Abstract
Background:
Anterior cruciate ligament (ACL) injuries are among the most distressing injuries for collegiate varsity athletes. Identifying easily attainable clinicodemographic risk factors in this subgroup can help screen for high-risk athletes who may benefit from proven ACL injury risk reduction programs.
Purpose:
To identify clinicodemographic risk factors for noncontact ACL injury among female and male collegiate varsity athletes from 10 different sports.
Study Design:
Cohort study; Level of evidence, 2.
Methods:
A total of 777 (276 female and 501 male athletes) collegiate varsity athletes from 3 consecutive seasons had an extended panel of clinicodemographic parameters recorded at their respective preseason physical sessions. The athletes were followed for 1 athletic season for noncontact ACL injuries.
Results:
Fifteen (6 female and 9 male athletes) athletes suffered a noncontact ACL injury during their season. Among all athletes, previous lower limb surgery and cutting sport participation were significantly associated with an increased risk of noncontact ACL injury. Among female athletes, previous ACL injury and previous lower limb surgery were significant risk factors. No significant clinicodemographic risk factors were identified in male athletes. Female sex was not a significant risk factor for noncontact ACL injury.
Conclusion:
The clinicodemographic risk factors for noncontact ACL injury identified in this study are easily attainable and may guide preseason screening for ACL injury risk in collegiate varsity athletes. The lack of association of these risk factors in male athletes may highlight the need to focus on other factors such as kinematics for these athletes.
Anterior cruciate ligament (ACL) injuries are common but devastating, with nearly a quarter of a million injuries occurring every year in North America. 45 Of these injuries, up to 70% occur by a noncontact mechanism. 8 ACL injuries are often season-ending, with up to 20% of athletes never returning to high-risk activities after injury and up to 1 in 5 athletes suffering reinjury. 4 In the long term, ACL injuries are associated with increased morbidity and incidence of osteoarthritis.16,25,27,30,38 Considering the suboptimal prognosis after ACL injury, high-risk athletes would benefit from early identification with targeted injury risk reduction interventions. 11
ACL injury risk is complex and multifactorial, with previous investigations highlighting an interplay between knee anatomy, landing kinematics, sex, type of sport participation, and hormonal factors.12,20,41,46,49 However, identifying many of these risk factors requires time and access to resources, which is not necessarily feasible for large-scale screening. Thus, risk factors that are easily identifiable with few resources—such as clinicodemographic parameters obtainable through history and chart review—are desirable. While there exists a notable body of literature investigating certain clinicodemographic parameters—including sex, sports participation, and previous ACL injury—as risk factors for noncontact ACL injury, study populations and risk findings vary between studies (elite athletes, collegiate athletes, high school athletes).9,34,57 Furthermore, no studies to our knowledge have performed a more comprehensive investigation that includes other easily attainable clinicodemographic parameters—such as history of previous knee injury (Ki), previous lower limb injury (LLi), or previous lower limb surgery (LLS)—exclusively in collegiate varsity athletes.
In the present study, we aimed to comprehensively investigate easily attainable clinicodemographic parameters in collegiate varsity athletes to identify pertinent ACL injury risk factors and sex-related differences that may guide and complement widespread screening for high-risk athletes.
Methods
Participants and Injury Surveillance
This investigation was a prospective 3-cohort study. Ethics approval was obtained from the research ethics board at our institution and informed consent was obtained from all participants in this study. Three cohorts of collegiate varsity athletes from consecutive seasons (2017-2018, 2018-2019, and 2021-2022) were included in the study. The 2019-2021 seasons were cancelled because of the coronavirus disease 2019 pandemic. Athletes were enrolled in the study in advance of their respective athletic seasons. Only athletes without an LLi at the time of enrollment were assessed for eligibility. The exclusion criteria were athletes <18 or >30 years old. Athletes initially enrolled in their respective varsity sports for multiple seasons were only included in the most recent season or in the season that they suffered an ACL injury if 1 occurred. Athletes were observed until the conclusion of their sports seasons with regular communications with team coaches and trainers. A total of 777 participants were enrolled in the study over the 3 seasons.
