Team Physician’s Corner
Anterior Cruciate Ligament and Knee Injury Prevention Programs for Soccer Players
M
A Systematic Review and Meta-analysis Nathan L. Grimm,*y MD, John C. Jacobs Jr,z BS, Jaewhan Kim,§ PhD, Brandon S. Denney,z BS, and Kevin G. Shea,|| MD Investigation performed at the Duke University Medical Center, Durham, North Carolina, USA Background: Soccer has one of the highest incidences of anterior cruciate ligament (ACL) injuries for both males and females. Several injury prevention programs have been developed to address this concern. However, an analysis of the pooled effect has yet to be elicited. Purpose: To conduct a systematic review and meta-analysis of ACL and knee injury prevention programs for soccer players, assess the heterogeneity among the studies, and evaluate the reported effectiveness of the prevention programs. Study Design: Systematic review and meta-analysis. Methods: A systematic search of the literature was conducted on PubMed (Medline), Embase, CINAHL, and Central-Cochrane Database. Studies were limited to randomized controlled trials (RCTs) of injury prevention programs specific to the knee and/or ACL in soccer players. The Cochrane Q test and I2 index were independently used to assess heterogeneity among the studies. The pooled risk difference, assessing knee and/or ACL injury rates between intervention and control groups, was calculated by random-effects models with use of the DerSimonian-Laird method. Publication bias was assessed with a funnel plot and Egger weighted regression technique. Results: Nine studies met the inclusion criteria as RCTs. A total of 11,562 athletes were included, of whom 7889 were analyzed for ACL-specific injuries. Moderate heterogeneity was found among studies of knee injury prevention (P = .041); however, there was insignificant variation found among studies of ACL injury prevention programs (P = .222). For studies of knee injury prevention programs, the risk ratio was 0.74 (95% CI, 0.55-0.89), and a significant reduction in risk of knee injury was found in the prevention group (P = .039). For studies of ACL injury prevention programs, the risk ratio was 0.66 (95% CI, 0.33-1.32), and a nonsignificant reduction in risk of ACL injury was found in the prevention group (P = .238). No evidence of publication bias was found among studies of either knee or ACL injury prevention programs. Conclusion: This systematic review and meta-analysis of ACL and knee injury prevention program studies found a statistically significant reduction in injury risk for knee injuries but did not find a statistically significant reduction of ACL injuries. Keywords: anterior cruciate ligament injury prevention; knee injury; injury prevention; meta-analysis; systematic review
Soccer is the most popular sport in the world, with approximately 265 million players as of 2006.1 The sport has seen tremendous growth over the past decade in the United States, especially in the youth and female populations, which has led to an associated increase in injuries sustained by soccer players.2,11 The lower extremity represents a majority of soccer injuries in both male and female athletes, with the knee being a frequently injured body part.8,36,46 In particular, female athletes have a 3 to 5 times higher risk of serious knee injury compared with male athletes for soccer, basketball, volleyball, and other sports.3-5,11,37 Many intervention programs have been designed to reduce the risk of injury to the knee and anterior
*Address correspondence to Nathan L. Grimm, MD, Department of Orthopaedic Surgery, Duke University Medical Center, 8 Intuition Circle, Durham, NC 27705, USA (e-mail: [email protected]). y Department of Orthopaedic Surgery, Duke University Medical Center, Durham, North Carolina, USA. z University of Utah School of Medicine, Salt Lake City, Utah, USA. § Division of Public Health, Study Design, & Biostatistic Center, University of Utah School of Medicine, Salt Lake City, Utah, USA. || St Luke’s Intermountain Orthopaedics, Boise, Idaho, USA. The authors declared that they have no conflicts of interest in the authorship and publication of this contribution. The American Journal of Sports Medicine, Vol. 43, No. 8 DOI: 10.1177/0363546514556737 Ó 2014 The Author(s)
2049
2050 Grimm et al
cruciate ligament (ACL) in predominantly young athletes.7,9,25,26,31,34 Studies that examine the efficacy of intervention programs are time consuming, expensive, and difficult to conduct. These limitations can lead to study design and methodology flaws. These flaws subsequently increase the risk for bias in the study and reduce the quality of evidence. The purpose of this systematic review and metaanalysis was to identify level 1 evidence studies of ACL and knee injury prevention programs for soccer players, assess the internal validity of these studies, and evaluate the quantitative pooled effectiveness of these prevention programs.
METHODS Criteria for Selecting Studies Types of Studies. Only evidence level 1 randomized controlled trials (RCTs) of injury prevention programs as defined by the Cochrane Handbook24 were included. Level of evidence assessments were made based on the criteria described by the Oxford Centre for Evidence-Based Medicine (http://www.cebm.net). Prospective, nonrandomized studies (level 2), retrospective studies (level 3), case series (level 4), and expert opinion (level 5) publications were not included in the analysis. Non-English articles were excluded from the study. Types of Participants. Only studies with participants who were limited to soccer players were included. Studies of cohorts containing multiple different sport athletes were excluded. Studies were not excluded because of any of the following factors: sex, skill level of athletes, or age group. Types of Interventions and Comparison Groups. The only studies included were those that used interventions to prevent knee and/or ACL injuries. These interventions included strengthening, stretching, proprioception, and neuromuscular exercises. Studies were excluded if they used an exogenous modality as used as a means of prevention (eg, bracing, taping). Outcome Measures. Studies were eligible for inclusion if either ‘‘ACL’’ and/or ‘‘knee’’ injury was reported as an outcome measure. We used the previously published definition of ‘‘knee injury’’ as any trauma, whether contact or noncontact, whether acute or overuse, whether ligamentous or nonligamentous, occurring to the knee. Acute injuries were those with sudden onset with obvious trauma, and overuse injuries were those occurring with an insidious onset.18
Search Method and Strategy The following major medical databases were searched from inception through February 25, 2014: PubMed (Medline), Embase, CINAHL, and Central (Cochrane Library). To develop a sensitive and comprehensive search strategy, a medical library search strategist (M.M.) was consulted,
The American Journal of Sports Medicine
who has participated in previous studies (see Acknowledgments).18 Briefly, the following search criteria were used: ((‘‘knee injuries’’[MeSH Terms] OR (‘‘knee’’[All Fields] AND ‘‘injuries’’[All Fields]) OR ‘‘knee injuries’’ [All Fields] OR (‘‘knee’’[All Fields] AND ‘‘injury’’ [All Fields]) OR ‘‘knee injury’’[All fields]) AND (‘‘prevention and control’’[Subheading] OR (‘‘prevention’’ [All Fields] AND ‘‘control’’[All Fields]) OR ‘‘prevention and control’’[All Fields] OR ‘‘prevention’’[All Fields])) AND Clinical Trial[ptyp] and (ACL[All Fields] AND (‘‘Inj Prev’’[Journal] OR (‘‘injury’’ [All Fields] AND ‘‘prevention’’[All Fields]) OR ‘‘injury prevention’’[All Fields])) AND Clinical Trial[ptyp]. The search included supplemental bibliographic reference searches of articles to identify potentially missed, relevant studies. For a more detailed list of the search strategy used, see the Appendix (available online at http://ajsm.sagepub.com/supplemental).
Data Collection and Analysis Study Selection Procedure. Two reviewers (N.L.G. and J.C.J.) independently conducted the article selection process. The reviewers initially screened these articles on the basis of title and abstract to determine whether the trial met inclusion criteria. The full text was retrieved and reviewed in detail if criteria were met. When disagreements were noted, a third reviewer (K.G.S.) facilitated group consensus agreement. Data Extraction. Two reviewers (N.L.G. and J.C.J.) independently extracted data from studies that met inclusion criteria as described above. The following data were extracted from each study: journal of publication, title, author(s), publication year, subject sex, subject age, number of subjects in both intervention and treatment arm, type of intervention(s), characteristics of intervention(s), study design, follow-up time, and outcome data. Furthermore, the following outcome data elements were extracted from each study: type of injury, frequency of injury occurrence, duration of follow-up, diagnostic methods for outcomes of interest, number of dropouts, and compliance rate. If any of the data elements of interest were missing or unclear, the study authors were contacted for clarification. Risk-of-Bias Assessment. All randomized and clusterrandomized controlled trials that met inclusion were assessed for internal validity, and therefore risk of bias, by 2 independent reviewers (N.L.G. and J.C.J.) using the assessment algorithm described by van Tulder et al,51 and each was assigned a risk-of-bias score based on its assessed internal validity (see Table 1). The van Tulder scale is one of many critical appraisal tools used as an assessment for RCTs developed and used by the Cochrane Spine Group.51 This assessment focuses on several methodological criteria, including randomization method, concealment of allocation, blinding, and intention-to-treat analysis, that have been shown to create an exaggeration of treatment effects and
Vol. 43, No. 8, 2015
Soccer Injury Prevention: Meta-analysis
2051
TABLE 1 Internal Validity Risk-of-Bias Assessment for Included Studies
(A) Was the method of randomization adequate? (B) Was the treatment allocation concealed? (C) Were the groups similar at baseline regarding the most important prognostic indicators? (D) Was the patient blinded to the intervention? (E) Was the care provider blinded to the intervention? (F) Was the outcome assessor blinded to the intervention? (G) Were the cointerventions avoided or similar? (H) Was the compliance acceptable in all groups? (I) Was the dropout rate described and acceptable? (J) Was the timing of the outcome assessment in all groups similar? (K) Did the analysis include an intention-to-treat analysis? Risk-of-Bias score a
Ekstrand et al12 (1983)
Soderman et al47 (2000)
Engebretsen et al14 (2008)
Gilchrist et al17 (2008)
Soligard et al48 (2008)
Steffen et al49 (2008)
Emery and Meeuwisse13 (2010)
van Beijsterveldt et al50 (2012)
Walden et al52 (2012)
Yesa
Yesa
Yesa
Yesa
Yesa
Yes
Yesa
Yes
Yes
Yesa
Yesa
Yesa
No
Yes
Yesa
Yesa
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
Yes
Yes
Yes
No
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
a
Yes
No
No
Yes
Yes
a
Yes
No
No
Noa
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yesa
No
Yes
No
Yes
Yes
Yes
Yes
Yes
6
6
5
5
9
8
8
8
9
Yes
Ascertained via direct communication with primary or secondary author.
