Ten-Year Epidemiology of Ankle Injuries in Men's and Women's Collegiate Soccer Players

This study was deemed acceptable by the Mayo Clinic Institutional Review Board and the NCAA Research Review Board (No. 16-006901). Deidentified injury data on NCAA ankle injuries from the NCAA Injury Surveillance Program (ISP) were used in the analysis. The Datalys Center for Sports Injury Research and Prevention, Inc (Indianapolis, IN), an independent, nonprofit research organization, received data from NCAA soccer ATs and physicians, analyzed them, and reviewed them to ensure quality.

We briefly describe the inclusion criteria and methods of the NCAA-ISP as they pertain to this research, but more detailed information can be found in “National Collegiate Athletic Association Injury Surveillance System: Review of Methods for 2004–2005 Through 2013–2014 Data Collection.”¹⁷ 

Athletic trainers reported injury data during organized practices and games from the participating NCAA sports programs. Each game or practice counted as 1 AE. During the 2004–2005 through 2008–2009 seasons, ATs used written questionnaires that they faxed or mailed to the NCAA. During the 2009–2010 through 2013–2014 seasons, ATs used electronic medical records to collect and organize the data. They could return and add or change data throughout the season as they saw fit. Common data elements were deidentified, collected, and added to a central aggregate research database. Data sent to the database were verified, evaluated for consistency, and flagged when invalid values were recorded. The ATs and quality-assurance staff worked together to resolve any concerns about data validity. Consistent and usable data were then added to the aggregate research database.

An AE was reported as 1 student-athlete participating in 1 NCAA-sanctioned practice or game. Injuries that were reported included those that occurred during a team-sanctioned practice or game and were evaluated by an AT or physician. The structure injured was determined via evaluation by an AT or team physician, with or without the aid of magnetic resonance imaging.

We analyzed injury rates using number of injuries per 1000 AEs; t tests were conducted to compare means, rate ratios (RRs) to compare injury rates, and injury proportion ratios (IPRs) to compare the proportions of injuries that had certain characteristics. Two examples of RR and IPR equations used to compare men and women are shown. Similar equations were used to compare the anatomical structures injured and the 2004–2005 through 2008–2009 versus 2009–2010 through 2013–2014 seasons. A 95% confidence interval (CI) and P value <.001 were used in the analyses to indicate statistical significance.

On average, 751 men's and 926 women's NCAA soccer programs participated each season during the 2004–2005 through 2008–2009 seasons and 802 men's and 926 women's programs participated per season during the 2009–2010 through 2013–2014 seasons.¹⁷  For the 2004–2005 through 2013–2014 seasons, the NCAA-ISP contains data for 1 459 186 AEs for male and female soccer players. During these 1 459 186 AEs, 2068 ankle injuries were reported, which equates to 1.42 ankle injuries for every 1000 AEs.

Ankle injuries among male and female soccer players are compared in Tables 1 through 3.

The 2004–2005 through 2008–2009 and 2009–2010 through 2013–2014 seasons are described in Tables 4 and 5 and the Figure. Compared with the 2004–2005 through 2008–2009 seasons, the injuries during the 2009–2010 through 2013–2014 seasons had better prognoses. Injuries during the 2009–2010 through 2013–2014 seasons resulted in more players missing <7 days and fewer players missing 7 to 13 days, 14 to 29 days, and ≥30 days postinjury. Also, the average number of days missed was different between these time periods: 11.06 ± 22.67 in 2003–2004 through 2008–2009 versus 5.09 ± 9.45 in 2009–2010 through 2013–2014 (P < .001). In addition, injuries during the 2009–2010 through 2013–2014 seasons were less likely to be recurrent.

