Contrast-Free Biomarkers for Cardiac Amyloidosis vs Ischemia based on Baseline Oxygen-sensitive cardiovascular MRI (OS-CMR) Scouts: A Pilot Study

Presenting Author(s)

• 

  Daniela Valderrama, MD

  Postdoctoral Researcher
  McGill University Health Center
  Montreal, Quebec, Canada

Primary Author(s)

• 

  Daniela Valderrama, MD

  Postdoctoral Researcher
  McGill University Health Center
  Montreal, Quebec, Canada

Co-Author(s)

• MC

  Michael Chetrit, MD

  Assistant professor
  McGill University Health Center
  Montreal, Quebec, Canada
• MB

  Mitchel Benovoy, PhD

  PhD
  Area 19 Medical Inc.
  Montreal, Quebec, Canada
• 

  Matthias G. Friedrich, MD, FSCMR

  Professor, Chief, Cardiovascular Imaging, McGill University Health Centre
  McGill University Health Center, Research Institute of the McGill University Health Centre
  Montreal, Quebec, Canada

Background: Oxygen-sensitive cardiovascular MRI (OS-CMR) uses the BOLD effect of deoxyhemoglobin to non-invasively evaluate myocardial oxygenation. Infiltrative and ischemic diseases differ in tissue composition and microvascular function, affecting the oxygenation-weighted signal intensity. Other studies use vasodilator stress or breath-hold maneuvers; however, the baseline scouts remain underexplored. We aimed to test whether scout-derived OS-CMR biomarkers and conventional tissue characterization (T1/T2 mapping) can discriminate among groups

Methods: Single-center pilot (n=21; 7 amyloidosis, 7 ischemic cardiomyopathy, 7 healthy controls) at the Royal Victoria Hospital (CMR10 protocol). Baseline OS-CMR scouts and T1/T2 mappings were performed on a 3T GE Premier Scanner and imaging analysis was performed with the cvi42 OxySense and Tissue T1/T2 mapping modules. Quantitative oxygenation biomarkers were extracted using a custom post-processing MATLAB (Area19 Medical). We perform statistical analysis using ANOVA and post-hoc tests with a p-value < 0.05 as a significant difference.

Results: Participants were predominantly male (amyloidosis 6/7, ischemia 7/7, controls 7/7); mean age 69±11.3, 68±8.6, and 60±7.8 years, respectively. Global mean OCG-SI (oxicardiogram) values were 0.344 ± 0.071 SI, 0.263 ± 0.025 SI, and 0.280 ± 0.065 SI, respectively. One-way ANOVA analysis shows the Global mean OCG-SI differed across groups (ANOVA F(2,18)=3.87, p=0.040*). Post-hoc analysis identified a single significant contrast between amyloidosis and ischemia (Δ=0.081SI, p=0.042*). Other biomarkers related to tissue heterogeneity as MeanEntropy and EndDiastolic-Entropy, were explored, showing differences between the groups (ANOVA F(2,18)=4.23, p=0.031* and F(2,18)=5.40, p=0.015*), and post-hoc analysis presented a significant contrast between controls and ischemic cases (Δ=0.257, p=0.047* and Δ=0.414, p=0.011*). Regarding the mapping analysis, Global T2 did not differ across groups (all p >0.33), whereas global T1 was higher in amyloidosis vs ischemia (+166 ms, p=0.001) and vs controls (+188 ms, p< 0.001); ischemia vs controls p=0.687.

Conclusion: Scout-derived OS-CMR biomarkers have the potential to discriminate myocardial phenotypes: global mean OCG-SI is higher in amyloidosis than in ischemia, and entropy (heterogeneity) is lowest in ischemic cases. Additionally, tissue-mapping results align with this pattern, supporting the hypothesis that oxygenation-weighted signal and entropy reflect tissue infiltration with accompanying microvascular dysfunction.

Figure 1. Right Panel. Oxycardiogram (OCG) across the cardiac cycle (blue = Healthy controls, red = Cardiac amyloidosis, yellow = Ischemic cardiomyopathy). Left Panel. Mean Entropy curves across the cardiac cycle