Paradoxical reduction of cerebral blood flow after acetazolamide loading:a hemodynamic and metabolic study with ^(15)O PET

Paradoxical reduction of cerebral blood flow （CBF） after administration of the vasodilator acetazolamide is the most severe stage of cerebrovascular reactivity failure and is often associated with an increased oxygen extraction fraction （OEF）. In this study, we aimed to reveal the mechanism underlying this phenomenon by focusing on the ratio of CBF to cerebral blood volume （CBV） as a marker of regional cerebral perfusion pressure （CPP）. In 37 patients with unilateral internal carotid or middle cerebral arterial （MCA） steno-occlusive disease and 8 normal controls, the baseline CBF （CBFb）, CBV, OEF, cerebral oxygen metabolic rate （CMRO2）, and CBF after acetazolamide loading in the anterior and posterior MCA territories were measured by （15）~O positron emission tomography. Paradoxical CBF reduction was found in 28 of 74 regions （18 of 37 patients） in the ipsilateral hemisphere. High CBFb （〉47.6 mL/100 mL/min, n = 7） was associated with normal CBFb/CBV, increased CBV, decreased OEF, and normal CMRO2. Low CBFb （〈31.8 mL/100 mL/min, n = 9） was associated with decreased CBFb/CBV, increased CBV, increased OEF, and decreased CMRO2. These findings demonstrated that paradoxical CBF reduction is not always associated with reduction of CPP, but partly includes high- CBFb regions with normal CPP, which has not been described in previous studies.

CBF/CBV maps in normal volunteers studied with ^(15)O PET:a possible index of cerebral perfusion pressure

Local cerebral perfusion pressure （CPP） is a primary factor controlling cerebral circulation and previous studies have indicated that the ratio of cerebral blood flow （CBF） to cerebral blood volume （CBV） can be used as an index of the local CPP. In this study, we investigated whether the CBF/CBV ratio differs among different brain structures under physiological conditions, by means of （15）O positron emission tomography. Nine healthy volunteers （5 men and 4 women; mean age, 47.0 ± 1.2 years） were studied by H2（15）O bolus injection for CBF measurement and by C（15）O inhalation for CBV measurement. The CBF/CBV ratio maps were created by dividing the CBF images by the CBV images after anatomical normalization. Regions of interest were placed on the CBF/CBV maps and comparing the regions. The mean CBF/ CBV ratio was highest in the cerebellum （19.3 ＋ 5.2/ min）, followed by the putamen （18.2 ± 3.9）, pons （16.4 ±4.6）, thalamus （14.5：1： 3.3）, cerebral cortices （13.2 ±2.4）, and centrum semiovale （1115 ± 2.1）. The cerebellum and putamen showed significantly higher CBF/CBV ratios than the cerebral cortices and centrum semiovale. We created maps of the CBF/ CBV ratio in normal volunteers and demonstrated higher CBF/CBV ratios in the cerebellum and putamen than in the cerebral cortices and deep cerebral white matter. The CBF/CBV may reflect the local CPP and should be studied in hemodynamicallycompromised patients and in patients with risk factors for small-artery diseases of the brain. Keywords： cerebral perfusion pressure; cerebra blood flow; cerebral blood volume; H2（15）O; C（15）O