What Are the Early Outcomes of True Reverse Periacetabular Osteotomy for Symptomatic Hip Overcoverage?

Acetabular overcoverage is associated with pincer-type femoroacetabular impingement (FAI). A subtype of acetabular overcoverage is caused by a deep acetabulum with a negatively tilted acetabular roof, in which acetabular reorientation may be a preferable alternative to rim trimming to uncover the femoral head. We introduced the true reverse periacetabular osteotomy (PAO) in 2003, which in contrast to an anteverting PAO, also flexes and abducts the acetabulum relative to the intact ilium to decrease anterior and lateral femoral head coverage and correct negative tilt of the acetabular roof. To our knowledge, the clinical results of the true reverse PAO have not been evaluated.

For a group of patients who underwent reverse PAO, (1) Do patients undergoing reverse PAO demonstrate short-term improvement in pain, function, and hip ROM, and decreased acetabular coverage, as defined by lateral and anterior center-edge angle and Tönnis angle? (2) Are there identifiable factors associated with success or adverse outcomes of reverse PAO as defined by reoperation, conversion to THA, or poor patient-reported outcome scores? (3) Are there identifiable factors associated with early complications?

Between 2003 and 2017, two surgeons carried out 49 reverse PAOs in 37 patients. Twenty-five patients had unilateral reverse PAO and 12 patients had staged, bilateral reverse PAOs. To ensure that each hip was an independent data point for statistical analysis, we chose to include in our series only the first hip in the patients who had bilateral reverse PAOs. During the study period, our general indications for this operation were symptomatic lateral and anterior acetabular overcoverage causing FAI that had failed to respond to previous conservative or surgical treatment. Thirty-seven hips in 37 patients with a median (range) age of 18 years (12 to 41; interquartile range 16 to 21) were included in this retrospective study at a minimum follow-up of 2 years (median 6 years; range 2 to 17). Thirty-four patients completed questionnaires, 24 patients had radiographic evaluation, and 23 patients received hip ROM clinical examination. However, seven patients had not been seen in more than 5 years. The clinical and radiographic parameters of all 37 hips that underwent reverse PAO in 37 patients from a longitudinally maintained institutional database were retrospectively studied preoperatively and postoperatively. Adverse outcomes were considered conversion to THA or a WOMAC pain score greater than 10 at least 2 years postoperatively. Patient-reported outcomes, radiographic measurements, and hip ROM were evaluated preoperatively and at most recent follow-up using a paired t-test or McNemar test, as appropriate. Linear regression analysis was used to assess for identifiable factors associated with clinical outcomes. Logistic regression analysis was used to assess for identifiable factors associated with adverse outcomes and surgical complications. All tests were two-sided, and p values less than 0.05 were considered significant.

At a minimum of 2 years after reverse PAO, patients experienced improvement in WOMAC pain (-7 [95% CI -9 to -5]; p < 0.001), stiffness (-2 [95% CI -3 to -1]; p < 0.001), and function scores (-18 [95% CI -24 to -12]; p < 0.001) and modified Harris Hip Score (mHHS) (20 [95% CI 13 to 27]; p < 0.001). The mean postoperative hip ROM improved in internal rotation (8° [95% CI 2° to 14°]; p = 0.007). Acetabular coverage, as defined by lateral center-edge angle (LCEA), anterior center-edge angle (ACEA), and Tönnis angle, improved by -8° (95% CI -12° to -5°; p < 0.001) for LCEA, -12° (95% CI -15° to -9°; p < 0.001) for ACEA, and 9° (95% CI 6° to 13°; p < 0.001) for Tönnis angle. The postoperative severity of radiographic arthritis was associated with worse WOMAC function scores such that for each postoperative Tönnis grade, WOMAC function score increased by 12 points (95% CI 2 to 22; p = 0.03). A greater postoperative Tönnis grade was also correlated with worse mHHS, with an average decrease of 12 points (95% CI -20 to -4; p = 0.008) in mHHS for each additional Tönnis grade. Presence of a positive postoperative anterior impingement test was associated with a decrease in mHHS score at follow-up, with an average 23-point decrease in mHHS (95% CI -34 to -12; p = 0.001). Nineteen percent (7 of 37) of hips had surgery-related complications. Four hips experienced adverse outcomes at final follow-up, with two patients undergoing subsequent THA and two with a WOMAC pain score greater than 10. We found no factors associated with complications or adverse outcomes.

The early clinical and radiographic results of true reverse PAO compare favorably to other surgical treatments for pincer FAI, suggesting that reverse PAO is a promising treatment for cases of pincer FAI caused by global acetabular overcoverage. However, it is a technically complex procedure that requires substantial training and preparation by a surgeon who is already familiar with standard PAO, and it must be carefully presented to patients with discussion of the potential risks and benefits. Future studies are needed to further refine the indications and to determine the long-term outcomes of reverse PAO.

Level IV, therapeutic study.