# **New Training Strategies and Evaluation Methods for Improving Health and Physical Performance**

Edited by Catarina Nunes Matias, Stefania Toselli, Cristina Monteiro, Francesco Campa

Printed Edition of the Special Issue Published in *International Journal of Environmental Research and Public Health*

www.mdpi.com/journal/ijerph

## **New Training Strategies and Evaluation Methods for Improving Health and Physical Performance**

## **New Training Strategies and Evaluation Methods for Improving Health and Physical Performance**

Editors

**Catarina Nunes Matias Stefania Toselli Cristina Monteiro Francesco Campa**

MDPI Basel Beijing Wuhan Barcelona Belgrade Manchester Tokyo Cluj Tianjin

*Editors* Catarina Nunes Matias Bettery Life Lab Universidade Lusofona ´ Lisbon Portugal

Stefania Toselli DIBINEM University of Bologna BOLOGNA Italy

Cristina Monteiro Laboratory of Physiology and Biochemistry of Exercise, Faculdade de Motricidade Humana Universidade de Lisboa Lisbon Portugal

Francesco Campa Department of Biomedical Sciences University of Padova Padova Italy

*Editorial Office* MDPI St. Alban-Anlage 66 4052 Basel, Switzerland

This is a reprint of articles from the Special Issue published online in the open access journal *International Journal of Environmental Research and Public Health* (ISSN 1660-4601) (available at: www.mdpi.com/journal/ijerph/special issues/New Training Strategies Evaluation Methods Improving Health Physical Performance).

For citation purposes, cite each article independently as indicated on the article page online and as indicated below:

LastName, A.A.; LastName, B.B.; LastName, C.C. Article Title. *Journal Name* **Year**, *Volume Number*, Page Range.

**ISBN 978-3-0365-4252-2 (Hbk) ISBN 978-3-0365-4251-5 (PDF)**

Cover image courtesy of Francesco Campa

© 2022 by the authors. Articles in this book are Open Access and distributed under the Creative Commons Attribution (CC BY) license, which allows users to download, copy and build upon published articles, as long as the author and publisher are properly credited, which ensures maximum dissemination and a wider impact of our publications.

The book as a whole is distributed by MDPI under the terms and conditions of the Creative Commons license CC BY-NC-ND.

## **Contents**


Performed by Competitive Bodybuilders: Implications for Resistance Training Reprinted from: *Int. J. Environ. Res. Public Health* **2021**, *18*, 772, doi:10.3390/ijerph18020772 . . . **169**

#### **Kelsey Denby, Ronald Caruso, Emily Schlicht and Stephen J. Ives**

The Impact of Wrist Percooling on Physiological and Perceptual Responses during a Running Time Trial Performance in the Heat Reprinted from: *Int. J. Environ. Res. Public Health* **2020**, *17*, 7559, doi:10.3390/ijerph17207559 . . . **181**

#### **Pablo Jorge Marcos-Pardo, Noelia Gonz´alez-G´alvez, Gemma Mar´ıa Gea-Garc´ıa, Abraham L´opez-Vivancos, Alejandro Espeso-Garc´ıa and Rodrigo Gomes de Souza Vale**

Sarcopenia as a Mediator of the Effect of a Gerontogymnastics Program on Cardiorespiratory Fitness of Overweight and Obese Older Women: A Randomized Controlled Trial Reprinted from: *Int. J. Environ. Res. Public Health* **2020**, *17*, 7064, doi:10.3390/ijerph17197064 . . . **199**

#### **MunHee Kim, Wi-Young So and Jiyoun Kim**

Relationships between Exercise Modality and Activity Restriction, Quality of Life, and Hematopoietic Profile in Korean Breast Cancer Survivors

Reprinted from: *Int. J. Environ. Res. Public Health* **2020**, *17*, 6899, doi:10.3390/ijerph17186899 . . . **215**

#### **Francesco Campa, Analiza M. Silva, Catarina N. Matias, Cristina P. Monteiro, Antonio Paoli and Jo˜ao Pedro Nunes et al.**

Body Water Content and Morphological Characteristics Modify Bioimpedance Vector Patterns in Volleyball, Soccer, and Rugby Players

Reprinted from: *Int. J. Environ. Res. Public Health* **2020**, *17*, 6604, doi:10.3390/ijerph17186604 . . . **227**

#### **Ozkan G ¨uler, Dicle Aras, Fırat Ak¸ca, Antonino Bianco, Gioacchino Lavanco and Antonio ¨ Paoli et al.**

Effects of Aerobic and Anaerobic Fatigue Exercises on Postural Control and Recovery Time in Female Soccer Players

Reprinted from: *Int. J. Environ. Res. Public Health* **2020**, *17*, 6273, doi:10.3390/ijerph17176273 . . . **241**

## **About the Editors**

#### **Catarina Nunes Matias**

Catarina N. Matias has a Masters degree in biochemistry and a Ph.D. (Summa cum laude) in Human kinetics in the Specialty of Physical Activity and Health at the Faculty of Human Kinetics at the University of Lisbon. Catarina has extensive experience in teaching, in the first cycle of studies at the University level within the curricular units of Biochemistry, Nutrition and Exercise, and Exercise Physiology, as well as in the advanced modules of the University masters and doctorates, either by the Faculty of Human Kinetics of the University of Lisbon or by the Faculty of Physical Education and Sport of the Lusofona University. This activity resulted in the supervision of seven ´ masters and one doctorate. Parallel to the academics, Catarina developed a scientific approach which resulted in collaborations and coordination of 15 research projects, 5 scientific peer awards and distinctions, 3 dissertations, 1 international book chapter, 2 research reports, 76 scientific articles published in international peered review journals with impact factor, 69 scientific abstracts published in international journals with impact factor, and in the book of abstracts of national and international scientific congresses of specialty, and 62 oral or panel communications at international scientific congresses. Catarina is currently a researcher at CIDEFES, a research center at the Faculty of Physical Education and Sport at the University of Lusofona, where she is also an Assistant Professor. In ´ addition, Catarina ensures the Coordination of Development & Research Laboratories, as well as Biochemistry, Body Composition and Metabolism, and Performance Laboratories at Bettery, S.A . The areas of expertise includes sports biochemistry, sports nutrition, metabolic regulation in exercise and immune response to exercise and training, with multiple scientific outputs and masters'thesis supervised within this fields.

#### **Stefania Toselli**

Stefania Toselli is associate professor of the Department of Biomedical and Neuromotor Sciences, Bologna, Italy. The main field of her research is represented by human biology, considering both adults and subjects during growth. In particular, the considered aspects regard auxology, secular trend, weight status, body composition (both considering methodological aspects and variability in different populations), somatotype, and body image, with specific reference to environmental and physical activity influence.

#### **Cristina Monteiro**

Cristina Monteiro has a degree in Biochemistry and a Ph.D. in Human Movement Sciences and is an Assistant Professor at the Faculty of Human Kinetics of the University of Lisbon teaching mainly exercise biochemistry and nutrition. Her scientific activity has been integrated in the Interdisciplinary Center for the Study of Human Performance (CIPER) of the Faculty of Human Kinetics. She was a member of the research team of three funded projects, two by the FCT, one by the Portuguese-French Scientific and Technical Cooperation Program. Additionally to this, her participation in several interdisciplinary investigations was central to the development and integration of her areas of expertise in the sports sciences and health fields, making the connection between them.

The areas of expertise includes sports biochemistry, sports nutrition, metabolic regulation in exercise and immune response to exercise and training, redox balance in exercise and disease and magnesium metabolism, with multiple scientific outputs and masters'thesis supervised within this fields. Recently she has supervised two Ph.D. thesis, one exploring the immune response to acute exercise, including the recovery period throughout 24 h, and training; and one other that sought to compare the effect of several leucine metabolites, namely two commercially available forms of beta-hydroxy-beta-methylbutyrate and leucic acid, on resistance training-induced changes induced on performance, body composition, and biochemical markers of muscle damage, inflammation and anabolic and catabolic hormones. These works have produced several research outputs in high level journals and show her previous experience in applying follow up studies with athletes in the field and randomized control trails to test the effects of nutritional supplements.

#### **Francesco Campa**

Francesco Campa is a researcher at the Department of Biomedical Sciences at the University of Padova (Italy), received his Ph.D. in pharmacology and toxicology, human development, and movement sciences from the University of Bologna. He has published one book on sports anthropometry, several book chapters, and over 70 peer-reviewed journal and conference papers (https://orcid.org/0000-0002-3028-7802), with an H-index of 16. His main interests are sports anthropometry, bioimpedance vector analysis, and body composition optimization applied to sports performance.

## **Preface to "New Training Strategies and Evaluation Methods for Improving Health and Physical Performance"**

Physical activity is among the most effective methods for improving health, body composition, and physical function, and its practice is suitable for every population. Its benefits are known for sedentary individuals who, by initiating sport, improve their physical condition by reducing risk factors. Active training is also encouraged for the general population who need to maintain an optimal level of fitness, as well as for athletes who want to achieve high performance during competitive periods. Even young people benefit from sports practice, growing into healthy young adults with important implications for their psychological and social development. In the last few years, the scope of research in sports has become very wide and detailed, laying the foundations for the development of innovative training methods and new evaluation approaches aimed at improving health, body composition, and performance. Contemporary researchers have contributed to the field of body composition research in the development of new measurement methods and training strategies. The aforementioned aspects have laid the foundations for the development of innovative techniques and new evaluation approaches aimed at improving and assessing body composition and sports performance. In these contexts, the bioelectrical impedance analysis was proposed as a valid method to quantify body composition elements (e.g., fat and fat-free mass, body fluids, muscle mass) and is based on predictive equations or the qualitative interpretation of the raw data. On the other hand, innovative training strategies aimed at improving body composition and performance have been presented. The aim of this Special Issue was to propose, on the basis of the evidence that the current literature provides, new training techniques and specific evaluation methods for the different populations practicing physical activity.

#### **Catarina Nunes Matias, Stefania Toselli, Cristina Monteiro, and Francesco Campa** *Editors*

## *Review* **Exercise Dose Equalization in High-Intensity Interval Training: A Scoping Review**

**Tom Normand-Gravier 1,2, Florian Britto 1,3, Thierry Launay 1,3, Andrew Renfree <sup>4</sup> , Jean-François Toussaint 1,2,5 and François-Denis Desgorces 1,2,\***

	- Institute Cochin, U1016 INSERM, 75014 Paris, France

**Abstract:** Based on comparisons to moderate continuous exercise (MICT), high-intensity interval training (HIIT) is becoming a worldwide trend in physical exercise. This raises methodological questions related to equalization of exercise dose when comparing protocols. The present scoping review aims to identify in the literature the evidence for protocol equalization and the soundness of methods used for it. PubMed and Scopus databases were searched for original investigations comparing the effects of HIIT to MICT. A total of 2041 articles were identified, and 169 were included. Of these, 98 articles equalized protocols by utilizing energy-based methods or exercise volume (58 and 31 articles, respectively). No clear consensus for protocol equalization appears to have evolved over recent years. Prominent equalization methods consider the exercise dose (i.e., energy expenditure/production or total volume) in absolute values without considering the nonlinear nature of its relationship with duration. Exercises resulting from these methods induced maximal exertion in HIIT but low exertion in MICT. A key question is, therefore, whether exercise doses are best considered in absolute terms or relative to individual exercise maximums. If protocol equalization is accepted as an essential methodological prerequisite, it is hypothesized that comparison of program effects would be more accurate if exercise was quantified relative to intensity-related maximums.

**Keywords:** training programs; physical activity; effort; patients; athletes

#### **1. Introduction**

Exercise is both described and prescribed on the basis of two main variables: intensity (i.e., level of muscular activity) and volume (e.g., duration, distance or number of repetitions of an interval or set, and of the entire session) [1,2]. Notably for the interval exercise modality, these major variables also depend on possible recovery pauses within the exercise bout, inducing a third exercise variable, called by some authors "exercise density" (i.e., work/recovery ratio but also intensity level of the recovery) [1,3,4]. For quantifying and designating the overall exercise performed, authors can use generic terms accounting for all exercise variables, such as exercise dose in exercise-induced health studies [1] or training load for athlete monitoring purposes [5–7]. Defining effort as what is required to achieve a task in line with individual maximal capacities [8], exercise dose and training load might refer to the quantity of exercise-induced effort [5,6].

The control and calibration of training protocols should be a prerequisite in exercise and sport science studies, and insufficient consideration of this may result in confusion regarding exercise program effects [9,10]. Manipulation of training variables (volume, intensity and density) might ensure that the effort level generated by two protocols being compared is similar, or in other terms, that their exercise dose is equalized. However,

**Citation:** Normand-Gravier, T.; Britto, F.; Launay, T.; Renfree, A.; Toussaint, J.-F.; Desgorces, F.-D. Exercise Dose Equalization in High-Intensity Interval Training: A Scoping Review. *Int. J. Environ. Res. Public Health* **2022**, *19*, 4980. https:// doi.org/10.3390/ijerph19094980

Academic Editor: Catarina Nunes Matias

Received: 21 March 2022 Accepted: 17 April 2022 Published: 20 April 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/).

methodologically these comparisons are not easy to conduct. Targeting large populations, recommendations for physical activity frequently use absolute values for intensity or, sometimes, exercise durations characterized by large intensity ranges (e.g., light, moderate, vigorous) that could complicate the quantification of an individualized and unique dose value [1]. Viana et al. suggested that conclusions about high-intensity interval training (HIIT) remain difficult to draw because of insufficient control of the numerous exercise variables [11]. Recently, the lack of protocol equalization in HIIT and moderate-intensity continuous training (MICT) has been suggested to represent a possible methodological bias limiting studies' conclusions [12]. Comments on this paper suggest that consensus was not reached in the methods used for equalizing protocols nor, more surprisingly, in the necessity for equalizing them [13]. Limits and issues raised by equalization methods based on energy expenditure, although largely recommended, have only been recently reported [14,15]. Similar debates on adequate terms to use and on quantification methods are currently in progress regarding the training-load concept [2,7,16]. Therefore, we suggest that equalization of training doses should be a methodological prerequisite before comparing the effects of different training protocols and is therefore a major challenge facing exercise physiologists and sport scientists.

