Incidence and risk factors for non-union of the superior ramus osteotomy when hip dysplasia is treated with periacetabular osteotomy

Ganesh Sivamurugan; Robert W Westermann; Natalie Glass; John C Davison; Aspen Miller; Jacob Henrichsen; Todd O McKinley; Michael C Willey

doi:10.1093/jhps/hnad006

Incidence and risk factors for non-union of the superior ramus osteotomy when hip dysplasia is treated with periacetabular osteotomy

J Hip Preserv Surg. 2023 Apr 12;10(2):80-86. doi: 10.1093/jhps/hnad006. eCollection 2023 Jul.

Authors

Ganesh Sivamurugan¹, Robert W Westermann¹, Natalie Glass¹, John C Davison¹, Aspen Miller¹, Jacob Henrichsen¹, Todd O McKinley², Michael C Willey¹

Affiliations

¹ Department of Orthopaedic Surgery and Rehabilitation, University of Iowa Hospitals and Clinics, 200 Hawkins Drive, Iowa City, IA 52242, USA.
² Indiana University Health, Methodist Hospital, 1801 N Senate Boulevard, Suite 535, Indianapolis, IN 46202, USA.

Abstract

Periacetabular osteotomy (PAO) is a well-established surgical treatment for hip dysplasia. Few studies report risk factors for the development of superior ramus osteotomy non-union. The purpose of this investigation was to document the incidence and risk factors for this complication. We identified 316 consecutive hips that underwent PAO for symptomatic acetabular dysplasia with a minimum 1-year radiographic follow-up. We developed and validated a technique to measure the superior ramus osteotomy location on anterior-posterior (AP) pelvis radiographs and computed tomography. Logistic regression with generalized estimating equations was used to evaluate the relationships between odds of non-union and potential demographic and radiographic predictor variables in univariate and multivariate analyses. Twenty-nine (9.2%) hips developed superior ramus non-union. Age {median [interquartile range (IQR)] 23 years (18-35) healed versus 35 years (26-40) non-united, P = 0.001}, pre-operative lateral center-edge angle (LCEA) [16° (11-20) healed versus 10° (6-13) non-united, P < 0.001] and the distance from the superior ramus osteotomy to the ilioishial line [15.8 mm (13.2-18.7) healed versus 18.1 mm (16.2-20.5) non-united, P < 0.001] varied significantly between groups. Using multivariate analysis, moderate-to-severe dysplasia [LCEA < 15°, odds ratio (OR) 5.95, standard error (SE) 3.32, 95% confidence interval (CI) 1.99-17.79, P = 0.001], increased age (5-year increase, OR 1.29, SE 3.32, 95% CI 1.105-1.60, P-value = 0.018) and distance from the ilioishial line (3-mm increase, OR 1.67, SE 0.22, 95% CI 1.29-2.18, P < 0.001) were at increased risk of developing non-union. Superior ramus osteotomy non-union is common after PAO. Older age, moderate-to-severe dysplasia, and more medial osteotomy location were independent risk factors for non-union. Consideration should be made in high-risk patients for a more lateral superior ramus osteotomy and adjuvant medical and surgical interventions.