The development of osteoarthritis following anterior cruciate ligament (ACL) injury reconstruction is a well-documented and vexing problem. Although much of the existing literature relies on the detection of radiographic signs of osteoarthritis to quantify osteoarthritis risk, the authors of the current study have utilized a different approach. Relying on billing data in a single-payer health-care system, they identified patients who had undergone cruciate ligament reconstruction and a matched control group. They then compared the risk of subsequent knee arthroplasty between these populations and attempted to identify additional risk factors for knee arthroplasty in the cruciate ligament reconstruction group.
Perhaps not surprisingly, given the abundance of literature demonstrating rather high osteoarthritis risk after ACL injury and reconstruction, Leroux and colleagues noted an increased (sevenfold) cumulative incidence of knee arthroplasty among those who had undergone cruciate ligament reconstruction. Unfortunately, because of the limitations of their study design and data set, they had no data regarding other factors that may contribute to the risk of subsequent knee arthroplasty beyond those that were matched (age, sex, income, address, and comorbidities). Perhaps the most important among these unknown factors is the degree of osteoarthritis present at the time of cruciate ligament reconstruction. Other limitations, such as the inclusion of some revision cases and posterior cruciate ligament reconstructions in the cohort, a lack of knowledge regarding the laterality of cruciate ligament reconstruction and knee arthroplasty, and limited knowledge of concurrent procedures at the time of cruciate ligament reconstruction, are discussed by the authors and should also be considered.
Even at seven times that of the control group, the cumulative incidence of knee arthroplasty at fifteen years was quite low (1.4%). This finding likely reflects the relatively short follow-up in what is generally a young patient population, perhaps too young for many surgeons to consider for knee arthroplasty, even the patients with advanced osteoarthritis. The literature tells us that radiographic signs of osteoarthritis are much more prevalent ten to fifteen years after ACL reconstruction1,2. The relatively low risk of knee arthroplasty in this series should not be equated to a low risk of osteoarthritis. It is interesting to note that although well over 50% of patients in the cohort were under the age of thirty years at the time of the cruciate ligament reconstruction, 76% of the patients who underwent knee arthroplasty were at least at the age of thirty-five years (median, forty-two years) at the time of the cruciate ligament reconstruction. Because it is these older patients who are driving the data regarding knee arthroplasty risk, one must be cautious in applying these findings to younger patients.
Within the reconstructed cohort, the authors found increased age, female sex, increased comorbidities, smaller surgeon volume, and academic health-care setting to be associated with increased knee arthroplasty risks. The authors have proposed explanations for their findings regarding associations between knee arthroplasty rates and surgeon and hospital factors; however, one must bear in mind the distinction between association and causation. Other unknown and uncontrolled factors may contribute to these associations and should be considered when reviewing these findings.
Interestingly, the authors found that meniscal resection was not associated with an increased risk of subsequent knee arthroplasty. This observation is in contrast to numerous published studies in which meniscectomy has been demonstrated to be among the strongest predictors of radiographic evidence of osteoarthritis following ACL reconstruction3,4. The length of follow-up and the choice of outcome measure are likely the key factors driving the different observed effect of meniscectomy in the study by Leroux et al. Although a fifteen-year duration of follow-up is relatively long, it is likely too short to adequately describe the progression from ACL injury and surgery to the development of osteoarthritis that is severe enough to warrant knee arthroplasty in the majority of patients, particularly those without preexisting osteoarthritis at the time of ACL reconstruction. The factors associated with increased knee arthroplasty risk in this study are thus best considered to be associated with early knee arthroplasty following cruciate ligament reconstruction. At-risk patients are likely those with preexisting osteoarthritis, those who are older, and those in whom reconstruction failed or a complication ensued. The strong association between meniscectomy and subsequent development of radiographic signs of osteoarthritis reported elsewhere in the literature suggests that meniscectomy will also be associated with increased knee arthroplasty risk with longer follow-up.
The authors should be commended on this interesting and novel study. Such work has great potential to identify possible risk factors and to direct and to inform future prospective studies specifically designed to assess risk factors for poor outcome after cruciate ligament injury and reconstruction.