Kurzfassung

Validation of quantification of pulmonary blood flow with dynamic, contrast-enhanced MRI is still missing. A possible reason certainly lies in difficulties based on the non linear dependence of signal intensity (SI) from contrast agent concentration. Both aspects were addressed in this study.
Nine healthy pigs were examined by first-pass perfusion MRI using Gd-DTPA and H215O PET imaging. Calculations of hemodynamic parameters were based on a one-compartment model (MR) and a two-compartment...Validation of quantification of pulmonary blood flow with dynamic, contrast-enhanced MRI is still missing. A possible reason certainly lies in difficulties based on the non linear dependence of signal intensity (SI) from contrast agent concentration. Both aspects were addressed in this study.
Nine healthy pigs were examined by first-pass perfusion MRI using Gd-DTPA and H215O PET imaging. Calculations of hemodynamic parameters were based on a one-compartment model (MR) and a two-compartment model (PET). Simulations showed a significant error when assuming linear relation between MR SI and contrast agent dose in the arterial input function even at low dosages of 0.025 mmol/kg bodyweight. To correct for non linearity a calibration curve was calculated on the basis of the signal equation. The required accuracy of equation parameters (like longitudinal relaxation time) was evaluated. Error analysis estimates less than 5% over /underestimation of the corrected SI.
Comparison of PET and MR flow values yielded a significant correlation (P<0.001) in dorsal regions where SNR was sufficient. Changes in PBF due to the correction method were significant (P<0.001) and resulted in a better agreement: mean values (standard deviation) in units of mL/min/100 mL lung tissue were 59(15) for PET, 112(28) for uncorrected MRI and 80(21) for corrected MRI.
» weiterlesen» einklappen