Abstract
Objectives:
Revision anterior cruciate ligament (ACL) reconstruction has been demonstrated to have outcomes inferior to those noted in primary ACL reconstructions in terms of patient-reported outcomes, return to activity and sports, and graft rupture rates. However, the long-term assessment of patients undergoing revision ACL reconstruction remains unknown, primarily due to the large number of patients needed to do an adequate analysis. A group was assembled to prospectively enroll and follow a revision ACL cohort to determine the risk factors for poor outcomes in a revision cohort, of which over 1,200 patients were enrolled from 83 surgeons and 52 sites and prospectively followed for 10 years. The overarching goal was to assess the long-term progression of outcomes following revision ACL reconstruction, and to determine how the initial factors at the time of revision surgery may influence and predict disease progression. The focus of this study is to characterize the 10-year natural history of this unique dataset, quantified by 3 complementary methodologies: patient-reported outcomes, radiologic measures, and physical examination measures. The purpose was to assess for clinical long-term outcomes, including incidence of graft failure, as well as signs and symptoms of knee osteoarthritis (OA) in this revision cohort at 10 years postoperatively.
Methods:
Patients were brought back to 6 sites for physical examination by an independent blinded sports medicine physician who was not involved with the revision ACL reconstruction. The physical exam included knee range of motion (ROM), ligamentous testing and bilateral KT-1000 assessment. A series of radiographs (standardized bilateral standing AP, synaflexer bent knee [similar to tunnel/Rosenberg], sunrise, and full extension lateral views) were obtained to assess for incidence of structural OA. Validated patient-reported outcome measures (PROMs) including International Knee Documentation Committee (IKDC) score, Knee injury and Osteoarthritis Outcome Score (KOOS), Western Ontario and McMaster Universities Arthritis Index (WOMAC), and Marx activity rating scale were also obtained. Two physicians independently graded the radiographs for severity of structural OA using the modified Kellgren-Lawrence (KL) grading system (grades 0 to 4). Structural OA was defined as a Kellgren-Lawrence grade of 3 or 4. Symptomatic OA was defined by the KOOS Pain subscale of < 70 points. Multivariate regression models were used to determine the predictors (risk factors) for both structural OA (KL grades 3 to 4) and symptomatic OA (KOOS pain < 70 points) at 10 years follow-up, controlling for the patient’s age, sex, body mass index (BMI), baseline PROMs, ACL graft choice, prior and current meniscal pathology and treatment at the time of revision surgery, chondral pathology at the time of revision surgery, and incidence of any subsequent surgeries.
Results:
205 patients (107 [52.4%] women) representing 40 surgeons returned at an average 12-year follow-up (range, 10-16 years). The mean (SD) age of this cohort at the time of their onsite evaluation was 40.2 (10.4) years with a BMI of 25.7 (range, 17.0-42.0).
Physical examination demonstrated loss of extension compared to the opposite knee in 118 (58%) and extension less than full (0 degrees) in 84 (41%). With regards to failure, a soft Lachman endpoint was noted in 41 patients (20%). A Lachman >5 mm was noted in 20 patients (10%), and 26 patients (13%) had a grade 2 pivot shift and 4 (2%) had a grade 3 pivot shift. KT-1000 measurements demonstrated 23 (11%) patients with 5 mm or greater involved versus uninvolved side-to-side difference. The blinded surgeons in their opinion detected 37 (18%) with a nonfunctional ACL. If all nonfunctional parameters were applied (Lachman ≥2, Pivot shift ≥2, KT-1000 side-to-side difference of ≥5mm, and surgeon determining failure) only 8 patients (4%) met all failure parameters.
Radiographs demonstrated that 115 (56%) exhibited joint space narrowing of the tibiofemoral joint (K-L grades 3 to 4) in their involved knee compared to 28 (14%) in their uninvolved knee (Table 1). The significant drivers of a higher KL grade (structural OA) at 10 years were found to be higher age, higher baseline BMI, having a medial meniscus excision performed either prior to or at the time of revision surgery, having a prior lateral meniscal excision, or having a subsequent surgery (p<0.05; Table 2). Sex, baseline activity level, graft choice, and chondral pathology at the time of revision were not significant.
PROMs of the onsite group reflected the PROMs of the overall cohort at 10 years (Figure 2). There were 40 onsite subjects (20%) that reported KOOS pain scores of < 70 points (defined as symptomatic OA), while 91 subjects (44%) reported KOOS pain scores over 90 points (defined as no pain). The predictors of increased pain (lower KOOS pain score) at 10 years were subjects who had higher pain scores at the time of revision surgery, grades 3 to 4 chondral pathology in the lateral compartment at the time of revision surgery, medial meniscus excision performed prior to the time of revision surgery or having a subsequent surgery (p<0.05; Table 3). Conversely, having a hamstring autograft or an allograft (compared to a bone-tendon-bone autograft) at the time of revision surgery predicted less symptomatic pain at 10 years. Age, sex, baseline BMI, activity level, and lateral meniscal pathology at the time of revision were not significant predictors of increased knee pain at 10 years.
Conclusions:
Outcomes in this first ever report of a revision ACL cohort at minimum 10 years follow-up demonstrates worrisome outcomes at a still young age. This study demonstrated a loss of ROM in 41-58% of the cohort, an 18% graft failure rate, 56% who exhibited KL grades of 3 to 4, and 20% who reported KOOS pain scores of less than 70 points, which collectively, all emphasize the challenge of managing the revision ACL reconstruction patient.
