*Article* **Effects of the Menstrual Cycle on Jumping, Sprinting and Force-Velocity Profiling in Resistance-Trained Women: A Preliminary Study**

**Felipe García-Pinillos 1,2 , Pascual Bujalance-Moreno 3,\* , Carlos Lago-Fuentes 4,5,† , Santiago A. Ruiz-Alias <sup>1</sup> , Irma Domínguez-Azpíroz 5,6 , Marcos Mecías-Calvo <sup>4</sup> and Rodrigo Ramirez-Campillo 7,8**


**Abstract:** The aim of this study was to examine the effects of the menstrual cycle on vertical jumping, sprint performance and force-velocity profiling in resistance-trained women. A group of resistancetrained eumenorrheic women (*n* = 9) were tested in three phases over the menstrual cycle: bleeding phase, follicular phase, and luteal phase (i.e., days 1–3, 7–10, and 19–21 of the cycle, respectively). Each testing phase consisted of a battery of jumping tests (i.e., squat jump [SJ], countermovement jump [CMJ], drop jump from a 30 cm box [DJ30], and the reactive strength index) and 30 m sprint running test. Two different applications for smartphone (My Jump 2 and My Sprint) were used to record the jumping and sprinting trials, respectively, at high speed (240 fps). The repeated measures ANOVA reported no significant differences (*p* ≥ 0.05, ES < 0.25) in CMJ, DJ30, reactive strength index and sprint times between the different phases of the menstrual cycle. A greater SJ height performance was observed during the follicular phase compared to the bleeding phase (*p* = 0.033, ES = −0.22). No differences (*p* ≥ 0.05, ES < 0.45) were found in the CMJ and sprint force-velocity profile over the different phases of the menstrual cycle. Vertical jump, sprint performance and the force-velocity profiling remain constant in trained women, regardless of the phase of the menstrual cycle.

**Keywords:** female athletes; ovarian cycle; plyometric exercises; testing; velocity

### **1. Introduction**

Lower-limb ballistic movements, or stretch-shortening cycle (SSC) muscle actions, have been identified as key determinants of physical performance in women [1], and jumping and sprinting tests are widely used to assess the mechanical capabilities of the lower-limbs and the efficiency of the SSC [1,2]. The individual force-velocity (*F-v*) relationship has been proposed as a valid marker of the athlete's mechanical profile [3], providing more useful information for training prescription and monitoring training adaptations than isolated jumping or sprinting tests [4]. Moreover, with the constant advances in

**Citation:** García-Pinillos, F.; Bujalance-Moreno, P.; Lago-Fuentes, C.; Ruiz-Alias, S.A.; Domínguez-Azpíroz, I.; Mecías-Calvo, M.; Ramirez-Campillo, R. Effects of the Menstrual Cycle on Jumping, Sprinting and Force-Velocity Profiling in Resistance-Trained Women: A Preliminary Study. *Int. J. Environ. Res. Public Health* **2021**, *18*, 4830. https:// doi.org/10.3390/ijerph18094830

Academic Editors: Richard B. Kreider and Antonio Sousa

Received: 27 March 2021 Accepted: 26 April 2021 Published: 30 April 2021

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

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

technology, more practical and easy-to-access alternatives have emerged in the past years, with different applications for jumping and sprinting assessment [5,6].

Traditionally, the adaptations to training and responses to exercise have been assumed as equal for both men and women [7]. However, a growing body of evidence [8,9] points to cyclical variations in steroids hormones (i.e., estrogen and progesterone) during an ovulatory menstrual cycle (MC). Despite the theoretical implications for physical performance, the available evidence about the influence of hormone variations related to MC on physical performance is controversial [7,9–11]. Some previous works have found cyclical variations in muscular performance parameters over an MC, such as handgrip, standing long jump and consecutive drop jumps [12,13], whereas other studies did not report differences in variables such as bench press or Smith machine squat, counter movement jump (CMJ), repeated sprint, or other no-muscular performance test [14–17]. Since an ovulatory MC implies alterations in the estrogen levels, and given that tendon and ligaments stiffness, and thereby SSC efficiency, has been shown to be impaired in the presence of high levels of estrogen [9], the use of single-joint isokinetic measurements, which minimize the SSC requirements, might overlook the influence of the MC and the hormone alterations.

Lastly, Thompson et al. [18] reviewed the effects of the MC on resistance training, suggesting that strength stimulus during the follicular phase (FP) will increase performance on this capacity. However, this study only registered 10 researches of acute responses, and most of them were focused only on biochemical parameters. Additionally, a recent review of this topic observed only a few studies that have analyzed performance associated to SSC requirements, due to most of them including single-joint isokinetic movements to test muscle strength and performance, or related to aerobic power [19–21]. Taking this into account, the lack of a consensus in the current scientific literature highlights the need to conduct further studies that analyze the relationship among the different phases of the MC and actions with SSC requirements.

Taken altogether, the purpose of this study is to examine the effect of the MC on vertical jump, sprint performance and *F-v* profiling in resistance-trained women. We hypothesized changes in vertical jump, sprint performance and some alterations in the *F-v* profile according to the phases of the MC with a better performance during the follicular phase.

#### **2. Methods**

In order to test our hypothesis, resistance-trained women completed a battery of loaded and unloaded vertical jump and linear sprint tests on three different days, according to their MC phases, including the early follicular phase, the late-follicular phase, and the mid-luteal phase.

#### *2.1. Participants*

A group of nine healthy eumenorrheic and trained women (age: 28.7 ± 3.6 years; height: 1.63 ± 0.05 m; body mass: 61.1 ± 5.6 kg) voluntarily participated in this study. The inclusion criteria were: (I) not to take any hormonal contraceptive; (II) to have a regular MC (i.e., 26 to 32 days of duration) for the last six months confirmed by the athletes through bleeding phase verification using the phone application "Clue"; (III) to train regularly (i.e., at least three times per week (over 200 min per week) for the last six months); (IV) to include resistance and endurance training in their training plan for the last six months. The sample size was selected by convenience and a post hoc analysis of the achieved power for this sample size was conducted (G\*Power software vs. 3.1), given α = 0.05, (1 − β) = 0.8, effect size = 0.5, statistical test = mean difference between matched pairs. This analysis revealed a low to moderate power (0.4). After receiving detailed information on the objectives and procedures of the study, each participant signed an informed consent form in order to participate, which complied with the ethical standards of the World Medical Association's Declaration of Helsinki, Finland (2013). The study was approved by the Institutional Review Board.

#### *2.2. Procedures 2.2. Procedures*

Participants performed a total of four testing sessions, held between 17:00 and 20:00 h to avoid the influence of the circadian rhythms during the months of June and July. The day before a testing protocol, participants were instructed to perform a low-intensity workout. During the training period, participants were encouraged to maintain their dietary routine. This procedure is based on a similar previous study [14]. Participants performed a total of four testing sessions, held between 17:00 and 20:00 h to avoid the influence of the circadian rhythms during the months of June and July. The day before a testing protocol, participants were instructed to perform a low-intensity workout. During the training period, participants were encouraged to maintain their dietary routine. This procedure is based on a similar previous study [14].

A preliminary session (i.e., session 1) was used to ensure that all participants were able to perform the vertical jumping and linear sprinting tests with a proper technique, despite all participants being experienced in loaded and unloaded plyometric jump and sprint exercises. The following testing sessions (i.e., sessions 2–4) were conducted in three different phases across the MC: (I) phase 1—bleeding or early follicular phase (i.e., testing between days 1–3); (II) phase 2—follicular or late-follicular phase (i.e., testing between days 7–10) and; (III) phase 3—luteal or mid-luteal phase (i.e., testing between days 19–21) (Figure 1). The selection of these phases were based on previous studies [11,14,19,22] and it has been suggested that those phases represent the main events during an MC, including menses, pre-ovulation and post-ovulation, respectively [11,15]. Phases of the MC were defined based on the first day of menses. A preliminary session (i.e., session 1) was used to ensure that all participants were able to perform the vertical jumping and linear sprinting tests with a proper technique, despite all participants being experienced in loaded and unloaded plyometric jump and sprint exercises. The following testing sessions (i.e., sessions 2–4) were conducted in three different phases across the MC: (I) phase 1—bleeding or early follicular phase (i.e., testing between days 1–3); (II) phase 2—follicular or late-follicular phase (i.e., testing between days 7–10) and; (III) phase 3—luteal or mid-luteal phase (i.e., testing between days 19–21) (Figure 1). The selection of these phases were based on previous studies [11,14,19,22] and it has been suggested that those phases represent the main events during an MC, including menses, pre-ovulation and post-ovulation, respectively [11,15]. Phases of the MC were defined based on the first day of menses.

*Int. J. Environ. Res. Public Health* **2021**, *18*, x FOR PEER REVIEW 3 of 10

proved by the Institutional Review Board.

