The Incidence of Anterior Cruciate Ligament Injury as a Function of Gender, Sport, and Injury-Reduction Programs

Published on 11/04/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 1366 times

Chapter 4 The Incidence of Anterior Cruciate Ligament Injury as a Function of Gender, Sport, and Injury-Reduction Programs

Chadwick C. Prodromos, Yung Han, Julie Rogowski, Brian T. Joyce, Kelvin Shi

Introduction

Although many studies have examined the incidence of anterior cruciate ligament (ACL) tears for given populations, an overall understanding of the real incidences is difficult to ascertain due to the breadth of the data and the disparate manner in which it is reported. The overall number of ACL tears appears to be increasing. This is caused in part by the increased participation of females in high-risk sports, as females clearly have an overall higher incidence of ACL tears than males. This realization has spawned the creation of training programs designed to decrease the incidence of ACL tears in females. The increase in ACL tears is also fueled by the increase in sports participation, from seasonal to yearlong, by athletes of both genders. This results in an increased number of exposures per year above that which was present for single-sport athletes in the past. Some sports appear to have higher risks of ACL tears than others, but these differences have not been well understood. Knowing the relative and absolute risks of ACL tear as a function of these parameters serves to focus attention on preventive strategies, where it is most needed. It also allows athletes and parents to understand the risks of participation in various sports.

Purpose

Our purpose was to acquire and review all of the relevant peer-reviewed published data on the incidence of ACL tears for the purpose of comparing the incidence of ACL tears in the following ways:

1 Among sports

2 Between females and males

3 Between those who have completed a program to decrease the incidence of ACL tears and those who have not

Methods

A computerized search of all papers in the peer-reviewed literature that had a possibility of dealing with the incidence of ACL tears was performed using a variety of indexing terms. Searches were then carried out by individual sports. This produced 793 articles that had some relation to knee or ACL injuries. These articles were reviewed, and bibliographies were cross-referenced for other papers, which were also reviewed for the purpose of identifying studies that had actual numerical incidences of complete ACL tears; this eliminated the overwhelming majority of papers. However, 33 papers were found that did have such quantitative data, and they form the basis for this chapter. Of these 33 papers, 25 had data that either used, or could be converted into, the preferred ACL injury incidence reporting method, namely: “ACL tears/1000 exposures.” An exposure is defined as either a practice or a game. These studies are listed in Table 4-1. They are divided by sport and then subdivided by level of competition, gender, and whether an ACL injury-reduction training program had been applied. The ratio of injury of females versus males is also listed for studies in which there were cohorts for both genders. These data form the basis for the analyses present in this chapter.

Table 4-1 ACL Tear Rate by Sport, Gender, and Injury Training*

Table 4-2 aggregates like subgroups from Table 4-1 and provides mean injury rates weighted according to the number of exposures. Table 4-3 lists the remaining studies, which use methods other than “tears/1000 exposures”¹^–⁸. Table 4-4 aggregates the like populations from studies that compared incidences by gender. Table 4-5 lists all the studies that involved training regimens designed to reduce ACL tear incidence.

Table 4-2 Weighted Means for Groups*

Table 4-3 Studies not Expressed in ACL Tears/1000 Exposures

Table 4-4 Ratios of Female to Male ACL Tear Rates*

Table 4-5 Effect of ACL Reduction Training Program on Tear Rate*

Exposures

In all of these studies it is important to remain cognizant of the number of exposures in the given study. The variance in the number of exposure between the studies is very large. The largest study has more than 8 million exposures and the smallest has only 600, a difference of more than 10,000 to 1 in the statistical power of the studies. These differences are so important that we have highlighted all the studies with exposures of more than 100,000 to make it easier for the reader to recognize those incidence studies of greatest statistical power.

Data Conversions

A number of studies report their incidence by dividing the number of ACL tears by hours of participation instead of practices. In these studies the hourly incidence was therefore used to calculate an incidence per 1000 exposures by converting hours to practices or games and adjusting the incidence accordingly. Doing so allowed these series to be used in the comparative analysis with the other studies, which used the tear per exposure methodology. Without this method, a large number of useful studies would have been lost from the analysis. If an exact practice length was not listed, we assumed a practice length of 2.25 hours. In the study by DeLee,²⁸ the data were presented in tears per hour, but exact data were given on number of practices and games and their lengths, so the data could be directly transformed into tears per 1000 exposures.

Individual Sports

Alpine Skiing

The alpine skiing data are notable for the large disparity in incidence between ski lodge employees, who are assumed to be expert skiers, and recreational skiers. The two studies of ski lodge employees by Oates and Viola show rates of 0.02 and 0.04, respectively.^9,¹⁰ The two general population studies by Warme and Deibert show tenfold higher rates of 0.63 and 0.40, respectively.^11,¹² Although the rate for the expert skiers is the lowest for any of the high-risk sports studied, the rate for the recreational skiers is overall one of the highest (P < 0.001). It is remarkable that this huge disparity is produced by two independent studies in each group, each with a very large number of exposures. These observations are thus of high reliability and statistical power.

The lower risk among the expert skiers is presumably a combination of increased skill and increased fitness in this group. The expert group is also remarkable for being one of only two cohorts for which the rate of injury is the same for males and females.

The lack of a gender difference in the large study by Viola ¹⁰ is also remarkable. Aside from lacrosse, alpine skiing is the only sport studied with a large enough number of exposures to generate reliable numbers to find this lack of a gender difference.

