Highlight
• The SEGA randomized trial (n=260) compared general anesthesia (GA) versus moderate sedation for endovascular therapy (EVT) in acute ischemic stroke due to large‑vessel occlusion (LVO).
• At 90 days, the ordinal modified Rankin Scale (mRS) distribution favored GA (OR 1.22; 95% credible interval [CrI] 0.79–1.87) with an 81% posterior probability of superiority.
• GA was associated with numerically higher rates of 90‑day independence (mRS 0–2; RR 1.2; 95% CrI 0.9–1.66) and slightly higher rates of successful reperfusion; credible intervals crossed unity and uncertainty remains.
Background: clinical context and unmet need
Endovascular therapy (EVT) is standard of care for acute ischemic stroke (AIS) caused by large‑vessel occlusion (LVO) of the anterior circulation. Optimal periprocedural management, including anesthetic strategy, remains debated. Two common approaches are moderate (conscious) sedation and general anesthesia (GA). Potential advantages of GA include immobilization, airway protection, and controlled ventilation; potential disadvantages include hemodynamic instability, induction‑related delays, and need for mechanical ventilation. Prior randomized trials and observational studies have produced mixed results, yielding persistent uncertainty and variable practice across centers.
Study design: SEGA randomized clinical trial
The SEGA trial was a pragmatic, multicenter randomized clinical trial conducted at 10 comprehensive stroke centers in the United States between July 2018 and August 2023. Adults with anterior circulation LVO (carotid artery, proximal middle cerebral artery, and anterior cerebral artery) who were candidates for EVT were randomized 1:1 to either moderate sedation or GA. The primary endpoint was the ordinal distribution of the modified Rankin Scale (mRS) at 90 days. Key secondary endpoints included the proportion of patients with good functional outcome (mRS 0–2) at 90 days, successful reperfusion rates, and safety outcomes such as symptomatic intracerebral hemorrhage (sICH).
Key results
Population and randomization: Of 1,931 patients screened, 260 were randomized (mean age 66.8 years, SD 13.3; 52% male), with 130 patients assigned to GA and 130 to sedation. This report uses intention‑to‑treat analysis.
Primary outcome (90‑day ordinal mRS): There was a shift in the distribution of mRS favoring GA with an odds ratio (OR) for improvement of 1.22 (95% CrI, 0.79–1.87). The Bayesian analysis reported an 81% posterior probability that GA was superior to sedation for the ordinal mRS.
Secondary outcomes:
• Good functional outcome (mRS 0–2 at 90 days): GA had a relative risk (RR) of 1.20 (95% CrI, 0.90–1.66) with an 89% posterior probability of superiority.
• Successful reperfusion: RR 1.01 (95% CrI, 0.96–1.08) with a 69% posterior probability favoring GA.
• Symptomatic intracerebral hemorrhage: 0.8% (1/125) with GA vs 2.4% (3/125) with sedation; RR 0.71 (95% CrI, 0.23–2.16) and posterior probability of GA superiority 72%.
• Other secondary outcomes (mortality, procedure time, and other adverse events) were broadly similar between groups.
Interpretation of effect sizes and uncertainty: Point estimates consistently favored GA across primary and several secondary endpoints, but credible intervals overlapped 1.0. In Bayesian terms, the posterior probabilities (ranging from ~69%–89% across outcomes) indicate moderate evidence favoring GA but stop short of the high certainty typically used to change practice decisively. The absolute magnitudes of benefit—particularly for functional independence and reperfusion—were modest and imprecise.
Why might GA influence outcome?
Biological plausibility and mechanistic considerations support several pathways by which GA could improve EVT outcomes: complete immobilization may facilitate faster and more precise device deployment; airway protection prevents aspiration and allows controlled ventilation and oxygenation; and deeper anesthetic control can permit consistent cerebral perfusion by controlling PaCO2 and addressing agitation that interferes with angiography. Conversely, GA induction may cause transient hypotension and procedural delay—both of which could worsen ischemic injury—unless managed by experienced anesthesia teams with strict hemodynamic protocols.
Strengths of SEGA
• Randomized design across multiple high‑volume centers increases internal validity compared with observational data.
• Intention‑to‑treat analysis preserves randomization benefits.
• Use of a clinically meaningful primary endpoint (ordinal mRS at 90 days) captures the full range of functional outcomes and aligns with prior EVT trials.
• Bayesian reporting of posterior probabilities provides a probabilistic measure of benefit that clinicians can weigh alongside credible intervals.
Limitations and cautions
• Generalizability: Only 260 of 1,931 screened patients were randomized. High exclusion rates may reflect strict eligibility, logistical constraints, or predefined protocol criteria; the enrolled cohort may not represent the broader stroke population seen in routine practice.
