Abstract
Purpose :
To compare perioperative outcomes in patients who undergo carotid endarterectomy (CEA) for retinal artery occlusion (RAO) versus for cerebrovascular accident (CVA) in the United States and to determine comparative predictors of 30-day perioperative outcomes.
Methods :
A retrospective, multicenter nationwide study of the American College of Surgeons National Surgical Quality Improvement Program database in the eight-year period 2010-2017 of all patients with a diagnosis of RAO or CVA who underwent CEA. Patients were selected based on primary international classification of diseases-9 (2010-2015) or 10 (2015 or later) codes corresponding to RAO or CVA and had a primary procedural current procedural terminology code of CEA. Ideal caliper width 1:1 propensity score (PS) matching was performed on age and sex to yield balanced cohorts. Demographic characteristics, preoperative and intraoperative variables, length of stay (LOS), and 30-day adverse events (AEs) were recorded. Nonparametric hypothesis tests and uni- and multivariate linear and logistic regressions were constructed to determine independent predictors of prolonged LOS or AE.
Results :
52 (1.9%) underwent primary CEA for RAO and 2,757 (98.1%) for CVA. After PS-matching the cohorts were balanced on age (p=1.0) and sex (p=1.0). The median age in both groups was 72.5 years (IQR: 63-78.5 years) and both were 38.5% female. RAO patients were more likely to have a history of COPD (26.8% vs. 4.0%, p=0.015) but there was no significant difference between the two groups in any other demographic characteristics, preoperative, or intraoperative variables. The median hospital LOS was significantly longer in the CVA group (3 [IQR 1-6] vs. 6 [3-8], p=0.005). The most common AEs were pulmonary, myocardial infarction, bleed, and stroke. There was no significant difference in AEs between the two groups, though subsequent stroke was more common in the CVA group (3.8% vs 1.9%). When controlling for other variables, CVA (RR 3.0, 95% CI 1.4-4.7 days, p<0.001) and history of COPD (RR 2.8, 0.3-5.3, p=0.030) significantly predicted prolonged LOS.
Conclusions :
In this comparative analysis of patients who received CEA for RAO vs. CVA, CVA and history of COPD were associated with significantly prolonged LOS. As other perioperative measures were similar between the groups, the prolonged LOS in the CVA group may reflect lengthier recovery from the CVA itself.
This is a 2021 ARVO Annual Meeting abstract.