Purpose: Pericardial adipose tissue may exert unique metabolic and cardiovascular risks in patients. The use of cardiac multidetector computed tomography (MDCT) imaging may enable the accurate localization and quantification of pericardial and intrathoracic adipose tissue. The reproducibility of electrocardiogram-gated high-resolution cardiac MDCT-based volumetric quantification of pericardial and intrathoracic adipose tissue has not been reported.
Methods: We included 100 consecutive patients (age 54.5 +/- 12.0 yr, 60% men) who underwent a standard contrast-enhanced coronary CT for the evaluation of coronary artery plaque and stenosis (64-slice MDCT, temporal resolution: 210 ms, spatial resolution: 0.6 mm, 850 mAs, 120, kvp) after a presentation of acute chest pain. Two independent observers measured intrathoracic (IAT) and pericardial adipose tissue (PAT) by using a semiautomatic segmentation algorithm based on three-dimensional analysis.
Results: Inter-reader reproducibility was excellent (relative difference: 7.35 +/- 7.22% for PAT and 6.23 +/- 4.91% for IAT, intraclass correlation 0.98 each). Similar results were obtained for intra-observer reproducibility (relative difference: 5.18 +/- 5.19% for PAT and 4.34 +/- 4.12% for IAT, intraclass correlation 0.99 each).
Conclusion: This study demonstrates that MDCT-based 3D semiautomatic segmentation for volumetric quantification of PAT and IAT is highly reproducible. Further research is warranted to assess whether volumetric measurements may substantially improve the predictive value of obesity measures for insulin resistance, type 2 diabetes mellitus, and cardiovascular diseases.