HIIT may be defined as repeated short-to-long exercise bouts performed at an intensity between 80% and 120% of maximum aerobic power (oxygen consumption or equivalent heart rate) [11]. Recently, the use of HIIT has been proposed as a method for improving quality of life of older people and for rehabilitation of patients suffering from several pathologies, such as cardiovascular diseases. As HIIT has become a real worldwide trend for exercise practice and sport sciences, this has increased the need for accurate equalization of training protocol doses in order to compare their efficiency [11,17,18]. Furthermore, we propose that HIIT studies display most of the characteristics necessary to understand the issues of exercise dose quantification and protocol equalization: (i) high number of studies published; (ii) changes in exercise variables; (iii) methods for equalization already developed and discussed.

The present scoping review aims to identify in the literature the evidence for protocol equalization and the soundness of methods used for it [19].

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

The latest methodological guidance for scoping reviews was followed, leading to completing the checklist of the Preferred Reporting Items for Systematic Reviews for scoping reviews (Supplementary File S1) [20–22].

#### *2.1. Search Strategy*

We analyzed published studies on electronic databases until 30 November 2020 without restriction set on the publication year. PubMed and SCOPUS databases were explored using a keyword search strategy for 'High-intensity interval training' with a first filter step used for including studies that were: written in English; randomized controlled trials, clinical trials or from journal articles; based on human subjects. A second step was based on abstract screening to select studies comparing HIIT to another type of training program and to retain only chronic training programs. When the information was missing in the abstract, the authors searched for it in the whole article. Because variables measured to control exercise do not correspond between sprint interval training and HIIT, the last step consisted of retaining studies focusing on HIIT (80–120% of VO2max or equivalent) and excluding sprint interval training (intensity higher than 120%) [11]. All duplicate studies and protocols were excluded; if the same experimental protocol was used for several articles, only the first published was retained. Finally, studies were sorted according to publication year and type of subjects observed: (i) patients or older people; (ii) untrained; (iii) trained. All search results were extracted and imported into a reference manager (Zotero, version 5.0.96.3). No included studies were authored by any of the review authors, thereby limiting possible conflicts of interest.

#### *2.2. Assessment of Reporting Quality*

The reporting quality of studies was assessed using items specific to the research field. Most of them originated from a modified version of the Downs and Black checklist, resulting in eight assessment criteria (Supplementary File S2) [23]. Studies reporting quality were scored on a scale from '0' (unable to determine, or no) to '1' (yes) for each item. Scores were allocated on the basis of good (6–8), moderate (3–5) and poor (0–2) methodological reporting quality.

#### *2.3. Terms Used and Methods Applied for Protocols Equalization*

Articles' methods sections were analyzed, and the proportion of studies that equalized doses of training protocols and methods used for equalizing were recorded. If any information necessary for protocol equalization was not included in a study's methods section, they were considered as not equalized. In line with this search, articles were analyzed to determine terms used to describe how exercise-induced effort was quantified (e.g., exercise dose or exercise volume) and the equalization process (e.g., equated protocols or matched training).

To assess the soundness of the methods used for protocol equalization, the exercise details were extracted from the articles, specifically exercise volume (duration, distance or number of repetitions for session and for each interval or set), intensity (varied metrics in absolute or relative values), recovery (duration and intensity if necessary) and exercise type (running, walking, cycling or resistance training).

#### *2.4. Statistical Analysis*

The present study is largely descriptive, and quantifies proportions (%) of studies that equalized training protocols and identified methods used for equalization. Differences in reporting-quality methodology between studies that equalized protocols and those that did not, and between subject populations were assessed by using one-way analysis of variance for total score and Pearson's Chi-2 test for each criterion assessed. The evolution of dose equalization over the years was observed by linear regression analysis of percentage of studies equalizing doses. Statistical analysis was performed with R software (version 3.6.2), and statistical significance was set at *p* < 0.05.

#### **3. Results**

The identification process described in Figure 1 resulted in 169 studies being included in the review. The complete list of articles retained is presented in the Supplementary File S3.

We aimed first to document the equalization of exercise protocols in studies comparing HIIT and other exercise types. We also aimed to highlight if protocol equalization was associated with a better-quality study design and/or if it was specific to recent studies.

The assessment of methodological reporting quality of these articles was moderate, but with poor quality for calculations of statistical power, and moderate for group homogeneity and for groups matched by physical condition (Table 1). Matching by "subjects' physical condition" was the only criteria that led to a significant difference between types of subjects observed by studies (*p* < 0.001)

**Score**

Equalized proto-

Non-equalized

Older people and

**Figure 1.** Phases of study selection during data collection. **Figure 1.** Phases of study selection during data collection.

We aimed first to document the equalization of exercise protocols in studies comparing HIIT and other exercise types. We also aimed to highlight if protocol equalization was associated with a better-quality study design and/or if it was specific to recent studies. **Table 1.** Reporting quality expressed through positive assessment of studies according to protocol equalization processes (middle of table) and population observed (bottom of table). Total score expressed as mean and standard deviations.


**in Same Ramdomiza-Condition direct Su-Control** \* significant differences with other groups of subjects (*p* < 0.05).

**Matching**

**tion**

**Population (%) (%) (%) (%) (%) ing (%) (%) (%)** Total (*n* = 169) 5.1 ± 1.5 47.6 70.8 53.0 73.8 88.7 61.3 84.5 29.8 cols (*n* <sup>=</sup> 98) 5.2 ± 1.5 46.4 71.1 51.5 75.3 91.7 61.9 89.7 28.9 protocols (*n* <sup>=</sup> 71) 5.0 ± 1.6 49.3 70.4 54.9 71.8 84.5 60.5 77.4 31.0 patients (*n* <sup>=</sup> 99) 5.1 ± 1.7 50.5 67.0 44.4 76.7 85.8 62.6 85.8 29.3 Untrained (*n* = 41) 5.3 ± 1.3 52.5 85.0 47.5 75.0 92.5 57.5 87.5 32.5 The most-frequently occurring terms used for the process of protocol equalization (total *n* = 98) were as follows: matched protocols (*n* = 44); equalized (or equated, equal, equivalent, *n* = 10); isocaloric (or isoenergetic, *n* = 8); The most-frequently used terms to designate what had been equalized were: total work (or external, mechanical, *n* = 26); workload (or training load, *n* = 29); exercise volume (or total volume, *n* = 13); exercise dose (or effort, *n* = 4). Protocol equalization did not evolve clearly over time, but there was a trend for a reduction in the proportion of equalizing studies (R<sup>2</sup> = 0.21, *p* = 0.06; Figure 2) and an increase in the absolute number of studies that equalized protocols (R <sup>2</sup> = 0.59, *p* = 0.01). No differences were observed in studies' reporting quality between those that equalized protocol doses and those that did not (*p* = 0.1).

**to Train-**

**Up**

**Power**

**pervision**

Trained (*n* = 29) 5.0 ± 1.2 34.5 62.1 89.6 \* 62.1 93.1 62.1 75.9 24.1 \* significant differences with other groups of subjects (*p* < 0.05). The most-frequently occurring terms used for the process of protocol equalization (total *n* = 98) were as follows: matched protocols (*n* = 44); equalized (or equated, equal, equivalent, *n* = 10); isocaloric (or isoenergetic, *n* = 8); The most-frequently used terms to The distribution of studies based on equalized and non-equalized protocols, and associated methods for quantifying exercise doses are shown in Figure 3. Studies observing patients and older people equalized protocols at 58.7%, compared to 62.5% in untrained subjects and 51.7% in trained, without significant differences between groups (*p* = 0.09). Training protocols differed between studies; however, typical HIIT exercises were identified among all studies (Figure 3).

designate what had been equalized were: total work (or external, mechanical, *n* = 26); workload (or training load, *n* = 29); exercise volume (or total volume, *n* = 13); exercise dose (or effort, *n* = 4). Protocol equalization did not evolve clearly over time, but there was a trend for a reduction in the proportion of equalizing studies (R2 = 0.21, *p* = 0.06; Figure 2)

ized protocol doses and those that did not (*p* = 0.1).

**Figure 2.** Percentage and absolute number of studies using equalized protocols (dashed blue line and black bars, respectively) and number of studies without using equalization of protocols (grey bars) from 1979 to November 2020. **Figure 2.** Percentage and absolute number of studies using equalized protocols (dashed blue line and black bars, respectively) and number of studies without using equalization of protocols (grey bars) from 1979 to November 2020. *Int. J. Environ. Res. Public Health* **2022**, *19*, x FOR PEER REVIEW 6 of 11

**Figure 3.** From article-selection process to equalization methods and exercise sessions. achieved anyway, thereby increasing the proportion of protocols actually equalized. **Figure 3.** From article-selection process to equalization methods and exercise sessions.

a substantial number of studies did not. For designating what was equalized, authors mainly focused on actual measures performed (e.g., total work, energy expenditure or exercise volume) rather than using a more generic term (e.g., exercise dose, training load or effort). Energy-based methods were prominently used for equalizing protocols, whereas methods based on exercise volume and perceived exertion appeared markedly

Among the 169 studies included in this review paper, most equalized their protocols (58%), whilst 42% did not. Consensus for protocol equalization is not apparent, and the protocol equalization rate has not evolved significantly since the first paper published in 1979. In addition, data did not show differences according to populations observed. This is in line with the assessment of reporting quality, which did not differentiate studies according to protocol equalization or populations observed. Satisfactorily, "exercise control" and "direct supervision" criteria of reporting quality achieved the highest assessments. Among studies that did not equalize protocols, twenty-one compared HIIT with typical MICT programs (Figure 3) that had been designated by previous studies to be equal based on energy expenditure or production [24]. Therefore, although protocol equalization was not reported in the methods section of these studies, it had possibly been

**4. Discussion**

less frequently.

#### **4. Discussion**

We aimed to determine whether researchers, when comparing HIIT to other types of programs, had utilized equalized protocols. Although most studies equalized protocols, a substantial number of studies did not. For designating what was equalized, authors mainly focused on actual measures performed (e.g., total work, energy expenditure or exercise volume) rather than using a more generic term (e.g., exercise dose, training load or effort). Energy-based methods were prominently used for equalizing protocols, whereas methods based on exercise volume and perceived exertion appeared markedly less frequently.

Among the 169 studies included in this review paper, most equalized their protocols (58%), whilst 42% did not. Consensus for protocol equalization is not apparent, and the protocol equalization rate has not evolved significantly since the first paper published in 1979. In addition, data did not show differences according to populations observed. This is in line with the assessment of reporting quality, which did not differentiate studies according to protocol equalization or populations observed. Satisfactorily, "exercise control" and "direct supervision" criteria of reporting quality achieved the highest assessments. Among studies that did not equalize protocols, twenty-one compared HIIT with typical MICT programs (Figure 3) that had been designated by previous studies to be equal based on energy expenditure or production [24]. Therefore, although protocol equalization was not reported in the methods section of these studies, it had possibly been achieved anyway, thereby increasing the proportion of protocols actually equalized.

Vollard and Metcalf [13] argued that the key advantage of HIIT is time efficiency. MICT requires more prolonged exercise duration than HIIT, and it could be presumed as self-evident that MICT is not as effective if exercise duration is short. However, for a given exercise duration, because of higher intensity, HIIT induces a greater exercise dose than MICT. If the aim is to demonstrate the positive effects of HIIT despite a short exercise duration, such demonstration could be achieved without requiring comparison with another training program. Conversely, when comparing programs' effects on performance improvement or biological parameters, if the higher exercise intensity of HIIT is not counterbalanced by a lower exercise volume, responses may have originated from the higher intensity, but also simply from a greater exercise dose. This methodological point was accounted for by 98 studies that attempted to equalize protocols.

In some studies, training protocols were partly equalized by prescribing similar total exercise durations. Such a method is in line with population-based studies that quantify physical activity through time spent in light/moderate/strenuous intensity ranges without aiming to compare the particular effects of these intensity levels [1]. Using session durations to equalize protocols corresponded to physical activity recommendations for health and wellbeing (e.g., three sessions of 30–45 min per week for HIIT and moderate intensities) [1]. During HIIT, high-intensity activity itself could not account for the entire 30–45 min of the session: 10–20 min of high-intensity exercise was paired with low-intensity exercise for the remaining 10–20 min. Therefore, protocols equalized by similar durations compared MICT to mixed MICT and HIIT, but studies did not describe the rationale underpinning the selection of exercise durations for different intensities. Equalization by total volume does not consider the slope of the relationship between intensity and duration, and even less the nonlinearity of this relationship. Consequently, the absolute value of exercise duration was equalized, but not the combination of the exercise variables. If expressed relative to respective maximums, durations prescribed by HIIT programs were markedly higher than for MICT. In these studies, responses to training might be due to changes in intensity or to changes in exercise dose. Furthermore, by proposing similar exercise durations, these protocols cancelled the time gains expected from HIIT [13].

