Predictors for Type 2 Endoleak Requiring Embolization: Large Inferior Mesenteric Artery, Multiple Lumbar Arteries, and Anticoagulation.

Abstract

IntroductionEndovascular abdominal aortic aneurysm repair (EVAR) is the primary treatment for abdominal aortic aneurysm (AAA). Despite favorable early outcomes, lifelong surveillance is essential as endoleaks remain a major cause of reintervention. Among these, type II endoleaks (T2E) remain controversial regarding optimal management. This study aimed to identify factors associated with intervention for T2E using time-to-event analysis and predefined anatomic thresholds. Secondary objectives included comparing outcomes between T2E patients with (intT2E) and without (nointT2E) intervention.MethodsA retrospective review of EVAR procedures from 2011-2024 was performed. Patients with newly diagnosed or persistent T2E were identified on completion and follow-up CT angiography. Patients were categorized as intT2E or nointT2E. Multivariable Cox regression evaluated time to first intervention, and logistic regression served as sensitivity analysis. Kaplan-Meier curves assessed freedom from intervention by inferior mesenteric artery (IMA) size.ResultsAmong 207 EVAR patients, 78 (37.6%) developed T2E over a mean 3.4 ± 2.4 years. Nineteen (24.3%) required intervention. IntT2E patients were younger (74.9 ± 6.8 vs 78.3 ± 7.9 years, P = 0.02), had more frequent anticoagulation use (47.4% vs 20.3%, P = 0.02), larger IMAs (4.2 ± 0.6 vs 3.3 ± 0.7 mm, P < 0.01), and more lumbar arteries (6.0 ± 1.2 vs 5.0 ± 1.3, P = 0.02). IMAs >4 mm occurred in 68.4% of intT2E vs 8.5% of nointT2E (P < 0.01). Mean sac growth was greater in intT2E (9.7 ± 5.3 mm vs -1.9 ± 6.5 mm, P < 0.01). Thirty-day complications following the index EVAR procedure, mid-term reinterventions, and overall mortality were similar between groups. Post-embolization, T2E resolved in 36.9%, with sac regression in 63.1%. Among embolized patients, 26.3% required a second embolization and one patient required 3 attempts without achieving resolution. IMA >4 mm independently predicted intervention (HR 7.18, 95% CI 1.97-26.16, P < 0.01). Logistic regression confirmed IMA >4 mm (OR 23.4, 95% CI 6.17-88.6, P < 0.01), >6 lumbar arteries (OR 4.2, 95% CI 1.10-15.98, P = 0.02), and anticoagulation (OR 3.4, 95% CI 1.17-10.6, P = 0.02) as predictors.ConclusionsApproximately one-quarter of T2E patients required embolization. IMA >4 mm was the strongest predictor of intervention, while anticoagulation and increased lumbar artery number also increased risk. Management should prioritize risk-stratified surveillance and selective intervention.