Clinical Biomechanics 20 (2005) 966972 www.elsevier.

com/locate/clinbiomech

Dierences in peak knee valgus angles between individuals with high and low Q-angles during a single limb squat
Kathleen J. Pantano
a b

a,b,*

, Scott C. White a, Louise A. Gilchrist a, John Leddy

Biomechanics Laboratory, Department of Physical Therapy, Exercise and Nutrition Sciences and Sports Medicine Institute, School of Medicine and Biomedical Sciences, University at Bualo, Bualo, NY 14214, USA Drexel University, Hahnemann Programs in Rehabilitation Sciences, 245 N. 15th Street, Mail Stop 502, Philadelphia, PA 19102, USA Received 7 July 2004; accepted 19 May 2005

Abstract Background. Dierences in anatomical alignment between genders have been suggested as causes of the disparity in anterior cruciate ligament injury rates. A larger Q-angle may be associated with increased knee valgus during movement resulting in anterior cruciate ligament strain. This study investigated whether healthy college-aged subjects with a large Q-angle display greater peak knee valgus during a single limb squat compared to those with a small Q-angle. The study also determined whether the high and low Q-angle groups displayed dierences in other select anatomical variables, and whether these anatomical variables were related to knee valgus. Methods. Twenty subjects, categorized as having a high Q-angle (P17) or a low Q-angle (68) were videotaped during the performance of a single leg squat. The peak valgus angles for the right knee were calculated. One-tailed independent measures t-tests were used to determine whether individuals with a large Q-angle exhibit (1) signicantly greater peak knee valgus during a single leg squat compared to those with a small Q-angle and, (2) greater pelvic width to femoral length ratios and greater static knee valgus than subjects with a small Q-angle. The Pearson productmoment correlation was used to establish the relationships between pelvic width to femoral length ratios and static knee valgus, pelvic width to femoral length ratios and dynamic knee valgus, and static knee valgus and dynamic knee valgus. Findings. Peak knee valgus during the single leg squat, and static knee valgus were not signicant greater in the high Q-angle group compared to the low Q-angle group (P = 0.09; P = 0.31). Subjects with a larger Q-angle, however, had a signicantly greater pelvic width to femoral length ratios (P = 0.015) compared to subjects with a small Q-angle. Pelvic width to femoral length ratios was related to both static and dynamic knee valgus (r = 0.47, P = 0.02; r = 0.48, P = 0.02), but static knee valgus was not related to dynamic knee valgus. Interpretation. The ndings suggest that pelvic width to femoral length ratios, rather than Q-angle, may be a better structural predictor of knee valgus during dynamic movement. 2005 Elsevier Ltd. All rights reserved.
Keywords: Anatomical alignment; Frontal plane kinematics; Anterior cruciate ligament injury; Closed-chain squat

1. Introduction Non-contact anterior cruciate ligament (ACL) injury rates in female athletes are reported to be three to eight times greater than that of male athletes (Arendt et al., 1999; Lindeneld et al., 1994; Moul, 1998). Skeletal alignment dierences between sexes, such as a greater

* Corresponding author. Address: Drexel University, Hahnemann Programs in Rehabilitation Sciences, 245 N. 15th Street, Mail Stop 502, Philadelphia, PA 19102, USA. E-mail address: kjp29@drexel.edu (K.J. Pantano).

