How to Improve the Reactive Strength Index among Male Athletes? A Systematic Review with Meta-Analysis

rength index (RSI) describes the individual's capability to quickly change from an eccentric muscular contraction to a concentric one and can be used to monitor, assess, and reduce the risk of athlete's injury. The purpose of this review is to compare the effectiveness of different training programs on RSI. Electronic searches were conducted in MEDLINE, PubMed, Scopus, SPORTDiscus, and Web of Science from database inception to 11 February 2022. This meta-analysis was conducted in accordance with the recommendations of the preferred reporting items for systematic reviews and meta-analyses (PRISMA). The search returned 5890 records, in

of this review is to compare the effectiveness of different training programs on RSI. Electronic searches were conducted in MEDLINE, PubMed, Scopus, SPORTDiscus, and Web of Science from database inception to 11 February 2022. This meta-analysis was conducted in accordance with the recommendations of the preferred reporting items for systematic reviews and meta-analyses (PRISMA). The search returned 5890 records, in which 39 studies were included in the systematic review and 30 studies were included in the meta- analysis. Results from the randomized studies with the control group revealed that plyometric training improved RSI in adult athletes (0.84, 95% CI 0.37 to 1.32) and youth athletes (0.30, 95% CI 0.13 to 0.47). Evidence withdrawn from randomized studies without a control group revealed that resistance training also improved the RSI (0.44, 95% CI 0.08 to 0.79) in youth athletes but not in adults. Interventions with plyometric training routines have a relatively large, statistically signi cant overall effect in both adult and youth athletes. This supports the implementation of this type of interventions in early ages to better cope with the physical demands of the various sports. The impact of resistance training is very low in adult athletes, as these should seek to have a more power-type training to see improvements on the RSI. More interventions with sprint and combined training are needed. Keywords:strength; power; reactive strength; players; plyometric training; resistance training 1. Introduction The reactive strength index (RSI) describes the individual's capability to quickly change from an eccentric muscular contraction to a concentric one [1]. In other words, the RSI was created to assess the athlete's reactive strength, and it was originally measured with the drop jump (DJ) test [1]. For this test, the athletes must perform a vertical jump as soon as they land on the ground from a speci c height [1]. The hands can stay on the athletes' hips throughout the test or not, as both methods have shown good levels of reliability [2]. This test should incorporate various drop heights to assess at what height the athlete can elevate more his centre of gravity, and it was already

as they land on the ground from a speci c height [1]. The hands can stay on the athletes' hips throughout the test or not, as both methods have shown good levels of reliability [2]. This test should incorporate various drop heights to assess at what height the athlete can elevate more his centre of gravity, and it was already proven to be a reliable and valid test to measure the RSI [3]. The RSI can be calculated by dividing the jump height by Healthcare2022,10, 593.

Healthcare2022,10, 593 2 of 17 the ground contact time, providing valuable information for coaches, regarding plyometric performance (i.e., jump height) and how each jump is performed (i.e., ground contact time) [4]. Jump height can be measured directly or can be derived from ight time with the following mathematical formula [5]: jump height (m) = (gravity ( ight time) 2 )/8, where gravity = 9.81 m/s and ight time is in seconds. Most recently, with advancements in technology, more tests have been developed to measure the RSI, such as the countermovement (CMJ), tuck jump, squat jump, weighted CMJ, single-leg jump [6], 10/5 [7], single rebound jump [8], vertical rebound for 5 repeti- tions [9], vertical rebound for 15 repetitions [10], and vertical rebound for 10 s tests [11]. In those cases where there is no drop or rebound jump, the RSI is designated explicitly by reactive strength index modi ed (RSI mod), since it is calculated by dividing the jump height by the time to take-off (time to produce force from the beginning of the eccentric muscular phase until the moment the athlete leaves the ground) [6]. To obtain these variables mentioned before (i.e., jump height, ight time, ground contact time, and time to take-off), three different methods can be used: (a) the ight time [12]; (b) the difference between the height of two marks during the jump [13]; and (c) the mathematical integration of the ground reaction force [14]. The rst one requires the use of contact mats [12,15,16], photocell mats [13,15], or accelerometers [16,17]. The second method uses different devices to calculate displacement (i.e., linear position trans- ducers) [18]. The third method is considered the best one, as its accuracy is extremely high if adequate sampling frequency methods are chosen and requires the use of one or two force plates [13,19]. The use of RSI is vital for high-performance sports professionals, as it can be used as a motivational tool, in a way that coaches can instantly deliver feedback to their athletes, ac- cording to their RSI value, in order to improve their physical performance [5]. Furthermore, both RSI

