Inciting Events of Lower-Limb Muscle Injury: A Systematic Review and Meta-analysis of Video-Based Studies
Sports medicine - open · 2026
Football hamstring injuries are mostly non-contact and high-speed, but without exposure data this review can't tell you which actions are actually riskiest.
De studie
Systematic review and meta-analysis of 12 video-based observational studies, 513 professional footballers (434 male, 79 female), 538 lower-limb muscle injuries.
Wat ze vonden
Non-contact mechanisms were far more common than indirect contact (OR 8.01, 95% CI 3.85 to 16.64, p<0.001), and injuries clustered at moderate to high horizontal running speed versus low speed (OR 19.56, 95% CI 2.55 to 150.12, p=0.004). Hamstrings accounted for 74.7% of injuries, occurring most often during running (29%), stretching/reaching (24%) or kicking (22%), though these three did not credibly differ from each other.
Studieopzet
Systematic review and meta-analysis pooling observational video-analysis studies of lower-limb muscle injuries, almost entirely in men's professional football, with a small parallel female dataset (79 players, 86 injuries) too limited to pool. This sits above a single video-analysis cohort but is still built on retrospective, non-randomised observational data, so it can describe association and pattern, not cause.
Methodologie
Each included study relied on retrospective visual classification of injury mechanism from match or training footage, a method vulnerable to detection and misclassification bias since 'non-contact' versus 'indirect contact' and speed category are judgment calls made without access to biomechanical ground truth, and inter-study consistency in how reviewers coded these categories is unclear. Pooling across studies also assumes comparable video quality, camera angles and coding definitions, which is rarely fully true and can inflate or dilute the pooled ORs, most visible in the very wide CI around the speed finding (2.55 to 150.12). Because eligible studies needed usable broadcast-quality footage, the sample skews toward elite, well-televised men's leagues, a selection bias that further explains why female data couldn't be pooled and why these findings shouldn't be generalised beyond male professional football.
De beoordeling
The non-contact association is precise and statistically solid (OR 8.01, 95% CI 3.85-16.64). The high-speed association is significant but imprecise (OR 19.56, 95% CI 2.55-150.12, a range wide enough that the true effect could be modest or huge), so treat that one as a directional signal rather than a firm number. Both confirm what hamstring injury mechanism work has suggested for two decades, and both describe men's football only, sufficient data for meta-analysis existed only for male players. The finding that running, stretching/reaching and kicking are statistically indistinguishable as injury contexts is arguably the more useful (if less headline-grabbing) result, because it undercuts any claim that one specific action is the dominant target for prevention. Without exposure data this remains descriptive epidemiology, not a basis for causal claims about relative risk of specific actions.
De beperking
No exposure-adjusted denominators, and no pooled female data. We know what players were doing when injured, not how often they do those things relative to injury-free play, so you cannot conclude that sprinting or reaching is 'riskier' than any other action, only that it's common when injuries happen. The review included 79 female players and 86 female injuries, but the authors state there was insufficient data to meta-analyse female outcomes, so every OR and pooled proportion above describes men's professional football only.
Referentiestudie
This extends the tradition of single-cohort video mechanism studies in professional football, most notably the English FA's video audit of hamstring injuries by Woods and colleagues in the early 2000s, by pooling across multiple leagues and studies rather than relying on one dataset. It sits alongside, but does not replace, the large-scale epidemiological surveillance work (e.g. the UEFA Elite Club Injury Study) that quantifies incidence and burden rather than mechanism.
Wat te doen op maandag
This doesn't hand you a new intervention for Monday, and it speaks to men's professional football specifically, female data was insufficient to analyse. It reinforces that hamstring injury prevention programming should keep covering high-speed running, deceleration mechanics and reaching/overstretch actions (including kicking follow-through) as roughly equal-weight contexts, rather than assuming one specific match action is the dominant culprit.
In practice
This applies to male professional (and by extension, high-sprint-exposure sub-elite) footballers with a hamstring or adductor strain history, that's the population this data actually describes, don't extrapolate confidently to female or amateur players. In clinic, don't let your mechanism history default to 'sprint story every time': ask specifically about sprinting, reaching/overstretch and kicking follow-through, since this review shows all three produce injuries at similar rates, so a kicking-leg strain deserves the same workup rigour as a sprint one. On the S&C floor, this is a reason to keep hamstring prevention programming broad rather than narrowing to Nordics and top-speed sprint exposure alone, keep eccentric strength, deceleration mechanics and kicking-leg range/control all in the program, since nothing here tells you sprinting is a bigger threat than kicking, only that both show up. No exposure data means none of this changes return-to-sport criteria or dosing, it just argues against over-indexing prevention work on any single mechanism.
De vijf studies die ertoe doen — wekelijks, gratis.
Wij lezen de hele literatuur over sportgeneeskunde, revalidatie en prestatie en beoordelen de studies die je tijd waard zijn.
Ontvang de gratis wekelijkse editieGerelateerde beoordelingen
Lees in je taal: English · Français · Deutsch · Español · Italiano · Nederlands · Dansk · Svenska