Virtual stenting and coregistration of instantaneous wave-free ratio to predict the physiological effect of percutaneous coronary intervention in patients with multilevel coronary artery disease

Abstract

Aim. To evaluate the accuracy of virtual stenting in predicting the physiological effect of percutaneous coronary intervention (PCI) for multilevel coronary artery disease (CAD).Material and methods. In order to define PCI strategy, 34 patients with multilevel CAD underwent assessment of instantaneous wave-free ratio (iFR) with mapping. Using the virtual stenting, predicted iFR (priFR) was assessed. After stent implantation, the factual iFR (fiFR) was re-assessed. A discrepancy between the priFR and fiFR by 0,03 was considered the threshold level of a significant difference and "poor agreement" criterion between measurements.Results. Mean iFR before PCI was 0,77±0,11; after PCI, this indicator increased significantly and amounted to 0,94±0,04 (p<0,001). Comparison of the predicted and factual iFR revealed that the difference did not significantly depend on the value (rxy=-0,183; p=0,300), and mean difference between measurements was 0,013 (standard deviation, ±0,019), which indicates no systematic discrepancy and good comparability of the studied methods. In addition, correlation analysis of priFR and fiFR revealed a significant strong (Chaddock scale) direct relationship (r=0,854; p<0,001). Independent predictors of "poor agreement" were heart rate and systolic blood pressure before surgery. Discrepancy in decisions on the choice of stented lesions and the required number of stents based on coronary angiography (CAG) and MRI occurred in 21 patients (62%) and 16 patients (47%), respectively. After iFR measurement, there was a significant decrease in the mean length of affected segment (from 61,82±19,68 mm (CAG) and 49,15±19,19 mm (iFR)), which made it possible to significantly reduce the number of implanted stents from 1,91±0,57 (CAG) to 1,50±0,56 (p=0,001).Conclusion. Mapping the iFR and virtual stenting technology makes it possible to revise the classification of coronary lesions by length, which leads to a significant reduction in the number of implanted stents and the length of stented segment. High correlation between priFR and fiFR achieved in our study indicates the high accuracy of virtual stenting in predicting the physiological effect of stenting.