Clinicodemographic parameters were prospectively collected at preseason physical sessions, including previous ACL injury, previous Ki, previous LLi, previous LLS, previous concussion, number of previous concussions, varsity sport, and cutting sport (CS). The following anthropometric measurements were also obtained: height, weight, body mass index (BMI), and sex. Previous Ki and previous LLi were defined as any ligamentous sprain, muscle strain, or fracture involving the knee joint or lower limb, respectively, before enrollment. Previous LLS was defined as any major surgery involving any bone or ligament in the lower extremity—including the hip, knee, or ankle joint. A previous ACL injury was also recorded as a previous Ki and LLi, and a previous ACL reconstruction surgery was recorded as a previous LLS. The severity of previous injuries, in the form of missed training time or otherwise, was not specified. Cutting sports were defined as sports involving frequent changes of direction and/or pivot movements and included the following: basketball, soccer, football, rugby, and lacrosse. Prophylactic bracing was not routinely used by football athletes at the institution where this study was performed, given the limited definitive data regarding benefits. 7 Suspected ACL injuries were confirmed using magnetic resonance imaging and, for those undergoing ACL reconstruction, through direct visualization with arthroscopy. Noncontact ACL injuries were defined as those occurring without a direct impact on the affected limb.
Statistical Analysis
Statistical analyses were performed using SPSS Version 27 (IBM). The Shapiro-Wilk test was used to determine whether continuous parameters had parametric distributions. Parametric parameters were compared between ACL-injured athletes (athletes who suffered noncontact ACL tears during the study period) and uninjured athletes using the independent-samples t test with Glass delta effect sizes. Nonparametric parameters were compared using the Mann-Whitney U test with Mann-Whitney U effect sizes. Binary and nonbinary categorical parameters were compared between groups using the Fisher exact test and the Pearson chi-square test, respectively. Cases with missing data were addressed with pairwise deletion. These analyses were performed on the entire cohort and female and male athlete subgroups, respectively. Statistical significance was set as P < .05.
Results
A total of 777 participants were included in this study, including 276 female and 501 male collegiate varsity athletes (Figure 1). The follow-up time was 1 varsity athletic season for each participant. Participant characteristics are presented in Table 1 and the distribution of varsity sports participation is summarized in Table 2. At the end of the study period, 15 athletes (1.93%) had sustained noncontact ACL tears. All ACL tears were confirmed using magnetic resonance imaging or by direct visualization during arthroscopy.
STROBE flowchart. STROBE, Strengthening the Reporting of Observational Studies in Epidemiology.
Participant Characteristics a
Data are presented as median (range) unless otherwise indicated. ACL, anterior cruciate ligament; BMI, body mass index.
Height, weight, and BMI data were unavailable for 1 female athlete.
Distribution of Varsity Sports a
ACL, anterior cruciate ligament. Dashes indicate that no varsity team existed for the sex listed in that column.
Continuous Clinicodemographic Parameters
All continuous clinicodemographic parameters were nonparametrically distributed. In all athletes, female athletes, and male athletes, no significant difference was observed in height (P = .419; P = .523; P = .258), weight (P = .777; P = .496; P = .479), or BMI (P = .242; P = .728; P = .091) between ACL-injured athletes and uninjured athletes (Table 3).
Comparison of Continuous Variables Between ACL-injured and Uninjured Athletes a
ACL, anterior cruciate ligament, BMI, body mass index. Data are reported as the mean, unless otherwise specified.
Categorical Clinicodemographic Parameters
In all athletes, a history of previous LLS and participation in a CS was significantly associated with increased ACL injury risk (P = .044; P = .027) (Table 4). Female sex was not significantly associated with increased risk for ACL injury in all athletes and in sex-comparable sports exclusively (P = .787; P = .311) (Table 4). In female athletes, a history of previous ACL injury and previous LLS were both significantly associated with increased ACL injury risk (P = .020; P = .010) (Table 4). In male athletes, no significant difference was found in categorical clinicodemographic parameters between ACL-injured athletes and uninjured athletes.
Comparison of Categorical Variables Between ACL-injured and Uninjured Athletes a
Data are presented as n (%) or n (range). ACL, anterior cruciate ligament; ACLi, anterior cruciate ligament injury; Ki, knee injury; LLi, lower limb injury; LLS, lower limb surgery. Dashes were utilized for “Sport” as this data was nominal.
Data were unavailable for 1 female and 1 male participant.
