Endovascular versus open repair of ruptured descending thoracic aortic aneurysms: A nationwide risk-adjusted study of 923 patients

Objective: Recent studies support the use of endovascular treatment for ruptured abdominal aortic aneurysms, but few studies have examined the use of thoracic endovascular aortic repair (TEVAR) for ruptured descending thoracic aortic aneurysm. We evaluated nationwide data regarding short-term outcomes of TEVAR and open aortic repair (OAR) for ruptured descending thoracic aortic aneurysm. Methods: From US Nationwide Inpatient Sample data, we identified 923 patients who underwent ruptured descending thoracic aortic aneurysm repair in 2006-2008 and who had no concomitant aortic disorders. Of these patients, 364 (39.4%) underwent TEVAR and 559 (60.6%) underwent OAR. Multivariable regression was used to assess the effect of TEVAR versus OAR after adjusting for potential confounding factors. Outcomes assessed were in-hospital mortality, complications, failure to rescue (defined as the mortality among patients in whom a complication develops), and disposition. Backward stepwise logistic regression was used to identify independent predictors of outcomes for each approach. Results: Patients undergoing TEVAR were older (72 +/- 12 years vs 65 +/- 15 years; P < .001) and had a higher Deyo comorbidity index (4.19 +/- 1.79 vs 3.14 +/- 2.05; P < .001) than patients undergoing OAR. Unadjusted mortality was 23.4% (85/364) for TEVAR and 28.6% (160/559) for OAR. After risk adjustment, the odds of mortality, complications, and failure to rescue were similar for TEVAR and OAR (P > .1 for all), but patients undergoing TEVAR had a greater chance of routine discharge (odds ratio [OR] = 3.3; P < .001). An interaction was identified that linked hospital size and operative approach with risk of complications (P < .001). In smaller hospitals, TEVAR was associated with lower complication rates than OAR (OR = 0.21; P < .05). Regression analysis revealed that smaller hospital size predicted significantly higher rates of mortality (OR = 2.4; P < .05), complications (OR = 4.0; P < .005), and failure to rescue (OR = 51.12; P < .001) in those undergoing OAR but not in those undergoing TEVAR. Preexisting renal disorders substantially increased mortality risk (OR = 10.81; P < .001) and failure to rescue (OR = 309.54; P < .001) in patients undergoing TEVAR. Conclusions: Nationwide data for ruptured descending thoracic aortic aneurysm reveal equivalent mortality, complication rates, and failure to rescue for TEVAR and OAR but more frequent routine discharge with TEVAR. Unlike OAR outcomes, TEVAR outcomes were not poorer in smaller hospitals, where TEVAR produced fewer complications than OAR. Therefore, TEVAR may be an ideal alternative to OAR for ruptured descending thoracic aortic aneurysm, particularly in small hospitals where expertise in OAR may be lacking and immediate transfer to a higher echelon of care may not be feasible. (J Thorac Cardiovasc Surg 2011;142:1010-8)