Over 200,000 people suffer from anterior cruciate ligament (ACL) injury in the US every year [1], of which 38,000 are female [2]. The likelihood of injuring the ACL is greater in females than in males. In the early 1990s, a study showed that when males and females engaged in the same sports, a greater proportion of females suffered from non-contact ACL injury than males [3]. In the US, a statistical survey by the National Collegiate Athletic Association showed that if a female athlete engaged in sports to the same extent as a male athlete, the female athlete was 2-8 times more likely to injure her ACL than a male athlete [4]. A meta-analysis showed that this gender disparity in the likelihood of ACL injury existed in various sports. In soccer, females were as 2.67 times more likely to suffer from ACL injury than males, in basketball 3.5 times, in wrestling 4.05 times, and in alpine skiing 1.00 times more likely [5]. The participation of females in sports at US collegiate level had increased tenfold between 1971 (3.7%) and 1998 (33%) [6]. The combined percentage of females in sports at collegiate and at university level was 40% in 1998 [6]. This increase in females in sports over the years has inevitably led to the increase in sports-related injuries in females [3].

Although the absolute number of ACL injuries is lower in females than in males, the relative prevalence is significantly greater in females than in males, especially within a type of sports, such as basketball, soccer, and handball. These sports repeatedly require specific set of motions such as abrupt deceleration, cutting, and changing directions. Thus, special attention should be paid on female athletes participating in such sports to prevent ACL injuries [7]. Furthermore, since the prevalence of the preventable non-contact ACL injury is greater than the non-preventable contact ACL injury, taking prior actions to minimize ACL injuries is all the more important [8]. Various risk factors may increase the likelihood of female athletes injuring the ACL, such as genetic, environmental, anatomical, hormonal, neuromuscular, and biomechanical factors. The diversity of risk factors may mean that formulating a gold standard approach for prevention that is effective for everyone may be unfeasible.

In this study, we assessed the risk factors that contribute to an ACL injury in females (Table 1) and by doing so we aimed to begin developing a prevention program against ACL injuries. Where possible, the association of a certain risk factor and the outcome of ACL reconstruction were noted.

A few studies have implicated environmental factors in non-contact ACL injuries [9]. However, none of the environmental factors suggested so far seems to be a gender-specific risk factor. Examples of environmental factors that may contribute to ACL injuries include climate condition, ground surface, footwear, resistance of footwear on surfaces, and prophylactic knee braces.

Climate conditions

Unpredictability of climate conditions has meant research investigating this factor is limited compared to, for example, research on hormonal and neuromuscular factors. ACL injury has been shown to be more prevalent in Australian football players than players from other countries [11]. This is thought to be a result of frequent playing on dry and hard ground, brought about by the dry weather in Australia, and thus players are constantly exposed to high friction and torsional resistance between the shoe sole and ground surface. For every 1,000 Australian football and American soccer players, 0.82 Australian football players compared to just 0.12 American soccer players is likely to suffer from ACL injury [12]. Despite a clear influence of climate condition on the prevalence of ACL injuries in certain regions, climate condition has not been shown to influence gender differently.

Ground surface

Ground surface with increased friction limit and resist the movement of players, increasing the likelihood of ACL injuries. A dry ground surface causes friction between the shoe sole and ground surface than when the ground is damp, thus ACL injury occurs more easily on dry over damp surfaces [12]. Conversely, it is not yet agreed whether artificial surfaces are more or less harmful than natural surfaces. However a recent study of American soccer players suffering from ACL injuries at the collegiate level showed a 50% greater prevalence of non-contact ACL injuries in those who played on natural grass than in those who played on artificial turf [13]. Yet, another study has shown natural grass to be less harmful than artificial turf [14]. In another study, artificial indoor floors have been shown to be a less favorable ground surface than natural wood floors, although there was no gender disparity in terms of the occurrence of ACL injury. Whilst no difference in the prevalence of ACL injury between the two types of ground surfaces was seen in males, this study did find that ACL injury was more frequent in artificial indoor floors than natural wood floors in females [15].