chosen and requires the use of one or two force plates [13,19]. The use of RSI is vital for high-performance sports professionals, as it can be used as a motivational tool, in a way that coaches can instantly deliver feedback to their athletes, ac- cording to their RSI value, in order to improve their physical performance [5]. Furthermore, both RSI and RSI modcan be used as variables to potentially monitor athlete's neuromuscu- lar readiness [20]. Moreover, the RSI has been shown to have a strong relationship with change of direction speed, acceleration speed [21], and agility [22]. Additionally, maximal strength, especially relative to body mass, appears to have a very strong relationship with RSI mod, indicating that stronger athletes tend to have better reactive strength [23]. However, there is no clear information on which type of training would produce better improvements on RSI. Besides that, to the best of our knowledge, the scienti c literature does not review this topic. Therefore, the aim of this study was to analyse the strategies that can improve the RSI of male athletes, through a systematic review of experimental research and meta-analysis. 2. Methods 2.1. Protocol and Registration This meta-analysis was conducted in accordance with the recommendations of the preferred reporting items for systematic reviews and meta-analyses (PRISMA) [24]. The study protocol was registered with PROSPERO (CRD42020176616). 2.2. Information Sources The literature search on ve electronic databases (i.e., MEDLINE, PubMed, Scopus, SPORTDiscus, and Web of Science) started on 5 July 2020 and was conducted from database inception to 11 February 2022. 2.3. Search Strategy All retrieved papers were exported to CADIMA software, a tool designed to increase the ef ciency of the evidence synthesis process and facilitate reporting of all activities to maximize methodological rigor [25]. Duplicates were automatically removed. Titles and abstracts of potentially relevant papers were screened by two reviewers (A.R. and J.R.P.). Disagreements between authors were solved through discussion and, when necessary, three other authors (D.M., J.P.D., and J.V.-d.-S.) were involved. Full-text copies were acquired for

rigor [25]. Duplicates were automatically removed. Titles and abstracts of potentially relevant papers were screened by two reviewers (A.R. and J.R.P.). Disagreements between authors were solved through discussion and, when necessary, three other authors (D.M., J.P.D., and J.V.-d.-S.) were involved. Full-text copies were acquired for

Healthcare2022,10, 593 3 of 17 all papers that met title and abstract screening criteria. Full-text screening was performed by two reviewers (A.R. and J.R.P.). Again, any discrepancies were discussed, until the authors reached an agreement and consulted the three other authors, when required. The comprehensive search strategy is available in the Supplementary File (Supplemental Material S1). 2.4. Inclusion Criteria Scienti c peer-reviewed published papers written in English, Portuguese, French, and Spanish were eligible for the present systematic review. The review sought to identify all studies reporting exercise interventions to improve the RSI in both male adult and youth athletes. Therefore, studies were eligible if: (1) subjects were male athletes; (2) subjects had between 11 and 45 years old; (3) the study included at least two moments of evaluation, with a baseline RSI measurement and post-intervention RSI measurement; (4) the study included a training program that aimed to improve the RSI. 2.5. Exclusion Criteria Studies that do not describe a protocol to induce effects on RSI or that used RSI during a recovery program were excluded from the present study. 2.6. Categorisation of Studies We identi ed six categories of exercise interventions, through the process of reviewing the included studies. The de nitions of these exercise interventions are provided in Table. Table 1.De nition of types of interventions and comparators. Type De nition Intervention Plyometrics Exercises that are designed to enhance neuromuscular performance on the lower limbs. This involves application of jump, hopping, and bounding training. Resistance training Training program that aims to improve strength, power, or hypertrophy with resistances (e.g., elastic bands, barbells, dumbbells, kettlebells, or body weight). Sprint training Acceleration or maximal velocity training either resisted or unloaded. Change of direction (COD) or sprint or plyometric or a combination of those COD: Any exercise that enforces the participant to accelerate, decelerate and do a COD. This type of intervention is de ned by a combination of one or more of sprint training, COD training, or plyometric training. Sports-speci c training Sports-speci c exercises training (e.g., small-sided games in soccer). Control Maintained training routines Sport training routines 2.7. Data Extraction

