Results

Search Results

Database searches returned a total of 5526 articles. After removal of duplicates, 2961 were checked for inclusion. Two studies reported on the same data set [30, 31]; thus, one [31] was excluded from the meta-analysis. The final sample of eligible papers consisted of 39 articles (Fig. 1).

Fig. 1: Flow diagram of the literature search and study inclusion process. Numbers in brackets indicate the number (n) of the initial search and the updated search, resulting in the overall number (n = initial search + updated search = overall number)

Characteristics of Included Studies

Overall, 34 of the 39 studies reported ACL injuries in a single sport (Table 1; Fig. 2). Of these, 15 focused on football (soccer) [17, 19–21, 23, 32–41], 6 on American Football and Australian Rules football or rugby (summed as contact-oriented football [COF]) [22, 26, 42–45], 9 on basketball [25, 27, 46–53], 2 on netball [18, 54], and 1 each on judo [55] and handball [24] (Table 1). The remaining five articles reported ACL injuries in multiple sports (Table 1) [30, 56–59]. The level of play varied from 1 (recreational) to 5 (world-class). At the same time, most studies (33 of 36) reported results of level 3 (highly trained/national level) or higher (Table 1). Three articles did not report on the expertise level (Table 1). A total of 33 studies reported data collection/observation duration, with a median duration of 9 years and a range of 1–47 years (Table 1). In addition, 23 studies included male athletes only, 10 included female athletes only, and 6 included both male and female athletes (Table 1).

