Summary: Fourteen patients were studied by positron emission tomography (PET) within 48 h of onset of a hemispheric ischemic stroke and again 7 days later. After the first set of PET scans, the patients were randomized to receive either nimodipine (n = 7) or a carrier solution (n = 7) by intravenous infusion. The infusions were main tained until the end of the second PET studies. CBF, cerebral blood volume (CBV), oxygen extraction ratio (OER), CMR02, and CMRglc were measured each time. These metabolic and perfusion measurements were per formed by standard methods. A surface map of each met abolic and perfusion measurement in the cortical mantle was generated by interpolating between the available slices. The various surface maps representing the physi ological characteristics determined in the same or subse quent studies were aligned so that all data sets could be analyzed identically using an array of square regions of interest (ROIs). The functional status of each ROI was recorded at the two intervals following the cerebrovascu- 523 lar accident to characterize the evolution of the infarct, penumbra, and normal brain regions. We presumed the ischemic penumbra to be cortical regions in the proximity of the infarct and perfused at CBF values between 12 and 18 ml/loo g/min on the first PET scan, while densely isch emic regions had CBF of <12 nlllOO glmin and normally perfused brain> 18 ml/loo g/min. In the densely ischemic zone, CBF increased more in the nimodipine-treated group than in the carrier group. As well, in this region nimodipine reversed the decline in CMR02 noted in the carrier group, the difference in the changes being signif icant. In the penumbra zone, comparable trends were noted in OER and CMR02 but the difference in the changes between the two groups did not reach statistical significance. Changes in CMRglc and CBV were compa rable between the two groups in both cortical regions.

In prolonged fasting, the brain derives a large portion of its oxidative energy from the ketone bodies, beta-hydroxybutyrate and acetoacetate, thereby reducing whole body glucose consumption. Energy substrate utilization differs regionally in the brain of fasting rat, but comparable information has hitherto been unavailable in humans. We used positron emission tomography (PET) to study regional brain glucose and oxygen metabolism, blood flow, and blood volume in four obese subjects before and after a 3-wk total fast. Whole brain glucose utilization fell to 54% of control (postabsorptive) values (P less than 0.002). The whole brain rate constant for glucose tracer phosphorylation fell to 51% of control values (P less than 0.002). Both parameters decreased uniformly throughout the brain. The 2-fluoro-2-deoxy-D-glucose lumped constant decreased from a control value of 0.57 to 0.43 (P less than 0.01). Regional blood-brain barrier transfer coefficients for glucose tracer, regional oxygen utilization, blood flow, and blood volume were unchanged.

Twelve patients were studied within 48 hours of stroke using positron emission tomography to determine cerebral blood flow (CBF), cerebral metabolic rate for oxygen (CMRO2), oxygen extraction fraction, cerebral blood volume, cerebral pH (CpH), and cerebral metabolic rate for glucose (CMRGlc), the last calculated using published normal rate constants (CMRGlc[N]) and those for severe ischemia. In these studies, a cortical region of severe ischemia (I) was outlined, its metabolic and perfusion properties evaluated, and its length measured. The contralateral uninvolved cortical rim (C) in these patients and the cortical rim in 5 older normal patients were used for comparison. The length of region I correlated with the neurological deficit measured independently by a clinical scoring method. The 12 patients fell into two groups: Group I (8 patients), whose CBF in I was 9.3 +/- 2.5 ml/100 gm/min (mean +/- SEM) and was in every patient lower than that in C (33.1 +/- 2.2), and Group 2 (4 patients), whose CBF in I was 42.1 +/- 8.5 ml/100 gm/min and was in every case higher than that in C (28.2 +/- 1.5). In Group I, region I showed a CMRGlc(N)/CMRO2 ratio of 0.22 +/- 0.06 and a CpH of 6.83 +/- 0.06. In Group 2, the same ratio in the region I was 0.58 +/- 0.09 and the CpH was 7.12 +/- 0.05. The CMRGlc (N)/CMRO2 ratio for the contralateral hemisphere was comparable in the two groups. Our data suggest that, within 48 hours of the clinical onset of stroke, the ischemic cortex is already reperfused in one third of patients. Those ischemic regions with persistent hypoperfusion appear acidotic, whereas in the reperfused regions, despite evidence of an increased CMRGlc/CMRO2 ratio, acidosis is not evident. The implications of these findings for therapies proposed in acute human cerebral ischemia are discussed.

Initial experience is reported with the Scanditronix PC 2048-15B, a 15-slice positron emission tomography (PET) system using multicrystal/multiphoto-multiplier modules to obtain high spatial resolution. Random and scattered events are reduced using an orbiting (68)Ge rod source for transmission scans by only accepting coincidence lines which intersect the instantaneous position of the source. Scatter correction of the emission data is removed with a deconvolution kernel, random and dead-time correction by the use of observed singles rates. The peak count rates are 11.7/20.0 Kcps for the direct cross slices at concentrations of 4.5/5.1 muCi/cc. respectively. Over the radial range 0-9 cm from the ring center, radial transverse resolution is 4.6-6.4 mm, aid tangential transverse resolution is 4.6-5.1 mm using a Hanning filter. Over the same range, axial resolution varies from 6.1-6.2 mm in direct slices and from 5.4-7.1 mm in cross slices. This near-isotropic resolution allows collection of image volume data with no preferred direction for signal averaging errors.