Footwear

Smaller and fewer cleats on the sole of training shoes of soccer players has been associated with less occurrence of knee and ankle injuries in the US [16]. No studies have been carried out in females to see what type of shoes protect them from ACL injury but simply that females, like males, should avoid long and irregular cleat designs in footwear [17]. Long and irregular cleat designs increase the torsional resistance in both artificial and natural grass, which is why they may be associated ACL injury [17].

Prophylactic knee brace

Prophylactic knee braces do not affect the likelihood of an ACL injury in a normal athlete [18]. Although wearing a double-hinged, single, upright, off-the-shelf knee brace (Donjoy Inc., Carlsbad, CA, USA) decreases medial collateral ligament and ACL injury, the associated decrease in ACL injury was not statistically significant [19,20]. These studies did not address gender disparities.

Gender disparity in the difference in the prevalence of ACL injuries is thought to be partly down to differences in anatomical structures between males and females. Sexual dimorphism in anatomical structures includes bone length, Q-angle, intercondylar notch width and shape, ACL size and mechanical property, posterior tibial slope, body mass index (BMI), and generalized ligament laxity. However, it is difficult to find an exact characteristics or property of an anatomic structure that implicates ACL injury in females.

Bone length

In children, the knee torque increases as the tibial and femoral bone length increases, this increases instability of the knee [21]. In males, this instability can be in part stabilized through muscle strength and stiffness. In females, this partial stabilization is not achieved due to smaller muscle mass, and this is thought to increase the likelihood of ACL injury in females [21,22]. Hip width to femoral length ratio has been shown to be a better marker for ACL injury than the absolute length and width of the lower extremity limbs [7]. In other studies, however, this ratio was not significantly different between the sexes, being 0.73 in males and 0.77 in females, suggesting that this ratio cannot explain the gender disparity in ACL injuries [23].

Q-angle and a valgus knee

Q-angle is the angle formed when the line connecting the center of anterior superior iliac spine and the patella intersects with the line connecting the center of the patella and the tibial tuberosity. Males in general have larger Q-angles than females [7,23]. A larger Q-angle means that the lateral pulling force exerted by the quadriceps femoris muscle of the patella on the medial knee is larger. Abnormal stress on the medial knee has been shown to increase the likelihood of ACL injury and other injuries of the knee joint [24]. Further, the knee in valgus position has been used as a diagnostic predictor for ACL injury [9]. However, as the static Q-angle in the valgus or dynamic knee position cannot be used in the same way to predict ACL injury, there is still controversy as to whether the Q-angle and a valgus knee can be used as an indicator [25,26].

Intercondylar notch width and shape

In general, the taller you are the larger the total intercondylar notch is. Although the intercondylar notch width also increases as you become taller in men, this width does not necessarily increase in relation to your height in females [27]. Rather, females with a narrow intercondylar notch (< 13 mm) is 16.8 times more likely to injure their ACL than those with a broader intercondylar notch [28]. This association between a narrow intercondylar notch and ACL injury has been further confirmed by that fact that narrower the notch, the more severe the ACL injury was [29,30]. Females with unilateral ACL injuries were shown to have a narrower intercondylar notch width than those without ACL injuries, and those with bilateral ACL injuries were shown to have an even narrower intercondylar notch than those with unilateral ACL injuries [31]. However, there are cases when the difference between the width of intercondylar notch of ACL injured knee and the contralateral unaffected knee is insignificant. As the contralateral unaffected knee, despite having a notch width that predisposed the affected knee to ACL injury, is still free of ACL injury, controversies still exist whether intercondylar notch width can really be an indicator for ACL injury [32].

It has been reported that notch width index (NWI) can be a predictor of ACL injury by assessing the size of distal femoral bone at the popliteal groove level. However reports are contradictory as Souryal and Freeman [30] found that a significant difference in the size of the distal femoral bone was not seen between the ACL injured and the uninjured cohort. Whereas Griffin et al. [33] found that NWI was larger in males than females, a different study found no relationship between NWI and ACL injury and no difference between gender [29,32]. The shape of the intercondylar notch varies but differences in the shape is unlikely to contribute to ACL injury [34,35].

