Effect of insulin hypoglycemia upon cerebral energy metabolism and EEG activity in the rat

Chronic Hyperinsulinaemic Hypoglycaemia in Rats Is Accompanied by Increased Body Weight, Hyperleptinaemia, and Decreased Neuronal Glucose Transporter Levels in the Brain

The brain is vulnerable to hypoglycaemia due to a continuous need of energy substrates to meet its high metabolic demands. Studies have shown that severe acute insulin-induced hypoglycaemia results in oxidative stress in the rat brain, when neuroglycopenia cannot be evaded despite increased levels of cerebral glucose transporters. Compensatory measures in the brain during chronic insulin-induced hypoglycaemia are less well understood. The present study investigated how the brain of nondiabetic rats copes with chronic insulin-induced hypoglycaemia for up to eight weeks. Brain level of different substrate transporters and redox homeostasis was evaluated. Hyperinsulinaemia for 8 weeks consistently lowered blood glucose levels by 30-50% (4-6 mM versus 7-9 mM in controls). The animals had increased food consumption, body weights, and hyperleptinaemia. During infusion, protein levels of the brain neuronal glucose transporter were decreased, whereas levels of lipid peroxidation products were unchanged. Discontinued infusion was followed by transient systemic hyperglycaemia and decreased food consumption and body weight. After 4 weeks, plasma levels of lipid peroxidation products were increased, possibly as a consequence of hyperglycaemia-induced oxidative stress. The present data suggests that chronic moderate hyperinsulinaemic hypoglycaemia causes increased body weight and hyperleptinaemia. This is accompanied by decreased neuronal glucose transporter levels, which may be leptin-induced.

Effect of insulin-induced hypoglycaemia on the central nervous system: evidence from experimental studies

Insulin-induced hypoglycaemia (IIH) is a major acute complication in type 1 as well as in type 2 diabetes, particularly during intensive insulin therapy. The brain plays a central role in the counter-regulatory response by eliciting parasympathetic and sympathetic hormone responses to restore normoglycaemia. Brain glucose concentrations, being approximately 15-20% of the blood glucose concentration in humans, are rigorously maintained during hypoglycaemia through adaptions such as increased cerebral glucose transport, decreased cerebral glucose utilisation and, possibly, by using central nervous system glycogen as a glucose reserve. However, during sustained hypoglycaemia, the brain cannot maintain a sufficient glucose influx and, as the cerebral hypoglycaemia becomes severe, electroencephalogram changes, oxidative stress and regional neuronal death ensues. With particular focus on evidence from experimental studies on nondiabetic IIH, this review outlines the central mechanisms behind the counter-regulatory response to IIH, as well as cerebral adaption to avoid sequelae of cerebral neuroglycopaenia, including seizures and coma.

Neurologic damage in hypoglycemia

Hypoglycemia occurs in diabetic patients as a consequence of treatment with hypoglycemic agents, in insulinoma patients as a result of excessive insulin production, and in infants as a result of abnormal regulation of metabolism. Profound hypoglycemia can cause structural and functional disturbances in both the central (CNS) and the peripheral nervous system (PNS). The brain is damaged by a short and severe episode of hypoglycemia, whereas PNS pathology appears after a mild and prolonged episode. In the CNS, damaged mitochondria, elevated intracellular Ca2(+) level, released cytochrome c to the cytosol, extensive production of superoxide, increased caspase-3 activity, release of aspartate and glutamate from presynaptic terminals, and altered biosynthetic machinery can lead to neuronal cell death in the brain. Considering the PNS, chronic hypoglycemia is associated with delayed motor and sensory conduction velocities in peripheral nerves. With respect to pathology, hypoglycemic neuropathy in the PNS is characterized by Wallerian-like axonal degeneration that starts at the nerve terminal and progresses to a more proximal part of the axon, and motor axons to the muscles may be more severely damaged than sensory axons. Since excitatory neurotransmitters primarily involve the neuron in the CNS, this "dying back" pattern of axonal damage in the PNS may involve mechanisms other than excitotoxicity.

Role of corticosterone on sleep homeostasis induced by REM sleep deprivation in rats

Sleep is regulated by humoral and homeostatic processes. If on one hand chronic elevation of stress hormones impair sleep, on the other hand, rapid eye movement (REM) sleep deprivation induces elevation of glucocorticoids and time of REM sleep during the recovery period. In the present study we sought to examine whether manipulations of corticosterone levels during REM sleep deprivation would alter the subsequent sleep rebound. Adult male Wistar rats were fit with electrodes for sleep monitoring and submitted to four days of REM sleep deprivation under repeated corticosterone or metyrapone (an inhibitor of corticosterone synthesis) administration. Sleep parameters were continuously recorded throughout the sleep deprivation period and during 3 days of sleep recovery. Plasma levels of adrenocorticotropic hormone and corticosterone were also evaluated. Metyrapone treatment prevented the elevation of corticosterone plasma levels induced by REM sleep deprivation, whereas corticosterone administration to REM sleep-deprived rats resulted in lower corticosterone levels than in non-sleep deprived rats. Nonetheless, both corticosterone and metyrapone administration led to several alterations on sleep homeostasis, including reductions in the amount of non-REM and REM sleep during the recovery period, although corticosterone increased delta activity (1.0-4.0 Hz) during REM sleep deprivation. Metyrapone treatment of REM sleep-deprived rats reduced the number of REM sleep episodes. In conclusion, reduction of corticosterone levels during REM sleep deprivation resulted in impairment of sleep rebound, suggesting that physiological elevation of corticosterone levels resulting from REM sleep deprivation is necessary for plentiful recovery of sleep after this stressful event.