The injured anatomical structures are compared in Tables 6 through 8. The most common ankle injury was an LLC tear (partial or complete), which accounted for 65.67% (1358/2068) of ankle injuries. Other frequent ankle injuries were tibiofibular ligament (high ankle) sprains (10.3%, 212/2068), ankle contusions (10.1%, 208/2068), and medial (deltoid) ligament tears (9.77%, 202/2068). Further ankle injuries were reported but not included in the data analysis due to their rarity; when combined, they accounted for 4.3% (88/2068) of the total. Examples of these injuries were synovitis (n = 7), lateral malleolar fracture (n = 5), bursitis (n = 5), medial malleolar fracture (n = 4), subluxation (n = 2), dislocation (n = 1), talar fracture (n = 1), and “other ankle injury” (n = 28).

When compared with the other 3 most common ankle injuries, high ankle sprains resulted in a greater proportion of athletes missing ≥30 days of participation (IPR = 1.90, 23/202 versus 112/1866, 95% CI = 1.24, 2.90) and a lower proportion of athletes who returned to activity in <7 days (IPR = 0.61, 78/206 versus 1158/1862, 95% CI = 0.51, 0.73).

Ankle injuries occurred at a rate of 3.05 per 1000 AEs during games and 0.93 per 1000 AEs during practices. When compared with practices, games resulted in an ankle-injury RR of 3.33 (95% CI = 3.05, 3.63). New injuries constituted 84.3% (n = 1767) of total ankle injuries, whereas 15.7% (n = 328) were recurrent.

For every 1000 AEs, injuries occurred at a rate of 1.30 during the preseason, 1.51 during the season, and 0.95 during the postseason. A player was more likely to injure the ankle during the season than during the preseason (RR = 1.17; 95% CI = 1.05, 1.29) and more likely to injure the ankle during the preseason than during the postseason (RR = 1.37; 95% CI = 1.09, 1.71). However, when we looked only at practice injuries, the rate was 1.19 injuries per 1000 AEs during the preseason and 0.76 injuries per 1000 AEs during the season. The RR of injury during preseason practices compared with season practices was 1.57 (95% CI = 1.39, 1.79).

From 2004–2005 through 2013–2014, a total of 943 ankle injuries occurred during games. The second half accounted for 56.8% (536/943) and the first half for 43.2% (407/943), suggesting that ankle injuries were more likely to occur during the second half (RR = 1.32; 95% CI = 1.20, 1.44).

The 3 most common mechanisms of ankle injuries were noncontact (44.87%, 928/2068), contact with another player (44.73%, 925/2068), and contact with an apparatus, presumably the ball or goalpost (9.48%, 196/2068).

Of the 2068 total ankle injuries, days lost were reported for 1982. Of these, 58.93% (1168/1982) resulted in 0 to 6 days missed, 22.20% (440/1982) resulted in 7 to 13 days missed, 13.22% (262/1982) resulted in 14 to 29 days missed, and 5.65% (112/1982) resulted in ≥30 days missed.

Our research showed how the time period (2004–2005 through 2008–2009 versus 2009–2010 through 2013–2014 seasons), anatomical structure injured, and sex of the athlete affected the injury rate, mechanism, and prognosis.

These findings correlate with those of many other investigators^3,4,7,12,18  who found that ankle injuries during games occurred at a rate 3 times greater than during practices (RR = 3.33; 95% CI = 3.05, 3.63). Furthermore, these results align with data from other studies^19,20  suggesting that injuries occurred more often during the second half of soccer games than during the first half (RR = 1.32; 95% CI = 1.20, 1.44). The increased injury rate during the second half could be attributed to athlete fatigue.^19–21 

Regarding the time of year in which injuries occurred, soccer players were more likely to be injured during preseason practices than during regular season practices (RR = 1.57; 95% CI = 1.39, 1.79). This value is consistent with the results of other authors^3,4,7  and could indicate that players beginning the preseason are less conditioned and, therefore, more likely to be injured.

