What Radiologists Should Do About Incidental Findings

Imaging technology allows radiologists to view bodily structures much more clearly than in the past. The increase and quality of cross-sectional imaging, especially CT scans, means that along with better clarity of the target structures, radiologists are detecting more incidental findings.

For patients with clinically significant surprise findings, that’s appreciated. But the rest? It opens a Pandora’s box: what to report, how to frame it, whether to recommend further work-up, and how best to follow up with the physician or patient on the findings.

“A lot of times we’re in a difficult situation because the quality of our imaging and sensitivity to the test has risen faster than our knowledge of how to handle things,” said Renda Soylemez Wiener, MD, MPH, a pulmonologist and assistant professor of medicine at Boston University. “We’re in that grey area of having this information and not really knowing how to handle it.”

While ventilation perfusion (VQ) scanning was once the primary test for pulmonary emboli, that took a back seat when CT pulmonary angiography was introduced in 1998, according to Wiener’s study, “When a test is too good: how CT pulmonary angiograms find pulmonary emboli that do not need to be found,” published in the July 2013 British Medical Journal.

[[{"type":"media","view_mode":"media_crop","fid":"15965","attributes":{"alt":"Renda Soylemez Wiener, MD, MPH","class":"media-image media-image-right","id":"media_crop_7249598373022","media_crop_h":"0","media_crop_image_style":"-1","media_crop_instance":"883","media_crop_rotate":"0","media_crop_scale_h":"0","media_crop_scale_w":"0","media_crop_w":"0","media_crop_x":"0","media_crop_y":"0","style":"float: right; margin: 5px;","title":"Renda Soylemez Wiener, MD, MPH","typeof":"foaf:Image"}}]]“In the eight years after it was introduced and rapidly adopted, diagnoses exploded more than 80 percent, but the death rate changed very little,” she said. “We also saw the case fatality among those with the emboli went down dramatically,” implying that many of­­­­ the emboli treated in the hospital were subclinical.

How should a radiologist determine whether a pulmonary embolus -  or any incidental finding for that matter - is significant or warrants additional work-up?

Use evidence-based recommendations, when available, said Stephen Brown, MD, staff radiologist at Boston Children’s Hospital and assistant professor of radiology at Harvard Medical School. Two that Brown relies on are the American College of Radiology’s white paper on reporting abdominal imaging incidental findings, and the Fleischner Society’s guidelines for managing small pulmonary nodules detected on CT scans.

Even when the guidelines exist, though, reporting practices are variable. “Many radiologists conform to the guidelines, but many don’t. When there are no guidelines, the variability is likely even greater,” Brown said.

Since small pulmonary nodules are such a common incidental finding on CT, identified in up to a quarter of patients undergoing a chest CT, Wiener said she’s often faced with the difficult decision of trying to determine their clinical significance. Additional work-up is stressful to the patient and potentially harmful as well. Depending on the size, the nodule might be reevaluated on serial CT scan surveillance over one to two years.

“This exposes patients to radiation, distress and uncertainty about whether they have a cancer,” she said. Larger nodules may require a biopsy, exposing patients to the risk of physical complications like collapsed lung or bleeding. For what? “Less than five percent turn out to be lung cancer,” she said.

One outcome of Wiener’s study is the admonition that not every pulmonary embolus needs treatment.

“In general doctors are still at the point where if they see a pulmonary embolus, they feel obligated to treat it,” she said. When doctors find an incidental pulmonary embolus, they may think they found something important, whereas they should consider questioning its clinical significant, especially if the patient lacks symptoms.

This is where radiologists can provide expertise. As Brown noted in his recent JACR article, “Professional Norms Regarding How Radiologists Handle Incidental Findings,” working up these unexpected (but ultimately benign) findings and treating incidental quiescent cancers, may outweigh the benefit of finding and treating incidental findings clinically beneficial to the patient.

Radiologists are thinking about the stress that these findings may have on patients, as well as the legal implications of whether to include them in the report.

“Many radiologists sitting at the view box probably have both of these concerns in mind,” Brown said, adding that there’s no published research available looking at radiologists’ motivation or behavior around incidental findings. “Whether radiologists can rest more comfortably about malpractice fears I think depends on adherence to evidence-based criteria when possible, and on remembering that, in reality, the likelihood of being sued for a missed incidental finding is quite low.”

Documenting the findings and then determining the best way to convey that information to the ordering physician isn’t always as simple as a phone call or flagged report. “It does put the treating physician or person ordering the test in a difficult situation trying to decide what to do with that information,” Wiener said.

Giving the ordering physician guidance about the incidental finding is part of a diagnostic radiologist’s responsibility, said Brown, though where that line is drawn is open to interpretation. It’s best to be as unequivocal as possible and discuss why there might be equivocation, Brown said, adding that the report can include recommended next steps to consider, as well as documenting how the findings were communicated to the ordering physician.

Where does the radiologist draw the line on reasonably discharging their responsibilities? Should the physician just write the report and discuss [[{"type":"media","view_mode":"media_crop","fid":"15964","attributes":{"alt":"Stephen Brown, MD","class":"media-image media-image-right","id":"media_crop_5258462519950","media_crop_h":"0","media_crop_image_style":"-1","media_crop_instance":"882","media_crop_rotate":"0","media_crop_scale_h":"0","media_crop_scale_w":"0","media_crop_w":"0","media_crop_x":"0","media_crop_y":"0","style":"margin: 5px; float: right;","title":"Stephen Brown, MD","typeof":"foaf:Image"}}]]findings with the ordering physician? What if that specialist is less familiar with the patient or the type of finding? When should the radiologist consider approaching the patient’s primary physician versus the specialist ordering the exam, or initiating a conversation with the patient?

Brown gives an example of an orthopedic surgeon ordering a scoliosis series, which shows an incidental finding outside of the surgeon’s range of expertise and experience. Is the radiologist’s responsibility handed off by just informing the surgeon about the finding?  “Radiologists are of different minds on this,” Brown said.

A specialist may not understand the significance of the finding or may not follow up on it, Brown said, and every radiologist should ask themselves where their responsibility to a patient ends. Brown said he follows up on every incidental finding he’s concerned about, or ones where he’s unsure the referring physician appreciates the significance. He thinks he’s in the minority of radiologists that do this, though.

While there are many downsides to reporting incidental findings that are ultimately benign, Brown concluded in his study that radiologists have limited choice about sharing the results.

“In a culture in which patient autonomy is valued, standards are inconsistent, and malpractice fears abound, little room exists for individual physicians to decide unilaterally whether reporting or disclosing incidental findings will result in better or worse decisions for any given patient,” he said in the study. Using evidence-based guidelines and framing the reports with these guidelines in mind can assist not only the ordering physician, but ultimately the patient.