Anterior cruciate ligament size and mechanical property

ACL size is smaller in females than men even when weight is considered for [36]. In many studies, a smaller ACL was associated more with injury of the ACL [27,28]. Although it is true that a smaller ACL receives more external stress compared to a larger ACL, it is unsure whether such forces are enough to cause an injury [9]. Another possible explanation for the association of ACL injury is that as a small ACL also predisposes you to a small intercondylar notch, motions such as jumping, landing, and cutting that require extension of the knee may lead to greater collision between the ligaments [27,37].

The mechanical property of ACL is important in situations where high force is applied. Cadaveric studies have shown that female ACLs have lower mechanical property than male ACLs meaning that females are inherently more prone to injury than males [38].

Posterior tibial slope

When the posterior tibial slope increases, the tibial bone becomes more anteriorly placed in relation to the femoral bone during quadriceps femoris muscle contraction. This leads to increased loading on the ACL. The posterior tibial slope has been shown to be greater in females who suffered from ACL injuries than those who did not. Between individuals with ACL injuries, the angle was shown to be greater in males than in females [39], but similar studies have also found no difference between the genders [40]. As reports on posterior tibial slope increase, a standardized system is required to aid comparison of values of posterior tibial slope, as well as other factors such as meniscal slope angle, across papers.

Body mass index

In females, the prevalence of generalized ligament laxity has increased while the prevalence of non-contact ACL injury has not. The knee instability has been shown to increase 2.7 fold after ACL injury [28]. As well as knee instability, hamstring laxity has been shown to be slightly higher in athletes with injured ACL than those with normal ACL, though the same could not be said when a bilateral ACL injury occurred [42].

Estrogen is thought to be one of the causative factors of ACL injury in females [25,43]. Receptors for estrogen and progesterone is present in human ACL fibroblasts [44,45]. These cells produce collagen, which is critical for the load-bearing capacity of ACLs [37]. Estrogen inhibits the formation of collagen by ACL fibroblasts, reducing the load-bearing capacity of ACLs [44]. This therefore increases the probability of ACL injury [46]. Increasing the concentration of estrogen in vitro has shown an interactive, dose-dependent, time-dependent effect on ACL metabolism such as collagen formation [47,48].

As well as decreasing collagen formation, high estrogen concentration decreases the neuromuscular activity of the knee [44] and increases knee joint laxity [49-51]. However, whether the probability of ACL injury at different stages of the menstrual cycle changes when estrogen levels are also changed remains to be elucidated [49-54].

So far, the effect of oral contraception, which decreases estrogen levels, on knee function and ACL injury in female athletes is unknown [7,44]. Although it has been shown that athletes who take oral contraception are associated with less injury [4,52], further use of oral contraception by college athletes did not increase non-contact ACL injuries [55].

In contrast to females, when males enter puberty, the neuromuscular development matches the rapid speed at which the body develops [21]. However, in females, puberty comes earlier and neuromuscular growth cannot accommodate the growth spurt, and this vulnerable state may increase the likelihood of ACL injury in females [21].

The coactivation of the hamstring and quadriceps femoris muscle is required for the protection from dynamic valgus, extreme anterior drawer, and varus positions [44]. However, in females, dominant contraction of quadriceps femoris muscles during landing or cutting [9] lead to increased anterior movement of the femur relative to the tibia, which increases the propensity to ACL injury [44]. Female athletes show a low medial-to-lateral quadriceps recruitment and enhanced lateral hamstring firing [26]. These neuromuscular systems exert force on the lateral joint, and open up the medial joint leading to increased anterior tibial movement and thus ACL injury [56,57]. Furthermore, when females land after jumping, the quadriceps femoris muscle contraction is greater and gluteal muscle contraction is lower than compared to males [58]. This means that hip muscles and the lower extremity in general are used less, which could explain the cause of the valgus collapse [44]. Decreasing the hip muscle activity decreases in the activities of the quadriceips femoris and hamstring muscle, altogether causing a change in load-bearing capacity of the ACL. Such changes may increase the predisposition of ACL to injury [33]. Ultimately, insufficient hip joint activity may be the cause of valgus collapse at the transverse plane [44].