lead to systematic bias.35,39,41-44,54 These instruments are recommended by PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses)29 and for reporting systematic reviews in the orthopaedic literature.55 If any of the specific criteria of the van Tulder scale were not explicitly stated in a study’s manuscript, authors were contacted for clarification. If disagreements arose, a third reviewer was consulted (K.G.S.). This process of assessment is consistent with the PRISMA statement for conducting systematic reviews.29 Meta-analysis and Statistical Procedures. All studies’ injury rates concerning intervention and control groups were evaluated and summarized in tabular form for knee and/or ACL injuries. Estimates of the intervention effect on the incidence of knee and ACL injuries were expressed as a relative risk (RR), which compared the knee/ACL incidence rate in the intervention group (numerator) with the rate in the comparison group (denominator). The analysis of publication bias was performed with Stata statistical software (Release 13; StataCorp LP) using the Egger test, Harbord test, Peter test, and Begg test, which are general approaches to test publication bias in the meta-analysis. The Egger test has been widely used and has become a standard procedure.24,33 The first 3 tests are the regression-based tests, which are parametric. However, the Begg test is based on a rank correlation method, which is nonparametric. This was represented qualitatively using a funnel plot derivation. With Stata, an inverse variance method was used to calculate the weight for each study included using random-
effects meta-analyses with the DerSimonian-Laird method to estimate the between-study variance.10 The Cochrane Q test, with a P value of .10 being considered significant, and the Higgins and the Thompson I2 index were independently used to assess heterogeneity among the studies.23,24 These analyses were conducted separately for knee and ACL injury. The estimated prospective statistical power analysis was calculated to reach a power greater than 90%.
RESULTS Search Findings Using an a priori inclusion and exclusion criteria, our search yielded 9 RCTs of knee/ACL injury prevention programs for soccer players from major medical literature databases. We screened a total of 3377 articles based on title and abstract, of which 79 full-text articles were retrieved and reviewed in detail. This resulted in 9 RCTs that met both inclusion and exclusion criteria (Figure 1). Primary reasons for exclusion included nonrandomized study design, no comparison group, and inclusion of subjects who were non–soccer-playing athletes.
Study Characteristics All studies and sample sizes included in this systematic review and meta-analysis are provided in Table 2. A total of 11,562 athletes were included in the 9 studies analyzed.
2052 Grimm et al
The American Journal of Sports Medicine
statistically significant individually (Figure 2B). Likewise, the pooled effect size showed a nonsignificant protective effect against ACL injuries (RR, 0.66; 95% CI, 0.33-1.32; P = .238) (Figure 2B). For sensitivity analyses, the data were analyzed with fixed-effects meta-analysis. In those analyses, similar results in RR and CI were found.
Publication Bias
Figure 1. PRISMA flow diagram for study selection. Five studies included only females,17,47-49,52 3 studies included only males,12,14,50 and 1 study included both males and females13 (Table 2). All study designs were either prospective cluster-randomized controlled trials or prospective RCTs. Of the studies that met the inclusion criteria, 5 reported on knee injury alone and 4 reported both knee and ACL injury (Table 3).
Risk of Bias The average risk-of-bias score for the included studies was 7 of 11 (range, 5-9) (Table 1). Although all internal validity elements of the van Tulder scale51 were obtained from each study, we contacted authors from 7 of the 9 included studies to clarify at least 1 element that was not readily available or clear in the manuscript. Two studies adequately reported all elements of internal validity within their respective manuscript.50,52 Methodological elements that were not performed included cointervention avoidance, intention-to-treat analysis, care provider blinding, outcome assessor blinding, and subject blinding. However, the latter is logistically difficult with this particular study design.
Study-Specific and Quantitative Analysis For overall knee injuries, the estimated RR was less than 1 (which is consistent with a protective effect) in 7 studies, of which 2 were statistically significant individually (Figure 2A). The pooled effect size was statistically significant in favor of injury prevention interventions showing a protective effect against overall knee injuries (RR, 0.74; 95% CI, 0.55-0.98; P = .039) (Figure 2A). For ACL injuries, the estimated RR was less than 1 in 3 studies, which is consistent with a protective effect. However, no study was
Using the methods for publication bias described above, for studies that reported knee injuries, the estimated bias coefficient from the Egger test is 20.97 with a standard error of 1.14, giving a P value of .422. Other tests also showed no small-study effects (P = .537 [Harbord test], .463 [Peter test], and .532 [Begg test]). The test thus provides no evidence of publication bias and can be seen qualitatively in the funnel plot (Figure 3A). Similarly, for studies that reported ACL injuries, the estimated bias coefficient is 3.17 with a standard error of 0.89, giving a P value of .071 with the Egger test and thus no evidence of publication bias. Other tests provided similar results except the result based on the Peter test (P = .131 [Harbord test], .022 [Peter test], and .174 [Begg test]). This can be seen qualitatively in the funnel plot (Figure 3B).
DISCUSSION For this systematic review and meta-analysis, we found 9 high-quality RCTs designed to prevent knee and/or ACL injuries in athletic soccer players. The quantitative pooled effect suggests a significant protective effect against overall knee injuries in this population (RR, 0.74; 95% CI, 0.55-0.98; P = .039); however, a nonsignificant protective effect for ACL injuries was found (RR, 0.66; 95% CI, 0.33-1.32; P = .238). Given the relatively small number of studies meeting inclusion for our meta-analysis, we did not perform subgroup analyses based on sex, skill level, or specific intervention type. It has been shown that multiplicity of analysis increases the risk of type 1 error, resulting in spurious results, and should be performed and interpreted with caution.28,45 Our study was designed a priori to have the statistical power to detect a difference in knee and ACL effects using a random-effects model, and therefore the modifying effects of individual covariates would not have sufficient power for detection. Therefore, we cannot say which elements of the intervention were more or less effective. Neither are we able to clearly comment on the effects of the interventions relative to gender. Outside of the studies analyzed in this meta-analysis and systematic review, several injury prevention studies have been published regarding handball,31,32,53 soccer,9,20,22,34 basketball,22,34 volleyball,22,34 football,7 and skiing.15 However, given the logistical, methodological, and financial complexity of conducting an injury prevention study in athletes, relatively few have been conducted at the highest methodological rigor—the RCT. To date, all of the published systematic reviews and meta-analyses on injury prevention programs have included both
Vol. 43, No. 8, 2015
Soccer Injury Prevention: Meta-analysis
2053
TABLE 2 Summary Table of Study Characteristicsa
Study
Publication Subjects’ Age, y, Year LOE Sex Mean (Range)
Journal
Ekstrand et al12
American Journal of Sports Medicine
1983
1
Male
24.0 (17-37)
Soderman et al47
Knee Surgery, Sports Traumatology, Arthroscopy American Journal of Sports Medicine American Journal of Sports Medicine
2000
1
Female
20.4 (NR)
2008
1
Male
NR
2008
1
Female
19.9 (NR)
Soligard et al48
British Medical Journal
2008
1
Female
15.4 (13-17)
Steffen et al49
Scandinavian Journal of Medicine and Science in Sports British Journal of Sports Medicine
2008
1
Female
15.4 (13-17)
2010
1
Male and female
NR (12-18)
van Beijsterveldt et al50
British Journal of Sports Medicine
2012
1
Male
24.8 (20-29)
Walden et al52
British Medical Journal
2012
1
Female
14 (12-17)
Engebretsen et al14 Gilchrist et al17
Emery and Meeuwisse13
Program Exercises
No. of Subjects Followat Follow-up up
Study Design
Multifaceted approach: Prospective, clusterprotective gear, taping, randomized and warm-up and controlled study b flexibility program Proprioceptive (balance Prospective, clusterboard) randomized controlled study
6 mo
180
6 mo
140
Balance training
Prospective randomized controlled study Prospective, clusterrandomized controlled study
8 mo
131
3 mo
1435
Prospective, clusterrandomized controlled study
8 mo
1892
Prospective, clusterrandomized controlled study
6 mo
2020
Prospective, clusterrandomized controlled study
12 mo
744
Prospective, clusterrandomized controlled study
1 season
456
Prospective, clusterrandomized controlled study
1 season
4564
Multifaceted approach: warm-up, stretching, strength, plyometrics, and agility training Multifaceted approach: warm-up, stretching, plyometrics, and balance training Multifaceted approach: core stabilization, balance, plyometrics, and strength training Dynamic stretching, eccentric strength, agility, jumping, balance Core stability, proprioception, dynamic stabilization, plyometrics, eccentric muscle training Neuromuscular exercises, jumplanding technique
a b
LOE, level of evidence; NR, not reported (data not reported and unable to contact primary or secondary author). The use of protective gear and taping only applied to the shin and ankle. No exogenous interventions were applied to the knee.