Study | Sport | Country | Level of play [28] | Duration of data collection | Number of included ACL injuries (female/male) | ACL injuries
Contact | Indirect | Non-contact
Johnston et al. [43] | American Football | USA | 3–5 | 2013–2016 | 69 (0/69) | 19 | 34 | 16
Schick et al. [26] | American Football | USA | 3–5 | 2007–2016 | 53 (0/53) | 20 | 11 | 22
Vargas et al. [59] | American Football, basketball, football, Australian Rules football, baseball, rugby | USA, ESP, AUS | 3–5 | 2010–2017 | 26 (0/26) | N/A | 0 | 26
Boden et al. [57] | American Football, basketball, football, netball | USA | 2–5 | N/A | 23 (7/16) | 8 | N/A | 15
Cochrane et al. [42] | Australian Rules football | AUS | 3–5 | 1992–1998 | 34 (0/34) | 11 | 4 | 19
Rolley et al. [45] | Australian Rules football | AUS | 3–5 | 2016–2020 | 21 (21/0) | N/A | 8 | 13
Axelrod et al. [46] | Basketball | USA | 3–5 | 1997–2019 | 10 (10/0) | 1 | N/A | N/A
Gill et al. [47] | Basketball | USA | 3–5 | 2006–2022 | 38 (0/38) | N/A | 29 | 9
Krosshaug et al. [48] | Basketball | USA | 2–5 | N/A | 39 (22/17) | 4 | 7 | 28
Petway et al. [25] | Basketball | USA | 4–5 | 1975–2022 | 35 (0/35) | 9 | 21 | 5
Saito et al. [49] | Basketball | USA | 4–5 | 2011–2022 | 27 (0/27) | 0 | 11 | 15
Tosarelli et al. [50] | Basketball | EU | 3–5 | 2013–2020 | 37 (0/37) | 1 | 21 | 14
Heder Ternell et al. [51] | Basketball | FRA, GER, ITA, ESP, SWE | 3–5 | 2018–2023 | 41(41/0) | 0 | 23 | 18
Costello et al. [52] | Basketball | USA | 3–5 | 2006–2022 | 31 (0/31) | 2 | 17 | 12
Hurley et al. [53] | Basketball | USA | 3–5 | 2009–2020 | 23 (0/23) | 0 | 13 | 10
Sheehan et al. [58] | Basketball, football, American Football, handball | N/A | N/A | N/A | 20 (13/7) | N/A | 0 | 20
Boden et al. [56] | Basketball, handball, football, American Football, cheerleading, gymnastics | USA | 3–4 | 1995–2007 | 29 (18/11) | N/A | 8 | 21
Achenbach et al. [17] | Football | GER | 4–5 | 2016–2017 | 37 (37/0) | 6 | 14 | 17
Brophy et al. [32] | Football | N/A | 1–4 | N/A | 55 (23/32) | 31 | N/A | N/A
D’Hooghe et al. [19] | Football | QAT, ITA | 3–5 | 2014–2018 | 19 (0/19) | 4 | 6 | 9
De Carli et al. [20] | Football | ITA, FRA, ESP, GER, UK | 3–5 | 2010–2020 | 128 (0/128) | 36 | 36 | 50
Della Villa et al. [21] | Football | ITA | 3–5 | 2008–2018 | 134 (0/134) | 16 | 59 | 59
Grassi et al. [33] | Football | Worldwide | N/A | 1981–2015 | 34 (0/34) | 12 | 7 | 15
Grassi et al. [34] | Football | ITA | 3–5 | N/A | 21 (0/21) | N/A | 5 | 16
Lucarno et al. [23] | Football | USA, GER, FRA, UK, ESP, ITA | 3–5 | 2017–2020 | 35 (35/0) | 4 | 12 | 19
Rekik et al. [35] | Football | QAT | 3–4 | 2013–2019 | 15 (0/15) | 3 | 4 | 8
Waldén et al. [36] | Football | EU | 3–5 | 2001–2011 | 39 (0/39) | 6 | 8 | 25
Buckthorpe et al. [37] | Football | Spain | 3–5 | 2010–2022 | 115 (0/115) | 16 | 49 | 50
Della Villa et al. [38] | Football | England | 3–5 | 2010–2021 | 124 (0/124) | 24 | 52 | 47
Ranzini et al. [39] | Football | UK, FRA, GER, ESP, International | 3–5 | 2020–2023 | 27 (0/27) | N/A | 9 | 18
Zago et al. [40] | Football | UK, FRA, GER, ESP, International | 3–5 | 2020–2022 | 33 (33/0) | N/A | 9 | 24
Gokeler et al. [41] | Football | Italy | 3–5 | 2008–2018 | 47 (0/47) | N/A | N/A | 47
Olsen et al. [24] | Handball | NOR | 3–4 | 1988–2000 | 20 (20/0) | 1 | 6 | 13
Koga et al. [30] | Handball, basketball | N/A | N/A | N/A | 10 (10/0) | N/A | 6 | 4
Akoto et al. [55] | Judo | EU | 3–4 | 2010–2017 | 17 (8/9) | 11 | 6 | 0
Belcher et al. [18] | Netball | AUS, NZL | 3–4 | 2011–2019 | 21 (21/0) | N/A | 7 | 14
Stuelcken et al. [54] | Netball | AUS, NZL | 3–5 | 2008–2015 | 16 (16/0) | N/A | 8 | 8
Della Villa et al. [22] | Rugby | Worldwide | 3–5 | 2015–2019 | 57 (0/57) | 18 | 15 | 24
Montgomery et al. [44] | Rugby | N/A | 3–5 | 2014–2015 | 35 (0/35) | 10 | 8 | 15

Fig. 2: Pie charts (scaled to the number of ACL injuries) displaying the Quality Appraisal for Sports Injury Video Analysis Studies (QA-SIVAS) score of the included studies, stratified by the type of sport

Methodological Quality Characteristics

The quality of the studies was high (81–100%) in 2 studies, good (71–80%) in 17, moderate (60–70%) in 10, and low (< 60%) in 10 (Fig. 2, Table 2). QA-SIVAS scores ranged from 33 to 83%, with an average quality of 66% across all studies (Table 2). All articles clearly stated the study’s objectives. A total of 18 studies used video recordings from a representative sample, 8 provided sample information, and 4 specified video source/quality information. Detailed methodologies were described in 33 articles, and 35 reported systematic video analysis approaches (Table 2). Medical report information was included in 15 articles, while raters’ background/expertise was noted in 20. In 32 articles, multiple researchers evaluated the recordings. Three and two articles reported a control group and validated methods for quantitative biomechanical analysis, respectively. The main results were clearly presented in 38 articles, with clear injury case reporting in 37. The injury context was assessed in 37 articles, and 29 provided example screenshots or video frames. All articles discussed results in the context of current literature, with 36 addressing clinical/practical implications and all discussing study limitations.