Fortunately, when ankle injuries occur in these athletes, they often result in only a brief period of time being missed. In fact, many soccer players returned to the event in which they sustained the ankle injury.^7,9  Furthermore, of the 1982 injuries for which we have a record of days lost, 58.93% (n = 1168) resulted in 0 to 6 days missed, and only 5.65% (n = 112) resulted in ≥30 days missed. However, the time lost varied significantly depending on the anatomical structure injured.

Of the 4 most common ankle injuries, contusions resulted in the smallest number of days missed (males = 4.33 days, females = 6.58 days) and high ankle sprains resulted in the most days missed (males = 12.86, females = 16.97; Table 3). Lateral ligamentous complex tears and medial (deltoid) tears were associated with similar numbers of days missed after injury (Table 3). Among the 4 frequent ankle injuries, high ankle sprains had the worst prognosis.^14,15  In our study, when compared with the other 3 common ankle injuries, high ankle sprains were most likely to cause an athlete to miss ≥30 days (IPR = 1.90; 95% CI = 1.24, 2.90). Also, only 37.9% (n = 78) of athletes who had a high ankle sprain returned within a week of injury. High ankle sprains are usually caused by eversion or external rotation (or both).^14,15  The prolonged period of pain and instability from these injuries can be attributed to the anterior tibiofibular ligament, which is typically injured during high ankle sprains. This ligament is a syndesmosis that keeps the tibia and fibula connected and functioning in unison.^14,15  As an athlete runs and cuts, large forces traverse the tibia and fibula, applying significant stress to the relatively small anterior tibiofibular ligament.^14,15  These large forces cause lingering periods of pain and instability after high ankle sprains.^14,15 

Similar to the current literature,^3,4,7  we found that the LLC was the most commonly injured anatomical structure, accounting for 65.67% (n = 1358) of ankle injuries. The mechanism of LLC injuries is inversion of the ankle joint.^7,14,18  Structurally, the fibula, which lies on the lateral aspect of the ankle, extends farther distally than the tibia, allowing the ankle to invert more readily than evert.⁹  This relatively easy inversion allows athletes to “roll” the ankle in supination, often resulting in an LLC tear.⁹  The LLC includes the anterior talofibular, calcaneofibular, and posterior talofibular ligaments.⁷  The anterior talofibular ligament is the weakest of the LLC ligaments and, therefore, the most likely to be injured during inversion.^9,14  Also, the anterior talofibular is the primary lateral ligament stressed in plantar flexion and inversion, which occurs when the ankle is in its most vulnerable position due to the narrower geometry of the posterior talus.^9,14 

Although some LLC tears resulted from contact with another player, they were more likely to occur without contact (Table 7). This could be due to hyperinversion, as when an athlete landed from a jump, cut, or stepped on uneven ground.^9,14  Conversely, medial ligament ankle tears were more likely to be due to contact with another player (Table 7). This can be explained, once again, by the relative length of the fibula compared with the tibia.⁹  Because the fibula extends farther than the tibia, hypereversion of the ankle is more difficult and more likely to require an external force, such as contact with another player.

Lateral ligament complex tears, deltoid tears, contusions, and high ankle sprains were more frequent during games than during practices, but contusions showed the highest RR during games (Table 6). This is probably because contusions were more likely to result from contact than noncontact (Table 7). Given that games are often associated with greater intensity than practices, games may also lead to increased contact, thereby causing a higher incidence of contact injuries.^3,4 

When we categorized the 4 most common ankle injuries as new or recurrent, all were typically described as new (Table 8). However, it is important to recognize that these data may be inaccurate because ATs and team physicians do not always know an athlete's medical history and may incorrectly classify a recurrent injury as new. Still, advances in physical therapy, strengthening exercises, taping, and bracing probably contributed to the lower number of recurrent injuries.

