General Anesthesia May Negate Endovascular Benefit in Stroke

February 17, 2015

NASHVILLE, TENNESSEE — A new subgroup analysis of the MR CLEAN stroke trial has shown that patients who underwent the endovascular clot retrieval procedure with local anesthesia did much better than those who received general anesthesia.

Patients who had the procedure under local anesthetic had reduced disability and were more likely than those who did not receive endovascular therapy to have a good functional outcome, while those who had general anesthetic showed almost no incremental benefit with the endovascular procedure vs medical therapy alone, basically missing out on the benefit shown in the overall trial.

The new analysis on general anesthetic use was presented here at the International Stroke Conference (ISC) 2015 by Olvert A. Berkhemer, MD, Academic Medical Center, Amsterdam, the Netherlands.

"This is not randomized data so it is difficult to know exactly what is happening here," Dr Berkhemer commented. "It would be good to have a randomized study on this, but our data back up what is already being reported in the literature, and many centers that used to routinely use general anesthesia are now trying to avoid it."

Dr Olvert A Berkhemer (left) and Dr Diederik W Dippel (right)

"All signs are going in the direction of not using general anesthesia," senior MR CLEAN author, Diederik W. Dippel, MD, PhD, Erasmus MC University Medical Center, Rotterdam, the Netherlands, added. "We do not use it in our center and while I would like to see some randomized data on this, I would not want to enter patients in such a trial."

Chair of the ISC press conference at which the study was discussed, Bruce Ovbiagele, MD, Medical University of South Carolina, Charleston, and vice-chair of ISC 2015, said, "This is another line of support that general anesthesia may be less beneficial than local anesthesia for patients undergoing endovascular therapy for stroke. But this is nonrandomized data so it is not conclusive."

Nonrandomized Data

The main results of MR CLEAN, published December 17, 2014, in the New England Journal of Medicine, showed that stroke patients with an occluded major cerebral artery who underwent endovascular intervention (mostly after thrombolysis) were left less disabled.

The primary outcome — the odds ratio of achieving a lower score on the modified Rankin Scale (mRS) (shift analysis) at 90 days with endovascular therapy — was 1.67. In addition, patients who underwent the endovascular procedure had double the likelihood (odds ratio, 2.0) of achieving functional independence (mRS score, 0 to 2) vs those who received thrombolysis.

Dr Berkhemer explained that a recent systematic review of the literature has suggested that patients undergoing such endovascular procedures did better if they were not given general anesthesia, and the MR CLEAN investigators decided to look at this issue in their data.

Of the 216 patients who underwent endovascular therapy in MR CLEAN, 79 received general anesthesia and 137 underwent the procedure without general anesthesia. There was some crossover, with 6 patients who started out not receiving general anesthesia requiring the anesthetic during the procedure.

Dr Berkhemer explained that most hospitals have a policy on whether to use general anesthesia for these procedures, and in the MR CLEAN study it was left to individual centers and operators to make this decision.

Results showed that patients who did not receive general anesthesia had double the likelihood of reduced disability at 90 days with endovascular therapy vs no endovascular therapy, and were almost 3 times more likely to achieve functional independence (mRS score, 0 to 2), whereas those who received general anesthesia showed "almost no benefit" from the procedure, he noted.

Odds of Reduced Disability/Independent Functional Outcome With Endovascular Therapy

Endpoint Odds Ratio With No General Anesthetic (95% Confidence Interval) Odds Ratio With General Anesthetic (95% Confidence Interval)
Reduced disability (mRS shift analysis) 2.13 (1.46 - 3.11) 1.09 (0.69 - 1.71)
Functional independence (mRS score, 0 - 2) 2.79 (1.70 - 4.59) 1.09 (0.56 - 2.12)


Major safety or complication rates did not differ between the two groups.

Asked whether the results could have been due to general anesthesia being used in sicker patients, Dr Berkhemer replied that pretreatment National Institutes of Health Stroke Scale scores were the same in the two groups, which would suggest similar baseline status. In addition, he noted, most hospitals have fixed protocols on whether general anesthesia is used, making it unlikely that this could be the reason for the results.

On whether the difference in outcomes could have been caused by a time delay due to anesthesia being given, Dr Berkhemer pointed out that although patients who received anesthesia had a slightly longer time to the start of endovascular treatment (64 minutes from randomization vs 50 minutes for no general anesthetic), time to reperfusion did not differ.

Dr Ovbiagele told Medscape Medical News that most institutions prefer to do these procedures without general anesthesia because it takes time to call an anesthetist and any delay is harmful. "These data are probably going to cause more operators to think this way."

As a rule, observational data should not drive change. Dr. Larry Goldstein

Another outside commentator, Larry Goldstein, MD, Duke University, Durham, North Carolina, also highlighted the difficulties of interpreting such post hoc observations.

"There are always issues related to residual confounding in these analyses," he said. "Yes, there have been other case series showing similar suggestions, but there are too many variables which have not been accounted for, such as which specific general anesthetics were used, how and by whom were they administered, and comorbidities of patients."

He added: "It is hard to know if this will change practice. As a rule, observational data should not drive change."

Dr Goldstein noted that the policy at Duke at present is to use general anesthetic for these procedures. "Having people lie still is not the easiest."

Dr Ovbiagele elaborated: "There are challenges of doing such interventional procedures in awake patients. If patients are lying still access is normally faster and if they can't lie still, we have to use some sedation."

Speculating on why general anesthesia may have an adverse effect, Dr Ovbiagele pointed out that general anesthesia tends to lower blood pressure. "We want to avoid this in acute stroke as the brain needs perfusion. In addition, there may be a time delay factor and a higher risk of pneumonia from intubation."

MR CLEAN was funded by the Dutch Heart Foundation, with nominal unrestricted grants from AngioCare BV, Covidien/EV3, MEDAC Gmbh/LAMEPRO, Penumbra Inc, and Top Medical/Concentric. The authors have disclosed no relevant financial relationships.

International Stroke Conference (ISC) 2015. Abstract LB17. Presented February 13, 2015.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.