Comparison of shear stress patterns by the established and advanced reconstruction method incorporating side branches to predict plaque progression

Background: Complete vessel reconstruction (CVR) with incorporation of side branches is essential for accurate evaluation of wall shear stress (WSS) distribution. However, CVR is time consuming and blood flow simulation is computationally expensive, while there is no evidence that WSS computed by CVR, enables better prediction of disease progression compared to WSS derived from the conventional single vessel reconstruction (SVR). We aim to compare the WSS in models reconstructed using the CVR and SVR methods and examine its ability to predict disease progression. Methods: Patients who had baseline and 13-months follow-up intravascular ultrasound (IVUS) imaging (n=19 vessels), and with neoatherosclerotic lesions (n=13 vessels) on optical coherence tomography (OCT) were included in the present analysis. All the studied vessels had at least one side branch with diameter >1mm. 3-dimensional (3D) CVR and SVR were performed and time averaged (TAWSS) and multidirectional WSS were computed using pulsatile blood flow simulation and the performance of both methods in predicting disease progression in IVUS and OCT models were assessed. Results: The incorporation of side branches in 3D geometry resulted in lower TAWSS in the IVUS (0.821 vs 1.698Pa, p<0.001) and OCT-based reconstructions (0.682 vs 1.325Pa, p<0.001) and influenced the multidirectional WSS distribution. In native segments, WSS metrics estimated by the CVR enabled better prediction of the lumen and plaque area and burden at follow-up than SVR and disease progression defined as decrease in lumen area and increase in plaque burden (AUC CVR 0.712 vs SVR 0.554). In stented segments, multidirectional WSS was associated with neointima area in both CVR and SVR methods, but TAWSS was only a predictor of neointima area in the CVR method. Conclusions: The incorporation of side branches in vessel reconstruction influences WSS distribution and enables more accurate prediction of disease progression in native and stented segments than SVR modelling. ### Competing Interest Statement The authors have declared no competing interest.