Title: Validation of magnetic resonance imaging-derived venous oxygen saturation and oxygen consumption measurements during exercise
Authors: Skow, RJ
Foulkes, SJ
Mugele, H
Perkins, DR
Lawley, JS
Tomczak, CR
Nelson, MD
La Gerche, A
Haykowsky, MJ
Thompson, RB
Issue Year: 2025
Publisher AMER PHYSIOLOGICAL SOC
Series AMERICAN JOURNAL OF PHYSIOLOGY-HEART AND CIRCULATORY PHYSIOLOGY:
Abstract Accurate measurement of venous oxygen saturation (Svo(2)) is fundamental for quantifying oxygen consumption (Vo(2)) and its Fick determinants. This study evaluated the validity of a novel magnetic resonance imaging (MRI) approach using susceptometry-based oximetry (SBO) to noninvasively measure Svo(2) during exercise. First, Svo(2) was measured from an antecubital vein in 14 adults (26 +/- 5 yr; 8 males) via portable blood analyzer (iSTAT) at rest and during two 5-min bouts of rhythmic handgrip exercise at 30% of maximum voluntary contraction, with temporally matched SBO measures of Svo(2) during an identical in-scanner exercise protocol. Separately, 21 adults (27 +/- 6 yr; 8 males) completed incremental step ergometry exercise with simultaneous pulmonary Vo(2) measurement (indirect calorimetry) and MRI-derived Vo(2) from the Fick determinants (MRIFick), including inferior vena cava Svo(2)(SBO) and cardiac output (cardiac MRI). Agreement between methods was assessed by interclass correlation coefficients (ICCs), linear regression, and Bland-Altman plots. There was good to excellent agreement between iSTAT and SBO Svo(2) for arm exercise (ICC: 0.86, 95% CI: 0.71-0.93, P < 0.01; mean bias: +5.2%). Test-retest reliability demonstrated excellent agreement for Svo(2) measures between trials using iSTAT (ICC: 0.97; 95% CI: 0.93-0.96, P < 0.01; mean bias: -2.7%) and SBO (ICC: 0.90, 95% CI: 0.80-0.95, P < 0.01; mean bias: -0.3%), Vo(2) calculated from MRIFick demonstrated excellent agreement with pulmonary Vo(2) (ICC: 0.98, 95% CI: 0.97-0.99, P < 0.01; mean bias: +0.10 L/min). MRI-SBO-derived Svo(2) measurements demonstrated excellent reliability and strong agreement with both venous measures and respiratory gas analysis, validating this technique across a wide range of Svo(2) values during exercise. NEW & NOTEWORTHY Noninvasive MRI susceptometry-based Svo(2) measurements demonstrated excellent repeatability and validity against established direct blood sampling during localized handgrip exercise. The technique's accuracy for determining Vo(2) during step ergometer exercise was validated through strong agreement between combined cardiac MRI and susceptometry-based Fick calculations and reference pulmonary measures, establishing this method as a reliable tool for noninvasive assessment of oxygen uptake and its Fick determinants across varying exercise intensities.
URI: https://publications.svi.edu.au/publications/10152
Other Identifiers 10.1152/ajpheart.00134.2025
Publication type Article