Abstract
Introduction:
Femoro-acetabular impingement reduces the range of motion of the hip joint and is thought to contribute to hip osteoarthritis. Surgical treatments attempt to restore hip motion through resection of bone at the head-neck junction. Due to the broad range of morphologies of FAI, the methodology of osteochondroplasty has been difficult to standardize and often results in unexpected outcomes, ranging from minimal improvement in ROM to excessive head resection with loss of cartilage and even neck fracture. In this study we test whether a standardized surgical plan based on a pre-determined resection path can restore normal anatomy and ROM to the CAM-impinging hip.
Methods:
Computer models of twelve femora with classic signs of cam-type FAI were reconstructed from CT scans. The femoral shaft and neck were defined with longitudinal axes and the femoral head by a sphere of best fit. Boundaries defining the maximum extent of anterior resection were constructed: (i) superiorly and inferiorly along the anterior femoral neck at 12:30 and 5:30 on the clock face, approximating the locations of the vascularized synovial folds; (ii) around the head-neck junction along the edge of the articular cartilage; and (iii) at the base of the neck, perpendicular to the neck axis, 20–30 mm lateral to the articular edge. All four boundaries were used to form 3 alternative resection surfaces that provided resection depths of 2 mm (small), 4 mm (medium), and 6 mm (large) at the location of the cam lesion. Solid models of each femur after virtual osteochondroplasty were created by Boolean subtraction of each of the resection surfaces from the original femoral model. For each depth of neck resection, we measured the following: (i) alpha angle, (ii) anterior offset of the head-neck junction, and (iii) volume of bone removed. Before and after each resection, we also measured the maximum internal rotation of the hip in 90° flexion and 0° abduction.
Results:
The initial alpha angles of the twelve femora averaged 63.8°, with corresponding average anterior head-neck offset of 5.8 mm and average maximum internal rotation of 16.3°. Impingement prevented one specimen from attaining the initial position of 90° flexion and 0° abduction. Implementation of pre-operative plans demonstrated that normal alpha angles (<55°) could be achieved using resection depths of 2 mm, 4 mm, and 6 mm (small: 48.8°, medium: 40.8°, large: 35.3°). The corresponding changes in internal rotation were +7.7° (to 24.0°; p < 0.001), +11.8° (to 28.1°; p < 0.001), and +14.7° (to 31°; p < 0.001), with anterior offsets of 8.0 mm, 9.9 mm, and 11.2 mm, respectively. The corresponding volume of resected bone ranged from 0.57 cm3 to 3.20 cm3.
Conclusions:
Our study shows that a standardized method of pre-operative planning may enable surgeons to restore normal hip ROM, alpha angles, and anterior offsets through pre-determined bony resection. This method shows how osteochondroplasty can be customized to each deformity, thus removing only the necessary amount of bone to correct each abnormality. We believe implementation of our boundaries and method will enable surgeons to consistently and quantitatively reproduce and teach osteochondroplasty, and that this method is readily adaptable to computerized machining of the femur.
