Safety of Thoracic Aortic Surgery in the Present Era

Background: Advances in graft materials, hemostasis, and surgical techniques have facilitated surgery on the thoracic aorta. We investigate the current safety level of these operations-for the purposes of enabling risk/benefit decisions for surgery and also to serve as a benchmark for comparison with emerging endovascular approaches. Methods: Five hundred six consecutive patients (315 male, 191 female; aged 14 to 91 years [mean, 61]) underwent surgery on the thoracic aorta at one institution from 1995 to 2004. In all, 360 operations involved the ascending and arch (71.1%) and 130 (25.7%) involved the descending or thoracoabdominal aorta, or both, and 16 (3.2%) were classified as miscellaneous aortic operations. Clinical data collected prospectively were analyzed retrospectively using χ² and multivariable logistic regression statistics for the outcomes reoperation for bleeding, perioperative (hospital or 30-day) mortality and stroke. Midterm survival was assessed by Kaplan-Meier methodology. Results: Mortality for elective operations on the ascending/arch was 3.0%; mortality for elective operations on the descending aorta was 2.9%. Mortality for elective thoracoabdominal operations was 11.9%. Mortality for all operations was 8.6%. Probability of stroke was 3.0% for ascending/arch, 4.2% for descending, and 2.1% for thoracoabdominal operations. The paraplegia rate was 7.3% for all descending and thoracoabdominal operations. Age and emergency operation predicted increased risk of death, stroke, and reoperation for bleeding. For young patients (less than 55 years old) having elective ascending/arch operations, freedom from permanent complications of operation (death, stroke, paraplegia) was 98%. Overall survival at 1, 3, and 5 years was 84.7%, 78.3%, and 72.5%, respectively. Conclusions: Aortic surgery is quite safe in the current era and leads to good long-term survival for this patient group. These data support prophylactic replacement of the thoracic aorta in patients with poor expected natural history (based on aneurysm size or symptoms). As catheter-based therapies proliferate, surgical data provide a benchmark that must be equaled or exceeded by newer endovascular approaches.