Risk factors for mortality within 5 years of carotid endarterectomy for asymptomatic stenosis

Objective: The current Society for Vascular Surgery guidelines for the treatment of patients with asymptomatic carotid stenosis recommend endarterectomy for patients with >70% stenosis and acceptable surgical risk. The reduced rate of stroke with modern medical therapy has increased the importance of careful selection in deciding which patients should undergo elective carotid endarterectomy (CEA) for asymptomatic disease. It would, therefore, be very prudent to investigate preexisting variables predictive of 5-year mortality for patients meeting the criteria to undergo CEA.

Methods: The Vascular Quality Initiative was queried from 2003 onward for all cases of CEA. Inclusion in the study required the following: (1) documentation of survival status; (2) complete data on all incorporated demographic study variables; and (3) asymptomatic neurologic status. The variables present at surgery were investigated using binary logistic regression to identify multivariate predictors of 5-year mortality. The highest risk variables were then interrogated for an additive effect regarding long-term mortality. A subanalysis was performed for patients aged >80 years.

Results: A total of 30,615 patients met the inclusion criteria, 5414 (18%) of whom had died within 5 years. The highest risk variables were classified as those that had had an adjusted odds ratio >1.25, P < .001, and beta coefficient of ≥0.25. These included a body mass index <20 kg/m², diabetes mellitus, a history of congestive heart failure, renal insufficiency, end-stage renal disease, chronic obstructive pulmonary disease, living status other than home, prior lower extremity bypass, prior major amputation, Black race relative to other races combined, hemoglobin <10 mg/dL, a history of neck irradiation, and a history of smoking. Age had an annual odds ratio of 1.04 (P < .001). Other variables that achieved a statistically significant (P < .05) association with 5-year mortality were coronary artery disease, a positive stress test or the occurrence of myocardial infarction within 2 years, lower extremity arterial intervention, aneurysm repair, and P2Y12 inhibitor therapy at surgery. The use of statin and aspirin therapy at surgery were both protective against 5-year mortality (P < .001).

Conclusions: We identified 12 particularly high-risk variables, which, in combination, progressively predicted for increasing mortality within 5 years of CEA performed for asymptomatic stenosis. Special attention should be given to patients aged >80 years and patients with any history of congestive heart failure regardless of current symptoms, chronic obstructive pulmonary disease, renal insufficiency or end-stage renal disease, peripheral artery disease, diabetes, and variables associated with frailty (BMI under 20, anemia, assisted living status).