of those COD: Any exercise that enforces the participant to accelerate, decelerate and do a COD. This type of intervention is de ned by a combination of one or more of sprint training, COD training, or plyometric training. Sports-speci c training Sports-speci c exercises training (e.g., small-sided games in soccer). Control Maintained training routines Sport training routines 2.7. Data Extraction Pre-established data extraction criteria were created with seven items: (a) general information (authors name and year of the study), (b) sample characteristics (size and age), (c) sport, (d) training program, (e) measurement equipment, (f) methodology, and (g) results. 2.8. Risk of Bias Assessment The revised Cochrane risk of bias tool for randomized trials (RoB 2.0) scale was used to quantify the risk of bias in eligible, individually randomized, parallel-group trials and provide information on the general methodological quality of studies. The RoB 2.0 scale rates internal study validity and the presence of replicable statistical information on a scale from low risk of bias to high risk of bias [26]. The risk of bias in non-randomized studies of interventions (ROBINS-I) tool was used to quantify the risk of bias in eligible

Healthcare2022,10, 593 4 of 17 non-randomized trials and provide information on the general methodological quality of the studies. The ROBINS-I scale rates internal study validity and the presence of replicable statistical information on a scale from 1 (low risk of bias) to 4 (high risk of bias) [27]. Inter-rater agreement was calculated using Cohen's kappa coef cient (k). Using different tools to assess the risk of bias on randomized and non-randomized studies was supported elsewhere [28]. 2.9. Statistical Models Meta-analyses were conducted to estimate the overall effects of the intervention programs to improve RSI for all available data, as well as for studies that only included randomized samples. Within the previous categorization, meta-analyses were also divided by studies with (intervention vs. control) and without a control group (pre-intervention vs. post-intervention). Additionally, subgroup meta-analyses were performed to identify the effects of each training program (“plyometrics”, “resistance training”, “sprint training”, “sprint or plyometric or a combination of those”, and “sports-speci c training”) on adults ( 18 years old) and youth (<18 years old) athletes. Additionally, pre- and post-intervention results from all the included studies (randomized and non-randomized) were ranked on two different meta-analyses to understand the effectiveness of each training program on adult and youth athletes. Meta-analysis, and the respective forest plots, were calculated when the mean, stan- dard deviation, and sample sizes were introduced on the Cochrane collaboration's review manager computer program (RevMan version 5.4.1, Oxford, UK). When this data was impossible to retrieve from the manuscript [29–38], authors were contacted to provide the missing information. Most of the authors replied to the request [29–34]; therefore, the data was included on the meta-analyses. In addition, when the same study reported different RSI measurements (i.e., from different heights and/or tests), the method that resulted in the highest positive performance change was selected for the meta-analysis. The pooled data for each outcome were reported as standardized mean differences (SMD), with a 95% con dence interval (CI). Each meta-analysis was performed using the random-effects model, and heterogeneity was assessed using I 2 statistic and chi-square (Q) tests. A Q value with a signi cance