For males and females, the injury rates (Table 1) and the time to return to activity after ankle injury (Table 2) were similar. Yet males were more likely to be injured by contact with another player (IPR = 1.28; 95% CI = 1.16, 1.41) and while being slide tackled (IPR = 1.90; CI = 1.30, 2.78) and less likely to incur a noncontact ankle injury (IPR = 0.86; CI = 0.78, 0.95) or an injury due to contact with an apparatus (IPR = 0.57; 95% CI = 0.43, 0.76). Also, males were more likely to be injured during games (RR = 1.14; 95% CI = 1.01, 1.29) and to sustain contusions (IPR = 1.34; 95% CI = 1.03, 1.73). The similar rates of ankle injuries among NCAA soccer players have been well described in the literature.^3,4  The increase in game injuries, contact injuries, and contusion injuries may be due to the increased contact in men's sports compared with women's sports.^22,23 

In comparing the anatomical structure injured, males and females showed similar IPRs for high ankle sprains and medial (deltoid) tears (Table 2), but males were less likely to experience an LLC tear (IPR = 0.92; 95% CI = 0.86, 0.98). The slight increase in LLC tears among females could result from the increased LLC laxity that has been described previously.^22,23 

Prior authors^2,4,8,16  showed discrepancies in joint laxity, hormones, height, weight, strength, mechanical axis, and level of contact among males and females that may contribute to ankle injuries. Recognizing the differences in injuries between sexes may lead to the creation of and changes in sex-specific rule revisions and ankle exercises.

The rates of injuries were similar during the 2 time periods. However, the time missed and the proportions of recurrent injuries were different. Injuries during the 2009–2010 through 2013–2014 seasons resulted in fewer players missing ≥30 days (IPR = 0.3; 95% CI = 0.16, 0.57) and more players returning to play within 1 week (IPR = 1.31; 95% CI = 1.21, 1.41) than those during the 2004–2005 through 2008–2009 seasons. Injuries during the 2009–2010 through 2013–2014 seasons were also less likely to be recurrent (IPR = 0.55; 95% CI = 0.41, 0.74). During the 2004–2005 through 2008–2009 seasons, more days were missed because of injury (11.06 ± 22.67) compared with the 2009–2010 through 2013–2014 seasons (5.09 ± 9.45; P < .001). Decreases in the number of recurrent ankle injuries and the days lost per injury may reflect continuing advances in rehabilitation and injury prevention.^6,10,19,24  Athletes were able to return to soccer more quickly in the most recent time period, perhaps because ATs and team physicians prescribed early proprioceptive training and earlier use of physical therapy as opposed to immobilization, which was commonly prescribed in the past.^6,10,19,24  It is interesting that even though athletes were returning to play more quickly, they sustained fewer recurrent injuries. Advances in prevention such as taping, bracing, and injury-specific weight-training regimens may be responsible.^6,10,19,24 

Previous researchers²  showed similar decreases in injury severity and recurrence when comparing the 2004–2009 seasons with the 1990–1996 seasons. Also, fewer ankle injuries occurred during the 2004–2009 seasons than the 1990–1996 seasons.²  The rate of ankle injuries from 1990–1996 in NCAA soccer was 1.73 per 1000 AEs versus 1.44 per 1000 AEs during the 2004–2009 seasons.²  Although our data demonstrated a decreased injury rate of 1.35 per 1000 AEs during the 2009–2010 through 2013–2014 seasons, this value was not statistically different from the rate of 1.44 per 1000 AEs during the 2004–2009 seasons (RR = 0.94; 95% CI = 0.85, 1.04). However, as stated earlier, the rate of recurrence and time loss both decreased significantly.

In addition, the 2004–2005 through 2008–2009 seasons resulted in more preseason injuries than the 2009–2010 through 2013–2014 seasons (Table 4). One possibility for this decrease is that coaches, ATs, and team physicians were more aware of the increased preseason injury rates and attempted to decrease the intensity and full-contact periods during preseason practices. Another potential reason for this decrease is the use of off-season strength and conditioning programs to help the athletes begin preseason workouts in better condition.

