Quantitative assessment of cerebral blood flow measurement using the microsphere model with N-isopropyl-p-[123I]iodoamphetamine SPECT. Simulation analysis for the influence of washout from brain tissue

The microsphere model and the continuous withdrawal of arterial blood have been commonly used in clinical studies when measuring regional cerebral blood flow (rCBF) by N-isopropyl-p-[123I]iodoamphetamine (IMP) SPECT. The method is considered to underestimate rCBF because of the washout of the tracer from brain tissue with increasing time after injection; however, the extent of this underestimation is not known. To assess whether this underestimation can be quantitatively determined and until how long after injection the microsphere model can be applied to obtain accurate measurements of rCBF, we performed a simulation analysis. Simulation was based on the microsphere model and the 2-compartment model using the standard time-activity course of 123I-IMP in arterial blood (standard input function), the accuracy of which could be validated. The values of rCBF calculated by the microsphere method following the injection of 123I-IMP were compared with those given in the 2-compartment model [influx; K1(rCBF) and outflux; k2 (washout)]. As results, with the microsphere method, rCBF values became lower with the time after injection. The rate of underestimation of rCBF was determined to be 4.6% at 5 min, 10.2% at 10 min and 15.1% at 15 min. Since 4.6% is considered negligible in clinical studies, we concluded that the microsphere model could be applied to obtain accurate measurements of rCBF up to approximately 5 min, regardless of washout. The results of this simulation were confirmed clinically in 11 patients with various cerebral diseases. (author)