What the study measured
The COMBO trial, led by Giulia Vatteroni and colleagues at IRCCS Humanitas Research Hospital in Milan and published in Insights into Imaging, followed breast cancer survivors who received contrast-enhanced mammography (CEM) instead of standard digital mammography for routine post-surgical surveillance between January 2023 and February 2025. Of 969 eligible women, 936 completed an anonymous 16-question survey about discomfort, pain, exam duration, anxiety, and satisfaction immediately after the exam — a 96.6% response rate in its own right, and a mean participant age of 58.2.
CEM adds a step standard mammography doesn't have: an iodinated contrast injection, plus extra imaging time. That's a reasonable setup for patients to report more discomfort than they're used to. They largely didn't. 99.6% said they would recommend CEM to others, and 96.2% (900 of 936) reported overall satisfaction, as covered by Radiology Business.
Comparable discomfort, not zero discomfort
The more interesting numbers sit underneath the headline. When asked to compare CEM with their prior digital mammograms, 67.4% rated discomfort as similar, 77.4% rated pain as comparable, 76.6% found positioning similar, and 64.3% rated the longer exam duration as comparable. None of those numbers are 100% — roughly a quarter to a third of patients did notice the exam was harder in some way. They recommended it anyway.
Anxiety tells a similar story. 46.3% of participants reported moderate-to-high anxiety before the exam, and the researchers found that higher anxiety was significantly associated with greater perceived discomfort, pain, and lower satisfaction. But — and this is the detail worth sitting with — the authors reported that anxiety did not reduce patients' willingness to undergo CEM. Patients who were nervous beforehand still rated the exam as worth recommending afterward.
| Measure | Result |
|---|---|
| Would recommend CEM to others | 99.6% |
| Overall satisfaction | 96.2% (900/936) |
| Discomfort comparable to prior mammogram | 67.4% |
| Pain comparable to prior mammogram | 77.4% |
| Positioning comparable to prior mammogram | 76.6% |
| Exam duration comparable to prior mammogram | 64.3% |
| Moderate-to-high pre-exam anxiety | 46.3% |
The pattern: explained technology earns tolerance
Pull back from mammography specifically and the shape of the finding is bigger than one exam. Patients in the COMBO trial accepted a procedure that was objectively more involved — contrast injection, longer time, a real anxiety response in nearly half of them — and still endorsed it at a rate that's effectively unanimous. What makes that tolerable is context: survivors already knew why CEM was being used (closer surveillance after a breast cancer diagnosis), and the exam was explained to them as part of their care, not sprung on them.
That same pattern shows up in research on AI disclosure in imaging. A separate 2026 survey of more than 1,000 imaging patients, published in RSNA's Radiology, found that about 96% wanted to be told when AI was used to help read their scan, and more patients preferred that disclosure documented in writing than preferred it given verbally — detail covered in xAID's earlier look at that data. Patients in both studies aren't resistant to more technology in their care. They're responsive to whether it's explained and whether a clinician stays accountable for the result.
Where this fits with how AI CT reporting works
The lesson generalizes past breast imaging: transparency, not concealment, is what earns patient trust in new imaging technology. That principle is exactly why AI-assisted CT reporting should be disclosed rather than treated as invisible back-office tooling — the model AI draft, radiologist-reviewed, ready-to-sign report is built on the same logic the COMBO trial's survivors demonstrated: patients tolerate — and even endorse — more technology in their care once they understand what it is and who remains responsible for the outcome.
Frequently asked questions
What percentage of breast cancer survivors would recommend contrast-enhanced mammography?
In the COMBO trial, a prospective study of 936 breast cancer survivors (from 969 eligible women) undergoing contrast-enhanced mammography for post-surgical surveillance between January 2023 and February 2025, 99.6% said they would recommend the exam to others, and 96.2% reported overall satisfaction.
Is contrast-enhanced mammography more uncomfortable than standard mammography?
Most patients rated it comparable. In the COMBO trial, 67.4% found discomfort similar to prior digital mammography, 77.4% rated pain comparable, 76.6% found positioning similar, and 64.3% rated exam duration as comparable — despite the added contrast injection step.
Did pre-exam anxiety affect how patients experienced contrast-enhanced mammography?
Yes, but it didn't change their willingness to have the exam. 46.3% of participants reported moderate-to-high pre-examination anxiety, and higher anxiety was significantly associated with greater perceived discomfort, pain, and lower satisfaction. Even so, the researchers found anxiety did not reduce patients' willingness to undergo contrast-enhanced mammography.
What does patient trust in contrast-enhanced mammography suggest about disclosing AI in imaging?
The COMBO trial shows patients tolerate an added step — a contrast injection, more time, more procedure — at very high acceptance rates when they understand why it's being done and what it adds. The same logic supports disclosing AI involvement in reporting rather than treating it as a detail to bury: separate survey data on AI in imaging has found patients overwhelmingly want to be told when AI is used, provided a radiologist remains accountable for the final report.
Source: Vatteroni G, et al., "Patient acceptance and experience with contrast-enhanced mammography in women with a personal history of breast cancer: findings from the COMBO trial," Insights into Imaging (2026), DOI: 10.1186/s13244-026-02434-x, as reported by Radiology Business. AI-disclosure survey figures from the 2026 RSNA Radiology patient survey as covered in xAID's earlier article. Figures are rounded as reported.