Predictors of Radiographic Knee Osteoarthritis After Anterior Cruciate Ligament Reconstruction.
Li RT, Lorenz S, Xu Y, Harner CD, Fu FH, Irrgang JJ. Am J Sports Med. 2011 Oct 21. [Epub ahead of print]
Over the past few months, SMR had several posts highlighting that individuals with a history of knee injury are at greater risk for knee osteoarthritis (OA) even if they undergo surgeries like anterior cruciate ligament (ACL) reconstructions (see below). As clinicians we must strive to reduce the risk of knee OA for our injured patients. Unfortunately, it is unclear which variables might predict individuals who will develop knee OA after an injury. Therefore, Li et al. tried to determine the prevalence of knee OA after a single-bundle ACL reconstruction and to identify factors that predict the development of knee OA. This study was a secondary analysis derived from a study that assessed outcomes among 422 patients that underwent a single-bundle ACL reconstruction (Kowalchuk et al 2009). The current study focused on 249 patients that were entered in a research database and had physical examination and radiographic data. All of the ACL reconstructions were performed by one of two surgeons using the same surgical method. One orthopaedic surgeon scored the radiographic knee OA severity in medial and lateral tibiofemoral compartments as well as patellofemoral joints. The authors evaluated numerous potential risk factors: 1) patient characteristics (e.g., age, sex, body mass index [BMI], occupation), 2) surgical variables (e.g., concurrent meniscal tears, chondral lesions, graft type/placement), and 3) other factors (e.g., length of time between injury and surgery, length of follow-up time, need for revision surgery). The average time between surgery and follow-up was 7.9 years (range: 2.1 to 20.3 years) and on average patients were 26.4 years of age at the time of surgery. Between 25% and 32% of participants had radiographic evidence of knee OA in the medial or lateral tibiofemoral compartment or patellofemoral joint. In the final statistical models, four predictors were associated with increased odds of developing knee OA: pre-operative high BMI, medial chondral lesions, concurrent medial meniscectomy, and length of follow-up. Overall, predicting knee OA development based on these four predictors had a sensitivity of 60.2% (chance of positively diagnosing a patient who has pathology) and a specificity of 70.4% (chance of correctly identifying a healthy patient).
This is a helpful study for identifying predictors to identify patients that may develop knee OA after a single-bundle ACL reconstruction. While the authors state that this model is not ready to be used in clinics to predict knee OA development it does provide us helpful information. Not surprisingly concurrent injuries (i.e., cartilage lesions and meniscal damage) were risk factors for knee OA since these injuries further complicate the adverse biomechanical and biochemical changes associated with ACL injuries. It is also not surprising that the length of follow-up was related to knee OA development. It is possible that all joints with a history of trauma may be on a path towards knee OA but some might get to OA faster than other joints. If provided a long enough time perhaps all of these joints would progress to knee OA without a disease modifying or risk modifying intervention. Finally, there is the issue of high BMI which has been suggested as risk factor for knee OA. It is important for future studies to untangle whether the risk is associated with high body mass or high body fat percentage (or both). Regardless, BMI is the only predictor from this study that is modifiable. This highlights the need for us to educate our patients with a history of knee injuries that they are at increased risk for knee OA and therefore should reduce their other risk factors for knee OA by maintaining a healthy lifestyle as they age (e.g., remain physically active, maintain a healthy body weight). Have you ever discussed the risks of knee OA with your patients with a history of knee injuries? If so, how do you advise them?
Written by: Jeffrey Driban
Reviewed by: Stephen Thomas
Related Posts:
Articular Cartilage Damage and Long-term Changes After an ACL Tear
ACL Injured Knees Have Compromised Cartilage
Structural Changes Occur in Knees After ACL Tears
Knee Injuries and Knee Osteoarthritis: A Meta-Analysis
Related Posts:
Articular Cartilage Damage and Long-term Changes After an ACL Tear
ACL Injured Knees Have Compromised Cartilage
Structural Changes Occur in Knees After ACL Tears
Knee Injuries and Knee Osteoarthritis: A Meta-Analysis
Li RT, Lorenz S, Xu Y, Harner CD, Fu FH, & Irrgang JJ (2011). Predictors of Radiographic Knee Osteoarthritis After Anterior Cruciate Ligament Reconstruction. The American Journal of Sports Medicine PMID: 22021585
Wow. That's really real huge statement after reading this concept. Here everything important feedback about sport medicine. Thanks!
I would wonder if you took out BMI what the predictive values would look like?
Nicole, I agree. The authors did not report the results of entering the variables one step at a time into the predictive model (which can be a lot of work to check each model). It would have been interesting to see BMI measured a surgery and at follow-up as well.
Identification of predictive risk factors following ACL reconstruction for OA is an important aspect of creating successful treatment plans for these patients. The 4 predictive factors found in combination (medial meniscus tear, BMI of overweight to obese, medial compartment chondral lesion, and length of follow-up) were relatively high for sensitivity (60.2%) and specificity (70.4). This algorithm can help guide clinicians to determine treatment courses that would optimize the likelihood of a positive outcome. In this case the outcome of interest is the reduced progression or prevention of knee joint OA. These factors in combination with clinical expertise and an individualized healthcare plan provide the best option for sustained recovery. Aspects of the regression study I found intriguing were the behavioral factors measured, i.e. activity level, smoking, and BMI. Even though smoking and activity level did not make the list of factors that have the highest percentage of predicting OA following ACL reconstruction, they are still important factors in the recovery process. Along with BMI these modifiable factors should be addressed as aggressive as strength and motion, however these factors tend to be the lowest in compliance.
Brandon, I think you worded it perfectly with "these factors in combination with clinical expertise and an individualized healthcare plan". I agree with the study authors that the model is not sensitive or specific enough to deploy in the clinical setting. There are probably other variables that we can detect with our "clinical expertise" (e.g., neuromuscular control) that we need to incorporate in the models. There may also be genetic components that influence the resilience of the tissues or the amount of inflammatory/catabolic response the tissues are exposed to after an injury.
I agree that we need to emphasize a long-term healthy lifestyle in our injured patients to help reduce the risk of knee osteoarthritis. The study supports maintaining a healthy body weight and previous systematic reviews have shown that physical activity may be beneficial to the joint. There are also a lot of very basic questions that we need answers to including how to determine the optimal time to return to play. Over the next few years I think we'll be seeing some very interesting work exploring the early onset of osteoarthritis after joint trauma. There are a lot of groups starting to examine this using various types of analyses (e.g., biomechanics, biochemical, imaging).
Nice post there is. Osteoarthritis knee is a common problem of old age but it could be treated by surgeries and alternative treatments as Yoga , acupuncture also.