TABLE 3 Injury Data for Included Studiesa Knee Injuries, % (n/N) Study
Prevention 12
Ekstrand et al Soderman et al47 Engebretsen et al14 Gilchrist et al17 Soligard et al48 Steffen et al49 Emery and Meeuwisse13 van Beijsterveldt et al50 Walden et al52 a
1.1 (1/90) 12.9 (8/62) 10.8 (7/65) 6.9 3.1 3.4 0.8
(40/583) (33/1055) (37/1073) (3/380)
10.8 (24/223) 1.9 (48/2479)
Control 18.9 (17/90) 7.7 (6/78) 12.1 (8/66) 6.8 5.6 3.2 2.2
(58/852) (47/837) (30/947) (8/364)
17.2 (40/233) 2.1 (44/2085)
ACL Injuries, % (n/N) P Value
Prevention
Control
P Value
\.05 NR .93
NR 6.5 (4/62) NR
NR 1.3 (1/78) NR
NR NR NR
.86 .08 ..05 .232
1.2 (7/583) NR 0.4 (4/1073) NR
2.1 (18/852) NR 0.5 (5/947) NR
.2 NR .73 NR
NR
NR
NR
NR
.71
0.3 (7/2479)
0.7 (14/2085)
.02
ACL, anterior cruciate ligament; LOE, level of evidence; NR, not reported (data not reported and unable to contact primary or secondary author).
2054 Grimm et al
The American Journal of Sports Medicine
Figure 2. (A) Random-effects model with DerSimonian-Laird weighting method showing a statistically significant risk ratio in favor of injury prevention programs for reducing knee injuries. (B) Random-effects model with DerSimonian-Laird weighting method showing no statistically significant benefit for ACL injury prevention.
Figure 3. (A) Funnel plot with 95% confidence interval showing no evidence of publication bias for injury prevention programs evaluating knee injury prevention. (B) Funnel plot with 95% confidence interval showing no evidence of publication bias for injury prevention programs evaluating anterior cruciate ligament injury prevention. randomized and nonrandomized studies.16,19,21,56 The data of these meta-analyses should therefore be evaluated with caution given the mixing of several study designs, cohorts, and individual knee/ACL injury risks. For example, the recent meta-analysis by Gagnier et al16 on ACL injury prevention programs showed a statistically significant reduction of ACL injuries through the use of injury prevention programs. However, the difference in findings may be explained by noting that Gagnier et al16
included several different athletic populations in their analysis (eg, soccer, handball, basketball, volleyball) and they included both evidence level 1 randomized studies and levels 2 and 3 nonrandomized studies. It is well published that these individual sports have variable ACL and knee injury risks.46 Additionally, it is well understood that nonrandomized trials can lead to an exaggeration of treatment effects and greater chance of an introduction of bias into a study.27,54 This exaggeration of treatment effects is
Vol. 43, No. 8, 2015
perhaps suggested by the subgroup analysis by Gagnier et al16 showing that the pooled effect of the randomized studies was not statistically significant; however, nonrandomized studies demonstrated significant benefit. In our study we attempted to control for this shortcoming by including only level 1 studies as they pertained to a single sport (soccer) and limiting the multiplicity of analyses. Although we adhered to a strict protocol, following the PRISMA guidelines,29 for collection, interpretation, and analysis, our study is not without limitations. First, we did not blind the reviewers (N.L.G. and J.C.J.) to study author, institution, or journal in which the study was published. Although the extra step of blinding the reviewed manuscripts has been performed in other reviews, it is an onerous extra step with little evidence to support its ability to protect against bias.6 Furthermore, with a small number of studies (n \ 10), the assessment of publication bias should be interpreted with caution.24 Although there are no strict guidelines on the absolute number of studies needed to detect a true publication bias, the larger the sample the more accurate the assessment of bias will be. Nonetheless, we used recommended analytical techniques for detection of publication bias.24,33,38 Additionally, one of the authors (K.G.S.) has published on knee injury prevention programs in a cohort of female soccer, basketball, and volleyball players.34 This study showed no treatment benefit, and an argument could be made that this could introduce selection bias. However, we attempted to control for this by removing this author as a reviewer of selected studies and involving this author only when consensus was needed for disagreement between the primary reviewers (N.L.G. and J.C.J.).
CONCLUSION To our knowledge, we have conducted the first level 1 meta-analysis of injury prevention programs in soccer athletes. While this analysis supports the use of injury prevention programs for preventing overall knee injuries in this athletic population, a nonsignificant reduction in ACL injuries was observed through meta-analytical techniques. Similar to the authors of previous studies,16 we cannot say with much confidence what components of the injury prevention programs are most effective; however, neuromuscular training has showed promising results.30 Given the logistical challenges of conducting highquality RCTs on injury prevention in this population, we applaud the researchers who have performed these much-needed scientific studies and urge future investigators to follow the standards set forth by the CONSORT (Consolidated Standards of Reporting Trials) statement40 for reporting this research in our athletic population.
ACKNOWLEDGMENT The authors thank Mary McFarland of the Eccles Health Science Medical Library for assisting with the development of a sensitive search strategy for this study.
Soccer Injury Prevention: Meta-analysis
2055
An online CME course associated with this article is available for 1 AMA PRA Category 1 CreditTM at http://ajsm-cme.sagepub.com. In accordance with the standards of the Accreditation Council for Continuing Medical Education (ACCME), it is the policy of The American Orthopaedic Society for Sports Medicine that authors, editors, and planners disclose to the learners all financial relationships during the past 12 months with any commercial interest (A ‘commercial interest’ is any entity producing, marketing, re-selling, or distributing health care goods or services consumed by, or used on, patients). Any and all disclosures are provided in the online journal CME area which is provided to all participants before they actually take the CME activity. In accordance with AOSSM policy, authors, editors, and planners’ participation in this educational activity will be predicated upon timely submission and review of AOSSM disclosure. Noncompliance will result in an author/editor or planner to be stricken from participating in this CME activity.
REFERENCES 1. Big count. FIFA Magazine. FIFA; 2007. http://www.fifa.com/mm/document/fifafacts/bcoffsurv/emaga_9384_10704.pdf. Accessed January 1, 2014. 2. Agel J, Evans TA, Dick R, Putukian M, Marshall SW. Descriptive epidemiology of collegiate men’s soccer injuries: National Collegiate Athletic Association Injury Surveillance System, 1988-1989 through 2002-2003. J Athl Train. 2007;42(2):270-277. 3. Agel J, Olson DE, Dick R, Arendt EA, Marshall SW, Sikka RS. Descriptive epidemiology of collegiate women’s basketball injuries: National Collegiate Athletic Association Injury Surveillance System, 1988-1989 through 2003-2004. J Athl Train. 2007;42(2):202-210. 4. Agel J, Palmieri-Smith R, Dick R, Wojtys EM, Marshall SW. Descriptive epidemiology of collegiate women’s volleyball injuries: National Collegiate Athletic Association Injury Surveillance System, 19881989 through 2003-2004. J Athl Train. 2007;42:295-302. 5. Arendt E, Dick R. Knee injury patterns among men and women in collegiate basketball and soccer. NCAA data and review of literature. Am J Sports Med. 1995;23(6):694-701. 6. Berlin JA. Does blinding of readers affect the results of metaanalyses? University of Pennsylvania Meta-analysis Blinding Study Group. Lancet. 1997;350(9072):185-186. 7. Cahill BR, Griffith EH. Effect of preseason conditioning on the incidence and severity of high school football knee injuries. Am J Sports Med. 1978;6(4):180-184. 8. Caine D, Caine C, Maffulli N. Incidence and distribution of pediatric sport-related injuries. Clin J Sport Med. 2006;16(6):500-513. 9. Caraffa A, Cerulli G, Projetti M, Aisa G, Rizzo A. Prevention of anterior cruciate ligament injuries in soccer: a prospective controlled study of proprioceptive training. Knee Surg Sports Traumatol Arthrosc. 1996;4(1):19-21. 10. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177-188. 11. Dick R, Putukian M, Agel J, Evans TA, Marshall SW. Descriptive epidemiology of collegiate women’s soccer injuries: National Collegiate Athletic Association Injury Surveillance System, 1988-1989 through 2002-2003. J Athl Train. 2007;42(2):278-285. 12. Ekstrand J, Gillquist J, Liljedahl SO. Prevention of soccer injuries: supervision by doctor and physiotherapist. Am J Sports Med. 1983;11(3):116-120. 13. Emery CA, Meeuwisse WH. The effectiveness of a neuromuscular prevention strategy to reduce injuries in youth soccer: a clusterrandomised controlled trial. Br J Sports Med. 2010;44(8):555-562.