The primary methods used for protocol equalization were energy-based. Most studies measured exercise-induced energy expenditure through oxygen consumption, while some others measured external work based on power output and exercise duration [12,24,25]. Energy expenditure methods typically incorporated both exercise and recovery periods, while methods based on external work only considered exercise bouts. That is quite

surprising as the typical HIIT exercise utilized in studies based on external work (i.e., 8–10 × 1 min at 90–95% HRmax, 1–2 min recovery) were characterized by short–moderate recovery pauses, allowing maintenance of a high level of physiological stress [26]. Furthermore, exercise-induced excess post-oxygen consumption is largely influenced by exercise intensity and may be prolonged for many hours [27]. Although some authors suggest that exerciseinduced energy expenditure should also account for exercise-induced excess post-oxygen consumption, this point may require more careful attention in HIIT studies that focus on the effects of changes in both intensity and interval volumes [1,12].

Energy-based methods for quantifying exercise consider the human ability for energy expenditure or external work to be similar whatever the exercise intensity. For several decades, models of the intensity–volume relationship have described a hyperbolic pattern, with maximal exercise volumes dramatically decreasing with increases in intensity [28–30]. By extension, maximal energy expenditure/external work follows the same pattern [31]. Thanks to recovery pauses, for a given intensity level, interval exercise allows accumulation of more exercise than continuous exercise and, consequently, greater energy expenditure [26]. The typical 4 × 4 min session is likely to be performed at a higher intensity level than a 16 min exercise performed in continuous modality [30]. Seiler et al. reported that the maximal tolerable intensity for 4 × 4 min was 94 ± 2% of maximal heart rate when interspersed with 2 min passive recovery [32]; in HIIT studies, an active recovery (3 min at 70% HRmax) was added to this maximal effort. Conversely, because of the nonlinear relationship between exercise intensity and energy expenditure, typical MICT exercise appears to be far from the exercise dose performed during typical HIIT. In fact, in typical MICT, 30–45 min is prescribed at 65–75% HRmax, an intensity that can be maintained for several hours before exhaustion. It may be assumed that the typical HIIT exercise resulting from the energy-based equalizing method reached a maximum of energy expenditure and was exhausting, while MICT represented relatively easy training. This assumption is supported by significantly higher ratings of perceived exertion (RPE) following HIIT sessions [25,33,34], and some authors argued that energy-based methods for equalization underestimate the work that athletes are able to perform at lower intensities [32,35]. Such differences in session-induced exertion should be considered as a possible methodological bias that is likely to become more pronounced with increases in intensity differences between programs. HIIT-induced dose could represent the maximum tolerable (or excessive) training stimulus, whereas MICT dose could be low or insufficient. Finally, despite the popularity of equalization methods based on energy expenditure, its soundness and relevancy are still questioned [13,15].

Finally, six studies used RPE to equalize protocols, and only one used the session-RPEbased method for training-load quantification (i.e., duration × RPE of the session). It seems that studies equalizing protocols by using RPE were composed of varied exercise modalities (e.g., running, resistance exercise or skating) [36–38]. RPE is not only influenced by exercise intensity [39], as exercise duration [40,41], interval volume [42], exercise modality [43] and recovery periods [44] have also been reported to significantly influence RPE. Finally, RPE appears to be influenced by all exercise variables and, consequently, might represent a subjective assessment of the exercise dose. Previous studies have shown that it provides similar session assessments to exercise volume expressed relative to maximum for the considered intensity level [4,40]. Conversely, training load based on RPE might account twice for the exercise volume (i.e., in duration and RPE itself), inflating the calculated load for prolonged sessions [4,5]. In line with studies that have used RPE for protocol equalization, some authors have suggested that RPE alone is therefore preferable for exercise quantification, thereby avoiding the overexpression of volume [4,45].

We acknowledge that the present study may have overlooked some published papers, as it was only conducted on two literature databases and only considered original experimental investigations. Based on the numerous studies utilizing equalization of protocols and researcher support for equalization, it seems that, although the need for equalization is not debated per se, the soundness of methods for equalizing is [1,5,7,12]. In addition, generic terms that designate the quantity of exercise-induced effort (i.e., exercise dose, internal training load) and associated quantification methods (e.g., RPE) may be considered to account for individual maximal capacities in the exercise considered [5–8]. In essence, this is not the case for energy expenditure/production or exercise volume. Finally, the main methodological issue is whether to quantify the exercise—whatever the method—in absolute values or relative to individual maximums for the considered exercise. Lack of consideration of the slope and nonlinearity of the energy–duration or of the intensity–duration relationship is questionable. As proposed recently for training-load quantification and by one study among the 169 retained [4,46], we hypothesize that exercise quantified relative to maximum energy expenditure/external work, or exercise volume, for specified intensity levels will allow more precise program comparisons. This may also be the case when dose is assessed via perceived exertion.

Although scoping reviews can be the first step before systematic review or metaanalysis on the topic, and even if only equalized protocols were retained, results of studies comparing HIIT vs. MICT should be interpreted carefully because of the uncertain accuracy of equalization methods mainly used [19].

#### **5. Conclusions**

In HIIT studies, no clear consensus for protocol equalization appears to exist, and there has been no evolution in practices over time. If the scientific community supports this methodological prerequisite, it may assist with the assessment of methodology reporting quality.

Equalization based on exercise duration does not consider all the variables composing exercise-induced effort. Primary equalization methods consider energy expenditure/external work in raw values without considering the slope and the nonlinear nature of its relationship with duration. Exercises resulting from these quantification methods induced maximal exertion in HIIT exercises but low exertion in MICT. Evidently, the main issue is whether to consider exercise dose in absolute values or relative to individual exercise maximums. It is hypothesized that comparison of program effects would be more accurate if the exercise (e.g., exercise volume, energy) was expressed relative to intensity-related maximums (e.g., perceived exertion, exercise volume relative to maximum).

**Supplementary Materials:** The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/ijerph19094980/s1, Supplementary File S1: PRISMA checklist for scoping reviews. Supplementary File S2: Checklist for the assessment of the methodological quality for HIIT studies, adapted from Downs and Black (1998). Supplementary File S3: Complete list of articles retained according to dose equalization methods.

**Author Contributions:** T.N.-G. and F.-D.D. conceived the research question and designed the study protocol. T.N.-G. conducted the initial database search and screening and synthesized and analyzed relevant data. T.N.-G. and F.-D.D. conducted the full-text screening and drafted the paper. F.-D.D., T.L. and F.B. conducted the study quality assessment and data extraction. T.L., F.B., A.R. and J.-F.T. contributed throughout the review, starting from conceptualizing to editing subsequent drafts of the 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:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

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

#### **References**


## *Review* **Sports Performance Tests for Amputee Football Players: A Scoping Review**

**Agnieszka Magdalena Nowak \* , Jolanta Marszalek and Bartosz Molik**

Faculty of Rehabilitation, Jozef Pilsudski University of Physical Education in Warsaw, 00-968 Warsaw, Poland; jolanta.marszalek@awf.edu.pl (J.M.); bartosz.molik@awf.edu.pl (B.M.)

**\*** Correspondence: agnieszka.nowak@awf.edu.pl

**Abstract:** Background: This scoping review aims to identify sports performance tests for amputee football players and to critically analyze the methodological quality, validation data, reliability, and standardization of sport-specific tests to indicate the best-fitting tests. Methods: Electronic database searches were conducted between January 2019 and October 2021. Twelve articles met the inclusion criteria. Qualitative assessment of each study was conducted by STROBE checklist. Results: Twentynine sports performance tests were identified. No sports performance test fully met all three criteria associated with the qualitative assessment of tests. The critical appraisal of the articles demonstrates a gap in study design, settings, and main results description. Some inconsistencies were found in the methodological descriptions of tests assessing the same motor skill. A STROBE score of 13 points was considered a satisfactory score for the article (it was obtained by 8 of the 12 studies). The weakest point of the analyzed studies was the description of how the test group size was accessed and later obtained. Conclusions: No test was found that was simultaneously presented as valid, reliable, and standardized. The authors can recommend the use of the two-sports performance tests that are the closest to ideal: the L test and the YYIRT1.

**Citation:** Nowak, A.M.; Marszalek, J.; Molik, B. Sports Performance Tests for Amputee Football Players: A Scoping Review. *Int. J. Environ. Res. Public Health* **2022**, *19*, 4386. https:// doi.org/10.3390/ijerph19074386

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

Received: 21 February 2022 Accepted: 5 April 2022 Published: 6 April 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/).

**Keywords:** field-based tests; amputee soccer; assessment; disability; impairment; athletes; adapted sport

#### **1. Introduction**

Amputee football (amputee soccer; AF) is an impairment-specific football for people with an amputation or limb deficiency (US Soccer Federation). The major part of AF rules is based on regular soccer rules, while the few paragraphs consider the physical impairment of players [1]. Accordingly, two halves are being played (2 × 25 min) on a smaller pitch (from 60 × 30 to 70 × 55 m) by seven players (six field players, one goalkeeper). Singleleg amputees (either above or below the knee) play without prosthesis on aluminum wrist crutches (field players). Goalkeepers must be single-arm amputees [2]. AF is still developing and has become a point of interest for many researchers since it is a non-Paralympic sport discipline that is applying to enter the Paralympic Games. AF has become greatly popular as a recreational and elite sport. It is also recommended as a continuation of the rehabilitation process for amputees to improve the level of functional fitness, as well as a form of physical activity that allows people to realize themselves as athletes. What is more, AF has a positive impact on body composition and quality of life, and it gives a sense of belonging to society [3–5].

It is assumed that AF is classified as a high-intermittent sport with periods of highintensity activity [6,7]. AF requires from its players a high level of many physical attributes, such as power, speed, strength, balance, agility as well as endurance [8,9]. Short bursts of high-intensity power production and aerobic capacity play a major role in AF performance [2,9]. Some studies have confirmed this, indicating that athletes spend the majority of their playing time in a heart rate zone above 80% of their maximum heart rate

(HRmax) [6,7]. Given the high intensity of the game, it is important to emphasize that AF players should be in excellent condition to easily perform the entire spectrum of activities with and without the ball and while moving on crutches [9]. Studies underline the fact that using crutches is quite exhausting for AF players [10]. Therefore, it can be assumed that players should not only be well prepared technically and tactically but also, most importantly, physically for the game, as is the case with able-bodied soccer players [11]. Coaches should be obliged to evaluate the motor performance of players to notice progress or weaknesses in the training process.

In the literature, many different sports performance tests have been reported [2,3,9,12–20]. Moreover, by reviewing the literature and observing the various nomenclature of motor abilities used and the different descriptions of the same tests, we decided to organize the sports performance tests for assessing the motor performance of amputee football players to make them transparent and understandable for researchers in the field, coaches, and people interested in this type of sport. Considering how important the periodic assessment of athletes' motor performance is to both sport-specific and non-sport-specific tests related to AF, the fundamental aim of this study is to identify sports performance tests for amputee football players in a literature review and to critically analyze the methodological quality, validation data, reliability, and standardization of sport-specific tests to indicate the best-fitting tests. Furthermore, the quality of the reviewed articles is checked to indicate the quality of the studies' descriptions.

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

#### *2.1. Search Strategy, Study Selection, and Data Extraction*

Reporting of this scoping review was guided by the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) statement standards. The review protocol was registered with PROSPERO (CRD42021286911), and the review itself was conducted in January–October 2019 with no restrictions on the date of study publication. It was then regularly updated until November 2021. Electronic databases (EBSCO (SPORTDiscus with Full Text, Academic Search Ultimate, Teacher Reference Center, Health Source: Nursing/Academic Edition, MasterFILE Premier), Web of Science, and PubMed (Medline)) were searched. Database settings were customized for each database (option to search all fields, scientific journals, peer-reviewed articles). The keywords used in the search were divided into three groups: amputee OR amputation AND physical AND soccer OR football and were conducted by the Boolean AND/OR. More specific keywords were not needed due to the small number of publications in the field. The keyword combinations were used according to the databases' capabilities and were presented in an online repository.

In the examination process, the title and the abstract were first checked for compatibility with at least one keyword. If an article met the inclusion criteria, it was carefully selected for this review by making sure that it: was available in an online database in full text (1), was written in English (2), was an original study (cohort, case–control, cross-sectional) (3), involved amputee football players (4), and used sports performance tests as research tools (5). The criteria according to which an article could not be included in the examination were as follows: no keyword in the title and/or abstract, papers of other types (reviews, case reports, conference reports, chapters in books), written in a language other than English, not related to amputee football players, and did not include sports performance tests. We used Microsoft Excel to collect the data and uploaded them to an online repository. The PRISMA flowchart was used to describe the review process (Figure 1). Two researchers (A.M.N., J.M.) independently conducted the process.

#### *2.2. Studies Description*

First, the studies included in the review are described in a table pointing out the type of research conducted, the purpose of the study, and the characteristics of the study group. A summary description of the included studies is presented in Table 1.

**Figure 1.** Study selection flow diagram.

#### *2.3. Sports Performance Tests Description*

The sports performance tests identified in the literature were analyzed in terms of the type of test and the entire procedure for conducting the test, including athlete preparation, warm-up, how to do the test, number of repetitions, intervals between repetitions of the test or between tests, and variables that are test results. The methodology of the identified sports performance tests is described in Table 2.


**Table 1.** General description of included studies (*n* = 12; studies arranged in chronological order).

AB—able-bodied; AF—amputee football players; AMP—individuals with amputation; C—cohort; CC—case– control; CG—control group; CS—cross-sectional; CS-C—cross-sectional control; mths—months; ND—no data; SG—study group; yrs.—years.


**Table 2.** A detailed description of the sports performance tests in included studies.






17




18

c-RPE—central rate of perceived exertion; p-RPE—peripheral rate of perceived exertion; SJ—squat jump; T10, T20, T30—10, 20, 30 m sprint test; VE—ventilation equivalent; WR—work

rate; YYIRT1—the Yo-Yo intermittent recovery test—level 1.