Data were unavailable for 1 male participant.
Discussion
The most important findings of the present study were that previous LLS and CS participation were significant ACL injury risk factors when female and male athletes were grouped, previous ACL injury and previous LLS were significant risk factors in female athletes, and no clinicodemographic risk factors were identified in male athletes. Female sex, anthropometric features, and concussion history were not significantly associated with noncontact ACL injury.
Previous LLS as an ACL Injury Risk Factor
The presence of a previous LLS as a statistically significant risk factor for noncontact ACL injury in all athletes and female athletes presents an interesting preliminary finding in ACL injury risk factors. While some studies have reported an increased risk of non-ACL leg injuries after ACL injury and reconstruction,35,53,56 there is a scarcity of research investigating the risk of ACL injuries in relation to non-ACL surgical history. A 2020 systematic review by Vereijken et al 55 found no studies reporting on functional performance after return to performance for lower extremity injuries other than after anterior or posterior cruciate ligament reconstruction. To our knowledge, the only other study to report on non-ACL surgery and injury as potential ACL injury risk factors was by Beynnon et al, 6 who reported that previous non-ACL Ki increased ACL injury risk in female athletes. In the present study, 3 of the 4 ACL-injured athletes with a history of LLS were female athletes, and 2 of the 4 athletes had a history of ACL reconstruction surgery. Further, this association may be statistically fragile given the P value of .044, which becomes nonsignificant with the removal of 1 ACL-injured athlete with a history of LLS (P = .133). With the limited number of cases in this study, we propose a potential association between previous LLS and ACL injury risk that could be mediated by altered landing mechanics after non-ACL LLSs,2,15,47,51,59,62 psychological distress after surgeries,14,39 premature return to sports, or insufficient rehabilitation. However, further studies with larger cohorts are required to confirm or characterize this potential relationship.
Sport Participation
Five out of the 10 sports included in the present investigation (basketball, soccer, football, rugby, and lacrosse) were classified as cutting sports, based on the quick directional changes and pivoting movements that are involved. Fourteen out of the 15 ACL-injured athletes participated in a CS, while the other 1 athlete participated in hockey. Accordingly, CS participation was a significant risk factor for noncontact ACL injury. These cutting sports are often reported to be the highest risk for ACL injury in the literature but with notable sex differences. In the epidemiologic study on collegiate varsity athletes by Hootman et al, 22 men's spring football, women's gymnastics, women's soccer, women's basketball, and men's football had the highest rates of ACL injury. Similarly, Joseph et al 24 reported the highest ACL injury rates per athlete exposure in girls’ soccer, football, and girls’ basketball. Interestingly, boys’ basketball had among the lowest ACL injury rates in both studies.22,24 In their systematic review, Gornitzky et al 17 reported the highest rates of ACL injury in soccer and basketball for women and football and lacrosse for men. Contrary to their study, no lacrosse athlete suffered an ACL injury in our study group. The incidence of ACL injury per season in the present study ranged from 1.52% in men's soccer to 8% in women's basketball, with the second highest incidence being 3.47% in men's football. These values are elevated in comparison with previously reported values, ranging from 0.1% in women's volleyball to 1.1% in women's soccer. 17 We hypothesized that the elevated incidence can be explained, in part, by the documented increase in rates of ACL injury in collegiate basketball, football, and volleyball. 1 While CS participation did not remain a significant risk factor in sex-specific subgroups and specific sport participation was not identified as a risk factor at all, these analyses were limited by a small number of injuries per sport. Nonetheless, the present study confirms previously reported patterns of sport-specific ACL injury risk.
