A protocol utilizing computer-aided detection (CAD) to perform real-time assessment of fluid-attenuated inversion recovery (FLAIR) noncontrast images reduced unnecessary gadolinium use in patients with multiple sclerosis (MS).
The multidisciplinary effort piloted at Hospital of the University of Pennsylvania in Philadelphia, paired in-house CAD software and 3D lab radiology technologists to identify whether or not patients with MS would benefit from contrast injection.
The method spared nearly 90% of patients from contrast and additional imaging sequences, reported authors of a May 16 study published in the Journal of the American College of Radiology.
“Patients with MS routinely undergo serial contrast-enhanced MRIs,” wrote Jeffrey D. Rudie, MD, PhD, of Penn Medicine’s Department of Radiology, and colleagues. “Given concerns regarding tissue deposition of gadolinium-based contrast agents (GBCAs) and evidence that enhancement of lesions is only seen in patients with new disease activity on noncontrast imaging, we set out to implement a prospective quality improvement project whereby intravenous contrast would be reserved only for patients with evidence of new disease activity on noncontrast images.”
Before implementing their protocol, Rudie et al. performed a retrospective analysis of 138 patients with MS follow-up imaging to see if it was possible to determine who would and would not benefit from IV contrast. Of those evaluated, 24% had new or enlarging lesions on FLAIR imaging; those patients were only identified within the group of patients with new disease activity found during FLAIR imaging. The researchers concluded 76% of those undergoing follow-up MRs had stable disease and no enhancing lesions, proving it was possible to identify who would benefit from IV contrast and who wouldn’t.
Rudie and colleagues conducted a two-month trial period, imaging 153 patients under the new protocol. When using the CAD software, the technologists identified patients with new or enlarging lesions on FLAIR images with 95% accuracy and 97% negative predictive value compared to neuroradiologists’ interpretations.
Importantly, contrast and additional imaging sequences were avoided in 87% of patients.
“The real-world performance of these preliminary assessments, although imperfect, showed high specificity for appropriately selecting patients that would likely receive little benefit from undergoing contrast-enhanced MR sequences,” the researchers concluded.
The new approach not only improved patient care, but allowed for “significant” savings for the healthcare system, spared costs associated with using contrast agents, time to administer contrast and time for image generation and interpretation, the authors noted.
A majority of radiology departments do not use CAD or AI to help interpret images, which makes it harder to compare two 3D MRI time points. But as AI becomes more integrated into radiology, using such tools may become a feasible solution, according to Rudie et al.
“These results are still considered preliminary given the relatively small time period, thus continued careful evaluation of the advantages and disadvantages of this protocol on patient outcomes is warranted,” the authors concluded.
Matt joined Chicago’s TriMed team in 2018 covering all areas of health imaging after two years reporting on the hospital field. He holds a bachelor’s in English from UIC, and enjoys a good cup of coffee and an interesting documentary.