2056 Grimm et al
14. Engebretsen AH, Myklebust G, Holme I, Engebretsen L, Bahr R. Prevention of injuries among male soccer players: a prospective, randomized intervention study targeting players with previous injuries or reduced function. Am J Sports Med. 2008;36(6):1052-1060. 15. Ettlinger CF, Johnson RJ, Shealy JE. A method to help reduce the risk of serious knee sprains incurred in alpine skiing. Am J Sports Med. 1995;23(5):531-537. 16. Gagnier JJ, Morgenstern H, Chess L. Interventions designed to prevent anterior cruciate ligament injuries in adolescents and adults: a systematic review and meta-analysis. Am J Sports Med. 2013;41(8):1952-1962. 17. Gilchrist J, Mandelbaum BR, Melancon H, et al. A randomized controlled trial to prevent noncontact anterior cruciate ligament injury in female collegiate soccer players. Am J Sports Med. 2008;36(8):1476-1483. 18. Grimm NL, Shea KG, Leaver RW, Aoki SK, Carey JL. Efficacy and degree of bias in knee injury prevention studies: a systematic review of RCTs. Clin Orthop Relat Res. 2013;471(1):308-316. 19. Grindstaff TL, Hammill RR, Tuzson AE, Hertel J. Neuromuscular control training programs and noncontact anterior cruciate ligament injury rates in female athletes: a numbers-needed-to-treat analysis. J Athl Train. 2006;41(4):450-456. 20. Heidt RS Jr, Sweeterman LM, Carlonas RL, Traub JA, Tekulve FX. Avoidance of soccer injuries with preseason conditioning. Am J Sports Med. 2000;28(5):659-662. 21. Hewett TE, Ford KR, Myer GD. Anterior cruciate ligament injuries in female athletes: part 2, a meta-analysis of neuromuscular interventions aimed at injury prevention. Am J Sports Med. 2006;34(3):490-498. 22. Hewett TE, Lindenfeld TN, Riccobene JV, Noyes FR. The effect of neuromuscular training on the incidence of knee injury in female athletes: a prospective study. Am J Sports Med. 1999;27(6):699-706. 23. Higgins JP, Thompson SG. Quantifying heterogeneity in a metaanalysis. Stat Med. 2002;21(11):1539-1558. 24. Higgins JPT, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions, Version 5.1.0. The Cochrane Collaboration; 2011. www.cochrane-handbook.org. Updated March 2011. Accessed January 1, 2014. 25. Junge A, Rosch D, Peterson L, Graf-Baumann T, Dvorak J. Prevention of soccer injuries: a prospective intervention study in youth amateur players. Am J Sports Med. 2002;30(5):652-659. 26. Kiani A, Hellquist E, Ahlqvist K, Gedeborg R, Michaelsson K, Byberg L. Prevention of soccer-related knee injuries in teenaged girls. Arch Intern Med. 2010;170(1):43-49. 27. Kunz R, Oxman AD. The unpredictability paradox: review of empirical comparisons of randomised and non-randomised clinical trials. BMJ. 1998;317(7167):1185-1190. 28. Magee CD, Byars LA, DeZee KJ. Limitations of subgroup analyses in meta-analysis of cardiac resynchronization therapy by QRS duration. Arch Intern Med. 2012;172(4):375; author reply 375-376. 29. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol. 2009;62(10):1006-1012. 30. Myer GD, Sugimoto D, Thomas S, Hewett TE. The influence of age on the effectiveness of neuromuscular training to reduce anterior cruciate ligament injury in female athletes: a meta-analysis. Am J Sports Med. 2013;41(1):203-215. 31. Myklebust G, Engebretsen L, Braekken IH, Skjolberg A, Olsen OE, Bahr R. Prevention of anterior cruciate ligament injuries in female team handball players: a prospective intervention study over three seasons. Clin J Sport Med. 2003;13(2):71-78. 32. Olsen OE, Myklebust G, Engebretsen L, Holme I, Bahr R. Exercises to prevent lower limb injuries in youth sports: cluster randomised controlled trial. BMJ. 2005;330(7489):449. 33. Peters JL, Sutton AJ, Jones DR, Abrams KR, Rushton L. Comparison of two methods to detect publication bias in meta-analysis. JAMA. 2006;295(6):676-680. 34. Pfeiffer RP, Shea KG, Roberts D, Grandstrand S, Bond L. Lack of effect of a knee ligament injury prevention program on the incidence of noncontact anterior cruciate ligament injury. J Bone Joint Surg Am. 2006;88(8):1769-1774.
The American Journal of Sports Medicine
35. Poolman RW, Struijs PA, Krips R, et al. Reporting of outcomes in orthopaedic randomized trials: does blinding of outcome assessors matter? J Bone Joint Surg Am. 2007;89(3):550-558. 36. Powell JW, Barber-Foss KD. Injury patterns in selected high school sports: a review of the 1995-1997 seasons. J Athl Train. 1999;34(3): 277-284. 37. Renstrom P, Ljungqvist A, Arendt E, et al. Non-contact ACL injuries in female athletes: an International Olympic Committee current concepts statement. Br J Sports Med. 2008;42(6):394-412. 38. Rothstein HR, Sutton AJ, Borenstein M. Publication Bias in Metaanalysis—Prevention, Assessment and Adjustments. West Sussex, UK: John Wiley & Sons; 2005. 39. Schulz KF, Altman DG, Moher D. Allocation concealment in clinical trials. JAMA. 2002;288(19):2406-2407; author reply 2408-2409. 40. Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. J Clin Epidemiol. 2010;63(8):834-840. 41. Schulz KF, Chalmers I, Altman DG. The landscape and lexicon of blinding in randomized trials. Ann Intern Med. 2002;136(3):254259. 42. Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias: dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA. 1995;273(5): 408-412. 43. Schulz KF, Grimes DA. Allocation concealment in randomised trials: defending against deciphering. Lancet. 2002;359(9306):614-618. 44. Schulz KF, Grimes DA. Blinding in randomised trials: hiding who got what. Lancet. 2002;359(9307):696-700. 45. Schulz KF, Grimes DA. Multiplicity in randomised trials II: subgroup and interim analyses. Lancet. 2005;365(9471):1657-1661. 46. Shea KG, Grimm NL, Ewing CK, Aoki SK. Youth sports anterior cruciate ligament and knee injury epidemiology: who is getting injured? In what sports? When? Clin Sports Med. 2011;30(4):691-706. 47. Soderman K, Werner S, Pietila T, Engstrom B, Alfredson H. Balance board training: prevention of traumatic injuries of the lower extremities in female soccer players? A prospective randomized intervention study. Knee Surg Sports Traumatol Arthrosc. 2000;8(6):356363. 48. Soligard T, Myklebust G, Steffen K, et al. Comprehensive warm-up programme to prevent injuries in young female footballers: cluster randomised controlled trial. BMJ. 2008;337:a2469. 49. Steffen K, Myklebust G, Olsen OE, Holme I, Bahr R. Preventing injuries in female youth football—a cluster-randomized controlled trial. Scand J Med Sci Sports. 2008;18(5):605-614. 50. van Beijsterveldt AM, van de Port IG, Krist MR, et al. Effectiveness of an injury prevention programme for adult male amateur soccer players: a cluster-randomised controlled trial. Br J Sports Med. 2012; 46(16):1114-1118. 51. van Tulder M, Furlan A, Bombardier C, Bouter L. Updated method guidelines for systematic reviews in the Cochrane Collaboration back review group. Spine. 2003;28(12):1290-1299. 52. Walden M, Atroshi I, Magnusson H, Wagner P, Hagglund M. Prevention of acute knee injuries in adolescent female football players: cluster randomised controlled trial. BMJ. 2012;344:e3042. 53. Wedderkopp N, Kaltoft M, Lundgaard B, Rosendahl M, Froberg K. Prevention of injuries in young female players in European team handball: a prospective intervention study. Scand J Med Sci Sports. 1999;9(1):41-47. 54. Wood L, Egger M, Gluud LL, et al. Empirical evidence of bias in treatment effect estimates in controlled trials with different interventions and outcomes: meta-epidemiological study. BMJ. 2008;336 (7644):601-605. 55. Wright RW, Brand RA, Dunn W, Spindler KP. How to write a systematic review. Clin Orthop Relat Res. 2007;455:23-29. 56. Yoo JH, Lim BO, Ha M, et al. A meta-analysis of the effect of neuromuscular training on the prevention of the anterior cruciate ligament injury in female athletes. Knee Surg Sports Traumatol Arthrosc. 2010;18(6):824-830.
