Secondary Logo

Brain Trust: CTP Improves Outcomes in Large Vessel Occlusion Acute Stroke

Marcolini, Evie MD

doi: 10.1097/01.EEM.0000512770.00506.95
Brain Trust

Dr. Marcolini is an assistant professor of emergency medicine and neurology in the department of emergency medicine and the division of neurocritical care and emergency neurology and the medical director of SkyHealth Critical Care Air Medical Transport at Yale University School of Medicine. Follow her on Twitter @eviemarcolini. Read her past columns at



A 68-year-old woman was shopping and suddenly fell, also experiencing right arm and leg weakness and left facial droop, according to her friend. She was brought to your emergency department by EMS, and a stroke code was activated en route. You and the stroke team calculate an NIHSS of 17, and she is whisked off to CT. What studies will you order? CT? CTA? CTP? How will a CTP help your decision-making, and how does it work?

The recent groundbreaking trials (MR CLEAN, REVASCAT, ESCAPE, SWIFT PRIME, AND EXTEND-IA) showed us that endovascular therapy can improve outcomes in patients with anterior large vessel occlusion acute stroke. How did they do this? They used better tools (stent retrieval devices), for one thing, but they also used functional imaging technology to figure out which patients would benefit most from revascularization. After years of failing to show that thrombectomy can improve outcomes, these tools helped us by narrowing the studied patient population with practice-changing results.

Your patient's CT shows no hemorrhage, and she gets tPA, with mild improvement in her symptoms. Will she benefit from mechanical thrombectomy?

CT perfusion (CTP) can show us which patients have a relatively small core infarct (dead tissue) and large penumbra (salvageable tissue). These are the patients who will benefit most from revascularization. Conversely, if the imaging shows a large core infarct and small penumbra, the risk of hemorrhagic conversion of that large infarcted tissue will outweigh the benefit of saving the relatively smaller volume of ischemic tissue with reperfusion therapy.

Perfusion imaging has been available to us in the form of diffusion-weighted MRI, but it's not very practical to obtain MR imaging in most settings in a timely fashion. CTP techniques are newer and more accessible to most emergency departments. We need to understand some basic definitions when looking at these images. Cerebral blood volume (CBV) is the total volume of blood in a given region of the brain, including the parenchyma and large vessels, measured in ml/100 g. Cerebral blood flow (CBF) is the volume of blood that moves through a given region of the brain in a minute, measured as ml/100 g/min. Mean transit time (MTT) is the average time for blood to get through a given region of the brain, measured in seconds. This is calculated by CBV/CBF. Tmax is the time it takes contrast to arrive at the tissue.

Back to Top | Article Outline

A Thousand Words

The brain does not store energy and requires a constant flow of glucose, using 20 percent of total cardiac output. Autoregulatory mechanisms keep CBF at 50-60 ml/100g/min; and if CBF drops, vasodilation and increased blood pressure kick in to protect against hypoxia and low perfusion. On a cellular level, CBF values less than 40ml/100g/min are considered oligemic and less than 10ml/100g/min are considered the threshold for irreversible ischemic damage.

If CBF drops, vasodilation and hypertension result in increased CBV and MTT as the brain's compensatory mechanisms go to work. If CBF gets as low as 20 percent of its normal levels, autoregulation fails, or the pressure from a large stroke can collapse cerebral arteries, and we see a reduced CBV as well as CBF. This constitutes infarcted tissue. If CTP shows a decrease in CBF with stable or increased CBV, this constitutes reversible ischemia, or penumbra.

If you are like me, a picture makes everything much easier to understand. Recently developed software takes all of these components, and with the help of measured time density curves (difference between arterial input and venous outflow), produces a graphic representation of our parameters.

The top panel in the photograph shows a patient with L MCA occlusion and a large area of reduced CBV and CBF with increased MTT and Tmax. This represents a core of infarction and penumbra of equal size, therefore not a great candidate for reperfusion. The bottom panel shows a patient with an R MCA occlusion, and a very small area of reduced CBV and CBF, with insignificant increase in MTT, but a significant increase in Tmax (perfusion delay). This represents a penumbra that is much larger than the core infarct. These images are processed to show graphic representations of core (pink) and penumbra (green) in the bottom images. Because the penumbra is large and the core is small, this patient would be a great candidate for reperfusion, with very little tissue at risk for hemorrhagic conversion but a lot of tissue that can be salvaged.

The technology has some limitations. Perfusion maps can be inaccurate in atrial fibrillation, severe proximal arterial stenosis, or poor cardiac output. Placement of the measurements for time density curves can also over- or underestimate MTT. Small infarcts have poor resolution, and seizures and other mimics can show false-positive hypervascularity. It's important to have neuroradiology experts interpret this imaging.

This technology has been tremendously helpful in determining which patients to select for mechanical thrombectomy, and will likely be helpful in the future with patients whose last normal times are unknown, such as the “wake-up” strokes. With tools like these, we continue to improve the outcome for many stroke patients in the emergency department with rapid assessment and treatment.

Share this article on Twitter and Facebook.

Access the links in EMN by reading this on our website or in our free iPad app, both available at

Comments? Write to us at

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.