August 2014

Improving Stroke Assessment
By Kathy Hardy
Radiology Today
Vol. 15 No. 8 P. 26

One study suggests MRI may be a better predictor of when to use clot-busting drugs than CT. More data are needed, and there are practical issues to consider about MRI’s role in stroke assessment.

The time window during which the IV clot-busting drug tissue plasminogen activator (tPA) can help ischemic stroke patients may be gaining some clarity in light of research from the Johns Hopkins University School of Medicine. Researchers there developed a new technique utilizing MRI was able to identify with 95% accuracy which patients benefited from tPA treatment and which suffered potentially lethal brain bleeding.

“This technique reanalyzes standard MRI to determine how much contrast is leaking into the brain during the procedure, which is a sign of damage to the blood-brain barrier,” says Richard Leigh, MD, an assistant professor of neurology and radiology at Johns Hopkins. “Further tests of this method could form the basis of the expanded and more precise use of tPA.”

Leigh, the study’s lead investigator, says the researchers’ findings present a new technique for more accurately predicting which stroke victims may benefit from tPA, raising the question of whether CT, the current gold standard for immediate stroke treatment, or MRI should be the go-to imaging modality for stroke. The study reveals what could be movement toward a revised standard of care for ischemic stroke patients.

The time window for using tPA in stroke treatment, the only FDA-approved treatment for ischemic stroke, is at its best up to three hours but not more than 4.5 hours after a patient shows signs of stroke. After that, there may be too much damage to the blood-brain barrier and because the drug can cause bleeding in the brain those patients currently go untreated. Imaging measures the degree of damage in the completed stroke meaning the area where tissue may be ischemic to the point of infarction, making it more likely to suffer complications when treated with medical therapy or IA catheter therapy. The Johns Hopkins study findings suggest that MRI may be better for this measurement than CT because it is more specific for identifying damaged areas where tPA could pass through the blood-brain barrier.

 “We can look at the blood-brain barrier with MRI,” Leigh says. “That barrier can be damaged with stroke, as the tissue becomes ischemic due to blood flow issues. The barrier becomes open, and that opening can enlarge over time. This is the danger with tPA application. We knew this prior to our study, but no one was able to show the difference between what happens when there is a little damage to the barrier and when there is large damage to the barrier.”

Leigh says that in most cases, physicians order CT scans of stroke patients before administering tPA to determine whether they have any visible bleeding. In contrast, his computer program, which works with an MRI scan, can find subtle changes to the blood-brain barrier that may otherwise not be detected. This method allows physicians to see how much gadolinium leaks into brain tissue from surrounding blood vessels.

In the course of the study, by quantifying such brain-barrier damage in 75 stroke patients, Leigh identified a threshold for determining how much leakage was dangerous. After that, he and his team applied this threshold to those 75 records to determine how well it would predict who had or had not suffered a brain hemorrhage. The new test correctly predicted the outcome with 95% accuracy. “If these findings are supported by ongoing studies, we should probably be doing MRI scans in every stroke patient before we administer tPA,” Leigh says.

Time is one major obstacle to the adoption of MRI in ischemic stroke treatment. As Leigh notes, the longer it takes for a stroke patient to be treated, the less likely the patient will experience a full recovery. He acknowledges that, in most cases, an MRI scan does take longer to conduct than a CT scan. In addition, he says CT is more widely available. However, he still is a proponent of weighing the benefits before reaching a definitive conclusion. “If the benefits of getting tPA into the right people, as well as protecting the wrong people from getting it, outweigh the harms of waiting a little longer to obtain MRI results, then maybe doctors should consider changing the course of treatment,” he says. “CT doesn’t tell us as much, other than whether or not the patient is hemorrhagic. MRI can show us more specifically where the damage has occurred.”

Debates continue regarding the thinking process for which modalities play the best roles in timely stroke treatment. The ACR Appropriateness Criteria for Cerebrovascular Disease, last reviewed in 2011, refer to language in the FDA’s approval for tPA regarding MRI use for predicting the potential hemorrhagic complications that can occur with tPA therapy. However, the criteria go on to state that “there is currently insufficient widespread clinical experience to recommend MRI over CT” for determining whether it’s safe to use tPA or whether the patient has exceeded the window of opportunity for achieving benefits from the standard IV treatment for removing blood clotting.

MRI is “exquisitely sensitive to acute infarction within minutes” of the initial stroke, according to the ACR Appropriateness Criteria, with high numbers for both sensitivity and specificity. With this, this modality could help to identify “potentially salvageable ischemic tissue,” particularly once the patient is past three hours from symptom onset.

