Chest CT fat and muscle readings reveal who carries more epicardial fat

Routine chest CT scans may quietly hold a warning sign that radiologists rarely read: the amount of fat wrapped around the heart. A retrospective cross-sectional study from a general teaching hospital in Iraq, published in the Egyptian Journal of Radiology and Nuclear Medicine, shows that epicardial fat volume measured on ordinary non-ECG-gated chest CT is independently associated with age, sex, and the amount of subcutaneous fat visible on a single slice at the level of the sternal angle. Pectoralis major muscle measurements taken at the same anatomical landmark, by contrast, showed no independent association with the fat surrounding the myocardium.

Epicardial fat is a visceral fat depot situated between the pericardium and the heart itself. Unlike other fat stores in the body, it lacks a fascial layer separating it from the myocardium and the major coronary arteries, which allows it to release proinflammatory cytokines directly onto cardiac tissue. The authors of the new study note that this contributes to localized inflammation and may promote atherosclerotic plaque formation. Epicardial fat is already considered an established indicator of cardiovascular risk, and increased epicardial fat volume has been linked in prior research to a higher prevalence of coronary artery disease, heart failure, and dysrhythmias. It has also been proposed as a factor in the prognosis of COVID-19 pneumonia.

The volume of this fat varies widely between individuals, and several determinants have been described, including demographic characteristics and clinical status. A decrease in lean body mass and an increase in fat mass during aging may contribute to higher epicardial fat volume, and these changes occur at different rates in men and women, with the accumulation reported to be more pronounced in older women. Computed tomography, both ECG-gated and non-ECG-gated, along with echocardiography and magnetic resonance imaging, have all been used to assess it, and some authors have advocated including epicardial fat volume in routine reporting of coronary artery calcium scoring CT.

The research team, led by Israa Sadiq and Saman Nooruldeen, set out to test whether body composition measurements taken opportunistically at the sternal angle, an easily recognizable landmark on routine chest CT that generally corresponds to approximately the T4–T5 vertebral level, are associated with epicardial fat volume. Measurements of fat and muscle areas at the L3 vertebral level on abdominal CT are well-established proxies for whole-body fat-free mass and adipose tissue volume, and previous studies have explored thoracic levels, including T4, as alternatives when abdominal imaging is unavailable. Pectoralis muscle area and attenuation have been evaluated at the T4 level on routine chest CT, demonstrating feasibility and potential clinical relevance in patients with COVID-19. Whether these thoracic measurements relate to epicardial fat, however, had not been specifically investigated before.

The study was conducted between April 2025 and February 2026. The researchers initially identified 144 consecutive adult patients who had undergone routine chest CT for standard clinical indications, including evaluation of respiratory symptoms such as cough or dyspnea, suspected or confirmed infection, oncologic assessment, and follow-up of known thoracic conditions. Patients older than 65 years were excluded to minimize confounding from age-associated changes in skeletal muscle mass and body composition. Additional exclusions covered conditions affecting chest wall assessment, such as neuromuscular diseases or prior mastectomy, factors impairing epicardial fat evaluation such as massive pericardial effusion, geometric distortion from severe scoliosis, and missing data or image artifacts that precluded accurate measurement. After these criteria were applied, 110 patients formed the final study population.

All scans were acquired on a 64-detector-row Philips scanner at 120 kVp with automatic tube-current modulation, reconstructed at a slice thickness of 4.0 mm with a 2.0-mm increment using a hybrid iterative reconstruction algorithm and a soft-tissue kernel. Non-contrast images were analyzed in 3D Slicer software by a single radiologist with 15 years of CT experience. Epicardial fat volume was quantified by three-dimensional semi-automatic segmentation of adipose tissue within a Hounsfield unit threshold range of −190 to −30, applied within the pericardial sac across contiguous slices. The superior boundary was set at the axial slice corresponding to the center of the right pulmonary artery, identified on a sagittal reconstruction, and segmentation extended inferiorly to the level of the diaphragm, where the pericardial line was no longer clearly visualized, with care taken to exclude non-pericardial fat.

