Can automated plaque analysis help the case for CCTA?

An automated algorithm that evaluates coronary computed tomography angiography (CCTA)-derived plaque can better predict major adverse cardiovascular events (MACE) compared to clinical risk factors.

CCTA has proven capable of noninvasively diagnosing patients with high-risk plaque, a potential precursor for coronary artery disease (CAD). However, as authors of a recent study published in the European Journal of Radiology study pointed out, the modality hasn’t reached its full potential. 

“For CCTA to enter the mainstream of diagnostic clinical care, it is necessary to decrease observer variability and automate key parts of the interpretive process to manage the subjectivity, time-consuming nature, and variability of reader interpretation,” wrote first author Marly van Assen, MSc, department of radiology and radiological science, Medical University of South Carolina in Charleston, and colleagues.

In their study, Assen et al. retrospectively included 45 patients with suspected CAD, of which 16 (33%) experienced MACE within one year. All patients underwent CCTA between January 2006 and September 2014 across two centers in the U.S. and Europe.

The researchers used commercially available plaque quantification software to automatically extract quantitative plaque morphology including: lumen area, wall area, stenosis percentage, wall thickness, plaque burden, remodeling ratio, calcified area, lipid rich necrotic core (LRNC) area and matrix area. Assen and colleagues analyzed these markers and traditional risk factors for predicting MACE.

Regression analysis using only clinical risk factors achieved a prognostic accuracy of 63% and area under the ROC curve (AUC) of 0.587. Using the automated morphological features alone, the model notched a 77% accuracy and AUC of 0.94. Both methods combined boosted prognostic accuracy to 87% and led to an AUC of 0.924.

Assen and colleagues noted that their results are comparable to a prior study which used “highly skilled manual assessment” to achieve similar prognostic accuracy.

“In this case however, the calculations were performed with software which has the benefit of being potentially more efficient in clinical workflow, and readers with less specific experience in the assessment may be able to reach the same level of performance as experts,” the authors added.

A strength of this study, according to the group, was the use of automation itself in that it allowed the authors to avoid pre-specified thresholds which often produce results that lack histological validation.

“In conclusion, an automated model based algorithm to evaluate CTA-derived plaque features and quantify morphological features of atherosclerotic plaque increases the ability for MACE prognostication significantly compared to the use of clinical risk factors alone, while also having the potential to reduce observer variability and the time to evaluate patients relative to expert assessment,” the authors wrote.