Mechanical Testing and Motion Tracking

The biomechanical team (MCA, ST, AC) was blinded to the experimental group of each specimen at the time of testing. Five 1-cm diameter spherical reflective markers were used for motion tracking on each specimen. These markers were placed at the superior and inferior ends of the proximal and distal pieces of the fractured patella along its anteromedial surface using cyanoacrylate glue to track gap formation. An additional marker was placed on the medial femoral epicondyle and proximal tibia to determine the knee angle during flexion. During specimen testing, three Qualysis motion capture cameras (Qualysis, Highland Park, IL, USA) collected the marker spatial coordinates at a sampling rate of 1000 Hz. A stationary recording was conducted to determine the distance between the patellar surface markers. Subsequent marker positions were recorded relative to this initial position by the processing software.

Mechanical loading of the specimens was modeled after previous work by our laboratory [1]. The quadriceps tendon and its attached nylon webbing was connected to a cable-pulley system driven by an Instron 5866 testing apparatus (Instron, Norwood, MA, USA) with a 10 kN load cell. The femur was immobilized via a custom fixture that only allows translation of the patella and subsequent extension of the tibia. A 5-pound dumbbell was affixed to the tibia 33 cm distal to the axis of knee rotation, which simulated the moment arm of a 70-kilogram individual with an intact extremity [1]. Each cycle was conducted by translating the quadriceps at 1 mm/s until the knee had extended from its resting 90^o position to 5^o, and then returning to the 90^o position. This process was repeated for 40 cycles, and gap formation was determined for each cycle. Phosphate-buffered solution (Sigma-Aldrich, St. Louis, MO, USA) was sprayed to avoid tissue dehydration.

Upon completion of cyclic testing, the knee was immobilized in 45^o flexion via webbing tie-down. To determine load to failure, the pulley fixture was pulled at a displacement rate of 1 mm per second [6] until the construct achieved the ultimate failure load. Failure load was defined as the maximum force sustained by the construct before force was no longer measured by the recording apparatus. The patella was then sharply dissected from the leg along the distal quadriceps and proximal patellar tendons.

Areal bone mineral density (BMD) of the patellae was determined to assess its effect on ultimate load to failure. BMD was determined from dual energy x-ray absorptiometry (DEXA) scans conducted using an UltraFocus Digital Radiography system (Faxitron, Tuscon, AZ, USA) at 1.05 X magnification. BMD was recorded within a 3-mm by 3-mm region of bone directly inferior to the fracture site. The groups were pooled during testing to ensure that the biomechanical team remained blinded to the groups. Of note, BMD was obtained from seven pairs, rather than 10 pairs, of cadaveric specimens because three were used for an alternate bone quality assessment (data not shown or reported). There was no difference in mean BMD between the patellae in the recessed (1.06 ± 0.262 g/cm²) and prominent (1.03 ± 0.197 g/cm²) screw groups (p = 0.846; 95% confidence interval difference of means, -0.294 to 0.245 g/cm²).