Furthermore, decreases in preseason and practice injuries between the 2004–2005 through 2008–2009 seasons and the 2009–2010 through 2013–2014 seasons may also be due to changes in NCAA rules and guidelines, including a document released May 13, 2009, “Defining Countable Athletically Related Activities.”²⁵  This document was released between the 2 time periods we analyzed and specifically describes which team activities are considered organized team practices.²⁵  These rule changes may have reduced the number of hours of practice, possibly leading to a lower rate of fatigue-related injuries during the 2009–2010 through 2013–2014 seasons.

Another effort to protect athletes addressed by the NCAA was the update and implementation of rules regarding shin guards, which were first mandated for NCAA soccer players in 1991.³  Since then, the size and use of shin guards have undergone changes.³  The most current update³  regarding shin-guard rules (established in 2007) stated that they must “meet the standards established by the National Operating Committee on Standards for Athletic Equipment (NOCSAE).” Shin guards prevent contusions and fractures. Because of the small number of fractures among NCAA soccer players (n = 9), these injuries were not included in our study. However, shin guards have failed to decrease the number of contusions. In fact, the proportion of contusions increased slightly from the 2004–2005 through 2008–2009 seasons to the 2009–2010 through 20013–2014 seasons (IPR = 1.25; 95% CI = 0.94, 1.66), though the confidence interval suggests that the difference was not significant. This finding correlates with previous results that showed no changes in contusion rates with the implementation of shin guards.³ 

Previous investigators²⁶  determined that in the years 2005–2007, of the programs that provided data to the NCAA-ISP, 88.5% of NCAA soccer injuries were reported. Thus, the data we used in this study represent a majority but not all of the injuries that occurred during these seasons. Our data only accounted for injuries reported by ATs and medical staff. It is possible that injuries were reported incorrectly; for example, a new AT who had not previously worked with a particular athlete could have marked an injury as new when it was actually recurrent. The immobility and rehabilitation measures used for recovery in these athletes were not described, and time-out-of-play data did not control for these extenuating variables. Also, LLC tears, medial (deltoid) ligament tears, and high ankle sprains were not differentiated as partial or complete. Another inherent limitation of the data used in this study was that the diagnoses were made by many different ATs and physicians, and we do not know whether physical examinations or magnetic resonance imaging were undertaken.

Many more schools participated in the NCAA-ISP for the 2004–2005 through 2008–2009 seasons versus the 2009–2010 through 2013–2014 seasons, though generalizations are still possible due to the persistently high volume in the latter group. Also, only the number of AEs was measured and not participation time per event. Therefore, if a practice lasted twice as long as a game, it was measured as only 1 AE. Similarly, if an athlete played in only the last few minutes of a game, it was still counted as an AE. Had AEs been defined in minutes, the increased RR of game injuries would have likely been higher. Also, there was a discrepancy in the recording between the 2004–2009 seasons and the 2009–2014 seasons. The 2009–2014 seasons included injuries that resulted in no time loss. However, the 2004–2009 seasons did not include injuries that resulted in no time loss. Therefore, when comparing average days lost between these time periods, the analysis may be biased.

Although the rate of ankle injuries did not change between the 2004–2005 through 2008–2009 and 2009–2010 through 2013–2014 seasons, the prognoses improved. Of the 4 most common ankle injuries, high ankle sprains resulted in the worst prognosis. Overall, male and female NCAA soccer players injured their ankles at similar rates; however, males were more likely to have contact injuries.

This article contains materials created, compiled, or produced by the Datalys Center for Sports Injury Research and Prevention, Inc, on behalf of the NCAA. The NCAA-ISP data were provided by the Datalys Center, and the ISP was funded by the NCAA. The content of this manuscript is solely the responsibility of the authors and does not necessarily represent the official views of the Datalys Center or the NCAA. We thank the many ATs who volunteered their time and efforts to submit data to the NCAA-ISP. Their efforts are greatly appreciated and have had a tremendously positive effect on the safety of collegiate athletes.