 | Objective stated | A representative sample was chosen | Information about sample is included | Information about video source and quality of the footage are included | Applied methods are described comprehensively | A systematic approach to video analysis was chosen | Medical report information are included | Background/expertise of raters is stated | Findings are observed by more than one researcher | A control group is included | A quantitative biomechanical analysis was conducted using validated methods | The main results of the study are clearly described | Absolute numbers or proportions of injury cases for each/the main outcome are reported | Details about the injury context are included | Example screenshots/video frames are included | Findings are discussed within the context of the current evidence | Clinical/practical implications of the results are discussed | Limitations of the study are addressed | Score (%)
Achenbach et al. [17] | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 83
Akoto et al. [55] | 1 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 72
Axelrod et al. [46] | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 1 | 0 | 1 | 44
Belcher et al. [18] | 1 | 0 | 0 | 0 | 1 | 1 | 0 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 67
Boden et al. [57] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 1 | 1 | 1 | 33
Boden et al. [56] | 1 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 1 | 0 | 1 | 0 | 1 | 1 | 1 | 0 | 1 | 50
Brophy et al. [32] | 1 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 61
Buckthorpe et al. [37] | 1 | 1 | 0 | 0 | 1 | 1 | 0 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 72
Cochrane et al. [42] | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 61
Costello et al. [52] | 1 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 56
D'Hooghe et al. [19] | 1 | 0 | 1 | 0 | 1 | 1 | 1 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 72
De Carli et al. [20] | 1 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 56
Della Villa et al. [21] | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 78
Della Villa et al. [22] | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 78
Della Villa et al. [38] | 1 | 1 | 0 | 0 | 1 | 1 | 0 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 72
Gill et al. [47] | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 78
Gokeler et al. [41] | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 78
Grassi et al. [33] | 1 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 61
Grassi et al. [34] | 1 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 61
Heder Ternell et al. [51] | 1 | 1 | 1 | 0 | 1 | 1 | 0 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 78
Hurley et al. [53] | 1 | 1 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 56
Johnston et al. [43] | 1 | 1 | 0 | 0 | 1 | 1 | 0 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 67
Koga et al. [30] | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 61
Krosshaug et al. [48] | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 72
Lucarno et al. [23] | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 78
Montgomery et al. [44] | 1 | 1 | 0 | 0 | 1 | 1 | 0 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 78
Olsen et al. [24] | 1 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 72
Petway et al. [25] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 44
Ranzini et al. [39] | 1 | 0 | 0 | 0 | 1 | 1 | 1 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 67
Rekik et al. [35] | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 78
Rolley et al. [45] | 1 | 1 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 56
Saito et al. [49] | 1 | 0 | 1 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 61
Schick et al. [26] | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 50
Sheehan et al. [58] | 1 | 0 | 0 | 0 | 1 | 1 | 1 | 0 | 0 | 1 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 56
Stuelcken et al. [54] | 1 | 1 | 0 | 0 | 1 | 1 | 0 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 72
Toserelli et al. [50] | 1 | 0 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 78
Vargas et al. [59] | 1 | 0 | 0 | 0 | 1 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 61
Waldén et al. [36] | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 83
Zago et al. [40] | 1 | 1 | 1 | 0 | 1 | 1 | 1 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 78
Sum (%) | 100 | 46 | 21 | 10 | 85 | 90 | 38 | 51 | 82 | 8 | 5 | 97 | 95 | 95 | 74 | 100 | 92 | 100 | 66