0268-0033/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2005.05.008

K.J. Pantano et al. / Clinical Biomechanics 20 (2005) 966972

967

hip width to femoral length ratio (Horton and Hall, 1989), and a larger quadriceps angle (Q-angle) (Woodland and Francis, 1992; Horton and Hall, 1989) are thought to contribute to excessive amounts of knee valgus (lateral angulation or abduction of the tibia with respect to the femur). The combination of knee valgus (tibial abduction) and external rotation positions contribute to ACL impingement and injury (Fung and Zhang, 2003; Lephart et al., 2002). A Q-angle greater than 1520 is more prevalent in women (Hvid et al., 1981). This structural variation may be a factor in the disparity in ACL injury rates between male and female athletes (Huston et al., 2000; McLean et al., 1999; Moul, 1998). The Q-angle represents the direction of pull of the quadriceps muscles through the patella to its insertion on the tibial tuberosity (Huberti and Hayes, 1984). Measured statically, it is the angle formed by the intersection of lines drawn from the anterior-superior iliac spine (ASIS) to the mid-patella, and from the mid-patella to the tibial tuberosity. Therefore, it is strongly inuenced by the position of the patella and that of the tibial tubercle. It has been proposed that structural factors such as large Q-angles may be associated with greater genu valgus (Hutchinson and Ireland, 1995; Horton and Hall, 1989). Conventional wisdom has suggested that females have wider pelvises, which contribute to larger Q-angles (Zeller et al., 2003; Grin et al., 2000; Harmon and Ireland, 2000; Huston et al., 2000). A growing body of evidence, however, dismisses the fact that women have wider pelvises than men (Kernozek and Greer, 1993; Woodland and Francis, 1992; Horton and Hall, 1989). Rather, Horton and Hall (1989) suggest that females exhibit a greater hip width to femoral length ratio, which likely contributes to greater knee valgus. While Q-angle has been suggested to be a structural risk factor for ACL injury, it is primarily a measure of patellofemoral alignment. Q-angle may be increased simply by having a laterally displaced tibial tubercle. Therefore, a large Q-angle can be present without any genu valgus. To date, there have been no studies of the relationship between Q-angle and genu valgus. Therefore, the purpose of this study was threefold. The rst aim was directed at examining whether individuals with a large Q-angle have an increased angle of knee valgus during a single limb squat compared to subjects with a small Q-angle. It was hypothesized that there would be a signicantly greater amount of peak knee valgus for a group of subjects with large Q-angles compared to those with a small Q-angle, based on the theory that larger Q-angles would create greater angulation of the lower extremity resulting in increased valgus angles during the squat. A second aim was to determine whether subjects with a large Q-angle exhibit greater pelvic width to femoral length (PW/FL) and static knee valgus than those with a small Q-angle. It was hypothe-

sized that subjects with a large Q-angle would exhibit greater PW/FL and static knee valgus that those subjects with a small Q-angle. Lastly, the relationships between PW/FL and static knee valgus, PW/FL and dynamic knee valgus, and static knee valgus and dynamic knee valgus were examined. It was hypothesized that moderate to strong relationships (r > 0.50) (Portney and Watkins, 2000) would exist between PW/FL and static knee valgus, PW/FL and dynamic knee valgus, and static knee valgus and dynamic knee valgus during a single leg squat.

2. Methods 2.1. Subjects Twenty subjects (11 males, 9 females; mean age 22.9 (SD 2.7) years, range 1829 years; mean mass 71.6 (SD 15.6) kg; mean height 170 (SD 11.7) cm from the general college population were classied into high and low Q-angle groups, according to the degree of Q-angle that each subject exhibited (Table 1). Subjects having any right knee pathology or surgery within the last ve years and/or any prior right limb injury in the last year were excluded from the study. The number of subjects required for the study was determined from a power analysis conducted a priori, based on an expected large eect using a minimally clinical important dierence of 5 between subject groups with a projected variability of 3.3 and power greater than 0.80. All subjects gave written consent, approved by an Institutional Review Board, to participate in the study. The criteria used to place the subjects into high and low Q-angle groups were based on data taken from 204 male and female college students whose mean Qangle was 12.5 (SD 4.6). The mean Q-angle values established for our population sample were comparable to mean values reported by Horton and Hall (1989) (13.5 (SD 4.5)). Accordingly, the low Q-angle group was dened as being 8 or less, which represents the group mean minus one standard deviation (12.54.6). Subjects exhibiting a Q-angle of P17 (the mean plus one standard deviation) were placed in the high
Table 1 Description of subjects reported as means (SD) Groups (n = 20) Q-angle () Pelvic width/femoral length (cm) Static knee valgus angle () Age (years) Height (cm) Weight (kg)
a b

High Q-anglea 22 (3) 0.63 (0.08) 2.9 (2.7) 22 (3) 166.1 (12.0) 68.7 (20.2)

Low Q-angleb 7 (1) 0.56 (0.05) 2.9 (2.9) 24 (3) 173.9 (10.5) 74.5 (9.4)

High Q-angle is dened as P17 (7 females and 3 males). Low Q-angle is dened as 68 (8 males and 2 females).

