Interobserver Agreement on Assessing Endovascular Aneurysm Repair Feasibility in Patients With Ruptured Abdominal Aortic Aneurysms Using Computed Tomography Angiography

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Objective: This study was designed to investigate agreement between vascular surgeons and expert team consensus assessment, and inter-observer variability for endovascular aneurysm repair (EVAR) feasibility in patients with a ruptured abdominal aortic aneurysm (rAAA) on computed tomography angiography (CTA). Methods: Sixteen vascular surgeons assessed EVAR feasibility in 20 rAAA patients based on CTA. The majority of patients had rAAA with challenging anatomy. Surgeons were observed, and time to treatment decision (EVAR or open surgical repair [OSR]) was monitored to mimic an acute setting. Reference measurements were determined by an expert team consensus assessment in an elective setting using central luminal line-generating software without time pressure, following the instructions for use (IFU) of the endografts available to the participants. Primary outcomes were: (1) agreement between vascular surgeons in the simulated emergency setting and the expert team on feasibility of EVAR and (2) time to decision on treatment by the 16 surgeons. Results: Out of 20 rAAA cases, 10 (50%) were deemed feasible for EVAR according to the expert team. There was slight agreement between surgeons and the expert team on feasibility for EVAR (Cohen’s kappa = 0.13, interquartile range [IQR] 0.25). In a median of 12 rAAA (60%) cases, surgeons chose the same treatment as the expert team. Interobserver variability comparing observers with each other was fair (Fleiss’ kappa = 0.28, 95% CI, 0.24-0.32). Median time to treatment decision was 2 minutes and 26 seconds (IQR 2 minutes and 10 seconds). Agreement on main body endograft size between individual surgeons and the expert team occurred in a median of 2 (33%, IQR 1) cases. In 16 (80%) of the cases, EVAR was deemed the preferred treatment option, irrespective of IFU according to the expert team. Conclusion: Vascular surgeons showed minimal consensus when assessing the feasibility of EVAR for patients with rAAA with a challenging anatomy using CTA. The time taken to decide upon EVAR feasibility was notably brief. Clinical Impact In patients with ruptured abdominal aortic aneurysms and complex anatomy, vascular surgeons showed limited agreement when assessing EVAR feasibility using CTA compared with reference measurements determined by an expert team consensus assessment, and considerable interobserver variability. Agreement on main body endograft choice was low. Median time to treatment decision was 2 min and 26 s. Granting additional time for comprehensive assessment of EVAR feasibility, multidisciplinary team-decision making, the use of AI-supported tools such as CLL generating software, and higher levels of clinical experience in complex rAAA cases might improve interobserver agreement in both future research and clinical context.