that resulted in the highest positive performance change was selected for the meta-analysis. The pooled data for each outcome were reported as standardized mean differences (SMD), with a 95% con dence interval (CI). Each meta-analysis was performed using the random-effects model, and heterogeneity was assessed using I 2 statistic and chi-square (Q) tests. A Q value with a signi cance ofp 0.05 was considered signi cant heterogeneity, while, for the I 2 value, 25% was considered low, 50% was considered moderate, and 75% was considered high heterogeneity [39]. 3. Results 3.1. Study Selection The search strategy returned 5890 records, and the PRISMA ow diagram [40] is shown in Figure. Records were excluded based on the included participants (not male athletes), intervention or comparator (not a training program), post-intervention data (not reporting RSI measurements), or testing on surfaces other than the oor (i.e., force sledge). In addition, for the quantitative synthesis (meta-analysis), ineligible studies were excluded for reporting RSI data only in gures and/or percentage [35–38] or for being unique, in terms of physical training method [41,42] and, therefore, not being able to pair it with other studies to conduct a meta-analysis. In total, thirty-nine studies were included in the systematic review and thirty-three were included in the meta-analysis. 3.2. Risk of Bias Assessment Inter-rater agreement for the risk of bias assessment, using the RoB 2.0, was = 0.933; for the ROBINS-I, it was = 1.0. Thus, overall, the risk of bias within individual studies assessed using the RoB 2.0 scale ranged between low and high risk of bias (Supplemental Material S2), whereas the ROBINS-I scale ranged from a serious to critical risk of bias (Supplemental Material S3).

Healthcare2022,10, 593 5 of 17Healthcare 2021, 9, x 5 of 17 Figure 1. PRISMA statement flow chart. RSI, reactive strength index. 3.2. Risk of Bias Assessment Inter-rater agreement for the risk of bias assessment, using the RoB 2.0, was κ = 0.933; for the ROBINS-I, it was κ = 1.0. Thus, overall, the risk of bias within individual studies assessed using the RoB 2.0 scale ranged between low and high risk of bias (Supplemental material 2), whereas the ROBINS-I scale ranged from a serious to critical risk of bias (Supplemental material 3). Of the thirty-three randomized studies assessed with the RoB 2.0, twelve (36%) [35,43–53] had a high risk of bias. Randomisation process (18%) and selection of reported results (15%) were the most common sources of high risk of bias. Of the six non-randomized studies assessed with the ROBINS-I, one (17%) [54] was of critical risk of bias assessment, due to their departures from the intended interventions, participants being excluded because of missing data, and selection of participants based on their characteristics observed after the start of the intervention. The remaining five studies (83%) [36,42,55–57] were of serious risk of bias assessment. 3.3. Study Characteristics Study characteristics (including training program characteristics) for all 39 included studies are presented in Supplemental material 4. Overall, the most common tests used to quantify the RSI are DJs (79%) [29–36,38,41–43,45,46,49–54,56–66], vertical hops (21%) [44,47,48,67–71], and CMJs (8%) [37,55,72]. The most popular materials used to quantify Figure 1.PRISMA statement ow chart. RSI, reactive strength index. Of the thirty-three randomized studies assessed with the RoB 2.0, twelve (36%) [35,43–53] had a high risk of bias. Randomisation process (18%) and selection of reported results (15%) were the most common sources of high risk of bias. Of the six non-randomized studies assessed with the ROBINS-I, one (17%) [54] was of critical risk of bias assessment, due to their departures from the intended interventions, participants being excluded because of missing data, and selection of participants based on their characteristics observed after the start of the intervention. The remaining ve studies (83%) [36,42,55–57] were of serious risk of bias assessment.

six non-randomized studies assessed with the ROBINS-I, one (17%) [54] was of critical risk of bias assessment, due to their departures from the intended interventions, participants being excluded because of missing data, and selection of participants based on their characteristics observed after the start of the intervention. The remaining ve studies (83%) [36,42,55–57] were of serious risk of bias assessment. 3.3. Study Characteristics Study characteristics (including training program characteristics) for all 39 included stud- ies are presented in Supplemental Material S4. Overall, the most common tests used to quantify the RSI are DJs (79%) [29–36,38,41–43,45,46,49–54,56–66], vertical hops(21%) [ , and CMJs (8%) [37,55,72]. The most popular materials used to quantify the RSI are contact mats (46%) [ 29–34,36,38,45,46,49,53,61,62,68,69,71,72], force plates (36%) [35,37,41–43,48,50,52,54–56,60,65,66], and photoelectric systems (15%) [44,47,57,58,67,70]. Soccer is the most common sport studied (56%) [29–34,36,38,44–48,50,52,53,61,62,67–70], fol- lowed by rugby (18%) [35,43,50,52,58,60,65] and basketball (8%) [50,53,72]. Intervention dura- tion ranged from four weeks [

This systematic review compares the effectiveness of different training programs on the reactive strength index in male athletes.