PurposeThe purpose of this study was to compare the biomechanical characteristics and patient outcomes after either isolated intraarticular ACL reconstruction or intraarticular reconstruction with lateral extra-articular tenodesis. In addition, we aimed to evaluate biomechanical parameters of the entire uninjured, contralateral knee as a baseline during the analysis.MethodsEighteen patients were evaluated at an average of 9.3 years after ACL reconstruction. Twelve patients had an intraarticular reconstruction (BTB), and six had an additional lateral extraarticular procedure (BTB/EAR). Patients were selected for the additional procedure by the operating surgeon based on clinical and radiological criteria. At the time of review, each patient was assessed using subjective patient questionnaires, manual laxity testing, and instrumented laxity testing. Each knee was also evaluated using a robotic lower leg axial rotation testing system. This system measured maximum internal and external rotations at 5.65 Nm of applied torque and generated load deformation curves and compliance data. Pointwise statistical comparisons within each group and between groups were performed using the appropriate paired or unpaired t test. Features were extracted from each load deformation curve for comparative analysis.ResultsThere were no significant differences between the two groups with respect to the patient satisfaction scores or to laxity testing (manual or instrumented). Robotic testing results for within-group comparisons demonstrated a significant reduction in maximum external rotation (8.77°) in the reconstructed leg when compared to the healthy leg (p < 0.05) in the BTB/EAR group, with a non-significant change in internal rotation. The slope of the curve at maximum internal rotation was also significantly greater in the reconstructed legs for the BTB/EAR group (p < 0.05), indicating reduced endpoint compliance or a harder endpoint. Finally, the leg that received the extra-articular tenodesis had a trend towards a reduced total leg axial rotation. Conversely, patients in the BTB group demonstrated no significant differences between their legs. For between-group comparisons, there was a significant increase in maximum internal rotation in the healthy legs in the BTB/EAR group compared with the healthy legs in the BTB group (p < 0.05). If the injured/reconstructed legs were compared, the significant difference at maximum internal rotation disappeared (p < 0.10). Similarly, the healthy legs in patients in the BTB/EAR group had a significantly more compliant or softer endpoint in internal rotation, greater maximum internal rotation, and more internal rotation at torque 0 in their healthy legs compared with the healthy legs in the BTB group (p < 0.05). These same differences were not noted in the reconstructed knees. The only identifiable significant difference between the injured/reconstructed legs was rotation at 0 torque (p < 0.05).ConclusionsIn this group of patients who were at an average of 9 years from surgery, the addition of a...

IV.

PurposeThe purpose of this study was to determine the test–retest reliability and the repeatability over multiple days of a robotic testing device when used to measure laxity of the lower leg during a simulated dial test.MethodsTen healthy subjects were evaluated using an instrumented robotic lower leg testing system over 4 days. Three testing cycles were performed each day. Each leg was rotated into external and then internal rotation by servomotors until a torque threshold of 5.65 N m was reached. Load–deformation curves were generated from torque and rotation data. Both average-measure and single-measure intraclass correlation coefficients (ICC) were compared across the curves. ICC scores were also compared for features of the curves including: maximum external rotation at −5.65 N m of torque, maximum internal rotation at 5.65 N m of torque, rotation at torque 0, compliance (slope of load–deformation curve) at torque 0, endpoint compliance in external rotation, endpoint compliance in internal rotation, and play at torque 0. Play at torque 0 was defined as the width of the hysteresis curve at torque 0.ResultsAverage-measure ICC scores and test–retest scores were >0.95 along the entire load–deformation curve except around zero torque. ICC scores at maximum internal and external rotation ranged from 0.87 to 0.99 across the left and right knees. ICC scores for the other features of the curves ranged from 0.61 to 0.98. The standard error of the mean ranged from 0.0497 to 1.1712.ConclusionsThe robotic testing device in this study proved to be reliable for testing a subject multiple times both within the same day and over multiple days. These findings suggest that the device can provide a level of reliability in rotational testing that allows for clinical use of test results. Objective laxity data can improve consistency and accuracy in diagnosing knee injuries and may enable more effective treatment.

BackgroundThe presence of a positive pivot shift after surgical repair of the ACL is considered an important indicator of a failed reconstruction. The ability to predict the result of a pivot shift test after an ACL reconstruction using variables that can be measured prior to surgery could provide an indication of which patients may be at-risk of a poor surgical outcome.The purpose of this study was to determine whether structural characteristics of the femur and tibia, measured using plain radiographs, were associated with the result of the pivot shift test in unilateral ACL reconstructed patients.MethodsSixteen patients who had undergone unilateral ACL reconstruction were divided into two groups based on the results of manual pivot shift testing: 1) Pivot group; and 2) No pivot group. All patients had standing true lateral radiographs of both knees. Structural measurements of the tibia and femur were made on both knees. In addition, two new variables were created to describe the tibiofemoral mismatch: 1) Femur Tibia Size Ratio (FTSR); and 2) Tibia to Posterior Femoral Condyle Ratio (TPFCR). These measures were compared within groups and between groups.ResultsNone of the individual structural characteristics were significantly different when compared between groups. No individual structural characteristics had a significant association with the presence of a positive pivot shift. When a between-group analysis was performed, both the FTSR (p < 0.03) and the TPFCR (p < 0.01) were significantly different between the Pivot group and the No Pivot group. A larger FTSR ratio, or a larger femur relative to the tibia, was associated with a positive pivot shift. A smaller TPFCR ratio, or a smaller tibial depth relative to the depth of the lateral posterior femoral condyle, was associated with a positive pivot shift.ConclusionsStructural characteristics in the lateral femoral condyle and lateral tibial plateau were found to be associated with the presence of a positive pivot shift. These characteristics could separate between patients in the Pivot group and the No Pivot group. Two indices, the FTSR and the TPFCR, provided better predictive value than individual characteristics in identifying patients with a knee that was structurally “at-risk” for developing a positive pivot shift.

PurposeThe purpose of this study was to collect knee laxity data using a robotic testing device. The data collected were then compared to the results obtained from manual clinical examination.MethodsTwo human cadavers were studied. A medial collateral ligament (MCL) tear was simulated in the left knee of cadaver 1, and a posterolateral corner (PLC) injury was simulated in the right knee of cadaver 2. Contralateral knees were left intact. Five blinded examiners carried out manual clinical examination on the knees. Laxity grades and a diagnosis were recorded. Using a robotic knee device which can measure knee laxity in three planes of motion: anterior–posterior, internal–external tibia rotation, and varus–valgus, quantitative data were obtained to document tibial motion relative to the femur.ResultsOne of the five examiners correctly diagnosed the MCL injury. Robotic testing showed a 1.7° larger valgus angle, 3° greater tibial internal rotation, and lower endpoint stiffness (11.1 vs. 24.6 Nm/°) in the MCL-injured knee during varus–valgus testing when compared to the intact knee and 4.9 mm greater medial tibial translation during rotational testing. Two of the five examiners correctly diagnosed the PLC injury, while the other examiners diagnosed an MCL tear. The PLC-injured knee demonstrated 4.1 mm more lateral tibial translation and 2.2 mm more posterior tibial translation during varus–valgus testing when compared to the intact knee.ConclusionsThe robotic testing device was able to provide objective numerical data that reflected differences between the injured knees and the uninjured knees in both cadavers. The examiners that performed the manual clinical examination on the cadaver knees proved to be poor at diagnosing the injuries. Robotic testing could act as an adjunct to the manual clinical examination by supplying numbers that could improve diagnosis of knee injury.Level of evidenceLevel II.