Identical testing protocols were performed in sessions 2–4. First, the anthropometric characteristics of the participants were measured, including standing height, body mass and measurements needed to determine the push-off distance (leg length and initial height) [3,23]. Then, after a standardized 10-min warm-up protocol based on dynamic stretching and preparatory exercises, including jumping and sprinting exercises, the participants performed a battery of vertical jumping tests, and an incremental vertical loadedjump protocol, and two 30 m linear sprints (see below for further details). Every testing session was conducted on one specific day, including anthropometric measurements, jumping and sprinting test—in that order. The testing protocols were performed indoors, and weather conditions were registered for the subsequent analysis. Participants were encouraged to achieve maximum performance during each test, and the personal best at-Identical testing protocols were performed in sessions 2–4. First, the anthropometric characteristics of the participants were measured, including standing height, body mass and measurements needed to determine the push-off distance (leg length and initial height) [3,23]. Then, after a standardized 10-min warm-up protocol based on dynamic stretching and preparatory exercises, including jumping and sprinting exercises, the participants performed a battery of vertical jumping tests, and an incremental vertical loadedjump protocol, and two 30 m linear sprints (see below for further details). Every testing session was conducted on one specific day, including anthropometric measurements, jumping and sprinting test—in that order. The testing protocols were performed indoors, and weather conditions were registered for the subsequent analysis. Participants were encouraged to achieve maximum performance during each test, and the personal best attempt for each test was selected for the subsequent analysis.

tempt for each test was selected for the subsequent analysis. Anthropometric measurements*.* A stadiometer (Seca 202, Seca Ltd., Hamburg, Germany), a weighing scale (Seca 803, Seca Ltd., Hamburg, Germany) and a non-stretchable tape (Seca 201, Seca Ltd., Hamburg, Germany) were used to measure height, body mass Anthropometric measurements. A stadiometer (Seca 202, Seca Ltd., Hamburg, Germany), a weighing scale (Seca 803, Seca Ltd., Hamburg, Germany) and a non-stretchable tape (Seca 201, Seca Ltd., Hamburg, Germany) were used to measure height, body mass and push-off distance.

and push-off distance. Jumping tests. The participants performed a battery of vertical jumping tests including two maximal attempts for the squat jump (SJ), countermovement jump (CMJ) and drop jump from a 30 cm box (DJ30). Jumping tests were performed in that order. The resting period lasted 15 s between repetitions and 4 min between tests. As described by a previous study [24], during SJ, participants were instructed to adopt a flexed knee position

(approximately 90 degrees) for 3 s before jumping, while during the CMJ no restriction was imposed over the knee angle achieved before jumping. Jumping tests were executed with arms akimbo. Takeoff and landing were standardized to full knee and ankle extension on the same spot. During the DJ30, participants were instructed to maximize jump height and to minimize ground contact time after dropping down [25]. Jump heights (m) were registered for each test. Additionally, the reactive strength index (RSI) was obtained from DJ30 (i.e., RSI = flight time [ms]/contact time [ms]).

All these assessments were performed through the My Jump 2 app (v.5.0.5). This is based on high-speed video analysis (i.e., 240 fps) and it has been shown as valid and reliable to determine jump height during CMJ, SJ and DJ tests [5], as well as to determine related parameters such as temporal variables and RSI [26]. The instructions provided by the developer were followed for collecting data [5]. A researcher, laying prone on the ground, held the smartphone (iPhone; Apple, Inc., Cupertino, CA, USA) and recorded each jump from a frontal plane, at approximately 1.5 m.

CMJ *F-v* profiling. Thereafter, the participants were assessed for the *F-v* profiling, performing two maximal CMJs under an unloaded and loaded condition, including at least three loads from 10 to 45 kg [23]. Loads were increased until the jump height was shorter than 10 cm [27]. The resting period lasted 15 s between repetitions and 4 min between sets. The same equipment and protocol previously described (i.e., My Jump 2 app, v.5.0.5) was used for these measurements.

This method has been shown as valid and reliable for computing mechanical parameters, and thereby *F-v* profiles, from the CMJ [23]. As previously recommended [4], the variables extracted from the *F-v* protocol included the theoretical maximal force at null velocity (F0), the theoretical maximal velocity of lower-limbs extension under zero load (v0), maximal power output against different loads (Pmax), and the slope of the linear *F-v* relationship (Sf*<sup>v</sup>* ).

30 m sprint test. The participants performed two maximal-effort 30 m linear sprints, with 5 min rest in between, on a synthetic indoor track. An application for smartphone (i.e., My Sprint app) was used to record and analyze (i.e., split times: 5, 10, 15, 20, 25 and 30 m) the trials. The system is based on high-speed video analysis (i.e., 240 fps) and it has been shown as valid and reliable to evaluate linear sprint performance in relation to two different reference systems such as timing photocells and radar gun [6]. The testing protocol was based on the procedures described by the developer in a previous paper [6]. The recording was conducted through an iPhone 7, which was mounted to a tripod and located 10 m perpendicular to the sprint direction, just in front of the 15 m marker.

Sprint *F-v* profiling. Video analysis provided split times during the linear 30 m sprint test. Following a simple method proposed by Samozino and colleagues [3], those data along with anthropometric characteristics and weather conditions let the researchers obtain power, force and velocity properties as well as mechanical effectiveness during linear sprint running. This method has been examined and it has been shown as valid to determine mechanical parameters during linear sprint [3]. As identified by a previous work [4], the variables of interest from this profile are the theoretical maximal horizontal force (HZT-F0), maximal running velocity (HZT-v0), associated maximal power output (Pmax), the slope of the *F-v* relationship that determined the mechanical profile (F*v*slope), the maximum value of ratio of force (RFmax) and the rate of decrease in the ratio of force with increasing speed during sprint acceleration as a measure of the index of the effectiveness of ground force orientation (DRF).

#### *2.3. Statistical Analysis*

All data are presented as mean and standard deviation (mean ± SD). The normality assumption was confirmed by the Shapiro–Wilk test (*p* > 0.05). One-way repeated measures analysis of variance (ANOVA) with Bonferroni post-hoc tests were conducted to compare the outcome variables at three different phases during the MC (i.e., bleeding, follicular and luteal phases). The *F-v* relationships were established by means of least squares linear

regression models [4,28]. The Hedges g effect size (ES) was also calculated to determine the magnitude of differences, interpreted as follows: trivial (<0.2), small (0.2–0.59), moderate (0.60–1.19), large (1.20–2.0), and extremely large (>2.0) [29]. Statistical significance was set at α < 0.05. The SPSS software (version 25.0, SPSS Inc., Chicago, IL, USA) was used.

#### **3. Results**

Table 1 shows the vertical jumping and linear sprinting performance of participants at the three different phases of the MC. The repeated measures ANOVA reported no significant differences between phases in CMJ, DJ30, RSI nor sprint performance (*p* ≥ 0.05, ES < 0.25). However, differences (*p* = 0.033, ES = −0.22) were found in SJ, with the post-hoc test revealing a greater performance during the follicular phase compared to the bleeding phase.

**Table 1.** Comparison of the vertical jumping and 30 m sprint performance parameters obtained from three different phases of the menstrual cycle.


Values as mean (± standard deviation); \* denotes significant differences between phases (*p* < 0.05); ˆ indicates where the between-phase difference is. ES: g Hedges effect size; phase 1: bleeding phase; phase 2: follicular phase; phase 3: luteal phase; CMJ: countermovement jump; SJ: squat jump; DJ30: drop jump from a 30 cm box; RSI: reactive strength index (flight time [ms]/contact time [ms]).

> The CMJ *F-v* relationship parameters at the three different phases of the MC are indicated in the Table 2. No significant differences (*p* ≥ 0.327, ES < 0.45) were found between phases in any parameter (i.e., F0, *v*0, Pmax and Sf*<sup>v</sup>* ).

**Table 2.** Comparison of the *F-v* relationship parameters obtained from three different phases of the menstrual cycle during an incremental loading protocol for the countermovement jump (CMJ) test.


Values as mean (± standard deviation); ES (g): g Hedges effect size; phase 1: bleeding phase; phase 2: follicular phase; phase 3: luteal phase; *F-v*: force-velocity; F0: the theoretical maximal force at null velocity; *v*0: the theoretical maximal velocity of lower-limbs extension under zero load; Pmax: maximal power output against different loads; Sf*<sup>v</sup>* : slope of the linear *F-v* relationship.

> Table 3 shows the mechanical parameters associated to the *F-v* relationship during the 30 m linear sprint test in different phases of the MC. No between-phase significant differences (*p* ≥ 0.340, ES < −0.36) were found in any parameter.



Values as mean (± standard deviation); ES: g Hedges effect size; phase 1: bleeding phase; phase 2: follicular phase; phase 3: luteal phase; HZT-F0: theoretical maximal horizontal force; HZT-v0: maximal running velocity; Pmax: associated maximal power output; F*v*slope: slope of the *F-v* relationship that determined the mechanical profile; RFmax: maximal ratio of force; DRF: rate of decrease in the ratio of force with increasing speed during sprint acceleration.

#### **4. Discussion**

This study aimed to examine the effects of three different phases of the MC on vertical jumping, linear sprinting performance and *F-v* profiling in resistance-trained women. The main finding rejects our initial hypothesis, as athletic performance in these explosive tasks (i.e., jumping and sprinting) and the *F-v* profiling requiring SSC muscle actions suffer no significant variation in trained women over the course of their ovarian MC.

