Session IV - Pelvis

Computed Tomography Analysis of Posterior-Wall Fractures of the Acetabulum Treated Operatively

Berton R. Moed, MD; Seann E. Willson Carr, MD; Konrad I. Gruson, MD; J. Tracy Watson, MD; Joseph G. Craig, MD; Wayne State University, Detroit, Michigan, USA

Purpose: A number of studies have demonstrated that a disparity exists between the accuracy of the surgical reduction of posterior-wall acetabular fractures, as determined by plain radiographs, and clinical outcome. The purpose of this study was to evaluate the results of the operative treatment of posterior-wall fractures of the acetabulum in relationship to quality of fracture reduction as assessed by postoperative two-dimensional computed tomography.

Methods: The results were analyzed of 67 patients who had open reduction and internal fixation of an unstable posterior-wall fracture of the acetabulum and postoperative two-dimensional computed tomography. Sixty-one patients were followed for a mean of 5 years (range, 2 to 14) after the injury. The remaining six patients, with clearly poor results, were followed for less than 2 years. All patients were studied preoperatively and postoperatively with three standard plain radiographs (an anteroposterior and two Judet 45° oblique pelvic radiographs) and two-dimensional computed tomography. The patients' functional outcome was evaluated with use of the clinical grading system adopted by Letournel, incorporating modifications by Matta. The radiographs were graded according to the criteria described by Matta. The two-dimensional computed tomograms were examined by use of fracture gap and offset measurements. Additional patient, fracture, and radiographic variables were collated in an attempt to identify possible associations with functional outcome. These variables included patient age, gender, time to reduction of the dislocation (categorized as less than 12 hours, 12 to 24 hours, and more than 24 hours), impaction injury to the femoral head, involvement of the weight-bearing acetabular dome, intraarticular fracture comminution (defined as three or more fragments), presence of marginal impaction, radiographic evidence of osteonecrosis of the femoral head, and radiographic evidence of severe heterotopic ossification.

Results: Clinical outcome was graded as excellent in 31 patients (46%), very good in 20 (30%), good in 8 (12%), and poor in 8 (12%). Final radiographic results were graded as excellent in 53 hips (79%), good in 4 (6%), fair in 3 (5%), and poor in 7 (10%). There was a strong association between clinical outcome and the final radiographic grade. Fracture reduction was graded as anatomic in 65 and imperfect in 2, as determined by plain radiography, and did not correlate with clinical outcome. However, postoperative computed tomography revealed incongruency (offset) of more than 2 mm in 11 and fracture gaps (negative defect) of 2 mm or more in 52. Fracture gaps of 10 mm or more in any dimension or a total gap area of 35 sq mm or more were associated with a poor result. The main risk factors for a poor result were a residual fracture gap width of 10 mm or more and osteonecrosis of the femoral head.

Discussion and Conclusions: The disparity between the accuracy of posterior wall fracture reduction, as determined by plain radiographs obtained postoperatively, and clinical outcome has been well described. Postoperative computed tomography detects the degree of residual fracture displacement more accurately than do plain radiographs. The accuracy of surgical reduction as assessed on postoperative computed tomography is predictive of clinical outcome.