Noninvasive predictors of residual coronary microvascular dysfunction after elective percutaneous coronary intervention

Abstract Backgrounds Coronary computed tomographic angiography (CCTA) provides noninvasive assessments of coronary plaque morphologies and perivascular inflammation. Morphological features associated with residual coronary microvascular dysfunction (CMD) after elective percutaneous coronary intervention (PCI) remains unknown. Purpose This study aimed to examine prevalence and predictors of residual CMD and its phenotypes after elective PCI. Methods We investigated consecutive patients who were scheduled for elective fractional flow reserve (FFR)-guided PCI for de novo single lesions with pre-PCI CCTA data. Pre- and post-PCI comprehensive physiological assessments including coronary flow reserve (CFR) and index of microcirculatory resistance (IMR) were performed by using pressure temperature sensor-tipped guidewire in all patients. Patients with periprocedural myocardial infarction were excluded. Residual CMD was defined as lesions with post-PCI coronary flow reserve (CFR) <2.0, which was further categorized as functional (IMR<25) or structural (IMR ≥25) CMD. Results A total of 198 de novo lesions in 198 patients were included in the final dataset. Pre- and post-PCI mean CFR were 1.79 [1.27, 2.53] and 2.72 [2.00, 4.04], respectively. Residual CMD was observed in 24.7% (49/198) of patients, categorized as functional (n=26) or structural (n=23). Patients with residual CMD showed older age, less frequent diabetes mellitus, current smoking and a culprit lesion in left anterior descending artery (LAD), and higher CCTA-defined calcium score in the culprit vessel compared with those in patients without residual CMD. Multivariable logistic regression analysis showed that age (adjusted odds ratio [OR]:1.04; 95% confidence interval [CI]: 1.00 to 1.08; p=0.04) and non-LAD (adjusted OR: 2.18; 95%CI: 1.03 to 4.63; p=0.04) were independently associated with residual CMD. In patients with residual CMD, patients with structural CMD tended to have more frequent diabetes mellitus (43.5% vs 19.2%, P=0.13) and atrial fibrillation (26.1% vs 7.7%) and showed significantly lower pericoronary adipose tissue attenuation (PCATA) in right coronary artery (RCA) by CCTA (-79.9 HU vs -74.6 HU, P=0.04) compared with patients with functional CMD. CCTA-detected epicardial fat volume and attenuation showed no significant differences between patients with residual functional CMD, those with residual structural CMD, and those without residual CMD. Multivariable analysis revealed that diabetes mellitus (adjusted OR: 5.18; 95%CI: 1.14 to 23.7; p=0.03) and PCATA in RCA (adjusted OR: 0.93; 95% CI: 0.85 to 1.00; p=0.05) were independently predictive of residual structural CMD. Conclusions Residual CMD was observed in approximately 25% of patients undergoing elective PCI. CCTA-defined calcium score and PCATA, along with clinical factors could be potential predictors of post-PCI residual CMD and its phenotypes.