This study explored the acute effects of different warm-up methods on lower limb explosive strength from an evidence-based medical perspective. The results of a net meta-analysis showed that SDS and DS were able to have a positive effect on explosive strength overall, while SS showed a negative effect, with a smaller but still statistically significant effect size and no results yet proving the effectiveness of other warm-up methods. This finding is largely in line with previous research findings 47,48. However, a recent meta-analysis was inconsistent with the results of this study 49, and the analysis suggests that this inconsistency might be related to the study methodology on the one hand. This study used a traditional meta-analysis, which was only able to analyse the relative validity of the effects of different warm-up methods on the acute effects of lower limb explosive strength, and it lacked a comprehensive comparative analysis between studies, whereas this study used a continuum analysis to test the stability and reliability of the relative evidence between different warm-up methods, and the indirect comparative modelling approach provided more accurate statistical analysis 50, which would allow relevant coaches and athletes alternatives to different warm-up methods, rather than in a traditional meta-analysis of individual warm-up method studies. Therefore, it is speculated that the study method might have influenced the results. Another aspect could be related to publication bias in the literature. Publication bias in meta-analyses can lead to lower actual effect sizes and can also increase the risk of no and negative effect sizes 51, so-called false 'positive' or false 'negative' results. The study did not test for the risk of publication bias, so there is a potential risk of publication bias. The study used a qualitative (funnel plot) test of bias to ensure the accuracy of the results to a certain extent.
Analysis of the effect of different warm-up methods on the acute effect of lower limb explosive power
Dynamic stretching affects the effect
The results of the study showed that the DS group had a better effect size than the control group for both CMJ performance [MD = 1.60, 95% CI: (0.67, 2.60)] and sprint performance [MD = -0.08, 95% CI: (-0.15, -0.008)], and the results of the SUCRA ranking chart indicated that, although slightly lower than the SDS group for CMJ performance, its effect on explosive power was an acute effect that was significantly better than the other warm-up methods. Regarding the mechanism by which DS promotes explosive strength, some scholars believe that it might be related to DS increasing body and muscle temperature. During stretching, muscles actively contract and stretch, increasing temperature while decreasing viscosity; at the same time, increased muscle temperature can cause increased neuroreceptor sensitivity, and this increased sensitivity suggests that neuromuscles might show stronger motor unit activation through increased motor unit recruitment 52, thereby improving muscle contractile performance. For example, Fletcher et al. showed 53 that DS was able to significantly elevate muscle temperature and ultimately reverse longitudinal jump height compared to SS. Furthermore, some scholars have found that the ability of DS to enhance subsequent explosive performance could be related to the preactivation of movement patterns 54, 55. Studies have shown that it might stimulate the muscle shuttle to increase muscle reflex activity, thus allowing the muscle to better complete active contractions based on the characteristics of the subsequent movement and ultimately inducing a postactivation potentiation effect (PAP) 56.
DS has long been recommended as an essential component of warm-up activities, and its effects on improving joint mobility and preventing sports injuries are well established, but the effect on subsequent explosive power remains controversial, and the reasons for this controversy might be influenced by the duration of stretching, the study population and other factors. The results of the study showed significant heterogeneity (I2 = 75.6%) in the results of the comparison between the DS and control groups, suggesting that the variation caused by real differences in effect sizes accounted for 75.6% of the total variation, which can be considered a large degree of dispersion in the effect sizes of the individual studies; therefore, it is necessary to introduce moderating variables to investigate the heterogeneity in depth.
(1) In terms of stretch time. The test found that a stretch time of 7–10 min produced the largest effect size. This finding is largely in line with previous findings. A study by Mcmillian et al. 57 found that DS with a total time duration of 10 min was able to enhance subsequent jump performance compared to the SS and control groups. Similarly, Behm et al. 58 clearly suggested that a 7 min and 10 min DS could have a positive effect on subsequent explosive power. The reason for this finding might be that shorter periods of stretching do not make effective use of the compliance effects on tendon units, while longer periods can cause fatigue in the body, thus not maximising the benefits of dynamic stretching.
(2) Study population and mean age. In terms of the study population, fitness enthusiasts produced a larger effect size than athletes. The analysis suggests that athletes have better muscle mass and physical function than fitness enthusiasts due to long-term systematic training and that there could be a saturation effect on their physical function. This age group is in the adolescent stage, a sensitive period for neuromuscular development 59, and is at the peak of natural growth in both speed and explosive ability; therefore, the stimulatory response to dynamic stretching could be higher in this stage than in the adult stage.
