Differences in patient selection and outcomes based on abdominal aortic aneurysm diameter thresholds in the Vascular Quality Initiative

Abstract

ObjectiveRandomized trials have shown no benefit for repair of small abdominal aortic aneurysms (AAAs), although repair of small AAAs is widely practiced. It has also been suggested that repair of large-diameter AAAs may incur worse outcomes. We sought to examine differences in patient selection, operative outcomes, and survival after elective endovascular aneurysm repair (EVAR) based on AAA diameter thresholds. MethodsElective EVARs for asymptomatic AAAs in the Vascular Quality Initiative were studied from 2003 to 2017. AAAs were classified by diameter as small (<5 cm in women, <5.5 cm in men), medium (5-6.5 cm in women, 5.5-6.5 cm in men), and large (≥6.5 cm). Patient characteristics and operative factors were compared using univariate analyses and established risk prediction models. Effects of AAA diameter on reintervention and mortality were assessed using Kaplan-Meier and multivariable Cox regression analyses. ResultsOf 22,975 patients undergoing EVAR, 41% (9353), 47% (10,842), and 12% (2780) had small, medium, and large AAAs, respectively. Patients with small AAAs were younger and had fewer comorbidities. Consequently, patients with small AAAs were more likely to have low predicted operative mortality risk and 5-year mortality risk based on risk models (P < .001 for both). For operative outcomes, 30-day mortality was significantly different across diameter categories (small, 0.4%; medium, 0.9%; large, 1.6%; P < .001). EVAR for large AAAs had the highest rates of multiple medical complications, including myocardial infarction (P < .001), respiratory complications (P = .001), and renal complications (P < .001). In contrast, EVAR for small AAAs had the lowest rates of type I endoleak at completion and reoperation during index hospitalization, shortest operative times, and shortest hospital length of stay (P < .001 for all). Aneurysm diameter was associated with differential 1-year reintervention-free survival (92% small vs 89% medium vs 82% large; P < .001) and 5-year overall survival (88% small vs 81% medium vs 75% large; P < .001). Multivariable models showed that compared with medium AAAs, small AAAs had an independent protective effect against 1-year reintervention or death (hazard ratio [HR], 0.82; P = .003) and 5-year mortality (HR, 0.78; P = .001). Conversely, compared with medium AAAs, large AAAs carried an independent increased risk of 1-year reintervention or death (HR, 1.75; P < .001) and 5-year mortality (HR, 1.50; P < .001). ConclusionsSmall AAAs represent >40% of elective EVARs in the Vascular Quality Initiative. Patients with small AAAs selected for repair are younger and have fewer comorbidities. Consequently, EVAR for small AAAs carries lower risk of operative and 5-year mortality. Aneurysm diameter is independently associated with reinterventions and mortality after EVAR, suggesting that AAA diameter may have an important clinical effect on outcomes.