968

K.J. Pantano et al. / Clinical Biomechanics 20 (2005) 966972

Q-angle group. Q-angle measurements were taken for each subjects right lower extremity using a standard goniometer by a single physical therapist (19 years experience) (ICC (1, 1) = 0.85; SEM = 1.5). In a relaxed standing position, the center of the patella and the tibial tuberosity were identied by palpation and marked with a pen on the subjects right limb. A string was then stretched from the right anterior-superior iliac spine to the mid-patella, and from the mid-patella to the tibial tuberosity. The Q-angle was measured by the goniometer at the intersection of the two lines. To ensure that subjects were placed in the appropriate category, a second tester (blinded to the rst examiners records) measured the subjects Q-angle in the same fashion. The second tester was a physical therapist with 12 years of clinical experience. Subjects were included in the study if there was 100% agreement between the testers that the subject belonged to one of two Qangle groups (ICC (3, 1) = 0.97; SEM = 1.8) and the measured Q-angles did not dier more than 5 between the two testers. 2.2. Motion analysis Subjects were videotaped as they performed a series of single leg squats on a set of adjustable stairs. A video camera was positioned in front of the subject, perpendicular to the plane of motion (frontal plane) to collect two-dimensional (2-D) video data. The subjects right foot maintained contact with an upper step, while they stepped down onto a lower step with their left foot. To maintain consistency in the position of the weightbearing extremity (right foot) during the squat and to control for the degree of lower limb (knee/ankle) rotation as a result of foot position, each subject was asked to place their foot on an imprint on the step such that the foot was pointing straight ahead. This was done to avoid excessive rotation of the foot in either a medial (internal) or lateral (external) position. Only heel contact with the lower step was allowed so that the subject did not push-o with the left limb upon return from the squat. The right knee exion angle was limited to a maximum of 45 of knee exion by varying the step height. The rationale in limiting the angle of knee exion to 45 was based on evidence that the greatest amount of tension on the ACL has been found to occur with the knee positioned between 0 and 45 of knee exion (Blankevoort and Huiskes, 1996; Wilk et al., 1996; Markolf et al., 1995; Renstrom et al., 1986; Arms et al., 1984) with 22 being the average angle of knee exion at the time of injury (Boden et al., 2000; Colby et al., 2000). To determine knee valgus angles for each subject, the ankle, knee and hip joint centers were calculated from the spatial position of reective markers xed to the skin

over anatomical landmarks. Markers were placed on the right and left anterior-superior iliac spines, the proximal and distal anterior thigh, the medial and lateral femoral epicondyles, and the medial and lateral malleoli of the right lower extremity. The proximal and distal anterior thigh markers were placed at a distance from the greater trochanter that represented 30% and 80%, respectively, of the total length of the thigh. The 2-D spatial coordinates of the reective markers were determined by digitizing the 60 Hz video records (Peak Performance Technologies, Inc, Centennial, CO, USA) of the motion. The coordinates were smoothed using a fourth-order, zero-lag Butterworth lter with lter cut-o frequencies (range of 37 Hz) determined by the Jackson Knee Method (Jackson, 1979). The ankle and knee joint centers were determined by calculating the midpoint distance between the lateral and medial malleoli, and the lateral and medial femoral epicondyles, respectively. The hip joint center was dened as the instantaneous center of rotation calculated from a series of abductionadduction movements of the right limb (frontal plane motion) in a standing position. Subjects performed 6 trials of right hip abduction, within a 3040 range, at a constant rhythm set by a metronome. The six abductionadduction trials were used to calculate the average instantaneous center of rotation (ICR) coordinates for the thigh at the point of its maximum angular velocity (Winter, 1990). The ICR coordinates representing the hip joint center were expressed as a xed vector relative to the right ASIS. The position of the hip joint center point was then calculated on a frame-by-frame basis for the step-down motion based on frontal plane translation and rotation of the pelvis. Once the hip, knee, and ankle joint centers for each subject were calculated, their coordinates were used to compute knee valgus angles for the right leg during the static and single leg squat trials. The peak knee valgus angles for each subject during three single leg squat trials were calculated and averaged (Fig. 1).

