When can coronary computed tomography angiography in patients with calcified plaque be accurate?

To identify a new method to indicate when coronary computed tomography angiography (CCTA) in patients with calcified plaque can be accurate.

Prospective analysis on 105 cases of coronary artery stenosis with calcified plaque underwent both CCTA and invasive coronary angiography (ICA). The Hounsfield unit (Hu) values of calcified plaque and adjacent blood were measured, and then the ratio (R _Hu ) was subsequently calculated. The ICA data served as the gold standard for defining obstructive stenosis (≥ 50%) and were utilized to create a two-dimensional receiver operating characteristic (ROC) curve. The cut-off value was employed to categorize the CCTA data. Additionally, a Bland-Altman plot was used to analyze discrepancies in stenosis degree detection between CCTA and ICA. An in vitro experiment was designed to assess the practicability of R _Hu .

The R _Hu was correlated with the concordance of CCTA and ICA for stenosis evaluation (r = 0.509, p < 0.001). ROC analysis suggested a cut-off value of 0.36. The Bland-Altman plot indicated that stenosis evaluation by CCTA demonstrates good concordance when R _Hu exceeds 0.36; however, significant bias occurs when R _Hu is below 0.36 in comparison to ICA. In vitro experiments confirmed that the R _Hu parameter can be easily adjusted to enhance the accuracy of CCTA. In validation experiments, the R _Hu achieved a prediction accuracy of 74.0%.

Our study suggests that the accuracy of detection of stenosis with CCTA in calcified vessels is related to the difference in Hu values between calcified plaques and blood.