This study received institutional review board approval and was registered at (NCT00855023).
Study Population
We recruited healthy volunteer athletes and excluded those who had any of the following criteria: (1) contraindications to magnetic resonance imaging (such as pregnancy or implanted hardware); (2) relevant medical problems (such as connective tissue problems, paralyzed hemidiaphragm, morbid obesity, or claustrophobia); (3) clinical signs of an impairment or abnormality in the knee (such as abnormal range of motion, muscle weakness, or malalignment); (4) injury to the knee that required medical attention; (5) previous surgery on the knee; or (6) current pain in the knee.
The twenty-five athletes who had none of the exclusion criteria and who provided signed consent formed the study group. The mean age of the twelve men and thirteen women was twenty-five years (range, eighteen to forty-five years). One limb (randomized with regard to right or left) of each participant was designated for positioning, imaging, and measuring.
Positions, Imaging, and Measurements
On the basis of a previous study relating landing position to injury of the anterior cruciate ligament11, three positions were selected for assessment: one associated with no injury (safe position), one associated with anterior cruciate ligament injury (provocative), and one position that exaggerated the anterior cruciate ligament injury position (exaggerated provocative) (Fig. 1). For the safe position, the subject was positioned with the hip flexed 25°, the knee flexed 21°, and the ankle at 23° of plantar flexion. In the provocative position, the subject was positioned with the hip flexed 42°, the knee flexed 17°, and the ankle at 7° of plantar flexion. In the exaggerated provocative position, the subject was positioned with the hip flexed 50°, the knee flexed 10°, and the ankle at 7° of plantar flexion. For all positions, hip abduction was set at shoulder width and tibial rotation was set at neutral (0°).
Joint angles were measured with a goniometer, and the limb was supported with a specialized posterior fiberglass splint compatible with the magnetic resonance imaging scanner and preset at the correct knee angle (Fig. 2). The foot was secured on a firm foam positioning device to obtain the correct ankle angle. Although athletes in the safe position landed with forefoot contact on the ground, this wedge was used under the foot to replicate the ankle angle because subjects were unable to hold the forefoot position during the magnetic resonance imaging scan. In the sagittal plane, the ankle joint was measured as the angle between the axis of the lower limb and the plantar surface of the shoe11,16. The knee flexion angle was measured as the angle between a line connecting the superior tip of the greater trochanter to the midpoint of the lateral aspect of the knee at the joint line and a line connecting the midpoint of the lateral aspect of the knee at the joint line to the anterior point of the distal tip of the fibula. The hip angle was measured as the angle between the line from the superior tip of the acromioclavicular joint to the superior tip of the greater trochanter and the line from the superior tip of the greater trochanter to the midpoint of the lateral aspect of the knee at the joint line17. Neutral tibial rotation was achieved by placing the subject's patella and foot perpendicular to the plane of the body and the end of the foot platform.
For all three limb positions, each subject was placed in a standing position with partial weight support, provided by a small seat integral with the magnetic resonance imaging spine board. In total, nine magnetic resonance imaging scans (three in each position, with multiple images per each scan) were acquired in an open 0.6-T standing magnetic resonance imaging scanner (Fonar, Melville, New York). The first scan was a low-resolution scout image. The second was a two-dimensional gradient-echo axial scan (with a resolution of 0.67 mm by 0.67 mm by 5.0 mm) that assessed the area from the beginning of the femoral condyles to the tibial tuberosity. The image at the superior aspect of the femoral epicondyles was used to define the location and orientation of the final two-dimensional sagittal-oblique gradient-echo recall scan (with a resolution of 0.75 mm by 0.75 mm by 4.5 mm). In this third scan, the scan plane was perpendicular to the line connecting the posterior aspect of the condyles, and the number of images was sufficient to capture the full width of the femur from medial to lateral. All patient and position identifiers were removed from the data so that one author (F.T.S.) was blinded to the subject, the limb position, and the order of limb position sequences. The subject order and the order of limb position sequences were randomized before analysis. This author quantified all measures on the magnetic resonance images. This same author repeated all measures five months after the initial measurements were made. The images were reordered by an outside investigator so that the author was blinded to the subject, limb position, the order of the limb position, and the original measurements. One subject moved during the image acquisition, causing image blurring. Because of this blurring, the identification of anatomical landmarks was imprecise. Thus, this one dataset was eliminated.
One angular and three distance measurements were made from the reference slice, defined as the first sagittal-oblique slice containing the medial edge of the fibular head18. All measurements were quantified with use of ImageJ (National Institutes of Health, Bethesda, Maryland)19,20. To acquire these measures, two lines and four points were identified (Figs. 3-A through 3-D): the femoral shaft line, the tibial plateau line, the point of contact, the femoral sulcus point, the most anterior point on the circular posterior portion of the condyle, and the posterior tibial point. The point of contact was determined by drawing a line (contact line) where the femur and tibia were in contact. The midpoint of this line was used as the point of contact. The most anterior point on the circular posterior portion of the condyle was determined by first creating a circle to fit the circular posterior portion of the lateral femoral condyle. The most anterior point on the circular posterior portion of the condyle was selected as the first point on the anterior aspect of the circle that was no longer contacting the joint surface. With use of these markers, the distances from the posterior tibial point to the point of contact, the point of contact to the femoral sulcus point, and the point of contact to the most anterior point on the circular posterior portion of the condyle were measured. The last measure, the point of contact to the most anterior point on the circular posterior portion of the condyle, was calculated between the two points in the femoral anterior direction only, defined as the direction perpendicular to the femoral shaft line in the reference image. The tibial plateau angle was defined as the angle between the femoral shaft line and the tibial plateau line.
Statistical Analyses
An a priori power analysis showed that twenty subjects were required to detect a 1-mm difference in distance, assuming a two-sided Student t test, a power of 0.80, a significance level of 0.05, and a common variance that was twice the image resolution (1.5 mm). A two-way repeated-measures analysis of variance was used to assess main effects and interaction effects for two repeated-measurement trials (test one and test two) in each of three positions (safe, provocative, and exaggerated provocative). Thus, the two repeated factors were test and position. Three distances and one angle were measured twice in each position such that four variables were assessed in separate two-way analyses of variance. Main and interaction effects were further examined with post hoc Tukey tests (p < 0.05). Intraclass correlation coefficients were used to examine intrarater reliability for each distance or angle measurement in each of the three positions.
Source of Funding
No external funding was received for the study.
Note: The authors thank Nancy Epstein, Abrahm J. Behnam, the volunteers who made this study possible, and Joseph S. Torg, MD, for his help in developing the concepts for this project. Washington Open MRI donated the use of the magnetic resonance imaging scanners.