Knee Valgus Angle (deg)

30 25 20 15 10 5 0 0.0 0.5 1.0 Time (sec) 1.5 2.0

Fig. 1. Illustration of knee valgus during a trial of single leg squats for one individual. The peak knee valgus angles for each subject during three single leg squat trials were calculated for subsequent data analysis.

K.J. Pantano et al. / Clinical Biomechanics 20 (2005) 966972

969

Static knee valgus (deg)

From the video data, pelvic width was determined by measuring the distance between the right and left ASIS frontal plane coordinates; femoral length and static knee valgus were calculated from the frontal plane coordinates of the joint centers. These measurements were taken with the subject in double-leg stance. Pelvic width was divided by femoral length resulting in a pelvic width to femoral length ratio (PW/FL). 2.3. Data analysis A 1-tailed independent measures t-test was used to determine whether individuals with a large Q-angle would exhibit signicantly greater peak knee valgus during a single leg squat compared to those with a small Q-angle (P < 0.05), (n = 20). One-tailed t-tests were also used to examine whether subjects with a large Q-angle display greater PW/FL and greater static knee valgus than subjects with a small Q-angle (P < 0.05), (n = 20). The Pearson productmoment correlation was used to establish the relationships between PW/FL and static knee valgus (n = 20), PW/FL and dynamic knee valgus (n = 20), and static knee valgus and dynamic knee valgus (n = 20).

7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 High QA Subject groups
Fig. 3. Static knee valgus for the high Q-angle group (n = 10) compared to the low Q-angle group (n = 10). The values shown are the group mean and the bars represent the standard deviation from the mean.

2.94

2.89

Low QA

Pelvic width to femoral length (PW/FL) (cm)

3. Results There was no signicant dierence (P = 0.09) in the peak knee valgus angle between the high Q-angle group (mean 12.7 (SD 5.3)) and low Q-angle group (mean 9.8 (SD 3.7)) during the single-leg squat trials (Fig. 2). There was no signicant dierence (P = 0.31) in static knee valgus between the high and low Q-angle groups (Fig. 3). However, PW/FL was signicantly greater (P = 0.015) in the high Q-angle group, compared to the low Q-angle group (Fig. 4). PW/FL was related to static knee valgus (r = 0.47, P = 0.02), and dynamic
Peak mean valgus angles (deg) 20.0

0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00

0.63

0.56

High QA Subject groups

Low QA

Fig. 4. Pelvic width to femoral length ratios (PW/FL) for the high Qangle group (n = 10) and the low Q-angle group (n = 10). The values shown are the group mean and the bars represent the standard deviation from the mean. *P = 0.015.

knee valgus (r = 0.48, P = 0.02) (Fig. 5). In contrast, static knee valgus was not related to dynamic knee valgus (r = 0.18, P = 0.22).

15.0

Pelvic width to femoral length (PW/FL) (cm)

0.80 0.75 0.70 0.65 0.60 0.55 0.50 0.45 0.40 0 5 10 15 20 25 Dynamic knee valgus P= 0.016 r= 0.48

10.0 12.7 9.8

5.0

0.0 High QA Subject groups Low QA

Fig. 2. Peak knee valgus angle during a single leg squat for the high Q-angle group (n = 10) and the low Q-angle group (n = 10). The values shown are the group mean and the bars represent the standard deviation from the mean.

Fig. 5. The Pearson productmoment correlation regression line describing the strength of the relationship between PW/FL and dynamic knee valgus during the single leg squat (n = 20).

