I'd like to thank the Arthroscopy Association and AOSSM for the opportunity to give this talk today, as well as to thank my co-authors for making this study possible. I have no relevant financial disclosures. So the native NPFL has been shown to remain constant in length from zero to 60 degrees of knee flexion in biomechanical studies. We know that anisometry and inappropriate graft length have been associated with abnormal patellar kinematics and contact pressures, increased graft strain, and increased risk of failure. Surgical techniques have focused on minimizing length changes to recreate the isometric function of the NPFL. We know from cadaveric studies that morphologic abnormalities such as patella alta and lateralization of the tibial tuberosity can lead to anisometry of the graft. However, length changes in the intact NPFL in patients with morphologic risk factors have not been described. So our aim was to assess the NPFL length change patterns in asymptomatic knees of patients with contralateral patellar instability, and to assess the influence of morphologic risk factors on these changes. Patients with unilateral recurrent patellar instability with no history of symptoms or procedures on the contralateral knee were included in this study between the ages of 12 and 40. Dynamic CT imaging of the bilateral knees was performed based on an adaptation of protocol previously described by Cosgaria and Carino. Active knee extension was performed in a sitting position, performed over 10 seconds, and all scans were performed on the GE Revolution CT scanner in sitting mode. For 2D measurements, we measured trochlear dysplasia, as seen on this slide, using sulcus angle, lateral trochlear inclination, and trochlear depth. Patella alta was quantified using the Catan-Deschanes index with the ratio B over A, as shown in this picture. Contralateralization was quantified using TTTG distance. Volume-rendered 3D digital knee models were derived from each frame and grouped into 10-degree intervals of knee flexion. NPFL length was calculated at each flexion angle based on the distance between the known anatomic attachment points. For the patella, we used the radiographic midpoint of the anterior attachment on the medial border of the patella, which we know to be 19% of the articular length from the superior pole. On the femur, a .1 centimeter distal to the prominence of the adductor tubercle was used. The distance between the two points was calculated using straight-line representation of the fibers allowed to wrap around the femoral condyle to ensure accurate length measurements between the selected patellar and femoral points. Measurements were performed at each flexion angle, and all 3D measurements were performed using Rhinoceros 3D software. NPFL length was described at each knee flexion angle and assessed for significant differences. We used stepwise multiple linear regression analysis to identify the independent relationships between NPFL length changes and morphologic measurements. The knees were then categorized by the number of morphologic risk factors present using trochlear depth of 4 or less as an indication of trochlear dysplasia, Catan-Deschanes index of 1.2 or greater as indicating patella alta, and TTTG distance greater than 20 millimeters indicating malalignment. The cumulative number of risk factors was then used to categorize each knee from zero to three risk factors, and linear regression analysis was performed to assess the relationship between percent change in NPFL length and the number of morphologic risk factors present. Seventeen patients underwent dynamic CT scanning, and six were excluded for lack of adequate images or lack of full motion for the study. Eleven knees were included in this study with a mean age of 24.8. The NPFL length was found to be greatest at zero degrees with significant differences when compared to 20 to 50 degrees of knee flexion. The length varied from 74 millimeters to 64 millimeters on average with a 15.5 percent difference in length overall. Stepwise multiple regression analysis showed that there was an independent relationship between NPFL length change and TTTG distance. We also found that using risk factors to categorize the knees, knees with zero or one anatomic risk factors had a mean length change of less than five millimeters, whereas those with two or three risk factors had a mean length change of 13.9 millimeters, and this difference between the two groups was significant. We found a strong relationship between the number of risk factors and the percent change in length. In conclusion, NPFL length was significantly greater at zero degrees when compared to 20 to 50 degrees of knee flexion, and the number of morphologic risk factors present in the knee was associated with greater anisometry of the NPFL, and this was up to 22.6 millimeters greater in extension than 50 degrees in one case in an asymptomatic intact knee. The isometric function of the intact NPFL in these patients may not reflect previously described findings in anatomically normal patients, and further studies are needed to understand the path of anatomy related to these changes, as well as the implications for graph placement and assessment of isometry and NPFL reconstruction techniques. I'd like to acknowledge Dr. Andy Kosgaria and John Elias for introducing me to dynamic CT and starting me along this line of study. Thank you.

native patella femoral ligament

asymptomatic knees

contralateral patellar instability

morphologic risk factors

NPFL length changes