the sum of average time it took male judokas to come to grips (8.4 ± 3.1 s), establish a grip and control their opponent (6.1 ± 3.5 s) and execute a throw (1.3 ± 0.5 s). The 45 s rest period duration was chosen as a compromise between what is observed in a typical judo match (2:1 work-to-rest ratio) [1,2], and the typical work density that has been reported in several RSA studies (1:6-8 work-to-rest ratio) [4,6], to ensure that the work period matched what typically occurs throughout a match, as exercise intensity and duration are the main determinants of energy system specificity [43], while allowing participants to sustain the power output over the course of several high-intensity bouts. A set of variables were computed, in order to characterize overall RSA performance:

Dec-PPO (Decrease in PPO) = PPO 1st sprint − PPO 4th sprint

Dec-MPO (Decrease in MPO) = MPO 1st sprint − MPO 4th sprint

ΣWork (Accumulated work) = ∑ 4*th sprint* <sup>1</sup>*st sprint* Work performed

Throughout each sprint, the [HHb] signal was monitored, and the data collected were used to compute the maximal [HHb] attained in each sprint (Max. A [HHb]).

#### *2.5. Statistical Analysis*

The results are presented as means ± SD. The Shapiro–Wilk test was used to verify the normal distribution of the data for each variable [44]. Unpaired t-tests were used to compare the differences between groups regarding each variable. Pearson product correlations were used to determine the correlation between variables. In order to determine if a linear relationship could be established between a variable of interest and other independent variables, a stepwise regression analysis was performed, using Peak VO2, MAP, VT1\_VO2, Aphase II, τphase II, the effective slow component amplitude (A'SC), A [HHb], τ' [HHb], Max. A [HHb] 1, Max. A [HHb] 2, Max. A [HHb] 3 and Max. A [HHb] 4 as independent variables, and Σ Work as the dependent variable. The collected data regarding each individual variable were analyzed as a whole, considering all participants as a single heterogeneous group, and were analyzed separately by groups. The effect size for the differences between groups was calculated based on the ratio between the difference in the mean values and the weighted pooled SD. The threshold values for Hedges' effect size (ES, g) statistics were characterized according to the following scale [45]: <0.20 = negligible effect, 0.20–0.49 = small effect, 0.50–0.79 = moderate effect, ≥0.80 = large effect.

#### **3. Results**

#### *3.1. Incremental Step Test*

Mean and standard deviation of the variables obtained by each group of participants in the incremental test are depicted in Table 1.


**Table 1.** Physiological responses attained by UT and JT participants in the incremental step test.

UT, untrained participants; JT, judo athletes; Peak VO2, peak oxygen consumption; MAP, maximal aerobic power; VT1\_VO2, oxygen consumption rate at the onset of the first ventilatory threshold; VT1\_W, workload at the onset of the first ventilatory threshold; 20% ∆W, Workload corresponding to the sum of VT1\_W plus 20% of the difference between the MAP and VT1\_W; Peak HR, Peak heart rate achieved during the incremental test; \* Significant differences between groups for *p* < 0.05.

The JT group displayed higher Peak VO2, MAP, W\_VT<sup>1</sup> and 20%∆ than the UT group. A large effect size was observed for Peak VO<sup>2</sup> (g = 1.8), MAP (g = 3.0), W\_VT1 (g = 1.7) and 20%∆ (g = 1.9). VT<sup>1</sup> \_VO<sup>2</sup> and peak HR were not different between groups.

#### *3.2. Square-Wave Transitions*

Table 2 shows the VO<sup>2</sup> kinetics variables obtained by the two groups of participants in the heavy-intensity square-wave transitions.

**Table 2.** VO<sup>2</sup> kinetics parameters in the heavy-intensity square-wave transitions for each group.


UT, untrained participants; JT, judo athletes; VO2baseline, baseline oxygen consumption rate; Aphase II, Amplitude of the primary phase; τphase II, Time constant of the primary phase; TDphase II, Time delay of the primary phase; Asc, Amplitude of the slow component phase; A'sc, Effective amplitude of the slow component; TDSC, Time delay of the slow component phase; τsc, Time constant of the slow component phase; EE VO2, oxygen uptake rate observed at the end of the square-wave transitions; A'SC/EE VO2, Effective amplitude of the slow component relative to the oxygen consumption rate observed at the end of the square-wave transitions; Sum of residuals, Discrepancy in a dataset that is not explained by the model. \* Significant differences between groups for *p* < 0.05.

The JT group presented significantly lower τphase II and higher τSC than UT. None of the other VO<sup>2</sup> kinetics parameters were different between groups. Large effect sizes were observed for τphase II (g = 1.2) and τSC (g = 1.2).

The normalized parameters of the response of [HHb] in the heavy-intensity exercise transitions for the two groups are presented in Table 3.

**Table 3.** Observed [HHb] kinetics during the heavy-intensity exercise square-wave transitions.


τ' [HHb], Effective time constant of [HHb] kinetics; A [HHb], Amplitude of response of hemoglobin/myoglobin deoxygenation. \* Significant differences between groups for *p* < 0.05.

The JT group presented significantly higher A [HHb] than the untrained participants, whereas τ' was not significantly different between groups. A large effect size was observed for A [HHb] (g = 1.1)

#### *3.3. Upper Body RSA Protocol*

The Dec-PPO, Dec-MPO and ΣWork of the RSA test obtained for each group are shown in Table 4.



