Comparative effectiveness of the treatments for thoracic aortic transection [corrected].

ABSTRACT

OBJECTIVES: To synthesize the available evidence regarding the outcomes associated with nonoperative management, open repair, and endovascular repair of thoracic aortic transection. METHODS: We searched electronic databases (MEDLINE, EMBASE Cochrane, Web of Science, and Scopus) for studies that enrolled patients with aortic transection and measured the outcomes of interest. Two reviewers determined study eligibility and extracted data. We estimated the event rate associated with the different approaches from case series and the relative risk from comparative studies. Estimates from each study were pooled using the random effects model. RESULTS: We found 139 studies that fulfilled the inclusion criteria, the majority of which were noncomparative surgical case series, retrospective, and none were randomized. Studies included 7768 patients, the majority of which were males. The mortality rate was significantly lower in patients who underwent endovascular repair, followed by open repair and nonoperative management (9%, 19%, and 46%, respectively, P < .01). No significant difference in event rate across the three groups was noted for the outcomes of anterior stroke, posterior stroke, or any stroke. The risk of spinal cord ischemia and end-stage renal disease were higher in open repair compared with the other 2 groups (9% vs 3% and 3%, P = .01 for spinal cord ischemia and 8% vs 5% and 3%, P = .01 for end-stage renal disease). Compared with endovascular repair, open repair was associated with an increased risk of graft infection and systemic infections. Meta-analyses of comparative studies demonstrated that compared with open repair, endovascular repair is associated with reduced mortality and spinal cord ischemia (relative risk, 0.61; 95% confidence interval, 0.46-0.80; and relative risk, 0.34; 95% confidence interval, 0.16-0.74; respectively). Inferences are limited by methodological quality, survival, and publication biases. CONCLUSIONS: Very low-quality evidence suggests that, compared with open repair or nonoperative management, endovascular repair of thoracic aortic transection is associated with better survival and decreased risk of spinal cord ischemia, renal injury, and graft and systemic infections. Nonoperative management is associated with the least favorable outcomes.