For reprints and permission queries, please visit SAGE’s Web site at http://www.sagepub.com/journalsPermissions.nav
Anterior Cruciate Ligament and Knee Injury Prevention Programs for Soccer Players
M
A Systematic Review and Meta-analysis Nathan L. Grimm,*y MD, John C. Jacobs Jr,z BS, Jaewhan Kim,§ PhD, Brandon S. Denney,z BS, and Kevin G. Shea,|| MD Investigation performed at the Duke University Medical Center, Durham, North Carolina, USA Background: Soccer has one of the highest incidences of anterior cruciate ligament (ACL) injuries for both males and females. Several injury prevention programs have been developed to address this concern. However, an analysis of the pooled effect has yet to be elicited. Purpose: To conduct a systematic review and meta-analysis of ACL and knee injury prevention programs for soccer players, assess the heterogeneity among the studies, and evaluate the reported effectiveness of the prevention programs. Study Design: Systematic review and meta-analysis. Methods: A systematic search of the literature was conducted on PubMed (Medline), Embase, CINAHL, and Central-Cochrane Database. Studies were limited to randomized controlled trials (RCTs) of injury prevention programs specific to the knee and/or ACL in soccer players. The Cochrane Q test and I2 index were independently used to assess heterogeneity among the studies. The pooled risk difference, assessing knee and/or ACL injury rates between intervention and control groups, was calculated by random-effects models with use of the DerSimonian-Laird method. Publication bias was assessed with a funnel plot and Egger weighted regression technique. Results: Nine studies met the inclusion criteria as RCTs. A total of 11,562 athletes were included, of whom 7889 were analyzed for ACL-specific injuries. Moderate heterogeneity was found among studies of knee injury prevention (P = .041); however, there was insignificant variation found among studies of ACL injury prevention programs (P = .222). For studies of knee injury prevention programs, the risk ratio was 0.74 (95% CI, 0.55-0.89), and a significant reduction in risk of knee injury was found in the prevention group (P = .039). For studies of ACL injury prevention programs, the risk ratio was 0.66 (95% CI, 0.33-1.32), and a nonsignificant reduction in risk of ACL injury was found in the prevention group (P = .238). No evidence of publication bias was found among studies of either knee or ACL injury prevention programs. Conclusion: This systematic review and meta-analysis of ACL and knee injury prevention program studies found a statistically significant reduction in injury risk for knee injuries but did not find a statistically significant reduction of ACL injuries. Keywords: anterior cruciate ligament injury prevention; knee injury; injury prevention; meta-analysis; systematic review
Soccer is the most popular sport in the world, with approximately 265 million players as of 2006.1 The sport has seen tremendous growth over the past decade in the United States, especially in the youth and female populations, which has led to an associated increase in injuries sustained by soccer players.2,11 The lower extremity represents a majority of soccer injuries in both male and female athletes, with the knee being a frequently injured body part.8,36,46 In particular, female athletes have a 3 to 5 times higher risk of serious knee injury compared with male athletes for soccer, basketball, volleyball, and other sports.3-5,11,37 Many intervention programs have been designed to reduce the risk of injury to the knee and anterior
*Address correspondence to Nathan L. Grimm, MD, Department of Orthopaedic Surgery, Duke University Medical Center, 8 Intuition Circle, Durham, NC 27705, USA (e-mail: [email protected]). y Department of Orthopaedic Surgery, Duke University Medical Center, Durham, North Carolina, USA. z University of Utah School of Medicine, Salt Lake City, Utah, USA. § Division of Public Health, Study Design, & Biostatistic Center, University of Utah School of Medicine, Salt Lake City, Utah, USA. || St Luke’s Intermountain Orthopaedics, Boise, Idaho, USA. The authors declared that they have no conflicts of interest in the authorship and publication of this contribution. The American Journal of Sports Medicine, Vol. 43, No. 8 DOI: 10.1177/0363546514556737 Ó 2014 The Author(s)
2049
2050 Grimm et al
cruciate ligament (ACL) in predominantly young athletes.7,9,25,26,31,34 Studies that examine the efficacy of intervention programs are time consuming, expensive, and difficult to conduct. These limitations can lead to study design and methodology flaws. These flaws subsequently increase the risk for bias in the study and reduce the quality of evidence. The purpose of this systematic review and metaanalysis was to identify level 1 evidence studies of ACL and knee injury prevention programs for soccer players, assess the internal validity of these studies, and evaluate the quantitative pooled effectiveness of these prevention programs.
METHODS Criteria for Selecting Studies Types of Studies. Only evidence level 1 randomized controlled trials (RCTs) of injury prevention programs as defined by the Cochrane Handbook24 were included. Level of evidence assessments were made based on the criteria described by the Oxford Centre for Evidence-Based Medicine (http://www.cebm.net). Prospective, nonrandomized studies (level 2), retrospective studies (level 3), case series (level 4), and expert opinion (level 5) publications were not included in the analysis. Non-English articles were excluded from the study. Types of Participants. Only studies with participants who were limited to soccer players were included. Studies of cohorts containing multiple different sport athletes were excluded. Studies were not excluded because of any of the following factors: sex, skill level of athletes, or age group. Types of Interventions and Comparison Groups. The only studies included were those that used interventions to prevent knee and/or ACL injuries. These interventions included strengthening, stretching, proprioception, and neuromuscular exercises. Studies were excluded if they used an exogenous modality as used as a means of prevention (eg, bracing, taping). Outcome Measures. Studies were eligible for inclusion if either ‘‘ACL’’ and/or ‘‘knee’’ injury was reported as an outcome measure. We used the previously published definition of ‘‘knee injury’’ as any trauma, whether contact or noncontact, whether acute or overuse, whether ligamentous or nonligamentous, occurring to the knee. Acute injuries were those with sudden onset with obvious trauma, and overuse injuries were those occurring with an insidious onset.18
Search Method and Strategy The following major medical databases were searched from inception through February 25, 2014: PubMed (Medline), Embase, CINAHL, and Central (Cochrane Library). To develop a sensitive and comprehensive search strategy, a medical library search strategist (M.M.) was consulted,
The American Journal of Sports Medicine
who has participated in previous studies (see Acknowledgments).18 Briefly, the following search criteria were used: ((‘‘knee injuries’’[MeSH Terms] OR (‘‘knee’’[All Fields] AND ‘‘injuries’’[All Fields]) OR ‘‘knee injuries’’ [All Fields] OR (‘‘knee’’[All Fields] AND ‘‘injury’’ [All Fields]) OR ‘‘knee injury’’[All fields]) AND (‘‘prevention and control’’[Subheading] OR (‘‘prevention’’ [All Fields] AND ‘‘control’’[All Fields]) OR ‘‘prevention and control’’[All Fields] OR ‘‘prevention’’[All Fields])) AND Clinical Trial[ptyp] and (ACL[All Fields] AND (‘‘Inj Prev’’[Journal] OR (‘‘injury’’ [All Fields] AND ‘‘prevention’’[All Fields]) OR ‘‘injury prevention’’[All Fields])) AND Clinical Trial[ptyp]. The search included supplemental bibliographic reference searches of articles to identify potentially missed, relevant studies. For a more detailed list of the search strategy used, see the Appendix (available online at http://ajsm.sagepub.com/supplemental).
Data Collection and Analysis Study Selection Procedure. Two reviewers (N.L.G. and J.C.J.) independently conducted the article selection process. The reviewers initially screened these articles on the basis of title and abstract to determine whether the trial met inclusion criteria. The full text was retrieved and reviewed in detail if criteria were met. When disagreements were noted, a third reviewer (K.G.S.) facilitated group consensus agreement. Data Extraction. Two reviewers (N.L.G. and J.C.J.) independently extracted data from studies that met inclusion criteria as described above. The following data were extracted from each study: journal of publication, title, author(s), publication year, subject sex, subject age, number of subjects in both intervention and treatment arm, type of intervention(s), characteristics of intervention(s), study design, follow-up time, and outcome data. Furthermore, the following outcome data elements were extracted from each study: type of injury, frequency of injury occurrence, duration of follow-up, diagnostic methods for outcomes of interest, number of dropouts, and compliance rate. If any of the data elements of interest were missing or unclear, the study authors were contacted for clarification. Risk-of-Bias Assessment. All randomized and clusterrandomized controlled trials that met inclusion were assessed for internal validity, and therefore risk of bias, by 2 independent reviewers (N.L.G. and J.C.J.) using the assessment algorithm described by van Tulder et al,51 and each was assigned a risk-of-bias score based on its assessed internal validity (see Table 1). The van Tulder scale is one of many critical appraisal tools used as an assessment for RCTs developed and used by the Cochrane Spine Group.51 This assessment focuses on several methodological criteria, including randomization method, concealment of allocation, blinding, and intention-to-treat analysis, that have been shown to create an exaggeration of treatment effects and
Vol. 43, No. 8, 2015
Soccer Injury Prevention: Meta-analysis
2051
TABLE 1 Internal Validity Risk-of-Bias Assessment for Included Studies
(A) Was the method of randomization adequate? (B) Was the treatment allocation concealed? (C) Were the groups similar at baseline regarding the most important prognostic indicators? (D) Was the patient blinded to the intervention? (E) Was the care provider blinded to the intervention? (F) Was the outcome assessor blinded to the intervention? (G) Were the cointerventions avoided or similar? (H) Was the compliance acceptable in all groups? (I) Was the dropout rate described and acceptable? (J) Was the timing of the outcome assessment in all groups similar? (K) Did the analysis include an intention-to-treat analysis? Risk-of-Bias score a
Ekstrand et al12 (1983)
Soderman et al47 (2000)
Engebretsen et al14 (2008)
Gilchrist et al17 (2008)
Soligard et al48 (2008)
Steffen et al49 (2008)
Emery and Meeuwisse13 (2010)
van Beijsterveldt et al50 (2012)
Walden et al52 (2012)
Yesa
Yesa
Yesa
Yesa
Yesa
Yes
Yesa
Yes
Yes
Yesa
Yesa
Yesa
No
Yes
Yesa
Yesa
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
Yes
Yes
Yes
No
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
a
Yes
No
No
Yes
Yes
a
Yes
No
No
Noa
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yesa
No
Yes
No
Yes
Yes
Yes
Yes
Yes
6
6
5
5
9
8
8
8
9
Yes
Ascertained via direct communication with primary or secondary author.
