AI could give mammography screening a second purpose: In addition to early detection of breast cancer, the images can also reveal vascular calcifications that indicate an increased risk for cardiovascular disease. A large cohort study in the European Heart Journal suggests this possibility. 

“Evidence that calcifications on mammography are associated with increased cardiovascular risk has existed for some time. But it has not been studied at this scale with automated AI analysis,” said Stephan Achenbach, MD, who is the chairman of the Department of Cardiology and Professor of Medicine at the Friedrich-Alexander University Erlangen-Nuernberg, Erlangen, Germany, in an interview with Medscape Medical News. “Now it is clear that these calcifications are indeed associated with later heart disease.”

A Known Gap in Care

Study leader Hari Trivedi, MD, of the Department of Radiology at Emory University, Atlanta, sees the approach as a chance to address a known care gap. In a press release, he said that cardiovascular disease is the leading cause of death in women worldwide, yet it is consistently underdiagnosed and undertreated compared with men. Using existing mammography programs could help identify women at increased risk without extra effort.

The retrospective study included 123,762 women without known cardiovascular disease who underwent breast cancer screening in two US health systems — Emory University in Atlanta and the Mayo Clinic in Rochester, Minnesota — where an AI model quantified the area of breast arterial calcifications (BAC) on mammograms and classified them into four categories: none (0 mm²), mild (> 0-10 mm²), moderate (> 10-25 mm²) and severe (> 25 mm²).

Severe Calcifications are Extremely Rare

Calcifications were detected in 16.1% of women in the Emory cohort and in 20.6% of women in the Mayo Clinic cohort. Severe calcifications were rare: in both cohorts they affected only about 1% of women. Another 1% to 3% had moderate calcifications and 14% and 21% had mild calcifications, while the majority — 84% and 74%, respectively — had no findings.

Median follow‑up duration was 7 years. During that period, the two cohorts experienced more than 9000 major adverse cardiovascular events (MACE), defined as myocardial infarction, heart failure, stroke, and cardiovascular death.

The extent of BAC was strongly associated with the incidence of MACE and with all‑cause mortality. Both cohorts showed a clear dose‑response relationship, with stepwise increases in event rates. Compared with women without calcifications, the risk for cardiovascular events was about 30% higher with mild calcification, about 70% higher with moderate calcification, and roughly two to three times higher with severe calcification. The association also held when BAC was analyzed as a continuous variable: For each additional 1 mm² of calcified area, the risk for cardiovascular events rose by about 1%-2%.

Association Visible Even in Younger Women

“We found that the more calcium visible in the breast arteries on the mammogram, the higher a woman’s risk of developing cardiovascular disease,” Trivedi reported. “This was true even for women under 50, and it persisted after adjustment for traditional risk factors such as diabetes and smoking.” The authors concluded that mammography could be used as an “opportunistic screening” tool for cardiovascular risk — without additional radiation exposure or separate tests.

The accompanying editorial explicitly supports the approach. “Breast arterial calcifications have long been considered incidental findings, but the evidence of their prognostic significance continues to grow,” wrote Lori B. Daniels, MD, of the Sulpizio Cardiovascular Center at the University of California San Diego, in the European Heart Journal. She emphasized the study’s practical approach: Quantifying calcification as an area in mm² is “an intuitive and reproducible metric” that could ease standardization and implementation.

Possibility of Dual Prevention 

Daniels argues that breast vessel calcification should not be mistaken for a surrogate for coronary atherosclerosis, but rather regarded as a marker of vascular aging and chronic stress from hypertension, diabetes, chronic kidney disease and obesity. That leads to a “dual prevention” approach that, in addition to classic atherosclerotic cardiovascular prevention, also brings heart‑failure prevention into focus. Her conclusion: “It is time to move BAC from an incidental finding to clinical use — and to leverage mammography as a point of contact that women already trust.”

Achenbach urges a realistic interpretation of the study results: “You shouldn’t overstate relative risk increases,” he said. “Most women have no calcifications, and even with mild to moderate calcification the absolute risk often remains low.” As an example, he cited a 55‑year‑old nonsmoker with otherwise unremarkable risk: “If her 10‑year risk is 2% and it increases by 30%, that’s 2.6%. That is not a dramatic increase.”

A Few Women Could Benefit

Still, Achenbach sees practical potential: “If a few women have pronounced calcifications on mammography and an automated alert prompts the radiologist — ‘Alert: consider cardiovascular risk assessment’ — then that would be useful, and would not involve much additional effort when using AI.” It is important, he added, not to cause unnecessary alarm: “We must not worry women who have only minor calcifications and only a slightly increased risk.”

How the AI‑based procedure could be integrated into existing screening programs and what formal recommendations for clinicians and patients should look like remain open questions. “We are currently planning a clinical study to test exactly these steps,” Trivedi said.

This story was translated from Medscape’s German edition.