Abstract
BACKGROUND: To investigate the cost-effectiveness of endovascular aortic repair (EVAR) versus open aortic repair (OAR) for abdominal aortic aneurysm (AAA) using incremental costs per decreased in-hospital mortality rate gained through our patients' cohort. METHODS: Medical records and healthcare costs of patients with AAA hospitalized between 2010 and 2015 were extracted from the National Health Insurance Research Database (NHIRD) of Taiwan. Multiple regression analysis was applied to adjust for confounding factors and to compare the differences in postoperative clinical outcomes between patients who received EVAR and OAR. The incremental cost-effectiveness ratio (ICER) of EVAR was determined based on the healthcare cost obtained from the analyzed data. RESULTS: A total of 2803 AAA patients were identified (n = 559 with ruptured AAA and n = 2244 unruptured AAA). Patients with ruptured AAA who underwent EVAR compared with OAR patients had shorter hospital and intensive care unit (ICU) stays (all p < 0.05). For patients with unruptured AAA, those who received EVAR compared with OAR, the adjusted odds ratio (aOR) of postoperative complications and in-hospital mortality were 0.371 and 0.447 (all p < 0.05). The total direct surgical costs and medical expenses during hospitalization in all AAA patients were higher for the EVAR group; however, ICER was <1 per capita gross domestic product. Stratification by age groups further suggested that ICER for patients with unruptured AAA who received EVAR, compared with OAR, decreased with age. CONCLUSION: Total direct medical costs were higher for AAA patients receiving EVAR regardless of rupture status; however, the cost is offset by lower odds of postoperative complications and in-hospital mortality. The observed decrease in ICER with age and EVAR use warrants further analysis. Our findings further validate the use of EVAR over OAR. These results provides supporting evidence for physicians and patients with AAA to inform shared decision making regarding endovascular or OAR options.