The purpose of this investigation was to clarify the reproducibility of the values of cerebral blood flow (CBF) measured by the
133Xe inhalation method. Two measurements of regional CBF were made at 13 locations in each hemisphere in 19 subjects with an age range of 19 to 83 years (mean age 51), with an interval of 10 to 70 minutes between the serial runs in each subject. The blood flow of the fast compartment (gray matter) of the brain (F
1) was calculated as well as the initial slope index (ISI), the relative weight (W
1) and the fractional flow (FF
1) of the fast compartment. Because it was found that the maximal count rate per 5 seconds recorded from the region (MC) is an important factor for good reproducibility, only the data obtained from detectors with values of MC in the both measurements of 1000 counts or more were analyzed in this study.
The mean value of hemispheric F
1 of the second measurement was significantly smaller (-7.6 ml) than that of the first measurement (79.3±18.4 ml per 100 gm brain per minute). The mean value of ISI of the second measurement was also significantly smaller (-5.7) than that of the first measurement (57.3±13.8). This decrease of the mean hemispheric F
1 or ISI was especially large in young subjects or subjects with high flow values. There was actually no difference of W
1 and FF
1 between the two measurements. There was no significant difference in PaCO
2 or mean arterial blood pressure in the serial measurements. The reasons why F
1 and ISI decreased are discussed.
The values of variation coefficient (V.C.) of the change of mean hemispheric value from the first to the second measurement were 7.6% (F
1), 13.9% (ISI), 4.1% (FF
1), and 6.4% (W
1). The V.C. of the changes of regional values were 13.8% (F
1), 14.6% (ISI), 5.4% (FF
1), and 9.3% (W
1). The intermeasurement correlation coefficients of mean hemispheric F
1 and ISI were 0.95 and 0.85, respectively. Analysis of the intermeasurement correlations of regional F
1 and ISI also revealed significant correlations, though the degrees of correlation (F
1 0.82, ISI 0.78) were less than those of the mean hemispheric values. From these data we conclude that the large variation of regional CBF values as measured by the
133Xe inhalation method must be taken into account, especially in assessing the effects on CBF of various factors, such as activation, sleep or vasoactive agents.
View full abstract