The mechanical receptors of ACL recognize the rotational momentum and elongation when hamstring muscles stretch in a flexed knee. This could be a marker for the anterior movement of the femur relative to the tibia [44]. In individuals without ACL injury, it has been shown that females have a reduced single-leg sway than males, but if ACL is damaged, this swaying is increased [59]. This phenomenon is thought to arise specifically in females as the proprioception system is impaired in females with ACL injury, and thus impairment of proprioception is also thought to increase the likelihood of ACL injury [59].

Fatigue may also alter the landing and cutting movements and thereby increase the likelihood of ACL injury. However no studies have actually shown whether fatigue comes faster in females than males or whether fatigue affects females disproportionately compared to males [60]. Fatigue of the lower extremity can increase the chances of anterior dislocation of the tibial bone by 32.5% [61]. Further, in a fatigue state, executing stop-jumps can decrease the flexion angle of the knee and increase the anterior dislocation of the proximal tibial bone, thereby exacerbating the valgus knee.

Non-contact ACL injury has been triggered commonly by pivoting and cutting (29%), landing with knees in slight flexion (28%), one-step stop landing with an overextension (26%) [62]. Therefore, lower extremity pose in females during these activities can be a useful indicator of whether the female concerned may suffer ACL injury. Females pose a more standing position than males during a cutting motion, which means the flexion of the knee and hip joints is reduced leading to a valgus knee and enhanced activity of quadriceps femoris muscles [63]. ACL injury may be prevented if females bend down more during these poses [63].

Foot pronation

Navicular drop has been associated with ACL injury, which is more prominent in males with ACL injury than in females with ACL injury [64-66]. In those with ACL injury, a navicular drop of 2.5 mm was seen in males and 2 mm in females [67]. Postpuberty, females become more flexible than males, and thus the generalized laxity of the body is greater [44]. The laxity of the foot is thus greater in females. Increased ligament laxity and navicular drop have both been implicated in ACL injury [44,66,68]. As the navicular drop increases the anterior translation of the tibial bone increases, this translation causes internal rotation of the tibia and places strain on the ACL [9,66]. Therefore, an excessive foot pronation can be a possible risk factor for ACL injury.

Change in the angle of ankle joint can influence the strength, momentum, muscle activity pattern of the knee [33,69]. Female athletes, compared to male athletes, have higher ankle eversion upon cutting exercises, leading to greater valgus stress and tibial rotation, accumulating to strain on ACL [70].

Knee

Internal rotation and external rotation momentum increases more in females [71] upon landing due to the extension of the knee and weak muscle strength [72]. On coronal place, this brings changes to the dynamic neuromuscular control of the lower extremity [70]. The momentum and angle of the external rotation of the knee have been shown to be important diagnostic factors for ACL injury (73% sensitivity and 78% specificity) [22]. Further, bending of the knee at 10o-20o seems to elicit severe anterior dislocation by the quadriceps femoris muscle leading to extreme tension on the ACL [33]. However, it is unclear whether female athletes have in general greater knee angle at landing and cutting than male athletes [44,73,74]. When only looking at the point of surface contact at landing, the knee flexion angle was found to decrease from the age of 12 in females, whereas this decrease was not seen in males [75].

Hip

At landing, the biomechanics of the hip predisposes females to ACL injury. At saggital plane, peak external hip flexion moments is seen in female athletes with ACL injury [22]. During landing, hip adduction can occur to greater extent in female than in males [76]. Gluteal muscle use in females decrease [58], asymmetrical neuromuscular activity and flexibility can induce a varus hip and a valgus knee, and thereby increase probability of ACL injury [44].

Core stability

The core refers to abdominal, back extensor, and pelvic floor muscle strengths and function, which together contribute to the lumbopelvic-hip stability. It is unknown whether ACL is more vulnerable to injury in conditions of greater core instability. In both sexes, it was found that core instability is not associated ACL injury, but weak hip abduction and external rotation leads to increased risk for lower extremity injury [7].

Skill and level of exposure

The level of exposure to sport activities is one way to judge how skilled an athlete is [77]. When looking at athletes of the same gender in a particular type of sports, even if athletes were differently skilled, the extent of ACL injury was similar among athletes. In male athletes of a soccer league, where many skilful players existed, the prevalence of ACL injury was seen to be high. Although the number of research is limited, a direct association between the level of experience or exposure to sports and ACL injury does not seem to exist in female athletes [78,79].

