Fatty Liver and CT-FFR Each Signal Heart Risk in Suspected Coronary Disease

Patients undergoing coronary CT angiography for suspected coronary artery disease may carry a second, quieter warning sign: fatty liver. A retrospective cohort study published in BMC Cardiovascular Disorders followed 614 such patients, each of whom also had an unenhanced CT scan covering the liver and spleen within one week of their angiography, and found that non-alcoholic fatty liver disease was independently associated with major adverse cardiac events over a median follow-up of 62.5 months.

The researchers defined non-alcoholic fatty liver disease using a mean liver-to-spleen CT attenuation ratio of 1.0 or lower, after excluding substantial alcohol consumption and other causes of hepatic steatosis. Of the 614 patients, 189, or 30.8 percent, met this definition. Hemodynamically significant coronary stenosis was assessed with CT-derived fractional flow reserve, with a value of 0.80 or below considered significant; 185 patients, or 30.1 percent, fell into this category. The primary endpoint, major adverse cardiac events, combined cardiac death, heart failure hospitalization, nonfatal myocardial infarction, and late coronary revascularization. During follow-up, 61 patients, or 9.9 percent, experienced such an event.

In the final hierarchical Cox regression model, two factors emerged as independently associated with major adverse cardiac events: obstructive plaque, with an adjusted hazard ratio of 3.01 (95 percent confidence interval, 1.27 to 7.11; P = 0.012), and non-alcoholic fatty liver disease, with an adjusted hazard ratio of 2.29 (95 percent confidence interval, 1.34 to 3.91; P = 0.002). These associations are observational and come from a single retrospective cohort, so they describe risk markers rather than proven causal pathways.

The picture for CT-derived fractional flow reserve was more nuanced. A value of 0.80 or below showed a statistically significant time-varying association with major adverse cardiac events (P = 0.027 for the time-varying effect), meaning its association with events was strongest earlier in follow-up and attenuated over time. Sequential likelihood-ratio testing in nested models showed that adding CT-FFR improved model fit beyond clinical and anatomic variables (χ2 = 16.44, df = 1, P < 0.001), and adding fatty liver status improved fit further (χ2 = 9.26, df = 1, P = 0.002).

A key question was whether the two markers interacted, that is, whether fatty liver changed the prognostic meaning of an abnormal fractional flow reserve value or vice versa. The NAFLD-by-CT-FFR interaction term was not statistically significant (P for interaction = 0.071), so the study does not establish a combined or synergistic effect. In a sensitivity analysis restricted to hard endpoints, event-free survival differed significantly according to CT-FFR status (log-rank P < 0.001), but the difference according to fatty liver status did not reach statistical significance (P = 0.052), a borderline result the authors report without overinterpreting.

The study’s design shapes how its findings should be read. Because it was retrospective and restricted to patients who happened to have both coronary CT angiography and a suitable unenhanced CT within a week, the cohort may not represent all patients with suspected coronary artery disease. Fatty liver was identified from attenuation ratios on scans acquired for other reasons, a practical but indirect measure. Follow-up captured events over roughly five years at the median, and the attenuating association of CT-FFR over time suggests its prognostic value is concentrated in the earlier years after imaging.

What the study does establish is that, in this population, a liver finding visible on an ordinary CT scan added prognostic information beyond coronary anatomy and physiology. Non-alcoholic fatty liver disease has long been linked to cardiovascular disease in broader epidemiological work, but its role alongside modern CT-based coronary assessment had remained uncertain. By modeling both markers together and testing their interaction formally, the authors showed that each contributed to risk prediction in this cohort without the two overlapping into statistical redundancy.

For clinicians, the results suggest that an incidental hepatic steatosis finding on a CT scan in a patient being worked up for suspected coronary disease may carry prognostic weight, while CT-derived fractional flow reserve remains a meaningful, though time-dependent, marker of event risk. The authors caution, through their own sensitivity analyses, that the fatty liver signal was weaker for hard endpoints such as cardiac death and myocardial infarction than for the composite endpoint. As an accepted early-release article carrying a permanent DOI, the paper’s final version of record may include further edits, and its conclusions apply to the studied cohort rather than establishing screening or treatment recommendations.

Subject of Research: Prognostic value of non-alcoholic fatty liver disease and CT-derived fractional flow reserve in suspected coronary artery disease

Article Title: Prognostic association of non-alcoholic fatty liver disease and CT-derived fractional flow reserve in suspected coronary artery disease

Article References: Li, S., Wang, Q., Zhang, S., Wang, X., & Yuan, X. (2026). Prognostic association of non-alcoholic fatty liver disease and CT-derived fractional flow reserve in suspected coronary artery disease. BMC Cardiovascular Disorders. https://doi.org/10.1186/s12872-026-06919-x

Image Credits: AI Generated

DOI: 10.1186/s12872-026-06919-x

Keywords: non-alcoholic fatty liver disease, CT-derived fractional flow reserve, coronary CT angiography, coronary artery disease, major adverse cardiac events, retrospective cohort, Cox regression, prognosis, hepatic steatosis, cardiovascular risk, Prognostic, association