Spotting Stroke Markers Helps Radiologists Ward off Risks for Future Adverse Events
White matter hyperintensities, microbleeds, infarcts and vessel wall features provide useful clues
In patients who have had a stroke, common imaging findings aren’t only useful to confirm a diagnosis, they could also make a difference in identifying risk for dementia or preventing another stroke in the future.
“Imagine it’s a regular day at the workstation, and you’re reading a brain MRI on a patient who came to the emergency department with stroke-like symptoms,” said Hediyeh Baradaran, MD, MS. “You did not see an acute infarct on the diffusion-weighted imaging, but you see this focal area of encephalomalacia and gliosis in the left thalamus—of course compatible with a chronic lacunar infarct.”
“Obviously, this is a very commonly encountered finding—something we see every day,” Dr. Baradaran continued. “But what if, instead of the thalamus, you see an area of encephalomalacia and gliosis in the right parietal cortex. Does this change how we think about the case?”
Dr. Baradaran is an assistant professor of neuroradiology in the Department of Radiology and Imaging Sciences at the University of Utah in Salt Lake City. She was recently named chief of the Division of Neuroradiology for the Department of Radiology at the Columbia University Vagelos College of Physicians and Surgeons in New York City and will assume her new role in September.
Dr. Baradaran explored the question of diagnostic reasoning during RSNA 2024, moderating an educational course focused on critical findings that can help classify stroke and associated risks.
In addition to assessing infarcts, Dr. Baradaran noted that radiologists should also pay close attention to other imaging markers that may predict risk, such as white matter intensities.
Assessing White Matter Hyperintensities
According to Dr. Baradaran, white matter hyperintensities of presumed vascular origin typically appear in the deep white matter, and they’re a quintessential manifestation of cerebral small vessel disease.
When grading the extent of these hyperintensities, most radiologists tend to use subjective classifiers such as “mild,” “moderate,” and “severe/marked/extensive.” Increasingly, evidence demonstrates that the more “severe” categories are associated with negative outcomes such as dementia and stroke.
Radiologists can familiarize themselves with some qualitative assessment tools that use visual rating scales, and with qualitative methods that employ semi-automated and automated techniques, Dr. Baradaran noted. The visual rating scales demonstrate high inter- and intraobserver reliability and correlate well with the volumetric assessments.
Using a consistent method to grade the burden of white matter hyperintensities can be an important means to identify patients at higher risk for future stroke or dementia, Dr. Baradaran summarized.
White matter changes are just part of the story. Microbleeds and covert infarcts—often subtle and crucial findings—offer additional clues to a patient’s future risk.
Tracking Microbleeds and Infarcts
As the use of quantitative susceptibility-weighted imaging becomes more widespread, radiologists are more commonly encountering another marker of cerebral small vessel disease: cerebral microbleeds.
While it might not be feasible to use standard grading scales in a clinical setting for MRI reports, Dr. Baradaran recommends that radiologists record the number of bleeds. “This is becoming increasingly critical, as many guidelines are based on the number of microbleeds present,” she said. “Then, consider their location. Primarily lobar cerebral bleeds are more likely associated with cerebral amyloid angiopathy as opposed to deep lesions, which are more likely associated with hypertensive microangiopathy. We also want to describe whether it’s supratentorial versus infratentorial.”
Dr. Baradaran added that the finding of superficial siderosis, especially when seen adjacent to the cortex or at the vertex, and in focal locations without a history of subarachnoid hemorrhage, is particularly important to note. “It is a very strong predictor of future events in cerebral amyloid angiopathy, with increased risk of functional decline and dementia,” she said.
Reports of covert brain infarcts can represent a source of frustration for clinicians who aren’t sure what to do with the information, Dr. Baradaran acknowledged. But the presence of these brain infarcts may represent a five-fold increased risk of recurrent stroke.
