Abstract
OBJECTIVES: High-level basketball athletes at the collegiate level and in the National Basketball Association (NBA) have a greater proportion of knee cartilage pathology than the non-athlete population. However, little is known as to whether identified pathology on knee magnetic resonance imaging (MRI) correlates with clinical symptoms or orthopaedic history in these players. The purposes of the study were: 1) To evaluate knee MRIs in a cross-sectional population of NBA Draft Combine players to establish prevalence of knee pathology, including that of articular cartilage and meniscus in professional basketball players, and 2) To identify independent variables including demographics, playing history, prior knee injury, and symptoms that correlate to these MRI findings. METHODS: Players from the NBA Draft Combine over a two-year period voluntarily participated by undergoing bilateral knee MRIs and completing confidential, validated clinical questionnaires. MRIs were evaluated blindly by two independent radiologists for cartilage assessment (using Modified Noyes score) and meniscal pathology. The Modified Noyes score attributes a score from 0 to 10 to assess chondral pathology ranging from mild chondromalacia (score of 0) to complete exposure of the subchondral bone in a lesion > 15 mm (score of 10). Separate scores were assigned to the patella, trochlea, medial femoral condyle (MFC), medial tibial plateau (MTP), lateral femoral condyle (LFC), and lateral tibial plateau (LTP). Modified Noyes Scores and MRI findings were summarized using descriptive statistics. Intraclass correlation coefficients were determined for the radiologists’ evaluation of articular cartilage scoring. Furthermore, imaging scores were correlated to VAS pain and IKDC questionnaires using multivariate linear regression analyses to determine the association between clinical symptoms and imaging findings. Predictors of absolute articular cartilage pathology (i.e. a Modified Noyes score > 0) were determined using multivariate logistic regression analyses. All statistical analyses were performed with R (RStudio Version 2023.09.1+494). An a priori alpha level of 0.05 was utilized for statistical significance. RESULTS: Forty-three players (80 knees) were included. Intraclass correlation coefficients averaged 0.827 for agreement and 0.831 for consistency for the Modified Noyes score. Cartilage pathology was identified in 48.3% of knees, with the most common locations being the patella and trochlea (Table 1). On linear regression analysis of pain and function scores versus Modified Noyes scores, there was no significant association between cartilage pathology and clinical symptoms (Table 2). Pain and function scores did not correlate to Modified Noyes scores in any region of the knee. On logistic regression analysis, there was minimal predictivity of pathology by pain and function scores or a prior surgical history of the knee (Table 3). Reduced function scores were only predictive of cartilage pathology on the lateral femoral condyle (OR 0.55, p = 0.03). Increased pain scores were predictive of non-zero total Modified Noyes score (OR 6.11, p < 0.01). Additionally, a prior orthopaedic surgical history of the knee was predictive of cartilage pathology on the lateral tibial plateau (OR 6.58, p = 0.04). The overall incidence of meniscal pathology was 10.0% (Table 4). There was an equal tear prevalence in the medial meniscus (n = 4, 5.0%) and the lateral meniscus (n = 4, 5.0%). No players with ongoing medial meniscal tears had concomitant medial compartment cartilage defects. Players with ongoing lateral meniscal tears had a 75% (n = 3) incidence of concomitant LFC pathology and 50% (n = 2) incidence of concomitant LTP pathology. CONCLUSIONS: The incidence of pathology identified on knee MRIs in players at the NBA Draft Combine without clinical symptoms is high. There is little correlation between the objective imaging findings and subjective pain, function, and orthopaedic injury history. Clinicians treating high-level athletes should focus on treating clinically significant findings and be cautious about treating pathology found on imaging that may not explain or correlate with symptoms. It remains critical to study the natural history of imaging findings in these athletes to better ascertain the ultimate impact of sport and physiologic load on disease progression.