What Proportion of Patients Undergoing Bernese Periacetabular Osteotomy Experience Nonunion, and What Factors are Associated with Nonunion?

• Published in: Clinical orthopaedics and related research 2020-07, Vol.478 (7), p.1648-1656
• Main Authors: Selberg, Courtney M., Davila-Parrilla, Ariel D., Williams, Kathryn A., Kim, Young-Jo, Millis, Michael B., Novais, Eduardo N.
• Format: Article
• Language: English
• Subjects: Acetabulum; Acetabulum - diagnostic imaging; Acetabulum - physiopathology; Acetabulum - surgery; Activities of daily living; Adolescent; Adult; Age; Biomechanical Phenomena; Bone dysplasia; Bone grafts; Chondrodystrophy; Clinical Research; Female; Hip; Hip Dislocation - diagnostic imaging; Hip Dislocation - physiopathology; Hip Dislocation - surgery; Humans; Male; Medical records; Nonunion; Osteoarthritis; Osteotomy; Osteotomy - adverse effects; Pain; Patients; Postoperative Complications - diagnostic imaging; Postoperative Complications - etiology; Postoperative Complications - physiopathology; Quality of life; Radiography; Recovery of Function; Retrospective Studies; Risk Assessment; Risk Factors; Surgeons; Surgery; Time Factors; Treatment Outcome; Vitamin D; Wound Healing; Young Adult
• ISSN: 0009-921X; 1528-1132
• DOI: 10.1097/CORR.0000000000001296

The Bernese periacetabular osteotomy (PAO) is one of the most-used surgical techniques to treat symptomatic acetabular dysplasia. Although good functional and radiographic short-term and long-term outcomes have been reported, several complications after PAO have been described. One complication that may compromise clinical results is nonunion of an osteotomy. However, the exact prevalence and risk factors associated with nonunion are poorly elucidated. (1) What proportion of patients have complete bony healing versus nonunion during the first year after PAO? (2) What is the clinical and functional impact of nonunion at a minimum of 1 year after PAO, as assessed by the modified Harris hip score (mHHS) and the Hip Disability and Osteoarthritis Outcome Score (HOOS)? (3) What patient-specific or surgery-specific factors are associated with nonunion at 6 months and at a minimum of 1 year postoperatively? Between January 2012 and December 2015, we retrospectively identified 314 patients who underwent PAO at our institution. During this period, 28 patients with a diagnosis different from symptomatic acetabular dysplasia (reverse PAO for acetabular over-coverage: n = 25; PAO for skeletal chondrodysplasia: n = 3) underwent PAO but were ineligible to participate. Hence, 286 patients underwent PAO to treat symptomatic acetabular dysplasia during the study period and were considered eligible. Inclusion criteria were patients with a complete set of postoperative radiographs (AP, Dunn lateral, and false-profile) at 12 months or more postoperatively. Eighteen percent (51 of 286) of the patients underwent staged, bilateral PAOs, but we only included the first PAO. Finally, 14% (41 of 286) of the patients were excluded because they had an incomplete set of postoperative radiographs at 12 months or more. The study comprised 245 patients. Eighty-five percent (209 of 245) of the patients were female and the mean age at surgery was 24 years ± 9 years. The healing status (complete healing vs. nonunion) was recorded for ischial, superior pubic, supraacetabular, and posterior column osteotomies at each subsequent visit. Nonunion was defined as noncontiguous osseous union with a persistent radiolucent line across any osteotomy site and was recorded at 3 months, approximately 6 months, and approximately 12 months postoperatively. Calculation of Cohen's kappa statistic coefficients showed the classification had perfect interobserver agreement (0.53; 95% confidence interval, 0.12-0.9