Although there are theoretical implications for athletic performance, there is no conclusive evidence about cyclical variations during the MC in sportswomen [7,9,10,14]. Focused on the influence of the MC on muscular performance and muscle strength, the lack of consensus is remarkable, with previous studies reporting opposing findings. As mentioned earlier, some previous works have found variations in muscular performance parameters over an MC [12,13], whereas other studies did not find differences [14,20,30,31]. The authors suggest that between-study differences might be attributed to some methodological issues (e.g., exercise testing, timing of measurements or definition of the MC phases). Related to this, the last review about this topic matches with this finding, due to the differences among phases analyzed and level of participants, including non-homogeneous participant groups, among others. However, our study presents a small but homogeneous sample, similar to the mean of previous studies, among 10 to 15 trained women [32]. Nonetheless, more studies should analyze muscular performance with the same methodology to ensure the existence, or not, of differences among the different phases of MC, including the bleeding, late-follicular and mid-luteal phases [32].

In this context, it is noteworthy that some previous studies have used maximal voluntary contraction or maximal voluntary isometric force through isokinetic measurements [19,20] to examine the influence of the MC on muscular performance. Variations in steroid hormones affect tendons and ligaments, with a high level of estrogen decreasing musculotendinous stiffness [9]. Therefore, it is possible that the effect of the MC may be modulated by the type of muscle action being performed, with those requiring high levels of musculotendinous stiffness (i.e., SSC muscle actions) more prone to be affected by the MC. However, the current study provides some insights into the influence of the MC phases on explosive tasks with high SSC requirements (i.e., jumping and sprinting), with both athletic performance and mechanical parameters showing no differences over different phases of the MC. Regarding this, the rise of estrogen during the follicular phase has been signaled as the main influencing factor to affect the muscular performance [32]; meanwhile, during the luteal phase the CK concentrations increases and decreases the strength levels [18]. However, its influence on short–high intensity efforts such as jumps, among others, is not clear. The body composition has also been described as a potential influencing factor on athletic performance. Traditionally, it has been suggested that the fat oxidation increases during the follicular phase because of the anabolic effect of estrogen, meanwhile fluid retentions can influence the lowest performance during the luteal

phase. However, a recent study with trained women compared different body composition variables throughout the three main MC phases and found no significant differences in these parameters [33]. So, this factor should be cautiously considered as an influencing parameter to explain plausible differences on athletic performance over the ovarian cycle.

Regarding jumping performance, few studies have analyzed the effect of the MC phases on this physical fitness outcome, and conflicting results have been reported. Whereas Davies et al. (1991) reported an improvement in standing long jump test in the bleeding phase compared to the follicular phase, other previous works did not find differences in performance over the course of an ovarian MC in the CMJ-comparing only the early follicular phase vs. the mid-luteal phase in soccer players [16], SJ [34] nor in the DJ-comparing only the follicular phase vs. the ovulation phase in active women [13]. Nevertheless, the participants of these studies did not report previous experience in resistance training, which might be relevant for this type of efforts. Of note, any of the aforementioned studies provided mechanical parameters related to vertical jumping, which might better describe the differences among different MC phases. Recently, another study performed with a similar sample (i.e., trained athletes with six months of experience in resistance training) registered three different variables to determine each MC phase to test force, velocity and power output in the concentric phase in a Smith machine half squat exercise [15]. This work neither found differences in performance comparing these three phases (bleeding, follicular and luteal phases), which concur with our main findings for similar outcomes. Lastly, our results are also in line with a recent study with high-level team sport players, which did not show differences among MC phases in CMJ performance in eumenorrheic athletes, analyzed with serum hormonal levels by blood sample [35]. Therefore, the current study confirms the lack of MC effect on vertical jumping performance and, as a novelty, provides information about the dynamic of the *F-v* relationship parameters over the different phases of MC.

Concerning the linear sprint performance, previous studies have considered the effect of the MC in cycling sprints [17,34,36] and linear running sprints [16,30,31], with all those works reporting no effect of the MC in maximal anaerobic performance. With the focus on studies testing linear running sprint, some works [30,31] have used a 30-s nonmotorized treadmill sprinting test in different phases of the MC, reporting no differences in performance (i.e., in terms of mean and peak power output and sprint total time) in trained women. Likewise, in an experiment performed outdoors (i.e., on field testing) [16], the authors found no differences in 30 m linear sprint time during the different phases of MC in female soccer players. Another recent study found differences in 20 m linear sprint with high-level team sport players [35]. However, as previously indicated, the authors of these studies only compared the follicular phase with the luteal phase, dismissing the bleeding phase, one of the most important physiological moments of the MC due to the lower concentrations of estrogen and progesterone [8,9]. In fact, a recent systematic review reinforces the need to include this phase in further studies on the relationship between athletic performance and MC [32]. This study observed that performance could be reduced during the bleeding phase compared with the rest of the MC phases. Due to the small number of studies which performed CMJ or sprinting tests including this stage, our findings provide a novel result which suggests that performance does not increase during the first days of the MC. That is, the current study provides support about the lack of effect of MC phases on linear running sprint performance comparing the three main phases, and it builds up the available information on the MC influence on sprint mechanical parameters and *F-v* profile.

Finally, some limitations must be taken into consideration to properly interpret these findings: first, the verification of the MC phases with no hormone concentration measurements [37]; second, the limited number of participants (*n* = 9) with low to moderate statistical power for this sample size; third, the performance level of the subjects; fourth, the methods for assessing jump performance parameters based on high-speed video analysis and with no information regarding jump strategy (e.g., time to take-off or CMJ displacement) [38]. Notwithstanding these limitations, the current study provides some insights into the effects of MC on jumping, sprinting and *F-v* profiling in women by using low-cost and easy-to-access tools and measures.

#### *Practical Applications*

The current study confirms the lack of MC effect on vertical jumping performance, but a small difference was found in SJ, with a greater performance during the follicular phase compared to the bleeding phase. Likewise, the linear running sprint performance was not influenced by MC phases, supporting the aforementioned research projects. From a practical standpoint, given the lack of differences in muscular performance (in terms of *F-v* profile) during different MC phases, the hormone variations over the course of an ovarian cycle do not seem to play a key role for athletic performance in high-intensity muscle activities such as jumping or sprinting for non-competitive eumenorrheic trained women.

Considering the lack of consensus, the authors claim the convenience of further studies to highlight, on the one hand, if the training response (i.e., internal load to an external load) might change over the course of an ovarian MC and, on the other hand, if training programmes based on MC phases are more efficient and safer for eumenorrheic women than traditional plans based on training outcomes.

#### **5. Conclusions**

Vertical jumping, linear sprinting performance and the *F-v* profiling requiring stretchshortening cycle muscle actions suffer no significant variation in eumenorrheic sportswomen over the course of an ovarian MC.

**Author Contributions:** F.G.-P. participated in the design of the study, data reduction/analysis and interpretation of results; P.B.-M. participated in the design of the study and contributed to data collection; C.L.-F. contributed to data analysis and interpretation of results; R.R.-C., I.D.-A., S.A.R.- A. and M.M.-C. participated in the design of the study and interpretation of results. All authors contributed to the manuscript writing. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by the Pre-competitive Projects for Early Stage Researchers Programme from the University of Granada (ref: PPJIA2020.03)

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Ethics Committee of University of Jaen, Jaen, Spain (protocol code ABR.19/13.PRY and date of approval 29 April 2019).

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

**Data Availability Statement:** Data will be available by request.

**Conflicts of Interest:** The authors declare that they have no competing interests.

#### **References**


### *Article* **Exploring the Determinants of Repeated-Sprint Ability in Adult Women Soccer Players**

**Lillian Gonçalves 1,\*, Filipe Manuel Clemente 2,3 , Joel Ignacio Barrera <sup>4</sup> , Hugo Sarmento <sup>4</sup> , Francisco Tomás González-Fernández <sup>5</sup> , Markel Rico-González 6,7 and José María Cancela Carral <sup>1</sup>**


**Abstract:** This study aimed to explore the main determinants of repeated-sprint ability (RSA) in women soccer players considering aerobic capacity, sprinting performance, change-of-direction, vertical height jump, and hip adductor/abductor isometric strength. Twenty-two women soccer players from the same team participating in the first Portuguese league were observed. Fitness assessments were performed three times during a 22-week cohort period. The following assessments were made: (i) hip abductor and adductor strength, (ii) squat and countermovement jump (height), (iii) change-of-direction test, (iv) linear sprinting at 10- and 30-m, (v) RSA test, and (vi) Yo-Yo intermittent recovery test level 1. Positive moderate correlations were found between peak minimum RSA and adductor and abductor strength (r = 0.51, *p* < 0.02 and r = 0.54, *p* < 0.01, respectively). Positive moderate correlations were also found between peak maximum RSA and adductor and abductor strength (r = 0.55, *p* < 0.02 and r = 0.46, *p* < 0.01, respectively). Lastly, a moderate negative correlation was found between fatigue index in RSA and YYIR1 test performance (r = −0.62, *p* < 0.004). In conclusion, abductor and adductor isometric strength-based coadjutant training programs, together with a high degree of aerobic endurance, may be suitable for inducing RSA in female soccer players.

