AimsChronic total occlusion (CTO) percutaneous coronary intervention (PCI) is characterized by a low success rate and an increase in complications. This study aimed to explore a new and simple classification method based on plaque composition to predict guidewire (GW) crossing within 30 min of CTO lesions. MethodsThis study consecutively enrolled individuals undergoing attempted PCI of CTO who underwent coronary computed tomographic angiography (CCTA) within 2 months. Lesions were divided into soft and hard CTO groups according to the necrotic core proportion. ResultsIn this study, 207 lesions were divided into soft (20.3%) and hard CTO (79.7%) groups according to a necrotic core percentage cutoff value of 72.7%. The rate of successful GW crossing within 30 min (57.6 vs. 85.7%, p = 0.004) and final success (73.3 vs. 95.2%, p = 0.001) were much lower in the hard CTO group. For patients with hard CTO, previous failed attempt, proximal side branch, bending > 45 degrees calcium >= 50% cross-sectional area (CSA), and distal reference diameter <= 2.5 mm were demonstrated to be associated with GW failure within 30 min. For patients with soft CTO, only blunt entry was proved to be an independent predictive factor of GW failure within 30 min. ConclusionsGrouping CTO lesions according to the proportion of necrotic core is reasonable and necessary in predicting GW crossing within 30 min. A soft CTO with a necrotic core is more likely to be recanalized compared with a hard CTO with fibrous and/or dense calcium. Different plaque types have variable predictive factors.