Introduction

Upper-body strength-related measures (e.g., maximal strength; rate of force development), and anthropometric characteristics, have been shown to differentiate sub-elite and elite athletes [1, 2]. Research has also demonstrated the positive relationship between upper-body power measures, such as bench press throws and upper-body Wingate test [3, 4], and strength measures, such as the bench pull, bench press, and pulldown test [5,6,7,8], with athletic performance in various sports, including golf, rugby, handcycling, (sprint) swimming, double poling, and (sprint) kayaking. Furthermore, markers of upper-body strength, such as one repetition maximum (1RM) for the bench press and bench row or handgrip strength, have proven to be useful for identifying talented athletes in kayaking, cross-country skiing, and tennis [9,10,11]. In addition to its relevance to performance, upper-body strength-related measures can impact training dosage, as weaker individuals may require longer rest intervals between sets [12, 13]. Moreover, upper-body strength, particularly handgrip strength, is a prognostic indicator of morbidity and all-cause mortality [14,15,16], and can predict fall severity in older adults [17, 18]. Therefore, upper-body strength-related measures are important from both performance and health-related perspectives. The development of safe, effective, and convenient training methods that can improve these outcomes is necessary.

Various methods of upper-body resistance training are commonly used by strength and conditioning specialists, coaches, and athletes [19]. In this context, upper-body plyometric training (UBPT) offers some advantages over other training approaches. UBPT is effective in improving measures of power [20] and strength [21] through exercises that maximize the storage and use of elastic energy through the muscle stretch–shortening cycle, typically with a brief transition period between the stretching and shortening phases of muscle actions [22,23,24]. UBPT exercises, such as bench press throws and medicine ball throws, can be added to heavy resistance training, such as the bench press and dumbbell shoulder press, to improve power-related performance, such as medicine ball throwing speed [25], and maximal upper-body strength such as 1RM bench press [26]. This can be particularly relevant in sports where powerful upper-body actions are more common, such as golf, rugby, handcycling, and baseball. Therefore, UBPT is considered a time-efficient training method [27, 28]. Additionally, UBPT is well-suited for sports, such as handball and volleyball, that require powerful upper-body actions as it mimics the motor patterns demanded by these sports more closely than non-plyometric strength training methods.

Research studies have demonstrated that plyometric training can significantly improve sport-specific performance in sports like soccer (e.g., kicking velocity) [29,30,31] and swimming [32]. However, the transferability of these findings into practical settings has been limited due to the small sample sizes of most published studies [33]. Studies investigating UBPT often have small sample sizes (e.g., n < 13) [34, 35], which can be a drawback in the sport science literature [36]. To overcome this issue, systematic reviews with meta-analyses can be useful in evaluating the results of comparable studies and helping practitioners make evidence-based decisions [37, 38]. While previous systematic reviews on plyometrics have mainly focused on lower body training [33, 39], some conflicting findings have been reported regarding the effectiveness of UBPT in enhancing physical performance parameters like strength, power, and throwing speed [40, 41].

A prior systematic review, which included a database search up to August 2017 [42] assessed the impact of UBPT on strength, ball throwing speed, distance and power in healthy individuals, including only six randomized controlled trials. Given the rapid growth in the field of plyometric training research in the recent years, with the rate of yearly publications increasing almost 16-fold between 1980–1999 and 2000–2019 [33], it is crucial to re-evaluate the scientific literature. In rapidly emerging research fields, 25–50% of systematic reviews may be obsolete within 2–5 years [43]. Therefore, the primary objective of this systematic review with meta-analysis was to examine the effects of UBPT, compared with active/passive controls, on the physical fitness of healthy youth and young adult participants.

Methods

Registration

The systematic review with meta-analysis protocol was registered on the Open Science Framework (OSF) platform, in January 13, 2023 (registration code: XQ25T; registration https://doi.org/10.17605/OSF.IO/XQ25T).

Procedures

The procedures were in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines [44] and the Cochrane Handbook for Systematic Reviews of Interventions [45].

