Thoracic endovascular aortic repair (TEVAR) is now the treatment of choice for most diseases of the thoracic aorta. Current evidence and guidelines for practice are based on pooled analysis of data that are of variable quality. Questions remain about the applicability of TEVAR to different pathologies, peri-procedural neurological complications, the subsequent durability of the procedure and mid-term survival of patients. Methods and Results A systematic review of the literature was performed to describe the incidence of peri-operative and mid-term adverse events following TEVAR. For the main investigation, data from five phase IV trials was combined with that from St George's Vascular Institute to form a composite dataset of 1010 patients that underwent TEVAR from 2002-2010. A high rate of mid-term all cause mortality was observed in patients treated for thoracic aortic aneurysm, with less than 60% of those treated living beyond 5 years. Factors predicting neurological complications were determined using logistic regression modelling. Coverage of the left subclavian artery was the most important predictor of post-operative stroke, specifically in posterior vascular territories. Patients undergoing elective repair of thoracic aortic aneurysms were stratified into those at high, intermediate and low risk of a death wi!hin five years using regression models based on pre-operative factors. Those at high and low risk of aortic re-intervention were similarly identified. The models were validated externally using a dataset obtained from the United States. Conclusions Post-operative outcomes following TEVAR are partly dependent on presenting pathology, especially with regards to mid-term all-cause death. Coverage of the left subclavian artery is one of the few potentailly modifiable risk factors for stroke. Patients can risk-stratified according to the risk of midterm all cause death following TEVAR, and careful consideration should be given before treating patients who may not benefit in terms of overall life expectency.