This study was undertaken to identify factors that lead to improvements in the results of endovascular aneurysm repair, with particular focus on new endograft design.

We analyzed data for patients enrolled in the European Collaborators on Stent Graft Techniques for Abdominal Aortic Aneurysm Repair (EUROSTAR) registry, and compared those for endografts now withdrawn from the market with those for endografts currently in use. Patients in whom a variety of endograft types were used in small numbers were excluded. Postoperative and long-term outcomes were initially compared with univariate analyses, and subsequently multivariate tests were used to adjust for baseline differences between the 2 groups. The main outcome measures were freedom from a variety of secondary interventions, aneurysm rupture, and death.

Some 1224 patients received “withdrawn” endografts, and 2768 patients received “current” endografts. The 2 groups were generally similar, but patients with current devices were more often men, significantly older, more frequently unfit for open surgery, and had larger aneurysms with wider necks. Of no surprise, current endografts were also more often used by experienced (>60 previous cases) surgical teams (44% vs 20%; P < .0001). Thirty-day clinical outcomes were comparable in the 2 groups, although patients with withdrawn devices were less likely to have type II endoleak (9.2% vs 5.5%; P < .0001), and those with current devices had a shorter mean hospital stay (5.4 vs 6.8 days; P < .0001). At 3 years more patients with current devices were free from secondary transfemoral intervention (88.4% vs 76%; P < .0001) and conversion to open repair (95.4% vs 93.4%; P = .007). Aneurysm-related mortality at 3 years, defined as death due to aneurysm rupture or within 30 days of a secondary intervention, was also less frequent with current endografts (2.7% vs 4.4%; P = .02). Aneurysm rupture at 3 years was infrequent (0.8% vs 1.8%; P = .07). At multivariate analysis the use of current devices was a protective factor against late conversion to open repair (hazard ratio, 0.49; 95% confidence interval, 0.28-0.86; P = .014) and aneurysm-related death (hazard ratio, 0.51, 95% confidence interval, 0.34-0.75; P = .0008). Larger aneurysm or neck diameter and shorter neck length were also associated with late conversion to open repair; larger aneurysm diameter, older age, and unfitness for open surgery were predictive of aneurysm-related death.

Modern endograft design has improved the results of endovascular aneurysm repair.