However, the criteria go on to state that “although MRI is potentially more sensitive than CT, it is not generally used as a substitute for CT to detect acute intracranial hemorrhage.” In addition, “Advanced MRI and other metabolic imaging techniques may improve acute cerebral ischemia evaluation and assessment of long-term disability but are not currently used in general clinical practice.”

The criteria continue, “It is also important to emphasize the issue of availability of MRI in the context of the 3-hour therapeutic window, the difficulty of managing medically unstable patients in the MRI machine, and potential contraindications: patients with pacemakers, cerebral aneurysm clips, ocular foreign bodies, or cochlear implants and those suffering from claustrophobia or morbid obesity.”

As a member of the ACR neuroradiology commission, Alexander Norbash, MD, MHCM, FACR, is part of a team that’s continually reviewing the best approaches for visualizing what’s happening with the brain of a stroke patient. Whether the tool is CT or MRI, he describes it as a delicate process. “The greater the amount of time that passes beyond initial stroke symptoms, the more likely there is irreversible and increasing damage and the less likelihood for successful treatment of that damage,” he says. “You may not be able to reestablish blood flow and reanimate the brain. Re-establishing flow to the brain is not simply an engineering problem; the brain is a unique biologic organ. As a stroke ages over a period of only hours, the blood vessels of the brain become progressively leakier, so that even if you do re-establish flow, once the vessels are leaking blood into the brain substance a bad situation becomes much, much worse. We need better techniques to see if there’s damage of the brain, increasing leakiness of blood vessels and whether or not the stroke is reversible. Otherwise, we run the risk of accidentally harming patients while trying to cure them.”

Norbash, a practicing interventional neuroradiologist and chairman and professor of radiology at Boston University Medical Center, says that physicians have mixed opinions about whether CT or MRI is the better imaging alternative in determining stroke damage to the brain. While many physicians believe that MRI is better than CT, there are challenges to the widespread adoption of this modality for that purpose. “MRI is not as easily available as CT,” he says. “There’s also the matter of what to do when the MRI machine is in use. Do you remove that patient? You also need to screen the stroke patient for metal. It’s easier to image with CT in everyday situations, but maybe MRI is worth the extended time and risk.”

Norbash says the decision could come down to how aggressive a physician wants to be with a patient’s treatment. “Strokes may appear to be identical, but this is a case where the challenge is to determine what’s best for each individual patient,” he says. “We all have a little different build when it comes to our collateral circulation and how our brains are irreversibly damaged following seemingly identical strokes.”

As an example, Norbash cites cases where a patient receives tPA within three hours, the blockage is opened, and blood flow resumes, yet the patient still doesn’t recover. “We need to try to understand how aggressive we can be in each individual patient since the stakes are so high,” he says.

Leigh says the Johns Hopkins study is just the first step in establishing what could be new guidelines for the treatment of ischemic stroke patients. He currently is analyzing data from patients who received other treatments for stroke beyond the typical window, in some cases several hours after the FDA-approved cutoff for tPA. He says some people who arrive at the hospital many hours after suffering a stroke could still benefit from tPA. “It will be important to replicate this study with a cohort of patients similar to those in the first study but different people,” Leigh says. “That would confirm that the threshold identified is not just a fluke.”

After that, a clinical trial would be beneficial, Leigh says. However, he notes the potential difficulties when it comes to establishing clinical trials for areas such as stroke treatment, where time is of the essence in the difference between permanent damage and even death. “Clinical trials in stroke treatment need to be done quickly,” he says. “You need to have good workflow with results in real time.”

Norbash adds that clinical trials involving stoke treatment need to be large and well designed and should involve the collection of large amounts of data in order for everyone to be on the same page when it comes to results. He says this is where IT developments in PACS and cloud storage can help advance the decision-making process in this area. “Every patient could be a learning opportunity if we were recording the appropriate data from each and every case,” he says. “We would have a better understanding of how the new techniques are working. Everyone wants to use the new and best tools available, but we don’t know the science. Learning from occasional and individual patients isn’t rather than all patients isn’t helping the overall field.”

Specialized training can bring even more clarity in which techniques are best for stroke treatment, Norbash adds, coming as more specialized training for physicians and fellowships in this field spreads from the large medical centers to facilities large and small throughout the country.

“The goal is to help patients in the community by developing methods to better identify patients who will benefit from timely and accurate stroke treatment,” Leigh says.

— Kathy Hardy is a freelance writer based in Phoenixville, Pennsylvania.