SCMR 22nd Annual Scientific Sessions
Differential right ventricle (RV) to left ventricle (LV) oxygen saturation (ΔSaO2) measures the oxygenation difference between the blood in RV and LV, which can be an important indicator of cardiopulmonary function for assessing cardiac function in heart failure patients, and to measure shunt fractions in patients with congenital or acquired heart disease. Cardiac quantatitiave susceptiblity mapping (QSM) is an emerging technique for non-invasive qualtifacation of ΔSaO2. Here, we present our preliminary data in validation of QSM derived ΔSaO2 via comparison with right heart catheterization (RHC) measurements in patients.
QSM was acquired in 22 patients via an ECG-triggered navigator-gated free-breathing multi-echo 3D GRE sequence. QSM was performed on a clinical 3T scanner (GE 750W). QSM maps were reconstructed by first preparing the total field via graph cut phase analysis and chemical shift update methods, and then computing the susceptibility map using a total field inversion method. In this work, a method, termed TFI+0, was introduced that used 2 additional regularizations terms, one for each of the heart ventricles, to restrict the susceptibility variations within the right and left ventricle blood pools. Once the QSM map was reconstructed, the susceptibility difference between RV and LV blood pools can then be converted to ΔSaO2 by scaling such susceptibility difference with heme concentration in blood and molar susceptibility of deoxyheme. In 5 patients, an RHC based ΔSaO2 measurement was performed as a reference.
82% (18/22) of cases yielded interpretable QSM maps (acquisition time 433±142.8s, typical resolution1.5x1.5x5 mm3). In a representative patient case, the QSM map generated with TFI+0 showed a stronger and more uniform contrast between RV and LV. The ΔSaO2 from QSM generated with TFI+0 (35.7%) also agree better with the RHC measurement (40%) than the from QSM generated with TFI (27%) (Figure 1A). In the 5 patients with RHC reference, ΔSaO2 from TFI+0 (34.0±2.1%) matched more closely with the RHC ΔSaO2 (36.6±3.9%, p=0.31) than the ΔSaO2 from TFI (24.9±5.6%, p=0.013) (Figure 1B). The average ΔSaO2 measured in the rest of 13 successful cases using TFI+0 was 23.8±5.0%, which is within the physiologically viable range.
In this preliminary validation study, cardiac QSM was successfully obtained in a substantial proportion of patients undergoing contrast-enhanced CMR. Using TFI+0 reconstruction, the QSM yielded differential RV-to-LV blood pool oxygen saturation calculations that agreed well with invasive measurements and physiologic norms.