Study: CT Alone Noninferior to CT Plus MRI for Stroke Outcomes

Marlene Busko

July 28, 2022

Among patients with acute ischemic stroke, diagnostic imaging with CT alone was noninferior to initial CT plus MRI for discharge and 1-year clinical outcomes, in a new study.

The rates of death or dependence at hospital discharge, and of recurrent stroke or death at 1 year, were not worse in the patients who only had a cranial CT scan.

Therefore, "the value of MRI added to CT in patients such as these should not be presumed," Heitor Cabral Frade, MD, University of Texas, Galveston, and colleagues write in their study, published July 21 in JAMA Network Open.

The addition of MRI to CT has greatly increased, but it is not clear if the added MRI, which is more expensive, improves outcomes, senior author William J. Powers, MD, University of North Carolina, Chapel Hill, explained to theheart.org | Medscape Cardiology .

From 1999 to 2008, the use of MRI to evaluate patients with ischemic stroke in the United States increased from 28% to 66%, and more than 90% of patients who had a brain MRI first had a CT scan, the researchers note. Unnecessary medical imaging is a major cause of preventable waste in the American healthcare system.

"Many physicians believe that more data lead to better patient outcomes, but that's not always true," said Powers. With MRI, "you see more stuff and you make decisions based on that, but does that mean people do better? That's the implicit assumption, but that's not always true."

When you come up with a new diagnostic test, he continued, unlike with a new drug, you don't have to show the US Food and Drug Administration (FDA) that using it improves patient outcomes.

"Maybe [this study] will get people to think that we really do need more data and more research to determine which patients hospitalized with acute ischemic stroke benefit from MRI in addition to initial CT," he said.

"Pause and Reconsider"

"Given the pervasiveness of routine MRI in addition to CT in clinical stroke practice, the implications" of this study are "substantial," Michael Teitcher, MD, and Jose Billar, MD, from Loyola University, Chicago, write in an accompanying editorial.

"As stewards of healthcare resources, clinicians should be asking whether the additional information provided by diagnostic tests meaningfully affects patient outcomes," they advise, and "the answer should be data-driven rather than anecdotal."

There are circumstances in which additional MRI is still justified, Teitcher and Billar acknowledge. "But at a minimum, these results should give the healthcare practitioners reason to pause and reconsider routine use of CT plus MRI."

"Hopefully, the present study paves the way for future prospective studies that would provide additional data on this common clinical question," the editorialists write, echoing Powers.

Current American Heart Association/American Stroke Association guidelines state that it's reasonable to obtain additional MRI after initial head imaging in cases where initial imaging did not demonstrate infarction.

Some researchers and practitioners recommend that all hospitalized patients with acute ischemic stroke undergo brain imaging with MRI, Billar told theheart.org | Medscape Cardiology in an email. This "may help in differentiating ischemic stroke subtypes (for example, large artery extracranial and intracranial atherosclerotic disease, cardioembolic disease, lacunar, and small vessel disease) within the continuum of ischemic cerebrovascular syndromes."

However, whether this imaging paradigm is associated with improved patient outcomes, he continued, remains unsubstantiated by either consensus or evidence review.

The routine use of brain MRI in addition to CT among hospitalized patients with acute ischemic stroke "requires verification in properly designed clinical trials," Biller said, adding, "let the data speak for itself!"

"In the meantime," he said, "it would be timely and sensible to rethink when to order brain MRIs for hospitalized patients with acute ischemic stroke."

Propensity-Matched Patients

For the propensity-score-matched study, 246 patients with acute ischemic stroke admitted to University of North Carolina Hospitals Comprehensive Stroke Center between January 2015 and December 2017 and were imaged with either initial CT alone or CT plus MRI.

Patients were classed as having dependence at hospital discharge if they had a modified Rankin Scale score of 3 to 6 (where 3 indicates needing some help but able to walk unassisted, and 5 indicates need for constant nursing care and attention and being bedridden and incontinent). Median age of the study participants was 68 years, and 53% were male.

Among the 123 patients with additional MRI, 42.3% of tests were ordered under the supervision of attending neurologists, 33.3% under the supervision of attending emergency physicians, and 24.4% by nurse practitioners or neurocritical care attending physicians.

Of the six attending neurologists caring for people with stroke during the study period, one never requested an MRI, another always requested one, and the others were in between.

For 111 of the 123 MRIs, there was no specified indication other than stroke or neurologic symptoms.

Death or dependence at hospital discharge occurred more often in patients who had MRI added to CT than in patients who had CT alone (48.0% vs 42.3%), which met the –7.5% margin for noninferiority.

Similarly, stroke or death in the year after discharge occurred more often in patients who had both types of imaging than in patients who had CT alone (19% vs 13%), meeting the 0.725 margin for noninferiority.

"Consider What Value It Will Add"

Bruce C.V. Campbell, PhD, Royal Melbourne Hospital, Australia, told theheart.org | Medscape Cardiology that at his center, "we order MRI selectively, in perhaps 20% to 30% of patients."

"We also often do diffusion-only MRI to characterize the infarct," said Campbell, who authored a second editorial that accompanies the article.

"We routinely do CT, CT-perfusion, and aortic arch to cerebral vertex CT-angiography, so we have a lot of vascular information already," he continued.

"The diffusion MRI," he explained, "confirms the diagnosis, indicates infarct volume (which is useful when considering timing of anticoagulation), provides hints to mechanism [such as] small vessel disease, cardioembolism if multiterritory infarcts, watershed patterns, and confirms whether a carotid stenosis is likely to be symptomatic."

"Like any investigation, it's good practice to consider what value it will add to [patient] management decision-making," Campbell summarized. "There are many situations where MRI is valuable after stroke, but it's not needed for everyone."

The authors and editorialists report having no relevant financial disclosures.

JAMA Netw Open. 2022;5:e2219416, e2223077, e2223074. Full text, Teitcher and Billar editorial, Campbell editorial

For more from theheart.org | Medscape Cardiology, follow us on Twitter and Facebook.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
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.

processing....