Good morning, good afternoon, or good evening, everyone. My name is Mathias Piper, and I'm an orthopedic surgery resident in the University Hospital of Ghent in Belgium, and today I'm going to present the Pipeline for Automatic Ligamentous Injury Prediction in Seismotic Lesions Based on Geometrical Morphometric Analysis. First of all, I have no disclosures. As a general introduction, the ankle joint is passively stabilized by different ligaments, whereas the ankle stenosmosis is built up by major ligament complexes, which stabilize the distal tibia and fibula to form a congruent mortise around the tibias. In the ankle, this congruence is of key importance for optimal load transfer on the cartilage, as well as stability of the ankle. Ankle sprains occur in up to 5% of all emergency department visits, while it has been shown that 24 of these ankle sprains include an injury to the ankle stenosmosis. Atomechanically, the external rotation movement of the talus directly pushes the distal fibula into supraphysiological external rotation and diastasis, sequentially rupturing the anterior inferior tibiofibular ligament, the interosseous ligament, and the posterior inferior tibiofibular ligament. Unfortunately, several drawbacks are encountered when using current imaging modalities. Firstly, radiographic measurements have been shown to be unreliable and highly dependent on the rotation of the XG. Conventional CT allows for three-dimensional investigation, but lacks the weight-bearing nature of the ankle. Thirdly, MRI allows for investigation of specific ligament injury, but also lacks dynamic information. Diagnosis of syndesmotic injuries could therefore greatly benefit from a weight-bearing imaging modality with inclusion of ligamentous information. The purpose of this study was therefore to construct a statistical shape model of the distal tibiofibular joint with inclusion of ligamentous morphometry, and subsequently, this model will be applied to the clinical entity of traumatic syndesmotic lesions by evaluating patterns of ligament injury using weight-bearing models. For the construction of the ligamentous shape model and the first MMR study, we started off with a previously validated shape model of the ankle, and additionally, cartilage and ligament insertions were segmented on 60 MRI patients of healthy patients. By combining these two input models, insertion sites of the cartilage and ligaments could be transferred to the statistical shape model, after which the cartilage and ligaments could be modeled. For our second aim, we included the distal tibiofibular joint anatomy of patients with syndesmotic ankle sprains, which was compared on arthroscopy. To investigate instability of the distal tibiofibular joint, as well as differences in length of the ligament parts, these patients were compared to our healthy cohort. After fitting of the statistical shape models, ligaments would be modeled on these input geometries. As a validation on MRI, the mean distance error of the predicted anterior inferior tibiofibular ligament compared to the actual ligament was 0.354 mm, whereas the mean distance error of the posterior inferior tibiofibular ligament was 0.063 mm. As a result, the mean length of the anterior inferior tibiofibular ligament was 12.23 mm for the control cases, in comparison to 14 mm for the cases with a syndesmotic injury. Notably significant differences were found between these two measurements, but no statistical significant differences were found between the posterior syndesmotic length, as well as for the interest of distance. As a result, a typical anterior open book lesion was observed in patients with a syndesmotic injury with an increased predicted length of the anterior inferior tibiofibular ligament corresponding to an anterior inferior tibiofibular ligament elongation injury or rupture. As a conclusion, an elongation injury of the AITFL was found in traumatic high-angle sprains, resulting in an external rotation of the distal fibula. Due to the intact PITFL, no diastasis was observed. While it has been shown that up to 5 feet 2 of the syndesmotic joints were malreduced after ankle fractures, our study could identify specific ligament patterns which causes this malreduction and help intraoperatively address these injuries. Future studies will include the prediction of contact stress in the ankle in healthy and syndesmotically malreduced ankles to show patterns of increased contact stress, which could lead to post-traumatic OA in the short or long term. Thank you very much for listening and have a great AOFAS conference. Bye.

ligamentous injuries

ankle joint

ankle sprains

ligamentous morphometry

post-traumatic osteoarthritis