Single Centre Evaluation of the Proposal of the European Society for Vascular Surgery Abdominal Aortic Aneurysm Guidelines to Stratify Surveillance after Endovascular Aortic Aneurysm Repair

Objective: 

The aim of this study was to evaluate and compare methods that identify patients at low risk of developing complications after endovascular aortic aneurysm repair (EVAR) and who would thus not require surveillance in the first post-operative years. 

Methods: 

This was a retrospective, single centre, cohort study including all patients after elective infrarenal EVAR with both immediate post-operative and one year computed tomography angiography (CTA) imaging. Patients were categorised by adherence to instructions for use (IFU), adequate seal, and absence of endoleak (method A1), and without high risk features (method A2) on the first post-operative CTA. Additionally, these patients were dichotomised based on aneurysm sac shrinkage at one year (> 5 mm maximum diameter reduction, method B). Outcomes were graft related adverse events and all cause death. Negative predictive value (NPV) was used to compare risk classifications. 

Results: 

Of 422 eligible patients, 297 underwent the required imaging for classification: 140 (47.1%) and 109 (36.7%) patients were classified as low risk based on methods A1 and A2, respectively, while 147 (49.5%) were assumed low risk based on method B. The five year cumulative incidence of adverse events in low risk patients according to method A1 was 14.7% (95% confidence interval [CI] 8.5 – 20.9%), similar to method A2 (16.1%, 95% CI 8.8 – 23.4%) and method B (15.4%, 95% CI 9.3 – 21.5%). The five year median NPV for adverse events for method A1 was 85.2% (95% CI 79.7 – 90.8%), comparable with method A2 (83.8%, 95% CI 76.9 – 90.3%; p = .37) and method B (84.7%, 95% CI 79.4 – 89.5%; p = .87). Significantly higher NPVs were found by combining method A1 or A2 with method B, with median values ≥ 95% up to four years after EVAR. The five year NPV for death did not differ between methods (five year NPV_{method A1}, 81.7%, 95% CI 76.6 – 86.5%). 

Conclusion: 

Refraining from imaging in the first five years after EVAR in patients treated within IFU and with a favourable post-operative CTA would have failed to detect important complications at an early stage. It is proposed to combine the post-operative CTA with sac shrinkage at one year in order to stratify post-EVAR surveillance. No benefit was found in considering the high risk features suggested in the European Society for Vascular Surgery (ESVS) guidelines.