Injury Characteristics

Collectively, the 39 primary studies investigated 1551 video-recorded ACL injuries. In addition, 20.8% of the investigated injuries in the studies were sustained by women, while 79.2% were sustained by men. Most injuries were classified as noncontact (n = 745), followed by indirect contact (n = 533) and direct contact (n = 273; Table 3). The precise relative shares among injury classifications could not be calculated, as some studies only reported pooled injury counts (e.g., for indirect plus noncontact injuries). Moderator analysis revealed a significant impact of the type of sport on direct contact proportions (p < 0.05; Table 3), which were more common in judo (64.7%; number of studies (n) = 1) than in COF (32.1%; 95% CI 26.4–37.9%; n = 5), football (22.5%; 95% CI 15.6–30.3%; n = 11), basketball (6.1%; 95% CI 1.7–12.8%; n = 8), or handball (5%; n = 1). Indirect contact ACL injury proportions were statistically significantly moderated by the type of sports, with 48.9% in basketball (95% CI 36.7–61.2; n = 7), showing higher rates than in judo (35.3%; n = 1), football 35.0% (95% CI 29.6–40.7%; n = 10), handball 30% (n = 1), or COF 26.8% (95% CI 15.4–40.0%; n = 5). The type of sport was not significantly associated with the occurrence of noncontact injuries (p > 0.05).

 | Estimated proportion (%) | p | 95% CI (%) | df | I2 (%) | Moderator p value | n
Non contact ACL injuries | 44.0 | 0.038 | (38.5–49.7) | 24 | 72.9 | 0.152 | 24
Indirect contact ACL injuries | 36.4 | < 0.001 | (31.1–41.8) | 23 | 71.6 | 0.017 | 24
Direct contact ACL injuries | 19.5 | < 0.001 | (14.2–25.5) | 26 | 84.2 | < 0.001 | 26
Non contact and indirect contact ACL injuries pooled | 80.5 | < 0.001 | (74.5–85.8) | 26 | 84.2 | < 0.001 | 26
Indirect contact and contact ACL injuries pooled | 56.7 | 0.020 | (51.0–62.3) | 23 | 72.6 | 0.152 | 24

Movement Patterns

When combining all ACL injuries (noncontact, indirect contact, and direct contact), the highest proportions occurred during ball actions (45.7%), pressing/tackling (40.9%), and cutting (36.6%, Table 4). The moderator analysis yielded significant effects for landing in general (p < 0.05). ACL injuries during landing (single and double leg landing combined) were more frequent in netball (81.2%, n = 2) than in basketball (32.5%; 95% CI 21.8–44.3%; n = 9), handball (20.0%; n = 1), COF (16.9%; 95% CI 5.9–39.5%; n = 5), or football (12.8%; 95% CI: 7.3–19.6%; n = 9). The moderator analysis yielded significant effects for pressing/tackling situations (p < 0.05). Injuries in soccer (44.3%; 95% CI 39.7–48.8%; n = 13) were more frequent than in COF (17.5%; n = 2).

