SRS345 - Digital twin modelling of spinal cord blood flow to predict paraplegia after aortic aneurysm repair

Abstract Aims Spinal cord ischaemia (SCI) and paraplegia may follow aortic aneurysm (AA) repair due to disruption of spinal cord blood flow. Current predictive tools are unreliable, limiting perioperative risk assessment. We aimed to develop an in-silico ‘digital twin’ of the aorta and spinal cord branches to characterise haemodynamic changes after AA repair. Methods CT angiograms obtained before and after endovascular thoracoabdominal aneurysm repair were reconstructed in SimVascular. Pulsatile flow simulations used patient-specific inlet and three-element Windkessel outlet boundary conditions. Haemodynamic changes after stent coverage of intercostal and lumbar arteries were evaluated with further assessment of surface metrics: time-averaged-wall-shear-stress (TAWSS), oscillatory-shear-index (OSI), relative-residence-time (RRT), and endothelial-cell-activation-potential (ECAP). Results Two patients with type IV thoracoabdominal AA repairs using non-customised 4-branched devices were modelled: one remained well, the other developed paraplegia. In the uncomplicated case, spinal cord flow fell 51.9% after exclusion of 19 branches, with increased TAWSS (+5.2%), reduced RRT/ECAP, and minimal OSI change. In the paraplegic patient, flow fell 66.1% after exclusion of 15 branches. The uncomplicated patient showed redistribution of flow away from the spine, with modest increases in leg (+6.1%), reno-visceral (+5.9%), and supra-aortic (+6.0%) vessels. Visceral arteries had the highest TAWSS and lowest RRT/ECAP, while leg arteries had the lowest TAWSS and highest RRT/ECAP. Conclusions This proof-of-principle study demonstrates the feasibility of predicting spinal cord haemodynamics after aortic repair using open-source software and routine imaging. Expansion to larger cohorts could yield a clinical decision-support tool to stratify SCI risk and address this unmet need.