The effects of age and ketogenic diet on local cerebral metabolic rates of glucose after controlled cortical impact injury in rats

Previous studies from our laboratory have shown the neuroprotective potential of ketones after TBI in the juvenile brain. It is our premise that acutely after TBI, glucose may not be the optimum fuel and decreasing metabolism of glucose in the presence of an alternative substrate will improve cellular metabolism and recovery. The current study addresses whether TBI will induce age-related differences in the cerebral metabolic rates for glucose (CMRglc) after cortical controlled impact (CCI) and whether ketone metabolism will further decrease CMRglc after injury. Postnatal day 35 (PND35; n = 48) and PND70 (n = 42) rats were given either sham or CCI injury and placed on either a standard or a ketogenic (KG) diet. CMRglc studies using (14)C-2 deoxy-D-glucose autoradiography were conducted on days 1, 3, or 7 post-injury. PND35 and PND70 standard-fed CCI-injured rats exhibited no significant neocortical differences in CMRglc magnitude or time course compared to controls. Measurement of contusion volume also indicated no age differences in response to TBI. However, PND35 subcortical structures showed earlier metabolic recovery compared to controls than PND70. Ketosis induced by the KG diet was shown to affect CMRglc in an age-dependent manner after TBI. The presence of ketones after injury further reduced CMRglc in PND35 and normalized CMRglc in PND70 rats at 7 days bilaterally after injury. The changes in CMRglc seen in PND35 TBI rats on the KG diet were associated with decreased contusion volume. These results suggest that conditions of reduced glucose utilization and increased alternative substrate metabolism may be preferable acutely after TBI in the younger rat.

Hypoglycemic brain damage

Hypoglycemia was long considered to kill neurons by depriving them of glucose. We now know that hypoglycemia kills neurons actively from without, rather than by starvation from within. Hypoglycemia only causes neuronal death when the EEG becomes flat. This usually occurs after glucose levels have fallen below 1 mM (18 mg/dl) for some period, depending on body glycogen reserves. At the time that abrupt brain energy failure occurs, the excitatory amino acid aspartate is massively released into the limited brain extracellular space and floods the excitatory amino acid receptors located on neuronal dendrites. Calcium fluxes occur and membrane breaks in the cell lead rapidly to neuronal necrosis. Significant neuronal necrosis occurs after 30 min of electrocerebral silence. Other neurochemical changes include energy depletion to roughly 25% of control, phospholipase and other enzyme activation, tissue alkalosis and a tendency for all cellular redox systems to shift towards oxidation. The neurochemistry of hypoglycemia thus differs markedly from ischemia. Hypoglycemia often differs from ischemia in its neuropathologic distribution, a phenomenon applicable in forensic practice. The border-zone distribution of global ischemia is not seen, necrosis of the dentate gyrus of the hippocampus can occur and a predilection for the superficial layers of the cortex is sometimes seen. Cerebellum and brainstem are universally spared in hypoglycemic brain damage. Hypoglycemia constitutes a unique metabolic brain insult.

Hypoglycemic seizures during transient hypoglycemia exacerbate hippocampal dysfunction

Severe hypoglycemia constitutes a medical emergency, involving seizures, coma and death. We hypothesized that seizures, during limited substrate availability, aggravate hypoglycemia-induced brain damage. Using immature isolated, intact hippocampi and frontal neocortical blocks subjected to low glucose perfusion, we characterized hypoglycemic (neuroglycopenic) seizures in vitro during transient hypoglycemia and their effects on synaptic transmission and glycogen content. Hippocampal hypoglycemic seizures were always followed by an irreversible reduction (>60% loss) in synaptic transmission and were occasionally accompanied by spreading depression-like events. Hypoglycemic seizures occurred more frequently with decreasing "hypoglycemic" extracellular glucose concentrations. In contrast, no hypoglycemic seizures were generated in the neocortex during transient hypoglycemia, and the reduction of synaptic transmission was reversible (<60% loss). Hypoglycemic seizures in the hippocampus were abolished by NMDA and non-NMDA antagonists. The anticonvulsant, midazolam, but neither phenytoin nor valproate, also abolished hypoglycemic seizures. Non-glycolytic, oxidative substrates attenuated, but did not abolish, hypoglycemic seizure activity and were unable to support synaptic transmission, even in the presence of the adenosine (A1) antagonist, DPCPX. Complete prevention of hypoglycemic seizures always led to the maintenance of synaptic transmission. A quantitative glycogen assay demonstrated that hypoglycemic seizures, in vitro, during hypoglycemia deplete hippocampal glycogen. These data suggest that suppressing seizures during hypoglycemia may decrease subsequent neuronal damage and dysfunction.