Primary Versus Secondary ACL Injury Risk
Interestingly, while previous ACL reconstruction surgeries were included as previous LLSs, a history of previous ACL injury itself was only a significant risk factor for noncontact ACL injury in female athletes. A recent systematic review and meta-analysis by Hong et al 21 reported a similar pattern, with a higher secondary ACL injury incidence rate in female soccer players (27%) compared with male soccer players (10%). The increased risk of secondary ACL injury versus initial ACL injury in female athletes may mediated by the persistence of intrinsic risk factors that contributed to initial ACL injury, which have been found to be more pronounced in female athletes. These include, but are not limited to, anatomical features such as greater femoral plateau angle and posterior tibial slope angle, as well as biomechanical properties such as decreased hip abduction strength, greater hip flexion and knee abduction during landing, and core instability.26,28,50,52,60 Ultimately, the true clinical significance and degree of sex-specificity of these various factors remains to be determined. Additional factors—including graft type 49 and lack of lateral extra-articular procedure23,43 for graft failures—may also play a part, although no clear sex differences have been demonstrated at this time. A hypothetical decrease in ACL reinjury rates may be further mediated by advancements in functional performance tests— including isokinetic testing—and the development of tools to guide decisions to return to sports—such as the Limb Symmetry Index.13,18,37 However, it is unclear whether these interventions impact male and female athletes differently, given the paucity of literature on the actual effect of return-to-play decision modalities on the rate of subsequent reinjury. Ultimately, further research is necessary to better understand why secondary ACL injury risk is more pronounced in female collegiate varsity athletes compared with their male counterparts to allow for targeted intervention at the time of ACL reconstruction and during postoperative rehabilitation.
Female Sex and ACL Injury Risk
The findings of the present study present a complex picture of ACL injury risk factors between female and male athletes. While a greater percentage of female athletes suffered noncontact ACL injuries in all sports combined (2.17% vs 1.80%) and sex-comparable sports (2.49% vs 1.01%), there was no significant association between female sex and increased risk of noncontact ACL injury in either case. In their epidemiologic study, Joseph et al 24 also found no significant sex difference in ACL injury rates when looking at all sports but found a significant difference when limiting analyses to sex-comparable sports. A similar pattern was reported by Mountcastle et al 36 in their epidemiologic study. In contrast, other studies have reported up to an 8-fold increased risk for noncontact ACL tears in female athletes.41,48,54 It has been thought that these differences may be secondary to anatomical and biomechanical differences,19,29 hormonal influences, 61 and more recently, even gendered environmental influences through the “entanglement” of sex and gender. 40 This divergence in findings underscores a complex, heterogeneous relationship between sex and ACL injury risk that depends on various contextual factors such as age, level of competition, and sports participation. Considering that a nonsignificantly greater percentage of female athletes suffered ACL injuries in the present study, the absence of sex as a significant risk factor is likely secondary to the relatively smaller cohort size, limited number of ACL injuries, and the fact that football accounted for 6 of the 9 noncontact ACL injuries in male athletes.
Male-Specific Risk Factors
While a history of LLS was a significant risk factor when female and male athletes were grouped, none of the clinicodemographic parameters were associated with an increased risk for noncontact ACL injury in the male athlete subgroup. Of note, while specific sport participation and CS participation were not significant risk factors in male athletes in the present study, all 9 male athletes who sustained a noncontact ACL injury participated in cutting sports (6 football athletes, 2 basketball athletes, and 1 soccer athlete). Thus, it is possible that the study was underpowered to detect a significant difference, resulting in a type 2 error. Nonetheless, while there is an expanding body of literature regarding increased ACL injury rates among female athletes, 48 the nature of male-specific risk factors for ACL injury remains more enigmatic. Terauchi et al 50 reported a similar trend in their cross-sectional study, where they found significantly greater femoral plateau angle and tibial posterior slope angle in female ACL-deficient athletes compared with negative controls, yet these differences were not seen in male athletes. A similar pattern was again reported by Todd et al. 52 Overall, the results of the present study indicate that factors beyond clinicodemographic parameters— such as jump landing kinematics 12 —likely play a greater role in noncontact ACL injury risk in male collegiate varsity athletes. As with female athletes, further investigation into male-specific ACL injury risk factors is also necessary to better understand their risk profiles and to facilitate screening for high-risk athletes who may benefit most from targeted injury prevention interventions.
