Oxygenation-Sensitive CMR Uncovers Distinct Myocardial Abnormalities in MINOCA: A Pilot Study

Background: Myocardial infarction with non-obstructive coronary arteries (MINOCA) is a heterogeneous condition, accounting for up to 15% of acute infarctions¹, yet its underlying mechanisms remain incompletely understood. Coronary microvascular dysfunction (CMD) has emerged as a central contributor to ischemia in the absence of obstructive disease.² However, definitive diagnosis of CMD relies on invasive coronary function testing.³ Although considered the reference standard, these procedures are technically demanding, not widely available, leaving a critical gap in the evaluation of MINOCA patients. Oxygen-sensitive cardiovascular magnetic resonance (OS-CMR) offers a promising non-invasive method to quantify myocardial oxygenation in vivo and has shown utility in detecting CMD.^3,4

Methods: We conducted a single-center prospective pilot study including 20 age-matched women (7 MINOCA, 7 healthy controls and 6 MI-CAD) with a mean age of 57 ± 9 years. All underwent OS-CMR at 3T using a standardized vasoactive breathing protocol consisting of a 60 second period of paced hyperventilation followed by a 60 second breath-hold.³ OS-CMR imaging was acquired in two short-axis slices (basal and mid-ventricular views). Image analysis was performed with the CVI42 OxySense module, and quantitative oxygenation biomarkers were extracted using a custom post-processing MATLAB code developed by Area19 Medical. Principal component analysis (PCA) was applied to identify the biomarkers with the most discriminative power.

Results: In total, 80 OS-CMR biomarkers showed significant group-level differences. PCA reduced dimensionality to 10 components, with the first 4 explaining 76.4% of total variance. The strongest discriminators were local variance features (maximum local variance = .97, 90th percentile of local variance = .96) and oxygenation-curve difference measures capturing the change from the end of hyperventilation to the beginning of breath-hold (area under the curve = .95, diastolic mean intensity = .93). These components reflect regional heterogeneity and dynamic oxygenation shifts, highlighting that MINOCA exhibits distinct myocardial oxygenation patterns compared to both healthy and MI-CAD patients.

Conclusion: This pilot study, OS-CMR combined with PCA revealed that distinct abnormalities in regional heterogeneity and dynamic oxygenation shifts may underlie MINOCA pathophysiology. While hypothesis-generating, this work highlights the potential of OS-CMR biomarker profiling to identify disease-specific signatures and advance mechanistic understanding of this complex condition.

OS-CMR biomarkers of regional heterogeneity and dynamic oxygenation shifts in Healthy, MINOCA, and MI-CAD patients