#### *2.4. Sports Performance Tests' Quality Assessment*

In this phase, we divided the sports performance tests according to their characteristics (motor abilities), which assess balance, aerobic capacity, strength, endurance, power, sprint performance, agility, and flexibility. Two researchers (A.M.N., J.M.) independently assessed all the papers and then consulted the results among themselves.

All found sports performance tests were analyzed for reliability, validity, and standardization based on the authors' descriptions in the methods section of the articles. Test reliability and validity were recognized based on information about the reliability and validity of the test in the study and whether test references or expert validity were used. Expert validity implies that the researcher, based on their knowledge and experience, selected a sports performance test to assess specific motor abilities, e.g., the 30 m sprint test was used to assess sprint performance. Standardization means that the researchers have written down all the information necessary to repeat the test (participant preparation, environment, methodology, number of repetitions, intervals, outcomes). A description of this assessment is provided below, and points were allocated for each parameter:


Table 3 presents the qualitative assessment of the sports performance tests identified through the literature review process.


**Table 3.** Quality assessment of sports performance tests in the review (*n* = 12).



V—valid; R—reliable; S—standardization; CMJ—countermovement jump; CPX—cardiopulmonary exercise test; F8W—figure-of-8 walk; MBT—medicine ball throw; PUT—push up test; SJ—squat jump; T10, T20, T30—10, 20, 30 m sprint test; YYIRT1—the Yo-Yo intermittent recovery test—level 1; "1"—presence of validity, reliability, standardization; "0"—absence of validity, reliability, standardization; \*—test performed with prosthesis; <sup>1</sup>—sprint refers to tests in which movement is as fast as possible in one line, movement speed refers to tests in which movement is as fast as possible with changing directions.

#### *2.5. Studies' Quality Assessment*

The studies included in the review were qualitatively assessed to highlight the value of the papers in terms of their methodological design. To accomplish this, the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement was used, which was created to improve the quality of reported observational studies, and such studies were included in our study. The STROBE statement allows the strengths and weaknesses of the observational studies to be identified and provides an opportunity to generalize the results of the report [21]. The STROBE checklist consists of a checklist of 22 items that relate to the title and abstract (1 item), introduction (2 items), methods (9 items), results (5 items), discussion section (4 items), and other information (1 item) in the articles. One point for each item in the paper was given [22]. Some of these items originally had sub-items. In that case, one point was awarded for more positive responses. The outcome was the score obtained when consensus was reached (A.M.N., J.M.). Discrepancies were resolved by consensus with a third researcher (B.M.).

#### **3. Results**

Twenty-nine sports performance tests were found in the 12 included studies to assess AF players. They assessed motor abilities such as balance, anaerobic performance (strength, power, sprint performance), aerobic performance (capacity), flexibility, and agility (speed performance) (see Table 3). Despite measuring the same motor ability, the identified tests had different methodologies. For example, the jump test was performed once with and once without a prosthesis, and, in the second case, there was no information about it.

Through 29 sports performance tests, no test met all three quality assessment criteria. In eight cases, participants performed tests with a prosthesis, as marked with an asterisk and presented in Table 3.

#### *3.1. Qualitative Assessment of Sports Performance Tests*

#### 3.1.1. Reliability

Five out of twenty-nine tests had confirmed reliability in the cited publications (handgrip test, CMJ, L test, YYIRT1, and Berg Balance Scale [2,3,16,19]), and the L test was tested for reliability among amputees.

Two tests, despite the references provided (isometric test of back extensors and trunk flexor test), were not described as reliable or used in the cited publications [3].

#### 3.1.2. Validity

In total, 18 sports performance tests were considered valid based on expert validity and 11 on literature reference; 4 of the 11 cited books were not available.

#### 3.1.3. Standardization

Although all the identified tests had a description of the procedure, only 28% of them met the standardization criteria. Sports performance tests that had complete instructions (subject preparation, environment, methodology, number of repetition, intervals, outcomes) were the T-square test, modified Thomas test, sit-and-reach test, vertical jump test, static balance test, and dynamic balance test [3,9,12,16,20]; 8 tests lacked information about participants' preparation, 9 tests lacked information about the warm-up, 16 tests lacked information about the number of test repetitions, and 6 tests lacked information about intervals between test attempts. The qualitative assessment of sports performance tests is presented in Table 3.

#### *3.2. Qualitative Assessment of Articles*

In this scoping review, we included observational studies available in the field of amputee football (5 case–control studies, 6 cohort studies, and 1 cross-sectional study). In total, 10 out of 12 articles met eligibility criteria and were from the past 10 years; 50% of the studies had a study group and a control group. Participants were AF players aged 24–32 years from local [2,9,12,15] or national teams [3,16–20]. Training experience ranged from two months to eight or more years. Two studies did not provide information on players' training experience [12,14] (see Table 1 for details).

The qualitative assessment of the studies resulted in STROBE scores ranging from 5 to 17 (mean 12.9 points; 65%). Two studies had the highest score of 17 points [17,18], while two different studies had the lowest possible score [14,19]. Six of the twelve studies had an appropriately constructed abstract and title, with two studies indicating the study design in the title or abstract and four studies indicating the study design in the methods section. All studies stated specific objectives, and 11 of 12 studies sufficiently explained the background of the study. In most cases, the participant description was correct. Simim et al. (2018) obtained the highest and the maximum score in the methods section. Providing information on how the study size was obtained in the methods section was the weakest aspect of the evaluation (only 2 of 12 authors reported this data [12,18]). In the results section, two articles met the requirements of item 13 (participants), eight articles met the requirements of item 14 (descriptive data), six articles met the requirements of item 15 (outcome data), three articles met the requirements of item 16 (main results), and six articles met requirements of item 17 (other analysis). In summary, in four studies, the key results concerning the study objectives were presented in the discussion section [3,14,15,20]. The items on limitations, interpretation, and generalizability were met by most of the included studies. Four studies provided information on the source of funding (item 22). The qualitative assessment of the included studies is presented in Table 4.


**Table 4.** The STROBE qualitative assessment of included studies.

20—interpretations; 21—generalizability; 22—funding.

#### **4. Discussion**

The purpose of this scoping review was to identify sports performance tests for amputee football (AF) players in the scientific papers and to critically analyze these tests for reliability (i), validity (ii), and standardization (iii) to indicate the best-fitting tests. Along this line, 29 sports performance tests used in AF were found in the current literature (12 studies). We found no sports performance test that would fully meet all three criteria associated with a qualitative assessment of sports performance tests.

When discussing the first parameter (i), the authors of the included studies did not conduct a test reliability examination. The reliability of five tests (YYIRT1, L test, handgrip test, CMJ, and Berg Balance Scale) has been confirmed by the authors of the included studies based on the references [2,3,16,19]. The reliability of only one test, the L test, was verified on amputees, which is an advantage of the reported study [3,23] compared to other tests in which reliability was verified on able-bodied individuals. The authors of the analyzed studies used reliable tools to assess muscle strength [19], lower limb power [17], aerobic capacity [16], and balance [2,3].

The PUT, the isometric back extension test, and the isometric trunk flexion test had inappropriate references to prove the reliability of these tests because the works cited were off-topic [3,17,18]. Consequently, we suggest that researchers and coaches pay attention to the reliability of sports performance tests applied to their groups of athletes.

In terms of validity (ii), from one point of view, the indispensable information was obtained in seven sports performance tests, which included the static balance one-leg test, dynamic balance test, handgrip test, L test, F8W test, YYIRT1, and Berg Balance Scale [2,3,16,19]. Whereas, in the case of four tests, such as the modified Thomas test, the sitand-reach test, the vertical jump test by the Lewis formula, and the isometric back extension test, we could not approve their validity due to the inability to find the reference cited by the authors [3]. Additionally, about the PUT, it was performed differently than reported in the original paper [24]. Consecutively, it also remains unknown whether the presented PUT is truly valid [16]. Moreover, in articles that used static (Kistler force platform) and dynamic balance tests, isokinetic trunk strength tests, PUT, isotonic sit-ups tests, isometric back extension and trunk flections tests, CMJ and SJ (force plate Sport Expert TM), MBT, CPX two-armed exercise tests, modified Thomas tests, sit-and-reach tests, T-square, and sprint tests, there was no information on validity and reliability verification [2,3,9,12–14,16–19]. It is probably the case that the authors of included studies, when selecting tests to assess the motor abilities of AF players, verified these tests based on their experience and general knowledge (e.g., sprint tests to assess speed or sprint performance); therefore, we decided to give them one point as an expert validation.

It must be admitted that in most sports performance tests, the standardization (iii) was clearly explained. Information regarding the starting and finishing positions, the number of repetitions and break times, and the type of movement (running, walking with or without prosthesis) was adequately introduced. This renders them easily repeatable and, thus, helpful for both researchers and coaches. When analyzed in detail, 8 of the 29 test descriptions met all standardization criteria (T-square test, modified Thomas test, sit-andreach test, vertical jump test, static balance test, and dynamic balance test [3,9,12,16,20]). For a test such as the YYIRT1, the only information about the number of repetitions of the test performed was missing, but we believe that this information is not necessary in this case because this type of aerobic capacity test is usually performed only once due to the maximal stimulation of the aerobic system, after which a long recovery is necessary [25]. The lack of descriptions regarding the warm-up and intervals between repetitions in sprint tests [9,14,19], in which a maximal effort is performed, deserves significant criticism since all these elements are crucial in the assessment of anaerobic performance. In the case of balance tests, information about the use of a familiarization session is important in the context of repeating and comparing the test in the future, and the question of whether and how this session affects test performance (learning process) and the final result is still unknown [26].

On the other hand, some of the tests might be misleading, e.g., the PUT, sprint tests, CMJ (by Myotest), MBT, and CPX two-armed exercise tests, in which it was not explained why and how the procedures were followed and how they were adapted for amputees [9,13,16–18]. The PUT did not have information about the position and the type of movement included in the description, as well as whether a prosthesis was used during this test and other tests [13,17,18]. Because of these confusions, we expected that performing the MBT in a seated position with or without a prosthesis might influence the stability of the trunk position, and, consequently, the final results might be different (athlete sits close to the wall vs. athlete performs a full backward and forward movement to complete the task). In some locomotion tests, participants used a prosthesis (L test and F8W test), while in others, they performed the tests on crutches without a prosthesis (T-square). At this point, it is worth asking ourselves under which conditions we want to evaluate the AF players, as it must be remembered that the athlete is moving on crutches during the match. The same dilemma regarding the use of a prosthesis or not has been noted in vertical jump tests [3,9] and balance tests [2,12]. Consequently, the reader does not know if these tests were performed in a single-leg standing position or if the athletes had three or four points of support. Moreover, we noted several discrepancies concerning the start of the tests. For the T10, T20, and T30 procedures [9], there was no information about the starting position or whether the starting signal was given by the researcher or whether the athlete decided to start the test. Then, in the MBT, it was not clear where the starting point was for measurement. Without such information, it is difficult to compare the results obtained by different groups of participants and then repeat and compare the tests with each other. The differences in results are likely due to erroneous measurements rather than the athletes' skills, making the ratings unreliable. Therefore, it is recommended that in future papers, authors describe their tests accurately.

The studies included in this review have many limitations in the clarity of the names of the motor abilities assessed in sports performance tests because of various wording. In other words, three different groups of researchers used different terms to match tests to the physical attribute; for instance, T30 was used to assess anaerobic performance, sprint performance, or movement speed [9,16,19]. It would be clearer for readers to use only one term. Surprisingly, the L test and the F8W have been classified as sprint tests, together with the T10, T20, and T30, which are speed tests [3,14,16,19]. It becomes obvious that the sprint tests were performed as fast as possible in a straight line, while the L test and the F8W were performed with changes in direction, which may affect the change in running speed and is more to assess agility than speed. Moreover, the result of the L test and the F8W may consist of the route execution technique, which is unlikely for the sprint tests.

A similar observation was made for the vertical jump tests and the MBT. The latter has been used as a power test, a muscle test, a neuromuscular performance test, and an anaerobic performance test and has been positioned as a test focused on strength assessment [3,9,17,18]. Given these achievements, we suggest classifying the MBT as a power assessment because it is the same physical attribute that vertical jump tests assess. We believe that future manuscripts should pay more attention to the terms and expressions used in the sports performance tests and to the description of the physical attributes. Maintaining this level of vocabulary clarity will be beneficial to both coaches and athletes in understanding which motor abilities are being tested in each sports performance test.

The articles included in this review had large discrepancies in scoring in the qualitative assessment. The authors of the current study believe that the methods and results sections of the included studies need the most correction and attention. First, providing the study design in the abstract and/or methods section is important because it gives the reader an understanding of what type of research they will be dealing with. Most studies correctly described the participants. The reader can read about: eligibility criteria and how participants were selected, outcomes, exposures, predictors, potential confounders, and details of assessment methods. The above-mentioned description is important because it indicates whether the study group was homogeneous and whether there were confounding factors.

In this review, the authors of the included studies did not mention any possible confounding factors or description of the test location (whether the tests were performed in the same setting, such as a gym, laboratory, or outdoor soccer pitch). Different conditions and environments can affect the results: e.g., headwind, a slippery floor in sprint tests, and low temperatures can cause poorer results in sprint or flexibility tests. In addition, researchers and coaches should be cautious when interpreting their results concerning the already existing results of others, as there have been times when the results of the same test have depended on different variables. For example, in the PUT, the duration of the test or the number of repetitions performed within a specified time was evaluated; in the sprint tests, the time or speed of the distance covered was evaluated; in the jump tests (vertical jump, CMJ, SJ), the height of the jump or power was evaluated.

