Abstract
Background
Rotational acetabular osteotomy (RAO) is an effective treatment option for symptomatic acetabular dysplasia. However, excessive lateral and anterior correction during the periacetabular osteotomy may lead to femoroacetabular impingement. We used preoperative planning software for total hip arthroplasty to perform femoroacetabular impingement simulations before and after rotational acetabular osteotomies.
Methods
We evaluated 11 hips in 11 patients with available computed tomography taken before and after RAO. All cases were female and mean age at the time of surgery was 35.9 years. All cases were early stage osteoarthritis without obvious osteophytes or joint space narrowing.
Radiographic analysis included the center-edge (CE) angle, Sharp's acetabular angle, the acetabular roof angle, the acetabular head index (AHI), cross-over sign, and posterior wall sign. Acetabular anteversion was measured at every 5 mm slice level in the femoral head using preoperative and postoperative computed tomography.
Impingement simulations were performed using the preoperative planning software ZedHip (LEXI, Tokyo, Japan). In brief, we created a three-dimensional model. The range of motion which causes bone-to-bone impingement was evaluated in flexion (flex), abduction (abd), external rotation in flex 0°, and internal rotation in flex 90°. The lesions caused by impingement were evaluated.
Results
In the radiographic measurements, the CE angle, Sharp's angle, acetabular roof angle, and AHI all indicated improved postoperative acetabular coverage. The cross-over sign was recognized pre- and postoperatively in each case. Acetabular retroversion appeared in one case before RAO and in three cases after RAO. Preoperatively, there was a tendency to reduce the acetabular anteverison angle in the hips from distal levels to proximal. In contrast, there was no postoperative difference in the acetabular anteversion angle at any level. In our simulation study, bone-to-bone impingement occurred in flex (preoperative/postoperative, 137°/114°), abd (73°/54°), external rotation in flex 0°(34°/43°), and internal rotation in flex 90°(70°/36°). Impingement occurred within internal rotation 45°in flexion 90°in two preoperative and eight postoperative cases. The impingement lesions were anterosuperior of the acetabulum in all cases.
Discussion
It is easy to make and assess an impingement simulation using preoperative planning software, and our data suggest the simulation was helpful in a clinical setting, though there were some remaining problems such as approximation of the femoral head center and differences in femur movement between the simulation and reality. In the postoperative simulation there was a tendency to reduce the range of motion in flex, abd, and internal rotation in flex 90°.
There was a correlation between acetabular anteversion angle and flex. Since impingement occurred within internal rotation 45°in flexion 90°in eight postoperative simulations, we consider there is a strong potential for an increase in femoroacetabular impingement after RAO.
