Computed tomography for prediction of stent underexpansion after percutaneous recanalization of calcified coronary chronic total occlusions

Abstract Background Coronary calcifications are the leading cause of stent underexpansion after percutaneous coronary intervention (PCI) and are commonly encountered in chronic total occlusions (CTO). Coronary computed tomography angiography (CCTA) is uniquely suited for pre-procedural assessment of coronary calcifications. Recently quantitative CCTA, providing exact volumes of each plaque component, has shown additional value in future risk assessment compared with conventional qualitative CCTA. Purpose The aim of this study was to evaluate whether CCTA predicts intravascular ultrasound (IVUS)-defined suboptimal stent expansion following percutaneous recanalization of calcified CTO lesions. Methods We retrospectively identified consecutive patients with a CCTA performed within 6-months prior to successful PCI of a calcified CTO (defined as calcification within the target lesion on CCTA). All patients had final IVUS of implanted stents (Figure 1). Patients requiring atherectomy were excluded. The primary outcome was IVUS-derived stent expansion index (SEI), defined as stent area divided by vessel area (both measured at the site of maximum calcium). SEI<0.4 indicated suboptimal stent expansion. The site of maximum calcium was a cross-section with greatest calcium arc on pre-stent IVUS within the target lesion. CCTA calcium analysis was performed by a researcher blinded to the IVUS and angiography results. Measurements included total-, vessel-, and lesion-specific calcium score (CAC), total calcium length, maximum calcium thickness and maximum CT calcium attenuation at the target lesion defined as occlusion site and adjacent severely diseased segments. Calcium quantification was performed using a semi-automated software to obtain volumes and burdens of calcified and non-calcified plaque components within the target lesion. Results The cohort included 43 patients (age 64.9±10.0 years, 88% males) with 43 CTO lesions. Mean final stent area at maximum calcium by IVUS was 8.2±2.1mm². Mean SEI was 0.47±0.1, and suboptimal SEI of <0.4 was found in 10 (23.3%) lesions. Lesions with suboptimal stent expansion (SEI<0.4) were characterized by numerically higher total-, vessel- and lesion-specific CAC, higher maximum calcium attenuation, greater calcium length and remodeling index compared with patients with optimal stent expansion (Table 1). However, only calcified plaque volume (105 [IQR 27-271] vs. 29 [4-96]mm³, P=0.02) and calcified plaque burden (16 [7-24] vs. 6 [1-10]%, P=0.03) were significantly greater in lesions with suboptimal stent expansion compared with lesions with optimal stent expansion, respectively. Calcified plaque volume ≥51 mm³ within the target lesion predicted suboptimal stent expansion with 70% sensitivity and 62.5% specificity (AUC 0.74). Conclusion CCTA-derived calcium volume identifies CTO lesions at greater risk of stent underexpansion. Such lesions require more aggressive plaque modification prior to stent implantation.