**Keywords:** football; athletic performance; anaerobic; aerobic; sports training

### **1. Introduction**

Soccer is a team sport practiced by many athletes throughout the world, with an estimated 4–26 million female participants [1–4] and approximately 238 million male participants [5]. The number of female soccer players has increased in the last years in approximately 50% considering the last report of FIFA [3,6]. Due to the challenges associated with this rapid increase in the number of participants, it is important to better understand the characteristics of these players, their physiological/physical demands, and their training processes [1,2,7].

As an intermittent exercise, a women's soccer match involves activities with different intensities, such as walking, jogging, moderate running, high-intensity running, and sprinting [8–10]. It is well-known that low-intensity movements are predominant during women's matches [9,11,12], although high-intensity activities are also considered important

**Citation:** Gonçalves, L.; Clemente, F.M.; Barrera, J.I.; Sarmento, H.; González-Fernández, F.T.; Rico-González, M.; Carral, J.M.C. Exploring the Determinants of Repeated-Sprint Ability in Adult Women Soccer Players. *Int. J. Environ. Res. Public Health* **2021**, *18*, 4595. https://doi.org/10.3390/ ijerph18094595

Academic Editor: Veronique Billat

Received: 25 March 2021 Accepted: 24 April 2021 Published: 26 April 2021

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

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

components of physical performance, and they are often crucial to the outcomes of matches because they are associated with offensive attacks [12,13]. Usually, women soccer players cover between 8.5 and 11 km in a match, of which 1.5–1.8 km are spent performing highspeed running and from 14.9 to 460 m are spent sprinting [4,9,10,14,15].

To support the demands of the match, a proper fitness status should be sustained. As an example, in previous research on women soccer players, a strong correlation was observed between Yo-Yo intermittent recovery test performance and the amount of highintensity running performed in games [9,16]. Additionally, a strong correlation between sprinting skills and high-intensity performance was found in a previous study [2]. In fact, many decisive phases during a soccer match require players to exercise at a high intensity [17]. Therefore, the ability of a soccer player to recover and to reproduce their performance in subsequent sprints is a vital fitness condition [10]. In the particular case of elite level, the intermittent high-intensity endurance and the ability to repeatedly sprint in short time intervals (RSA) are considered relevant fitness conditions for competitive soccer players [18–22].

As a multifactorial factor, the RSA can be influenced by anaerobic and aerobic metabolism [18,23–25]. From a physiological perspective, RSA is a complex quality that is correlated with motor unit activation and is essential to achieving maximal sprint speed and oxidate capacity for phosphocreatine (PCr) recovery and hydrogen (H+) buffering to provide the ability to repeated sprints [26]. Following the same line of thinking, other authors have shown that better sprinters use more of their accessible PCr stores than weaker sprinters [27]. This idea could be related to the strong relationship between PCr resynthesis and power output recovery following 30-s sprints [27,28].

The RSA test simulates intermittent exercise and identifies a player's capacity to maintain maximal effort and recovery during multiple successive high-speed running or sprinting efforts [20,29–31]. Therefore, RSA is an essential factor for determining success in soccer, alongside other qualities like technical and tactical skills, strength, explosive power, speed, and endurance [26,32]. When RSA is compared with aerobic capacities, it was concluded that players with a higher aerobic capacity and faster oxygen kinetics recover faster after high-intensity exercise [29]. These athletes also exhibited better overall RSA performance and recovery performance during the RSA test [29]. Another study showed that subjects with a higher maximal oxygen consumption (VO2max) value present smaller sprint decrements, suggesting that VO2max contributes to maintaining performance during repeated-sprint efforts [27].

Beside the metabolic perspective that supports RSA, physical capacities also play a determinant role in RSA. As example, a well-developed neuromuscular system allows a better activation of motor unit [26], while lower-limb strength and power support the acceleration and the maximal speed in the first repetitions and aerobic capacity sustain the performance over the last sprints [33]. The efficiency of RSA could also depend on the player's agility, as this factor is known to be correlated with linear sprint ability [34–36].

The ability to perform repeated sprints while requiring minimal recovery periods between efforts (RSA) appears to be an important aspect of field-based team sport [37]. However, it is difficult to understand which determinants are related to RSA. Thus, some doubts and non-consensual evidence remain in this regard in women's soccer. For that reason, it is important to identify which physical capacities could explain RSA in women's soccer. Such identification may help coaches define better strategies for improving RSA. Therefore, the purpose of this study was to analyze the determinants of RSA based on aerobic performance, linear sprinting and change-of-direction, vertical height jump, and abductor and adductor isometric strength. We hypothesize that strength and power will be determinants for maximum and minimum peak power RSA, while aerobic performance will be determinant for sustaining the performance (fatigue index) [33].

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

#### *2.1. Experimental Approach to the Study*

This study followed an observational analytic cohort design. The period of observation was 22 consectutive weeks. Fitness assessment were performed three times during the cohort (Figure 1). Between the first and second assessment occurred 4 weeks (pre-season) and between the second and third assessment 18 weeks (end of the first half of the season). The aim was to explore determinants of RSA based on the measures of aerobic capacity, sprinting performance, change-of-direction, vertical height jump and hip adductor/abductor isometric strength. From the initial twenty-five participants, twenty-two remained. Three were excluded based on the fact that did not participated in all the assessments.

**Figure 1.** Timeline of the study.

#### *2.2. Participants*

Twenty-two women soccer players from the same team participating in the first Portuguese league were observed during the study. In the beginning of the season, the participants presented a mean age of 22.7 ± 5.21 years old, 162.51 ± 7.08 cm of height, 59.1 ± 9.50 kg of body mass. In the second assessment the mean of body mass was 59.01 ± 9.31 kg and in the third evaluation the mean body mass was 61.10 ± 9.94 kg. The eligibility criteria for including in the final sample was: (i) participants were assessed in the three moments of the cohort; (ii) participants participated in, at least, 85% of the total number of training sessions during the cohort; (iii) players had injuries or illness no longer than 4 consecutive weeks; and (iv) players should had a minimum of two years of experience to volunteered for this study. Among the included participants, three were goalkeepers, four were external defenders, four were central defenders, six were midfielders, and five were attackers. The team had three training sessions per week plus an official match in the weekends. Before the cohort begin, all the players were informed about the study design and procedures. After that, each player signed an informed consent. The study was approved by the local university (code: CTC-ESDL-CE001-2021) and followed the ethical standards of Declaration of Helsinki for the study in humans.

#### *2.3. Data Collection*

In the three moments of assessment, the tests were made always at the same hour (7:30 p.m.) and days of the week, with a rest period of 48 h (considering the last match/training). Additionally, the assessments two and three were preceded by the same type of microcycle. In each moment of the assessment, the tests were split over three days (interspaced by 24 h). In the first training session of the week players were tested for their anthropometry and hip adductor and abductor strength. In the second training session it was assessed the vertical jump, changes of direction and linear speed. In the third session it were applied the repeated sprint ability test and the Yo-Yo intermittent recovery test. Before the first assessment of each day, a standard warm-up protocol was implemented, by group of players, since they were organized in groups of three to have the same duration between the end of warm-up and beginning of the test. All the players followed the same order. Between tests, there was a minimum of 3 min of rest. The anthropometry, abductor and adductor strength and squat and countermovement jump were performed in a private room, with a stable temperature of 23 ◦C and relative humidity of 55%. The sprinting tests, RSA test and the Yo-Yo intermittent

recovery test were executed in a synthetic turf with a mean temperature of 19.5 ± 3.4 ◦C and relative humidity of 63 ± 4%. No raining conditions occurred in the assessments.

#### 2.3.1. Anthropometry

There was collected the height and body mass in the three moments of evaluation, at the same hour and at the same day of the week. The evaluation of the height was executed by using the stadiometer (SECA 213, Birmingham, UK), players were asked to remove shoes and other accessories that influence the assessment, they also should be in a vertical and immobile position, with the arms extended along the body and keep a fixed stare, straight ahead and in an upright position. The evaluation of the body mass was executed with a digital balance (SECA 869, Birmingham, UK), it was asked to the players to be barefoot and only in light clothing. For each measure, only one measured was collected.

#### 2.3.2. Hip Adductor and Abductor Isometric Strength

Hip adductor and abductor isometric strength measurement was tested with the dynamometer (Smart Groin Trainer, Neuro excellence, Braga, Portugal). The dynamometer was positioned in the thigh area. Players were asked to lie down in the supine position, with 45◦ of hip flexion and around 90◦ of knee flexion [38]. Players were instructed to execute the maximum squeeze in accordance with a previous study [38], although with changes to 20 s for the present protocol. Three trials were made for abductor and adductor, with 10 s of rest between trials. Abductor were tested first (all the trials) and then adductor (all the trials). The highest strength in kilograms were extracted as the main outcome. The best score among trials was obtained for the data treatment.

