Back

Geometry Of The Carotid Artery At Baseline Improves Prediction Of Stenosis Severity At Follow-Up

Kamenskiy, A.; Poulson, W.; MacTaggart, J.

2025-12-02 cardiovascular medicine
10.64898/2025.11.28.25341216 medRxiv
Show abstract

IntroductionCarotid artery geometry may contribute to the development of atherosclerotic carotid artery disease (ACD). Prior reports associated ACD with narrow bifurcation angles, tortuous Common Carotid Arteries (CCA), and large bulbs. Our objective was to determine whether deviations in these geometric features at baseline can help predict stenosis severity at follow-up. MethodsA total of 20 human subjects (59{+/-}16 years old, range 31-85 years) with a 4.7{+/-}2.3 year (range 2.0-11.4 years) follow-up CTA imaging examination were identified, and n=36 carotid arteries that had no prior interventions were selected for analysis. Three-dimensional reconstructions of carotid arteries at baseline and follow-up were used to measure bifurcation angle, CCA tortuosity, bulb area, and stenosis severity, and these measurements were compared to previously reported values for healthy carotid arteries undergoing normal aging. Geometric factors were used along with clinical risk factors to predict severity of stenosis at follow-up using multivariate regression. ResultsStenosis averaged 39{+/-}26% at baseline, and 44{+/-}28% at follow-up. The presence of coronary artery disease, hypertension, female gender, and deviation of bifurcation angle and tortuosity at baseline explained 67% of stenosis severity at follow-up. Clinical risk factors positively contributed to stenosis severity but when used alone explained only 47% of variability. ConclusionsNarrow bifurcation angles and tortuous CCAs are associated with increased stenosis severity over time. Combining morphometric data with clinical risk factors may improve prediction of stenosis severity at follow-up, helping identify patients at higher risk for ACD and potentially allowing more accurate and personalized treatment recommendations.

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.