Evaluation of the redistribution in 123I-IMP SPECT in cerebrovascular disease

We evaluated the correlation between redistribution on IMP SPECT and the regional oxygen extraction fraction (rOEF) on PET. Ten lesions that had shown the redistribution phenomenon on IMP SPECT in five patients were selected. In four patient, IMP SPECT was performed within 2 months of the onset of cerebrovascular attack. In one patient, the examination was performed 3 years after the onset. The interval between IMP SPECT and PET was 45 days, except in one patient who was in the chronic phase. The ten lesions were classified into two groups according to rOEF values as follows: Group 1, rOEF≥50% and Group 2, rOEF<50%. In group 1, all lesions did not show low density on final X-ray CT. By contrast, in group 2, three lesions did not show low density on the X-ray CT and the other two lesions did. Redistribution rate (RD rate) was defined as follows, RD rate= {(B-A) /B- (B'-A') /B'} / (B-A) /Bx100 (%) where, A, B, A' and B' are count rate in each of the regions of interest (ROIs). Region of interest (ROI) A was positioned in the lesion on the early image of SPECT that showed redistribution on the delayed image and ROI B was in the opposite side or occipital lobe (normal brain). On the delayed image, ROI A' and B' were positioned at the same place as A and B. ROIs on PET were positioned in the lesion corresponding to ROI A and B on SPECT. In group 1, all RD rates increased more than 70%. By contrast, in group 2, most RD rates were not more than 70%. In group 1, the A'/B' ratio increased more than A/B and was always greater than 90%. It is suggested that the presence of the redistribution phenomenon on IMP SPECT dose not always show the tissue viability. But quantification of the phenomenon by calculating A'/B' or RD rate is useful in evaluating tissue viability of the lesions and in predicting the final outcome of the area. (author)