Hypoglycemia, brain energetics, and hypoglycemic neuronal death

Hypoglycemia is a common and serious problem among diabetic patients receiving treatment with insulin or other glucose-lowering drugs. Moderate hypoglycemia impairs neurological function, and severe hypoglycemia leads to death of selectively vulnerable neurons. Recent advances have shed new light on the underlying processes that cause neuronal death in hypoglycemia and the factors that may render specific neuronal populations especially vulnerable to hypoglycemia. In addition to its clinical importance, the pathophysiology of hypoglycemia is an indicator of the unique bioenergetic properties of the central nervous system, in particular the metabolic coupling of neuronal and astrocyte metabolism. This review will focus on relationships between bioenergetics and brain dysfunction in hypoglycemia, the neuronal cell death program triggered by hypoglycemia, and the role of astrocytes in these processes.

Central nervous system involvement in diabetes mellitus

Diabetic complications result in much morbidity and mortality and considerable consumption of scarce medical resources. Thus, elucidation of the risk factors and pathophysiologic mechanisms underlying diabetic complications is important. The effects of diabetes on the central nervous system (CNS) result in cognitive dysfunction and cerebrovascular disease. Treatment-related hypoglycemia also has CNS consequences. Advances in neuroimaging now provide greater insights into the structural and functional impact of diabetes on the CNS. Greater understanding of CNS involvement could lead to new strategies to prevent or reverse the damage caused by diabetes mellitus.

Effects of deprivation of oxygen or glucose on the neural activity in the guinea pig hippocampal slice--intracellular recording study of pyramidal neurons

The block of synaptic transmission and neural activity during deprivation of oxygen or glucose has been simply attributed to the lack of energy due to the disorder of energy production. To clarify the interrelation between neural activity and energy metabolism during hypoxia or glucose deprivation, we studied the changes in ATP levels and electrical events of pyramidal neurons in the CA3 region and [Ca2+]i mobilization of the dendritic and cellular region of CA3 area, using guinea pig hippocampal slices. The studies of field potentials and intracellular recording from the pyramidal cell of CA3 area during hypoxia or glucose deprivation revealed that the cessation of synaptic activity and the depolarization of resting potential occurred earlier than during glucose deprivation while the increase of [Ca2+]i was slow during hypoxia but rapid during glucose deprivation although the ATP level of CA3 area was maintained at its original level for 20 min during both conditions. When glucose was replaced by lactate, ATP concentration was not reduced but the electrical activity decayed and [Ca2+]i increased with the similar time course as observed during lack of glucose, only. These results suggest that different mechanisms underlie the block of synaptic transmission in the CA3 pyramidal neurons during hypoxia and glucose deprivation and that lactate cannot substitute for glucose in the maintenance of neural activity.

Acetylcholinesterase and Na+,K(+)-ATPase activities in different regions of rat brain during insulin-induced hypoglycemia

The activities of acetylcholinesterase (acetylcholine acetylhydrolase, EC 3.1.1.7), responsible for hydrolysis of acetylcholine and Na+,K(+)-ATPase (Mg(2+)-dependent ATP phosphohydrolase, EC 3.6.1.3), which plays a crucial role in neurotransmission, were determined in four brain regions after 1, 2, and 3 h of insulin administration. Significant decrease in the acetylcholinesterase and Na+,K(+)-ATPase activities was observed in the soluble and total particulate fractions from cerebral hemispheres, cerebellum, brain stem, and diencephalon + basal ganglia after 1, 2, and 3 h of insulin-induced hypoglycemia. Blood glucose level decreased significantly after 1 h of insulin administration and remained at low level for 2 h thereafter, whereas, the protein content in different subcellular fractions from four brain regions did not show any significant change under this physiological stress.

