Background:  Quantification of blood O₂ saturation is essential for the clinical evaluation of patients with heart failure and pulmonary diseases and is routinely performed via cardiac catheterization. In a recent pre-clinical study in Landrace pigs, we were able to preliminary validate a novel blood-oxygen-level-dependent (BOLD) effect for the non-invasive assessment of the O₂ saturation through T2 mapping. A validation against right heart catheterization (RHC) in clinical settings was not performed yet.

Objectives: We evaluated the accuracy the BOLD-T2 mapping to estimate O₂ saturation in the heart against cardiac catheterization in a cohort of cardiovascular patients undergoing a clinically indicated RHC.

Methods: BOLD-T2 mapping was prospectively validated in a clinical cohort of n=11 cardiovascular patients undergoing a clinically indicated RHC. Blood T2 was measured in the ventricles by means of T2 maps using different inter-echo pulse spacing (τ180 = 12, 15, 20, 25 ms) in the short axis view (see attached figure at τ180 = 12). A multiparametric Luz-Meiboom (L-M) model was used to jointly process the data and estimate O₂ saturation. During the invasive RHC central venous blood O2 saturation (ScvO2) was sampled from the right atrium and right ventricle for each patient.

Results: Linear regression analysis and Bland-Altman plot showed a good agreement between non-invasive and invasive O₂ saturation (p<0.05, r²=0.51), as well as narrow limits of agreement (-9.8% und 8.5%, respectively). A semi-automated BOLD-T2 imaging workflow was developed for the clinical translation and will support further clinical studies.

Conclusions: BOLD-T2 mapping is comparable to invasively measured ScvO₂ in a small cohort of cardiovascular patients. This novel technique may add important information in the clinical evaluation of patients with heart failure as well as pulmonary hypertension and further studies are granted.