Local cerebral glucose utilization during intracerebral pH changes.

The effect of ventriculocisternal perfusion with mock CSF with alkaline or acidic pH on the local CMRglu (LCMRglu) in the caudatoputamen was studied in artificially ventilated and relaxed rats. In control rats both lateral cerebral ventricles were perfused with mock CSF at pH 7.4. In the experimental series one cerebral ventricle was infused with normal mock CSF while the other was infused with mock CSF in which the pH was decreased or increased by changing [HCO-3]. LCMRglu was depressed in acidotic brain tissue while it was strongly increased in alkalotic brain tissue. The importance of these alterations in brain glucose metabolism for the homeostatic regulation of brain pH is discussed.

Experimental thromboembolic stroke studied by positron emission tomography: immediate versus delayed reperfusion by fibrinolysis.

Acute obstruction of the middle cerebral artery (MCA) was obtained by injecting a single autologous blood clot into the internal carotid artery of dogs. The technique induced very reproducible unilateral ischemic lesions in the MCA territory; hemorrhagic transformation of the lesions was often seen. The hemodynamic and metabolic effects of blood clot embolism were studied in 35 dogs with positron emission tomography (PET) and the 15O steady-state technique, and compared with a control group of seven intact animals. In the acute phase, the involved brain tissue still had a nearly normal oxygen consumption (-11%) despite the lowered tissue perfusion (-20%) caused by the vascular obstruction. The lowered oxygen availability was compensated by an increased oxygen extraction ratio (+11%). Twenty-four hours after the insult, the hemodynamic situation had barely changed, and the ischemic event had evolved into a brain infarct in which oxygen consumption was clearly lowered (-25%) and accompanied by a significant lowering (-22%) of the oxygen extraction ratio compared with the acute situation. Therapeutic thrombolysis by local administration of streptokinase (500,000 IU), starting 30 min after the insult, was not able to salvage any brain tissue or to ameliorate tissue perfusion despite angiographically confirmed clot lysis. However, when fibrinolytic therapy was started within the first 5 min after the insult, hemispheric blood flow was normalized, and most of the threatened brain tissue was salvaged, as was indicated by its normalized oxygen consumption and oxygen extraction ratio. Early fibrinolysis was accompanied by definite clinical improvement and substantial reduction in the severity of the morphological lesions that were never hemorrhagic.

Tomographic scintigraphy of regional myocardial perfusion.

Estimation of the extent of regional ischemia by scintigraphic methods has been hampered by the geometric constraints of two-dimensional imaging. Myocardial perfusion scintigraphy was performed using the Fresnel zone-plate tomographic camera after the injection of Tc-99m microspheres (20-40 micron) into a coronary artery. Coronary artery occlusion was performed in six dogs by embolization via a catheter guidewire system. Twenty milicuries of Tc-99m microspheres were injected into the left main coronary artery of the six occluded and three unoccluded dogs. Scintigraphy was performed in multiple projections in the living animal. Optical reconstruction of the holographic image provided tomographic gamma images of the heart. Scintigraphy was also performed with an Anger camera for comparison. The extent of the perfusion defect was measured by planimetry and expressed as a percentage of the ventricular area in that projection. The average of the right and left anterior oblique projections provided the most accurate estimate of the size of the perfusion defect (average error: 13.6%; range: 0-38.2%). Fresnel zone-plate imaging provided an accurate in vivo assessment of the extent of altered myocardial perfusion.

Changes in the brain surface pH and cisternal cerebrospinal fluid acid-base variables in respiratory arrest.