Hypoglycaemia: brain neurochemistry and neuropathology

The widespread use of insulin and oral hypoglycaemic agents has increased the incidence of hypoglycaemic brain damage due to accidental, suicidal, or homicidal overdose. Hypoglycaemia is capable of damaging the brain in the face of intact cardiac function, but neuronal necrosis occurs only when the electroencephalogram (EEG) becomes isoelectric. Neurochemical changes are distinct from ischaemia, and cerebral blood flow is actually increased, in contrast to cerebral ischaemia. Salient neurochemical changes include an arrest of protein synthesis in many but not all brain regions, a shift of brain redox equilibria towards oxidation, incomplete energy failure, loss of ion homeostasis, cellular calcium influx, intracellular alkalosis, and a release of neuroactive amino acids, especially aspartate, into the extracellular space of the brain. The metabolic release of aspartate, and to a lesser extent glutamate, into the interstitial space of the brain produces histopathological patterns of neuronal death that can be distinguished from ischaemic brain damage in experimental brain tissue and, occasionally, in brains from human autopsies after hypoglycaemic brain damage. The excitatory amino acids released during profound hypoglycaemia bind to neuronal dendrites and perikarya, but not to other cell types in the nervous system, thus giving rise to selective neuronal death. The absence of acidosis, and an adequate blood supply during hypoglycaemia, protect the brain against pan-necrosis or infarction. However, the neurons die more quickly during hypoglycaemic brain damage than after cerebral ischaemia. Hypoglycaemic brain damage thus falls into the newly defined class of cerebral 'excitotoxic' neuropathologies, where neurons are selectively killed by an extracellular overflow of excitatory amino acids produced by the brain itself. The pathogenesis of hypoglycaemic brain damage is thus rather more novel and intriguing than was thought even a decade ago, when it was believed that glucose starvation and simple energy failure resulted directly in neuronal catabolism.

Adenosine release is a major cause of failure of synaptic transmission during hypoglycaemia in rat hippocampal slices

Glucose-free medium (aglycaemia) caused a complete failure of CA1 population spikes (after 14.5 +/- 0.8 min) and field EPSPs (after 18.1 +/- 0.5 min). In the presence of the selective adenosine A1 antagonist, 8-(p-sulfophenyl)theophylline (10 microM), population spikes and EPSPs were decreased by only 13.8 +/- 11.9% and 32.4 +/- 11.6% at the end of 17.0 +/- 3.0 min and 19.8 +/- 1.7 min of aglycaemia, respectively. A similar effect was produced by caffeine (0.2 mM). The ATP-sensitive K+ channel blockers tolbutamide (1 mM) and glibenclamide (10 microM) had no significant effect on aglycaemic suppression of synaptic transmission. These observations indicate that endogenous adenosine, but not ATP-sensitive K+ conductance, plays a major role in hypoglycaemia failure of transmission.

Quantitative microdialysis of dopamine in the striatum: effect of circadian variation

Two quantitative microdialysis methods were used to determine the concentration of extracellular dopamine in the anterior striatum of the rat. In the first method, the slow perfusion flow rate method, perfusion was at 57 nl/min and dialysate samples were collected every 90 min for 18 h and assayed for dopamine (DA), DOPAC (3,4-dihydroxy-phenylacetic acid), homovanillic acid (HVA) and 5-hydroxy-indoleacetic acid (5-HIAA). There was a significant increase in the concentration of dopamine during the dark cycle compared with the light cycle (14.7 +/- 1 nM vs. 9.3 +/- 0.7 nM; mean +/- SEM; P less than 0.0001), indicating possible circadian variations in the extracellular concentration of DA. There was a steady decrease in the level of DOPAC and HVA, and no change in the level of 5-HIAA. For the point of no-net-flux method, animals were perfused with 4 concentrations of DA or DOPAC, bracketing the extracellular concentrations. The extracellular concentrations of DA and DOPAC using this method were 10.2 +/- 1.7 nM and 17.4 +/- 2.6 microM, respectively. The in vivo recoveries for DA and DOPAC as derived from the slope of the linear regression curves were 72 +/- 3% and 43 +/- 5%. These values were shown to be significantly different (P less than 0.001). Both methods gave similar results for the level of DA in the striatum.

The influence of hypothermia on hypoglycemia-induced brain damage in the rat

The effects of hypothermia on hypoglycemic brain damage were studied in rats after a 30-min period of hypoglycemic coma, defined as cessation of spontaneous EEG activity. The rats were either normothermic (37 degrees C) or moderately hypothermic (33 degrees C). Morphological brain damage was evaluated after various periods of recovery. Hypothermic animals with halothane anesthesia never resumed spontaneous respiration, thus requiring artificial ventilation during recovery (maximally 8 h). In contrast, when isoflurane was used as the anesthetic agent, all animals survived and were examined after 1 week of recovery. There was a tendency towards gradually higher arterial plasma glucose levels during hypoglycemia with lower body temperature. The time period from insulin injection until isoelectric EEG appeared was gradually prolonged by hypothermia, and was shorter when isoflurane was used for anesthesia. Brain damage was examined within the neocortex, caudoputamen and hippocampus (CA1, subiculum and the tip of the dentate gyrus). Damage to neurons was found to be of two types, namely condensed dark purple neurons (pre-acidophilic) and shrunken bright red-staining neurons (acidophilic). In the neocortex, no clear influence of temperature on the degree of injury was seen. In the caudoputamen, the number of injured neurons clearly decreased at lower temperature (33 degrees C, P less than 0.001) when halothane was used, while no such difference was seen when isoflurane was used as the anesthetic agent. Likewise, a protective effect of hypothermia was seen in subiculum (P less than 0.01) when halothane, but not isoflurane was used.(ABSTRACT TRUNCATED AT 250 WORDS)

