Liver and spleen enhancement after intravenous injection of carboxydextran magnetite: effect of dose, delay of imaging, and field strength in an ex vivo model.

It has been predicted that liver and spleen enhancement after administration of superparamagnetic contrast agents may be different, depending on the strength of the main magnetic field. With the use of an ex vivo model, we investigated at 0.3, 0.5, and 1.5 T the effects on liver and spleen signal intensity of 5, 15, and 45 mumol/kg body weight of dextran magnetite (SHU 555A) in 54 rats. Nine rats served as controls. At different time delays since injection, the animals were killed, and after perfusion with saline, the liver, brain, and spleen were fixed in formalin. The specimens were embedded in an agar gel matrix and imaged with inversion recovery T1-weighted, proton density spin echo, and T2*-weighted gradient recalled echo (GRE) sequences. At each magnetic field strength, peak liver and spleen signal loss increased with increasing dose of the contrast medium. Signal loss was significantly more conspicuous after a dose of 15 than 5 mumol/kg body weight, but not after a dose of 45 compared with 15 mumol/kg. No signal change was observed in the brain. GRE images showed higher enhancement than proton density-weighted spin echo and inversion recovery images but were noisier. The enhancement showed a plateau between 30 min and 24 hours. Only the signal decrease of the liver after a low dose of contrast medium on GRE images was significantly higher (p < 0.01) at 1.5 than at 0.5 and 0.3 T. Other differences in respect to the field strength were less significant (p < 0.05) or nonsignificant. Differences in the spleen enhancement were nonsignificant. SHU 555A at a dose of 15 mumol/kg is an efficient intracellular contrast agent for liver and spleen at low, mid, and high field strength. Proton density spin echo images are probably the sequence of choice to exploit SHU 555A contrast effects and a wide time window for imaging after its intravenous injection does exist.