Soccer

The soccer data are dominated by the three extremely large studies of Mihata, Agel, and Arendt.¹³^–²¹ These data are remarkable for their amazing similarity. The female rates in the Mihata, Agel, and Arendt studies are 0.32, 0.33, and 0.31, respectively; the male rates are 0.12, 0.12, and 0.13, respectively. The female–male ratios are all also in the 2.5 to 1 range. The overall female to male difference was highly significant (P < 0.0001). Soccer is also notable in that in all three high school studies with ACL injury-reduction training program cohorts, the programs were apparently successful. (These studies were carried out only in females.) In other words, the trained athletes had significantly lower ACL injury rates (P = 0.0001) than the untrained athletes. These data are dominated by the landmark Mandelbaum et al study,¹⁶ which showed a 24% reduction in ACL tear incidence. It should be noted that the one adult study showed no reduction in ACL tear incidence with training.^21a

Basketball

As was the case with soccer, the basketball rates are dominated by the three large studies of Mihata, Agel, and Arendt.* Also similar to soccer, the basketball numbers are amazingly similar among the studies. The female rates are 0.28, 0.29, and 0.30. The male rates are 0.08, 0.08, and 0.07. The female to male ratios are 3.5, 3.6, and 4.2. The overall difference in rate between females and males was highly significant (P < 0.0001). Interestingly, and unlike the case with soccer, the two studies that included a cohort in which athletes had completed a program to diminish the ACL tear incidence showed no reduction in ACL tear rate as a result of training. In fact, the trained athletes in both studies showed increased ACL tear rates versus the untrained athletes, although this difference was not significant.

Indoor Soccer

The two studies of indoor soccer are included for the sake of completeness, but with only about 3600 total exposures, they have negligible statistical power by comparison to the more than 10 million soccer exposures by the three large studies cited earlier for outdoor soccer.^27,²⁸ Thus the very high female incidence of 5.2 tears per 1000 exposures, more than 10 times the outdoor rate, should be interpreted with caution, although the difference between females and males was statistically significant (P = 0.04). Nonetheless it is of interest that two separate studies arrived at almost identically high rates. If these increased rates were real, there would be two obvious potential causes: first, the fact that this study included only games, not practices, implying a higher risk with competition. Second, the fact that indoor soccer is played on artificial turf, whereas outdoor soccer is played on grass, may upwardly influence the injury rate.

Volleyball

The two volleyball studies have only 28,000 exposures, again too small to make reliable incidence conclusions.^19,²¹ However, it is remarkable that no ACL tears were recorded in either study. Basketball and soccer are often included with volleyball as high-risk sports for females. The rates of the six cohorts for basketball and soccer from each of the three large cited studies are all clustered between 0.28 and 0.33. If the average rate of 0.30 is applied to the 28,000 volleyball exposures, about 9 ACL tears would be expected. The fact that none were recorded may support a lower incidence of ACL tears in volleyball than in soccer and basketball.

Football

The one large football study, all in high school males, produced an injury rate of 0.11.²⁹ This is very similar to the male rate found in college soccer (0.12) and basketball (0.08). With more than 331,000 exposures, this study is of high statistical power.

Rugby

The 0.35 and 0.36 rates for women and the 0.18 rate for men from the two published studies are very similar to the 0.32 rate for women and 0.12 rate for men found in soccer.^17,³⁰ This is not unexpected given the similarities of the sports, which involve running and pivoting on a grass surface. It is of interest that the higher level of contact in rugby did not produce a higher ACL tear rate. The 89,000 total exposures are a significant number, although far less than the soccer exposure numbers. Further supporting the validity of the data, however, is the fact that two separate rugby studies produced almost identical ACL tear incidences.

Wrestling

With just under 12,000 exposures, the one wrestling study is of low power.¹⁷ It showed a rate of 0.77 in females and 0.19 in males.

Lacrosse

There is one published study with usable data for our analysis.¹⁵ The 0.17 rate of ACL tear for men is similar to that of soccer, rugby, and basketball. The female rate of 0.18, however, is substantially lower than the female rates for these three sports. With 1,783,903 exposures, this study is of high statistical power. Lacrosse stands as the only sport aside from alpine skiing for which the rates for males and females are roughly the same. There is no obvious explanation. The argument has been made that the carrying of the stick is ACL tear protective and may be at least part of the reason for the lower injury rate in females. However, if this were true there would be no obvious explanation for the fact that such an effect does not serve to lower the rate in males compared with other similar sports.

Handball

The two published female cohorts both have very high incidences of 0.56 and 0.86.^30,³¹ The two male cohorts of 0.24 and 0.11 are relatively unremarkable.^31,³² The gender difference is significant (P < 0.0001).

Australian Rules Football

The one large published study with usable data showed a quite high ACL tear rate of 0.82 per 1000 exposures (games only in this study).³³ This is quite high in relation to other sports for males. The 100,000 exposures represent a substantial number, although not as large as some studies. It is interesting to note that the indoor soccer study, which was also a games-only study, also had an unusually high tear rate of 0.97. This suggests the possibility that games have a higher risk of ACL tear than practices, although there are far too little data to make this conclusion. If this is not a contributing factor, other explanations would include an intrinsically high rate for Australian Rules football or that the rate is high due to chance in this modest-sized study.