Hypothermia is critical for survival during prolonged insulin-induced hypoglycemia in rats

Hypothermia is a well-known concomitant of hypoglycemia in mammals. We tested the hypothesis that this hypothermia is an important adaptive response to hypoglycemia in 11 normal Sprague-Dawley rats. Twelve-hour fasted, conscious animals received primed, continuous insulin infusions for up to 8 hours. Plasma glucose was clamped between 30 and 40 mg/dL and core body temperature was monitored continuously during the insulin infusions. Five of the animals were maintained in a room temperature environment (22 to 24 degrees C) during the hypoglycemia; all became hypothermic (mean +/- SE nadir core temperature, 31 +/- 0.5 degrees C). Spontaneous activity was reduced in these animals, but they remained conscious and responsive to external stimuli. All five returned to normal behavior after euglycemia was restored at the end of the insulin infusions. In the remaining six animals, hypothermia was prevented during hypoglycemia by warming of the air in their cages (mean of hourly core temperatures, 37 +/- 0.1 degrees C). None of these animals survived more than 7 hours. The severity of the hypoglycemia was no greater in the euthermic than in the hypothermic group, as judged by the mean of individual nadir plasma glucose levels (25 +/- 1 v 24 +/- 1 mg/dl, respectively) and by the mean number of glucose values per animal that were less than 30 mg/dL (2 +/- 1 v 7 +/- 1). Plasma osmolality did not change significantly in either group during the period of hypoglycemia, suggesting that dehydration was not the cause of death in the euthermic animals.(ABSTRACT TRUNCATED AT 250 WORDS)

Cytosolic calcium during glucose deprivation in hippocampal pyramidal cells of rats

Glucose deprivation (GD) results in a hyperpolarization by turning on a potassium conductance (gK,GD) in hippocampal CA3 pyramidal cells. We used combined intracellular and microfluorometric recording techniques to evaluate whether gK,GD is activated by a rise in the concentration of intracellular calcium ([Ca2+]i). We found that the activation of gK,GD is only followed, but not preceded by a rise in [Ca2+]i. Furthermore, gK,GD is not blocked by the sulfonylurea glibenclamide, a blocker of ATP-regulated potassium conductance. We conclude that activation of gK,GD does not simply reflect breakdown of the calcium of ATP homeostasis, but on the contrary might represent an active restoring mechanism which delays the pathological consequences of sustained glucose deficiency.

Extracellular pH in the rat brain during hypoglycemic coma and recovery

It has previously been shown that hypoglycemic coma is accompanied by marked energy failure and by loss of cellular ionic homeostasis. The general proposal is that shortage of carbohydrate substrate prevents lactic acid formation and thereby acidosis during hypoglycemic coma. The objective of the present study was to explore whether rapid downhill ion fluxes, known to occur during coma, are accompanied by changes in extra- and/or intracellular pH (pHe and/or pHi), and how these relate to the de- and repolarization of cellular membranes. Cortical pHe was recorded by microelectrodes in insulin-injected rats subjected to 30 min of hypoglycemic coma, with cellular membrane depolarization. Some rats were allowed up to 180 min of recovery after glucose infusion and membrane repolarization. Arterial blood gases and physiological parameters were monitored to maintain normotension, normoxia, normocapnia, and normal plasma pH. Following depolarization during hypoglycemia, a prompt, rapidly reversible alkaline pHe shift of about 0.1 units was observed in 37/43 rats. Immediately thereafter, all rats showed an acid pH shift of about 0.2 units. This shift developed during the first minute, and pHe remained at that level until repolarization was induced. Following repolarization, there was an additional, rapid, further lowering of pHe by about 0.05 units, followed by a more prolonged decrease in pHe that was maximal at 90 min of recovery (delta pHe of approximately -0.4 units). The pHe then slowly normalized but was still decreased (-0.18 pH units) after 180 min when the experiment was terminated. The calculated pHi showed no major alterations during hypoglycemic coma or after membrane repolarization following glucose administration.(ABSTRACT TRUNCATED AT 250 WORDS)

Ingestion of carbohydrates varying in complexity produce differential brain responses

Variations in the complexity of carbohydrate substances were studied with respect to their differential effects on brain function. Glucose, sucrose, fructose and corn starch were each administered as part of an oral carbohydrate tolerance test during 17 test trials. Brain EEG changes and blood glucose levels were monitored concurrently throughout a 5 hour period. The glucose solution produced more substantial EEG effects than the other three carbohydrate solutions. Absolute blood glucose level was the primary determinant of electrocortical changes found predominantly in the left parietal-occipital and left temporal cortical regions. Implications for the study and evaluation of cognitive function were discussed.