(3) In terms of sample size, years of publication and quality of literature. In terms of sample size, samples with fewer than 30 cases were the most significant, consistent with the ease of good results in trials with fewer subjects found in the meta-analysis by Kang Yujie et al. 60. However, if the sample size is too small, potential errors due to random error factors cannot be excluded, regardless of whether the treatment effect survives as valid. Conversely, a sample size that is too large can result in a waste of resources. Therefore, the selection of the optimal sample size should consider both clinically significant and statistically significant differences in efficacy, as well as factors such as financial budget. In terms of the year of publication of the literature, the effect size of the studies between 2011 and 2015 was significant, consistent with the statement by Liu et al. 2 that "a large number of studies prior to 2016 demonstrated some facilitation of subsequent exercise performance by DS, after which many opposing views emerged that DS does not necessarily facilitate exercise performance". The reason for this outcome could be related to the relatively stable methodological quality of studies prior to 2016 61. In terms of the quality of the literature, only two of the 17 DS-related studies reported the use of blinding, which to some extent affects the quality of the literature, although a meta-analysis indicated that there was no significant correlation between the efficacy of physical methods and adequate blinding 62. However, adherence to the 'blinding principle' can improve the internal consistency of trials and reduce bias due to the expectations of subjects, intervention implementers or outcome evaluation.
Static stretching affects the effect
SS has become one of the most widely used warm-ups in sports due to its simplicity and controllability and low muscle damage 2. Although SS can significantly improve joint mobility 63, its effect on subsequent explosive power must be further confirmed 64. The results of this study showed that the effect size in terms of sprint performance was MD = 0.07 [95% CI: (0.002, 0.13)], which was significantly different from the control group, and the effect size in terms of CMJ performance was MD = -0.75 [95% CI: (-1.70, 0.18)], which was not significantly different from the control group, but from the SUCRA ranking graph, it can be seen that the percentage of its area under the curve was 15.6%, which was much smaller than that of the control group, indicating that SS was able to negatively affect subsequent explosive performance. The reason for the lack of significant differences between SS and controls in terms of CMJ performance could be that, compared to the complexity and coordination required for the short sprint, the measure of jumping ability is one dimensional; therefore, the effect of SS on relatively single-movement ability might be lower than that on relatively complex movement ability 65, and although the reverse vertical jump also requires a coordinated body effort, the time to complete the movement is relatively short, at least compared to the short sprint, and any change in the session might not have a significant effect on the outcome metrics.
Physiological and neurological studies have provided insight into the reasons for the negative impact of SS on explosive power. It has been found that prolonged static stretching of muscles affects the sensitivity of the muscle spindle (MS), which functions to encode information about the length of muscle extension as nerve impulses to the centre, reflexively generates and maintains muscle tension, and participates in the casual regulation of movement 66 through the coactivation of alpha-gamma motor neurons to ensure high sensitivity of the MS during muscle contraction 67. However, prolonged SS can cause deactivation of the sensitivity of the γ system in MS, resulting in the inability of MS to transmit the actual length of the muscle to the superior centre, reducing the number of motor units excited out of neuroprotective inhibition and ultimately leading to a decrease in explosive power 68. Another part of the study suggested that SS leads to a decrease in muscle-tendon unit (MTU) stiffness. The lower limb muscle-tendon union acts as a carrier of elastic energy storage and utilisation. In a state of constant muscle length, greater stiffness helps the muscle to generate more force during centripetal contraction 69. Therefore, it has been suggested that SS might reduce muscle length and tone 70, thus preventing the muscle from being in an activated state, in turn leading to a decrease in stiffness 71 and ultimately having a negative impact on explosive power.
Static combined with dynamic stretching to influence the effect
Since the strengths and weaknesses of SS are so obvious, it has been asked whether a combined approach of SS and DS could be used to take advantage of the improved joint mobility of SS while avoiding the detrimental effects of SS on explosive power through subsequent DS. The results of this study showed that, in terms of CMJ performance, the SDS effect size was MD = 1.80 [95% CI: (0.43, 3.20)], which was significantly different from the control group; the effect size in terms of sprint performance was MD = -0.06 [95% CI: (-0.16, 0.05)], suggesting that SDS can have some effect on subsequent CMJ performance. No results were available to demonstrate the effect of SDS on sprint performance. For the lack of significant differences between SDS and controls in terms of sprint performance, the analysis suggests that this difference could be related to the number of included papers. Of the 35 included papers, only three 15, 29, 42 examined the effect of SDS on sprint performance; in other words, the network comparison analysis was based on only these three papers, so the strength of the relevant evidence findings was significantly reduced. For example, only Chaouachi et al. 29 of the 3 papers showed no significant difference in SDS regarding subsequent sprint performance because the study was conducted with physical education students, and the trained population was less susceptible to the acute effects of stretching than the untrained population, as confirmed by Egan et al. 72.