What is more, we were concerned about the lack of explanation of how the study group size was obtained (only two articles reported this [12,18]). This issue is particularly relevant when judging null results, which might indicate that there was no real difference between the study groups or that the power of the statistical analysis was too low to detect a real difference. It is worth noting that some studies on AF players included relatively few participants (6–33 people). In the result section, the items were quite complex, and a study had to meet most of the criteria for each item to receive one full point. If a sub-criterion did not apply to the study in some cases, we did not count it. It seems worrying that most articles do not state the key results at the beginning of the discussion section (item 18). Another important point to indicate is if the purpose of the study was achieved in order to lead the discussion section fluently.

Although the STROBE checklist was designed for observational studies, it is important to keep in mind when using this tool that not all criteria are mandatory for every subtype of study, e.g., cohort studies usually do not have any follow-ups or reduction in the number of participants because they have only one group and the study is conducted over one or two days. The STROBE statement is a particularly detailed tool; on one hand, it can help in the preparation of the manuscript, but, on the other hand, it can cause difficulties in the evaluation of the study due to its precision. Considering the presented conclusions and the fact that most of the studies were single-case studies and that we could not give a positive score for some criteria (not because there was an error in the article but because the criteria did not apply to the study), we judged that 60% (13 points) was a satisfactory score, and, thus, 8 of the 12 articles achieved it.

#### *Limitations and Perspectives*

This is the first review to bring together all the sports performance tests used in AF and organize them in detail in terms of motor abilities and test descriptions. The available literature lacks a "gold standard", a battery of sports performance tests, or a compilation of which tests are dedicated to AF players (sport-specific tests). Our study indicates that some tests, based on their standardization, may be suitable for assessing sports performance in AF, and coaches may use them in their practice. However, further research is needed to investigate the tests' validity and reliability and characterize them for AF players.

We understand that the literature search performed in this research field may be conducted differently in future studies. This manuscript presents a structured way of literature review (keywords, inclusion/exclusion criteria). Other authors may search the literature using different methodology and other guidelines for reporting the main types of studies, such as the STROBE guidelines that were used in our study. However, in our opinion, recommendations for future studies seeking sports performance tests in a specific sport should be structured as a research review and a presentation of the advantages and disadvantages of the tests and research, such as was done in this study (quality of paper, presence of validity and reliability of tests, and the completeness of description of selected tests). A well-planned research review and manuscript organization will be important for the next steps in AF development as a future Paralympic sport, considering the development of sports classification based on evidence (evidence-based classification system) [27]. The International Paralympic Committee (IPC) has outlined the steps in this process, and the identification of tests in this manuscript is relevant to step 2 (identifying key activities and determinants) and step 3 (identifying appropriate tests to assess key determinants) of the IPC classification process [28]. Authors of future research may consider this rationale and address the need for an evidence-based classification approach as a purpose of their work.

#### **5. Conclusions**

Our study constitutes a practical and detailed description of the sports performance tests identified in the literature and includes a qualitative assessment of sports performance tests and a qualitative evaluation of the included articles. The authors of the studies included in this review have verified the reliability and validity of sports performance tests based on results from others' studies. Considering the final conclusions of the reviewed studies and our evaluation of these studies, we conclude that none of the 29 tests from the 12 research papers included in this review were simultaneously reported as valid, reliable, and standardized. We found few tests for amputee football players, which were only partially verified for validity and reliability; thus, we recommend verifying those tests using, for instance, the test–retest method [29]. Despite the deficiencies in the test descriptions, we recommend using two sports performance tests: the L test and the YYIRT1, to assess agility and endurance, respectively.

**Author Contributions:** Conceptualization, A.M.N.; data curation, A.M.N. and J.M.; formal analysis A.M.N., B.M. and J.M.; investigation, A.M.N., B.M. and J.M.; methodology, A.M.N. and J.M.; project administration A.M.N. and J.M.; supervision, J.M. and B.M.; visualization, A.M.N., J.M. and B.M.; writing—original draft, A.M.N.; writing—review and editing, A.M.N., B.M. and J.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by the Ministry of Science and Higher Education in the years 2021 and 2022 under Research Group No. 4 at Jozef Pilsudski University of Physical Education in Warsaw ("Physical activity and sports for people with special needs").

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are openly available in FigShare at: https://doi.org/10.6084/m9.figshare.16850194.v1 (accessed on 15 February 2022) and https://doi.org/ 10.6084/m9.figshare.16850185.v1 (accessed on 15 February 2022).

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

#### **References**


### *Systematic Review* **Biomechanical Performance Factors in the Track and Field Sprint Start: A Systematic Review**

**Maria João Valamatos 1,2,3,4,\* , João M. Abrantes 1,3, Filomena Carnide 1,2,3, Maria-José Valamatos <sup>1</sup> and Cristina P. Monteiro 1,2,5**


**Abstract:** In athletics sprint events, the block start performance can be fundamental to the outcome of a race. This Systematic Review aims to identify biomechanical factors of critical importance to the block start and subsequent first two steps performance. A systematic search of relevant English-language articles was performed on three scientific databases (PubMed, SPORTDiscus, and Web of Science) to identify peer-reviewed articles published until June 2021. The keywords "Block Start", "Track and Field", "Sprint Running", and "Kinetics and Kinematics" were paired with all possible combinations. Studies reporting biomechanical analysis of the block start and/or first two steps, with track and field sprinters and reporting PB100m were sought for inclusion and analysis. Thirty-six full-text articles were reviewed. Several biomechanical determinants of sprinters have been identified. In the "Set" position, an anthropometry-driven block setting facilitating the hip extension and a rear leg contribution should be encouraged. At the push-off, a rapid extension of both hips and greater force production seems to be important. After block exiting, shorter flight times and greater propulsive forces are the main features of best sprinters. This systematic review emphasizes important findings and recommendations that may be relevant for researchers and coaches. Future research should focus on upper limbs behavior and on the analysis of the training drills used to improve starting performance.

**Keywords:** track and field; sprinters; sprint start; block start; block velocity; biomechanics; kinematics; kinetics; sprint running; initial acceleration; sprint first stance; sprint first two steps

#### **1. Introduction**

The 100 m race is perhaps the highlight of the Olympic Games, as it defines who is the fastest man and woman in the world. In this type of event, the block start performance and the subsequent first two steps can be of critical importance since they have a direct influence on the overall 100 m time [1–8]. Given the importance of the sprint start, a new body of research has emerged in the past two decades that involved advanced technologies, highprecision methods, and sprinters of a higher performance level. For this reason, several technical (kinematic) and dynamic (kinetic) aspects are currently identified as determinant factors for starting block phase and initial sprint acceleration performances [1,4,6,9–25]. However, the concepts, outcomes, and findings between studies are sometimes inconsistent and difficult to interpret and conclude from. These inconsistencies may be accounted for

**Citation:** Valamatos, M.J.; Abrantes, J.M.; Carnide, F.; Valamatos, M.-J.; Monteiro, C.P. Biomechanical Performance Factors in the Track and Field Sprint Start: A Systematic Review. *Int. J. Environ. Res. Public Health* **2022**, *19*, 4074. https:// doi.org/10.3390/ijerph19074074

Academic Editor: Paul B. Tchounwou

Received: 14 February 2022 Accepted: 25 March 2022 Published: 29 March 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/).

by different study designs, methods, technologies of measure (e.g., external reaction forces under or on the blocks), statistical analyses, or more importantly, the ambiguity between samples of sprinters with different performance levels (e.g., elite, sub-elite, well-trained or trained) and/or between-group analyses based on the overall 100 m performance (i.e., personal best at 100 m—PB100m), and not on block performance. Although two important narrative reviews have already been published [26,27], to our knowledge, no previous review conducted a systematic search of literature exploring the inter-individual variability on block start performance across different performance levels. Thus, the main purposes of this systematic review were: (a) determine the biomechanical parameters of greatest influence on the sprint start, including the "set" position and push-off phase, and the first two steps of initial sprint acceleration and (b) identify the kinematic and kinetic biomechanical variables that best differentiate sprinters of different performance levels in each of those three phases of the sprint start. Considering the impact of the sprint in the sports field and the absence of systematic studies on the kinematics and kinetics factors that determine success in block starts and initial sprint acceleration, we hypothesized that this systematic review will have a relevant impact on researchers to better design experimental/intervention studies, as well as constituting relevant support for coaches and athletes in the definition of efficient strategies for performance in the 100 m race.

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

#### *2.1. Article Search, Eligibility, Inclusion, and Exclusion Criteria*

The systematic search of relevant articles was conducted based on PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) guidelines [28]. PubMed, Web of Science, and SPORTDiscus databases were searched for the following mesh terms: "Block Start" OR "Track and Field" OR "Sprint Running" OR "Acceleration" AND "Kinetics and Kinematics" pairing them with all possible combinations. In addition, filters for 'English' and 'articles' have been applied. The last search took place on 30 June 2021.

The inclusion criteria were: publications in English; original observational and experimental studies published in peer-reviewed journals; studies mainly focused on the block phase and/or one or two of the subsequent stance phases concerning kinematic and kinetic variables; and studies that included track and field sprinters with the indication of their PB100m. The following types of records were excluded: conference abstracts; studies focused exclusively on the acceleration phase (beyond the first two stance phases) or mainly focused on limitations imposed by motor and neurological impairments; studies reporting data referring to samples evaluated in previously published papers; studies not mentioning the performance level of the sprinters through their PB100m; case reports; and studies without reference to biomechanical variables.

The records identified from the databases with the aforementioned mesh terms were exported to the reference manager software EndNote X8 that eliminated duplicates. All articles' eligibility was then assessed independently by two reviewers' authors (JMA and FC). The articles identified were first screened by title and abstract for relevance. Studies that raised any uncertainty in exclusion were conservatively retained for subsequent fulltext review. The full text of the articles selected as relevant or having raised uncertainty in exclusion was read and further scrutinized for meeting the inclusion criteria and their quality was evaluated. Disagreements on final inclusion or exclusion of studies were resolved by consensus, and if disagreement persisted, a third reviewer (first author, MJV) was available for adjudication. Articles that did not meet the selection criteria or presented a quality score below 50% were excluded.

#### *2.2. Quality of the Studies*

The study quality of each publication was evaluated according to the guidelines of the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Initiative [29]. This analysis was based on 22 items. Title and abstract. Introduction: background and rationale. Methods: study design, setting, participants, variables, data sources,

bias, sample size, quantitative variables, and statistical methods. Results: participants, descriptive data, outcome data, main results, and other analyses. Discussion: key results, limitations, interpretation, and generalizability. Funding. These criteria were scored on a binary scale (1 = yes, 0 = no) independently by two of the authors, and a quality score was then calculated for each study by adding its binary scores and dividing the result by the maximum possible score the study could have achieved. This was then expressed as a percentage to reflect a measure of methodological quality. The quality scores were classified as follows (a) low methodological quality for scores < 50%; (b) good methodological quality for scores between 50% and 75%; and (c) excellent methodological quality for scores > 75%. The studies with a score lower than 50% [30] were excluded from the systematic review. The inter-rater reliability analysis was evaluated by the Cohen's Kappa for nominal variables (2 dimensions) [31]. Standards for strength of agreement for the kappa coefficient were: ≤0 = poor, 0.01–0.20 = slight, 0.21–0.40 = fair, 0.41–0.60 = moderate, 0.61–0.80 = substantial, and 0.81–1 = almost perfect [32].

#### *2.3. Data Extraction*

An Excel form was used for data extraction. Of each manuscript selected for review, the following information was extracted from each included study: (a) the primary focus of study, means the phase of sprint start, e.g., block phase, first stance, and study design; (b) the main purpose, e.g., associations between biomechanical variables of starting blocks and the sprint start performance, comparing athletes of different performance levels, comparing different footplate spacing and block angles; (c) type of kinematic and kinetic analyses systems used—two dimensional (2D) or three dimensional (3D) analysis and starting blocks instrumented or placed on force platforms; (d) study sample—the number per gender of participants, and per level of expertise of participants according with the authors, and their PB100m; (e) biomechanical measurement protocols—the variables used to characterize the biomechanical factors of sprint start, number and distance of repeated trials; and (f) key findings of sprint start kinematic and kinetic factors.

#### **3. Results**

#### *3.1. Search Results*

The initial search identified 756 titles in the described databases. With the reference manager software, 406 duplicates were eliminated automatically. The remaining 350 articles were then screened according to title and abstract for relevance, resulting in another 289 studies being eliminated from the database. The full text of the remaining 61 articles was read and another 22 were rejected for not meeting the inclusion criteria defined for the current study and 3 studies were excluded for not meeting the quality criteria (quality index < 50%). A total of 36 studies was fully reviewed.

Studies were excluded in the screening stage due to not including track and field athletes or sprint starts using starting blocks (n = 289). In the eligibility stage, there were several reasons for exclusion, namely studies with results focused exclusively on the acceleration phase (n = 8), case studies (n = 4), studies reporting data referring to samples of previously published papers (n = 3) or mainly focused on the limitations of disability (n = 3), lack of information about the PB100m (n = 2) and studies presenting only results for electromyography and reaction time data (n = 2). Figure 1 presents the complete flow diagram.

#### *3.2. Quality of Studies 3.2. Quality of Studies*

presented in Table 1.

In the evaluation of methodological quality, the inter-rater reliability analysis achieved a Kappa value of 0.91 (0.84–0.98), indicating almost perfect agreement between raters. The mean quality score of the included studies was 74.92%. None of the studies achieved the maximum score of 100% and 3 studies (excluded) scored below 50%. Sixteen studies were classified with good methodological quality (quality score between 50 and 75%), while 20 studies had excellent methodological quality (quality score > 75%). The main deficiencies in methodological quality were related to the estimation of sample size and study limitations discussion. In the evaluation of methodological quality, the inter-rater reliability analysis achieved a Kappa value of 0.91 (0.84–0.98), indicating almost perfect agreement between raters. The mean quality score of the included studies was 74.92%. None of the studies achieved the maximum score of 100% and 3 studies (excluded) scored below 50%. Sixteen studies were classified with good methodological quality (quality score between 50 and 75%), while 20 studies had excellent methodological quality (quality score > 75%). The main deficiencies in methodological quality were related to the estimation of sample size and study limitations discussion.

