Obesity and Heart Disease Link Spurs Structured Reporting Ideas
A study on the correlation between obesity and cardiovascular disease gave RSNA Research & Education (R&E) Grant recipient Marta Heilbrun, M.D., a better understanding of what imaging can convey beyond the original hypothesis and led her to a structured reporting initiative to track and quantify such outcomes.
 Marta Heilbrun, M.D. |
Dr. Heilbrun used a 2005 RSNA Presidents Circle Research Resident Grant to fund her study, “A Nonhuman Primate Model for Quantitative CT Volumetric Measures of Obesity and Cardiovascular Disease Risk Factors,” which demonstrated the feasibility and ease of performing high-volume CT imaging of the entire body in a relatively large number of non-human primates. The study was the first to show a significant correlation between CT-measured hepatic steatosis and atherosclerotic plaque at histology in any species. The research, which has not yet been published, was completed at Wake Forest University in Winston-Salem, N.C.
Dr. Heilbrun examined volumetric measures of abdominal fat compartments using whole-body multidetector CT scans of cynomolgus monkeys taken from a larger trial investigating the link between depression and cardiovascular disease risk. Results of Dr. Heilbrun’s research showed the monkeys had almost no CT demonstrable plaque.
“Yet at histology, there was clearly plaque in the specimens,” said Dr. Heilbrun, now a professor in the Department of Radiology at the University of Utah in Salt Lake City. “We found that they had varying amounts of visceral fat and varying amounts of subcutaneous fat compared to cardiac fat and liver fat.
“Factors that we hadn’t even thought about were part of the cardiovascular disease risk profile,” she said. “This was an incredible opportunity to understand much more about the spectrum of disease and what an image might tell us, separate from the original hypothesis of the experiment.
“It was valuable to realize that a multitude of questions can be asked and answered each time an image is obtained—especially in non-human primates with a controlled subject and controlled exposure, which is one of the problems with human epidemiologic studies,” she added. “Our exposures are so heterogeneous that it’s very hard to sort the signal from the noise.”
Future studies will be able to use CT to measure intermediate endpoints instead of sacrificing part of the study population or study animals, Dr. Heilbrun said.
More Standardized Data Sought
Her experience using monkey scans from a separate, larger study led Dr. Heilbrun to an initiative to identify more standardized information to be taken from images and reported for use by researchers. She became a member of the RSNA Structured Reporting Committee established to create clear and consistent report templates that contain reusable structured data.
“If I can get quantifiable and reliable data out of the reports we generate, we will be much more likely to be able to actually track and measure outcomes and determine what happens with patients when they interact with imaging.
“There was a lot more information there, once I started actually looking at the images and measuring,” Dr. Heilbrun said. “Right now we generate reports in clinical practice and there is a lot of information about the state of health in America that we don’t know how to measure. Maybe a structured report is a way to start measuring that.”
Along with her cost-effective analysis and outcomes-based research, Dr. Heilbrun’s work on structured reporting is especially significant to radiology—and research in general, said Kathryn A. Morton, M.D., a professor of radiology at the University of Utah and a long-time mentor to Dr. Heilbrun.
“Structured reporting is critical because it allows for combining or comparing radiology data from multiple institutions,” said Dr. Morton.
“This is critical for our profession, yet few people have the expertise necessary to accomplish this. Dr. Heilbrun has that expertise, acquired through her graduate program and a long-standing interest in medical informatics dealing with outcomes-based research.”
Evaluating Image-Guided Biopsy in Renal Cell Carcinoma
Since arriving in Utah in 2007, Dr. Heilbrun has been working on a paradigm to systematically evaluate the impact of imaging-based technologies on patient outcomes—an inquiry that she said has prompted her to investigate what is really on patients’ minds.
The first step was learning to use statistical modeling, decision analysis and cost-effectiveness tools to evaluate the role of image-guided percutaneous biopsy on outcomes in renal cell carcinoma. The research is funded through a GE-Association of University Radiologists Radiology Research Academic Fellowship that provides $70,000 for each of two years.
The work is based on the hypothesis that adding percutaneous biopsy to the diagnostic workup of renal masses that meet the imaging criteria of a T1a tumor—four centimeters or smaller and limited to the kidney without spread to lymph nodes or distant organs—will be cost-effective by preventing patients from undergoing unnecessary treatment, said Dr. Heilbrun.
“The current standard of care is not to do a biopsy except in certain indications, such as when a procedure is not going to yield a final whole specimen or there is significant risk involved with surgery,” said Dr. Heilbrun. “I am trying to understand if we should expand the biopsy to more patients and how to demonstrate whether it would be beneficial if we did so in terms of this disease.”
The project hinges on understanding a patient’s perspective once diagnosed with a particular disease, she said.
“A lot of it comes down to the willingness of the patient and caregiver to accept some uncertainty,” said Dr. Heilbrun. “If the patient is willing and there is a high risk associated with surgery, then the biopsy is probably a good choice. If the patient is unwilling to accept uncertainty, then neither watchful waiting nor doing a biopsy is a particularly good choice.
“If we develop a new imaging test, I want to understand how good would it have to be and what kind of additional information would be necessary in order to really change the outcome of patients with this disease,” she said. “That is why we are trying to quantify uncertainty about patient behaviors in order to inform the development of new tests.”
One limitation to her research lies with the patients themselves. Dr. Heilbrun said it could be difficult to find patients willing to undergo a new imaging test since someone with a renal mass may want it removed whether or not it is cancerous.
“Then there is no reason to develop a new MR imaging technique that can differentiate whether it is cancer or not,” she said. “That is what is driving my next question. I really need to start asking patients and real people how they would deal with this information.”