Rethinking the impact of amaurosis with contemporary outcomes of carotid endarterectomy among patients with transient monocular blindness.

Abstract

While historically amaurosis fugax was clinically categorized similarly to ipsilateral hemispheric transient ischemic attack (TIA) or stroke, it remains unclear whether transient monocular blindness confers comparable risk to patients in contemporary practice. The purpose of this analysis was to compare postoperative stroke risk among patients undergoing carotid endarterectomy (CEA) for amaurosis, compared with patients with TIA, stroke, or asymptomatic carotid stenosis. We studied patients who underwent CEA in the Vascular Quality Initiative (VQI) (2016-2024) and performed a subgroup analysis among patients with linked Medicare claims data (2016-2019). The primary exposure was baseline symptom status, classified as amaurosis, TIA, stroke, or asymptomatic. The primary outcome was stroke. Logistic regression was used to compare the likelihood of in-hospital stroke across the four groups for all examined VQI patients, and Kaplan-Meier analysis and Cox regression was used among patients with linked Medicare data. We identified 177,859 carotid stenosis patients who underwent CEA (amaurosis, 7.0%; TIA, 12.1%; stroke, 25.4%; and asymptomatic, 55.5%). The in-hospital stroke risk was 0.8% for amaurosis, compared with 1.4%, 2.0%, and 0.7% for TIA, stroke, and asymptomatic presentations, respectively. Compared with asymptomatic patients, the adjusted odds ratio of stroke was 1.17 (95% confidence interval [CI], 0.93-1.46; P = .18) for patients with amaurosis and was 1.76 (95% CI, 1.51-2.06; P < .01) and 2.30 (95% CI, 2.05-2.59; P < .01) for those with TIA and stroke, respectively. We identified 31,010 patients with linked Medicare claims data (amaurosis, 6.0%; TIA, 10.7%; stroke, 21.9%; and asymptomatic, 61.4%). The 3-year stroke risk was 6.1% for patients with amaurosis, compared with 12.4%, 16.0%, and 5.0% for patients with TIA, stroke, and those who were asymptomatic, respectively (log-rank: P < .001). Compared with asymptomatic patients, the adjusted hazard ratio of stroke was 1.15 (95% CI, 0.91-1.45; P = .21) for patients with amaurosis and was 2.03 (95% CI, 1.74-2.36; P < .01) and 2.80 (95% CI, 2.50-3.14; P < .01) for patients with TIA and stroke, respectively. Patients undergoing CEA in the setting of amaurosis fugax had a statistically similar perioperative and 3-year stroke risk compared with asymptomatic patients. By contrast, hemispheric TIAs or stroke conferred greater risks longitudinally. These findings suggest that it may be appropriate to reconsider the historical risk perception associated with transient monocular blindness to a less severe phenotype and can inform preoperative decision-making for patients considering CEA.