Regarding the mechanism by which SDS can enhance explosive performance, the present study suggested that it could be related to the reactivation of γ motor neurons in MS. Although the sensitivity of γ-motor neurons is inactivated after SS, the subsequent stimulation of the muscle shuttle by DS could "reawaken" the sensitivity of γ-motor neurons, and muscle contraction reverts to the coactivation pattern of α-γ-motor neurons 68, thus exploiting the advantages of SS and DS while avoiding their disadvantages. However, since few studies have been conducted, the results must be interpreted with caution, and further evaluation is needed.
Other warm-up methods affect the results
In addition to the above warm-up methods, there are no results showing that foam axis rolling, PNF stretching and bouncy stretching can have effects on subsequent explosive power.
(1) Foam axis rolling. Also known as self-fascial relaxation, foam axis rolling is an emerging warm-up and relaxation technique in recent years. The practitioner uses self-weight to give the target muscle a certain amount of pressure to roll back and forth on the foam axis to improve the stretch and flexibility of the outer connective tissue of the muscle fibres 73, promote blood circulation, and increase myocyte oxygen and energy metabolism. The results of the present study remain largely consistent with previous studies and with a meta-analysis of FR by Wiewelhove et al. 49 showing that FR does not significantly affect subsequent explosive performance and that it is more suitable for postexercise relaxation than preexercise warm-up since it has been shown to be effective in relieving exercise-related muscle soreness. The few studies that have shown FR to significantly improve exercise performance suggested that FR could disrupt myofascial trigger-points (MTrP), which are nodules produced by transitional stress in skeletal muscle that can lead to muscle fatigue and stiffness 74, while Huang Haojie et al. 24 suggested that the stress produced by the foam axis on the muscle activated the Golgi Tdon organs (GTOs), which when active inhibit the muscle shuttle, causing a muscle relaxation response, a decrease in muscle tension, a decrease in muscle adhesion and an increase in muscle performance. Other studies have suggested that it could be related to the psychological factors of the participants since they believed that FR would improve their exercise performance 75.
(2) PNF stretching. The full name is proprioceptive neuromuscular facilitation (PNF), which first originated in the field of rehabilitation medicine for the treatment of diseases such as neuromuscular paralysis by activating the autonomic and cross-inhibitory effects of muscles to improve the function of specific muscles and later began to be widely used in the field of competitive sports, mostly used to prevent and treat sports injuries and improve joint mobility 76. However, the stretching process requires the assistance of a professional to apply the force and is time-consuming and could be more suitable for professional athletes from a simple and economic point of view. The results of the MeSH meta-analysis showed that PNF stretching had a negative impact on explosive power, although the difference was not statistically significant, consistent with previous studies. 77 Therefore, Bradley et al. 78 suggested that PNF stretching should not be performed prior to explosive sports. The mechanism of the effect of PNF stretching on explosive strength has been considered by most scholars to be the same as that of static stretching and can be explained by affecting the sensitivity of sarcolemmal receptors.
(3) Stretching with elastic shock. BS has been gradually marginalised due to its special stretching mechanism -- forcing the target muscle to elongate by means of rapid rebound, which most scholars believe causes a strong stretch reflex that in turn causes the muscle to contract to a shorter length than before the stretch and is therefore more likely to trigger muscle damage 79. The results of the present study suggested that BS has a positive effect on subsequent explosive performance, although this effect was not statistically significant. Mariscal et al. 80 suggested that BS, by stimulating neuromuscular activity, could activate the stretch-shortening cycle (SSC), thereby improving subsequent sprint performance. The outcome indicators included in this study, both CMJ in situ and short-distance sprinting, required centrifugal muscle elongation to store energy, followed by centripetal contraction to improve subsequent performance, which could, to some extent, explain why BS could have a positive effect on subsequent explosive performance.
Limitations of the study. (1) The search of the literature for this study did not include unpublished literature, and some literature was not included due to the absence of a control group or incomplete data about outcome indicators, which might have affected the comprehensiveness of the information to some extent. (2) The small number of static combined with dynamic stretching and foam axis rolling related literature included might have weakened the argument to some extent. (3) Due to the warm-up method intervention short duration and other peculiarities, most studies did not use randomised, controlled studies but their own before-and-after controlled studies to avoid the influence of individual differences on the study results.