Anthropometric Measurements
No anthropometric measurements were associated with an increased risk for noncontact ACL injury when looking at all athletes combined or female and male subgroups. In a prospective cohort study on military cadets, Uhorchak et al 54 found that a body weight or BMI >1 standard deviation above the mean carried relative risks of 3.2 or 3.5, respectively, for noncontact ACL injuries in female athletes, while neither were significant risk factors in male athletes. Beynnon et al 6 reported that increased body weight was associated with ACL injury in female and male athletes; however, this was found in 65% of a younger cohort of high school athletes and 35% of collegiate athletes. While ACL injury rates have been shown to increase with age in all athletes, with notably higher rates in female athletes immediately after growth spurts, these trends are most pronounced in adolescent age groups.44,58 Renstrom et al 42 found that female athletes had the most ACL reconstructions in the 15 to 19 years age range. 42 The effect of age on ACL injury risk is not as clearly described in the adult age range. The present study highlights that height, weight, and BMI do not play a significant role in the risk of noncontact ACL injury in collegiate varsity athletes, which we believe to be a result of the stabilization of physical growth and musculoskeletal changes. 32
Concussion and ACL Injury Risk
The relationship between concussion and musculoskeletal injuries—including ACL injuries—is a growing field of research within sports medicine. Considering the relationship between altered biomechanics and ACL injury risk, concussion history is thought to increase ACL injury risk through various neurocognitive deficits. 5 In a recent study on a general population, McPherson et al 33 reported increased odds for ACL injury among people with a history of concussion in the 3 preceding years compared with people without such history. 33 However, ACL injury mechanisms were not assessed. Conversely, Buckley et al 10 reported that concussion did not increase the risk for lower extremity musculoskeletal injury in American collegiate athletes. In a study focused on female collegiate athletes from 5 different sports, only lacrosse athletes had an increased risk of ACL tear in the 1 year after a concussion. 31 In the present study, the time frame of the athletes’ concussion histories was not factored into the analyses. With this in consideration, a history of concussion and an increased number of concussions were not significantly associated with an elevated risk for noncontact ACL injury. However, there may have been large disparities between athletes with positive concussion histories, as those with more remote concussions may have been closer to their theoretical baseline risk for ACL injury. While the characterization of concussion history and ACL injury risk continues to grow, the literature at present remains sparse with diverse study populations, study periods, and analytical methods. Future cohort studies with large cohorts, detailed data on the time of concussion, and adequate follow-up periods are required to better understand the relationship between concussion and ACL injury risk and address the potential role of adapted return to play protocols. 3
Limitations
The present study was not without limitations. Despite the relatively large cohort of collegiate varsity athletes, the number of ACL injuries was small, potentially reducing statistical power to detect significant associations. Further, there may have been an element of bias introduced into the study by utilizing data from the most recent season for noninjured multiseason athletes and data from the injury year for ACL-injured athletes. Data for 5 of the 15 ACL-injured athletes were not from their most recent season. Given the small number of injuries, we believe that the risk of selection bias was minimal and acceptable in this context to allow for a more extensive comparison of clinicodemographic data between ACL-injured and uninjured athletes. As ACL injury data were collected for 1 season after each preseason assessment, some high-risk athletes may continue to be at risk after the study period and suffer an ACL injury in subsequent seasons. Ideally, future studies should aim to incorporate larger cohorts of collegiate varsity athletes and have longer follow-up periods to increase statistical power. Other risk factors that are feasibly screened for at preseason physicals—such as drop vertical jump parameters as demonstrated by our group 12 —should be studied in conjunction with clinicodemographic parameters to obtain more granularity in ACL injury risk profiles to ultimately develop a comprehensive ACL injury risk assessment tool.
Conclusion
Previous LLS and CS participation were significant risk factors for noncontact ACL injury when female and male athletes were grouped, while previous ACL injury and previous LLS were significant risk factors among female athletes. Clinicodemographic parameters likely play a smaller role in ACL injury risk among male athletes, as no significant risk factors were identified in this subgroup. Female sex, anthropometric features, and concussion history did not increase the risk of noncontact ACL injury with the numbers available in the current study. These results may ultimately guide the development of preseason ACL injury risk screening programs for collegiate varsity athletes. They also highlight the necessity for complementary risk assessment modalities—such as kinematic assessment—using motion capture systems, particularly in male athletes.
Footnotes
Acknowledgements
The authors thank MEDTEQ+, Emovi Inc, Consultation Semperform Inc, and the Montreal General Hospital Foundation for their support of this research endeavor.
Final revision submitted March 23, 2024; accepted May 14, 2024.
One or more of the authors has declared the following potential conflict of interest or source of funding: J.P.A.H.C. has received education payments from Saxum Surgical. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.
Ethical approval for this study was obtained from McGill University (A02-M35-17A).
