Add Structured Reporting to Chest X-rays for Better COVID-19 Prediction

When interpreting chest X-rays taken on patients suspected of COVID-19 infection, using structured reporting can increase the chance that a provider will correctly predict and anticipate positive RT-PCR test results, a new study has found.

Among imaging exams, chest CT has been the most broadly used scan when providers think a patient might have the virus. But, that modality is not always readily available in every healthcare facility, and the 24-hour turnaround time for the RT-PCR test can sometimes be too long. That is why a team of investigators from the University College of Dublin Ireland suggested turning to the more ubiquitous X-ray and using structured reporting.

“Given that there are often constraints on resources, such as personal protective equipment, ventilators, and healthcare staff, it is important to optimize decision-making on whether to manage a patient in a COVID-19-specific pathway,” said the team led by Andrew Yates, M.D., a radiology resident with Mater Misericordiae University Hospital.

For more coverage based on industry expert insights and research, subscribe to the Diagnostic Imaging e-Newsletter here.

The team published their findings and recommendations in a Nov. 11 study published in the European Journal of Radiology. From their work, they found that patients who had positive imaging results had an 88-percent probability of getting a positive RT-PCR result, as well.

“We have found that, using a COVID-specific structured approach to interpretation, a [chest X-ray] designated as ‘high suspicion’ or ‘characteristic’ in a patient presenting to the emergency department with clinically suspected COVID-19 pneumonia correlates with a very high likelihood of a positive RT-PCR.”

Related Content: Chest X-rays Can Predict COVID-19 Severity in Young and Middle-Age ER Patients​

Yates’ team conducted a retrospective study, examining the images of 582 patients who presented to the emergency department and were suspected of viral infection. All patients underwent both a chest X-ray at the RT-PCR test. Then, while blinded to RT-PCR results, two radiologists reviewed and interpreted the images, using a five-category system to divide the scans based on the likelihood of infection. “High suspicion” and “characteristic” were the two levels most indicative of severity, and the providers agreed 71.1 percent of the time.

Among the patients included in the study, approximately 25 percent ended up being COVID-19-positive. Based on their analysis, the team said, the structured reporting approach had an 88-percent positive-predictive value, as well as specificity of 98 percent. Consequently, this approach is a highly useful arrow to have in the COVID-19 diagnostic quiver.

Given the strength of these results, the team said, they feel confident this approach can be used to positively impact patient care.

“We have demonstrated very good agreement in the application of this structured approach in a single institution and feel that this practice is generalizable to other institutions,” they said.