Q-angle, Pelvic width, and Intercondylar notch width as predictors of knee injuries in women soccer players in South Africa

Objective : To investigate the association between the three anatomical factors of Q-angle (QA), pelvic width (PW) and Intercondylar notch width (INW) and knee injuries among the U-23 female soccer players of South Africa Methods : The study is a case-control prospective study design. Twenty four U-23 women soccer players of the South African team were purposively chosen to participate in this study. Participants were divided into two groups: group1 (Case) was those with knee injuries, while those without injuries were in group-2 (Control). PW and INW were measured after X-rays of the hip were taken while the QA was measured manually with the goniomenter. Association between anatomical factors and knee injuries were tested with ANOVA. Results: Q-angle ranged from 14 0 to 18 0 for both injured and non injured groups. PW was between 24 -29 cm for both injured and non injured groups. INW was between 1.3mm and 2.8mm for the right and between 1.4mm and 2.5mm for the left notch for the injured group, while INW for the right and left of the non injured group were between 1.7mm to 2.1mm and 1.8mm to 2.1mm, respectively No significant association between knee injuries and each of the anatomical factors was found QA (p= 0.74), PW (p=0.34), INW (right and left respectively) (p=0.142 & p=0.089). Conclusion: The three anatomical factors of QA, PW and INW could not be used to predict knee injuries amongst the U-23 female players in South Africa. Pelvic width; Q-Angle; Intercondylar Notch Width; Female Soccer; Knee Injuries


Introduction
Physical activity is very important for all human beings and the benefits are well-documented for both genders 1,2 The level of physical activity will differ in frequency, intensity, and duration depending on gender, age, and type of physical activity. Sport is defined as physical activity involving a structured competitive situation governed by rules 3 .Soccer is one such example and is considered one of the most popular sport with women making up to 22% of world's soccer player 4 . According to Strong et al 1 numerous benefits of sport to women have been documented. Amongst these are lifestyle improvements, low teenage pregnancy rate, low drug and alcohol use and abuse, higher graduation rate, improved self-esteem and body image, increased bone mass, cardiovascular function and weight control 1 .
Over the last twenty years, women soccer has experienced tremendous improvement and growth with a series of events organized by FIFA and the Olympics. Examples of these are; the first women's football cup in China in 1991, Olympic Games in 1994, U-19/ U-23 World Championship in 2002 5. Most countries including South Africa have national female teams and a local league. There are currently more than 300 women's soccer clubs registered by the South African Football Association (SAFA) and a pool of about 50 000 players. In Africa the first women's soccer tournament was held in 1995 6. Generally, injuries abound in soccer and increasing incidences of injuries have been reported in female soccer. The most common injuries among female athletes are knee injuries, especially injury to the anterior cruciate ligament (ACL) 7 Adolescent females suffer a disproportionate number of knee and anterior cruciate ligament (ACL) injuries compared to adolescent males. McAlindon 8 found that female athletes have four to ten times more ACL injuries than male athletes. This was further reiterated by Ireland and Otto 9 who stated that female athletes have an increased rate of ACL injuries and patelofemoral disorders compared with their male counterparts. Knee injuries in soccer also occur without contact from another person and most often occur while the player is participating in the training programme. These training programmes include deceleration, twisting, jumping, and other sport specific maneuvers 10 . There have been many studies comparing injury rates between male and female players 8,11,12 . It was found that women soccer players sustain more knee injuries than their male counterparts 11 with ACL, the medial collateral ligament (MCL) and acute injuries to the menisci as the most common 13 .
'More information on ACL injuries compared to other knee injuries have been documented [13][14][15] . ACL provides 85% of the total restraining force to anterior translation of the tibia 14 .Injuries to the ACL usually occur during a sudden deceleration, as it typically is a noncontact injury 14 . An ACL tear is a common injury in all sports. Epidemiological studies of the ACL injury among female athletes revealed that female soccer players sustained 2.29 times more ACL injuries than male soccer players 16 .Giza et al 17 analyzed 173 injuries that occurred during the 2001-2002 seasons of the Women's United Soccer Association (WUSA) and found that the most common site for injuries was the knee with (31.8%) and ACL injuries accounted for 4.8%. The reasons for the higher rate of knee injuries in women are not clear. Some theories have implicated internal factors such as differences in the anatomical configuration between genders, knee ligament, ligament laxity and muscle strength 18 and extern al factors, such as conditioning, type of training and the development of muscle coordination 19 .
The higher QA in females is reported to influence knee injuries in female athletes 20 . Horton and Hall 21 , reports a mean QA of 15.8 + 4.5 degrees for women and 11.2 + 3.0 degrees for men. Increase in QA might contribute to an increase in the contact pressure applied to the patellofemoral joint 22 .A valgus knee and a pronated foot are often implicated in knee injuries 23 . A potential factor in knee injury is intercondylar notch configuration 14 .Studies have shown that players with smaller intercondylar notch dimensions are at greater risk for ACL injury because it accommodates the ACL 14, 24-25 . On the contrary a few studies found no relationship between the anatomical factors and predisposition to knee injuries 26,27 . There is however, limited empirical information on the association between anatomical factors and knee injuries amongst female soccer players in South Africa. This study will attempt to establish the association between knee injuries and each of the following anatomical factors: QA, PW and INW.