#### 2.3.3. Squat and Countermovement Jump

The squat and countermovement jumps were performed. The squat jump (SJ) consisted in standing with the knees at 90 degrees, as the position of squat, with no movement, hand in the waist, with no help of the upper limbs the player should jump and extend the legs, falling in the same place. The players waited 3 s in squat position before each jump. The countermovement jump (CMJ) started in standing position with the hands in the waist, being realized with the flexion of the legs and immediately the extension with the jump, the legs will be in extension and they will fall in the same place. For each movement, three trials were executed, with a rest period of 30 s between. The SJ and CMJ were tested with an optical measurement system consisting of a transmitting and receiving bar (Optojump, Microgate, Bolzano, Italy). The Optojump allows a repeatable measurement of flight time as confirmed in a reliability experiment with an intraclass correlation test of 0.95 [39]. The outcome extracted in each trial was the jump height (cm). For each movement, it was considered the highest jump for data treatment.

#### 2.3.4. Change-of-Direction Test

Agility was assess by using the test zig-zag 20 m [40], this test consisted in four sections of 5 m each set out at 100◦ . The time was recorded using photocells timing gates (Photocells, Brower Timing System, UT, USA), with resolution of 1 thousandth of seconds. Typical error of the Photocells was between 0.04 and 0.06 (s), while the smallest worthwhile change was between 0.11 and 0.17 (s) [41]. This test was performed in the fields, before the training session. Subjects performed three trials of the test, with 3 min of rest between all trials and tests. The outcome extracted was the best time (lowest time in seconds) considering the trials.

#### 2.3.5. Linear Sprinting

Linear Sprint was assessed over 10-m and 30-m using photocell timing gates (Photocells, Brower Timing System, USA), with resolution of 1 thousandth of seconds. The participants started 0.5 m behind the initial timing gate in a two point split stance and were instructed to set off in their own time and run at the maximal speed until the last gate. Each

participant performed three trials at maximal effort. The outcome extracted was the time (seconds) for completing the run. The best score in each running distance was considered for the data treatment.

#### 2.3.6. Running Anaerobic Sprint test

The protocol used for testing the RSA consisted in 35 linear meters (no change-ofdirection), performed six times and with a recovery time between efforts of 10 s [42]. The participants started their sprint 0.5 m behind the starting timing gate. Photocell timing gate (Photocells, Brower Timing System, UT, USA), with resolution of 1 thousandth of seconds were positioned in the beginning and at the end lines to record the time of each sprint effort. The time (seconds) for each trial was collected. After that, the minimum and maximum peak power was determined by using the equation [43] Power = Body mass×Distance<sup>2</sup> Time<sup>3</sup> , as well as the fatigue index used the following equation [43] Fatigue index = maxpower−minpower Sum of 6 sprints (s) .

#### 2.3.7. Yo-Yo Intermittent Recovery Test—Level 1

The Yo-Yo IR1 test consisted of repeated 20-m runs back and forth between two markers with a progressive increase in speed, which was regulated by an audio player. Between each 40-m run, the athlete recovered with 10 s of jogging (shuttle runs of 2 × 5 m). Yo-yo level 1 starts at 10 km/h and level 2 at 13 km/h, with both levels progressively increasing in speed throughout the test. The test was completed when the athlete reached voluntary exhaustion or failed to maintain her running pace in synchrony with the audio recording. The number of completed levels and shuttles and the total distance covered were recorded at the end of the test. The total distance (meters) was extracted as the outcome. The maximal oxygen Uptake (VO2max in mL/min/kg) was estimated by the next equation [44]: VO2max = final distance (m) × 0.0084 + 36.4.

#### *2.4. Statistical Analysis*

For the treatment of the data, we use adequate statistical methods to calculate percentages and central and dispersion parameters (arithmetic mean and standard deviation). Descriptive statistics were calculated for each variable (See Table 1, for more information). In ADD and ABD two subjects missed the data collection, and they were excluded from the item analysis. Similarly occurred with one participant in YYIRT. Before any parametric statistical analysis was performed, the assumption of normality was tested with the Kolmogorov–Smirnov test on each variable. The changes over the season were determined by a one-way ANOVA with repeated measures. Significant main effects were subsequently analyzed using a Bonferroni post hoc test. Effect size is indicated with partial eta squared for Fs. To interpret the magnitude of the eta squared we adopted the following criteria: η <sup>2</sup> = 0.02, small; η <sup>2</sup> = 0.06, medium; and η <sup>2</sup> = 0.14 large. Pearson correlation coefficient r was used to examine the relationship between RSA (Pmax, Pmin, and Fatigue index) and the remaining variables (ADDs, ABDs, SJ, CMJ, 10 and 30 m sprint, COD and YYIRT1). To interpret the magnitude of these correlations we adopted the following criteria: r ≤ 0.1, trivial; 0.1 < r ≤ 0.3, small; 0.3 < r ≤ 0.5, moderate; 0.5 < r ≤ 0.7, large; 0.7 < r ≤ 0.9, very large; and r > 0.9, almost perfect [45]. Confidence intervals (95% CI) were calculated for each correlation. Multiple regression analysis was used to model the prediction of RSA from remaining variables. In this regression analysis, were examined separately all variables. Data were analyzed using software Statistica (version 10.0; Statsoft, Inc., Tulsa, OK, USA).


**Table 1.** Anthropometrical and fitness variables in the three moments of assessment (Mean ± SD).

ADD: adductor isometric strength; ABD: abductor isometric strength; SJ: squat jump; CMJ: countermovement jump; 10 m: 10-m sprint; 30 m: 30-m sprint; COD: change-of-direction; YYIRT1: Yo-Yo intermittent recovery test level 1; Pmin: peak power (minimum); Pmax: peak power (maximum); FI: fatigue index; cm: centimeters; s: seconds. \* denotes significance at *p* < 0.01.

#### **3. Results**

Descriptive statistics were calculated for each variable (See Table 1, for more information). Different repeated measures ANOVAs with participants' mean ADDs, ABDs, 10 m, 30 m, COD and FI, did not revealed any effect of moment F (1.16) = 0.00080, *p* = 0.97, η <sup>2</sup> = 0.001, F (1.16) = 0.00063, *p* = 0.98, η <sup>2</sup> = 0.001, F (2.28) = 1.39, *p* = 0.26, η <sup>2</sup> = 0.09, F (2.28) = 2.81, *p* = 0.07, η <sup>2</sup> = 0.16, F (2.26) = 1.18, *p* = 0.32, η <sup>2</sup> = 0.08, and F (2.26) = 0.99, *p* = 0.38, η <sup>2</sup> = 0.07, respectively. Continuing with the same type of repeated measures ANOVA analysis with participant´s mean SJ, CMJ, Pmin, Pmax, YYIR1 and VO2max revealed a significant effect of moment, F (2.26) = 7.03, *p* = 0.003, η <sup>2</sup> = 0.35, F (2.26) = 20.20, *p* = 0.001, η <sup>2</sup> = 0.60, F (2.26) = 12.41, *p* = 0.001, η <sup>2</sup> = 0.48, F (2.26) = 8.84, *p* = 0.001, η2 = 0.40, F (2.18) = 10.26, *p* = 0.001, η <sup>2</sup> = 0.53, and F (2.16) = 9.84, *p* = 0.001, η <sup>2</sup> = 0.55.

The correlation coefficients between RSA indices (Pmax, Pmin, and Fatigue index) and fitness variables are summarized in Table 2. No significant correlations were found between all RSA indices and SJ, CMJ, 10m, 30m and COD. However, positive moderate correlations were found between Pmin and ADDs and ABDs [r = 0.51, *p* < 0.02 and r = 0.54, *p* < 0.01, respectively (See Figure 2]. In the same line, positive moderate correlations were found between Pmax and ADDs and ABDs (r = 0.55, *p* < 0.02 and r = 0.46, *p* < 0.01, respectively (see Figure 3)). Last, other interest and negative moderate correlation was found between FI and YYIR1 test [r = −0.62, *p* < 0.004 (Figure 4)].

The regression analysis to predict RSA from physical fitness variables was in agreement with the correlation analysis (See Table 3). On the one hand, ADDs and ABDs were predictor variables of Pmin (r = 0.53 and r = 0.55, respectively). On the other hand, ABDs was predictor variable of Pmax (r = 0.48). Finally, YYIR1 test was a predictor variable of IF (r = −0.53).


**Table 2.** Pearson correlation coefficient between RSA indices and fitness variables (*n* = 22).

ADD: adductor isometric strength; ABD: abductor isometric strength; SJ: squat jump; CMJ: countermovement jump; 10 m: 10-m sprint; 30 m: 30-m sprint; COD: change-of-direction; YYIRT1: Yo-Yo intermittent recovery test level 1; Pmin: peak power (minimum); Pmax: peak power (maximum); FI: fatigue index; cm: centimeters; s: seconds. \* Denotes significance at *p* < 0.05, and \*\* denotes significance at *p* < 0.01.

**Figure 2.** Relationship between hip adductor and abductor isometric strength (ADDs and ABDs) and Pmin of RSA test.

**Figure 3.** Relationship between hip adductor and abductor isometric strength (ADDs and ABDs) and Pmax of RSA test.

**Figure 4.** Relationship between Yo-Yo IR1 test and fatigue index (FI) of repeated-sprint ability test.

**Table 3.** Values of regression analysis explaining, Pmax, Pmin and Fatigue index based on the remaining variables.


Pmin: peak power (minimum); Pmax: peak power (maximum); FI: fatigue index; s: seconds.