#### *3.3. Basic Characteristics of Included Studies*

*3.3. Basic Characteristics of Included Studies* Fifteen studies [2,3,10–12,17,20,21,23,25,33–37] focused specifically on the block phase, 18 studies [1,4–8,13–16,18,19,24,38–42] on the block phase and, at least one of the subsequent two flight and stance phases, and 3 studies [9,22,43] on the initial acceleration (the first and/or the second step). A summary of all the individual studies reviewed is Fifteen studies [2,3,10–12,17,20,21,23,25,33–37] focused specifically on the block phase, 18 studies [1,4–8,13–16,18,19,24,38–42] on the block phase and, at least one of the subsequent two flight and stance phases, and 3 studies [9,22,43] on the initial acceleration (the first and/or the second step). A summary of all the individual studies reviewed is presented in Table 1.


**Table 1.** Studies are listed in reverse-chronological order by year, followed by alphabetically for studies published in the same year. Samples (n) are restricted to total participating sprinters and are classified by performance level(s) according to the original authors.













38











*Int. J. Environ. Res. Public Health* **2022**, *19*, 4074

Slawinski, Bonnefoy [36]

Block phase. Threedimensional kinematic full-body model.

Measure the joint angular velocity and the kinetic energy of the different segments in elite sprinters

3D kinematics and 3D Euler angular velocities

M 8 Elite 10.30 ± 0.14

Highlights the importance of a 3D analysis of a sprint start. Joints such as shoulders, thoracic, or hips did not reach their maximal angular velocity with a movement of flexion-extension, but with a combination of flexion–extension, abduction–adduction and internal–external rotation.

67.73





EMG—electromyography; F—female sample; F-V—force-velocity; GRF—ground reaction forces; MTU—muscle-tendon unit; M—male sample; P-V—power-velocity; ROM—range of motion; WR—world record; (a) 100 m world record at the study time was 9.58 s; (b) 100 m U20 world record at the study time was 9.97 s; (c) all sample was divided into 3 groups according to the Cormic Index (12 brachycormic, 19 metricormic, and 11 macrocormic); (d) sample divided into two groups: 5 elite sprinters and remaining 52 sprinters; (e) all subjects included in a single experimental group; (f) sample divided into 2 experimental groups: adult/senior vs. junior sprinters; (g) sample divided into 4 experimental groups.

Study purposes included evaluation of specific block start and initial acceleration variables and their influence on block performance (14 studies) [2–4,6,10,11,14,18,23,24,33,36,40,43]; analysis of different "set" position or block configurations (11 studies): location [20] and modulation [35] of center of pressure (COP) on the starting block surface, different block spacing [8,12,37] and widened conditions [21], different block plate obliquities [19,25,34], changed "set" position knee angles [41] and block pre-tension [17]; and comparisons between sprinters of different performance levels, despite the subjectivity associated with the descriptor of the performance level of the athletes (11 studies) [1,5,7,9,13,15,16,22,38,39,42]. The ambiguity in the performance level descriptors includes categories such as: elite vs. sub-elite or well-trained [7,16,22], world-class vs. elite [38], faster vs. slower [5], adult well-trained vs. trained [9,15,42]; elite or well-trained senior vs. junior academy, elite junior, U18 or young well-trained [1,39]; and top sprinters [13]. All studies comparing groups of athletes included male sprinters, but only 4 [1,9,15,38] included women of different performance levels. The studies included in the systematic review presented a cross-sectional study design, except for one study that presented a follow-up design [16].

Twenty-one studies evaluated kinetic variables from blocks start placed on force platforms (12 studies) [5,10,17–21,23,33,35,39,42] or instrumented starting blocks sensors (9 studies) [1,4,11–13,16,24,25,34]. Twelve studies [4,6,9,14,15,18,19,22,24,39,42,43] used a large variety of force platforms arrangements to analyze the dynamic characteristics of the first steps of the initial acceleration.

Concerning kinematic variables, a bi-dimensional analysis, including one or two high-speed digital cameras, was applied in 9 studies [3,12,13,18,19,25,34,37,40], and a 3D kinematic analysis, including 3 [38], 6 [16], or 8 or more cameras [5–9,21,24,36,41] was applied in 11 studies.

Total participants are 766 track and field sprinters, including 179 women and 587 men, and 11 non-trained male subjects [42]. Regarding the sample size of the individual studies selected, Chen, Wu [37] and Debaere, Delecluse [14] are those with the smallest number, 7 participants, and Schrodter, Bruggemann [25] conducted the study with 84 subjects (the largest sample size). The sample sizes from the other studies ranged from 8 [18,36] to 67 [1] subjects, with a mean sample size of 20 participants per study. The mean age of the participants in the selected studies ranged from 15.3 years (under 16) to 28 years. For women, PB100m ranged from 11.10 s (world-class) to 13.10 s (university level), with more classification terms being used, such as "elite" (11.29 to 11.95 s), "well-trained" (11.87 to 12.20 s), "trained" (<11.90 s), or "national level" (11.45 to 12.66) sprinters. Men were classified as "worldclass" (10.03 to 10.98 s), "elite" (9.95 to 10.81 s), "sub-elite" (10.40 to 10.95 s), "well-trained" (10.65 to 11.77 s), "trained" (10.40 to 11.37 s), "national level" (10.58 to 11.22 s), "university level" (10.78 to 12.00 s), or just "sprinters" (10.50 to 11.24 s). Among studies, male PB100m ranged from 9.95 s to 12.00 s.

Through the analysis of the research setup protocols, it was possible to identify a "standard experimental setup". Sixty-nine percent of the studies used distances between 10 and 30 m, with distances shorter than 10 m used only in 4 studies [5,24,41,43] and distances greater than 30 m used in 7 studies [10,20,22,33,37–39]. The number of trials performed ranged between 3 and 10 in 86% of the studies, but in 3 studies [10,20,38] the participants performed 1 or 2 trials, and in 2 studies [40,41] more than 10 trials. Fifty-eight percent of the studies were carried out on an indoor track, 4 studies [12,37,38,40] on an outdoor track, 2 studies [24,41] in a laboratory context, and 9 studies [1,8,10,16,20,23,25,42,44] did not mention the measurement location.

#### *3.4. Data Organization and Analysis*

There was a very large diversity of kinematic and kinetic variables reported among selected studies. Since it is impossible to discuss them all, we will highlight those reported as explicative of high levels of the sprint start performance and that best differentiate faster from slower sprinters. Based on the main findings highlighted in Table 1, the explanatory variables of superior performance levels were identified and systematized in a sequence of tables in Appendixes A–C, related to the "Set" position (Appendix A Table A1), block phase (Appendix B Tables A2 and A3), and first two steps of the initial acceleration (Appendix C Tables A4 and A5). With this strategy of results presentation, it is expected that readers will have access to the primary data extracted from all the studies included in the systematic review. Therefore, Appendix A Table A1 summarizes the kinematic variables in the "Set" position, showing that anthropometry-driven block setting and muscle-tendon unit (MTU) length have an important role in the block start performance. Furthermore, faster sprinters tend to move their center of mass (CM) closer to the starting line and closer to the ground. Concerning joint angles, the knee angular position seems to be a greater performance predictor than any other lower limb joint. At the push-off phase (Appendix B Tables A2 and A3, for kinematic and kinetic variables, respectively) a rear hip extension range of motion (ROM) and a rapid extension of both hips appear to be positively associated with block performance. Moreover, greater average force production during the push against the blocks, especially from the rear leg and particularly the hip, appears to be important for performance. A posterior COP location on block surfaces can also improve sprint performance. Immediately after exiting the blocks, shorter first flight durations and longer first stance durations (allowing more time to generate propulsive force) are the kinematic features of best sprinters (Table A4). During the first two steps of initial acceleration, higher levels of performance seem to be associated with shorter flight times, longer contact times, and the ability to extend the knee throughout both stance phases (Table A5).

#### **4. Discussion**

This paper systematically reviews the kinematic and kinetic biomechanical variables of the block start and initial sprint acceleration phase that influence performance and best differentiate sprinters of different levels. Despite the large number of variables reported in the reviewed studies it was possible to identify some that effectively best describe the influential factors of these events as they are associated with better performance outcomes or best differentiate sprinters of different performance levels. However, notice should be made to the difficulty in analyzing data between studies as there are still no standards for reporting the data, such as measurement units (e.g., m vs. cm) [12,17,18,35], joint angular measurement norms and conventions [3,4,6,12,13,36,38] and/or data normalization methodologies (e.g., for fullheight/lower limb length, body mass/body weight) [2,4,17,22,24,25]. Additionally, there is some subjectivity associated with inconsistent descriptors of performance level [26], confirmed by the variability of the sprinter's classifications used (e.g., from just sprinters to well-trained sprinters, elite sprinters, world-class sprinters, or high-level sprinters) [5,7,16,22,36,38,42]. Another critical factor that somehow may influence data variability between studies is the period of the season in which the data collection took place (e.g., prior to the competition phase of the indoor season vs. during the competitive indoor season or beginning of the summer season) [18].

To better understand the determinant factors of sprint start, the findings from the reviewed studies have been organized into three focuses: (i) the "set" position, (ii) the push-off phase, and (iii) the first two steps of initial acceleration, according to the data presented in Appendixs A–C.

#### *4.1. The "Set" Position*

The "Set" position is the first performance key factor in the block start performance because it depends on block settings and the body posture assumed by sprinters. For the question: "Is there one optimal "Set" position which should be adopted by sprinters?" the answer seems to be no. The researched studies [3,38] showed that it is not an important differentiating factor of performance, since it does not present any correlation with PB100m or normalized block power [3]. However, there are some interesting aspects that sprinters should look out for in a more effective "Set" position [5,12]. The ideal "Set" position depends on the individual anthropometric features [12], strength [38], and morphologic characteristics and motor abilities [13].

#### 4.1.1. Block Settings

The "Set" position depends largely on the anteroposterior block distance, which defines the type of start used. There are three types of block starts based on inter-block spacing: bunched—less than 0.30 m; medium—0.30 to 0.50 m; and elongated—greater than 0.50 m [27,37].

Studies that reported block spacing based on the individual sprinter's preferences [5,12,13,18,35] reported distances between 23.5 ± 1.9 cm (for female sprinters; PB100: 11.97 ± 2.6 s) [13] and 32 ± 5 cm (for male sprinters; PB100m: 10.79 ± 0.21) [18]. This suggests that most sprinters adopt distances within or very close to the bunched start type, favoring CM positioning closer to the starting line [7,38]. Slawinski, Dumas [8] have demonstrated that elongated start settings increase the block velocity (i.e., horizontal CM velocity at the block clearing [7]), but linked to an increase in the pushing time on the blocks which implies a significantly worse performance at 5 and 10 m compared to the bunched start. The same authors showed that the medium start offers the best compromise between the pushing time and the force exerted on the blocks, allowing better times at 10 m [8]. Additionally, more recently, Cavedon, Sandri [12] have demonstrated that the anthropometry-driven block setting based on the sprinter's leg length has an important role in the block start performance leading to a postural adaptation that promotes several kinematic and kinetic advantages [12]. Adjusting inter-block spacing to the relative lengths of the sprinter's trunk and lower limbs (increasing 25.02% the usually bunched start inter-block spacing), allows greater force and impulse on the rear leg and greater total normalized average horizontal external power (NAHEP) [12], the latter one identified as the best descriptor of starting block performance [2].

Other blocks setting features that should be considered in the "set" position are the feet plate obliquity and the amount of pre-tension exerted on the blocks prior to the gunshot. The block inclination (relative to the track) affects the plantar flexor muscle-tendon units' (MTU) initial lengths and determines the muscle mechanics and the external force parameters during the block phase [19,25,34]. Faster sprinters presumably produce the peak torque at longer MTU lengths and adopting a more crouched position would allow them to produce a higher force on the block phase [38]. Research data shows that reductions in both footplates' inclinations (from 65 to 40◦ ), meaning more muscle-tendon pre-stretch, lead to acute increases in block velocity and higher peak joint moments and powers, especially in the ankle [19]. Reductions in front block inclination alone (from 70 to 30◦ ) also acutely increase block velocity without affecting push-off phase duration [34]. In another study [25], however, a greater mean rear block horizontal force was achieved by switching the rear foot to a steeper position (to 65◦ ). This potential conflict between evidence might have arisen from differences in the location of the COP and the length of the footplates' surface between studies since a better sprint start performance is accomplished with a higher and more to the rear COP on the starting block surface [20,35]. Conversely, a pre-tensioned start does not seem to yield a performance advantage over a conventional start, because the increase in the propulsive force of the lower limbs is reversed by an increase in the back force exerted through the hands during the same period [17].

#### 4.1.2. Sprinter Body Posture

Apart from block configuration, the choice of the sprinter's body posture also determines the effectiveness of the "Set" position on the subsequent block push-off phase. The horizontal distance between starting line and the vertical projection of the CM to the ground in the "Set" position (XCM) [7] is a factor that differentiates sprinters with different performance levels. As said before, faster sprinters tend to move their CM closer to the starting line [7,38] and closer to the ground [38]. Elite (PB100: 10.27 ± 0.14 s) and well-trained (PB100: 11.31 ± 0.28 s) male sprinters showed XCM of 22.9 and 27.8 cm, respectively [7]. Likewise, world-class (PB100: 11.10 ± 0.17 s) and elite (PB100: 11.95 ± 0.24 s) female sprinters presented XCM of 16.2 and 24.8 cm, respectively [38]. This more crouched position is only possible due to the high explosive strength of best sprinters, which allows them to produce higher levels of strength in the blocks [38] and reduce the horizontal travel distance of the CM. This body position is complemented by a more advanced shoulder position, putting more tension on the arms, allowing greater blocking speed during the subsequent phase [7].