 | Estimated proportion (%) | p value | 95% CI (%) | df | I2 (%) | Moderator p value | n
Movement patterns/players action |  |  |  |  |  |  | 
All ACL injuries pooled |  |  |  |  |  |  | 
Landing | 23.4 | < 0.001 | (17.5–29.8) | 29 | 83.4 | 0.016 | 25
Single-leg landing | 21.7 | < 0.001 | (14.0–30.5) | 15 | 76.9 | 0.287 | 11
Double-leg landing | 9.9 | < 0.001 | (3.1–20.0) | 13 | 87.8 | 0.142 | 10
Cutting | 36.6 | 0.015 | (26.5–47.4) | 23 | 90.8 | 0.896 | 21
Deceleration | 31.0 | 0.006 | (18.9–44.6) | 14 | 91.0 | 0.313 | 11
Acceleration | 10.4 | < 0.001 | (2.6–22.5) | 4 | 82.2 | 0.355 | 4
Ball action | 45.7 | < 0.001 | (31.0–60.9) | 15 | 93.8 | 0.099 | 15
Being tackled | 20.1 | < 0.001 | (15.4–25.2) | 13 | 62.7 | 0.075 | 14
Kicking | 9.3 | < 0.001 | (4.3–15.9) | 4 | 51.2 | 0.345 | 5
Heading | 7.4 | < 0.001 | (3.0–13.6) | 3 | 27.3 |  | 
Passing | 13.8 | < 0.001 | (3.1–30.6) | 5 | 90.3 | 0.348 | 6
Pressing/ Tackling | 40.9 | < 0.001 | (34.8–47.1) | 14 | 66.6 | 0.002 | 15
Ball protection | 9.6 | < 0.001 | (6.2–13.5) | 5 | 0.0 | 0.953 | 4
Non-contact and indirect contact ACL injuries pooled |  |  |  |  |  |  | 
Landing | 23.8 | < 0.001 | (17.6–30.6) | 28 | 83.1 | 0.036 | 25
Single-leg landing | 20.8 | < 0.001 | (13.6–29.0) | 12 | 64.6 | 0.763 | 10
Double-leg landing | 8.9 | < 0.001 | (2.6–18.6) | 10 | 82.5 | 0.114 | 8
Cutting | 38.0 | < 0.001 | (25.3–51.6) | 22 | 93.2 | 0.360 | 20
Deceleration | 28.8 | 0.012 | (14.8–45.2) | 11 | 88.1 | 0.212 | 10
Ball action | 34.8 | < 0.001 | (22.6–48.2) | 16 | 92.0 | 0.004 | 15
Being tackled | 21.1 | < 0.001 | (17.9–24.5) | 12 | 55.2 | 0.101 | 13
Pressing/ Tackling | 41.8 | 0.017 | (35.2–48.5) | 13 | 61.2 | 0.002 | 14
Indirect contact and contact ACL injuries pooled |  |  |  |  |  |  | 
Landing | 22.8 | 0.007 | (8.1–42.2) | 12 | 80.1 | 0.223 | 9
Single-leg landing | 18.3 | < 0.001 | (6.6–34.2) | 8 | 66.1 | 0.420 | 6
Double-leg landing | 5.8 | 0.076 | (1.5–12.7) | 7 | 22.8 | 0.175 | 5
Cutting | 12.7 | < 0.001 | (4.8–23.6) | 14 | 72.3 | 0.176 | 12
Deceleration | 12.5 | < 0.001 | (2.8–27.6) | 8 | 74.2 | 0.355 | 7
Ball action | 23.1 | 0.044 | (4.9–49.2) | 5 | 87.2 | 0.185 | 5
Being tackled | 37.2 | 0.048 | (25.4–49.9) | 10 | 80.8 | 0.856 | 11
Pressing/ Tackling | 22.3 | < 0.001 | (13.9–32.1) | 13 | 77.6 | 0.795 | 14

Noncontact injuries were most frequent during cutting (53.8%), followed by pressing/tackling (50.2%), decelerating (38.9%), and landing (30.1%, Table 5). The moderator analysis revealed that noncontact ACL injuries during pressing/tackling and deceleration situations were significantly affected by the type of sports (p < 0.05). For pressing/tackling situations, higher frequencies were reported in soccer (55.8%; 95% CI 49.8–61.8%; n = 9) than in COF, where one study [44] explicitly reported zero noncontact injuries. Noncontact injuries during deceleration were more frequent in netball (58.0%, n = 2) than in football (35.3%, n = 1), COF (8.6%, n = 2), or handball (7.7%; n = 1). No association was detected between the type of sport and ACL injury during other movement patterns (p > 0.05).