Dec-PPO, decrease in the peak power output between the first and fourth sprint, Dec-MPO, decrease in mean power output between the first and fourth sprint; ΣWork, accumulated work in the RSA test. \* Significantly different from UT values for *p* < 0.05.

There were significant differences between groups in the Dec-PPO, with the JT group displaying a lower Dec-PPO over the course of the upper body RSA protocol. Significant differences were also observed between groups in the ΣWork, with a larger mean value of ΣWork being observed in the JT group. A large effect size was observed for Dec-PPO (g= 1.7) and ΣWork (g = 2.5).

#### *3.4. Correlations between RSA and the Physiological Variables Obtained in the Square-Wave Transitions and Incremental Step Test*

Analysis considering the separate groups did not show any correlation between the performance parameters in the upper body RSA protocol and the parameters determined in the incremental test or square-wave transitions, neither in the JT group nor in the UT group. However, considering the heterogeneous group consisting of the whole sample of participants, significant correlations were found between MAP, peak VO<sup>2</sup> and VT1\_VO<sup>2</sup> and Dec-PPO or ΣWork (Figure 1). *Int. J. Environ. Res. Public Health* **2021**, *18*, x FOR PEER REVIEW 9 of 17

**Figure 1.** Relationships (correlations) between maximal aerobic power (MAP), peak oxygen consumption (peak VO2) and oxygen consumption at the first ventilatory threshold (VT1\_VO2) achieved in the incremental test and the decrement in peak power output (Dec-PPO) and accumulated work (ΣWork) over the course of the upper body repeated sprint (RSA) test, observed in the group of heterogeneous fitness level (whole sample). **Figure 1.** Relationships (correlations) between maximal aerobic power (MAP), peak oxygen consumption (peak VO<sup>2</sup> ) and oxygen consumption at the first ventilatory threshold (VT1\_VO<sup>2</sup> ) achieved in the incremental test and the decrement in peak power output (Dec-PPO) and accumulated work (ΣWork) over the course of the upper body repeated sprint (RSA) test, observed in the group of heterogeneous fitness level (whole sample).

Additionally, for the UT, a significant correlation was found between Aphase II and Dec-

PPO or ΣWork and between A [HHb] and ΣWork, as highlighted in Figure 2.

Additionally, for the UT, a significant correlation was found between Aphase II and Dec-PPO or ΣWork and between A [HHb] and ΣWork, as highlighted in Figure 2. *Int. J. Environ. Res. Public Health* **2021**, *18*, x FOR PEER REVIEW 10 of 17

**Figure 2.** Relationships (correlations) between A of the phase II VO2 kinetics, A[HHb] observed during the square-wave transitions and the decrement in peak power output (Dec-PPO) or accumulated work (ΣWork) over the course of the upper body RSA test, observed in the group of heterogeneous fitness level (whole sample). **Figure 2.** Relationships (correlations) between A of the phase II VO<sup>2</sup> kinetics, A [HHb] observed during the square-wave transitions and the decrement in peak power output (Dec-PPO) or accumulated work (ΣWork) over the course of the upper body RSA test, observed in the group of heterogeneous fitness level (whole sample).

No other significant correlations were observed between any of the performance variables and the VO2 kinetics and [HHb] kinetics variables, whether when we consider each group separately or when we analyze the whole sample as a single heterogeneous group. No other significant correlations were observed between any of the performance variables and the VO<sup>2</sup> kinetics and [HHb] kinetics variables, whether when we consider each group separately or when we analyze the whole sample as a single heterogeneous group.

#### *3.5. Predictive Model for ΣWork over the Course of the RSA Protocol*  Considering the whole sample as a single heterogeneous group, a significant regres-*3.5. Predictive Model for* Σ*Work over the Course of the RSA Protocol*

the whole group of participants.

sion equation was found (F (2.12) = 12.737; *p* < 0.001) with an r2 of 0.68, presented in Table 5, for which the main predictors were Peak VO2 and Max. A [HHb] 4. The model had a yintercept at 132.9 kJ, with the ΣWork increasing 0.8 kJ per each unit of increase in Peak VO2 and increasing 0.16 kJ per each unit of increase in Max. A [HHb]. These two variables were found to explain 68% of the ΣWork during the RSA protocol. **Table 5.** Predictors of accumulated work over the course of the upper body repeated sprint test for Considering the whole sample as a single heterogeneous group, a significant regression equation was found (F (2.12) = 12.737; *p* < 0.001) with an r<sup>2</sup> of 0.68, presented in Table 5, for which the main predictors were Peak VO<sup>2</sup> and Max. A [HHb] 4. The model had a y-intercept at 132.9 kJ, with the ΣWork increasing 0.8 kJ per each unit of increase in Peak VO<sup>2</sup> and increasing 0.16 kJ per each unit of increase in Max. A [HHb]. These two variables were found to explain 68% of the ΣWork during the RSA protocol.

**Accumulated Work over the Course of the Upper Body RSA Protocol (kJ) R R2 Adj. R2 SEE** *p*  ΣWork (kJ) = 132.9 + 0.8 Peak VO2 + 0.16 Max. A [HHb] 4 0.82 0.68 0.63 5.1 0.001 **Table 5.** Predictors of accumulated work over the course of the upper body repeated sprint test for the whole group of participants.


Peak VO2, the highest 30 s average VO<sup>2</sup> attained over the course of the incremental test, Max. A [HHb] 4, the maximal [HHb] achieved in the fourth repetition of the upper body RSA test. All other variables were excluded from the model.

When each group was analyzed separately, no significant regression equation to predict ΣWork was found.

