Does Periacetabular Osteotomy Affect the Load Distribution on the Knee?

Periacetabular osteotomy (PAO) treats developmental dysplasia of the hip (DDH) by reducing load on the hip and improving joint function. Untreated DDH affects lower extremity alignment and alters knee morphology, with valgus alignment more pronounced in hip osteoarthritis secondary to DDH. While PAO may influence knee mechanics, its association with subchondral bone density in the tibiofemoral joint remains unclear.

(1) To what degree is PAO associated with changes in the distribution of subchondral bone density in the knees of patients with DDH? (2) Is PAO associated with altered subchondral bone density distribution in patients with DDH such that they more closely resemble a control cohort of patients?

We conducted a retrospective chart review to evaluate the association of PAO with knees in female patients with DDH. From January 2015 to December 2021, 69 patients (≤ 49 years of age, center-edge angle ≤ 25°) underwent PAO. Of these, 38% (26) of patients were included after excluding patients for bilateral operations, lack of follow-up, or incomplete CT data. A power analysis required at least 20 hips per group based on the past study. For comparison, we reviewed 63 patients undergoing joint-preserving surgery for idiopathic osteonecrosis from January 2014 to December 2024, with 32% (20) of female patients meeting criteria (unilateral hip necrosis only) for the control group. PAO resulted in sufficient acetabular coverage and improved clinical scores in patients. Importantly, no change in lower limb alignment was observed postoperatively. The distribution and quantification of subchondral bone density in the proximal tibial articular surface were measured using CT osteoabsorptiometry (CT-OAM). This was achieved by assessing radiodensity variations in Hounsfield units (HUs) and mapping these as two-dimensional visualizations. The high-density area within these regions was defined as the top 20% of HUs. The medial and lateral tibial compartments were divided into three subregions of equal width in the coronal direction: lateral-lateral, lateral-central, lateral-medial, medial-medial, medial-central, and medial-lateral. Each subregion percentage represented by the high-density area was calculated (percentage of high-density area). Our primary study goal was to evaluate the association of PAO with changes in subchondral bone density distribution in the knees of patients with DDH. To achieve this, we utilized CT-OAM to map subchondral bone density patterns before and after surgery. Our secondary study goal was to determine whether PAO results in a subchondral bone density distribution in patients with DDH that more closely resembles that of a control cohort without DDH. For this goal, we analyzed radiographic and CT data to identify changes in high-density areas across tibial plateaus and compared preoperative and postoperative results within the PAO group and between the PAO and control groups.

Preoperatively, the mean ± SD percentage of high-density area of the medial region was lower in the PAO group compared with the control group (control versus PAO preoperative 61% ± 12% versus 50% ± 20%; p = 0.02). After PAO, the percentage of high-density area of the medial region increased (preoperative versus postoperative 50% ± 20% versus 58% ± 19%; p = 0.003) and was not different from the control group postoperatively (control versus PAO postoperative 61% ± 12% versus 58% ± 19%; p = 0.16).

Our findings suggest that DDH may cause a lateral shift in knee loading distribution. PAO appears to modify this loading pattern, based on subchondral bone density, making it more similar to one in a control cohort of knees. However, long-term follow-up studies are necessary to confirm whether early changes in subchondral bone density because of PAO are associated with subsequent knee degeneration.

Level III, therapeutic study.