Abstract
Background: Awake carotid endarterectomy (CEA) under local anesthesia demands an optimal sedation strategy that ensures patient comfort while preserving the ability for real-time neurological assessment. Dexmedetomidine (DEX) and remifentanil (REMI) are widely used agents, but direct comparisons in this setting remain scarce. Methods: Exploratory, retrospective, single-center study of awake CEA (March-July 2019). DEX or REMI infusions were titrated to a Richmond Agitation-Sedation Scale (RASS) of -1 to -2. Outcomes were sedation failure (RASS ≥ +2 despite maximum infusion rate), bradycardia, hypotension, and neurologic events. Statistical analyses used χ(2) test (categorical variables) and Student's t-test or Mann-Whitney test (continuous variables). Associations were assessed with Firth's logistic regression (univariable and bivariate models), reporting odds ratios (OR) with 95% confidence intervals (CI(95%)). Trends in the Bispectral Index (BIS), hemodynamic, and respiratory parameters were assessed using two-way repeated-measures Analysis of Variance (ANOVA). A p-value < 0.05 was considered significant. Results: Fifty-two patients were included (DEX = 25; REMI = 27). DEX group showed more frequent sedation failure (32.0% vs. 3.7%; p = 0.020), bradycardia (36.0% vs. 3.7%; p = 0.009), and hypotension (28.0% vs. 0%; p = 0.011). DEX was associated with increased risk in sedation failure (OR 8.58, CI(95%) 1.70-85.81), bradycardia (OR 10.17, CI(95%) 2.05-101.21), and hypotension (OR 22.30, CI(95%) 2.46-2959.60); the direction of associations remained consistent in bivariate models adjusted for baseline confounders. ANOVA showed group-by-time interactions for BIS, heart rate, mean arterial pressure, and end-tidal CO(2). No intraoperative complications or adverse outcomes were observed. Conclusions: In this retrospective cohort of awake CEA, DEX was associated with higher rates of sedation failure and hemodynamic adverse events compared with REMI, without an apparent impact on procedural success. Given non-random allocation and baseline imbalances, these findings are hypothesis-generating and warrant confirmation in larger, robust, and prospective studies.