Quantification of fluid flow in a model and in the cerebral aqueduct in volunteers investigated by MRI

Purpose: To study the feasebility o MRI for quantification of fluid flow in a tube model and the cerebral aqueduct (CA) in volunteers. Volunteers and Methods: All studies were performed on a 1.5 T MR scanner using a head coil and a FLASH 2D phase contrast sequence with a velocity encoding at 20 cm/s. Flow (real value, ml/sec) of a saline fluid was measured in a flexible tube model with different inside diameters: 0.75 - 3 mm. Three flow velocities were given (normal value). To test the reproducibility, three studies were done using a flow of 0.12 or 0.14 ml/sec and a tube diameter of 0.75 and 2.0 mm. The ratio of normal to real flow value was calculated (ideal ratio = 1). MRI of CA and flow quantification was done in 24 volunteers (28 ± 4 years). Results: Using tubes with a diameter of 0.75 and 1.5 mm the real flow was sometimes higher than the velocity encoding of the phase contrast sequences. Because of this measurements of the fluid flow and the flow velocities were impossible. There was agreement for fluid flow quantification in the tube of 3.0 mm and high agreement in the tube of 2.0 mm in diameter with reproducible results. The mean diameter of the CA in normal subjects was 2.0 ± 0.3 mm, the mean cerebral flow was 0.04±0.02 ml/sec and the peak velocity 3.06 ± 1.59 cm/sec. Conclusions: Reliable flow quantification with MRI is feasible if the diameter of the lumen is greater than 1.5 mm, and if the flow velocity is lower than the velocity encoding. In cases of smaller diameters and higher flow velocities the velocity encoding has to be changed. Because of this the quantification seems to be inaccurate in cases of aqueductal stenosis with the method we used. (orig.)