 | Estimated proportion (%) | p value | 95%CI (%) | df | I2 (%) | Moderator p value | n
Movement patterns/player’s action |  |  |  |  |  |  | 
Non-contact ACL injuries |  |  |  |  |  |  | 
Landing | 30.1 | 0.048 | (18.0–43.8) | 13 | 79.0 | 0.442 | 10
Single-leg landing | 19.7 | < 0.001 | (10.2–31.5) | 7 | 51.9 | 0.302 | 5
Double-leg landing | 8.8 | < 0.001 | (0.3–27.1) | 7 | 84.3 | 0.240 | 4
Cutting | 53.8 | < 0.001 | (40.4–67.0) | 13 | 74.8 | 0.132 | 11
Deceleration | 38.9 | < 0.001 | (18.4–61.2) | 7 | 85.0 | 0.004 | 6
Ball possesion | 23.4 | < 0.001 | (11.9–37.4) | 4 | 46.8 | 0.096 | 4
Being tackled | 1.7 | < 0.001 | (0.4–3.9) | 8 | 0 | 0.965 | 9
Pressing/tackling | 50.2 | < 0.001 | (37.7–62.6) | 9 | 75.9 | 0.001 | 10
Indirect contact ACL injuries |  |  |  |  |  |  | 
Landing | 30.0 | < 0.001 | (12.3–51.6) | 11 | 73.6 | 0.288 | 9
Single-leg landing | 25.7 | 0.006 | (11.8–42.7) | 7 | 50.0 | 0.416 | 6
Double-leg landing | 8.3 | < 0.001 | (2.5–17.3) | 6 | 17.9 | 0.187 | 5
Cutting | 22.5 | 0.003 | (8.7–40.5) | 12 | 73.5 | 0.199 | 11
Deceleration | 25.5 | 0.040 | (7.9–48.9) | 6 | 63.5 | 0.567 | 6
Ball possesion | 44.8 | 0.003 | (22.5–68.3) | 4 | 64.1 | 0.174 | 4
Being tackled | 56.1 | < 0.001 | (39.8–71.6) | 8 | 73.0 | 0.029 | 9
Pressing/tackling | 24.8 | < 0.001 | (12.8–39.3) | 8 | 70.5 | 0.537 | 9
Contact ACL injuries |  |  |  |  |  |  | 
Being tackled | 23.9 | 0.031 | (6.6–47.6) | 7 | 85.3 | 0.629 | 8
Pressing/tackling | 24.2 | 0.011 | (9.1–43.7) | 10 | 83.6 | 0.855 | 11
Movement Speed |  |  |  |  |  |  | 
Non-contact ACL injuries |  |  |  |  |  |  | 
Horizontal speed high | 70.7 | < 0.001 | (59.6–80.6) | 4 | 2.3 | 0.441 | 5
Horizontal speed low | 20.9 | < 0.001 | (11.5–32.2) | 3 | 0.0 |  | 
Horizontal speed zero | 4.4 | < 0.001 | (0.3–13.0) | 2 | 0.0 |  | 
Indirect contact ACL injuries |  |  |  |  |  |  | 
Horizontal speed high | 36.1 | 0.012 | (6.3–74.0) | 4 | 81.9 | 0.325 | 5
Horizontal speed low | 9.1 | 0.036 | (0.7–25.3) | 3 | 23.8 |  | 
Horizontal speed zero | 7.1 | 0.317 | (0.3–21.4) | 2 | 0.0 |  | 
Vertical speed zero | 50.8 | 0.041 | (2.5–98.1) | 2 | 89.4 | 0.225 | 3
Game situations |  |  |  |  |  |  | 
Non-contact ACL injuries |  |  |  |  |  |  | 
Offensive | 58.5 | < 0.001 | (35.7–79.6) | 5 | 84.7 | 0.376 | 5
Defensive | 37.6 | 0.004 | (16.3–61.8) | 5 | 86.2 | 0.476 | 5
Indirect contact ACL injuries |  |  |  |  |  |  | 
Offensive | 22.8 | 0.081 | (3.1–53.7) | 5 | 64.9 |  | 
Defensive | 42.7 | 0.007 | (8.0–77.9) | 5 | 72.5 |  | 
Contact ACL injuries |  |  |  |  |  |  | 
Offensive | 52.9 | < 0.001 | (43.2–62.4) | 6 | 0.0 | 0.175 | 7
Defensive | 46.1 | < 0.001 | (36.6–55.7) | 6 | 0.0 | 0.089 | 7

Indirect contact injuries were most frequent while being tackled (56.1%), followed by situations with ball possession (44.8%) and landing (single- and double-leg landing combined, 30.0%, Table 5). For injuries occurring while being tackled, the moderator analysis revealed a significant effect of the type of sport, with higher frequencies reported in COF (100%; eight indirect injuries, all of which occurred were while being tackled; n = 1 [44]) than in soccer (49.1%, 95% CI 36.1–62.3%; n = 8). No association was detected between the type of sport and any other movement pattern for indirect contact injuries.

