PREDICTORS OF RADIOLOGICAL OSTEOARTHRITIS AT FIVE YEARS AFTER ANTERIOR CRUCIATE LIGAMENT RECONSTRUCTION

To evaluate the predictive factors of post-traumatic knee osteoarthritis (OA) in the medial, lateral, and patellofemoral compartments at 5-years post-ACL reconstruction, within a large randomized clinical trial. Three-hundred-and-thirty patients (110 patients/group), ages 14–50 years, were randomized intra-operatively to ACL reconstruction with either a patellar tendon, single-bundle semitendinosus/gracilis tendon, or double-bundle semitendinosus and gracilis tendon autograft. Two- and 5-year clinical and quality-of-life outcomes have been previously published. Standardized radiographs (bilateral P-A weight-bearing, lateral, and bilateral skyline patella views) were performed on all patients at baseline, 2-years and 5-years follow-up. Radiographic OA was graded separately in the medial, lateral, and patellofemoral compartments, using the International Knee Documentation Committee (IKDC) radiographic grading scale. An independent, fellowship-trained orthopaedic surgeon graded the radiographs for absence (A–Normal, B–Nearly Normal) or presence (C–Abnormal, D–Severely Abnormal) of OA. Three separate multi-variable logistic regression analyses were performed with medial, lateral, and patellofemoral radiographic OA as the dependent outcome variable. The independent predictor variables included: demographics (age and sex), knee alignment, stability (Lachman and pivot shift tests), meniscal treatment and chondral condition at index surgery, graft type, graft failure, re-injury, and secondary surgery. A bivariate logistic regression was performed for each individual predictor. For each knee compartment, individual predictor variables with p<0.1 in a bivariate regression analysis were added to a multi-variable logistic regression model if the event rate was adequate. Odds ratios (OR) and 95% confidence intervals (95%CI) were reported. IKDC grades for radiographic OA were available for 302 patients (91.5%) at 5-years post-operative. Medial OA was assessed in 35 patients (10.6%). Varus alignment, medial meniscectomy, medial compartment chondral damage, age > 35 years and male sex were identified as individual predictors of radiographic OA. However, only medial meniscectomy (OR 5.8 (95%CI 2.7–12.6, p<0.01)) and varus knee alignment (OR 2.4 (95%CI 1.1–5.1; p=0.03)) significantly predicted medial OA. Lateral OA was assessed in 67 patients (20.3%). Graft type, pivot shift, valgus alignment, lateral meniscectomy, lateral meniscal repair, and lateral compartment chondral damage were identified as individual predictors of radiographic OA. However, only patellar tendon graft (OR 2.4 (95%CI 1.2–4.8 p=0.02)), lateral meniscectomy (OR 2.6 (95%CI 1.3–5.2, p≤0.01)), lateral meniscal repair (OR 3.3 (95%CI 1.5–7.4, p≤0.01)) and lateral compartment chondral damage (OR 2.0 (95%CI 1.0 – 3.9, p=0.04)) significantly predicted lateral OA. Based on a bivariate analysis with 9 patients (2.7%) with patellofemoral OA, only patellofemoral chondral damage (OR 5.7 (95%CI 1.5–21.8, p=0.01)) was a significant predictor of OA. Graft failure, traumatic re-injury, and secondary surgery did not predict presence of radiographic OA at 5-years post-operative in any compartment. Graft failure, traumatic re-injury, and secondary surgery did not predict presence of radiographic OA at 5-years post-operative in any compartment. At 5-years post-ACL reconstruction, varus alignment and medial meniscectomy significantly increase the risk of developing medial OA. Use of a patellar tendon graft, lateral meniscectomy, lateral meniscal repair, and chondral damage significantly increase the risk of developing lateral OA. Graft failure, traumatic re-injury, or secondary surgery were not predictive of radiographic OA in any compartment. Higher proportions of patients with OA are required to increase confidence in these associations.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.