Methods
The design for this study is a case-control prospective study design. All the 24 U-23 national female soccer players of South Africa were participants in this study. They were divided into two groups, according to the presence or absence of knee injuries. Those with injury were in group-1 (Case) while those without injuries were in group-2 (Control).

Data collection sheet, X-ray, and Goniometer were used to collect data
The data collection sheet was divided into four sections. The first section was for demographic data (Name, age, weight, height and position on the field, current club name, seasons played for the current club, and other clubs played for (if any)). The second section was for history of injury (if any). The third section sought information on management of the injury. The last section required information on other injuries.

Procedure for data collection
The objectives of this study were presented to the administrative section of the U-23 women soccer project of SAFA in March 2008 prior to obtaining permission to conduct this study. Approval to conduct the study was granted after the presentation. Two week later, participants were taken to Dr George Mukhari (DGM) hospital where all radiographs and anatomical measurements were taken.

Measurement of intercondylar noch width (INW)
Plain X-rays of both lower limbs were taken using Shimadzu, RAD speed Safire machine. The first Xray used a conventional film for both pelvis and knees, from the anterior superior iliac spine to the patella of both lower limbs with participants in standing position. While the second one used digital film for the intercondylar notch with participants in 'tunnel view' position (which is semi prone with knee flexed >80 0 ). Machine measurements were (66-68KV, 25 MAS, 0.12 Sec) for both exposures. The intercondylar noch width was then measured as seen in figure 1 as reported by Shelbourne et al 28

Measurement of Q Angle
The QA measurements were taken manually with the universal goniometer as described by Emami et al 29 and Smith et al 30 . All measurements were taking with participants in standing with knees exposed in full extension (figure 2). The proximal arm of the goniometer was aligned to the anterior superior iliac spine, the axis at the midpoint of the patella, while the distal arm was aligned to the tibial tubercle. The intrarater reliability of the goniometer measurements was 0.90.

Measurement of pelvic width
Tape measure was used to measure the distance between the left and the right anterior superior iliac spine and the width recorded in cm.

Pilot study
The data collection sheet was piloted on four players of the non-professional Garankuwa women soccer club side to test for clarity, ambiguity and time to complete.

Ethical consideration
This study was approved by the Research, Ethics and Publications Committee of the Faculty of Medicine, University of Limpopo (Medunsa Campus), DGM Hospital and the administration section of the women soccer project of SAFA. Informed consent was obtained from participants using the University of Limpopo (Medunsa Campus) consent form prior to the X-ray procedures. Participants were also provided with the information sheet which included the aims and objectives of the study and were given the opportunity to ask questions. It was made clear that the participation in this study was voluntary and that they may withdraw from it at any time without giving reasons.

Interpretation of the X-rays
A radiologist at the department of radiology at DGM Hospital interpreted all the X-rays in this study.

Data analysis
The data emanating from this study was analyzed with a statistical package of social sciences version 17. Descriptive statistics of means and standard deviation (SD), percentages were used to describe the demographic information of the participants while tables were used to present data. Inferential statistics of ANOVA was used test the relationship between the variables evaluated. Each was categorized based on the concept or construct that they represented between the groups. The level of significance was set at 0.05.