Cytosolic free calcium, NAD/NADH redox state and hemodynamic changes in the cat cortex during severe hypoglycemia

Using indo-1, a fluorescent Ca2+ indicator, in vivo fluorometric measurements were made of changes in cytosolic free Ca2+, NAD/NADH redox state, and hemodynamics directly from the cat cortex during and after severe insulin-induced hypoglycemia. Cytosolic free Ca2+ started to increase when the EEG became isoelectric, remained at a significantly high level (p less than 0.05) during the period of isoelectric EEG (IEEG), and recovered to the control level 6 min following an intravenous infusion of glucose. The NAD/NADH redox state oxidized significantly during IEEG and then recovered rapidly to the control level after the glucose infusion. Local cortical blood volume (LCBV) increased gradually during the progression of hypoglycemia, reaching the maximal level (146 +/- 7%) at the end of IEEG, and then started to recover. The mean transit time (MTT) through the cortical microcirculation was shortened during the IEEG (control: 3.84 +/- 0.41 s versus IEEG: 2.73 +/- 0.17 s, p less than 0.05), whereas it was prolonged during the 30-min recovery period (5.68 +/- 0.58 s, p less than 0.05). Local cortical blood flow calculated from the LCBV and MTT showed a twofold increase 5 min into IEEG (201 +/- 27% of control, p less than 0.05), recovered 15 min into the recovery period, and then decreased to 77% of control (p less than 0.05) by 30 min. The data support the hypothesis that hypoglycemic brain damage might be mediated by an elevation of cytosolic free calcium.

Effects of hypoglycaemia and hypoxia on the intracellular pH of cerebral tissue as measured by 31P nuclear magnetic resonance

Changes in high-energy phosphate metabolites and the intracellular pH (pHi) were monitored in cerebral tissue during periods of hypoglycaemia and hypoxia using 31P nuclear magnetic resonance spectroscopy. Superfused brain slices were loaded with deoxyglucose at a concentration shown not to impair cerebral metabolism, and the chemical shift of the resulting 2-deoxyglucose-6-phosphate (DOG6P) peak was used to monitor the pHi. In some experiments with low circulating levels of Pi, the intracellular Pi was visible and indicated a pH identical to that of DOG6P, an observation validating its use as an indicator of pHi in cerebral tissue. The pHi was found to be unchanged during moderate hypoglycaemia; however, mild hypoxia (PO2 = 16.4 kPa) and severe hypoglycaemia produced marked reductions from the normal of 7.2 to 6.8 and 7.0, respectively. Hypoglycaemia caused a fall in the level of both phosphocreatine (PCr) and ATP, whereas hypoxia affected PCr alone, as shown previously. However, the fall in pHi was similar during the two insults, thus indicating that the change in pH is not directly linked to lactate production or to the creatine kinase reaction.

Electrolyte shifts between brain and plasma in hypoglycemic coma

Hypoglycemia of sufficient severity to cause cessation of EEG activity (coma) is accompanied by energy failure and by loss of ion homeostasis, the latter encompassing a marked rise in extracellular fluid (ECF) K+ concentration and a fall in ECF Ca2+ concentration. Presumably, ECF Na+ concentration decreases as well. In the present study, the extent that the altered ECF-plasma gradients give rise to net ion fluxes between plasma and tissue is explored. Accordingly, whole tissue contents of Ca2+, Mg2+, K+, and Na+ were measured. The experiments were carried out in anaesthetized and artificially ventilated rats given insulin i.p.; cerebral cortical tissue was sampled at the stage of slow-wave EEG activity, after 10, 30, and 60 min of coma (defined as isoelectric EEG), as well as after 1.5, 6, and 24 h of recovery. In the precomatose animals (with a slow-wave EEG pattern), no changes in electrolyte contents were observed. During coma, tissue Na+ content increased progressively and the K+ content fell (each by 20 mumol g-1 during 60 min). During recovery, these alterations were reversed within the first 6 h. The Mg2+ content remained unchanged. In spite of the appreciable plasma to ECF Ca2+ gradient, no significant calcium accumulation was observed. It is concluded significant calcium accumulation was observed. It is concluded that hypoglycemia leads to irreversible neuronal necrosis in the absence of gross accumulation of calcium in the tissue.

Regional differences in brain glucose content in graded hypoglycemia

Graded hypoglycemia was induced with insulin in anesthetized and artificially ventilated rats. The brains were frozen in situ, and the regional glucose concentration was determined in different areas of the brain with the bioluminescent technique. In all nine brain structures analyzed, brain tissue glucose content assessed with the bioluminescent technique correlated closely with the plasma glucose levels; the tissue/plasma glucose concentration ratios approximating 0.3. There were, however, relatively marked regional differences. For example, whereas glucose concentrations in the neocortex, caudoputamen, hippocampus, and cerebellum were very low in rats having a plasma glucose concentration of less than 4 mumol/mL, higher glucose concentrations were present in these animals in the thalamus, hypothalamus, and brainstem. The lowest glucose content was found in the caudoputamen, which was depleted of glucose in animals with plasma levels below 3 mumol/mL. It is concluded that regional inhomogeneities in the glucose levels observed during hypoglycemia may, at least in part, explain differences in the vulnerability of different brain structures following reversible hypoglycemia.