#### **4. Discussion**

To our knowledge, this was the first study to date which analyzed the relationship between parameters of aerobic fitness and upper body RSA performance, both in a heterogeneous sample consisting of trained and untrained participants, and more specifically, in a group of trained judo athletes. The present study revealed that a shorter τphase II and τ' [HHb] were not correlated to a lower decrease in PO over the course of an upper body RSA protocol, nor with a higher ΣWork. However, other variables of aerobic fitness, namely Peak VO2, MAP, VT1\_ VO2, and A of the phase II of VO<sup>2</sup> kinetics, were inversely correlated with the decrease in PO and directly correlated with the ΣWork over the course of the upper body RSA protocol, in the sample comprising both UT and JT groups. A [HHb] was also directly correlated with a higher ΣWork over the course of the upper body RSA protocol, in the sample comprising both UT and JT groups.

It has been proposed that VO<sup>2</sup> kinetics influences high-intensity exercise performance [46–49]. Several authors proposed that faster VO<sup>2</sup> kinetics, as expressed by a shorter τphase II, are associated with the ability to support a given workload without tapping into O<sup>2</sup> deficit-related metabolic processes [7] and that faster VO<sup>2</sup> kinetics are related with faster [PCr] recovery kinetics following exercise [48], two potential aspects that may determine exercise tolerance during repeated high-intensity exercise.

The results observed in the current study indicate that there is no significant correlation between pulmonary τphase II and upper body RSA performance, namely between τphase II and the Dec-PPO, Dec-MPO or ΣWork over the course of the four sprints, either when we consider the sample of participants as a single heterogeneous group, or when we analyze the JT separately. These results contradict our main hypothesis, in which we proposed that a shorter τphase II, would be associated with improved RSA performance variables, namely a higher ΣWork and smaller Dec-PPO and Dec-MPO.

Dupont and associates [21] have previously reported a significant direct correlation between τphase II and relative decrease in speed and total work performed over the course of an RSA protocol in a group of soccer players. Rampini and associates [49] also found a direct and significant (r = 0.62; *p* < 0.05) association between τphase II and the relative decrease in sprint speed over the course of six 40 m (20 m run-and-back) shuttle sprints separated by 20 s of passive recovery.

However, in line with our results, Buchheit [50] found no correlations between RSA performance and τphase II, reporting that stepwise multiple regression analysis showed that mean repeated-sprint time, best sprint time and maximal aerobic speed were the only significant predictors of RSA performance. Accordingly, Christensen and associates [51] also found that τphase II was not associated with better RSA performance in a group of soccer players, although the changes in τphase II after a speed-endurance training program were associated with changes in RSA performance.

The studies mentioned above [21,50,51] involved running activities, utilizing a different set of muscle groups, in a different set of participants, exposed to very different training regimens compared to the individuals involved in the present study. However, they allow us to make some assertions regarding the observed results. The protocol used in the study by Dupont and associates [21] involved significantly more volume, and involved active recovery periods between sprints (15 × 40 m sprints, interceded with 25 s of active recovery), which may have biased the contribution of the aerobic energy system to the total work performed, and therefore the degree of association between RSA performance and τphase II, while Buchheit [50] using a set of lower volume RSA protocols (10 × 30 m; 6 × 2 × 15 m; 6 × 16 m; 6 × 16 m; 20 × 15 m; 6 × 25 m) did not find such correlations. It is possible that an association may only be found between τphase II and RSA performance involving a high volume of RSA activity, given the increased contribution of the aerobic system as the number of sprints increases [6]. It is possible that an association between

τphase II and RSA performance would be found if a protocol involving a higher volume of sprints had been used. Moreover, the possibility that upper body VO<sup>2</sup> kinetics variables may play a more important role in judo contests that drag over a longer period of time should also be considered.

The results of the present study indicate that there is an inverse correlation between the MAP attained in the incremental step test and the Dec-PPO for the whole group of participants. Given that MAP is associated with training status [52], these results reveal that participants who are "aerobically" trained to a greater extent display an increased ability to resist decreases in PO over the course of repeated high-intensity exercise. Interestingly, this association was not observed for the JT group. It may be that the MAP of the individuals of the JT group was too similar (low range or spread of values) for any significant correlation to be established. It seems that there is a certain fitness threshold for which this association is valid, and that above this fitness threshold, other variables are more important in determining upper body RSA performance.

An inverse correlation was also found between the peak VO<sup>2</sup> and the Dec-PPO and a direct correlation between peak VO<sup>2</sup> and ΣWork. Similar findings have been reported by other authors [4,6]. The importance of peak VO<sup>2</sup> to RSA performance seems to be two-fold: (1) Across multiple sprints, aerobic ATP provision progressively increases such that aerobic metabolism may contribute as much as 40% of the total energy supply during the final repetitions of an RSA protocol [53]; (2) Enhanced oxygen delivery to muscles post-exercise potentially accelerates the rate of PCr resynthesis, an oxygen-dependent process [53,54], facilitating a faster recovery from high-intensity exercise.