lead to systematic bias.35,39,41-44,54 These instruments are recommended by PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses)29 and for reporting systematic reviews in the orthopaedic literature.55 If any of the specific criteria of the van Tulder scale were not explicitly stated in a study’s manuscript, authors were contacted for clarification. If disagreements arose, a third reviewer was consulted (K.G.S.). This process of assessment is consistent with the PRISMA statement for conducting systematic reviews.29 Meta-analysis and Statistical Procedures. All studies’ injury rates concerning intervention and control groups were evaluated and summarized in tabular form for knee and/or ACL injuries. Estimates of the intervention effect on the incidence of knee and ACL injuries were expressed as a relative risk (RR), which compared the knee/ACL incidence rate in the intervention group (numerator) with the rate in the comparison group (denominator). The analysis of publication bias was performed with Stata statistical software (Release 13; StataCorp LP) using the Egger test, Harbord test, Peter test, and Begg test, which are general approaches to test publication bias in the meta-analysis. The Egger test has been widely used and has become a standard procedure.24,33 The first 3 tests are the regression-based tests, which are parametric. However, the Begg test is based on a rank correlation method, which is nonparametric. This was represented qualitatively using a funnel plot derivation. With Stata, an inverse variance method was used to calculate the weight for each study included using random-
effects meta-analyses with the DerSimonian-Laird method to estimate the between-study variance.10 The Cochrane Q test, with a P value of .10 being considered significant, and the Higgins and the Thompson I2 index were independently used to assess heterogeneity among the studies.23,24 These analyses were conducted separately for knee and ACL injury. The estimated prospective statistical power analysis was calculated to reach a power greater than 90%.
RESULTS Search Findings Using an a priori inclusion and exclusion criteria, our search yielded 9 RCTs of knee/ACL injury prevention programs for soccer players from major medical literature databases. We screened a total of 3377 articles based on title and abstract, of which 79 full-text articles were retrieved and reviewed in detail. This resulted in 9 RCTs that met both inclusion and exclusion criteria (Figure 1). Primary reasons for exclusion included nonrandomized study design, no comparison group, and inclusion of subjects who were non–soccer-playing athletes.
Study Characteristics All studies and sample sizes included in this systematic review and meta-analysis are provided in Table 2. A total of 11,562 athletes were included in the 9 studies analyzed.
2052 Grimm et al
The American Journal of Sports Medicine
statistically significant individually (Figure 2B). Likewise, the pooled effect size showed a nonsignificant protective effect against ACL injuries (RR, 0.66; 95% CI, 0.33-1.32; P = .238) (Figure 2B). For sensitivity analyses, the data were analyzed with fixed-effects meta-analysis. In those analyses, similar results in RR and CI were found.
Publication Bias
Figure 1. PRISMA flow diagram for study selection. Five studies included only females,17,47-49,52 3 studies included only males,12,14,50 and 1 study included both males and females13 (Table 2). All study designs were either prospective cluster-randomized controlled trials or prospective RCTs. Of the studies that met the inclusion criteria, 5 reported on knee injury alone and 4 reported both knee and ACL injury (Table 3).
Risk of Bias The average risk-of-bias score for the included studies was 7 of 11 (range, 5-9) (Table 1). Although all internal validity elements of the van Tulder scale51 were obtained from each study, we contacted authors from 7 of the 9 included studies to clarify at least 1 element that was not readily available or clear in the manuscript. Two studies adequately reported all elements of internal validity within their respective manuscript.50,52 Methodological elements that were not performed included cointervention avoidance, intention-to-treat analysis, care provider blinding, outcome assessor blinding, and subject blinding. However, the latter is logistically difficult with this particular study design.
Study-Specific and Quantitative Analysis For overall knee injuries, the estimated RR was less than 1 (which is consistent with a protective effect) in 7 studies, of which 2 were statistically significant individually (Figure 2A). The pooled effect size was statistically significant in favor of injury prevention interventions showing a protective effect against overall knee injuries (RR, 0.74; 95% CI, 0.55-0.98; P = .039) (Figure 2A). For ACL injuries, the estimated RR was less than 1 in 3 studies, which is consistent with a protective effect. However, no study was
Using the methods for publication bias described above, for studies that reported knee injuries, the estimated bias coefficient from the Egger test is 20.97 with a standard error of 1.14, giving a P value of .422. Other tests also showed no small-study effects (P = .537 [Harbord test], .463 [Peter test], and .532 [Begg test]). The test thus provides no evidence of publication bias and can be seen qualitatively in the funnel plot (Figure 3A). Similarly, for studies that reported ACL injuries, the estimated bias coefficient is 3.17 with a standard error of 0.89, giving a P value of .071 with the Egger test and thus no evidence of publication bias. Other tests provided similar results except the result based on the Peter test (P = .131 [Harbord test], .022 [Peter test], and .174 [Begg test]). This can be seen qualitatively in the funnel plot (Figure 3B).
DISCUSSION For this systematic review and meta-analysis, we found 9 high-quality RCTs designed to prevent knee and/or ACL injuries in athletic soccer players. The quantitative pooled effect suggests a significant protective effect against overall knee injuries in this population (RR, 0.74; 95% CI, 0.55-0.98; P = .039); however, a nonsignificant protective effect for ACL injuries was found (RR, 0.66; 95% CI, 0.33-1.32; P = .238). Given the relatively small number of studies meeting inclusion for our meta-analysis, we did not perform subgroup analyses based on sex, skill level, or specific intervention type. It has been shown that multiplicity of analysis increases the risk of type 1 error, resulting in spurious results, and should be performed and interpreted with caution.28,45 Our study was designed a priori to have the statistical power to detect a difference in knee and ACL effects using a random-effects model, and therefore the modifying effects of individual covariates would not have sufficient power for detection. Therefore, we cannot say which elements of the intervention were more or less effective. Neither are we able to clearly comment on the effects of the interventions relative to gender. Outside of the studies analyzed in this meta-analysis and systematic review, several injury prevention studies have been published regarding handball,31,32,53 soccer,9,20,22,34 basketball,22,34 volleyball,22,34 football,7 and skiing.15 However, given the logistical, methodological, and financial complexity of conducting an injury prevention study in athletes, relatively few have been conducted at the highest methodological rigor—the RCT. To date, all of the published systematic reviews and meta-analyses on injury prevention programs have included both
Vol. 43, No. 8, 2015
Soccer Injury Prevention: Meta-analysis
2053
TABLE 2 Summary Table of Study Characteristicsa
Study
Publication Subjects’ Age, y, Year LOE Sex Mean (Range)
Journal
Ekstrand et al12
American Journal of Sports Medicine
1983
1
Male
24.0 (17-37)
Soderman et al47
Knee Surgery, Sports Traumatology, Arthroscopy American Journal of Sports Medicine American Journal of Sports Medicine
2000
1
Female
20.4 (NR)
2008
1
Male
NR
2008
1
Female
19.9 (NR)
Soligard et al48
British Medical Journal
2008
1
Female
15.4 (13-17)
Steffen et al49
Scandinavian Journal of Medicine and Science in Sports British Journal of Sports Medicine
2008
1
Female
15.4 (13-17)
2010
1
Male and female
NR (12-18)
van Beijsterveldt et al50
British Journal of Sports Medicine
2012
1
Male
24.8 (20-29)
Walden et al52
British Medical Journal
2012
1
Female
14 (12-17)
Engebretsen et al14 Gilchrist et al17
Emery and Meeuwisse13
Program Exercises
No. of Subjects Followat Follow-up up
Study Design
Multifaceted approach: Prospective, clusterprotective gear, taping, randomized and warm-up and controlled study b flexibility program Proprioceptive (balance Prospective, clusterboard) randomized controlled study
6 mo
180
6 mo
140
Balance training
Prospective randomized controlled study Prospective, clusterrandomized controlled study
8 mo
131
3 mo
1435
Prospective, clusterrandomized controlled study
8 mo
1892
Prospective, clusterrandomized controlled study
6 mo
2020
Prospective, clusterrandomized controlled study
12 mo
744
Prospective, clusterrandomized controlled study
1 season
456
Prospective, clusterrandomized controlled study
1 season
4564
Multifaceted approach: warm-up, stretching, strength, plyometrics, and agility training Multifaceted approach: warm-up, stretching, plyometrics, and balance training Multifaceted approach: core stabilization, balance, plyometrics, and strength training Dynamic stretching, eccentric strength, agility, jumping, balance Core stability, proprioception, dynamic stabilization, plyometrics, eccentric muscle training Neuromuscular exercises, jumplanding technique
a b
LOE, level of evidence; NR, not reported (data not reported and unable to contact primary or secondary author). The use of protective gear and taping only applied to the shin and ankle. No exogenous interventions were applied to the knee.
