Spatial Heterogeneity of Coronary Microvascular Dysfunction in ANOCA: Potential Clinical Impact of Multivessel Concordance in CFR and MRR
T Watanabe, Y Kanaji, E Usui, H Ueno, K Sayama, T Watanabe, R Murakami, S Hiroi, K Hosokawa, T KakutaAbstract
Background
Coronary microvascular dysfunction (CMD) is a key diagnostic target in patients with ischemia and non-obstructive coronary arteries (INOCA/ANOCA). While invasive physiological assessment—including coronary flow reserve (CFR) and the microvascular resistance reserve (MRR)—is gold-standard, clinical evaluations are often limited to the left anterior descending artery (LAD).
Purpose
Whether CMD is a uniform or spatially heterogeneous process across different coronary territories remains poorly understood. We aimed to investigate the concordance of CMD between the LAD and the right coronary artery (RCA) and determine the clinical significance of multivessel (global) versus single-vessel (regional) dysfunction.
Methods
In this retrospective study, 206 consecutive ANOCA patients underwent comprehensive invasive physiological assessment of both the LAD and RCA between September 2022 and November 2025. Impaired microvascular function was defined as CFR <2.5, index of microcirculatory resistance (IMR) ≥25, and MRR <3.0. Patients were categorized into four groups based on the spatial distribution of CMD: concordant positive (both LAD and RCA), LAD-only, RCA-only, and concordant negative.
Results
The prevalence of reduced CFR and MRR in the LAD was 35.1% and 29.1%, respectively. Among these patients, less than half (41.7% for CFR; 45.1% for MRR) demonstrated concordant abnormalities in the RCA. Kappa analysis revealed only modest agreement between the LAD and RCA (κ = 0.34 for CFR; κ = 0.33 for MRR), highlighting significant spatial heterogeneity. Patients with concordant (multivessel) CFR positivity were significantly older and exhibited incrementally higher NT-proBNP levels compared to those with single-vessel or no CMD (157 vs. 110 vs. 85 vs. 55 pg/mL; P=0.006). Similar trends were observed for MRR (P=0.011). Multivariable analysis identified dyslipidemia (OR 2.79, 95% CI 1.10–7.08; P=0.030) and the E/A ratio (OR 5.59, 95% CI 1.56–20.0; P=0.008) as independent predictors of concordant multivessel CMD. Furthermore, in a logistic regression analysis evaluating patients who were positive for both CFR and MRR in either the LAD or RCA, multivariable analysis demonstrated that log NT-proBNP was an independent predictor (OR 1.42, 95% CI 1.04–1.93; P=0.025).
Conclusion
CMD in ANOCA is characterized by high spatial heterogeneity, with diagnostic findings in the LAD frequently failing to represent the entire coronary circulation. Patients with global (multivessel) microvascular involvement exhibit a more advanced clinical phenotype, including higher NT-proBNP levels and diastolic dysfunction. These findings underscore the necessity of multivessel physiological assessment to accurately risk-stratify ANOCA patients and avoid underdiagnosis of extensive microvascular disease.