Structured reporting improves accuracy of heart disease detection on non-cardiac CT

Implementing structured reporting templates for standard chest CT exams does not result in an increase in false-negatives and can improve the accuracy of reporting for coronary calcification and aid in early detection of coronary heart disease, according to study results published in the Journal of the American College of Radiology.

Heart disease continues to be one of the world’s leading causes of death, even in relatively young patients with no symptoms or major health issues. Coronary artery calcification, which is an indicator of the presence of heart disease, is often underreported on noncardiac CT, according to William Walter, MD, and his colleagues at the Montefiore Medical Center in New York. “Considering that coronary calcium scoring is a powerful risk stratification tool, if underreported, calcification may represent a missed opportunity to trigger the diagnosis and management of cardiovascular disease,” wrote Walter et al. “Conversely, reporting of such an incidental finding may trigger evaluation for CAD, allowing early lifestyle modification or pharmacotherapy to reduce the risk for cardiac events.”

Walter and his research team set out to determine whether including a field for coronary artery calcification on non-contrast chest CT structured templates would improve the accuracy of reporting regarding artery calcification. To do so, they mandated use of such template within the radiology department at their medical facility, which included the statement “There are no coronary artery calcifications.” They collected all noncardiac, noncontrast chest CT exams reported over a three-day period one month after the implementation of the new structured reporting template and compared them with a control group of reports representing a three-day period from one year prior to implementation. The reports were analyzed and designated either positive or negative for coronary calcifications, and CT images were scored for the presence or absence of coronary calcifications by a consensus of two radiologists.

Their results showed that reports obtained using the new structured template were more likely (96 percent) to accurately detect both the presence and absence of coronary artery disease than those from the control group (86 percent). Additionally, reports generated with the structured template were less likely to contain false-negative results (4 percent) than control group reports (12 percent).

Walter and his colleagues believe their findings should inspire radiologists who are tasked with creating standardized reports for a facility or practice to incorporate value-based fields for collecting additional imaging data as part of a structured template. “Structured radiology reporting promises to improve report quality because items are described in a regular and predictable order,” the researchers wrote. “Fields for entry of specific data within a structured report have the potential to direct radiologists’ search patterns and facilitate consistent reporting of findings that would not otherwise have elicited specific comment.”