Femoral Tunnel Malposition, Increased Lateral Tibial Slope, and Decreased Notch Width Index Are Risk Factors for Non-Traumatic Anterior Cruciate Ligament Reconstruction Failure

Purpose To identify risk factors for patients who sustain nontraumatic anterior cruciate ligament reconstruction (ACLR) failure. Methods A retrospective analysis was performed on patients undergoing primary or revision ACLR in our institution between 2010 and 2018. Patients sustaining insidious-onset knee instability without history of trauma were identified as nontraumatic ACLR failure and assigned to the study group. The control group of subjects who showed no evidence of ACLR failure with minimum 48-month follow-up were matched in a 1:1 ratio based on age, sex, and body mass index. Anatomic parameters including tibial slope (lateral [LTS], medial [MTS]); tibial plateau subluxation (lateral [LTPsublx], medial [MTPsublx]); notch width index (NWI); and lateral femoral condyle ratio were measured with magnetic resonance imaging or radiography. Graft tunnel position was assessed using 3-dimensional computed tomography and reported in 4 dimensions: deep-shallow ratio (DS ratio) and high-low ratio for femoral tunnel, anterior-posterior ratio and medial-lateral ratio for tibial tunnel. Interobserver and intraobserver reliability were evaluated by the intraclass correlation coefficient (ICC). Patients' demographic data, surgical factors, anatomic parameters, and tunnel placements were compared between the groups. Multivariate logistic regression and receiver operating characteristic curve analysis was used to discriminate and assess the identified risk factors. Results A total of 52 patients who sustained nontraumatic ACLR failure were included and matched with 52 control subjects. Compared to patients with intact ACLR, those who sustained nontraumatic ACLR failure showed significantly increased LTS, LTPsublx, MTS, and deceased NWI (all P < .001). Moreover, the average tunnel position in the study group was significantly more anterior (P < .001) and superior (P = .014) at the femoral side and more lateral (P = .002) at the tibial side. Multivariate regression analysis identified LTS (odds ratio [OR] = 1.313; P = .028), DS ratio (OR = 1.091; P = .002), and NWI (OR = 0.813; P = .040) as independent predictors of nontraumatic ACLR failure. LTS appeared to be the best independent predictive factor (area under the curve [AUC] = 0.804; 95% confidence interval [CI], 0.721-0.887), followed by DS ratio (AUC = 0.803; 95% CI, 0.717-0.890), and NWI (AUC = 0.756; 95% CI, 0.664–0.847). The optimal cutoff values were 6.7° for increased LTS (sensitivity = 0.615, specificity = 0.923); 37.4% for increased DS ratio (sensitivity = 0.673, specificity = 0.885); and 26.4% for decreased NWI (sensitivity = 0.827, specificity = 0.596). Intraobserver and interobserver reliability was good to excellent, with ICCs ranging from 0.754 to 0.938 for all radiographical measurements. Conclusions Increased LTS, decreased NWI, and femoral tunnel malposition are predictive risk factors for nontraumatic ACLR failure. Level of Evidence Level III, retrospective comparative study. To identify risk factors for patients who sustain nontraumatic anterior cruciate ligament reconstruction (ACLR) failure. A retrospective analysis was performed on patients undergoing primary or revision ACLR in our institution between 2010 and 2018. Patients sustaining insidious-onset knee instability without history of trauma were identified as nontraumatic ACLR failure and assigned to the study group. The control group of subjects who showed no evidence of ACLR failure with minimum 48-month follow-up were matched in a 1:1 ratio based on age, sex, and body mass index. Anatomic parameters including tibial slope (lateral [LTS], medial [MTS]); tibial plateau subluxation (lateral [LTPsublx], medial [MTPsublx]); notch width index (NWI); and lateral femoral condyle ratio were measured with magnetic resonance imaging or radiography. Graft tunnel position was assessed using 3-dimensional computed tomography and reported in 4 dimensions: deep-shallow ratio (DS ratio) and high-low ratio for femoral tunnel, anterior-posterior ratio and medial-lateral ratio for tibial tunnel. Interobserver and intraobserver reliability were evaluated by the intraclass correlation coefficient (ICC). Patients' demographic data, surgical factors, anatomic parameters, and tunnel placements were compared between the groups. Multivariate logistic regression and receiver operating characteristic curve analysis was used to discriminate and assess the identified risk factors. A total of 52 patients who sustained nontraumatic ACLR failure were included and matched with 52 control subjects. Compared to patients with intact ACLR, those who sustained nontraumatic ACLR failure showed significantly increased LTS, LTPsublx, MTS, and deceased NWI (all P < .001). Moreover, the average tunnel position in the study group was significantly more anterior (P < .001) and superior (P = .014) at the femoral side and more lateral (P = .002) at the tibial side. Multivariate regression analysis identified LTS (odds ratio [OR] = 1.313; P = .028), DS ratio (OR = 1.091; P = .002), and NWI (OR = 0.813; P = .040) as independent predictors of nontraumatic ACLR failure. LTS appeared to be the best independent predictive factor (area under the curve [AUC] = 0.804; 95% confidence interval [CI], 0.721-0.887), followed by DS ratio (AUC = 0.803; 95% CI, 0.717-0.890), and NWI (AUC = 0.756; 95% CI, 0.664–0.847). The optimal cutoff values were 6.7° for increased LTS (sensitivity = 0.615, specificity = 0.923); 37.4% for increased DS ratio (sensitivity = 0.673, specificity = 0.885); and 26.4% for decreased NWI (sensitivity = 0.827, specificity = 0.596). Intraobserver and interobserver reliability was good to excellent, with ICCs ranging from 0.754 to 0.938 for all radiographical measurements. Increased LTS, decreased NWI, and femoral tunnel malposition are predictive risk factors for nontraumatic ACLR failure.