For contact ACL injuries, injury frequencies while being tackled (23.9%) were similar to those during pressing/tackling (24.2%, Table 5) situations. Moderator analysis revealed no differences between sports. The supplemental digital content provides pooled estimates for combinations of direct and indirect contact as well as indirect and noncontact injuries.

Movement Velocity

With all injuries combined, the highest summary proportions were observed for situations with high (53.8%) and low horizontal velocity (34.8%), while injuries at no velocity (6.2%) were scarce (Table 6). For vertical velocity, an opposite pattern was observed, with a higher proportion of injuries sustained at no vertical (63.5%) versus low (19.0%) or high vertical velocity (11.5%). Moderator analyses revealed a significant impact of sports (p < 0.05): Injuries with high vertical velocity were highest for netball (75.0%; n = 1) but comparably low for basketball (19.6%; n = 2), COF (6.4%; 95% CI 3.0–10.9%; n = 3), or football (6.0%; 95% CI 4.0–8.4%; n = 5).

 | Estimated proportion (%) | p | 95% CI (%) | df | I2 (%) | Moderator p value | n
Movement Speed |  |  |  |  |  |  | 
All ACL injuries pooled |  |  |  |  |  |  | 
Horizontal speed high | 53.8 | < 0.001 | (43.6–63.9) | 14 | 88.7 | 0.696 | 15
Horizontal speed low | 34.8 | 0.007 | (24.5–45.9) | 15 | 91.1 | 0.627 | 16
Horizontal speed zero | 6.2 | < 0.001 | (4.3–8.4) | 12 | 29.0 | 0.298 | 13
Vertical speed high | 11.5 | < 0.001 | (6.3–18.1) | 10 | 81.5 | 0.017 | 11
Vertical speed low | 19.0 | < 0.001 | (10.3–29.6) | 9 | 89.6 | 0.145 | 10
Vertical speed zero | 63.5 | 0.029 | (51.4–74.8) | 10 | 89.7 | 0.057 | 11
Non-contact and indirect contact ACL injuries pooled |  |  |  |  |  |  | 
Horizontal speed high | 53.0 | < 0.001 | (37.8–67.9) | 9 | 88.5 | 0.874 | 10
Horizontal speed low | 41.6 | < 0.001 | (27.6–56.3) | 10 | 88.7 | 0.291 | 11
Horizontal speed zero | 5.1 | < 0.001 | (3.1–7.6) | 8 | 0.0 | 0.087 | 9
Vertical speed high | 17.1 | < 0.001 | (7.3–30.0) | 7 | 87.3 | 0.071 | 8
Vertical speed low | 25.9 | 0.001 | (13.4–40.8) | 5 | 85.9 | 0.044 | 6
Vertical speed zero | 56.4 | < 0.001 | (39.1–72.9) | 6 | 89.2 | 0.084 | 7
Indirect contact and contact ACL injuries pooled |  |  |  |  |  |  | 
Horizontal speed high | 28.3 | 0.038 | (2.4–67.6) | 4 | 90.1 | 0.330 | 5
Horizontal speed low | 7.1 | < 0.001 | (0.4–20.8) | 4 | 53.1 | 0.331 | 4
Horizontal speed zero | 6.3 | < 0.001 | (0.0–22.3) | 3 | 49.7 |  | 
Vertical speed zero | 52.1 | 0.957 | (0.2–99.9) | 2 | 95.1 |  | 

When considering noncontact injuries only, most injuries occurred at high horizontal velocity (70.7%). Indirect injuries also occurred most often at high horizontal velocity, although the proportion was lower (36.1%, Table 5). No pooling was possible for direct contact injuries.