Description of the participants
The participants were between 17and 22 years, with a mean age of 18.92 years ± 1.17 (SD), body weight of between 55kg and 63kg and a mean and SD of 58.5kg ± 2.28, and height of between of 157 and 172 cm and a mean and SD of 165cm +4.46 cm. Summary of measurements is presented in table 2.
They were no mean differences between the women who have sustained knee injuries (mean = 18.9 and SD = 1.19) and who have not sustained knee injuries (mean = 19.0 and SD = 1.49). Both have the same median of 19 and a significant value of p = 0.0023 as shown in table 1. Regarding playing position, a large number of respondents who were not injured were midfielder, 10 (71.4%), followed by defenders, 3 (21.1%). Large participants who sustained knee injury were forwards players 4(40%), 2 (20%) were goal keepers, defenders and midfielder. Please note that: Contact injuries are those sustained during tackles while non-contact are those sustained without contact/ tackle About 6 (60%) of the incidents of knee injuries were due to contact injuries and the rest of knee injuries were due to non contact. Four (40%) of the injured players continued playing despite injury. Four participants did not respond to this question as indicated in table 2.

Q-angle
The values of the Q-angles were scattered between 14 0 up to 18 0 for both injured and non injured groups,

Pelvic width
The different values of pelvic width of the players started from 24cm up to 29cm for both injured and non injured groups. The most standard value was shown in between 26 and 26.9 (table 3).  Group 0 = Players without knee injuries Group 1 = Players with knee injuries without contact Group 2 = Players with knee injuries with contact

Discussions
This study sought to establish the association between knee injuries and each of the three anatomical factors of QA, PW and INW. Q-angles ranged between14 0 to 18 0 for both injured and non injured groups. No significant association between Q-angle and knee injuries (p-value = 0.74) was found in this study. This finding agrees with those of Loudon et al 26 , who reported no clear link between knee injuries and Qangle. Until recently, it was believed that the joint hyper laxity predisposes individuals to musculoligamentous lesions, particularly in both ankle and knee, suggesting that increase of flexibility leads to increases laxity, which perhaps, increases the incident of knee injuries for both genders 31 . There were similarities in the three anatomical risk factors in both the injured and non injured groups. This findings contradicts those of Emami et al 29 and Tallay et al 32 found a significant association between higher Q-angle and the rate of knee injuries. However, higher QA alone might not be responsible for knee injuries as revealed by Emami et al 29 who found that 16% of the males and 20% of the females with abnormally high QA did not present with knee injuries. PW ranged between 24 -29 cm for both injured and non injured groups with no significant association between PW and knee injuries (P = 0.34). This finding is at variance with those of previous authors 16,19,32,33 who found a significant association between PW and the risk of knee injuries amongst women. Ireland et al 16 , and Hirst et al 34 , opined that the structural differences of the wider female pelvis increases the risk of knee injury by creating a greater coxa Vara/ geno valgum alignment with concurrent increase in tibio-femoral rotation force, thus imposing greater stress on the ACL.
The present study revealed scattered INW values of between 1.3mm to 2.8mm for the right and 1.4mm to 2.5mm for the left notch for the injured group, while the values for the right and left of the non injured group were between 1.7mm to 2.1mm and 1.8mm to 2.1mm, respectively. No significant association between knee injuries and INW (p-values of the right and left were p=0.14 and p=0.89, respectively) was found. Though, there was no attempt to compare INW with those of the male counterparts in this study, some other factors such as the differences in the ligamentous width geometry and materials of the ACL between genders might be linked to knee injuries 35,36 . It is therefore recommended that this should be further investigated.
A possible limitation in this study is the relatively small sample size and unequal sizes in all the groups. This might affect the generalization of the outcome to a larger population. Following a larger group for a longer period of time may produce a more generalisable outcome. Another possible factor that might affect the result of this study is participants' bias in recalling actual injury dates and time. This might have influenced the link between anatomical factors and the presence and /or absence of knee injuries.

Conclusion
Further studies should consider documenting only knee injuries identified by team doctors or physiotherapists during the season and not those that are obtained prospectively from the players. These knee injuries might also be identified pre-seasonally by performing standardized screening tests.