Literature Search: Administration and Update

A systematic literature search was performed without date restriction, and updated up to February 2023, in the electronic databases PubMed, Web of Science, and SCOPUS, using the Boolean operators AND/OR, in combinations with the keywords “plyometric”, “explosive”, “ballistic”, “upper body”, and “upper limb”. An exemplified combination and the search strategy (code line) for each database is described in the Additional file 1: Table S1. One author (EGC) conducted the initial search and removed duplicates. Two authors (EGC and RKT) independently screened the titles, abstracts, and full-texts of the retrieved studies. The search results were then analysed according to the eligibility criteria (Table 1). A third author (RRC) resolved potential disagreements between EGC and RKT. Although not considered in the original protocol, to expand the scope of our search and better abide by AMSTAR 2 guidelines [46], we contacted two experts in the field of UBPT (https://expertscape.com/ex/plyometric+exercise), sharing with them our inclusion/exclusion criteria and a detailed table of articles (i.e., authors, publication year, identifier, title, and journal) that we have identified as potential candidates for inclusion and asked if they were aware of any additional studies that could be added to the review.

Table 1 Eligibility criteria
Full size table

Inclusion and Exclusion Criteria

Original, peer-reviewed, full-text studies were selected, and included/excluded using a PICOS (participants, interventions, comparators, outcomes, and study design) framework [47] (Table 1). Additional exclusion criteria are provided as Additional file 1: Table S2. As 99.6% of the plyometric training literature is published in English [39], and due to limited resources, only articles written in English, French, German, Italian, Portuguese, and Spanish (i.e., authors' native languages), were considered for inclusion.

Data Extraction

The effects of UBPT on upper-body physical performance measures were assessed, compared to active (e.g., athletes in standard training), specific-active (e.g., a group performing high-load resistance training) and/or non-active controls. Performance measures included (but not limited to) different specific tests (e.g., bench press, medicine ball throw) and indices (e.g., kg; power [W]). Measures like the powerful medicine ball throw (intra-class correlation coefficient = 0.93–0.99) and the bench press throw (intra-class correlation coefficient = 0.94–0.85; coefficient of variation = 2.48%) have shown excellent reliability among competitive or physically active individuals [48, 49], thus favouring meta-analysis consistency [47].

Pre- and post-intervention means and standard deviations of the dependent variables were extracted from the included studies using Microsoft Excel (Microsoft Corporation, Redmond, WA, USA). For studies reporting values other than means and standard deviation (e.g., median, standard error), conversion was applied [50,51,52] or appropriate statistical software was used for different data formats (Comprehensive Meta-Analysis Software, Version 2, Biostat, Englewood, NJ, USA). When the required data were not clearly or completely reported, the authors of the respective studies were contacted for clarification purposes. If no response was obtained from the authors or the authors did not provide the requested data, the study outcome was excluded from further analysis. When needed, a validated [53] software (WebPlotDigitizer, version 4.5; https://apps.automeris.io/wpd/) was used to obtain numerical data from articles that reported results in figures. One author (EGC) performed data extraction, a second author (RKT) provided confirmation, and a third author (RRC) helped to resolve any discrepancies.

Risk of Bias of the Included Studies

Included studies were assessed with a valid and reliable tool (i.e., PEDro scale) [54,55,56], probably the most frequently used in the plyometric training literature [33, 57, 58]. Despite being called a "methodological quality" scale, its items assess aspects of the likelihood that research will be biased. Therefore, it is helpful to compare meta-analyses, especially when they used different risk of bias assessment methodologies [59, 60]. Considering that it is not possible to satisfy all scale items in UBPT interventions [61], the overall risk of bias of studies was interpreted using the following convention [57, 61,62,63]: ≤ 3 points was considered as “poor” quality (i.e., high risk of bias), 4–5 points was considered as “moderate” quality, while 6–7 points and 8–10 points was considered as “good” and “excellent” quality, respectively. For practical purposes and given the nature of the research field, we considered studies with ≥ 6 points to have low risk of bias. Two authors (EGC and RKT) independently assessed risk of bias, and a third author (RRC) helped to resolve discrepancies.

