Quantitative Measurements of Cerebral Blood Flow Using SPECT and [^99mTc]-d,l-HM-PAO Compared to Xenon-133

The uptake and retention in a 2 cm thick brain section was recorded serially by SPECT after i.v. injection of [99mTc]– d,l-HM-PAO (HM-PAO). In 16 patients, the fraction of the administered dose retained by the brain was 5.2 ± 1%, showing a peak after 40–50s, then decreasing by 10% within the first 10 min and then by only 0.4% per hour. The image contrast was measured in each patient as the regional hemispheric asymmetry difference in percent of the highest value of the two regions. It deceased from 31% at 30–40 s to 25% at 10 min. At 24 h, a value of 19% was reached. Using the images obtained at 10 min after injection, a region to region comparison of the original and corrected HM-PAO images to the xenon-133 regional cerebral blood flow (rCBF) images was performed. Forty-four patients with stroke, epilepsy, dementia, basal ganglia disease, and tumors and control subjects were included in this comparison. The algorithm proposed by Lassen et al. was used to correct the original images for back diffusion of tracer (brain to blood); a good correlation very close to the line of identity between the corrected HM-PAO and xenon-133 data was obtained when using a conversion/clearance ratio of 1.5 and when the noninvolved hemisphere was used as a reference region ( r = 0.86, p < 0.0001). Serial arterial and cerebral venous blood sampling was performed over 10 min following i.v. injection of HM-PAO in six patients. An overall brain retention fraction of 0.37 ± 0.03 (mean ± SEM) was calculated from the data. An average CBF of 0.62 ± 0.12 ml/g/min was determined on the basis of the Fick principle; this compared to a value of 0.59 ± 0.09 ml/g/min (mean ± SEM) measured by the xenon-133 inhalation method. The two sets of CBF values correlated linearly with a correlation coefficient of 0.97 ( p < 0.01). Inserting the average CBF value for the hemisphere as measured by the Fick principle into the algorithm described by Lassen et al. yields absolute rCBF values (ml/g/min) directly from the corrected HM-PAO images. We conclude that the xenon-133 inhalation method and [99mTc]– d,l-HM-PAO may supplement each other for SPECT studies of rCBF: xenon-133 inhalation for easy, repeatable, and quantitative measurements and [99mTc]– d,l-HM-PAO for high resolution static imaging in relative or absolute flow units.