A Biomechanically Based Acute Thoracic Aortic Dissection Potential Index

Acute thoracic aortic dissection, a separation of the layers of the aortic wall, is a serious med- ical condition with high mortality. Current clinical practice is a diameter based decision for elective surgery for patients with an orthogonal aortic diameter in excess of 5.5 cm. However, the utility of this metric is limited by the high number of patients experiencing dissection at diameters below this threshold. There thus exists a need for an improved metric for assessing dissection risk. As the dissection event ultimately involves biomechanical failure of the aortic wall, this thesis proposes a biomechanically based dissection potential index derived from both full aorta stress maps and microstructural failure behavior of the tissue. A structural model of the aortic wall, including the repeating lamellar structure of the aortic media was developed. This model was used to determine the aortic tissue biomechanical behavior in both the pre-failure and failure regimes. These model predictions were validated against experimentally determined uniaxial pre-failure elastic response as well as ultimate stress of the vessel wall for three different patient cohorts. The model was next utilized to quantify physiologic strength of the vessel wall subjected to the entire range of physiologically plau- sible biaxial loading conditions. Interestingly, we discovered that the vessel wall physiologic strength is not statistically different than its uniaxial longitudinal strength. We also found that the biomechanical integrity of the aorta is limited by the longitudinal strength of the aortic wall tissue in regions of high stress biaxiality, defined by the ratio of longitudinal and circumferential stress. We constructed a dissection risk metric, termed dissection potential index (DPI), from the ratio of the longitudinal stress to the longitudinal strength of the vessel wall. Using patient-specific aorta geometry and aortic stiffness, both determined from clinical CT images of the aorta, the stress map of the aortic wall was evaluated. The dissection potential index (DPI) for these patients was determined by interrogating these maps against cohort-specific longitudinal strength of the tissue. We found that DPI for dissected patients was significantly higher than for an age and BMI matched control cohort. We also observed an increase of the DPI for dissected patients over time. In addition, the model correctly predicted dissection initiation location marked by the surgeons, suggesting that the model captured essential aortic wall failure mechanisms. Surprisingly, the DPI for dissected patients did not correlate with their maximum aortic diameter, demonstrating the limitations of the current diameter based criterion as a predictive risk metric. Further studies with larger cohort of patients will lead to the establishment of DPI as an evidence based clinical metric for early management of dissection potential