Using flat-surface pH electrodes we continuously measured changes in the brain surface pH during respiratory arrest in anesthetized and paralyzed dogs which were previously ventilated with pure oxygen. Respiratory arrest was induced by halting the respirator. The mean arterial PO2 fell from 502.7 +/- 15.9 (1 SD) to 23.7 +/- 18.5, and the mean arterial PCO2 rose from 36.4 +/- 3.5 to 80.4 +/- 7.1 mm Hg, 10 min after asphyxia. The arterial blood pressure increased gradually over several minutes but fell relatively abruptly and profoundly at the end, due to circulatory failure. Initially, and as long as the arterial blood pressure and, therefore, cerebral blood flow were upheld (phase 1), changes in the brain surface pH were small (delta pH/delta t= -0.026 pH unit/min) in spite of severe hypercapnia. When cerebral perfusion pressure fell due to circulatory failure (phase 2), cerebral ischemia occurred and there was an abrupt fall in brain surface pH (delta pH/delta t= -0.067 pH unit/min). Changes in cisternal CSF [H+] grossly underestimated the magnitude of brain surface acidosis during the period of respiratory arrest; the initial difference between the mean brain surface fluid and cisternal CSF [H+] which was 8.9, rose to 15.1 and 47.4 nmol/L, respectively, 5 and 10 min after asphyxia. Changes in sagittal venous blood acid-base variables were more pronounced than those observed in the arterial blood or cisternal CSF; 5 min after respiratory arrest, arterial and sagittal venous blood and cisternal CSF and brain surface pH were 7.20, 7.09, 7.19 and 7.11, respectively. We conclude that (1) in the course of respiratory arrest cerebral outcome can potentially be determined by circulatory failure as evidenced by simultaneous changes in the arterial blood pressure and brain surface pH; (2) cisternal CSF acid-base changes lag behind those on the brain surface and CSF analyses provide unreliable information about the severity of brain acid-base changes during asphyxia; (3) changes in cerebral venous blood acid-base variables best represent the severity of metabolic aberrations in the brain during respiratory arrest.

Quantum noise in fresnel zone plate imaging.

A Fresnel zone plate, used as a coded aperture, offers a great advantage in geometric collection efficiency over a conventional pinhole or collimator. We present a detailed analysis of the signal-to-noise ratio (SNR) of a quantum-limited zone plate camera. The magnitude and spatial distribution of the noise field and its dependence on the source distribution and the characteristics of the optical processing system are derived. It is shown that the largest SNR advantage occurs for a point source, while for very large, uniform sources there may be a slight net disadvantage to using a zone plate. It is also shown that optical processing does not give the highest possible SNR.

Effect of non-respiratory alkalosis on brain tissue and cerebral blood flow in rats with damaged blood-brain barrier.

Acute alterations in plasma bicarbonate concentration have minimal effects on intracerebral pH and cerebral blood flow, perhaps due to blood-brain barrier mechanisms. To test this hypothesis, the consequences of an acute rise in the plasma bicarbonate concentration were studied in anesthetized rats previously subjected to an acute pressure pulse in the carotid system with unilateral damage to the blood-brain barrier. In rats subjected to a "heavy" hypertensive insult, the hemisphere on the side of the lesion showed a lactic acidosis, edema, and a depression of cerebral blood flow. An increase in the plasma bicarbonate concentrations of 15--20 mEq/1 during 35 minutes provoked a marked rise in the total CO2 content of this hemisphere, and a further increase in the lactate concentration, but did not later the brain edema nor affect further the already very low cerebral blood flow. An increase in the lactate concentration and a decrease of cerebral blood flow in the "reference" hemisphere indicated that the lesion was not completely unilateral. In rats subjected to a "moderate" hypertensive insult the changes were less pronounced and statistically not significant for all the parameters. There results illustrate the importance of an intact blood-brain barrier for the maintenance of intracerebral pH in the face of acute alterations in plasma [HCO3]. The impaired cerebral blood flow after an acute hypertensive insult did not appear to be influenced by the intracerebral [HCO3].

Correction of CSF HCO(-3) after its experimental increase in normocapnia. Role of plasma HCO(-3).

In anesthetized normocapnic dogs CSF [HCO(-3)] was increased to 33 mmol/l by perfusing the brain ventricles for 45 min with a mock CSF containing a high [HCO(-3)]. In dogs with normal plasma [HCO(-3)], CSF [HCO(-3)]fell by ca. 7 mmol/l in 2 h following the end of the perfusion. Lowering plasma [HCO(-3)] to 11 mmol/l by infusing HCl intravenously was without effect but increasing plasma [HCO(-3)] to 36 mmol/l by infusing Na2CO3 limited the CSF [HCO(-3)] fall to 2.8 mmol/l. It is concluded that correction of CSF [HCO(-3)] is partially dependent on a sufficiently low plasma [HCO(-3)]. The small and persistent fall of CSF [HCO(-3)] which at high plasma [HCO(-3)] occurs against a concentration gradient with blood suggests moreover the contribution of more specific mechanism(s) for lowering CSF [HCO(-3)] after its experimental increase.