Related to sprinter joint angles configuration in the "set" position, Milanese and Bertucco [41] have shown that horizontal CM velocity at the block take-off and along the first two steps increases significantly when the rear knee angle is set to 90◦ instead of 135◦ or 115◦ . A 90◦ rear knee angle allows for a better push-off of the rear leg than larger angles, showing such condition may be a strategy that allows some elite sprinters to maximize their strength capacity [41]. A more flexed front knee may facilitate the optimal joint moment production, but only in sprinters with exceptionally high levels of explosive strength [38].

#### *4.2. The Push-Off Phase*

The "block-phase" or "push-off phase" in the starting blocks initiates immediately after the gunshot and is considered a complex motor task that helps to determine sprint start performance [1]. Reaction time is the first factor in the time sequence of the block phase and it is the period from the gun signal to the first measurable change of pressure detected in the instrumented blocks [16]. While a sprinter's ability to react is undeniably important, it is related to the information-processing mechanisms that do not seem to correlate with the performance level [7,45] and, therefore, is beyond the scope of our review (for a review of factors that affect response times, see Milloz, Hayes [46]). Having reacted, the aim of the block phase is to maximize horizontal velocity in as little time as possible. The motion variables during the block phase are, therefore, the focus of this section.

#### 4.2.1. Push-Off Kinematics Analysis

The efficiency of the starting action depends mainly on the compromise between horizontal start velocity (or block velocity) and the block time (referring to the time elapsing from the first movement at the "set" position to the exiting from the block [7]), resulting in the horizontal start acceleration [13]. Despite the horizontal block velocity could be considered the main parameter for an efficient sprint start [13], it cannot be used solely [2] because an increased block velocity could be due to either an increase in the net propulsion force generated or to an increased push-off duration [2,18]. Thus, best sprinters tend to present higher block velocity and greater block acceleration than slower sprinters [1,5,7,13,16,22,39,42], because they are able to produce a greater impulse in a shorter time [2,5,36] and optimize their force production on the blocks [16,19]. In fact, if sprinters increase their anteroposterior force impulse (FI = force × time) from a longer block time, they decrease their block acceleration [2,42] and the performance at 5 and 10 m [8]. Studies comparing data between sprinters of different performance levels mostly show higher block velocities (3.38 ± 0.10 vs. 3.19 ± 0.19 m·s −1 ; 3.48 ± 0.05 vs. 3.24 ± 0.18 m·s −1 ; 3.61 ± 0.08 vs. 3.17 ± 0.19 m·s −1 ; and 3.36 ± 0.15 vs. 3.16 ± 0.18 m·s −1 ) [5,7,22,33] and greater block accelerations (9.5 vs. 8.8 m·s −2 ; 8.2 vs. 7.9 m·s −2 ; 9.72 vs. 8.4 m·s −2 ; and 7.47 vs. 7.35 m·s −2 ) [1,5,7,42] for faster sprinters. Furthermore, higher performance levels also appear to be slightly related to lower block vertical velocities [38] and more horizontal CM projection angles (i.e., resultant direction from the CM horizontal and vertical block exit velocities) [33,39].

Lower limbs joints pattern during the pushing phase (i.e., from movement onset until block exit) is mostly associated with extension movements, especially on the hips and knees [3,4,6,25,36]. The front leg joints typically extend through a considerable ROM in a proximal-to-distal extension pattern [3], reaching their maximum at the beginning of the flight phase (e.g., hip: 183.2 ± 6.8◦ , knee: 177.4 ± 5.2◦ , and ankle: 133.1 ± 6.7◦ ) [6]. Contrarily, the rear leg does not exhibit the same proximal-to-distal extension strategy, with the knee reaching its peak angular velocity before the hip and the ankle [3,36]. This happens

perhaps due to considerably less ROM of the rear knee compared to the front knee [3], as it starts from a more extended angle in the "set" position (e.g., rear knee: 120.7 ± 9.7◦ ; front knee: 91.0 ± 9.8◦ ). The movement of the ankles is more complex because it involves first a dorsiflexion and after an extension resulting in a stretch-shortening cycle of the triceps surae muscle [3,6,25,36]. The duration of the ankle's flexion is greater for the rear ankle (50% of the block phase) than for the front ankle (20% of the block phase) [36]. Experimental manipulations on footplates' inclinations [19,34] have shown an inverse association between block angles and muscle-tendon lengths of the gastrocnemius and soleus, highlighting that block angles steeper than 65◦ could have disadvantageous effects on plantar flexor function [19]. Peak angular velocities at both hips are reached by a combination of flexion– extension, abduction–adduction, and internal–external rotation [23,36], reinforcing the importance of a 3D analysis of the sprint start [36]. Whilst there is a consistent trend among sprinters in the joint angular velocity sequence during the block phase, the lack of comparative data between sprinters of different performance levels does not allow to highlight the technical aspects critical to success. However, a rapid hip extension should be one of the first aspects to consider on a sprinter's technique during the start, as peak angular velocities at both hips and rear hip range of extension are positively associated with block power (*r* = 0.49) [3].

Although upper body kinematics in the push-off phase has been the focus of a small number of studies, some important findings are noteworthy. The action of the upper limbs is more variable between sprinters than that observed for the lower limbs [36]. Despite this, it is possible to recognize a 3D movement pattern for shoulders and trunk with a combination of flexion–extension, abduction–adduction, and internal–external rotation movements, while the elbows exhibit an extension and pronation movement [36]. The velocity of the rear shoulder tends to be slightly greater than that of the other joints, but the peak resultant angular velocities at the upper limb joints are comparable to those at lower limbs during the push-off phase, particularly that of both knees and front ankle [36]. However, there is no evidence linking different upper limb kinematic patterns with any block phase performance predictor, and further research is needed to compile relevant recommendations for athletes and coaches.

#### 4.2.2. Push-Off Kinetic Analysis

According to Newton's second law of motion, horizontal CM acceleration requires net propulsive forces to be applied to the athlete's body in the sprinting direction. Therefore, as said before, the horizontal force impulse, made up by the mean horizontal force and push-off time, is the determining factor of the horizontal velocity at block exit [2,5,36,42]. The relationship between these factors (i.e., horizontal force and push-off time) shows that the application of a greater amount of horizontal force is a key performance factor [42], as an increase in the time action (block time) conflicts with the criterion for 100 m performance: 'shortest time possible'. Thus, best sprinters generate greater average forces [10,22], higher rates of force development [7,25], and larger net [7] and horizontal [5] block impulses than their slower counterparts. Likewise, Graham-Smith, Colyer [39] comparing senior to junior athletes also showed that sprinters with faster PB100m (senior athletes) exhibit higher relative horizontal force during the initial block phase and higher forces during the transition from bilateral to unilateral pushing [39]. The evident importance of the force generated against the blocks for proficient execution of the starting block phase has encouraged researchers to gain a deeper understanding of the kinetic determinants of such a crucial phase of sprinting. Bezodis, Salo [2] tried to find the push-off performance measure that was more adequate, objective, and possible to quantify in the field. From their analysis, the NAHEP was identified as the most appropriate measure of performance because it objectively reflects, in a single measure, how much sprinters are able to increase their velocities and the associated length of time taken to achieve this, whilst accounting for variations in morphologies between sprinters [2]. Later, the identification of the magnitude of the force applied to both blocks and their optimal orientation as major determinants

of performance encouraged researchers to gain a deeper understanding of the push-off forces applied against each block separately. Consequently, some studies support the importance of the force generated by the front leg for forwards propulsion [6,42] and show that faster sprinters are able to produce higher force impulses in the front block than slower sprinters [5,33] (for example: 221.3 ± 15.8 N·s vs. 178.3 ± 13.1 N·s for faster and slower sprinters, respectively [5]). Colyer, Graham-Smith [33] reinforce this feature highlighting that higher front block force production during the transition (when the rear foot leaves the block, 54% of the block push) and a more horizontally orientated front block force vector in the block phase (81–92%) are important performance-differentiating factors. However, other evidence ensures that the rear block force magnitudes are the most predictive external kinetic features of block power [10,33] and sprint performance [5,7,12,16]. For example, Coh, Peharec [5] found that a faster group of sprinters (PB100m = 10.66 ± 0.18 s; 913 ± 89.23 N) produced greater total forces against the rear block than a group of slower sprinters (PB100m = 11.00 ± 0.06 s; 771 ± 55.09 N). A longer relative rear leg push (i.e., as a percentage of the total push-off phase) is also positively associated (*r* = 0.53 [3]) with greater block power [3,10] and is present in sprinters with faster PB100m [5,7,33]. Modulations of the COP on the starting block surface showed that COP location may also be related to initial sprint performance [20,35]. Better sprint start performance appears to be achieved with a higher and more to the rear COP during the force production phase [20]. Thus, athletes and coaches should keep in mind that pushing the calcaneus onto the block (posterior location) may improve the 10 m time and/or horizontal external power for some individuals [35].

Forces under the hands have been reported in relatively few studies [10,33,42], showing somewhat contradictory results. While some point to a primary support role [42], others point out that the best athletes produced less negative horizontal impulse under hands compared with their slower counterparts [33]. Therefore, the importance of the hands' kinetics during the push-off phase remains unclear and should be the subject of future research.

In addition to external kinetic analyses, which provide valuable insight into starting block performance, the analysis of internal kinetics (i.e., joint kinetics) helps to increase the understanding of the segment motions that are responsible for CM acceleration. Recent research of joint kinetics has shown that 55% of the variance in NAHEP of a group of sprinters with a PB100m of 10.67 s was mainly accounted for by rear ankle joint moment (23%), front hip joint moment (15%), and front knee joint power (15%). The remaining 2% was shared by the remaining lower limbs joint kinetic variables [11]. In the rear block, the magnitude of the horizontal force produced is determined by the rear hip extensor moment and the rear hip extensor power coupled with large ankle joint plantarflexion moment [4,11,19], without any significant knee joint contribution [4,11]. At the front block, a proximal–distal pattern of peak joint power is evident [4], highlighting a strategy often adopted in power demanding tasks, with the main periods of positive extensor power at the front ankle and knee occurring after the rear foot has left the block [4]. In a study with 12 sprinters from the University of Tokyo team (PB100m: 10.78 ± 0.19 s), Sado, Yoshioka [23] showed that the peak lumbosacral extension moment was significantly larger than any other lumbosacral and lower-limb moment, being positively correlated with the starting performance. This peak value appeared in the double-stance phase where both hip joints exerted extension moments. The aforementioned evidence supports the findings of Slawinski, Bonnefoy [36] who showed that the lower limbs and the head–trunk segments are the two main segments that contribute to the kinetic energy of the total body. Upper limbs contribute 22% to the total body kinetic energy, demonstrating that their actions in the pushing phase on the blocks are not negligible [36].

#### *4.3. The First Two Steps*

The primary goal of the first steps is to generate a high horizontal velocity [40]. However, the transition between block start and the first steps represents a specific biomechanical paradigm: integrate temporal and spatial acyclic movements into a cyclic action [5]. The

efficiency of this transition depends on the biomechanical demands of the first stances after block clearance, which are very different from the other stances during acceleration [14]. The sprinter aims to generate maximal forward acceleration during the transition from start block into sprint running [2,14,22,42] while generating sufficient upward acceleration to erect itself from a flexed position in the start blocks to a more extended position [6,14]. Specific technical (kinematic) and dynamic (kinetic) skills are therefore needed to successfully achieve this transition, and they are the focus of this section.

#### 4.3.1. First Two Steps Kinematic Analysis

The primary goal of the initial steps of a sprint running is to generate a high horizontal sprint velocity, which results from the product of the length and frequency of the sprinter's steps [22,40]. Spatiotemporal parameters have shown that the sprinter's step length increases regularly during the acceleration phase, while step frequency is almost instantaneously leveled to the maximum possible [22]. Typically, the step frequency reaches the maximal values very quickly (80% at the first step and about 90% after the third step) [22], achieving around 4 Hz immediately after block exit [26,40]. The length of the first steps is more variable between sprinters, ranging from 0.82 to 1.068 m (senior females) [1,38] or 0.85 to 1.371 m (senior males) [1,7] on the first step, and from 1.06 to 1.30 m (senior females) [1,13] or 1.053 to 2.10 m (senior males) [7,37] on the second step. Despite this variability, step length tends to be longer in faster sprinters, particularly in the first step (e.g., 1.371 ± 0.090 vs. 1.208 ± 0.087 m [7]; 1.30 ± 0.51 vs. 1.06 ± 0.60 m [5]; 1.135 ± 0.025 vs. 0.968 ± 0.162 m [38]), exhibiting an increase of about 14 cm for every 1 s less in PB100m [38]. This may be a consequence of the lower vertical velocity of the CM at the block clearing shown by faster sprinters, allowing them to travel a longer distance despite shorter flight times [38]. Indeed, the kinematics of faster sprinters is also characterized by a tendency to assume long ground contact times in the first two steps (e.g., mean first contact duration for Diamond League sprinters is 0.210 s for males and 0.225 s for females, which is greater than those of lower-level Italian junior sprinters: 0.176 and 0.166 s, respectively), associated to short flight times (0.045 and 0.064 s, for the first flight of world-class and elite male sprinters, respectively) [38]. This strategy allows the high-level sprinters to optimize the time during which propulsive force can be generated, minimizing the time spent in flight where force cannot be generated. Combined with this, best sprinters have their CM projected further forward [7] at the first touchdown, putting the foot behind the vertical projection of the CM [3], and minimizing the braking phase. At the takeoff of the first and second steps, the CM horizontal position is also greater in elite than well-trained sprinters [7]. This means that the CM resultant and horizontal velocity in the first two steps are generally greater in high-level sprinters [7,15]. Slawinski, Bonnefoy [7], for example, reported that elite sprinters have a CM resultant velocity 5.8% higher than well-trained sprinters, at the end of the first step (4.69 ± 0.15 vs. 4.42 ± 0.11 m·s −1 for elite and welltrained sprinters, respectively). Furthermore, high-level sprinters also show slightly lower vertical velocities [7,39] and more horizontal CM projection angles at the end of the first two support phases [39].

