The human body contains a whopping 60,000 miles of blood vessels, and all of those tubes contain useful details about how well the body is functioning. For certain disease states, such as brain tumors, the size, shape, and number of vessels can reveal critical data on a tumor's progress or regression.
Describing the body's blood distribution system isn't easy. Angiography, the most common technique used to image vessels, only provides pictures (angiograms) of the vessels. It is difficult to obtain quantitative data from angiograms (obtained by injecting a dye and taking images with x-rays, computer tomography, or magnetic resonance). However, by combining magnetic resonance angiography (MRA) with a set of powerful algorithms (mathematical equations), a research team led by Elizabeth Bullitt has hit on a method that defines blood vessels by describing their number and shape. The approach could lead to a noninvasive method of determining whether a tumor is malignant and of tracking how a tumor responds to treatment. It also may yield clues on how the brain ages.
"Blood vessels can provide an amazing amount of information," says Bullitt, the Van L. Weatherspoon Jr. Distinguished Professor of neurosurgery and head of the Computer-Assisted Surgery and Imaging Laboratory at the University of North Carolina, Chapel Hill. Bullitt has worked for over a decade to extract meaningful information from the often stunning images of blood vessels. She has been particularly interested in how cancer affects blood vessel networks because cancer can cause blood vessels to undergo tremendous changes. Although many groups have studied how to define vessels from three-dimensional images, Bullitt’s approach gives flexibility when assessing the brain’s blood vessel network. It can provide quantitative details for blood vessels over the entire brain, over a large or small region of interest, or over a connected set of vessels.
Homing in on Wiggle Patterns
Bullitt analyzes ultrahigh-resolution three-dimensional images with specially designed software to make a computer model of the vessels that is so specific that she can count the number of vessels in a given location, map how vessels connect in their branching structures, and home in on each vessel’s tortuosity – its wiggliness – a prominent characteristic in disease states.
Tortuosity, or wiggliness, refers to the vessel's shape pattern rather than vessel movement. "It’s like looking at a snake in formaldehyde. The snake is dead but still retains its 'S' pattern," she explains. Smooth blood vessels that resemble cooked spaghetti are associated with healthy tissue. Vessels in and around diseased tissue are a different story. These vessels exhibit different patterns of wiggliness. Vessels associated with cancer, for instance, possess a high degree of wiggliness. "In cancer, the vessels are jagged and quite irregular. Cancer is at work doing something to the vessel wall," Bullitt says.
Tracking Squiggly Vessels
In several different studies, Bullitt has assessed her computer-assisted MRA technique's ability to determine tumor malignancy and track a drug therapy's effect on tumor growth. "We’ve known for a long time that tumors have squiggly vessels," says Keith Smith, an associate professor of radiology at UNC Medical School and a Bullitt collaborator. The algorithm-enhanced MRA technique is "a way to put numbers on those squiggles."
In one study of 30 hard-to-diagnose cases, Bullitt’s technique correctly identified whether a tumor was malignant in 29 of the cases. "These cases, which were all scheduled for gross tumor resection, were all scanned before surgery and included some really difficult cases where the presence of a malignancy could not be determined using conventional imaging," says Bullitt. Two of the tumors were just 0.3 cm3, about the size of a coffee bean.
Bullitt's technique also may give clinicians a more quantitative way of deciding whether a tumor is responding to treatment. In a recently completed study with Dr. David Reardon, associate deputy director of The Preston Robert Tisch Brain Tumor Center at Duke University Medical Center, Bullitt used her technique to assess tumor progression in patients with recurrent glioblastoma, a highly aggressive brain tumor. Patients were first treated with the drug Avastin, which attacks the agent that promotes tumor blood vessel growth. Conventional contrast imaging studies showed that each patient's tumor was shrinking because very little brightness was seen in the tumor area. When Bullitt analyzed the images with her MRA technique, she found that only one-half of the tumors were actually responding to treatment. In the other patients, she found the vessels' jagged patterns became more profound, signaling a lack of response.
In another multicenter study, Bullitt and her colleagues discovered that vessel wiggliness can predict treatment success or failure about 1–2 months sooner than traditional imaging methods. The study involved 22 women being treated for brain tumors as a result of breast cancer.
"One challenge in treating brain tumors is that you don’t have a lot of time," says Smith. "It’s not okay to wait six months to see if a treatment is working. You need to figure out if a treatment is working and then if it isn’t, move as quickly as possible to something else. The ability to analyze a tumor with the touch of a button would help a lot."
Brain Vessel Wiggliness Increases With Age
Bullitt's research may also give some insight into how the brain ages. Her group has created a database (available at http://hdl.handle.net/1926/594) that contains the MRAs of 100 healthy volunteers. Divided into 5 age groups with 20 patients per decade and an equal distribution of men and women, the data reveals that older brains have wigglier vessels than younger brains. The researchers also discovered that, in older brains, smaller vessels tend to stop functioning over time. In addition, more pronounced wiggle patterns are often associated with conditions such as high blood pressure and diabetes.
Bullitt’s technique offers radiologists another tool to assess healthy and diseased tissue. Whether an aging brain, a cancerous tumor, or a brain condition such as Alzheimer’s, the technique provides a noninvasive way to gather meaningful data on healthy and disease states. Because changes in blood vessel networks may directly affect or be caused by disease progression, accurately measuring those changes may make a difference in patient outcomes.
This work is supported in part by the National Institute of Biomedical Imaging and Bioengineering grant 5R01EB000219-11.
Bullitt E, Rahman FN, Smith JK, Kim D, Zeng D, Katz LM, Marks BL. The effect of exercise on the cerebral vasculature of healthy aged subjects as visualized by MR angiography. Am J Neuroradiol. 2009 July 9. [Epub ahead of print]
Bullitt E, Zeng D, Ghosh A, Aylward SR, Lin W, Marks BL, Smith K. The effects of healthy aging on intracerebral blood vessels visualized by magnetic resonance angiography. Neurobiol Aging 2008 May 7. [Epub ahead of print]
Bullitt E, Lin NU, Smith JK, Zeng D, Winer EP, Carey LA, Lin W, Ewend MG. Blood vessel morphologic changes depicted with MR angiography during treatment of brain metastases: A feasibility study. Radiology. 2007 Dec;245(3):824–30.