PREDICTING ANTERIOR CRUCIATE LIGAMENT REINJURY VIA MRI USING POSTEROLATERAL TIBIAL PLATEAU GEOMETRY

Anterior cruciate ligament (ACL) injuries frequently result from non-contact mechanisms, involving valgus force and internal tibial rotation.1 Following ACL injury, the posterior tibial plateau is critical in stabilizing knee biomechanics by limiting anterior tibial translation.2 Prior research has shown a strong association between increased tibial slope and ACL injury, using geometric measurements such as the lateral meniscus height and the cartilaginous tibial plateau slope.3–4 However, little is known regarding their implication in cases of ACL reinjury. This study aims to compare posterolateral tibial plateau geometry in patients with ACL reconstruction failure against healthy controls to help identify risk factors for ACL reinjury. This retrospective case-control study mathematically analyzed the posterolateral tibial slope in ACL-reinjured knees and matched control knees. Magnetic resonance imaging scans were used to plot ten points along the posterolateral tibial plateau cartilage, and their coordinates were fitted to a power function (y = a*xn) using a Python code. Lateral meniscal height and lateral anteroposterior (AP) length were further measured. Data was compared using the Mann-Whitney U test, and interobserver reliability was assessed with the interclass correlation coefficient (ICC) and 95% confidence intervals. A p-value of < 0.05 was deemed statistically significant. The study included 30 ACL-reinjured knees and 28 control knees. Overall, ACL-reinjured knees had larger power function coefficients than matched controls: coefficient a (5.94 vs. 4.97, p = 0.042) and coefficient n (0.47 vs. 0.46, p = 0.610). Lateral meniscal height (LMH) was significantly lower in the ACL-reinjured group (0.69 vs. 0.77, p = 0.003), whereas lateral AP length was slightly larger (3.62 vs. 3.50, p = 0.219). Moreover, ICCs showed good to excellent interobserver reliability: coefficient a (0.91, 95% CI 0.86–0.95), coefficient n (0.88, 95% CI 0.80–0.93), LMH (0.91, 95% CI 0.85–0.95), and lateral AP length (0.96, 95% CI 0.93–0.97). Patients with ACL reconstruction failure exhibited significant differences in posterolateral tibial geometry with strong interrater reliability, suggesting that these mathematical models may serve as valuable risk stratification tools. Hence, their practical implementation may help identify high-risk patients and optimize surgical planning to improve postoperative outcomes.