The association between angiographically derived radial wall strain and the risk of in-stent restenosis

In-stent restenosis (ISR) remains a major stent failure post-PCI. This study aimed to explore the predictive capability of pre-PCI baseline angiographically-derived radial wall strain (RWS) for the development of ISR in self-controlled patients. In this retrospective self-controlled study, we included 64 patients (146 stented lesions) with multivessel PCI in whom at least one vessel developed ISR and at least one remained free of ISR, allowing within-patient comparison. Pre-PCI baseline RWS was measured for all stented lesions in each patient. Lesion-level maximal RWS (RWS_max) was defined as the highest RWS in the stenotic segment. The incremental discriminant and reclassification abilities for ISR prediction were compared between 2 models (Model 1: Angiography Characteristics + Stent Characteristics; Model 2: Model 1 + RWS_max). Among all lesions, 76 (52.1%) lesions developed ISR (ISR group), while 70 (47.9%) lesions maintained luminal patency (non-ISR group) over a median follow-up duration of 21 months (IQR: 13.0-39.3). The baseline RWS_max was significantly higher in the ISR group compared to the non-ISR group (15.0% vs. 12.4%, P < 0.001). RWS_max demonstrated a strong predictive value for ISR, with an area under the curve (AUC) of 0.82 (95% CI: 0.75-0.88, P < 0.001) and an optimal cutoff of > 14.5%. RWS_max >14.5% was independently associated with increased risk of ISR, with an odds ratio of 8.74 (95% CI: 6.73-24.52, P < 0.001). Compared with Model 1, Model 2 exhibited superior discriminative ability (C-index: 0.852 vs. 0.713; P < 0.001) and higher reclassification ability (NRI: 0.471, P < 0.001; relative IDI: 0.347; P = 0.018) for predicting ISR. In this retrospective, self-controlled study, baseline angiographically-derived RWS is a reliable, independent biomechanical marker that can effectively predict the risk of ISR.