#### **4. Discussion**

The present study aimed to analyze the determinants of RSA based on aerobic performance, linear sprinting and change-of-direction, vertical height jump, and abductor and adductor isometric strength. The main findings were as follows: (i) power in repeated sprints can be improved and predicted through exercises of ABD´s and ADD´s strength, and (ii) RSA can be improved and forecasted through aerobic endurance-based exercises such as the YYIR1 test. Additionally, it was found that RSA, and cardiorespiratory fitness had meaningfully improved over the season, while vertical jump had decreased.

The assessments performed repeatedly over the season revealed a meaningful improvement in RSA and aerobic performance. On the other hand, vertical jump decreased over the season, possibly due to the lack of reactive strength training or oriented training for this physical quality. Usually, both RSA and aerobic performance are key determinants of physical performance in soccer and match running-performance is associated with those capacities [16,46,47], thus it can be expected that over the season the training and match load may explain positive changes in RSA and aerobic performance [48,49].

High-intensity efforts, such as sprints, are essential components explaining soccer players´ behavior [50]. However, in addition to sprints in isolation, players perform repeated high-intensity efforts in short intervals, drawing from their aerobic endurance to do so [10]. In this sense, RSA seems to be a suitable method for inducing optimal improvements in anaerobic and aerobic metabolism [18,23–25], thus giving a team an advantage over the opponent during moments in matches characterized by high-speed efforts.

It seems that RSA can be improved through any soccer-specific training program [51], supporting the improvements found in this study over four- and 18-week female soccer training programs. However, the main determinants of RSA test performance among

female soccer players are still unclear. Therefore, the authors of the present study tried to analyze the most relevant variables of RSA by comparing indicators from the RSA test (power and fatigue index) with aerobic endurance (YYIR1 test), linear sprinting with COD, vertical height jump (SJ and CMJ test), and ABDs and ADD strength. The primary moderate correlations were established between power and ABD/ADD isometric strength (Pmin and ADD [r = 0.51], Pmin and ABDs [r = 0.54], Pmax and ADDs [r = 0.55], and Pmax and ABDs [r = 0.46]) and between fatigue and YYIR1 test outcomes [r = −0.62]. No other correlations were found between RSA parameters and other tests.

The relationship between power and ABD/ADD isometric strength may be due to the implication of these muscle groups in sprinting efforts [52]. From an anatomy-based or biomechanic viewpoint, ADD assists hip flexion and neutralizes the abduction and external rotation caused by tensor fascia latae and Sartorius. In addition, during the midto-late swing, when the hip is flexed, adductors work as synergists of the gluteus maximus, helping with hip extension and counterbalancing external rotation [52]. On the other hand, the ABDs stabilize the femoral head during high-speed running efforts. They lengthen eccentrically while helping to stabilize the pelvis and control femoral adduction in the transverse plane [53].

Researchers have tried to analyze the implications of hypertrophy of these muscles during sprints. For example, Nuell et al. [52] and Tottori [54] highlighted the implications of and the close relationship between ADD and sprint performance, while Fredericson and Weir [53] highlighted the implications of ABD in gait and sprints. Interestingly, it seems that the implication of ADD correlates with sprinting time [52] and with sprinting distance [54]. These findings encourage physical fitness and conditioning coaches to design coadjutant training programs based on ADD and ABD isometric strength to improve female soccer players' sprint performance. However, no relationship was found between RSA parameters and exercises with significant quadriceps implications (i.e., SJ, CMJ, and COD).

This idea seems consistent among experts in this topic [55,56], who have concluded that the quadriceps are not related to sprint performance. Instead of the quadriceps, it may be anatomically due to the anterior and middle parts of the gluteus medius, which have a stronger vertical pull and help initiate abduction, which is then completed by the tensor fascia lata [53].

Nevertheless, in addition to strength exercises, aerobic endurance remains crucial during female soccer matches. In the present study, the authors found correlations between fatigue and YYIR1 test results (*p* < 0.004; Figure 4). This finding is supported by Gabrys et al. [57], who concluded that the anaerobic glycolytic system is more sensitive to long, repetitive sprints, highlighting that RSA is a suitable strategy for avoiding insufficient aerobic energy systems, which lead to early decreases in performance [57]. All of these results indicate the value of forecasting Pmax from ADD and ABD isometric strength values (r = 0.53 and r = 0.55, respectively), Pmax from ABD values (r = 0.48), and fatigue from YYIR1 (r = −0.53).

This study had some limitations. The force platforms were not used to calculate the rate of force development during vertical jumps, and this could be interesting. Additionally, an isometric mid-thigh pull test would be interesting to associated with RSA. Future studies should consider analyzing the influence of each physical capacity in different number of sprints, and also consider analyzing COD deficit and asymmetries trying to understand if this can be related with RSA ability. Other limitation is associated with small sample and the specificity of being conducted in women, thus not being possible to generalize for other populations. More research should be conducted to test the replication of results in different scenarios (other competitive contexts, age-groups and populations).

As practical applications, the coadjutant training program—mainly based on these determinants (ABD and ADD isometric strength exercises and YYIR1)—may induce improvements in female soccer players' RSA and better outcomes during critical moments of matches. Although it was declared that straight sprinting is the most frequent action taken before goals, both for scoring and assisting players [51], the current trend highlights that

sprints during soccer games are curvilinear [58–60]. As such, they may lead to different demands than straight sprints [57]. Therefore, further studies should assess the main determinants of curvilinear sprinting performance during RSA tests.

#### **5. Conclusions**

Power and fatigue are notable RSA-related parameters. Power during RSA is mainly determined by ABD and ADD isometric strength, while fatigue is related to YYIR1. Therefore, physical fitness and conditioning coaches are encouraged to improve ABD and ADD isometric strength alongside aerobic endurance. Doing so may lead to improvements in RSA, subsequently giving the player an advantage over the opponent during critical game situations. However, since it seems that most sprint efforts are made in a curvilinear trajectory, future studies should replicate the present study, focusing on these efforts.

**Author Contributions:** L.G., F.M.C. and J.M.C.C. lead the project, established the protocol and wrote and revised the original manuscript; J.I.B. and H.S. collected the data, wrote and revised the original manuscript; F.T.G.-F. and M.R.-G. wrote and revised the original manuscript. 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 Institutional Review Board (or Ethics Committee) of Polytechnic Institute of Viana do Castelo, School of Sport and Leisure (code: CTC-ESDL-CE001-2021).

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

**Acknowledgments:** Filipe Manuel Clemente: This work is funded by Fundação para a Ciência e Tecnologia/Ministério da Ciência, Tecnologia e Ensino Superior through national funds and when applicable co-funded EU funds under the project UIDB/50008/2020. Hugo Sarmento gratefully acknowledge the support of a Spanish government subproject Integration ways between qualitative and quantitative data, multiple case development, and synthesis review as main axis for an innovative future in physical activity and sports research [PGC2018-098742-B-C31] (Ministerio de Economía y Competitividad, Programa Estatal de Generación de Conocimiento y Fortalecimiento Científico y Tecnológico del Sistema I+D+i), that is part of the coordinated project 'New approach of research in physical activity and sport from mixed methods perspective (NARPAS\_MM) [SPGC201800X098742CV0]'.

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

#### **References**


### *Article* **Training Habits of Eumenorrheic Active Women during the Different Phases of Their Menstrual Cycle: A Descriptive Study**

**Felipe García-Pinillos 1,2 , Pascual Bujalance-Moreno <sup>3</sup> , Daniel Jérez-Mayorga <sup>4</sup> , Álvaro Velarde-Sotres 5,6 , Vanessa Anaya-Moix 6,7, Silvia Pueyo-Villa 6,7 and Carlos Lago-Fuentes 5,\***


**Abstract:** The purpose of this study was to examine the training habits of eumenorrheic active women during their menstrual cycle (MC), and its perceived influence on physical performance regarding their athletic level. A group of 1250 sportswomen filled in a questionnaire referring to demographic information, athletic performance and MC-related training habits. Of the participants, 81% reported having a stable duration of MC, with most of them (57%) lasting 26–30 days. Concerning MC-related training habits, 79% indicated that their MC affects athletic performance, although 71% did not consider their MC in their training program, with no differences or modifications in training volume or in training intensity for low-level athletes (LLA) and high-level athletes (HLA) with hormonal contraceptive (HC) use. However, LLA with a normal MC adapted their training habits more, compared with HLA, also stopping their training (47.1% vs. 16.1%, respectively). Thus, different training strategies should be designed for HLA and LLA with a normal MC, but this is not so necessary for HLA and LLA who use HC. To sum up, training adaptations should be individually designed according to the training level and use or non-use of HC, always taking into account the pain suffered during the menstrual phase in most of the athletes.

