Philippe Charbonneau1, Kiattisak Hongku2, Christine R Herman3, Mohammed Habib1, Elie Girsowicz1, Robert J Doonan1, Luc Dubois4, Sajjid Hossain4, Heather L Gill1, Kent S Mackenzie1, Jason P Bayne1, Daniel Obrand1, Oren K Steinmetz5. 1. Division of Vascular Surgery, McGill University, Montreal, Quebec, Canada. 2. Division of Vascular Surgery, McGill University, Montreal, Quebec, Canada; Division of Vascular Surgery, Mahidol University, Bangkok, Thailand. 3. Division of Vascular Surgery, McGill University, Montreal, Quebec, Canada; Divisions of Cardiac and Vascular Surgery, Dalhousie University, Halifax, Nova Scotia, Canada. 4. Division of Vascular Surgery, Western University, London, Ontario, Canada. 5. Division of Vascular Surgery, McGill University, Montreal, Quebec, Canada. Electronic address: oren.steinmetz@muhc.mcgill.ca.
Abstract
BACKGROUND: Randomized controlled trials of long-term survival for infrarenal abdominal aortic aneurysms have compared open surgical repair (OSR) with endovascular aneurysm repair (EVAR) in patients with suitable aortic anatomy for EVAR. However, in clinical practice, patients who do not meet instructions for use (IFU) criteria are often still treated by EVAR despite that some studies show higher graft-related adverse events. The goal of this study was to compare the long-term survival of EVAR and OSR in patients with anatomy outside IFU criteria for EVAR. METHODS: This multicenter retrospective cohort study included patients with at least one anatomic IFU violation for EVAR undergoing either elective EVAR or elective OSR for abdominal aortic aneurysm. Demographics, anatomic data, and follow-up data of patients were collected from three academic centers from 2003 to 2016. Device-specific IFU were used for EVAR patients, whereas generic IFU for EVAR were applied to the OSR patients. The primary outcomes were 30-day mortality and long-term all-cause mortality. Secondary outcomes were aneurysm-related mortality and perioperative complications at 30 days. Kaplan-Meier survival and Cox proportional hazards modeling were performed. Inverse propensity score weights were used to adjust for differences in treatment selection. RESULTS: The study population included 202 EVAR patients and 224 OSR patients with at least one anatomic IFU violation for EVAR. EVAR patients were older (78.1 ± 7.3 vs 70.9 ± 7.0 years; P < .001) and less likely to be hypertensive (69.3% vs 79.0%; P = .02) compared with OSR patients. OSR patients were more likely to have proximal aortic neck IFU violations (75.0% vs 47.1%; P < .001) and were less likely to have iliac IFU violations (65.2% vs 79.2%; P < .001). All-cause mortality was 37.6% in the EVAR group and 24.1% in the OSR group with a median follow-up time of 5.2 (3.5-7.2) and 5.4 (2.8-9.3) years, respectively (P < .002). Kaplan-Meier survival analysis revealed a significant association between patients undergoing OSR and increased long-term survival (log-rank P < .0001). When adjusted for possible confounders and weighted for propensity for treatment through Cox hazard modeling, the association remained significant (hazard ratio, 0.6; 95% confidence interval, 0.4-0.9). Aneurysm-related mortality was 3.5% in the EVAR group and 2.2% in the OSR group during long-term follow-up (P < .001). CONCLUSIONS: Our study identified that patients with IFU violations have higher overall long-term survival with open surgery compared with EVAR. Caution should be applied in considering standard EVAR for patients with anatomy outside of IFU.
BACKGROUND: Randomized controlled trials of long-term survival for infrarenal abdominal aortic aneurysms have compared open surgical repair (OSR) with endovascular aneurysm repair (EVAR) in patients with suitable aortic anatomy for EVAR. However, in clinical practice, patients who do not meet instructions for use (IFU) criteria are often still treated by EVAR despite that some studies show higher graft-related adverse events. The goal of this study was to compare the long-term survival of EVAR and OSR in patients with anatomy outside IFU criteria for EVAR. METHODS: This multicenter retrospective cohort study included patients with at least one anatomic IFU violation for EVAR undergoing either elective EVAR or elective OSR for abdominal aortic aneurysm. Demographics, anatomic data, and follow-up data of patients were collected from three academic centers from 2003 to 2016. Device-specific IFU were used for EVAR patients, whereas generic IFU for EVAR were applied to the OSR patients. The primary outcomes were 30-day mortality and long-term all-cause mortality. Secondary outcomes were aneurysm-related mortality and perioperative complications at 30 days. Kaplan-Meier survival and Cox proportional hazards modeling were performed. Inverse propensity score weights were used to adjust for differences in treatment selection. RESULTS: The study population included 202 EVAR patients and 224 OSR patients with at least one anatomic IFU violation for EVAR. EVAR patients were older (78.1 ± 7.3 vs 70.9 ± 7.0 years; P < .001) and less likely to be hypertensive (69.3% vs 79.0%; P = .02) compared with OSR patients. OSR patients were more likely to have proximal aortic neck IFU violations (75.0% vs 47.1%; P < .001) and were less likely to have iliac IFU violations (65.2% vs 79.2%; P < .001). All-cause mortality was 37.6% in the EVAR group and 24.1% in the OSR group with a median follow-up time of 5.2 (3.5-7.2) and 5.4 (2.8-9.3) years, respectively (P < .002). Kaplan-Meier survival analysis revealed a significant association between patients undergoing OSR and increased long-term survival (log-rank P < .0001). When adjusted for possible confounders and weighted for propensity for treatment through Cox hazard modeling, the association remained significant (hazard ratio, 0.6; 95% confidence interval, 0.4-0.9). Aneurysm-related mortality was 3.5% in the EVAR group and 2.2% in the OSR group during long-term follow-up (P < .001). CONCLUSIONS: Our study identified that patients with IFU violations have higher overall long-term survival with open surgery compared with EVAR. Caution should be applied in considering standard EVAR for patients with anatomy outside of IFU.