LAUSR

Objective Quality metrics were developed to improve outcomes after carotid artery revascularization; however, few studies have evaluated regional differences in perioperative outcomes. This study aimed to evaluate regional variation in mortality and perioperative outcomes after carotid endarterectomy (CEA) and carotid artery stenting (CAS). Methods We identified all patients who underwent CEA or CAS from 2009 to 2016 in the Vascular Quality Initiative. Patients were analyzed on the basis of their symptom status. We assessed variation in perioperative outcomes using χ2 analysis, Fisher exact test, and t‐test, where appropriate. Results A total of 78,467 carotid interventions were identified; 85% were CEAs, with 69% of those asymptomatic. Within CAS, 39% were asymptomatic. Perioperative stroke/death varied across regions within both CAS groups (asymptomatic, 0%‐5.8% [P = .03]; symptomatic, 2.4%‐8.1% [P = .1]), and several regions did not meet the American Heart Association (AHA) guidelines of 3% for asymptomatic patients and 6% for symptomatic patients, which persisted after risk adjustment. For CEA, the stroke/death rates fell within the standards set by the AHA guidelines in all regions for both the unadjusted and risk‐adjusted models; however, there was significant regional variation in the cohorts (asymptomatic, 0.9%‐3.1% [P < .01]; symptomatic, 1.3%‐4.9% [P < .01]). Variation in 30‐day mortality was significant in symptomatic patients (asymptomatic: CEA, 0%‐1.3% [P = .2], CAS, 0%‐2.4% [P = .2]; symptomatic: CEA, 0%‐1.8% [P < .01], CAS, 0%‐4.6% [P = .01]). Rates of in‐hospital stroke, postoperative myocardial infarction, prolonged length of stay (>2 days), and use of intravenous blood pressure medications all varied significantly across the regions. After CEA, there was significant variation in the rates of cranial nerve injuries (asymptomatic, 0.9%‐4.9% [P < .01]; symptomatic, 1.5%‐7.7% [P < .01]), return to the operating room (asymptomatic, 0.9%‐3.4% [P < .01]; symptomatic, 0.6%‐3.4% [P = .02]), and discharge on antiplatelet and statin (asymptomatic, 75%‐87% [P < .01]; symptomatic, 78%‐91% [P < .01]). After CAS, significant variation was found in the rates of access site complications (asymptomatic, 2.3%‐18.2% [P < .01]; symptomatic, 1.4%‐16.9% [P < .01]) and discharge on dual antiplatelet therapy (asymptomatic, 79%‐94% [P < .01]; symptomatic, 83%‐93% [P < .01]). Conclusions Unwarranted regional variation exists in outcomes after carotid artery revascularization across the regions of the VQI. Significant variation was seen in a number of outcomes for which quality metrics currently exist, such as length of stay and discharge medications. In addition, after CAS, several regions failed to meet the AHA guidelines for stroke and death. Given these results, quality improvement projects should be targeted to improve adherence to current guidelines to promote best practices.