Lower limb joints pattern during the first two steps is associated with a proximal-to-distal sequence of the hip, knee, and ankle of the stance leg [4,9,43]. During both first and second steps, the ankle joint undergoes dorsiflexion during the first half of stance (e.g., 17 ± 3 ◦ and 18 ± 3 ◦ for the first and second steps, respectively [43]) and subsequently a plantarflexion movement (e.g., 45 ± 6 ◦ and 44 ± 5 ◦ for the first and second steps, respectively [43]).

The hip performs extension for the entire stances, the knee extends until the final 5% of stances, and the ankle is dorsi-flexed during the first half of stances before the plantar flexing action [6]. After leaving the rear block, there is a small increase in ankle joint dorsiflexion during the swing phase, preceding the plantarflexion that occurs just before touchdown [6]. Although the ankle plantar-flexes slightly at the end of the flight, the ankle is in a dorsi-flexed position at initial contact (e.g., first stance: 70.6 ± 5.8◦ and second stance: 72.4 ± 7.1◦ [6]). During both first and second steps, the ankle joint dorsi-flexes

during the first half of stance (e.g., 17 ± 3 ◦ and 18 ± 3 ◦ for the first and second steps, respectively [43]) and subsequently performs a plantarflexion movement (e.g., 45 ± 6 ◦ and 44 ± 5 ◦ for the first and second stance, respectively [43]). Note that a reduction in the range of dorsiflexion during early stance, requiring high plantar flexor moments, has already been associated with increases in first stance power [47]. Maximal plantarflexion occurs immediately following takeoff reaching, for example, 111.3◦ at the first stance and 107.1◦ at the second stance [6]. The extension of both knees occurs just after the block exit and reaches its maximum at the beginning of the flight phase, with larger extension in the front compared with the rear leg (e.g., rear: 134.9 ± 11.2◦ ; front: 177.4 ± 5.2◦ ) [6]. From a flexed position at initial contact, the knee extensors generate power to induce extension throughout stance and to attain maximal extension at takeoff, achieving peak extension angles of around 160–170◦ (not full extension; e.g., first stance: 165.2 ± 20.6◦ ; second stance: 163.6 ± 17.7◦ [6]). This extension action of the knee during stances on its own may play a role in the rise of the CM during early acceleration [26]. The hip joints extend during block clearance to reach maximal extension during the beginning of the flight phase. During stance, the hips are in a flexed position at initial contact and continue to extend throughout stance, achieving maximal extension immediately following takeoff (e.g., first stance: 180.6 ± 20.9◦ ; second stance: 181.1 ± 20.0◦ [6]). There is also a considerable ROM in hip and pelvis rotation during stance as well as abduction. Although there are detailed descriptions of the lower limb angular kinematics during the first two stances and flight phases [3,6], there seems to be no clear evidence about the joint kinematic features that differentiate faster from slower sprinters. Furthermore, there is also a lack of experimental data on arm actions during early acceleration and its relationship to performance descriptors, making necessary future research in this area to help identify the most important performance features.

#### 4.3.2. First Two Steps Kinetic Analysis

As said before, fast acceleration is a crucial determinant of performance in sprint running, where a high horizontal force impulse in a short time [13] is essential to reach high horizontal velocity [43]. Thus, as the highest CM acceleration during a sprint occurs during the first stances [7,9,14] (e.g., first stance: 0.36 ± 0.05 m·s −2 ; second stance: 0.23 ± 0.04 m·s −2 [14]), the ability to generate during this phase greater absolute impulse [7,18], maximal external power [39,42], and a forward-leaning force oriented in the sagittal plane [21,22,24,42] is linked to an overall higher sprint performance. Larger propulsive horizontal forces are particularly important during early acceleration, being a discriminating factor for superior levels of performance [48]. Experienced male sprinters (PB100m: 10.79 ± 0.21 s) can produce propulsive horizontal forces of around 1.1 bodyweight during the first stance [18]. However, a negative horizontal force has also been reported during the first contact after the block exit, even if the foot is properly placed behind the vertical projection of the CM [18]. During the first stance, for example, the braking phase represents about 13% of the total stance phase and the magnitude of the braking forces can reach up to 40% of the respective propulsive forces [18].

Furthermore, 3D analysis studies also highlight a lower body motion outside the sagittal plane during the first few ground contact phases [6,21,22,24,36,42]. In fact, during the first steps of a sprinter, a stance medial deviation is often observed that results from an impulse in the transverse plane. Although the medial impulse is the smallest of the three orthogonal stance impulses [21,22,42], the fact that it is non-zero can have an effect on the motion of the CM and on step width. However, it has been shown that well-trained sprinters present similar step widths in the early acceleration to those of the trained and non-trained sprinters [42]. Moreover, manipulations of both "set" position [21] and first step [24] widths have shown no effect on block-induced power nor braking force or net anteroposterior impulse, showing that smaller step width is not a discriminator factor of superior performance levels. Therefore, the perception that the adoption of a widened stance during initial acceleration (referred to as "skating style") is detrimental to performance is not at all proven, and further research is needed to clarify the joint and muscular factors that contribute to the sprinters' lateral motion in the initial phase of acceleration.

At joint level, the hip, knee, and ankle joints generate energy during stance leg extension [6], although it appears that the ankle joint is the main contributor to CM acceleration [14]. However, experimental and simulation studies highlight that the knee plays an important role during the first stance, being decisive for forward and upward CM acceleration [4,6,14,15]. The importance of power generation at the knee seems to be specific for the first stance when the knee is in a more flexed position and the sprinter is leaning forward. From the second stance onwards, the knee becomes less and the ankle more dominant since the plantar flexors are in a better position to contribute to forward progression [6]. As the knee is in a flexed position during the first step, the sprinter favors the immediate power generation of the knee extensors rather than preserving a stretch-shortening cycle [6]. In contrast, a stretch-shortening mechanism can be confirmed at the hip and ankle [4,6,14,15]. Hip extensors maximal power generation occurs near touchdown [4,6] where the hip extensors actively pull the body over the touchdown point [6]. The hip can effectively generate large joint moments and power [14], but only contributes minimally to propulsion and body lift during the first two stances [14]. Ankle plantar flexors act throughout both the first and second stances under a stretch-shortening cycle. There is therefore an initial phase of power absorption preceding the forceful power generation at take-off [4,14]. As a major contributor to CM acceleration, the ankle joint can generate up to four times more power than it absorbs during the first two stances [43]. Nevertheless, the importance of ankle stiffness during the first two stances remains unclear. While Charalambous, Irwin [49], in a case report, found a correlation between greater ankle stiffness and greater horizontal CM velocity at take-off (*r* = 0.74), Aeles, Jonkers [9] did not, still highlighting the lack of differences between faster (senior) and slower (junior) sprinters. Future work is therefore needed to further clarify this issue. Furthermore, it remains unclear whether ankle stiffness is influenced by foot structure and function (e.g., planus, rectus cavus, clubfoot) as well as other important performance variables such as greater maximal power, a forward-leaning force oriented in the sagittal plane, or COP location during push-off.

Concerning kinetic factors differentiating senior and junior athletes, Graham-Smith, Colyer [39] reported that, contrarily to the block phase where there are marked differences between groups, the force and power waveforms relating to the first two steps did not differ considerably across groups. Still, senior sprinters are able to produce greater horizontal power during the initial part (10–19% of the stance phase) of the first and second ground contact (first step: 25.1 <sup>±</sup> 3.6 W·kg−<sup>1</sup> vs. 23.1 <sup>±</sup> 6 W·kg−<sup>1</sup> and second step: 26.7 <sup>±</sup> 3.6 W·kg−<sup>1</sup> vs. 24.9 <sup>±</sup> 4.5 W·kg−<sup>1</sup> , forsenior and junior sprinters, respectively), and also exhibit a higher proportion of forces immediately after braking forces are reversed (from 9% to 15% and 25% to 29% of stance phase) [39]. Furthermore, Debaere, Vanwanseele [15] also highlight that adult sprinters are able to generate more joint power at the knee during the first step compared to young sprinters, inducing longer step length and therefore higher velocity [15]. Younger sprinters tend to prioritize a different technique: the hip contributes more to total power generation, while the knee contributes far less [15]. This indicates that younger sprinters lack the specific technical skills observed in adult sprinters, likely due to less musculature than adults [1,9,15]. However, there is no evidence of differences in ankle joint stiffness, range of dorsiflexion, or plantar flexor moment between young and adult sprinters [9]. This indicates that the technical performance-related parameters of the first stances are not likely to explain the better 100 m sprint times in adult compared to young sprinters [9].

#### *4.4. Strengths, Limitations, and Recommendations*

A strength of this review was that it allowed us to identify a body of knowledge that provides fundamental information for athletes and coaches as relevant data that can contribute to improving the training and/or preparation strategies for better performance, supported by scientific evidence.

A possible limitation of this systematic review is that it only includes studies written in English, thereby potentially overlooking other relevant publications in other languages. Additionally, the present article reviewed only studies with mention to sprinters' PB100m, eventually precluding publications with relevant samples that could also add knowledge. Furthermore, extending the biomechanical analysis to muscular features beyond the simple kinematic and kinetic approach might have allowed a further understanding of the discriminating factors of superior performance levels. Another obvious limitation is the limited amount of research with female sprinters. Indeed, in the reviewed studies, there is a clear imbalance between the amount of female and male sprinters included (179 females vs. 587 males), questioning whether the biomechanical characteristics of the sprint start previously associated to female sprinters are attributable to sex-related aspects, or, rather, to aspects related to the 100 m time. Moreover, some of the studies included in this review were based on a relatively small sample size, especially when elite or world-class sprinters were included. This problem reflects the difficult access to high-level athletes, preventing the clear identification of discriminatory factors of superior performance levels. Finally, the conflicting classifications of sprinters level and the scarcity of information on effectively high-level or world-class sprinters, makes it difficult to compare sprinters of different performance levels. Considering entry standards for 100 m sprint event at the 2022 European Athletics Championships (10.16 s for men and 11.24 s for women), it can be said that a very small percentage of elite and/or world-class sprinters [50] was included in the reviewed studies.

Research on the biomechanics of the block and/or first stance phases has been the subject of growing interest in the past few years. Nonetheless, there are some unclear features in the studies published so far, which should be investigated in future studies for a better understanding of: (i) the association between different upper limb patterns and the main block start performance predictors; (ii) the influence of foot type (e.g., planus, rectus cavus, clubfoot on sprint start performance; (iii) the association between ankle stiffness during dorsiflexion and the horizontal CM velocity at take-off; (iv) the specificity characteristics of training drills, utilizing temporal organization and intra-limb joint coordination analyses, to help the process of exercise selection to enhance block starting performance; (v) how technical and/or physical training can improve ankle and knee function during first steps and increase horizontal velocity in the early acceleration; (vi) the influence of sex (such as physical or muscle structures and/or anthropometric characteristics) on sprint start performance descriptors. A major challenge for researchers is to align these research lines with the need for greater information on world-class sprinters during competition. Whenever possible, research based on a marker-less methodology and obtained during official top-level sprint competitions, during which the sprinters are supposedly more motivated to produce their best performance, should be encouraged.

It is worth mentioning two new studies [51,52] published after the date of this systematic review, which, meeting the defined inclusion criteria, could have added important knowledge on some of the issues mentioned above.

#### **5. Conclusions**

Based on this review, some important conclusions and recommendations to help athletes and coaches can be made, namely: (i) the choice of an anteroposterior block distance relative to the sprinter's leg length may be beneficial for some individuals, promoting greater block start performance (greater normalized average horizontal external power); (ii) the use of footplate inclinations that individually facilitate initial dorsiflexion should be encouraged—footplate angles around the 40◦ are recommended and block angles steeper than 65◦ should be avoided; (iii) pushing the calcaneus onto the block (posterior location) may be beneficial for some individuals, improving the 10 m time and/or horizontal external power; (iv) short block exit flight times and optimized first stance contact times should be encouraged, as they maximize the time during which propulsive force can be generated; (v) focus attention on the magnitude of force applied on the rear block, as it is considered to be a primary determinant of block clearance; (vi) rapid hip extension during the push-off

phase should be a priority in sprinter focus and coach feedback; (vii) the large role played by the hips on the push-off phase and by both the knee and ankle at the early stance must be acknowledged within physical and technical training to ensure strength and power are developed effectively for the nature of the sprint start.

**Author Contributions:** Conceptualization, M.J.V., J.M.A. and F.C.; methodology, M.J.V., J.M.A., F.C. and C.P.M.; formal analysis, M.J.V., J.M.A. and F.C.; writing—original draft preparation, M.J.V., J.M.A. and M.-J.V.; writing—review and editing, M.-J.V., F.C. and C.P.M.; visualization, M.J.V., M.-J.V., F.C. and C.P.M.; supervision, M.J.V., F.C. and C.P.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was partly supported by a national grant through the FCT—Fundação para a Ciência e Tecnologia within the unit I&D 447 (UIDB/00447/2020).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

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