Moderator analysis for noncontact and indirect contact injuries did not reveal any statistically significant effects. No moderator analysis was possible for the remaining movement speeds, or for noncontact, indirect, or direct contact injuries owing to insufficient data.

Game Situation

Considering all injuries, the highest ACL injury proportions were reported for situations with ball possession (67.5%) and offensive game situations (55.4%, Table 7). Concerning the time point of injury, the distributions across quarters were relatively similar (Table 7). Moderator analyses revealed no impact of the type of sport with regards to ball possession and tactical situation (p > 0.05) but an impact in terms of the time point of injury. The highest number of injuries during the first quarter were sustained in netball (37.8%; n = 1) and football (37.5%; n = 2), followed by COF (25.9%, 95% CI 15.4–41.8%; n = 3) and basketball (16.0%; n = 2).

 | Estimated proportion (%) | p | 95% CI (%) | df | I2 (%) | Moderator p value | n
Game situations |  |  |  |  |  |  | 
All ACL injuries pooled |  |  |  |  |  |  | 
Offensive | 55.4 | < 0.001 | (46.6–64.1) | 25 | 88.9 | 0.137 | 24
Defensive | 40.4 | 0.037 | (31.7–49.4) | 25 | 89.5 | 0.062 | 24
Quarter 1 | 28.8 | < 0.001 | (21.8–35.8) | 8 | 56.3 | 0.019 | 8
Quarter 2 | 24.8 | < 0.001 | (16.4–33.1) | 8 | 76.1 | 0.074 | 8
Quarter 3 | 20.3 | < 0.001 | (13.1–27.5) | 8 | 70.8 | 0.086 | 8
Quarter 4 | 29.0 | 0.002 | (10.7–47.3) | 8 | 96.3 | 0.261 | 8
Ball possesion | 67.5 | < 0.001 | (52.7–80.6) | 9 | 90.2 | 0.292 | 9
Non-contact and indirect contact ACL injuries pooled |  |  |  |  |  |  | 
Offensive | 51.7 | < 0.001 | (41.4–60.9) | 17 | 83.7 | 0.059 | 17
Defensive | 38.1 | < 0.001 | (26.9–50.0) | 18 | 89.8 | 0.030 | 17
Quarter 1 | 26.4 | < 0.001 | (18.3–34.5) | 7 | 38.3 | 0.021 | 7
Quarter 2 | 26.7 | < 0.001 | (14.9–38.4) | 7 | 76.5 | 0.084 | 7
Quarter 3 | 15.2 | < 0.001 | (6.9–23.5) | 7 | 67.9 | 0.326 | 7
Quarter 4 | 32.8 | 0.006 | (9.5–56.2) | 7 | 95.5 | 0.399 | 7
Ball possesion | 47.0 | < 0.001 | (29.8–64.6) | 8 | 76.7 | 0.277 | 6
Indirect contact and contact ACL injuries pooled |  |  |  |  |  |  | 
Offensive | 24.9 | < 0.001 | (15.2–36.2) | 8 | 72.8 | 0.626 | 8
Defensive | 20.5 | < 0.001 | (13.4–28.9) | 8 | 56.5 | 0.252 | 8
Quarter 1 | 26.6 | < 0.001 | (15.4–37.8) | 2 | 0.0 |  | 
Quarter 2 | 17.7 | < 0.001 | (8.0–27.3) | 2 | 0.0 |  | 
Quarter 3 | 14.1 | 0.013 | (3.0–25.3) | 2 | 26.2 |  | 
Quarter 4 | 12.9 | 0.111 | (0.0–28.7) | 2 | 61.1 |  | 

Noncontact injuries (58.5%) mainly occurred during offensive play, while indirect contact injuries had the highest frequency during defensive play (42.7%; Table 5). Pooling of estimates was either not possible or revealed no significant effect.