• Unreported or limited procedural detail: The summary provided does not specify periprocedural blood pressure targets, incidence and magnitude of hypotension, times from randomization to groin puncture or recanalization, rates of crossover (patients randomized to sedation converted to GA), or details on anesthetic agents and ventilation strategies. These factors strongly influence outcomes and may vary across centers.
• Statistical uncertainty: Credible intervals for key outcomes included the null, and although posterior probabilities favored GA, they do not reach a high probability threshold (for example, >95%) that many clinicians would consider definitive.
• Potential center effect: Outcomes may depend on local expertise in neuroanesthesia and EVT; centers with experienced anesthesiologists and rapid intubation workflows may achieve benefits not reproducible in lower‑volume settings.
• Sedation heterogeneity: “Moderate sedation” encompasses a range of agents and depths; without standardized sedation protocols, the comparator arm may be heterogeneous, potentially biasing results toward GA in settings where sedation was suboptimal.
How SEGA fits with prior evidence and guidelines
Prior randomized trials (for example, SIESTA, GOLIATH, and ANSTROKE) and meta‑analyses produced mixed results—some showing no difference between GA and conscious sedation, others suggesting potential benefits of GA when blood pressure is strictly managed. The American Heart Association/American Stroke Association (AHA/ASA) guidelines have emphasized individualized decision‑making, recommending that the choice of anesthetic approach be based on patient factors and local expertise rather than a universal rule (Powers et al., 2018).
SEGA strengthens the randomized evidence base by suggesting that GA may be associated with improved reperfusion and functional outcomes in contemporary EVT practice, but it does not provide unequivocal proof that GA is superior in all settings. Rather, SEGA argues that GA is at least noninferior and potentially advantageous when delivered within well‑organized neuroanesthesia and EVT systems with protocols to minimize hypotension and procedural delay.
Clinical implications and practical takeaways
• Individualize anesthetic choice: SEGA supports the permissibility of GA as a frontline strategy for EVT, particularly at centers with experienced neuroanesthesia teams and established blood pressure and airway management protocols.
• Prioritize hemodynamic management: Regardless of sedation strategy, meticulous prevention and treatment of hypotension is critical during EVT to preserve penumbral tissue.
• System readiness matters: If GA is adopted, workflows must minimize induction time and ensure rapid door‑to‑groin and groin‑to‑reperfusion metrics are maintained.
• Monitor and audit outcomes locally: Centers should track procedural times, reperfusion rates, and functional outcomes according to anesthetic strategy, and adapt protocols accordingly.
• Shared decision-making: When feasible, discuss anesthetic options with patients or surrogates, balancing airway/aspiration risks, agitation, comorbidities, and institutional expertise.
Research implications and unanswered questions
• Need for pooled analyses: Individual patient‑level meta‑analyses pooling SEGA with prior randomized trials can refine estimates and identify subgroups most likely to benefit from GA (for example, severe agitation, high aspiration risk, high NIH Stroke Scale scores).
• Mechanistic studies: Trials that standardize blood pressure targets, specify anesthetic agents, and measure physiologic variables (e.g., continuous blood pressure, end‑tidal CO2) will clarify whether benefits stem from reperfusion mechanics or physiologic stability.
• Implementation studies: Research into workflows that minimize induction delays and optimize neuroanesthesia staffing may help translate trial results into broader practice.
Conclusion
The SEGA randomized clinical trial provides important contemporary evidence suggesting that general anesthesia during EVT for anterior circulation LVO may modestly improve 90‑day functional outcomes and increase successful reperfusion rates compared with moderate sedation. The findings are suggestive rather than definitive: credible intervals include the null and important procedural details are needed to inform broader implementation. Clinicians should view SEGA as supportive of GA as a valid and potentially advantageous option—especially in centers experienced with neuroanesthesia—but continue to individualize anesthetic choices based on patient factors, hemodynamic management capacity, and local expertise. Further pooled and mechanistic studies will help define which patients derive the greatest benefit from GA and how to implement it safely at scale.
Funding and clinical trial registration
Trial registration: ClinicalTrials.gov Identifier: NCT03263117. Funding sources and detailed acknowledgments are reported in the original JAMA Neurology publication (Chen et al., 2025).
References
1. Chen PR, Artime CA, Sheth SA, et al.; SEGA Investigators. Sedation vs General Anesthesia for Endovascular Therapy in Acute Ischemic Stroke: The SEGA Randomized Clinical Trial. JAMA Neurol. 2025 Oct 13. doi:10.1001/jamaneurol.2025.3775. Epub ahead of print. PMID: 41082222.
2. Powers WJ, Rabinstein AA, Ackerson T, et al. 2018 Guidelines for the Early Management of Patients With Acute Ischemic Stroke: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association. Stroke. 2018;49:e46–e110.