Progress review: hypoglycemic brain damage

The central question to be addressed in this review can be stated as "How does hypoglycemia kill neurons?" Initial research on hypoglycemic brain damage in the 1930s was aimed at demonstrating the existence of any brain damage whatsoever due to insulin. Recent results indicate that uncomplicated hypoglycemia is capable of killing neurons in the brain. However, the mechanism does not appear to be simply glucose starvation of the neuron resulting in neuronal breakdown. Rather than such an "internal catabolic death" current evidence suggests that in hypoglycemia, neurons are killed from without, i.e. from the extracellular space. Around the time the EEG becomes isoelectric, an endogenous neurotoxin is produced, and is released by the brain into tissue and cerebrospinal fluid. The distribution of necrotic neurons is unlike that in ischemia, being related to white matter and cerebrospinal fluid pathways. The toxin acts by first disrupting dendritic trees, sparing intermediate axons, indicating it to be an excitotoxin. Exact mechanisms of excitotoxic neuronal necrosis are not yet clear, but neuronal death involves hyperexcitation, and culminates in cell membrane rupture. Endogenous excitotoxins produced during hypoglycemia may explain the tendency toward seizure activity often seen clinically. The recent research results on which these findings are based are reviewed, and clinical implications are discussed.

Insufficient supply of reducing equivalents to the respiratory chain in cerebral cortex during severe insulin-induced hypoglycemia in cats

The ability of endogenous substrates in brain to substitute for glucose as sources for energy metabolism during insulin-induced hypoglycemia was studied. The ratio of the arteriovenous difference of glucose to the arteriovenous difference of oxygen in the cerebral cortex was measured during progressive hypoglycemia in paralyzed, artificially ventilated cats that were anesthetized with pentobarbital sodium and nitrous oxide. The ratio did not change when blood glucose fell from a mean of 7.68 to approximately 2 mumol/ml. Below 2 mumol/ml the ratio decreased, indicating that substrates other than the glucose supplied by the blood were being utilized. In another series of experiments, changes in the redox state of respiratory chain NAD were monitored from the cerebral cortex using microfluorometry during the onset of hypoglycemia and the recovery. Hypoglycemia severe enough to produce isoelectric EEG was accompanied by an oxidation of NADH, demonstrating that the supply of reducing equivalents to the respiratory chain was decreased. Recovery from hypoglycemia, produced by intravenous glucose injections, was accompanied by an increase in blood glucose concentrations, the return of EEG activity, and a decrease in the NAD/NADH ratio. When blood glucose concentration reached 2.23 during the recovery, further increases in blood glucose had no effect on the redox state of NAD. Although alternative substrates appear to be utilized for energy metabolism during severe hypoglycemia, they cannot fully replace glucose as the source of reducing equivalent to the respiratory chain.

Effects of hypoglycemia on rat brain polyribosome sedimentation pattern

Insulin-induced hypoglycemia provokes polyribosome disaggregation and accumulation of monomeric ribosomes in the brain of rats with hypoglycemic paresis and coma. The extent of brain polyribosome disaggregation depends on the decrease of blood glucose concentration, and in comatose animals on the duration of hypoglycemia. Cycloheximide prevents the disaggregation of brain polyribosomes induced by hypoglycemia, indicating that hypoglycemia affects brain protein synthesis, decreasing the rate of initiation relative to the rate of elongation of polypeptide chain synthesis.

Effect of hypoglycemic encephalopathy upon amino acids, high-energy phosphates, and pHi in the rat brain in vivo: detection by sequential 1H and 31P NMR spectroscopy

Metabolic alterations in amino acids, high-energy phosphates, and intracellular pH during and after insulin hypoglycemia in the rat brain was studied in vivo by 1H and 31P nuclear magnetic resonance (NMR) spectroscopy. Sequential accumulations of 1H and 31P spectra were obtained from a double-tuned surface coil positioned over the exposed skull of a rat while the electroencephalogram was recorded continuously. The transition to EEG silence was accompanied by rapid declines in phosphocreatine, nucleoside triphosphate, and an increase in inorganic orthophosphate in 31P spectra. In 1H spectra acquired during the same time interval, the resonances of glutamate and glutamine decreased in intensity while a progressive increase in aspartate was observed. Following glucose administration, glutamate and aspartate returned to control levels (recovery half-time, 8 min); recovery of glutamine was incomplete. An increase in lactate was detected in the 1H spectrum during recovery but it was not associated with any change in the intracellular pH as assessed in the corresponding 31P spectrum. Phosphocreatine returned to control levels following glucose administration, in contrast to nucleoside triphosphate and inorganic orthophosphate which recovered to only 80% and 200% of their control levels, respectively. These results show that the changes in cerebral amino acids and high-energy phosphates detected by alternating the collection of 1H and 31P spectra allow for a detailed assessment of the metabolic response of the hypoglycemic brain in vivo.