970

K.J. Pantano et al. / Clinical Biomechanics 20 (2005) 966972

4. Discussion The primary goal of this study was to determine whether subjects with a large Q-angle exhibit greater peak knee valgus than subjects with a small Q-angle while performing a single leg squat. While the peak knee valgus angle was approximately 3 higher in the high Q-angle group, it was not statistically signicant. Therefore, the idea that high Q-angles are related to higher peak valgus angles may not be valid. Since peak valgus was not signicantly higher in the high Q angle group, Q-angle may not be an important structural risk factor for ACL injuries as has previously been suggested. Our ndings are supported by a retrospective study by Huegel et al. (1999) showing no signicant dierences in Q-angle between ACL injured and uninjured female athletes. In a related study, Zeller et al. (2003) compared genders during the performance of a single leg squat, using three-dimensional kinematics. The study found that women displayed greater knee valgus at the initiation of squat and remained in more valgus alignment throughout the squat than men. If increased valgus is indeed associated with increased ACL (Fung and Zhang, 2003), then knee valgus, rather than Q-angle, may be a more relevant risk factor for ACL injuries. A secondary aim of our study was to determine whether individuals with a large Q-angle display greater a PW/FL ratio and static knee valgus than individuals with a small Q-angle. Interestingly, we found that subjects with a large Q-angle did exhibit a greater PW/FL ratio, but not a greater static knee valgus angle. This nding may be explained by the fact that the anatomical reference used to measure static knee valgus was dierent than that used to measure Q-angle and PW/FL. Both the Q-angle and PW/FL provide a measure of the frontal plane angulation of the femur referenced to the ASIS, whereas the hip joint center (not the ASIS) was used as the proximal reference for static knee valgus. While this study was not designed to answer this question, it would be interesting to know whether subjects, prescreened to have high PW/FL ratios exhibited signicantly higher dynamic knee valgus compared to those with low PW/FL ratios. In assessing correlations between structural and dynamic factors, a moderate relationship was found between PW/FL and both static knee valgus, and dynamic knee valgus. Since subjects with high Q-angles were not found to have signicantly higher valgus, this suggests that PW/FL might be a better predictor of static and dynamic knee valgus than Q-angle during the performance of a single leg squat. The outcomes of our study emphasize the need to focus more on the relationship between PW/FL and knee valgus during a single leg squat (rather than Q-angle and knee valgus) when considering potential risk factors for ACL injury.

The weak correlation between static knee valgus and peak dynamic valgus during the single leg squat was surprising. It indicates that knee valgus measured statically is not related to the amount of knee valgus displayed during a single leg squat and should not be used to predict knee valgus during this motion. This suggests that dynamic knee valgus may be more related to neuromuscular control as opposed to static structure. Neuromuscular control dierences have been reported as contributing to gender dierences in the ACL injury rate (Huston and Wojtys, 1996; Hewett et al., 1996; Zeller et al., 2003). Although our subjects were grouped according to Q-angle and not gender, we discovered that the majority of subjects in the high Q-angle group were women and the majority of subjects in the low Q-angle group were men. The number of subjects in each group was too small, however, to eliminate gender as a confounding factor. Therefore it is possible that gender, due to dierences in neuromuscular control strategies, may have had a confounding eect on our results. Our ndings that static knee valgus and peak dynamic valgus during the single leg squat were not related were in contrast to those of Zeller et al. (2003). These investigators reported that greater knee valgus in stance found in the female group was related to greater peak valgus during a single leg squat. Their study however utilized a 3-D motion analysis system compared to the 2-D system used in our study, which may have accounted for some of the dierence between study results. A 2-D model was used to measure knee valgus in our study due to limitations in laboratory equipment available. Although we tried to control for limb rotation during the step-down motion, we cannot ignore that fact that rotation of the limb ultimately contributes to the resulting knee valgus angle and must be taken into consideration. The 2-D representation of the knee valgus angle does not account for the degree of rotation contributing to knee valgus as it does in the 3-D analysis. This fact is recognized as a limitation of our study. Recent evidence has been shown that a combination of transverse and frontal plane positioning increases ACL strain and may place someone at higher risk of ACL injury. Potential injury to the ACL may not only occur when the tibia is internally rotated (direct loading) but can also occur during combined loading of the tibia in external rotation and abduction. This occurs by means of impingement against the intercondylar notch (Fung and Zhang, 2003). These ndings reinforce the fact that combined transverse and frontal plane motion needs to be considered when attempting to understand the risk factors associated with ACL injury. The lack of signicant group dierences observed in our study suggests that large Q-angles do not aect the degree of knee valgus during a single leg squat. It