Response of local blood flow in the caudate nucleus of the cat to intraventricular administration of histamine.

The effect of intraventricular histamine on blood flow in the caudate nucleus of the cat was studied by means of the hydrogen clearance technique. Bilateral ventriculo-cisternal perfusion was installed. After a control period during which both lateral ventricles were perfused with mock CSF with the same composition, the drug under study was added to one side (experimental side) while the other side was perfused further with the control mock SCF (control side). At each point in time, blood flow at the experimental side was compared to that at the control side. Histamine (10(-3) M) caused a severe vasodilatation and this effect was completely antagonised by the H2-receptor blocker cimetidine (10(-2) M). Cimetidine had no vasoactive effects of itself in the concentration used. The H2-receptor agonist Dimaprit (10(-3) M) had a vasodilator effect although less important than histamine. Indirect evidence was gained that H1-receptors are not active in the vascular bed under study.

Response of local blood flow in the caudate nucleus of the cat to intraventricular administration of carbachol.

The effect of perfusion of the cerebral ventricles with artificial cerebrospinal fluid containing carbachol on the blood flow in the caudate nucleus of the cat and the possibility to inhibit this effect by anticholinergic drugs was studied by means of the hydrogen clearance technique. After a control period during which both lateral ventricles were perfused with artificial CSF of identical composition, the drug under study was added on one side (experimental side) while the other side continued to be perfused with the control artificial CSF (control side). The blood flow on the experimental side and on the control side were compared. A dose dependent response to carbachol was observed. Lower concentrations of carbachol (10(-6) up to 10(-4)M) caused vasodilatation whereas high concentrations (10(-3)M) caused local vasoconstriction. The increase in the local blood flow caused by the low carbachol concentrations was reduced by both atropine (10(-5)M) and hexamethonium (10(-3)M). The fall in CBF observed with the high carbachol concentration was prevented by atropine (10(-5)M). It may be concluded that low, physiologically more meaningful, carbachol concentrations cause a local vasodilatation due to interaction with both muscarinic and nicotinic receptors.

Correction of CSF HCO-3 after its experimental increase in normocapnia: inhibition by acetazolamide.

In anaesthetized normocapnic dogs CSF [HCO-3] was increased to ca 33 mmol/l by perfusing the brain ventricles for 45 min with a mock CSF containing a high [HCO-3] which in addition contained 2.5 mg/ml acetazolamide to inhibit central carbonic anhydrase. In dogs with normal plasma [HCO-3], CSF [HCO-3] fell by 5.4 mmol/l in 2 h following the end of the perfusion. Lowering plasma [HCO-3] to 11 mmol/l by infusing HCl intravenously increased the CSF [HCO-3] fall to 7.5 mmol/l. Increasing plasma [HCO-3] to 36 mmol/l completely impeded the fall in CSF [HCO-3]. It is concluded that in these experiments clearing of HCO-3 from the CSF is critically dependent on plasma [HCO-3]. When the data are compared to those of comparable experiments without intraventricular administration of acetazolamide (Weyne et al. 1982), they indicate that acetazolamide impedes clearing of HCO-3 from CSF at high and at normal plasma [HCO-3] but not at low plasma [HCO-3]. The experiments therefore suggest a dual contribution for the clearing of HCO-3 from the CSF after its experimental increase: diffusion along the CSF-plasma gradient for HCO-3 and a carbonic anhydrase dependent clearing of HCO-3.

Local cerebral glucose utilization in systemic acidosis.