Bishop and associates [55] observed a significant negative correlation (r = −0.50; *p* < 0.05) between VO2max and % decrease in work over the course of 5 x 6 s sprints in a group of female basketball athletes. The authors proposed that athletes with greater VO2max would be able to achieve a higher VO<sup>2</sup> rate throughout each sprint, reducing the contribution of substrate-level phosphorylation to ATP resynthesis, and therefore allowing more work to be done over the course of the RSA protocol [56]. Aguiar and associates [57] also reported a significant negative correlation (r = −0.58, *p* < 0.05) between VO2max and the decrease in performance over the course of an RSA protocol (10 × 35 m sprints, 20 s recovery between sprints) in a heterogeneous group composed of endurance runners, sprinters and healthy individuals. Collectively, these results seem to emphasize the relationship between VO2max and the ability to maintain a high power output over the course of several RSA efforts. This observed association between VO2max and the decrease in work capacity over the course of an RSA task may also be associated with a higher cardiac output (Q) and subsequent increase in muscle blood flow, which may aid post-exercise recovery [58].

Furthermore, since a positive correlation was found between A [HHb] and ΣWork, it seems that repeated sprint performance is enhanced in individuals with higher oxygen extraction during heavy intensity exercise. The NIRS-derived [HHb] signal has been considered to reflect the ratio between muscle O<sup>2</sup> delivery and demand, and therefore has been considered an index of muscle O<sup>2</sup> extraction [27,59]. A higher A [HHb] has been associated with a greater muscle oxygen extraction following the onset of exercise [60,61] and has been shown to increase following training [62]. This indicates that a greater oxygen extraction at the onset of exercise and in repeated sprints may have an important role in the ability to maintain a constant performance over the course of several upper body high-intensity efforts.

Moreover, peak VO<sup>2</sup> and maximal [HHb] achieved in the fourth sprint were found to be significant predictors of ΣWork over the course of the four sprints, which seems to indicate that these aerobic fitness variables contribute to performance in repeated sprints. Specifically, both central and peripheral determinants of oxygen uptake contribute to performance during high intensity efforts where the anaerobic component is predominant. These associations were not observed when we consider the JT group separately, probably due to the more homogeneous response in these parameters in this restricted sample. In light of the observed results, it seems that aerobic fitness variables are associated with and

increased upper body RSA performance in a group of individuals of heterogenous fitness. However, it is possible that once a certain level of upper body aerobic fitness is attained, other physiological and fitness variables may play a more important role in determining upper body RSA performance.

Complementary to this, the present study seems to indicate that there are significant differences between JT and UT participants in regard to upper body VO<sup>2</sup> kinetics parameters. No previous study has reported upper body VO<sup>2</sup> kinetics parameters in a group of JT. The τphase II values observed in this group of athletes are similar to those found by Koppo and associates [20] in a group of physically active males (τphase II = 48 ± 12 s). The values of VO<sup>2</sup> kinetics parameters for the group of UT participants observed in this study are similar to those reported by Schneider and associates [35] in a group of untrained participants (τphase II = 66 ± 3 s). Both mentioned studies analyzed the VO<sup>2</sup> kinetics response across the same exercise intensity range that was used in the present study. By comparison, the τphase II values observed by Invernizzi and associates [18] in a group of specifically upper body trained participants (elite competitive swimmers; τphase II = 34.3 ± 8.5 s), determined in an arm crank ergometer test, were much shorter than those which were observed in the JT group.

It is possible that judo-specific training may have induced sufficient training adaptations that resulted in a faster VO<sup>2</sup> kinetics response to exercise relative to untrained participants. However, given that judo-specific drills involve a different skeletal muscle function regimen, where isometric muscular actions of the upper body are emphasized, and muscle actions are performed in an intermittent way, the physiological adaptations that occur may involve very different mechanisms than those which are associated with the performance of high-volume, continuous exercise training of moderate–heavy exercise intensity, typical of swimming. This may also explain the similar results observed in the upper body VO<sup>2</sup> kinetics in the JT group compared to a group of physically active males [20].

Several studies have observed that different training programs, performed at different training intensities have the potential to induce adaptations compatible with shorter τphase II of VO<sup>2</sup> kinetics. Studies have observed improvements in VO<sup>2</sup> kinetics with training protocols ranging from low-intensity work at 60% VO<sup>2</sup> max. [62] to sprint-interval training performed at supramaximal intensities [60]. Nevertheless, these observations have been reported for studies involving dynamic, running or cycling exercise, which involve a different set of muscle groups and muscle action regimen compared to judo-specific training. As it has already been noted, judo-specific modalities seem to rely more on upper body musculature [2]. Given that upper body exercise has been associated with different hemodynamic [63] and metabolic responses [64], physiological adaptations may vary considerably compared to other forms of exercise.

McNeil and associates [65] have observed that performing isometric dorsiflexions at 100% of the maximal voluntary contraction (MVC) resulted in significant decreases in NIRS-derived tissue oxygenation compared to performing isometric dorsiflexions at 30% of MVC. Even though these authors reported no significant changes in tibial artery mean blood flow during the course of 60 s of sustained contraction, they suggested that the capillary mean blood flow might have been severely compromised over the course of the sustained exercise, and that this may have compromised tissue oxygenation dynamics throughout the 100% MVC exercise periods [66]. Given that, over the course of judo training and competition drills, athletes are likely to be exposed to similar conditions, muscle oxygen uptake dynamics might be compromised, and in turn, this may influence the type of physiological adaptations that take place.

The current study presents some limitations, which may limit the degree to which we can generalize the observations that were made. The sample size was relatively small, which affects the statistical meaningfulness of the observed results, as well as the degree to which we can extrapolate our conclusions. Nonetheless, it satisfies the requirements of statistical power determined a priori. Moreover, the athletes that participated in the present

study were mostly national-level athletes, although it included some international-level athletes, and was therefore quite heterogenous. It is possible that different associations may have been found if the study had included judo athletes with a higher performance level, and therefore, further conclusions might be drawn regarding the physiological and fitness parameters which may be associated with upper body RSA performance in this group of athletes.