K.J. Pantano et al. / Clinical Biomechanics 20 (2005) 966972

971

should be noted, however, that the possibility of a type II statistical error exists. While an a priori power analysis suggested that the sample size was adequate, a post hoc analysis revealed a moderate eect size of 64 with 42% power. In order to have achieved 80% power using the observed values a larger sample size was required. Although it is unclear whether a 3 dierence in knee valgus is clinically signicant it should be emphasized that the post hoc analysis indicated the results should be viewed with caution when interpreting our studys ndings and planning future studies. In addition, although we attempted to present a logical explanation for the results observed for the correlations between PW/FL, static knee valgus and dynamic knee valgus the reader should be cautioned against applying these ndings to the general population. One of the limitations of analyzing the relationship between these anatomical variables in our study is that our results can only be generalized to those individuals that exhibit a Q-angle of 68 and P17. These were our criteria for inclusion of subjects in the high and low Qangle groups but this group may not be representative of the general population. Future studies should investigate how PW/FL inuences dynamic knee valgus using a 3-D kinematic motion analysis to assess combined transverse and frontal plane motions and to allow comparisons with similar studies. 5. Conclusions In conclusion, the results of this study suggest that peak knee valgus during a single leg squat, and static knee valgus, were not signicantly greater in subjects with high Q-angles compared to those with low Q-angles. Subjects with a large Q-angle, however, had a signicantly greater PW/FL compared to subjects with a small Q-angle. In addition, for subjects classied as having a Q-angle of 68 and P17, we found that PW/FL was moderately related to both static and dynamic knee valgus, but static knee valgus was not related to peak knee valgus exhibited during a single leg squat. The role that PW/FL plays in predicting dynamic knee valgus and how it is related to ACL injury needs further exploration. Acknowledgments This work was supported in part by the New York State Chapter of the American Physical Therapy Associations Research Designated Fund. Special thanks to Michael Kempke, MS, PT at the University of Bualo Sports Medicine Institute, for his assistance in the collection of knee measurements used in this study.