**Keywords:** gender; training load; health surveys; sport participation

#### **1. Introduction**

The presence and popularity of physical activity and sport for women has considerably increased in the last few decades, and therefore, the need to improve knowledge about their physiology and adaptations to exercise has become crucial. Traditionally, the physiological responses to exercise were assumed to be equal between the sexes [1], and so the recommendations about sport practice and prescription for women have been generalized for decades without even testing whether these guidelines were correct. A potential rationale for this underrepresentation is the complexities and methodological difficulties related to the menstrual cycle (MC) [1]. Based on that argument, sports sciences have been focused on men, or have included women without considering the MC's influence [2].

A typical MC lasts around 28 days and consists of a follicular phase (i.e., characterized by 12–14 days' duration, high levels of estrogen and low progesterone), ovulation phase (i.e., 1–3 days' duration and preceded by a second increase in estrogen), and a luteal

**Citation:** García-Pinillos, F.; Bujalance-Moreno, P.; Jérez-Mayorga, D.; Velarde-Sotres, Á.; Anaya-Moix, V.; Pueyo-Villa, S.; Lago-Fuentes, C. Training Habits of Eumenorrheic Active Women during the Different Phases of Their Menstrual Cycle: A Descriptive Study. *Int. J. Environ. Res. Public Health* **2021**, *18*, 3662. https:// doi.org/10.3390/ijerph18073662

Academic Editors: Filipe Manuel Clemente and Ana Filipa Silva

Received: 13 March 2021 Accepted: 28 March 2021 Published: 1 April 2021

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

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

phase (i.e., 12–14 days' duration, with high levels of progesterone and medium levels of estrogen) [3]. Plentiful information has been made available about the role of estrogen and progesterone in the female physiology during the last few decades. Regarding estrogen, it has been demonstrated that it increases the muscle glycogen storage capacity as well as increasing free fatty acid availability, and it is used as a fuel source through the use of oxidative pathways [4]. From a physiological standpoint, this is very important because it leads to decreased carbohydrate use, or glycogen sparing [5], and it therefore decreases the dependence on anaerobic pathways for Adenosine triphosphate (ATP) production. In practical terms, high estrogen levels are associated with lower blood lactate levels and longer times to exhaustion [5]. Progesterone is known to have a sympathetic effect (e.g., it increases resting heart rate [6], basal body temperature and ventilation) [7]. This effect plays a key role during exercise because it seems to increase the perceived exertion and decrease athletic performance, especially in hot and/or humid environments [8]. Beyond these physiological effects, what is crucial to understand about progesterone is its ability to antagonize estrogenic effects [1,9]. In this regard, a previous study showed that high progesterone levels can inhibit the enhancement of the carbohydrate metabolism promoted by estradiol, which is a primary estrogen [10].

Taking into consideration the above information, it seems clear that both hormones have different target organs and create diverse biochemical and physiological environments, which have been suggested to be determinant in exercise capacity and thus adaptations to training [8,11,12]. Nevertheless, there is no consensus about the influence of hormone variations induced by the MC's phases on physical performance, with other works reporting no effect [13,14].

Once the physiological findings are revealed, the next question is to learn the influence of these changes in athletes. For this, the use of questionnaires to analyze vast populations is a common strategy in sports science. In this regard, some previous studies have used self-reported questionnaires to examine topics related to the MC [15,16]. Martin et al. [16] aimed to determine the prevalence of hormonal contraceptive (HC) use and the side effects experienced by users and non-users in elite female athletes, whereas Bruinvels et al. [15] focused on heavy menstrual bleeding (HMB) and its perceived effects on training and performance. However, to the best of the authors' knowledge, no previous studies have examined the influence of their MC and their phases on training habits in eumenorrheic active women.

Since female athletes keep training and competing whilst having to manage hormone alterations and their effects, a description of what female athletes are doing and how they manage the potential effects of the MC might be of interest for coaches, athletes and sports scientists. Although research based on a questionnaire can show some limitations, this is the best option to make a survey with a large sample size, with information directly collected by sportswomen regarding their own MC and its influence on their training habits. For these reasons, the purpose of this study is to examine the training habits of eumenorrheic active women during their MC and its perceived influence on physical performance. The current study also aims to investigate the differences in MC-related training habits regarding the athletic level and the use or non-use of hormonal contraceptives (HC).

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

#### *2.1. Subjects*

One thousand, two hundred and fifty eumenorrheic active women voluntarily participated in this study (age range: 18-45 years; age mean: 27.8 ± 6.6 years). All participants met the inclusion criteria: (1) older than 18 and younger than 45 years old, and (2) with three or more training sessions per week [17]. This study meets the ethical standards of the World Medical Association's Declaration of Helsinki (2013), and it was approved by the Institutional Review Board (Universidad de La Frontera, Temucho, Chile, 005\_19).

#### *2.2. Procedure*

A cross-sectional study with a descriptive purpose was performed. An ad-hoc questionnaire was designed for a massive mailshot to physically active women through an online Google Form (https://drive.google.com/open?id=1Pw5AecF3Wqhn-1dn13IhqZxZW2 QjiyxTtkmsz55Bs28) (accessed on 15 September 2019). This research project was conducted according to the European General Data Protection Regulation (2018).

After receiving ethical approval from the Institutional Review Board, pilot tests were conducted with a small sample of participants (*n* = 40) to evaluate the clarity and content of the online Google Forms. All participants involved in the pilot test indicated that the questionnaire was appropriate and suitable. Subsequently, sports centers, sports clubs, federations and sports institutes in Spain were contacted through their administrators and asked to publicize the study to their athletes and clients as long as those sports organizations were in line with the current data protection regulations, implying that athletes were informed about the potential use of their personal data for research purposes when they provided such information. Then, athletes who were willing to participate in the study were given a link to the online questionnaire.

According to online informed consent procedures, participants were told of the purpose and details of the study through a participant information sheet. Participants were informed that all responses would be kept strictly confidential and would only be used for the purposes of the study. Having consented to participation in the study, participants filled in seventeen items split into four sections: (i) demographic information (i.e., age), (ii) information about athletic performance in the last 6 months (i.e., to have a coach, to be federated, sport modality and athletic level), (iii) information about their training habits in the last 6 months (i.e., hours and sessions per week), and (iv) information about their MC and MC-related training habits (i.e., age at menarche, duration and regularity of their MC, use of HC, the perceived influence of their MC on physical performance, MC-related modifications in training plans, and pain influence).

#### *2.3. Statistical Analysis*

Descriptive data are presented as means and standard deviation for interval variables, and as frequency and percentage for nominal variables. Six level groups were determined according to self-reported athletic level (non-competitive level or level 0, local or level 1, autonomic or level 2, national or level 3, international or level 4, and elite or level 5). To compare differences in the adaptation of the parameters of training during the MC, the six groups were dichotomized into two according to their athletic level: lower-level athletes (LLA) (groups 0, 1 and 2), and higher-level athletes (HLA) (groups 3, 4 and 5). The chi-squared test was conducted to determine differences between level groups with and without HC. All statistical analyses were performed using the software package SPSS (IBM SPSS version 22.0, Chicago, IL, USA). The effect size was calculated following previous studies [18].

#### **3. Results**

As general information about this group of 1250 women, the results obtained indicate that participants trained 6.7 ± 3.4 h per week, distributed over 4.3 ± 2.1 sessions per week (Table 1). Among these women, 64.7% had a coach and 33.6% were federated. The most practiced sport modalities were team sports (25.9%), athletics (24.6%) and fitness-related activities (19.7%), whereas other modalities such as CrossFit (9.8%), cycling and dancing (5.8%) also showed moderate levels of practice among active women. The self-reported age at menarche was 12.7 ± 1.7 years.


**Table 1.** Descriptive data of active women related to sport modality.

Concerning the athletic level, regardless of sport modality, most of the surveyed women (60.7%) indicated a non-competitive level (level 0), whereas the rest of the participants reported a competitive level (at local level or level 1–9.8%; autonomic or level 2–7.7%; national or level 3–18.7%; international or level 4–2.4%; and elite or level 5–0.6%).

Regarding the profile of the MC in relation to the different athletic levels (Table 2), 81% reported having a stable duration of MC with most of them (57%) lasting 26–30 days. No statistical differences were found in the length of the MC when comparing different athletic levels, except with the elite level (*n* = 8, *p* < 0.001, effect size (ES) = 0.42). Regarding the regularity of the MC, international athletes reported a lower percentage (60% regularity) compared with the rest of the groups. In general, 28% reported using an HC method, showing that 43.8% of the level 2 group used HC.


**Table 2.** Profile of the menstrual cycle (MC) related to different athletic levels.

Note: percentages are calculated according to the number of sportswomen per level group. ES: effect size, MC: menstrual cycle; HC: hormonal contraceptives; LG0: non-competitive level; LG1: local level; LG2: autonomic level; LG3: national level; LG4: international level; LG5: elite level.

> Table 3 includes a comparison between different athletic levels (HLA vs. LLA) with a normal cycle (no HC use). Of the LLA, 80.3% indicated that their MC affects athletic performance, with statistical differences compared to HLA. However, almost 70% did not consider the MC in their training program for both groups. Regarding training volume and intensity, LLA affected both variables more during their menstrual phase (MP) compared with HLA (*p* < 0.005 and *p* < 0.001, respectively). Almost 50% of the LLA stopped their training during their MP, while <20% of HLA did so (*p* < 0.001; ES = 0.84). Lastly, among

the sportswomen, 55% to 63% reported that they suffered pain during their MP, without differences between both groups.