TABLE 3 Injury Data for Included Studiesa Knee Injuries, % (n/N) Study
Prevention 12
Ekstrand et al Soderman et al47 Engebretsen et al14 Gilchrist et al17 Soligard et al48 Steffen et al49 Emery and Meeuwisse13 van Beijsterveldt et al50 Walden et al52 a
1.1 (1/90) 12.9 (8/62) 10.8 (7/65) 6.9 3.1 3.4 0.8
(40/583) (33/1055) (37/1073) (3/380)
10.8 (24/223) 1.9 (48/2479)
Control 18.9 (17/90) 7.7 (6/78) 12.1 (8/66) 6.8 5.6 3.2 2.2
(58/852) (47/837) (30/947) (8/364)
17.2 (40/233) 2.1 (44/2085)
ACL Injuries, % (n/N) P Value
Prevention
Control
P Value
\.05 NR .93
NR 6.5 (4/62) NR
NR 1.3 (1/78) NR
NR NR NR
.86 .08 ..05 .232
1.2 (7/583) NR 0.4 (4/1073) NR
2.1 (18/852) NR 0.5 (5/947) NR
.2 NR .73 NR
NR
NR
NR
NR
.71
0.3 (7/2479)
0.7 (14/2085)
.02
ACL, anterior cruciate ligament; LOE, level of evidence; NR, not reported (data not reported and unable to contact primary or secondary author).
2054 Grimm et al
The American Journal of Sports Medicine
Figure 2. (A) Random-effects model with DerSimonian-Laird weighting method showing a statistically significant risk ratio in favor of injury prevention programs for reducing knee injuries. (B) Random-effects model with DerSimonian-Laird weighting method showing no statistically significant benefit for ACL injury prevention.
Figure 3. (A) Funnel plot with 95% confidence interval showing no evidence of publication bias for injury prevention programs evaluating knee injury prevention. (B) Funnel plot with 95% confidence interval showing no evidence of publication bias for injury prevention programs evaluating anterior cruciate ligament injury prevention. randomized and nonrandomized studies.16,19,21,56 The data of these meta-analyses should therefore be evaluated with caution given the mixing of several study designs, cohorts, and individual knee/ACL injury risks. For example, the recent meta-analysis by Gagnier et al16 on ACL injury prevention programs showed a statistically significant reduction of ACL injuries through the use of injury prevention programs. However, the difference in findings may be explained by noting that Gagnier et al16
included several different athletic populations in their analysis (eg, soccer, handball, basketball, volleyball) and they included both evidence level 1 randomized studies and levels 2 and 3 nonrandomized studies. It is well published that these individual sports have variable ACL and knee injury risks.46 Additionally, it is well understood that nonrandomized trials can lead to an exaggeration of treatment effects and greater chance of an introduction of bias into a study.27,54 This exaggeration of treatment effects is
Vol. 43, No. 8, 2015
perhaps suggested by the subgroup analysis by Gagnier et al16 showing that the pooled effect of the randomized studies was not statistically significant; however, nonrandomized studies demonstrated significant benefit. In our study we attempted to control for this shortcoming by including only level 1 studies as they pertained to a single sport (soccer) and limiting the multiplicity of analyses. Although we adhered to a strict protocol, following the PRISMA guidelines,29 for collection, interpretation, and analysis, our study is not without limitations. First, we did not blind the reviewers (N.L.G. and J.C.J.) to study author, institution, or journal in which the study was published. Although the extra step of blinding the reviewed manuscripts has been performed in other reviews, it is an onerous extra step with little evidence to support its ability to protect against bias.6 Furthermore, with a small number of studies (n \ 10), the assessment of publication bias should be interpreted with caution.24 Although there are no strict guidelines on the absolute number of studies needed to detect a true publication bias, the larger the sample the more accurate the assessment of bias will be. Nonetheless, we used recommended analytical techniques for detection of publication bias.24,33,38 Additionally, one of the authors (K.G.S.) has published on knee injury prevention programs in a cohort of female soccer, basketball, and volleyball players.34 This study showed no treatment benefit, and an argument could be made that this could introduce selection bias. However, we attempted to control for this by removing this author as a reviewer of selected studies and involving this author only when consensus was needed for disagreement between the primary reviewers (N.L.G. and J.C.J.).
CONCLUSION To our knowledge, we have conducted the first level 1 meta-analysis of injury prevention programs in soccer athletes. While this analysis supports the use of injury prevention programs for preventing overall knee injuries in this athletic population, a nonsignificant reduction in ACL injuries was observed through meta-analytical techniques. Similar to the authors of previous studies,16 we cannot say with much confidence what components of the injury prevention programs are most effective; however, neuromuscular training has showed promising results.30 Given the logistical challenges of conducting highquality RCTs on injury prevention in this population, we applaud the researchers who have performed these much-needed scientific studies and urge future investigators to follow the standards set forth by the CONSORT (Consolidated Standards of Reporting Trials) statement40 for reporting this research in our athletic population.
ACKNOWLEDGMENT The authors thank Mary McFarland of the Eccles Health Science Medical Library for assisting with the development of a sensitive search strategy for this study.
Soccer Injury Prevention: Meta-analysis
2055
An online CME course associated with this article is available for 1 AMA PRA Category 1 CreditTM at http://ajsm-cme.sagepub.com. In accordance with the standards of the Accreditation Council for Continuing Medical Education (ACCME), it is the policy of The American Orthopaedic Society for Sports Medicine that authors, editors, and planners disclose to the learners all financial relationships during the past 12 months with any commercial interest (A ‘commercial interest’ is any entity producing, marketing, re-selling, or distributing health care goods or services consumed by, or used on, patients). Any and all disclosures are provided in the online journal CME area which is provided to all participants before they actually take the CME activity. In accordance with AOSSM policy, authors, editors, and planners’ participation in this educational activity will be predicated upon timely submission and review of AOSSM disclosure. Noncompliance will result in an author/editor or planner to be stricken from participating in this CME activity.
REFERENCES 1. Big count. FIFA Magazine. FIFA; 2007. http://www.fifa.com/mm/document/fifafacts/bcoffsurv/emaga_9384_10704.pdf. Accessed January 1, 2014. 2. Agel J, Evans TA, Dick R, Putukian M, Marshall SW. Descriptive epidemiology of collegiate men’s soccer injuries: National Collegiate Athletic Association Injury Surveillance System, 1988-1989 through 2002-2003. J Athl Train. 2007;42(2):270-277. 3. Agel J, Olson DE, Dick R, Arendt EA, Marshall SW, Sikka RS. Descriptive epidemiology of collegiate women’s basketball injuries: National Collegiate Athletic Association Injury Surveillance System, 1988-1989 through 2003-2004. J Athl Train. 2007;42(2):202-210. 4. Agel J, Palmieri-Smith R, Dick R, Wojtys EM, Marshall SW. Descriptive epidemiology of collegiate women’s volleyball injuries: National Collegiate Athletic Association Injury Surveillance System, 19881989 through 2003-2004. J Athl Train. 2007;42:295-302. 5. Arendt E, Dick R. Knee injury patterns among men and women in collegiate basketball and soccer. NCAA data and review of literature. Am J Sports Med. 1995;23(6):694-701. 6. Berlin JA. Does blinding of readers affect the results of metaanalyses? University of Pennsylvania Meta-analysis Blinding Study Group. Lancet. 1997;350(9072):185-186. 7. Cahill BR, Griffith EH. Effect of preseason conditioning on the incidence and severity of high school football knee injuries. Am J Sports Med. 1978;6(4):180-184. 8. Caine D, Caine C, Maffulli N. Incidence and distribution of pediatric sport-related injuries. Clin J Sport Med. 2006;16(6):500-513. 9. Caraffa A, Cerulli G, Projetti M, Aisa G, Rizzo A. Prevention of anterior cruciate ligament injuries in soccer: a prospective controlled study of proprioceptive training. Knee Surg Sports Traumatol Arthrosc. 1996;4(1):19-21. 10. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7(3):177-188. 11. Dick R, Putukian M, Agel J, Evans TA, Marshall SW. Descriptive epidemiology of collegiate women’s soccer injuries: National Collegiate Athletic Association Injury Surveillance System, 1988-1989 through 2002-2003. J Athl Train. 2007;42(2):278-285. 12. Ekstrand J, Gillquist J, Liljedahl SO. Prevention of soccer injuries: supervision by doctor and physiotherapist. Am J Sports Med. 1983;11(3):116-120. 13. Emery CA, Meeuwisse WH. The effectiveness of a neuromuscular prevention strategy to reduce injuries in youth soccer: a clusterrandomised controlled trial. Br J Sports Med. 2010;44(8):555-562.