Furthermore, accessing body composition variables (% fat and fat-free mass), and other physical fitness variables (maximal upper body strength, anaerobic power) could have provided further insight regarding the determinants of upper body RSA performance and help explain the differences observed between groups. Although we designed the RSA test according to the observations reported by Soriano and associates [42], where the sum of time to kumi kata (8.4 s), kumi kata (6.1 s) and throwing time (1.3 s) corresponded to approximately 15 s, the duration chosen for the working periods, we recognize that the number of sprints may have been insufficient to match the number of sequences of attacks observed in more prolonged judo matches. It should also be noted that these tests were undertaken in an arm-crank ergometer, which is a general form of upper body dynamic exercise, and therefore the observed performance achieved by each individual in this task bears little resemblance to what actually happens over the course of a judo match. However, given that judo, as a grappling sport, relies heavily on the upper body musculature, a greater ability to preserve the work capacity of these muscle groups throughout a match seems to be a relevant aspect for potentially achieving a greater performance in the context of a judo match.

No studies to date had attempted to study the fitness variables underlying upper body RSA performance, particularly involving judo athletes. Therefore, the present study provides valuable information for further research regarding the variables that determine upper body RSA performance, particularly in this group of athletes, which may shed light regarding the factors that contribute to maintaining a high activity/attack profile over the course of a match.

Future research should consider using different repeated-sprint protocols with more repetitions and possibly a different work:rest period in order to reveal which factors may be associated with improved repeated-sprint ability under different match conditions. Moreover, future studies should include a larger sample of individuals, whether in a group of judokas of heterogenous fitness level/competitive status or in specific groups (international/elite vs. national level athletes), in order to better understand the factors that may determine RSA performance capacity in different groups. Future studies should also seek to include female individuals, in order to understand if different fitness variables influence upper body RSA performance for both sexes.

#### *Practical Applications for Coaches*

In lower-level athletes, developing a higher upper body aerobic fitness through higher volume general or specific exercises/drills would benefit their ability to maintain a higher performance over the course of the several high-intensity sequences of activity that take place during a match. However, for higher level athletes who already have reasonable upper body aerobic fitness, it may be more pertinent to devote time to developing other fitness variables in order to improve the ability to sustain a higher performance over the course of a judo match. The variables that ought to be developed in order to improve upper body RSA ability in higher-level judo athletes warrant further investigations.

#### **5. Conclusions**

The main conclusions of the present study are the following: (1) There seems to be no significant correlation between τphase II or τ' [HHb] and upper body RSA performance in a group of judokas or in a group of subjects with a heterogeneous fitness level; (2) No significant correlations seem to exist between peak VO2, MAP or VT1\_VO<sup>2</sup> and upper body RSA variables in a group of trained judokas; (3) Judokas displayed significantly faster VO<sup>2</sup>

kinetics and higher muscle oxygen extraction in heavy-intensity exercise than untrained participants and; (4) There seems to be a positive association between MAP, peak VO2, VT1\_VO2, Aphase II and A [HHb] and ΣWork over the course of an upper body RSA task in a group of participants of heterogeneous fitness level. Therefore, aerobic fitness variables seem to play an important role in upper body RSA performance in individuals with a heterogenous fitness level.

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

**Funding:** This project was supported by the Portuguese Foundation for Science and Technology, I.P., Grant/Award Number UID/CED/04748/2020. This study was also partially supported by Fundação para a Ciência e Tecnologia, under grant UIDB/00447/2020 to CIPER–Centro Interdisciplinar para o estudo da performance humana (Unit 447).

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of the Faculty of Human Kinetics of the University of Lisbon (approval code 42/2021).

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

**Data Availability Statement:** The data that support the findings of this study are available from the corresponding author upon reasonable request.

**Acknowledgments:** The authors gratefully acknowledge the participants for their time and effort and national judo coaches for helping recruit the participants.

**Conflicts of Interest:** The authors declare no conflict of interest that are directly relevant to the content of this manuscript.

#### **References**


## *Article* **Postactivation Performance Enhancement (PAPE) Increases Vertical Jump in Elite Female Volleyball Players**

**Lamberto Villalon-Gasch , Alfonso Penichet-Tomas , Sergio Sebastia-Amat , Basilio Pueo \* and Jose M. Jimenez-Olmedo**

> Physical Education and Sports, Faculty of Education, University of Alicante, 03690 Alicante, Spain; lamberto.villalon@ua.es (L.V.-G.); alfonso.penichet@ua.es (A.P.-T.); sergio.sebastia@ua.es (S.S.-A.); j.olmedo@ua.es (J.M.J.-O.)

**\*** Correspondence: basilio@ua.es

**Abstract:** The purpose of this study was to verify if a conditioning activity was effective to elicit postactivation performance enhancement (PAPE) and to increase the performance in vertical jump (VJ) in elite female volleyball players. Eleven national Superliga-2 volleyball players (22.6 ± 3.5 years) were randomly assigned to an experimental and control group. Countermovement jumps (CMJ) were performed on eight occasions: before (Pre-PAPE) and after activation (Post-PAPE), after the match (Pre-Match), and after each of the five-match sets (Set 1 to 5). ANOVA showed significantly increased jump performance for the experiment between baseline (Pre-PAPE) and all the following tests: +1.3 cm (Post-PAPE), +3.0 cm (Pre-Match), +4.8 cm (Set 1), +7.3 cm (Set 2), +5.1 cm (Set 3), +3.6 cm (Set 4), and +4.0 cm (Set 5), all showing medium to large effect size (0.7 < ES < 2.4). The performance of the control group did not show significant increases until Set 3 (+3.2 cm) and Set 5 (+2.9 cm), although jump heights were always lower for the control group than the experimental. The use of conditioning activity generates increased VJ performance in Post-PAPE tests and elicited larger PAPE effects that remain until the second set of a volleyball match.