Summary Measures, Synthesis of Results, and Publication Bias

Meta-analyses can be conducted with as little as two studies [64]. However, meta-analyses were performed if ≥ 3 studies were available due to the reduced number of participants commonly included in plyometric training studies [65,66,67]. Meta-analysis was performed using the DerSimonian and Laird random-effects model [68, 69], with the means and standard deviations from pre and post values taken to compute effect sizes (ES, i.e., Hedges' g, presented with 95% confidence intervals [95% CIs]) for performance parameters in the UBPT compared to the control groups. Data were standardised using post-intervention standard deviation values. Calculated ES were interpreted as trivial (< 0.2), small (0.2–0.6), moderate (> 0.6–1.2), large (> 1.2–2.0), very large (> 2.0–4.0), and extremely large (> 4.0) [70]. A proportional division treatment to the control group n was applied in studies including multiple intervention groups and a single control group [71]. The I2 statistic was used to assess the impact of heterogeneity, with values of < 25%, 25–75%, and > 75% representing low, moderate, and high levels of heterogeneity, respectively [72]. The extended Egger’s test was used to assess risk of publication bias for continuous variables (≥ 10 studies per outcome) [73,74,75] and a sensitivity analysis was conducted with the trim and fill method for adjustments [76], with L0 as the default estimator for the number of missing studies [77]. All analyses were carried out using the Comprehensive Meta-Analysis Software (Version 2, Biostat, Englewood, NJ, USA). Statistical significance was set at p ≤ 0.05.

Additional Analyses

Sensibility Analyses

The robustness of the summary estimates (e.g., p-value, ES, I2) for each outcome was analysed with each study deleted from the model (automated leave-one-out analysis).

Certainty of Evidence

Two authors (JA and RRC) rated the certainty of evidence (i.e., high; moderate; low; very low) using the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) [78,79,80]. The evidence started at a high level of certainty (per outcome), but was downgraded based on the following criteria: (i) Risk of bias in studies: judgments were downgraded by one level if the median PEDro scores were moderate (< 6) or by two levels if they were poor (< 4); (ii) Indirectness: low risk of indirectness was attributed by default due to the specificity of populations, interventions, comparators and outcomes being guaranteed by the eligibility criteria; (iii) Risk of publication bias: downgraded by one level if there was suspected publication bias; (iv) Inconsistency: judgments were downgraded by one or two levels when the impact of statistical heterogeneity (I2) was moderate (≥ 25%) or high (> 75%); (v) Imprecision: one level of downgrading occurred whenever < 800 participants were available for a comparison [81] and/or if there was no clear direction of the effects. When both were observed, certainty was downgraded by two levels.

Adverse Effects

The potential adverse health effects, derived from the inadequate implementation of UBPT interventions, were qualitatively assessed.

Results

Study Selection

The search process in the databases identified 5587 records. Figure 1 provides a flow chart illustrating the study selection process.

Fig. 1
figure 1

Flow diagram of the systematic search process. (*): denotes that 2 studies were identified through other sources (i.e., previous systematic review)

Full size image

Duplicate records were removed (n = 3617). After titles and abstracts were screened, 1853 records were removed and 117 full texts were evaluated. Thereafter, 35 studies were considered eligible for the systematic review, and 30 studies for the meta-analyses. The exclusion reasons from the meta-analysis for the five studies [20, 82,83,84,85] are indicated in Fig. 1. From the 30 studies eligible for meta-analyses 29 were written in English [21, 26, 28, 40, 41, 86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109] and one in Spanish [110].

Risk of Bias of the Included Studies

According to the PEDro checklist results (Table 2), the median (i.e., non-parametric) score was 6.0 (low risk of bias—good quality), with nine studies attaining 4–5 points (some risk of bias—moderate quality), and 26 studies attaining 6 points (low risk of bias—good quality).

Table 2 Rating of studies according to the Physiotherapy Evidence Database (PEDro) scale
Full size table

Study Characteristics

The characteristics of the participants and the UBPT programs of the included studies are detailed in Table 3.

