Early results of the Bernese periacetabular osteotomy: The learning curve at an academic medical center

Background: Most reports on the results of the Bernese periacetabular osteotomy for the treatment of developmental dysplasia of the hip have been by the originators of the procedure. In 1997, we began to use this osteotomy without direct training from the originators of the procedure. Methods: Seventy-three patients (eighty-three hips) underwent a Bernese periacetabular osteotomy between 1997 and 2003 and were followed prospectively with use of the Harris hip score to assess clinical results and with use of anteroposterior pelvic and false-profile lateral plain radiographs to assess radiographic results. The three-dimensional position of the acetabulum was recorded preoperatively and postoperatively. The mean duration of follow-up was forty-six months. Results: The average Harris hip score improved from 54 to 87 points (p < 0.001). Three hips (three patients) had a conversion to total hip arthroplasty at two, three, and four years after the periacetabular osteotomy. Preoperatively, fifty-four of the eighty-three acetabula were anteverted, and twenty-nine were either retroverted or had neutral wall relationships. Postoperatively, sixty-five hips (78%) were anteverted. Radiographically, in preoperatively anteverted hips, the average center-edge angle improved from 3 degrees to 29 degrees (p < 0.0001), the average anterior center-edge angle improved from 5 degrees to 31 degrees (p < 0.0001), and the acetabular index improved from 25 degrees to 5 degrees (p < 0.0001). In preoperatively retroverted or neutral hips, the average center-edge angle improved from 13 degrees to 33 degrees (p < 0.0001), the average anterior center-edge angle improved from 15 degrees to 36 degrees (p < 0.0001), and the acetabular index improved from 19 degrees to 2 degrees (p < 0.0001). Complications included four hematomas, three transient femoral nerve palsies, two deep wound infections, and one transient sciatic nerve palsy. Nine of the ten major complications and all four of the failed osteotomies occurred in the first thirty hips in which the index procedure was performed. Conclusions: In our experience, the early results of the Bernese periacetabular osteotomy have been encouraging, with a 92% survival rate at thirty-six months. The occurrence of complications demonstrates a substantial learning curve. Recognition of the true preoperative acetabular version and reorientation of the acetabulum into an appropriately anteverted position have become important factors in surgical decision-making. Level of Evidence: Therapeutic Level IV See Instructions to Authors for a complete description of levels of evidence.