Development and external validation of a nomogram prediction model based on quantitative coronary angiography for...

Area of Science:

• Cardiology
• Medical Imaging
• Predictive Modeling

Background:

• Quantitative coronary angiography (QCA) is crucial for diagnosing coronary artery disease.
• Accurate prediction of ischemic lesions is essential for effective patient management.

Purpose of the Study:

• To construct and validate a QCA-based nomogram for predicting ischemic lesions (FFR ≤ 0.80).
• To assess the predictive performance of the nomogram in independent patient cohorts.

Main Methods:

• A multi-center study involving 220 patients and 303 vessels.
• Development of a nomogram using Least Absolute Shrinkage and Selection Operator (LASSO) regression on a development set (n=113).
• External validation of the nomogram on an independent set (n=107).

Main Results:

• The nomogram incorporated lesion length, minimal lumen diameter, stenosis flow reserve, visual percent diameter stenosis, and weight.
• High predictive performance was observed, with an AUC of 0.922 (per-vessel) and 0.912 (per-patient) in the development set.
• The nomogram achieved an AUC of 0.915 (per-vessel) and 0.912 (per-patient) in the validation set, with accuracies ranging from 82.2% to 88.2%.

Conclusions:

• The QCA-based nomogram demonstrates significant potential for predicting ischemic lesions.
• External validation confirms its utility in real-world cardiology settings.
• This predictive model can aid clinicians in managing coronary artery disease.