At the sternal angle level, the team measured pectoralis major muscle area in square centimeters using semi-automatic threshold segmentation of muscle tissue between −29 and 150 HU with manual refinement of both sides, recording mean muscle attenuation to assess muscle quality. Subcutaneous fat area was segmented using the same adipose tissue threshold of −190 to −30 HU, excluding intrathoracic structures and encompassing the entire thoracic circumference. Because breast adipose tissue contributes to measured subcutaneous fat in females, sex-stratified analyses were performed to account for this anatomical difference. Reproducibility was tested by repeating measurements in 15 randomly selected cases after a two-week interval, with the same radiologist blinded to the initial results. Intraclass correlation coefficients indicated excellent agreement: 0.95 for epicardial fat volume, 0.92 for subcutaneous fat area, 0.88 for pectoralis muscle area, and 0.90 for muscle attenuation.

The results paint a consistent picture. On unadjusted comparison, there was no significant difference in epicardial fat volume between males and females, and age was comparable between the sexes. Males had significantly larger pectoralis major muscles with higher attenuation, while females had significantly larger subcutaneous fat area. Spearman correlation analysis showed a moderate positive correlation between age and epicardial fat volume, with a coefficient of 0.59. Total pectoralis muscle area showed essentially no correlation, at 0.01, while muscle attenuation showed a weak but statistically significant inverse correlation of −0.20. Subcutaneous fat area correlated positively with epicardial fat volume in both sexes, with a coefficient of 0.37 in males and 0.31 in females, of comparable strength.

In the multivariable linear regression model, which included age, sex, pectoralis muscle attenuation, pectoralis muscle area, and subcutaneous fat area, sex emerged as independently associated with epicardial fat volume, with males showing higher volumes after adjustment, a coefficient of 15.17 with a 95% confidence interval of −28.13 to −2.22 for the female coding and a p value of 0.022. Age carried a coefficient of 1.26 per year, with a 95% confidence interval of 0.88 to 1.65 and a p value below 0.001, indicating a progressive increase with advancing age. Subcutaneous fat area remained independently associated with a coefficient of 0.03, a 95% confidence interval of 0.01 to 0.06, and a p value of 0.002. An age-by-sex interaction term was tested and was not statistically significant, at p = 0.127, so it was excluded from the final model. Pectoralis muscle attenuation, despite its significant univariable inverse association, was not independently associated after adjustment, with a coefficient of −0.02 and a p value of 0.948, and muscle area likewise was not, with a coefficient of 0.07 and a p value of 0.381. The overall model was statistically significant, explaining approximately 41.2% of the variance in epicardial fat volume, with an R-squared of 0.412 and an adjusted R-squared of 0.384.

The authors interpret the subcutaneous fat finding as suggesting that overall adiposity contributes to ectopic fat deposition around the heart, possibly reflecting shared pathways of lipid overflow and low-grade inflammation that promote fat accumulation in multiple depots, including the epicardium. The disappearance of the muscle attenuation association after adjustment, they argue, indicates it was likely confounded by overall adiposity and age, consistent with prior reports that muscle quality is not consistently associated with cardiometabolic risk markers and that larger muscle mass can coexist with fat accumulation rather than exerting a protective effect. They also frame increased epicardial fat and skeletal muscle abnormalities as potentially parallel expressions of an underlying adverse metabolic profile rather than a direct causal relationship. The study carries clear limitations: the retrospective design limits causal inference, non-ECG-gated CT may introduce measurement variability, a single axial slice may not represent whole-body composition, only intraobserver variability was assessed, residual confounding from unmeasured factors such as smoking, physical activity, body mass index, and metabolic profiles cannot be excluded, and restricting the cohort to adults aged 18 to 65 limits generalizability to older adults. Even so, the work supports the idea that an ordinary chest CT, read with an extra set of measurements at the sternal angle, can yield cardiometabolic information that would otherwise go unreported.

Subject of Research: Associations between chest CT-derived body composition measurements at the sternal angle level and epicardial fat volume

Article Title: Association between chest CT-derived body composition measurements at the sternal angle level and epicardial fat volume: a retrospective cross-sectional study

Article References: Sadiq, I., & Nooruldeen, S. (2026). Association between chest CT-derived body composition measurements at the sternal angle level and epicardial fat volume: a retrospective cross-sectional study. Egyptian Journal of Radiology and Nuclear Medicine, 57(1), Article 211. https://doi.org/10.1186/s43055-026-01877-y

Image Credits: AI Generated

DOI: 10.1186/s43055-026-01877-y

Keywords: epicardial fat volume, chest CT, body composition, sternal angle, pectoralis major muscle, subcutaneous fat area, retrospective cross-sectional study, cardiometabolic risk, opportunistic imaging, radiology, multivariable regression, visceral fat