Table 3 Descriptive characteristics of participants and upper-body plyometric training interventions
Full size table

Ten studies [26, 40, 83, 85, 98, 99, 103, 105, 107, 108] examined non-athletes (including resistance-trained participants and physical education students). Other studies examined athletes from different sports such as handball (n = 5) [86, 88, 90, 91, 96], basketball [28, 82, 101], baseball [41, 89, 104], tennis [93, 94, 110], volleyball [20, 21, 106] (n = 3 for each sport), cricket [84, 102] (n = 2), golf [87], karate [97], rugby [100], softball [92], and table tennis [95] (n = 1 for each sport). Of note, one study [109] included participants from different sports (i.e., water polo, field hockey, gymnastics, and volleyball). A total of 1,412 participants, with an age range of 7.3–27.2 years, were analysed in this systematic review. Taking all included studies together, 773 individuals from 52 groups participated in the intervention programs and 639 participated in the control groups (35 groups). Among the 35 control groups, 22 groups were active controls (e.g., handball players), 6 groups were passive controls, and the other 7 groups were intervention control groups (e.g., high-load resistance training). Nineteen experimental groups (and their respective controls) involved participants with a mean age of < 18 years (Table 3). Regarding participants' sex, 6 studies reported a mixed sample of male and females (n = 287 [20.3% of total participants]), 13 groups involved females only (n = 216 [15.3% of total participants]), 44 groups involved males (n = 821 [58.2% of total participants]), and 6 groups involved unspecified participants' sex (n = 88 [6.2% of total participants]) (Table 3). All except six studies [21, 94, 97, 105, 109, 110] recruited experimental and control groups from the same sample through randomization procedures (e.g., groups with similar [probability] training-competition level, age, sport). According to one study [94], pre-tests were employed to regulate the initial status of players. Training duration in the intervention and control groups ranged from 4 to 12 weeks (Table 3), although most studies lasted 8 weeks (54.3%, n = 19). The frequency of weekly training sessions ranged from 2 to 4 sessions per week with a median (and mode) of 2 (Table 3). The testing protocols involved mostly medicine ball throws (n = 17 studies), maximal strength performance (n = 13), and sport-specific throwing performance (n = 10).

Results From the Meta-Analysis

Maximal Strength Performance

Thirteen studies provided data for maximal strength performance, involving 15 intervention groups and 15 control/comparator groups. Results showed a small significant effect for the intervention groups compared to the control (i.e., passive, active, specific-active) groups: ES = 0.39, 95% CI = 0.15–0.63, p = 0.002, I2 = 29.7%, total participants n = 363 (Fig. 2). Egger's two-tailed test revealed a p-value < 0.001, and after the Duval and Tweedie´s trim and fill adjustment method (with three studies trimmed to the right of the mean), the adjusted values indicated an ES = 0.52, 95% CI = 0.27–0.76. After the sensitivity analyses (automated leave-one-out analysis), the robustness of the summary estimates (i.e., p-value, ES and 95% CI) was confirmed.

Fig. 2
figure 2

Forest plot illustrating plyometric training-related improvements of the maximal strength performance in comparison to control (i.e., passive, active, specific-active) groups. Forest plot values are shown as effect sizes (Hedges’ g) with 95% confidence intervals (CI). Black squares: individual studies. The size represents the relative weight. White rhomboid: summary value. Rot dom: dominant torso rotational strength; Rot non-dom: non-dominant torso rotational strength

Full size image

Medicine Ball Throwing Performance

Seventeen studies provided data for medicine ball throwing (MBT) performance, involving 20 intervention groups and 20 control/comparator groups, using medicine balls with different weights: 1 kg [21, 96, 98, 99, 103], 1.5 kg [97], 2 kg [21, 94, 104, 110], 2.7 kg [107], 3 kg [21, 28, 96,97,98,99,100,101,102, 106], 3.6 kg [92], 4 kg [21], and 5 kg [21]. Results showed a significant effect for the intervention groups compared to the control (i.e., passive, active, specific-active) groups: ES = 0.64, 95% CI = 0.43–0.85, p < 0.001, I2 = 46.3%, total participants n = 819, Egger's test two-tailed = 0.229 (Fig. 3). After the sensitivity analyses (automated leave-one-out analysis), the robustness of the summary estimates (i.e., p-value, ES and 95% CI) was confirmed.