2056 Grimm et al
14. Engebretsen AH, Myklebust G, Holme I, Engebretsen L, Bahr R. Prevention of injuries among male soccer players: a prospective, randomized intervention study targeting players with previous injuries or reduced function. Am J Sports Med. 2008;36(6):1052-1060. 15. Ettlinger CF, Johnson RJ, Shealy JE. A method to help reduce the risk of serious knee sprains incurred in alpine skiing. Am J Sports Med. 1995;23(5):531-537. 16. Gagnier JJ, Morgenstern H, Chess L. Interventions designed to prevent anterior cruciate ligament injuries in adolescents and adults: a systematic review and meta-analysis. Am J Sports Med. 2013;41(8):1952-1962. 17. Gilchrist J, Mandelbaum BR, Melancon H, et al. A randomized controlled trial to prevent noncontact anterior cruciate ligament injury in female collegiate soccer players. Am J Sports Med. 2008;36(8):1476-1483. 18. Grimm NL, Shea KG, Leaver RW, Aoki SK, Carey JL. Efficacy and degree of bias in knee injury prevention studies: a systematic review of RCTs. Clin Orthop Relat Res. 2013;471(1):308-316. 19. Grindstaff TL, Hammill RR, Tuzson AE, Hertel J. Neuromuscular control training programs and noncontact anterior cruciate ligament injury rates in female athletes: a numbers-needed-to-treat analysis. J Athl Train. 2006;41(4):450-456. 20. Heidt RS Jr, Sweeterman LM, Carlonas RL, Traub JA, Tekulve FX. Avoidance of soccer injuries with preseason conditioning. Am J Sports Med. 2000;28(5):659-662. 21. Hewett TE, Ford KR, Myer GD. Anterior cruciate ligament injuries in female athletes: part 2, a meta-analysis of neuromuscular interventions aimed at injury prevention. Am J Sports Med. 2006;34(3):490-498. 22. Hewett TE, Lindenfeld TN, Riccobene JV, Noyes FR. The effect of neuromuscular training on the incidence of knee injury in female athletes: a prospective study. Am J Sports Med. 1999;27(6):699-706. 23. Higgins JP, Thompson SG. Quantifying heterogeneity in a metaanalysis. Stat Med. 2002;21(11):1539-1558. 24. Higgins JPT, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions, Version 5.1.0. The Cochrane Collaboration; 2011. www.cochrane-handbook.org. Updated March 2011. Accessed January 1, 2014. 25. Junge A, Rosch D, Peterson L, Graf-Baumann T, Dvorak J. Prevention of soccer injuries: a prospective intervention study in youth amateur players. Am J Sports Med. 2002;30(5):652-659. 26. Kiani A, Hellquist E, Ahlqvist K, Gedeborg R, Michaelsson K, Byberg L. Prevention of soccer-related knee injuries in teenaged girls. Arch Intern Med. 2010;170(1):43-49. 27. Kunz R, Oxman AD. The unpredictability paradox: review of empirical comparisons of randomised and non-randomised clinical trials. BMJ. 1998;317(7167):1185-1190. 28. Magee CD, Byars LA, DeZee KJ. Limitations of subgroup analyses in meta-analysis of cardiac resynchronization therapy by QRS duration. Arch Intern Med. 2012;172(4):375; author reply 375-376. 29. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol. 2009;62(10):1006-1012. 30. Myer GD, Sugimoto D, Thomas S, Hewett TE. The influence of age on the effectiveness of neuromuscular training to reduce anterior cruciate ligament injury in female athletes: a meta-analysis. Am J Sports Med. 2013;41(1):203-215. 31. Myklebust G, Engebretsen L, Braekken IH, Skjolberg A, Olsen OE, Bahr R. Prevention of anterior cruciate ligament injuries in female team handball players: a prospective intervention study over three seasons. Clin J Sport Med. 2003;13(2):71-78. 32. Olsen OE, Myklebust G, Engebretsen L, Holme I, Bahr R. Exercises to prevent lower limb injuries in youth sports: cluster randomised controlled trial. BMJ. 2005;330(7489):449. 33. Peters JL, Sutton AJ, Jones DR, Abrams KR, Rushton L. Comparison of two methods to detect publication bias in meta-analysis. JAMA. 2006;295(6):676-680. 34. Pfeiffer RP, Shea KG, Roberts D, Grandstrand S, Bond L. Lack of effect of a knee ligament injury prevention program on the incidence of noncontact anterior cruciate ligament injury. J Bone Joint Surg Am. 2006;88(8):1769-1774.
The American Journal of Sports Medicine
35. Poolman RW, Struijs PA, Krips R, et al. Reporting of outcomes in orthopaedic randomized trials: does blinding of outcome assessors matter? J Bone Joint Surg Am. 2007;89(3):550-558. 36. Powell JW, Barber-Foss KD. Injury patterns in selected high school sports: a review of the 1995-1997 seasons. J Athl Train. 1999;34(3): 277-284. 37. Renstrom P, Ljungqvist A, Arendt E, et al. Non-contact ACL injuries in female athletes: an International Olympic Committee current concepts statement. Br J Sports Med. 2008;42(6):394-412. 38. Rothstein HR, Sutton AJ, Borenstein M. Publication Bias in Metaanalysis—Prevention, Assessment and Adjustments. West Sussex, UK: John Wiley & Sons; 2005. 39. Schulz KF, Altman DG, Moher D. Allocation concealment in clinical trials. JAMA. 2002;288(19):2406-2407; author reply 2408-2409. 40. Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. J Clin Epidemiol. 2010;63(8):834-840. 41. Schulz KF, Chalmers I, Altman DG. The landscape and lexicon of blinding in randomized trials. Ann Intern Med. 2002;136(3):254259. 42. Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias: dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA. 1995;273(5): 408-412. 43. Schulz KF, Grimes DA. Allocation concealment in randomised trials: defending against deciphering. Lancet. 2002;359(9306):614-618. 44. Schulz KF, Grimes DA. Blinding in randomised trials: hiding who got what. Lancet. 2002;359(9307):696-700. 45. Schulz KF, Grimes DA. Multiplicity in randomised trials II: subgroup and interim analyses. Lancet. 2005;365(9471):1657-1661. 46. Shea KG, Grimm NL, Ewing CK, Aoki SK. Youth sports anterior cruciate ligament and knee injury epidemiology: who is getting injured? In what sports? When? Clin Sports Med. 2011;30(4):691-706. 47. Soderman K, Werner S, Pietila T, Engstrom B, Alfredson H. Balance board training: prevention of traumatic injuries of the lower extremities in female soccer players? A prospective randomized intervention study. Knee Surg Sports Traumatol Arthrosc. 2000;8(6):356363. 48. Soligard T, Myklebust G, Steffen K, et al. Comprehensive warm-up programme to prevent injuries in young female footballers: cluster randomised controlled trial. BMJ. 2008;337:a2469. 49. Steffen K, Myklebust G, Olsen OE, Holme I, Bahr R. Preventing injuries in female youth football—a cluster-randomized controlled trial. Scand J Med Sci Sports. 2008;18(5):605-614. 50. van Beijsterveldt AM, van de Port IG, Krist MR, et al. Effectiveness of an injury prevention programme for adult male amateur soccer players: a cluster-randomised controlled trial. Br J Sports Med. 2012; 46(16):1114-1118. 51. van Tulder M, Furlan A, Bombardier C, Bouter L. Updated method guidelines for systematic reviews in the Cochrane Collaboration back review group. Spine. 2003;28(12):1290-1299. 52. Walden M, Atroshi I, Magnusson H, Wagner P, Hagglund M. Prevention of acute knee injuries in adolescent female football players: cluster randomised controlled trial. BMJ. 2012;344:e3042. 53. Wedderkopp N, Kaltoft M, Lundgaard B, Rosendahl M, Froberg K. Prevention of injuries in young female players in European team handball: a prospective intervention study. Scand J Med Sci Sports. 1999;9(1):41-47. 54. Wood L, Egger M, Gluud LL, et al. Empirical evidence of bias in treatment effect estimates in controlled trials with different interventions and outcomes: meta-epidemiological study. BMJ. 2008;336 (7644):601-605. 55. Wright RW, Brand RA, Dunn W, Spindler KP. How to write a systematic review. Clin Orthop Relat Res. 2007;455:23-29. 56. Yoo JH, Lim BO, Ha M, et al. A meta-analysis of the effect of neuromuscular training on the prevention of the anterior cruciate ligament injury in female athletes. Knee Surg Sports Traumatol Arthrosc. 2010;18(6):824-830.
For reprints and permission queries, please visit SAGE’s Web site at http://www.sagepub.com/journalsPermissions.nav