**Keywords:** back squat; countermovement jump; sports performance; PAP; RM; training

#### **1. Introduction**

Vertical jump (VJ) is a good prognosticator of performance in numerous sports that involve explosive actions, including volleyball [1]. The jump height reached by players can be considered a key factor in volleyball. An improvement in height in VJ allows obtaining enhancements in technical actions such as sets, hits, services, or blocks [2] which are decisive to achieve success in a volleyball game [3]. Service, attack, and block effectiveness are the skills more correlated with winning games in volleyball [4–6].

In addition, jumping capacity is correlated to muscular strength [7] since greater muscular strength can lead to modifications in force–time profile resulting in better VJ performance. Numerous strength training methods have been used to improve VJ performance in volleyball, being most of them strength-based methods such as plyometrics, combined training methods as contrast and complex training [8], or routines based on weightlifting and powerlifting [9].

While these VJ improvement methods are long-term effect procedures, other practices are aiming at achieving acute effects on performance, on certain occasions during the competition, (e.g., warm-up). One of these short-term methods to enhance VJ performance is the Postactivation Performance Enhancement (PAPE) [10–12]. This concept has recently been proposed to be used when high-intensity voluntary conditioning contractions lead to enhancement in voluntary muscular performance, and therefore activation is produced in different ways as with postactivation potentiation (PAP) [10,11]. Although PAP and PAPE are related, they can be considered as a different phenomenon, since the mechanisms

**Citation:** Villalon-Gasch, L.; Penichet-Tomas, A.; Sebastia-Amat, S.; Pueo, B.; Jimenez-Olmedo, J.M. Postactivation Performance Enhancement (PAPE) Increases Vertical Jump in Elite Female Volleyball Players. *Int. J. Environ. Res. Public Health* **2022**, *19*, 462. https:// doi.org/10.3390/ijerph19010462

Academic Editors: Catarina Nunes Matias, Stefania Toselli, Cristina Monteiro and Francesco Campa

Received: 26 November 2021 Accepted: 29 December 2021 Published: 1 January 2022

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

**Copyright:** © 2022 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/).

that produce PAP are different from those for PAPE. PAP implies an enhancement in the effectiveness of contraction due to a better pairing of actin and myosin, and is generated by electrostimulation. On the other hand, PAPE is related to phenomena such as muscle temperature, the proportion of water in the muscle fibers, and the number of activated motor units among other causes [12]. Therefore, their effects may appear at different times and intensities [13]. The presence of PAP does not have to imply that PAPE is generated [14], even so, PAPE could be evoked by PAP, or occur simultaneously [12], and there have also been cases where PAPE is produced without PAP, which confirms that the mechanisms that generate these phenomena are different [13].

From an ecological point of view, it seems more precise to use PAPE than PAP to refer to the performance improvement in volleyball, since the dependent variables used to verify its existence are directly related to performance, such as force, speed, or jump [12]. Furthermore, its possible effects last longer and are more applicable to real volleyball match conditions, where, due to the game's rules [15], it is not possible to generate 8-min strength-related pre-activity before the start of the match, but activations before starting the warm-up that could elicit greater PAPE during the match are plausible. On the other hand, the effects of PAP can also increase sports performance, and in volleyball, this could be achieved by including PAP in resistance workouts that allow obtaining improvements in strength through complex training [16,17].

Since the magnitude of PAPE depends on the levels of fatigue and potentiation [18], the magnitude of the activation will depend on this relation, and therefore, the performance will be increased if the effect of the potentiation is larger than fatigue [19]. This relationship is influenced by other individual factors such as individual physiological characteristics of the subject, experience, age, type of muscular fibers' distribution (i.e., fast-twitch vs. slow-twitch fibers), maximum strength, strength to power ratio, level of training, among others [20].

The design of the activation protocols will greatly affect the result of the enhancement achieved. A resting period between activation and potentiation elicits better performance [21,22]. Similarly, other determining factors of PAPE are the intensity, volume, and protocols of the activation loads and the intensity of jumps or displacements after the potentiation [23].

It has also been suggested that the best increases in VJ are obtained with strength exercises such as the squat, with protocols of 1 to 3 sets of 1 to 5 repetitions and loads greater than 80% of 1RM, obtaining the best results in between 1 to 9 min after activation [24,25]. In the review carried out on vertical jump improvement, Suchomel et al. (2016) arrive at similar conclusions adding the cumulative fatigue of the athlete as individual factors to those already mentioned.

All of these studies have used both trained and untrained subjects as a sample [26]. In this meta-analysis, it was observed that the greatest effects of PAPE occur between 3 and 7 min in trained subjects, obtaining better results than studies for less than 3 min. Also, for studies carried out between 3 and 12 min or more, always for loads greater than 80% of 1RM and in trained subjects, the same authors noted that the longer times included in other meta-analysis are suitable for untrained subjects with smaller loads, where the effect of fatigue is greater.