Fig. 3
figure 3

Forest plot illustrating plyometric training-related improvements of the medicine ball throwing performance in comparison to control (i.e., passive, active, specific-active) groups. Forest plot values are shown as effect sizes (Hedges’ g) with 95% confidence intervals (CI). Black squares: individual studies. The size represents the relative weight. White rhomboid: summary value

Full size image

Sport-Specific Throwing Performance

Ten studies provided data for sport-specific throwing performance, involving 10 intervention groups and 10 control/comparator groups. Results showed a significant effect for the intervention groups compared to the control (i.e., passive, active, specific-active) groups: ES = 0.55, 95% CI = 0.25 to 0.86, p < 0.001, I2 = 36.8%, total participants n = 291 (Fig. 4). Egger's two-tailed test revealed a p-value = 0.029, and after the Duval and Tweedie´s trim and fill adjustment method (with five studies trimmed to the left of the mean) the adjusted values indicated an ES = 0.34, 95% CI = − 0.01–0.68. After the sensitivity analyses (automated leave-one-out analysis), the robustness of the summary estimates (i.e., p-value, ES and 95% CI) was confirmed.

Fig. 4
figure 4

Forest plot illustrating plyometric training-related improvements of the sport-specific throwing performance in comparison to control (i.e., passive, active, specific-active) groups. Forest plot values are shown as effect sizes (Hedges’ g) with 95% confidence intervals (CI). Black squares: individual studies. The size represents the relative weight. White rhomboid: summary value

Full size image

Muscle Volume

Three studies provided data for upper-body muscle volume, involving 3 intervention groups and 3 control/comparator groups. Results showed a significant effect for the intervention groups compared to the active control groups: ES = 0.64, 95% CI = 0.20–1.08, p = 0.005, I2 = 0.0%, total participants n = 78 (Fig. 5).

Fig. 5
figure 5

Forest plot illustrating plyometric training-related increase of the upper-body muscle volume in comparison to active controls. Forest plot values are shown as effect sizes (Hedges’ g) with 95% confidence intervals (CI). Black squares: individual studies. The size represents the relative weight. White rhomboid: summary value

Full size image

Certainty of Evidence

Results from the GRADE analyses are presented in Table 4. For medicine ball throwing and muscle volume the certainty of evidence was considered low. For maximal strength performance and sport-specific throwing, the evidence was rated as very low.

Table 4 GRADE analysis
Full size table

Adverse Effects

Most of the included studies did not report soreness, pain, fatigue, injury, damage, or adverse health effects related to the UBPT intervention. One study indicated that four participants from the control group were excluded because of unspecified injury and insufficient attendance at training/assessment sessions [88]. Another study reported that 6/30 players from the experimental group and 3/30 from the control group were excluded from the final analysis due to acute injuries (i.e., ankle sprain) produced during plyometric training (only 1 player) or during specific sport training (tennis, 8 players) [94].

Discussion

The primary goal of this systematic review with meta-analysis was to examine the effects of UBPT on the physical fitness of healthy youth and young adult participants, compared to active, specific-active, and passive matched controls. The main findings indicate that UBPT resulted in small to moderate improvements in maximal strength (ES = 0.52), medicine ball throwing performance (ES = 0.64), sport-specific throwing performance (ES = 0.34), and muscle volume (ES = 0.64) in healthy youth and young adult individuals.

Maximal Strength Performance

Compared to control conditions (i.e., passive, active, and specific-active), UBPT improved maximal strength performance in healthy individuals with small effect sizes ranging from 0.39 to 0.52. This observed improvement is consistent with several physiological adaptations related to maximal strength performance [111, 112], such as improved neuromuscular function (e.g., power, proprioception, and postural control) [113, 114] and increased muscle activation [115]. Moreover, improvements in muscle–tendon stiffness [116, 117] and architecture [116, 118], including increased muscle cross-sectional area, fibre type distribution, muscle mechanics of the upper and lower limbs (i.e., length. and muscle pennation angle), and neural adaptations such as increased firing rates, increased motoneuron excitability and decreased presynaptic inhibition [119] have also been noted after plyometric training.

Improvements in maximal strength are crucial for success in various sports [7, 120,121,122], especially those where athletes should overcome larger loads (e.g., throwing events, weightlifting) [123]. However, the percentage of athletes from maximal strength sports that incorporate UBPT regularly is relatively low, with ~ 14% in powerlifting and ~ 29% in strongman [124, 125]. Results from this systematic review can be helpful in evidence-based practice, but the limited number of studies available on the effect of UBPT on maximal strength performance precludes a robust analysis of the optimal prescription variables to maximize improvements.

