Regional variation exists in patient selection and treatment of abdominal aortic aneurysms.

OBJECTIVE: Significant regional variation in surgical rates has been identified following multiple surgical procedures. However, limited data have examined the regional variability in patient selection and treatment of abdominal aortic aneurysms (AAAs). This study aimed to evaluate regional variation in patient selection, perioperative management, and operative approach for the repair of AAAs.

METHODS: All patients undergoing open repair or endovascular aneurysm repair (EVAR) of an AAA in the Vascular Quality Initiative from 2009 to 2014 were identified. All regional groups were deidentified, and those with fewer than 100 open repairs were combined into a single region.

RESULTS: We identified 17,269 elective repairs (EVAR, 13,759; open, 3510) and 1462 ruptured AAAs (EVAR, 749; open, 713). There was significant regional variation in the use of EVAR for elective repair (range, 66%-88%; P < .01) and ruptured AAA repair (40%-80%; P < .01). The median diameter for elective repair was similar among regions (EVAR, 5.4 cm; open, 5.7 cm). There was wide variation in the treatment of small aneurysms in male patients (<5.5 cm) for EVAR (34%-49%; P < .01) and open repair (17%-38%; P < .01) and variation in the treatment of small aneurysms in female patients (<5 cm) for EVAR (14%-32%; P < .01) but not significant for open repair (6%-24%). For elective cases, preoperative aspirin (EVAR, 50%-75% [P < .01]; open, 49%-78% [P < .01]) and statin use (EVAR, 61%-75% [P < .01]; open, 56%-80% [P < .01]) varied widely. Among elective cardiac patients, preoperative management varied significantly, including beta-blocker use (EVAR, 66%-78% [P < .01]; open, 69%-88% [P = .01]) and the frequency of stress tests (EVAR, 33%-64% [P < .01]; open, 31%-73% [P < .01]). Among open repairs for aneurysms extending at or beyond the juxtarenal segment, there was wide variation in the use of retroperitoneal exposures (7%-70%; P < .01) and adjunctive renal protective measures (cold renal perfusion, 2%-43% [P < .01]; mannitol, 47%-92% [P < .01]).

CONCLUSIONS: Significant regional variation exists in patient selection, perioperative management, and operative approach for the repair of AAA. Definitive evidence is lacking in many aspects of operative care, including the use of the retroperitoneal approach and renal protective strategies. However, this variation emphasizes the importance of research to determine best practice in the areas of greatest variation. Furthermore, where current clinical process measures exist and data are clear, such as the use of statin and antiplatelet agents, the high degree of variation should serve as an impetus for regional quality improvement projects.