A. Laichetty, K. Zhou, W. C. Broaddus
Background. The optimal anesthetic strategy for mechanical thrombectomy remains contested; clinical trials and meta-analyses yield conflicting results. Emerging evidence has shown stress hyperglycemia ratio (SHR) to be among the strongest independent predictors of poor thrombectomy outcomes. General anesthesia (GA) is known to provoke a more pronounced neuroendocrine stress response than procedural sedation, yet no prior study has investigated perioperative glucose dynamics between GA and sedation in thrombectomy patients. Methods. Using the TriNetX federated research network, we conducted a retrospective propensity score-matched cohort study of non-diabetic adults undergoing mechanical thrombectomy for anterior circulation acute ischemic stroke under GA versus sedation. After propensity score matching 2,556 matched pairs remained. Results. At the population level, mean glucose divergence was modest in absolute terms (Day 0: 110.90 vs 106.77 mg/dL; ? = 4.13 mg/dL; p = 7.82 x 10-11; Day 1: 107.69 vs 103.25 mg/dL; ? = 4.44 mg/dL; p = 8.16 x 10-14). GA conferred an 83.2% relative risk increase to severe-range hyperglycemia, defined as blood glucose > 181 mg/dL on the day of the index procedure (Day 0: RR = 1.832, 95% CI 1.457 ? 2.303; NNH: 29), with effects persisting on day 1 following the thrombectomy (Day 1: RR = 1.66, 95% CI 1.228 ? 2.249). GA patients received significantly more exogenous insulin on both days 0 (GA: 21.24%, Sedation: 17.23%, RR = 1.231, p = 2.96 x 10-4) and day 1 (GA: 11.3%, Sedation: 8.69%, RR = 1.30, p = 0.00178). All primary outcomes survived Bonferroni correction. Conclusion. General anesthesia is strongly associated with persistent hyperglycemia following mechanical thrombectomy in non-diabetic patients. These findings identify anesthesia-modality as a potential modifiable risk and provides a mechanistic hypothesis in the ongoing GA versus CS debate for thrombectomy. Future glucose management protocols may need to be tailored to anesthetic strategy.