Implementing appropriate preventative strategies can help prevent injuries in athletes. Female athletes who had not participated in ACL prevention programs showed a 3.7-fold greater prevalence of ACL injury [71]. Further, female athletes who did not receive formal training were 4.6 fold more likely to injure the knee than when male athletes did not receive proper training. Strengthening of the knee can be achieved up to ten-fold by intensive training of the hamstring and gastocnemius muscle compared to when no training was done. Thus, specialized training programs tailored around the sexes may help reduce non-contact ACL injuries in females. Especially, by enhancing knee tension, improved balancing, minimizing problematic poses, and de-tensioning the ACL, it is important to help enhance neuromuscular control and ultimately help change the dynamic loading patterns at landing [9,44]. Further, plyometrics and agility exercises are preventative exercises that will increase muscle response time and hamstring activity, and improve hip muscle control [61,63,71]. Preventative protocols have been reported previously, such as the protocols provided by International Olympic Committee for neuromuscular and biomechanical training, plymetrics, agility, functional balance, and core stability [80,81]. In these protocols, emphasis is placed on knee-over-toe positions in exercises such as cutting and landing on both feet after jumping [81].

In females an intercondylar notch shape of A-type is the most predominant. U-type and W-type shapes are less common and smaller in size. Therefore if double-bundle ACL reconstruction is performed using U- or W-type notches, surgery may be difficult [82]. Whether females should undertake surgery of the intercondylar notch under these cases was not mentioned, but in case of a narrow intercondylar notch a standard notchplasty is recommended. Selection of the ligament for reconstruction is critical. In general a hamstring with a shorter diameter is used in females than males. However, it was found that height and BMI are not able to predict ligament diameter preoperatively [83]. Likewise, in Oriental Asians, the hamstring diameter is shorter in females, but in contrast to previous reports, height, wight, and BMI could be used to predict this [84]. If auto-bone-patellar tendon-bone (BPTB) is used instead of the hamstring for reconstruction in females, the rate of re-injury and laxity was lower, yet this did not lead to any significant improvements in the clinical outcomes [85-87].

It is known that the success rate for ACL reconstruction in females is not high [88]. Auto-BPTB reconstruction on ACL injured patients was followed up for 26 months. Complications and success-rates were compared between males and females but there was no difference. However, women required on average a 6 months longer period for rehabilitation than males [89]. In females, using hamstring for single-bundle reconstructions increased laxity [85,86], but in a double-bundle reconstruction, a prospective study did not show a difference in ligament laxity or clinical score between sexes [90]. Further, Laboute et al. [87] found re-injury was greater when a hamstring was used for reconstruction as opposed to using BPTB, although the increase was statistically insignificant. In females re-injury was seen for only those who had reconstruction using the hamstring. A Swedish study showed that after ACL reconstruction in females, one year, two year post-operative subjective scores were low, but a difference was no longer seen after two years [88].

The prevalence of ACL injury is greater in females than males. As the participation of women in sports is set to increase, the interest in ACL injury is also forecast to increase. In sum, according to literature, in females, Q-angle is larger, intercondylar notch width is narrower, ACL is smaller and mechanical property is lower. Posterior tibial slope is greater in females with ACL injury than those with intact ACL, and it is predicted that the change in estrogen levels contribute to ACL injury. Further, the dominant contraction of the quadriceps femoris muscle, valgus collapse of the hip and knee, weak hip muscles and vulnerability to fatigue, overall laxity of the ligament, quadriceps femoris muscle and hamstring torqueness, relatively weak core stability in females are all possible gender-biased risk factors for ACL injury.

Although external factors including footwear, ground surface, interaction between the surfaces of the ground and footwear, knee braces are all controllable parameters that can be prevented, these factors did not show any gender disparity. Further research should be done on factors that are considered to increase the risk of ACL injury to investigate preventative regimes. The possible approach of hormone suppression as one of the preventative approach should be also addressed experimentally. Lastly, to systematize the prevention program against ACL injury, trustworthy research highlighting the biomechanical basis of neuromuscular factors in ACL injury should be accumulated and assessed.