However, contradictory results were found in the reviewed literature: the improvements in VJ found by Dobbs et al. (2018) were not statistically significant. In addition, some authors did not find effects on jumping performance after a PAP protocol [14,26–30]. Furthermore, the persistence of PAP is significant only for a limited period of time from 28 s to less than 3 min [31], obtaining the performance peak improvement (PAPE) at 6–20 min [25,32].

After reviewing the studies of PAPE protocols applied to volleyball players, it was found that the samples in all the studies are mostly composed of university or college players [22,33–36], with most of them being male players. The physiological difference between sexes [37] may elicit different responses to PAPE. In general, male players have greater type II fibers cross-sectional area and shorter twitch contraction times, whereas female players show more fatigue resistance [38]. Therefore, the outcomes of PAPE may be different depending on gender [38]. Thus, there is a lack of studies on female athletes, particularly in elite female players [39]. Furthermore, none of the studies in the literature has been conducted in real game conditions with volleyball players.

Therefore, the purpose of this study was to observe the effects of PAPE throughout a match in professional female volleyball players. The initial hypothesis was that squat-based pre-activation can trigger PAPE which is displayed as an improvement of VJ height 8 min after the application of the activation and that PAPE lasts for several minutes in a volleyball match of female national Spanish Superliga 2 players.

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

#### *2.1. Subjects*

Twelve Superliga 2 players of University of Alicante volleyball team volunteered to participate in this study (Table 1). Informed consent was obtained from all subjects involved in the study, who read and signed the document before any action in the study was taken. The study was conducted according to the guidelines of the Declaration of Helsinki [40], and approved by the Ethics Committee of University of Alicante (UA-17 November 2018).

**Table 1.** Characteristics of the subjects aggregated by group (mean ± SD).


BMI: body mass index, *n*: number of subjects, Volleyball experience: years the subjects have been playing volleyball; strength experience: time that subjects have been doing specific workouts.

The inclusion criteria were to have 4 years of experience minimum in the practice of volleyball in a national competition and to have previous knowledge in both strength training and half-squat exercise. The exclusion criteria were not to participate in all the tests involved in the study or to suffer injury or illness that prevents the performance of the tests. A control group participant suffered an injury during the experimental process, therefore, she was excluded from the experimental procedure and subsequent analysis.

#### *2.2. Instruments*

For the determination of the force–velocity profile and the vertical jump height, a linear encoder was used (Chronojump-Boscosystem, Barcelona, Spain). To estimate the vertical jump height, a jump mat was used (Chronojump-Boscosystem, Barcelona, Spain), from which to measure the flight time and, thus, estimate the jump height [41]. Both instruments worked at 1000 Hz.

#### *2.3. Procedure*

The experimental design shown in Figure 1 consisted of three phases: individualization, activation, and match, which are described in more detail as follows.

**Figure 1**. Experimental design of the study¸ RM: Repetition maximum; CMJ: Countermovement jump; PAPE: Post-activation performance enhancement FIVB: Fedération Internationalle de Volleyball. **Figure 1.** Experimental design of the study¸ RM: Repetition maximum; CMJ: Countermovement jump; PAPE: Post-activation performance enhancement FIVB: Fedération Internationalle de Volleyball.

#### 2.3.1. Estimation of 1RM in the Half-Squat Exercise

2.3.1. Estimation of 1RM in the Half-Squat Exercise In order to determine the load corresponding to the 1RM percentage in the half-squat exercise for the PAP protocol, the relationship between force and velocity was analyzed, since the speed of execution and the percentage of 1RM are proportional to each other [41]. An incremental loading test was carried out, in which the initial load was established at 30 kg and was gradually increased in 10 kg steps until mean barbell velocity was below 0.50 m/s (i.e., around 80% of 1RM). Afterward, the load was increased from 5 kg and at the end of the test, with speeds close to 0.30 m/s, increments of 1 kg were made to reach 1RM in the most precise way [42,43]. The value of 1RM was considered the load interpolated in the force–velocity profile with the average acceleration velocity value for the halfsquat exercise of 0.30 m/s [44]. The players were refrained from performing physical activity 48 h previous to the test to ensure the absence of fatigue. In order to determine the load corresponding to the 1RM percentage in the half-squat exercise for the PAP protocol, the relationship between force and velocity was analyzed, since the speed of execution and the percentage of 1RM are proportional to each other [41]. An incremental loading test was carried out, in which the initial load was established at 30 kg and was gradually increased in 10 kg steps until mean barbell velocity was below 0.50 m/s (i.e., around 80% of 1RM). Afterward, the load was increased from 5 kg and at the end of the test, with speeds close to 0.30 m/s, increments of 1 kg were made to reach 1RM in the most precise way [42,43]. The value of 1RM was considered the load interpolated in the force–velocity profile with the average acceleration velocity value for the half-squat exercise of 0.30 m/s [44]. The players were refrained from performing physical activity 48 h previous to the test to ensure the absence of fatigue.

#### 2.3.2. Vertical Jump 2.3.2. Vertical Jump

To determine the possible effect of the activation on PAPE in the lower train, countermovement jump (CMJ) heights were measured using a jump mat [45]. CMJ was performed starting from the standing position, with their feet in the center of the jump mat and hands positioned at the hips in akimbo position. After an auditory signal, subjects performed a knee flexion before jumping vertically to maximum height and were instructed to land in the center of the jump mat. A video camera monitoring players in sagittal plane was used to control that knee flexion reached the right joint angle. Three attempts were carried out with a 60-s resting time [44] and the highest value was considered To determine the possible effect of the activation on PAPE in the lower train, countermovement jump (CMJ) heights were measured using a jump mat [45]. CMJ was performed starting from the standing position, with their feet in the center of the jump mat and hands positioned at the hips in akimbo position. After an auditory signal, subjects performed a knee flexion before jumping vertically to maximum height and were instructed to land in the center of the jump mat. A video camera monitoring players in sagittal plane was used to control that knee flexion reached the right joint angle. Three attempts were carried out with a 60-s resting time [44] and the highest value was considered for data analysis [45].