Nevertheless, effective and safe UBPT interventions, with an exercise frequency of 2–3 times per week and lasting between 4 and 12 weeks, can be a valuable addition to resistance training programs aimed at improving maximum strength. The exercises performed during the training sessions mainly consisted of medicine ball throws, push-ups, and bench press throws (at 40–55% 1RM [91, 109]). Although these exercises seem to be effective in improving maximal strength, the magnitude of the improvement was small (ES = 0.39). Future research should explore the use of other exercises that may have a greater impact on maximal strength. In this regard, resistance training performed at ~ 90% of 1RM with maximal volitional concentric contraction velocity has been shown to be a suitable method to focus on the nervous system, and investigations have shown that it is more effective in enhancing force generation than conventional resistance training performed at ~ 70–75% of the 1RM [126]. Therefore, the combination of resistance and plyometric training could be a time-effective method, potentially providing larger improvements in maximal upper-body strength [26, 104, 127]. Nevertheless, it is important to be cautious when interpreting the current results, as the GRADE analysis indicates a very low level of certainty concerning this outcome.

Medicine Ball Throwing Performance

After examining 17 studies that compared UBPT to various control conditions (i.e., passive, active, specific-active control groups), a moderate increase (ES = 0.64) in MBT performance was observed. Plyometric training can induce physiological adaptations associated with MBT performance, such as enhancement of the elastic properties of the musculotendon unit, neural sequencing optimization and firing rates of the motor units involved [57]. UBPT might also induce improvements in muscular fitness traits associated with MBT, including muscle power, explosiveness (e.g., rate of force development), and coordination [48]. However, the learning effect should be considered, as individuals demonstrated greater improvements in performance with weights similar to those used during training sessions which is in line with the principle of training specificity [97].

When using MBT as a field test in the studies included in this systematic review, variability was observed in testing protocols, including different throwing positions (seated, kneeling, standing), different numbers of maximal throwing trials, and the use of medicine balls with different weights ranging from 1 to 5 kg. Regardless of the testing protocol used, increased MBT performance after UBPT can be an important adaptation for athletes, as it can predict performance improvements in shot put [128] or handball throwing [129, 130].

It is noteworthy that adding plyometric training to resistance training programs was more effective than resistance training alone in improving upper-body power (measured by the seated medicine ball throw) [92], torso and sequential hip-torso-arm rotational strength [104]. However, resistance training has been observed to be more effective than MBT in increasing throwing velocity among individuals without prior experience in resistance training [41]. It appears that the total training workload is the most critical factor in increasing overhead throwing speed [131]. Despite this, the present meta-analysis's findings should be interpreted with caution due to the low certainty of evidence in the GRADE analysis.

Sport-Specific Throwing Performance

A systematic review and meta-analysis revealed a modest improvement (ES = 0.55) in sport-specific throwing performance after UBPT compared to control groups, including active, specific-active, and passive control groups. UBPT exercises can lead to physiological adaptations that may enhance throwing performance, such as increased firing rates of motor units, improved inter- and intra-muscular coordination, increased muscle fibre contraction velocity, and improvements in force and power generation capabilities [132,133,134,135]. Moreover, UBPT exercises predominantly target the velocity components of the force–velocity spectrum, which are important for sport-specific throwing [136]. Other neuromuscular adaptations induced by UBPT exercises may involve a better utilization of the stretch-reflex of the upper-body in a high-velocity ballistic manner, similar to the sport-specific throwing skill [137].

Included studies in this systematic review measured sport-specific throwing performance, including handball throwing velocity (standing, jumping, and three-step running), baseball throwing velocity, and ball speed during golf driving. Previous studies have reported the significant transference effects of plyometric exercises to sport-specific performance in other sports (e.g., soccer, swimming) [29, 31, 32]. Furthermore, UBPT-induced adaptations may lead to improved kinetic characteristics during throwing such as increased force, power, and rate of force development [88]. In addition, the increase in maximal strength (Fig. 2) and MBT performance (Fig. 3) discussed in previous sections may also contribute to sport-specific throwing performance [138]. For instance, a significant association between upper-body maximal strength and ball throwing velocity (r = 0.64–0.69) was previously noted [139, 140]. Results arising from the current meta-analysis should be interpreted with caution as the certainty of evidence is very low for this outcome in the GRADE analysis.

