Coronary CT Angiography Misses Over One Third of Calcified Plaques

Refining Atherosclerotic Risk Stratification in Coronary Artery Disease

Accurate assessment of coronary artery disease remains a cornerstone of cardiovascular risk stratification and the management of stable chest pain. Current clinical guidelines emphasize the importance of identifying atherosclerotic plaque burden to guide lipid-lowering therapies and primary preventive strategies [1, 2]. While coronary computed tomography angiography (CCTA) has become a primary diagnostic tool for evaluating luminal stenosis and plaque morphology, its ability to capture the full spectrum of coronary calcification is essential for predicting major adverse cardiovascular events [3, 4]. Total plaque volume and calcified plaque burden are established independent predictors of long-term outcomes in patients with suspected coronary artery disease [3]. However, the technical limitations of contrast-enhanced imaging may influence the sensitivity of plaque detection compared to non-contrast techniques. A recent study now evaluates the diagnostic performance of standard angiography against high-resolution non-contrast imaging to determine if clinically significant calcifications are being overlooked.

Comparative Analysis of Plaque Detection Sensitivity

The researchers conducted a retrospective analysis of 45 patients recruited from two clinical sites participating in the DISCHARGE trial. This patient cohort, which had a mean age of 62 ± 11 years and was 40% female, provided the basis for a head-to-head comparison between different computed tomography modalities. Each participant underwent a specialized imaging protocol where both 0.5 mm thin-slice non-contrast CT (NCCT) and standard coronary CT angiography (CCTA) were available. The use of 0.5 mm thin-slice non-contrast CT allowed for a high-resolution assessment of calcified structures without the potential interference of intraluminal contrast agents, which can sometimes obscure small, dense objects. To achieve a precise spatial comparison between the two types of scans, the study utilized deep learning-aided co-registration, a process using artificial intelligence to align images from different scans so they can be viewed in the same anatomical space. This technological approach ensured that the researchers could track the exact location of individual calcifications across both imaging sets. Within this analytical framework, plaques on CCTA were defined as missed if they showed no spatial overlap with NCCT-detected plaques after co-registration. This rigorous criterion allowed the authors to identify specific calcified lesions that were clearly visible on high-resolution non-contrast images but remained undetected during the contrast-enhanced angiographic phase.

Quantifying the Gap in Calcification Identification

The analysis of the 322 calcified plaques identified via thin-slice non-contrast CT revealed a significant diagnostic gap in standard angiography. The researchers found that coronary CT angiography missed 37.6% of calcified plaques (121 out of 322 plaques) that were clearly visible on the high-resolution non-contrast scans. This discrepancy suggests that the presence of iodinated contrast within the vessel lumen may obscure smaller or less dense calcifications, potentially leading to an underestimation of the total atherosclerotic burden during routine clinical evaluations. The failure to detect these calcifications has direct implications for patient risk stratification and subsequent management. In this study, the missing calcified plaques on coronary CT angiography misclassified 8.9% of patients (4 out of 45) as having no plaques at all. For these individuals, a reliance on angiography alone would result in a false-negative finding for coronary calcification. This misclassification is clinically significant because it could lead to the omission of primary prevention strategies, such as lipid-lowering therapy, in patients who actually harbor detectable atherosclerotic disease. The study also evaluated whether standard calcium scoring CT, the current clinical benchmark for quantifying calcification, could compensate for the limitations of angiography. However, the results indicated that standard techniques remain insufficient for detecting these subtle lesions. Only 44.0% of the plaques missed by coronary CT angiography (53 out of 121) were detected by standard calcium scoring CT, leaving more than half of these specific calcifications unidentified by either conventional method. These findings highlight the superior sensitivity of 0.5 mm thin-slice non-contrast CT in capturing the full spectrum of coronary calcified plaque that standard protocols frequently overlook.

