Prediction of no-reflow phenomenon in elective percutaneous coronary intervention for non-acute myocardial infarction

While the ‘no-reflow’ phenomenon is a well-recognized complication of acute myocardial infarction (AMI), its incidence, predictors, and clinical implications in patients undergoing elective percutaneous coronary intervention (PCI) for non-AMI remain insufficiently characterized. We investigated the clinical and procedural predictors of the no-reflow in patients undergoing elective PCI for non-AMI. Among 10,801 patients in a single-center prospective PCI registry (2013–2024), 3,904 patients with non-AMI who underwent elective PCI were included. No-reflow was defined as transient or persistent thrombolysis in myocardial flow < 3 without proximal obstruction. Independent predictors were identified using multivariable logistic regression with cluster-robust standard errors. Machine learning (ML) models were also developed to assess predictive performance. No-reflow was observed in 94 (2.4%) patients. No-reflow was associated with more complex lesions and higher rates of procedural complication rates. Independent predictors of no-reflow included atrial fibrillation (odds ratio [OR] 2.20, 95% confidence interval [CI] 1.08–4.49, p = 0.031), male sex (OR 2.08, 95% CI 1.15–3.74, p = 0.015), and a device-to-vessel ratio > 1.1 (OR 14.4, 95% CI 3.71–56.1, p < 0.001), while current or former smoking was inversely associated (OR 0.53, 95% CI 0.32–0.89, p = 0.017). Although ML models were developed to predict no-reflow, their performance was limited owing to the rarity of events. No-reflow is an infrequent but clinically relevant complication of elective PCI for patients with non-AMI. Atrial fibrillation, male sex, and stent oversizing were independent predictors, whereas smoking history was inversely associated. These findings underscore the importance of procedural planning and individualized risk assessment in non-AMI PCI.