For radiologists, the location of the covert brain infarction is key, she said. If present in the deep gray or white matter, it’s likely another manifestation of cerebral small vessel disease. However, when covert infarctions are found in the cortical gray matter, it is critical that radiologists search for a potential stroke source.
Vessel Wall Imaging in Stroke Workup
Intracranial vessel wall imaging can reveal hidden sources of ischemic stroke and guide patient management.
In an RSNA 2024 presentation, Jae W. Song, MD, MS, associate professor of radiology at the University of Pennsylvania in Philadelphia, described how T1-weighted turbo spin echo with variable flip angle pulse sequences can be optimized to identify atherosclerotic plaque, especially in cases of cryptogenic stroke.
According to Dr. Song, different vendor products have built-in protocols for vessel wall imaging, but these should be optimized for high spatial resolution—ideally at 0.5 millimeters—and intraluminal flow suppression.
“Atherosclerosis is a systemic disease, and it presents everywhere in the body,” Dr. Song explained. “Embolic mechanisms can occur, for instance, from the aortic arch, the cervical carotid arteries, or the intracranial arteries. Given the small caliber of the intracranial arteries, pulse sequences performed to evaluate the vessel wall need to be tailored for the relevant anatomy.”
When a patient with a focal neurologic deficit has an ischemic stroke, the role of the radiologist is not only to identify the diffusion restricting ischemic stroke but also to evaluate whether there is a potential etiology visible on imaging, she said.
In a meta-analysis performed at the Hospital of the University of Pennsylvania, Dr. Song and her colleagues found that plaque enhancement, positive vessel wall remodeling, vessel wall T1 hyperintensity, and plaque surface irregularity were associated on vessel wall imaging with downstream ischemic stroke, suggesting that these findings are potential imaging biomarkers of culprit intracranial plaque.
These features on vessel wall MR imaging can increase the vascular neurologist’s diagnostic confidence about intracranial plaque being the potential stroke etiology and manage the patient appropriately.
Vessel wall imaging is particularly useful for detecting non-stenotic or mildly stenotic intracranial plaques that may be missed on traditional angiographic studies, Dr. Song said. In cases of plaques with positive wall remodeling, the vessel expands outward to accommodate plaque burden while preserving the lumen and masking the disease on CT or MR angiography. By directly imaging the vessel wall itself, radiologists can glean valuable diagnostic information that may otherwise be occult on angiographic imaging.
“Vessel wall imaging is also valuable for detecting intracranial atherosclerotic disease at perforator origins, as seen in branch atheromatous disease,” she added. “This condition commonly involves the basal ganglia via lenticulostriate arteries from the middle cerebral artery or the pons via basilar artery perforators. By directly visualizing focal plaques that may compromise the origins of the perforators, vessel wall imaging offers a key diagnostic advantage over luminal imaging alone.”
When radiologists incorporate vessel wall imaging into their clinical protocols, the benefits are validated by measurable feedback, Dr. Song said. She and her team at the Hospital of the University of Pennsylvania established a Vessel Wall Imaging Program, with implementation steps published in the American Journal of Neuroradiology. Its impact was measured by a clinician survey, where 90% agreed with the statement, “Vessel wall imaging has changed and helped my clinical diagnostic confidence for diagnosing and managing my patients.”
The RSNA 2024 course also featured a perspective on embolic strokes of undetermined source from Hooman Kamel, MD, a vascular neurologist at Weill-Cornell imaging in New York City, and a dive into standardized risk reporting on extracranial carotid plaque and the Carotid Plaque-RADS classification by Tobias Saam, MD, from DIE Radiologie in Munich.
“Even though we encounter these imaging findings every day, we have to remember that they should be described in our reports, because they have very important clinical implications,” Dr. Baradaran said.
For More Information
Access the American Journal of Neuroradiology article.
Read previous RSNA News stories on stroke imaging:
- Research Highlights the Sex and Gender Disparities in Acute Ischemic Stroke
- Stroke in Young Adults: Occurrences on the Rise
- Deep Learning Assists with Stroke Evaluation and Management