Morphological Characteristics of Overlooked Plaques

The researchers conducted a detailed morphological analysis to determine why certain calcifications remained invisible during contrast-enhanced imaging. Using the Mann–Whitney U-test (a statistical method used to compare differences between two independent groups when the data are not normally distributed) to compare plaque volume, density, and local coronary luminal attenuation (the brightness of the blood pool within the vessel), the study identified distinct physical differences between detected and overlooked lesions. The data revealed that plaques missed by coronary CT angiography were significantly smaller in volume than those that were successfully detected, with a median volume of 3.0 mm³ (interquartile range [IQR], 1.5 to 4.9) compared to 9.2 mm³ (IQR, 4.3 to 21.9) for detected plaques (p < 0.001). This suggests that the partial volume effect, where the signal from small structures is averaged with surrounding tissue or contrast, significantly hinders the detection of low-volume calcifications when iodinated contrast is present. In addition to size, the mineral density of the plaques played a critical role in their visibility. The study found that plaques missed by coronary CT angiography had significantly lower density than detected plaques, measuring 212.7 Hounsfield units (HU) (IQR, 174.5 to 242.4) versus 292.7 HU (IQR, 243.2 to 361.3) for those that were identified (p < 0.001). For the clinician, these findings indicate that angiography is particularly prone to overlooking early-stage or less-dense calcifications. Because the attenuation of the contrast-filled lumen often exceeds 300 HU, it can effectively mask these smaller, lower-density lesions. This morphological profile explains why thin-slice non-contrast CT, which avoids the luminal brightening effect of iodine, provides a more comprehensive assessment of the total calcified plaque burden.

Reliability and Clinical Implications for Plaque Burden

To ensure the reliability of these measurements, the researchers evaluated the consistency between different readers using the intraclass correlation coefficient (a statistical measure of how closely different observers agree on a measurement) and Bland-Altman analysis (a method used to assess the agreement between two quantitative measurements by plotting their differences against their means). These analyses confirmed that the measurements were highly reproducible across different clinicians. Specifically, the interobserver agreement for calcified plaque volume on coronary CT angiography was excellent, with an intraclass correlation coefficient of 0.91. The consistency was even higher for non-contrast imaging, where the interobserver agreement for calcified plaque volume on non-contrast CT reached an intraclass correlation coefficient of 0.98. These high levels of reproducibility suggest that the detection gap between the two modalities is a result of the inherent limitations of contrast-enhanced imaging rather than observer error. For the practicing clinician, this highlights the potential role of thin-slice non-contrast CT as a supplemental tool for more accurate plaque quantification. By utilizing deep learning-aided registration to combine non-contrast and contrast-enhanced images, physicians can achieve a more comprehensive assessment of total plaque burden. This multimodal approach addresses the risk of underestimating disease in the nearly 9% of patients who might otherwise be classified as having no calcified plaques based on angiography alone. The study concludes that integrating thin-slice non-contrast CT into the diagnostic workflow could provide more accurate data to support clinical decision-making, particularly for risk stratification and the initiation of preventive therapies. While these findings demonstrate that angiography misses more than one third of calcified plaques, future studies are needed to determine how this improved detection directly influences long-term patient outcomes and treatment pathways. For now, the data suggest that thin-slice non-contrast CT is an underutilized resource that can refine the assessment of coronary artery disease beyond the capabilities of standard angiography.

References

1. Mach F, Baigent C, Catapano AL, et al. 2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk. European Heart Journal. 2019. doi:10.1093/eurheartj/ehz455

2. Visseren FL, Mach F, Smulders YM, et al. 2021 ESC Guidelines on cardiovascular disease prevention in clinical practice. European Heart Journal. 2021. doi:10.1093/eurheartj/ehab484

3. Bell JS, Weir-McCall J, Nicol E, Lip GYH, Nørgaard BL, Fairbairn TA. Plaque quantification from coronary computed tomography angiography in predicting cardiovascular events: A systematic review and meta-analysis.. Journal of cardiovascular computed tomography. 2025. doi:10.1016/j.jcct.2025.05.003

4. Windecker S, Windecker S, Kolh P, et al. 2014 ESC/EACTS Guidelines on myocardial revascularization. European Heart Journal. 2014. doi:10.1093/eurheartj/ehu278