Thank you. I'd like to thank my co-authors and the AOSSM and Anna for allowing us to present this today. So our disclosures are available on the website. So we know from Dr. Grickids and Dr. Volker's talk that ACL graft failures continue to occur at a high rate, especially in young patients returning to high-risk activities. We know from the MARS group that these are often multifactorial and there has been an increased focus on anatomic factors such as the sagittal inclination of the tibial plateau or posterior tibial slope. We know from biomechanical studies that an increased tibial slope is associated with increased anterior tibial translation as well as strain in the ACL. There's also been clinical studies that have shown elevated tibial slope both in ACL primary injuries, ACL graft failures and in multiple ACL graft failures. And Dr. DuJour and Sonari Kote both described their techniques for slope-reducing osteotomies. The purpose of our study was to compare posterior tibial slope on both lateral knee radiographs and MRI in patients who experience revision ACL graft failure versus age, sex, and graft match controls and identify cutoff values for posterior tibial slope measurements that predict the risk of graft failure. So we hypothesized that patients who experience revision ACL graft failure would exhibit an elevated or steeper posterior tibial slope compared to patients with successful revision ACL on both x-rays and MRI. And a posterior tibial slope greater than 12 degrees on lateral knee radiographs would be a predictive risk factor for revision ACL failure. So this was a retrospective review of a revision ACL database. Patients were included who had graft failure which was defined as re-revision ACL or clinical exam and MRI evidence of a re-tear. They needed to have acceptable lateral knee radiographs and an MRI available for measurements. And we excluded any patients who had a multiligament injury at any point in their primary or revision ACL or previous tibial fracture or previous knee infection. These were then matched one-to-one by age within five years, sex, and revision graft type to control patients with a minimum two years of follow-up. So slope measurements were made by a previously described technique. So we did five centimeters below the joint line in the lowest visualized portion to define the tibia anatomic axis and an angle formed with the tangent of the medial and lateral plateaus. MRI measurements were done by the Hashemi method shown here. So our statistical analysis, a power analysis was performed to detect a three degree difference on x-ray with a standard deviation of three degrees and we would need 17 patients per group to achieve adequate power. Optimal cutoff values were determined for predicting increased risk of revision ACL failure were determined by the ROC and a UDIN index was calculated to select the cutoff values with the highest sensitivity and specificity for each measurement. And then a conditional multivariable logistic regression was performed to assess the relative contributions of these cutoff values as predictors of revision failure. These were our results. There were no differences between the groups except for the time to re-tear versus follow-up and we had adequate length of follow-up within our control group of six years. All measurements were significantly different so we had higher tibial slope on both MRI and x-ray based on all measurements. Our ROC curves demonstrated they were moderately informative based off the AUC and these cutoff values were selected again to maximize the UDIN index. So this was the results of our conditional multivariable logistic regression and an odds ratio of 18.7 was found for a medial tibial slope greater than 14 degrees. So our radiographic medial posterior tibial slope had an 18 times greater odds of revision ACL failure. 19 of 22 patients or 86% with an elevated tibial slope greater than 14 degrees on the x-ray failed. So this was 92.1% specificity which is important when you're considering a slope reducing osteotomy in patients. You want to limit the number of patients who would get an unnecessary osteotomy. The sensitivity was lower which can be expected because patients with a lower slope can still experience graft failure as we know for multiple reasons. Limitations of our study was this was a retrospective study design, unable to control for other potential factors, confounding factors for ACL graft failure such as the pre-injury activity level, ACL tunnel position and the amount of meniscus removed at meniscectomy and we did not have objective clinical or arthrometric follow-up measurements at final follow-up. Though our control patients did have a pretty long length of follow-up at six years, again our failures were at an average of four years after the revision ACL and our measurements of posterior tibial slope. Unfortunately these measurements can be very highly dependent on the image rotation and observer experience and the MRI can be difficult when defining the anatomic tibial axis. So elevated tibial slope is a significant risk factor for revision ACL graft failure. An 18.7 times increased risk was a medial posterior tibial slope greater than 14 degrees and as Dr. Musall talked about earlier, should we be considering this in the revision setting and treating this as our other modifiable risk factors in a patient with a varus knee or medial meniscus deficiency where we're doing an HTO or meniscus transplants. Though more prospective studies are needed to determine what the best options are. Thank you.

posterior tibial slope

ACL graft failure

knee radiographs

MRI

revision ACL failure