#### for data analysis [45]. 2.3.3. Activation Protocol

2.3.3. Activation Protocol Considering that the sample were players with four years of minimum experience in volleyball training, they can be considered as trained subjects, so the guidelines set by Dobbs et al. (2018) in their meta-analysis were observed, as well as the corresponding 3 repetition activation protocol at 90% intensity of 1RM [35,46] with a resting time of 8 min. Such a protocol follows the margins indicated by these authors [26] and also the rest of Considering that the sample were players with four years of minimum experience in volleyball training, they can be considered as trained subjects, so the guidelines set by Dobbs et al. (2018) in their meta-analysis were observed, as well as the corresponding 3-repetition activation protocol at 90% intensity of 1RM [35,46] with a resting time of 8 min. Such a protocol follows the margins indicated by these authors [26] and also the rest of the studies consulted [23,24,32,34,45].

the studies consulted [23,24,32,34,45]. Prior to activation, a standardized warm-up was performed for both groups, control and experimental, consisting of 4 min of soft running followed by 4 min of dynamic stretching; then 2 min of speed and changes of rhythm and direction inside the playground, and 5 consecutive CMJ jumps to finish [23]. After the warm-up, 2-min rest

period was performed, followed by an initial evaluation of jump height before activation (Pre-PAPE test).

The experimental group performed the activation protocol, consisting of an approaching phase (12 repetitions with 20 kg, 3-min rest, 5 repetitions at 50% of 1RM, 3-min rest), followed by a conditioning phase (3 repetitions at 90% of 1RM). The control group executed the same approaching phase as the experimental group, but when the experimental group performed the conditioning phase control group executed a maintenance workout, consisting of smooth running interspersed with slight changes of direction and 3 vertical jumps. After 8 min, CMJ was measured (Post-PAPE test) to both groups with an identical methodology to that of the Pre-PAP data collection.

#### 2.3.4. PAPE Monitoring during a Volleyball Match

After warm-up was finished, both groups performed a CMJ test before starting the match (Pre-match) and also just at the end of every set of the match (Set 1 to Set 5). This procedure allows describing the evolution in the height reached for volleyball players in a match, as well as to check whether the experimental group shows enhancement derived from PAPE and how much this condition lasts.

#### *2.4. Statistical Analysis*

Descriptive data are presented as mean and standard deviation. Due to the small sample size, the Shapiro–Wilk normality test was used, which resulted in a normal distribution. The differences in jump height between Pre-PAPE, Post-PAPE, Pre-match, Set 1, Set 2, Set 3, Set 4, and Set 5, in regards to experimental and control groups were evaluated using a repeated-measures ANOVA, including the different tests in time points as an intragroup variable and group as a between-subjects factor. Variance homogeneity and homogeneity of the error variances were verified via the Mauchly's test (*p* = 0.304) and the Levene's Test of Equality of Error (*p* range between 0.192 and 0.892 for all comparisons). In addition, a t-test for independent samples was conducted to compare differences on improvement percentage between experimental and control groups. The level of significance was set at *p* < 0.05. The d index was analyzed to determine the magnitude of an effect independent of sample size [47] and to classify the effect size, the criteria of Rhea for elite trained athletes were applied (d < 0.25 trivial; 0.25 ≤ d > 0.50 low; 0.50 ≤ d > 1.0 moderate; d ≥ 1.0 Large) [48]. In this quasi-experimental study, the sample is composed of volleyball elite players competing at a national level. Power analysis conducted with G\*Power (v3.1.9.7, Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany) indicated a minimum sample size of *n* = 11 subjects in order to detect an effect size of Cohen's d = 1.6 with 80% power (α = 0.05, two-tailed) [49].

#### **3. Results**

Table 2 shows the compared results between control and experimental groups for CMJ. There were no significant differences between groups in Pre-PAPE tests in the height reached for both groups (*p*-value > 0.05), indicating that before the intervention the groups were homogeneous. Furthermore, significant intergroup differences and large ES can be observed in CMJ in Post-PAPE, Pre-Match, Set 1, Set 2, and Set 5 always being greater for the experimental group, therefore, the behavior is different for the groups until Set 2 and return to be different in Set 5 but with a reduced ES, Set 3 and Set 4 did not show significant differences. As well, there was a significant difference in the improvement percentage (∆%) between the control and experimental groups from Post-PAPE until Set 2 test, but in the tests Set 3, Set 4, and Set 5 no differences were found (*p*-value < 0.05).


**Table 2.** Vertical jump height performance (mean ± SD).

\* Significant difference between control and experimental groups at the same time point (*p* < 0.05); # Intragroup significant difference between Pre-PAPE and the other post-intervention tests.

As it can be observed in Table 3 the ES for the comparison between the pre-intervention and all post-intervention tests always are larger in the experimental group than in control, except in the Pre-PAPE test where both groups have moderate effect sizes, being slightly higher in the control group. However, in the control group, these values are negative, which occurred in a decrease in vertical jump performance.

**Table 3.** Effect size of intragroup differences in CMJ for Pre-PAPE and Pre-Match vs. the rest of the tests for control and experimental groups.