**Table 3.** Influence of the menstrual cycle (MC) on training habits, related to different athletic levels without contraceptive hormones (*n* = 905).

Note: percentages are calculated related to the number of sportswomen per level group. ES: effect size; MP: menstrual phase; MC: menstrual cycle; HLA: higher-level athletes, LLA: lower-level athletes.

Table 4 describes the influence of the MC on training habits, comparing HLA and LLA athletes. Both LLA and HLA athletes consuming HC reported their performance being affected. However, only 9.1% of HLA reported no effect, against 28% of LLA (*p* < 0.005; ES = 0.47). Regarding the rest of the variables, no statistical differences were reported when comparing both groups, considering the effect of the MC in their training (*p* = 0.669), neither adapting their volume or training, nor stopping training during their MP (*p* = 0.656, *p* = 0.09 and *p* = 0.143, respectively). Finally, both groups suffered pain during their MP, with between 55 and 63% of athletes reporting it, with no differences between groups.

**Table 4.** Influence of the menstrual cycle (MC) on training habits, related to different athletic levels with contraceptive hormones (*n* = 345).


Note: percentages are calculated related to the number of sportswomen per level group. ES: effect size; MP: menstrual phase; MC: menstrual cycle; HLA: higher-level athletes, LLA: lower-level athletes.

#### **4. Discussion**

This study aimed to examine the training habits of eumenorrheic active women during their MC and its perceived influence on their physical performance. The results obtained indicate that, despite a high percentage of the surveyed women confirming that

the MC affects physical performance and reporting feeling pain during the MP, most of them reported making no changes to their training programs during the MC, with no modifications in training volume or intensity during the MP. This information is of interest as it reinforces the importance of the MC in the training plans of eumenorrheic active women.

Until now, there has been no clear evidence about the relationship between MC and sports performance in different physical outcomes [13,16,19–21]. Regarding this, almost 80% of athletes indicated that the MC affects their performance, with more influence on LLA with a normal MC and HLA with HC consumption. Athletes also suffered pain during the MP, with no differences between groups, independent of the use or non-use of HC. These results match previous studies that indicated most athletes suffered negative side effects during the early days of the MP, affecting their sports performance throughout the cycle [15,16] because of HMB, stomach cramps, back pain and headaches, among other symptoms. Conversely, exercise has been shown to improve the symptoms of premenstrual syndrome, inter alia [22], so female athletes should consider managing training variables (e.g., volume and intensity) during the MC, especially during the MP, but not stop training.

Given the lack of consensus about the influence of the hormone variations associated with the MC, the results provided by the current work might be of interest for coaches and athletes. For instance, whereas it has been suggested that age at menarche might be influenced by the level of sports practice and specialization in the sport [23], the data provided by this current study indicate that age at menarche was similar for all the sports in the study. About this, a previous study suggested that, in gymnastics, menarcheal age was delayed compared with other athletes [23]. However, the authors could not justify this finding regarding training levels, so according to our results, menarcheal age could not be influenced by training levels [23]. Further longitudinal studies should be performed with young female athletes to analyze possible changes to the menarcheal age and the influence of training levels. In this context, another interesting finding is the lack of differences in the length of the MC between level groups. Martin et al. [16] showed that the length of the MC can vary regarding athletic level, with some negative side effects such as primary dysmenorrhea. However, this does not match our results, with more than 50% of HLA reporting a duration of 26 to 30 days. Furthermore, most groups reported a high regularity of the MC, which is key to female performance and the first step for designing training programs according to their different phases [19]. Lastly, only 27.5% used HCs, which is a low prevalence compared with previous studies, and almost 60% of the British elite athletes [16]. HC can influence the regularity of the MC as well as reducing dysmenorrhea, but it also affects sport performance [16].

Additionally, this study also aimed to investigate the differences in MC-related training habits concerning athletic level, comparing results with and without the use of HC. Regarding this, it has been suggested that athletic level can influence the MC [16]. Firstly, regarding athletes with a natural MC (no HC use), the HLA adapted their training much less according to their MC than did the LLA (i.e., the LLA skipped more training sessions than HLA). Intensity, one of the main factors to control during the MC [20], was modified in 48.6% of the LLA compared with only 30.8% in the HLA group during the MP. That is, the HLA did not modify their workouts, regardless of the negative side effects (such as pain) associated with the MP during their preparation. Okano et al. [24] showed that the athletic level influenced the prevalence of eating disorders in Japanese and Chinese athletes, revealing that HLA are more prone to suffering them. So, training at a high level without considering hormone fluctuations during the MC (and especially the negative side effects before and during the MP) could be dangerous for female health and also increase injury risk [25]. In fact, a recent study stated that periodizing the strength training according to the different phases of the MC improves the gaining of lean body mass compared with traditional training [26]. Secondly, comparing both athletic levels via HC use reported different findings. In general, both groups felt pain during the MP, without statistical differences. No differences were also reported regarding the adaptation of training habits

in either variable (intensity and volume) during the MP. Curiously, 90% of HLA with HC reported an effect on their performance, compared with only 72% of LLA. In this sense, a recent systematic review suggested that performance did not differ between the different phases of the MC in HC athletes, but their performance can be slightly inferior compared with a non-HC user [27]. Taking into account that HLA athletes using HC felt more effect on their performance compared with HLA with a natural MC, and the same percentage of pain was reported, HC might not be the best option for top athletes, due to the negative side effects of its use [27].

In summary, the results indicate that, despite a high percentage of the surveyed women confirming that the MC affects physical performance and feeling pain during their MP, most of them reported making no changes to their training programs during the MC, with no modifications in training volume nor intensity during the MP. Comparing HC use to non-use, LLA with a natural MC adapted their training variables more during the MP compared to HLA. On the other hand, athletes using HC did not differ in their training adaptations regarding their athletic level. Apart from these key points, it is relevant to take into account that more than 50% of athletes (in both groups) suffered pain during the MP. This information is essential for coaches and practitioners to understand and adapt training loads when pain is present during the MC. Further strategies should analyze different adaptations of training plans throughout the MC, and particularly during the MP, to as far as possible reduce the pain and optimize their sport performance. For these reasons, monitoring and programming training loads according to athletic level, their type of MC (use or non-use of HC), and the different phases of the MC might be relevant during the training process according to these results. Future studies should apply different intensities and frequencies during the MP to compare the effects on sport performance, especially in a natural MC.

#### **5. Conclusions**

The novel aspect of this research regards the level of influence of the menstrual cycle on the performance and training habits of eumenorrheic active women. Defining and knowing the side effects during each phase of the menstrual cycle at different performance levels and sports modalities should be relevant to adapting training programs properly and reducing non-practice times during the menstruation phase, both in HC and non-HC users. For this reason, staff and physical education coaches should be aware of the importance of managing and registering the menstrual phases of women to optimize their training and adapt training loads, especially during the bleeding or menstrual phase. Moreover, these results could also be useful to compare different performances regarding sports modalities.

To sum up, this study provides descriptive information about the MC of eumenorrheic active women, and the modifications performed in their training programs in relation to the different phases of the MC. Given the reported hormonal changes during the different phases of the MC, both sports scientists and coaches must pay special attention to the role of this factor, and also to their negative side effects in some phases, suggesting the need for an educational process and a constant dialog with their athletes about their feelings, negative side effects (if they exist) and pain during the MP, independently of the use or non-use of HC.

**Author Contributions:** Conceptualization, F.G.-P., P.B.-M. and C.L.-F.; methodology, F.G.-P., D.J.-M. and C.L.-F.; software, C.L.-F. and F.G.-P.; validation, S.P.-V. and V.A.-M.; formal analysis: F.G.-P., P.B.-M. and C.L.-F.; investigation, F.G.-P. and P.B.-M.; resources, F.G.-P.; data curation, F.G.-P. and P.B.-M.; writing—original draft preparation, F.G.-P., Á.V.-S. and D.J.-M.; writing—review and editing, F.G.-P., S.P.-V. and C.L.-F.; visualization, F.G.-P. and P.B.-M.; supervision, F.G.-P. and C.L.-F.; project administration, F.G.-P. and P.B.-M. All authors have read and agreed to the published version of the manuscript.

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

**Institutional Review Board Statement:** This study meets the ethical standards of the World Medical Association's Declaration of Helsinki (2013), and it was approved by the Institutional Review Board (Universidad de La Frontera, Temucho, Chile, 005\_19).

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

**Data Availability Statement:** Not applicable.

**Acknowledgments:** This research was supported by the Pre-competitive Projects for Early Stage Researchers Program from the University of Granada (ref: PPJIA2020.03). The authors would like to thank all the participants.

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

#### **References**


MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*International Journal of Environmental Research and Public Health* Editorial Office E-mail: ijerph@mdpi.com www.mdpi.com/journal/ijerph

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com ISBN 978-3-0365-3132-8