Cerebral glucose metabolism (CMRglu) is decreased during acute and prolonged hypercapnic acidosis and during prolonged metabolic (HCl) acidosis; it is increased in acute (hypocapnic) metabolic acidosis and is not changed in acute isocapnic metabolic acidosis. The alteration in CMRglu can be explained by the changes occurring in intracerebral pH under these experimental conditions. In pontine gray matter, n. tractus solitarii, and n. ambiguus, three structures participating in the neuronal regulation of ventilation, local CMRglu is increased in all acidotic groups, suggesting coupling of function and metabolism at the local level during acidosis-induced hyperventilation.

Effects of acetazolamide on ionic composition of cisternal fluid during acute respiratory acidosis.

We studied the effects of intravenous acetazolamide (50-200 mg/kg) on cerebrospinal fluid (CSF) electrolytes and pH regulation in 10 anesthetized and nephrectomized dogs (group II): acetazolamide was injected at -1 h, and respiratory acidosis was induced at zero time for 6 h. A control group of 10 animals (group I) was treated similarly except that an equal volume of 0.45% saline was injected intravenously instead of acetazolamide. The mean CSF PCO2 values in group I were 49.7 +/- 3.4 (SD), 50.2 +/- 3.6, 92.3 +/- 7.0, 100.3 +/- 8.1, and 97.8 +/- 7.3 Torr, respectively, at -1, 0, 3, 4.5, and 6 h; respective values in group II were 49.8 +/- 2.0, 55.2 +/- 5.2, 95.8 +/- 6.4, 103.1 +/- 16.7, and 104.9 +/- 14.1 Torr. During acute respiratory acidosis CSF [HCO3-] rose progressively with time in group I, and the mean values were 28.1 +/- 1.4 (SD), 29.2 +/- 1.7 and 30.1 +/- 1.9 mmol/l, respectively, 3, 4.5, and 6 h after induction of acidosis; respective values in group II were 28.2 +/- 1.1, 28.3 +/- 0.9, and 28.5 +/- 1.4 mmol/l. Acetazolamide at various doses administered inhibited any further rise in CSF [HCO3-] beyond the 3rd h of acidosis. The lower rise in CSF [HCO3-] in group II could not be ascribed to differences in CSF lactate concentration which changed similarly in both groups. Increments in CSF K+ and phosphate concentrations were significantly higher in the acetazolamide group than in the control group, the former presumably reflecting efflux of K+ from intracellular to extracellular fluid compartment. We conclude that in nephrectomized dogs during acute respiratory acidosis intravenously administered acetazolamide diminishes the rise in CSF [HCO3-], impairs CSF H+ regulation, and increases CSF K+ and phosphate concentrations.

Effects of SITS, an anion transport blocker, on CSF ionic composition in metabolic alkalosis.

Disulfonic stilbenes combine with the carrier protein involved in anion transport and inhibit the exchange of Cl- for HCO3- in a variety of biomembranes. Our aim was to determine whether such a mechanism is operative in the regulation of cerebrospinal fluid (CSF) [HCO3-] in metabolic alkalosis. In anesthetized, curarized, and artificially ventilated dogs either mock CSF (group I, 9 dogs) or mock CSF containing SITS, 4-acetamido-4'-isothiocyanostilbene-2,2'-disulfonic acid (group II, 7 dogs) was periodically injected into both lateral cerebral ventricles. During 6 h of isocapnic metabolic alkalosis, produced by intravenous infusion of Na2CO3 solution, plasma [HCO3-] was increased by approximately 14 meq/l in both groups. In SITS-treated animals the mean cisternal CSF [HCO3-] increased by 7.7 meq/l after 6 h, and this was significantly higher than the respective increment, 3.5 meq/l, noted in the control group. Increments in CSF [HCO3-] in both groups were reciprocated by decrements in CSF [Cl-] with CSF [Na+] remaining unchanged. Cisternal CSF PCO2 and lactate concentrations showed similar increments in both groups. It is hypothesized that in metabolic alkalosis a carrier transports HCO3- out of cerebral fluid in exchange for Cl- and that SITS inhibits this mechanism. The efflux of HCO3- out of CSF in metabolic alkalosis would minimize the rise in CSF [HCO3-] brought about by HCO3-] influx from blood into CSF and therefore contributes to the CSF [H+] homeostasis.