(IPC22) Propensity Score-Matched Analysis of Open Versus Endovascular Abdominal Aortic Aneurysm Repair in Standard Surgical-Risk Patients Using a Medicare-Matched Registry

Objectives: Open surgical repair (OSR) is the most durable option on the long-term for abdominal aortic aneurysm (AAA) repair compared to the endovascular approach (EVAR). Despite this, over 80% of AAA repairs in the US are performed using EVAR. Although EVAR is ideal for high-risk patients who cannot tolerate OSR, both OSR and EVAR are viable options in standard-risk patients, and the choice often depends on operator experience and patient preference. This study aims to investigate which technique is associated with superior long-term outcomes in standard-risk patients using a Medicare-matched database.

Methods: The Medicare-matched Vascular Implant Surveillance and Interventional Outcomes Network (VISION) database was queried from 2003 to 2019 for patients who underwent AAA repair. Our exposure groups were standard-risk patients undergoing OSR or EVAR. To include only standard-risk patients, those with high surgical risk features were excluded: age>70, BMI>35, GFR < 30, moderate-severe CHF, recent MI (within 6 months), positive stress test, on home oxygen, dialysis-dependent, and patients unfit for OSR based on surgeon’s judgement. Data on anatomical variables is not included in VISION. Patients presenting with rupture or with prior history of aneurysm repair were also excluded. Propensity score matching (PSM 1:1) was used to match baseline characteristics of patients in both groups. Matched covariates (26 variables) included demographics, comorbidities, aneurysm size, and medications. Cox regression, Kaplan Meier, and Log-Rank test analyzed long term outcomes which include 7-year mortality, aneurysm-related reintervention, and rupture.

Results: Before matching, the total sample included 7,409 standard-risk patients [EVAR 5,933(80.1%); OSR 1,476(19.9%)]. Patients in the EVAR group were more likely to be male, diabetic, non-smokers and normotensive. PSM produced two well balanced (SD < 0.1) cohorts of 1,017 patients in each group. There was no significant difference in 7-year mortality in both groups [EVAR 23.2% vs OSR 25.6%; HR=1.08,(0.85-1.38),p=0.511] (Table I). However, EVAR was associated with higher risk of reintervention at 7 years compared to OSR [20.8% vs 12.4%; HR=1.56,(1.17-2.07),p=0.002] (Fig-1). Furthermore, the hazard of rupture was almost 4-folds higher in the EVAR group [3.2% vs 0.4%; HR=3.87,(1.26-11.91),p=0.018].

Conclusions: In the current era where EVAR is the predominant choice even in low surgical risk patients, reasons to undergo OSR are limited. However, our study demonstrates superior durability of OSR in standard-risk patients with significantly lower rates of reintervention and rupture up to seven years. Our findings provide a strong rationale for performing OSR in physiologically fit patients with AAAs indicated for repair.