Purpose: Comparison of a breath-hold, velocity-encoded, phase-difference magnetic resonance (MR) sequence for intracardiac shunt flow measurement with the invasive shunt size determination by oximetry.
Patients and methods: 10 patients with different cardiac shunts (6 ASD/3 VSD/1 PDA) and four healthy individuals were investigated using a 1.5 Tesla Siemens Vision system. For flow measurements a breath-hold, velocity-encoded, phase-difference magnetic resonance (MR) sequence was used ('through plane', FLASH 2D-sequence, TR/TE 110/5 ms, 'velocity encoding' 250 cm/s). Mean flow (ml/R-R interval) in the ascending aorta, the pulmonary trunk, and the right and left pulmonary arteries was determined. The ratio of the mean flow in the pulmonary circulation (Qp: sum of the mean flows in the right and left pulmonary arteries) and the systemic circulation (Qs: mean flow in the proximal aorta) was compared with the Qp/Qs ratios determined by the invasive oxymetric technique. Oximetry was performed within 24 hours after MR imaging.
Results: In the 4 healthy individuals MR flow measurement yielded a Qp/Qs ratio of 0.96 +/- 0.15. In the 10 patients with the various shunt defects, the non-invasive shunt determination by MR gave a Qp/Qs ratio of 2.09 +/- 0.67. The percentage of the calculated shunt sizes was 47.05 +/- 17.45%. In the comparison with the results determined by the invasive oxymetric technique, the MR data showed a strong correlation of r = 0.87.
Conclusions: Breath-hold, velocity-encoded, phase-difference MR-technique enables a reliable quantification of cardiac shunts within a short acquisition time.