Life expectancy after endovascular versus open abdominal aortic aneurysm repair: Results of a decision analysis model on the basis of data from EUROSTAR

Abstract

Background/Objectives: Although endovascular abdominal aortic aneurysm (AAA) repair (ENDO) has decreased operative morbidity risks compared with open AAA repair (OPEN), risks of rupture and reintervention are higher after ENDO. We used decision analysis to examine the effect of these competing risks on quality-adjusted life expectancy (QALE) after ENDO and OPEN. Methods: We used a Markov decision-analysis model to simulate hypothetic cohorts of patients undergoing ENDO or OPEN. Patients moved through a multistate transition model according to probabilities derived from the literature, the EUROSTAR database (for ENDO) and Medicare claims data (for OPEN). Our primary outcome measure was QALE after surgery. We used sensitivity analysis to determine which factors most influenced this outcome. Results: In the base-case analysis of 70-year-old men, life expectancy after ENDO was 7.09 quality-adjusted life years compared with 7.03 quality-adjusted life years for OPEN, a difference of 3 weeks. Sensitivity analysis showed that at less than age 64 years, OPEN results in greater QALE. However, the difference in QALE was small (≤3 months) across the entire range of ages studied (60 to 85 years). The optimal strategy was sensitive to changes in ENDO and OPEN operative mortality rate, rupture rate after ENDO, late conversion to OPEN rate, ENDO revision rate, and OPEN reoperation rate. However, the difference between OPEN and ENDO strategies was small across the plausible range of most of these variables. Conclusion: For most patients who are candidates for AAA repair, ENDO and OPEN result in similar QALE. Decision analysis suggests that OPEN may be preferred for younger patients with low operative risk and ENDO may be preferred for older patients with higher operative risk. However, given the similarity in overall outcome, patient preference should be weighed heavily in decision making.