References
Arendt, E.A., Agel, J., Dick, R., 1999. Anterior cruciate ligament injury patterns among collegiate men and women. J. Athl. Training 34, 8692. Arms, S.W., Pope, M.H., Johnson, R.J., Fischer, R.A., Arvidsson, I., Eriksson, E., 1984. The biomechanics of anterior cruciate ligament rehabilitation and reconstruction. Am. J. Sports Med. 12, 818. Blankevoort, L., Huiskes, R., 1996. A mechanism for rotation restraints in the knee joint. J. Orthop. Res. 14, 676679. Boden, B.P., Dean, G.S., Feagin, J.A., Garrett Jr., W.E., 2000. Mechanisms of anterior cruciate ligament injury. Orthopade 23, 573578. Colby, S., Francisco, A., Yu, B., Kirkendall, D., Finch, M., Garrett Jr., W., 2000. Electromyographic and kinematic analysis of cutting maneuvers. Implications for anterior cruciate ligament injury. Am. J. Sports Med. 28, 234240. Fung, D.T., Zhang, L., 2003. Modeling of ACL impingement against the intercondylar notch. Clin. Biomech. 18, 933941. Grin, L.Y., Agel, J., Albohm, M.J., Arendt, E.A., Dick, R.W., Garrett, W.E., et al., 2000. Noncontact anterior cruciate ligament injuries: risk factors and prevention strategies. J. Am. Acad. Orthop. Surg. 8, 141150. Harmon, K.G., Ireland, M.L., 2000. Gender dierences in noncontact anterior cruciate ligament injuries. Clin. Sports Med. 19, 287302. Hewett, T.E., Stroupe, A.L., Nance, T.A., Noyes, F.R., 1996. Plyometric training in female athletes: decreased impact forces and increased hamstring torques. Am. J. Sports Med. 24, 765773. Horton, M.G., Hall, T.L., 1989. Quadriceps femoris muscle angle: normative values and relationships with gender and selected skeletal measures. Phys. Ther. 69, 897901. Huberti, H.H., Hayes, W.C., 1984. Patellofemoral contact pressures: the inuence of Q-angle and tendofemoral contact. J. Bone Joint Surg. 66A, 715724. Huegel, M., Meister, K., Rolle, G., Indelicato, P.A., Hartzel, J., 1999. The inuence of rotational alignment of the lower extremity on noncontact injuries to the ACL in the female athlete. Orthopedic Transactions. J. Bone Joint Surg. 22, 429. Huston, L.J., Wojtys, E.M., 1996. Neuromuscular performance characteristics in elite female athletes. Am. J. Sports Med. 24, 427436. Huston, L.J., Greeneld, M.L., Wojtys, E.M., 2000. Anterior cruciate ligament injuries in the female athlete. Potential risk factors. Clin. Orthop. Relat. Res. 372, 5063. Hutchinson, M.R., Ireland, M.L., 1995. Knee injuries in female athletes. Sports Med. 19, 289302. Hvid, I., Anderson, I.B., Schmidt, H., 1981. Chondromalacia patellae: the relation of abnormal joint mechanics. Acta Orthop. Scand. 52, 661666. Jackson, K.M., 1979. Fitting of mathematical functions to biomechanical data. IEEE Trans. Biomed. Eng., 122124. Kernozek, T.W., Greer, N.L., 1993. Quadriceps angle and rear foot motion: relationships in walking. Arch. Phys. Med. Rehab. 74, 407410. Lephart, S.M., Abt, J.P., Ferris, C.M., 2002. Neuromuscular contributions to anterior cruciate ligament injuries in females. Curr. Opin. Rheumatol. 14, 168173. Lindeneld, T.N., Schmitt, D.J., Hendry, M.P., Mangine, R.E., Noyes, F.R., 1994. Incidence of injury in indoor soccer. Am. J. Sports Med. 22, 364371. Markolf, K.L., Burcheld, D.M., Shapiro, M.M., Shepard, M.F., Finerman, G.A., Slauterback, J.L., 1995. Combined knee loading states that generate high ACL forces. J. Orthop. Res. 13, 930935. McLean, S.C., Neal, R.J., Myers, P.T., Walters, M.R., 1999. Knee joint kinematics during the side-step cutting maneuver: potential for injury in women. Med. Sci. Sports Exer. 31, 959968.

972

K.J. Pantano et al. / Clinical Biomechanics 20 (2005) 966972 and electromyographic activity during open and closed kinetic chain exercises. Am. J. Sports Med. 22, 518527. Winter, D.A., 1990. Biomechanics and Motor Control of Human Movement. John Wiley and Sons, New York, NY. Woodland, L., Francis, R., 1992. Parameters and comparisons of the quadriceps angle of college-aged men and women in the supine and standing positions. Am. J. Sports Med. 20, 208211. Zeller, B.L., McCrory, J.L., Kibler, W.B., Uhl, T.L., 2003. Dierences in kinematics and electromyographic activity between men and women during the single-legged squat. Am. J. Sports Med. 31, 449 457.

Moul, J.L., 1998. Dierences in selected predictors of anterior cruciate ligament tears between male and female NCAA Division I collegiate basketball players. J. Athl. Training 33, 118121. Portney, L.G., Watkins, M.P., 2000. Foundations of Clinical Research: Applications to Practice, third ed. Prentice Hall Inc., Upper Saddle River, NJ. Renstrom, P., Arms, S.W., Stanwyck, T.S., Johnson, R.J., Pope, M.H., 1986. Strain within the anterior cruciate ligament during hamstring and quadriceps activity. Am. J. Sports Med. 14, 8387. Wilk, K.E., Escamilla, R.F., Fleisig, G.S., Barrentine, S.W., Andrews, J.R., Boyd, M.L., 1996. A comparison of tibiofemoral joint forces