Generalized anesthesia is superior to conscious sedation, offering better functional outcomes in stroke patients with large vessel occlusion (LVO) undergoing endovascular therapy (EVT), results of a randomized clinical trial have shown.

The findings should serve as “a wake-up call” that hospital policy changes are needed, study investigator Peng Roc Chen, MD, professor and director, Cerebrovascular/Endovascular Program, Department of Neurosurgery, University of Texas McGovern Medical School in Houston, told Medscape Medical News.

The study suggests that all stroke patients undergoing EVT should receive general anesthesia unless contraindicated, but as it stands, “many, many centers don't have anesthesiologists” to support this, said Chen.

The results were published October 13, 2025, in JAMA Neurology. 

Optimal Approach Unclear 

EVT is the standard of care for selected patients with acute ischemic stroke (AIS) due to LVO. However, the optimal anesthetic technique during the procedure is unclear.

Research suggests that GA might lead to better functional outcomes because it improves oxygenation in the penumbra, the area surrounding the core infarct that is functionally impaired but still metabolically active. Neurons in this region are at risk of dying if blood flow is not restored promptly, Chen explained.

Current anesthesia EVT practices in the US are “all over the place,” with limited availability of anesthesiologists a contributing factor, said Chen. 

Conscious sedation can be managed by EVT interventionists with certified nurses, whereas GA requires the expertise of anesthesiologists.

The investigators set out to determine whether general anesthesia (GA) or moderate sedation affects 90-day functional outcomes in patients undergoing EVT for acute ischemic stroke (AIS) due to LVO.

The Sedation vs General Anesthesia for Endovascular Therapy in Acute Ischemic Stroke (SEGA) trial included 260 adult patients from 10 US centers with a pre-stroke modified Rankin Scale (mRS) score of 2 or less who were to undergo EVT for AIS due to LVO. The mean age of study participants was 66.8 years, 48% were female, and 50.2% were White.

Participants were randomized to receive either general anesthesia (GA) or conscious (moderate) sedation during the procedure. Patients under GA are fully unconscious, mechanically ventilated, and unresponsive, whereas those receiving conscious sedation have reduced awareness, breathe independently, and can respond to verbal commands.

The study’s anesthetic protocol outlined the recommended techniques and classes of agents, but the supervising anesthesiologist could select specific drugs and dosages within those guidelines.

GA Bests Sedation

The study design permitted patients to switch from their initially assigned group. A total of 19 patients crossed over, with all but one moving from moderate sedation to the GA group. The most common reason was severe agitation, though other causes included seizures, vomiting, and difficulty maintaining the airway.

Baseline characteristics were similar between groups. The median National Institutes of Health Stroke Scale (NIHSS) score was 15.

Median times from stroke onset to groin puncture were 488 minutes for GA and 508 minutes for sedation. Door-to-groin times were 78 vs 69 minutes, and angiography suite-to-groin times were 17 vs 13 minutes, respectively.

The primary outcome was 90-day modified Rankin Scale (mRS) scores, analyzed in 120 participants from each group. Researchers employed Bayesian analyses, which combine prior evidence — in this case, three studies from outside the U.S. — with data from the current trial to estimate the probability of treatment benefit and harm.

The posterior probability from this analysis reflects the likelihood that GA provides benefit (or harm) compared with sedation. Trial success was defined as a posterior probability greater than 80% that GA improves 90-day mRS scores.

In the primary analysis, GA was superior to sedation (odds ratio [OR] 1.22, 95% credible interval [CrI] 0.79-1.87; posterior probability of benefit from GA 81%).

Functional independence at 90 days (mRS 0-2) occurred more frequently in the GA group, with a relative risk (RR) of 1.20 (95% CrI, 0.90-1.66; posterior probability of benefit from GA 89%).

Time to Rethink Anesthesia Practices? 

Chen noted that better functional outcomes for patients receiving GA during EVT not only save the health care system money but also strengthen the case for hiring more anesthesiologists.

Successful reperfusion was achieved in 97% of patients receiving GA versus 95% receiving sedation, with a relative risk of 1.01, a 95% credible interval of 0.96 to 1.08, and a posterior probability of benefit from GA of 69%.

For safety outcomes, procedural complications occurred in 3.2% of the GA group and 4.0% of the sedation group, corresponding to a relative risk of 0.88, a 95% credible interval of 0.34 to 2.2, and a posterior probability of benefit from GA of 61%.

Symptomatic intracerebral hemorrhage was less frequent in the GA group at 0.8% compared with 2.4% in the sedation group, with a relative risk of 0.71, a 95% credible interval of 0.23 to 2.16, and a posterior probability of benefit from GA of 72%.

In-hospital mortality was slightly higher in the GA group at 8.5% compared with 7.0% in the sedation group, with a relative risk of 1.13 and a 95% credible interval of 0.57 to 2.27. Chen noted that this difference “is very small” and could be due to chance. The mean hospital stay was shorter for the GA group, averaging 6.6 days vs 7.8 days for the sedation group.

Looking ahead, Chen said he hopes these findings will “shift the paradigm” of anesthesia practices for this patient population. He and his team are also conducting post hoc analyses comparing anesthesia agents, noting that roughly half of the GA group received anesthesia via inhalation and the other half intravenously.

One potential limitation of the study is that both groups were managed by anesthesiologists, unlike many EVT patients who receive sedation from neuro-interventionists and nurses. The study only included patients treated within 16 hours of last known well, and while outcome assessments were blinded, the interventions were not.

A 'Valuable Contribution'

Commenting on the findings for Medscape Medical News, Nestor Gonzalez, MD, professor of neurosurgery, Cedars Sinai Medical Center in Los Angeles, California, said the study “is a valuable contribution to our understanding of the role of anesthetic management in patients undergoing EVT.”

The results “support the use of general anesthesia in institutions that have access to anesthesiologists for emergency EVT interventions,” said Gonzalez, who chaired the 2024 American Heart Association (AHA) Scientific Statement on “large-core ischemic stroke endovascular treatment.”

For stroke centers with such access, the results support organizing teams to ensure availability for emergency EVT interventions, said Gonzalez. He emphasized that clinical practice “should consider local resources and patient characteristics when deciding anesthetic approaches.”

He noted that 1931 patients were assessed for study eligibility but only 260 were randomized, which is a “substantial attrition” that underscores limited generalizability, he said. “The results cannot be generalized outside the population studied.”

Gonzalez noted that the views he expressed are his own and don’t necessarily reflect the position of the AHA.

The trial was funded by Stryker Neurovascular and The Weatherhead Foundation. 

Chen and Gonzalez report no relevant conflicts of interest.