The Source for Neurovascular News and Education

July 28, 2021


Despite clinicians having high levels of confidence in their decisions, there is scant consensus among them regarding use of open surgical and endovascular techniques.


Specialists largely do not agree on whether ruptured intracranial aneurysms should be managed surgically or endovascularly, according to a new study. This lack of consensus remains true regardless of operator and aneurysm characteristics.


In their paper published last month in the Journal of Neurosurgery, the authors emphasize that these findings are strong evidence in support of randomized trials to guide treatment selection.


Jean Raymond, MD (CHUM Research Center, Notre-Dame Hospital, Montreal, Canada), and colleagues constructed an electronic portfolio of 42 cases of ruptured intravascular aneurysms, each of which included both an angiographic image and a brief clinical vignette. All the aneurysms had been categorized based on International Subarachnoid Aneurysm Trial (ISAT) criteria, with 18 defined as ISAT (ie, small, anterior-circulation, non–middle cerebral artery location) and 22 defined as non-ISAT. Among the non-ISAT aneurysms, two were basilar apex aneurysms, for which a high number of endovascular choices was expected. The portfolios did not disclose ISAT status.


“The study was designed to transparently identify and measure the clinical uncertainty involved in the management of ruptured aneurysms … because … clinicians and patients alike should be aware that diverse options are actually being variably proposed for the management of similar patients, if only to make the alternative option available,” Raymond told Neurovascular Exchange in an email.


The portfolios, lacking ISAT information, were distributed to 132 clinicians who managed patients with ruptured aneurysms as well as to the 629 members of the cerebrovascular section of the American Association of Neurological Surgeons. The judges then chose whether each case was a candidate for surgical or endovascular management, indicated their level of confidence in the choice of treatment on a 0-10 scale, and indicated whether they would include the patient in a randomized trial in which both treatments are compared. In addition, 11 clinicians were asked to respond twice at least 1 month apart.


In total, 85 clinicians, including 58 cerebrovascular surgeons, 21 interventional neuroradiologists, and six interventional neurologists answered the questionnaire. Overall, endovascular management was chosen more frequently over surgical management, comprising 2,136 of 3,570 answers (58%). Clipping was selected more frequently for the treatment of non-ISAT aneurysms (50.8%), compared with ISAT aneurysms (26.2%; P = 0.0003).


Interjudge agreement was only fair, with a kappa score of 0.210 (95% CI 0.158-0.276) for all cases and judges. Despite this, judges typically reported high confidence levels in their decision-making, with mean confidence score > 8 for all cases. Agreement was no better within subgroups of clinicians with the same specialty, years of experience, location of practice, or across capability groups.


Interjudge Agreement by Specialty and Experience


Kappa Value (95% CI)

0-5 Years’ Experience

0.203 (0.138-0.293)

6-10 Years’ Experience

0.266 (0.206-0.335)

> 10 Years’ Experience

0.183 (0.131-0.265)

Cerebrovascular Surgery Specialty

0.207 (0.147-0.288)

Interventional Neuroradiology Specialty

0.234 (0.174-0.309)

Interventional Neurology

0.250 (0.118-0.388)


Interjudge agreement was similarly poor for both ISAT (kappa = 0.100; 95% CI 0.037-0.183) and non-ISAT aneurysms (kappa = 0.189; 95% CI 0.100-0.307).


“One interesting finding was that, among non-ISAT aneurysms, which represent aneurysms for which there is currently no randomized evidence, surgical clipping was the dominant choice selected for more than half of patients,” noted Raymond. “While we might have thought that physicians would intuitively extrapolate the results of ISAT and prefer endovascular treatment for a wider spectrum of aneurysms/patients, we found that surgical clipping remains a preferred option for many physicians in many cases.”


A Call for Randomized Trials


“This study is, to our knowledge, the first study focusing on treatment decision reliability in the case of ruptured aneurysms,” Raymond said. “This study demonstrates the lack of agreement among physicians for the optimal management of patients with ruptured aneurysms, likely due to the lack of randomized controlled studies on this subject.”


He pointed out that the current lack of consensus indicates that random allocation of treatment in the context of a clinical trial is a reasonable approach, adding that the treatment selected for any individual patient would currently likely differ based on which specialist manages their care.


“Optimal care is care that has been validated by showing better clinical outcomes in randomized pragmatic trials,” Raymond explained, “because consensus, without validation, is dangerous. It could mean that physicians are deluding themselves and acting more like choir members than as rigorous doctors—choosing to sing the same song, but they may collectively be doing harm. This has repeatedly happened in the past.” Optimized care, he continued should be validated in a randomized trial or a nonvalidated approach delivered in the context of a randomized trial.


Raymond pointed out that one randomized trial, ISAT-2, is already in recruitment. “We urge our colleagues to acknowledge the scientific gap in this field and to join efforts for the recruitment of more patients with ruptured aneurysms in ongoing and future randomized trials,” he concluded.



Darsaut TE, Fahed R, Macdonald RL, et al. Surgical or endovascular management of ruptured intracranial aneurysms: an agreement study. J Neurosurg. 2018;Epub ahead of print.



Raymond reports no relevant conflicts of interest.