Various observations on the cerebellar vasoreactivity in crossed cerebellar diaschisis (CCD) have previously been reported. The purpose of this study is to clarify the difference between oxygen-15 H2O positron emission tomographic (PET) and technetium-99m hexamethylpropylene amine oxime (HMPAO) single-photon emission tomograph (SPET) findings in CCD and to evaluate the effect of the absolute values of the cerebellar blood flow as measured by 15O-H2O PET on the 99mTc-HMPAO SPET findings. The subjects comprised 15 patients with a supratentorial infarct and CCD. The cerebellar blood flow increased by about 40% at 5 and 20 min after acetazolamide i.v. on both the CCD and the non-CCD side, as measured by 15O-H2O PET. The percentage differences in cerebellar blood flow between the CCD and the non-CCD side were -22.3%+/-5.7% in the resting state, -19. 6%+/-6.4% at 5 min after acetazolamide i.v. and 21.5%+/-6.7% at 20 min after acetazolamide i.v., as measured by 15O-H2O PET, while they were -10.6%+/-5.5% in the resting state and -5.6%+/-5.1% at 5 min after acetazolamide i.v., as measured by 99mTc-HMPAO SPET. After Lassen's linearization correction, the latter two measurements were -16.2%+/-7.7% and -9.6%+/-8.9%, respectively. The effect of acetazolamide did not differ between the CCD and the non-CCD side in 15O-H2O PET, while a greater response on the CCD side was observed in 99mTc-HMPAO SPET, even after Lassen's linearization correction. It is concluded that acetazolamide HMPAO SPET may overestimate the cerebellar vascular response on the CCD side (or underestimate it on the non-CCD side).