Upper-Body muscle Volume

The results of this meta-analysis suggest that UBPT increases the upper-body muscle volume compared to the control groups (ES = 0.64). No previous meta-analysis has reported the effects of UBPT on upper-body muscle volume. However, previous meta-analyses have reported plyometric training to be effective in muscle hypertrophy [141] and increased muscle architecture (e.g., muscle thickness, fascicle length, pennation angle) [142] of the lower limbs. Indeed, plyometric exercises have also been reported to increase the calf-girth in physically active adults [143]. Moreover, a systematic review suggested that plyometric training may produce similar effects on whole muscle hypertrophy compared to traditional resistance training methods for lower extremities muscles, at least in the short-term (i.e., 12 weeks) [141]. The physiological basis of such an increment of muscle volume may be attributed to specific muscle fibre recruitment during the plyometric exercises [144]. UBPT is characterized by short duration high-velocity movement that activates the motor units associated with fast-twitch muscle fibres [144] which are suggested to have greater potential for hypertrophy [138, 145]. However, such speculations about hypertrophy response being specific to muscle fibre type have been suggested to be controversial in the literature [138, 146]. Another mechanism that may be responsible for the increase in muscle volume is the increase in the rate of protein synthesis [147], which has the potential to increase the muscle volume. This is noteworthy because body composition can influence physical performance such as speed, change of direction, and upper limb explosive strength [148]. However, results of this meta-analysis should be interpreted with caution due to the low number of studies (n = 3) and the low certainty of evidence obtained in the GRADE analysis.

Limitations and Directions for Future Research

According to the GRADE analysis, the level of certainty of the evidence for most outcomes ranged from very low to low. Therefore, it is important to exercise caution when interpreting the results of the current meta-analysis. To enhance the reliability and applicability of the findings, future research should aim to overcome these limitations by utilizing larger sample sizes and conducting more randomized controlled trials. It is noteworthy, however, that this study is a significant addition to the field, as it has summarized the existing evidence and pinpointed areas requiring further investigation.

While a significant number of studies were included in this review, UBPT is still relatively understudied compared to lower-body plyometric training. Future studies on UBPT should focus on determining the optimal dose for participants based on their characteristics. Additionally, the extent to which UBPT transfers to sport-specific performance in other upper-body-dominated sports (e.g., badminton, tennis, boxing) remains to be determined. Furthermore, most studies included in this review involved male participants, with only six studies including females (as shown in Table 3). This sex disparity highlights the need for more inclusive and representative studies to fully understand the effects of UBPT on female physical fitness.

Practical Applications

The findings of this systematic review with meta-analysis have significant implications for coaches and practitioners involved in training programs, aiming to optimize training protocols targeting various aspects of physical fitness and sport-specific performance. Table 5 provides evidence-based practical recommendations for UBPT programming for trainers and practitioners seeking to improve maximal strength, medicine ball and sport-specific throwing performance, and upper-body muscle volume among youth and young adult population.

Table 5 Evidence-based practical recommendations for UBPT programming
Full size table

To maximize the potential benefits derived from UBPT on specific training goals (e.g., maximal strength; medicine ball throwing performance [e.g., maximal power]), practitioners may consider a set of evidence-based practical recommendations indicated in Table 5: (i) include a warm-up (≤ 15 min) before UBPT sessions, (ii) perform 2–4 UBPT sessions per week, (iii) continue for a duration of at least 4 weeks, (iv) include a variety of exercises (e.g., 4–12 per session), particularly for long-term interventions, (v) after adequate technique reached for a particular exercise consider a progression up to maximal-intensity exercises, and (vi) include 1–6 sets of 1–10 repetitions per exercise, for a total of 50–100 repetitions per session.

It should be noted that the practical recommendations presented in this systematic review are based on information from the individual studies, rather than the meta-analyses themselves. Unfortunately, few studies directly compared different variables of UBPT prescription (e.g., exercise type, intensity), which limits our ability to analyse these factors.

Conclusions

The current systematic review provides evidence to support the effectiveness of UBPT interventions in enhancing physical fitness and performance outcomes in healthy youth and young adults. The findings suggest that UBPT can improve maximal strength, medicine ball throwing performance, sport-specific throwing performance, and muscle volume compared to control groups. However, caution should be exercised when interpreting the results, as the level of certainty of the evidence was found to be between very low and low.