Regional acetylcholine metabolism in brain during acute hypoglycemia and recovery

Insulin-induced hypoglycemia in normothermic rats caused progressive neurological depression and differentially altered regional cerebral acetylcholine metabolism. Reductions of plasma glucose from 7.7 mM (control) to 2.5-1.7 mM (moderate hypoglycemia associated with decreased motor activity) or 1.5 mM (severe hypoglycemia with lethargy progressing to stupor) decreased glucose concentrations in the cerebral cortex, striatum, and hippocampus to less than 10% of control. Moderate hypoglycemia diminished acetylcholine concentrations in cortex and striatum (21% and 45%, respectively) and reduced [1-2H2, 2-2H2]choline incorporation into acetylcholine (62% and 41%, respectively). Severe hypoglycemia did not reduce the acetylcholine concentration or synthesis in cortex and striatum further. The concentrations of choline rose in the cortex (+53%) and striatum (+130%) of animals that became stuporous but a similar rise in [1-2H2, 2-2H2]choline left the specific activities of choline in these structures unchanged. Even severe hypoglycemia did not alter the hippocampal cholinergic system. In rats that developed hypoglycemic stupor and were then treated with glucose, the animals recovered apparently normal behavior, and the concentrations of acetylcholine and the incorporation of [1-2H2, 2-2H2]-choline into acetylcholine returned to control values in the striatum but not in the cerebral cortex. Thus, impaired acetylcholine metabolism in selected regions of the brain may contribute to the early symptoms of neurological dysfunction in hypoglycemia.

Effect of aging on cerebral cortex energy metabolism in hypoglycemia and posthypoglycemic recovery

Severe hypoglycemia, causing the cessation of spontaneous EEG, induced in cerebral cortex of rats of different ages, causes gross energy failure and extensive derangement of both carbohydrate and amino acid contents. During posthypoglycemic recovery of adult rats, there was moderate restitution of energy metabolism and both ATP concentration and adenine nucleotide pool remained still reduced, even if the creatine phosphate and ADP contents were close to normal. During recovery of adult rats there was a rise in glutamate and glutamine concentrations and the perturbated aspartate and gamma-aminobutyrate cerebral contents normalized. Ammonia content decreased to normal, while alanine content was markedly elevated. Aging does not affect the cerebral metabolic derangements occurring in severe hypoglycemia, but rather the metabolic changes that the brain tend to reverse during the posthypoglycemic restitution. In fact, there was lower restitution of the contents of cerebral cortical metabolites of "mature" and "senescent" rats in comparison with "adult" ones. Particularly, in older brains the contents of many amino acids and adenylate nucleotides remained largely abnormal.

Influence of severe hypoglycemia on brain extracellular calcium and potassium activities, energy, and phospholipid metabolism

In the cerebral cortices of rats, during insulin-induced hypoglycemia, changes in the concentrations of labile phosphate compounds [ATP, ADP, AMP, and phosphocreatine (PCr)] and glycolytic metabolites (lactate, pyruvate, and glucose) as well as phospholipids and free fatty acids (FFAs) were studied in relation to extracellular potassium and calcium activities. Changes in extracellular calcium and potassium activities occurred at approximately the onset of isoelectricity . The extracellular calcium activity dropped from 1.17 +/- 0.14 mM to 0.18 +/- 0.28 mM and the potassium activity rose from 3.4 +/- 0.94 mM to 48 +/- 12 mM (means +/- SD). Minutes prior to this ionic change the levels of ATP, PCr, and phospholipids were unchanged while the levels of FFAs remained unchanged or slightly elevated. Following the first ionic change the steady-state levels of ATP decreased by 40%, from 2.42 to 1.56 mumol/g. PCr levels decreased by 75%, from 4.58 to 1.26 mumol/g. Simultaneously, the levels of FFAs increased from 338 to 642 nmol/g, arachidonic acid displaying the largest relative increase, 33 to 130 nmol/g. The first ionic change was followed by a short period of normalization of ionic concentrations followed by a sustained ionic change. This was accompanied by a small additional decrease in ATP (to 1.26 mumol/g). The FFA levels increased to 704 nmol/g. There was a highly significant negative correlation between the levels of FFAs and the energy charge of the tissue. The formation of FFAs was accompanied by a decrease in the phospholipid pool. The largest relative decrease was observed in the inositol phosphoglycerides, followed by serine and ethanolamine phosphoglycerides.(ABSTRACT TRUNCATED AT 250 WORDS)

Role of drugs in recovery of metabolic function of rat brain following severe hypoglycemia

Severe hypoglycemia with isoelectric EEG induced extensive deterioration of the energy state and gross alteration of amino acid contents on the rat cerebral and cerebellar cortex. During recovery, tissue glucose concentration returned to normal, while both lactate and pyruvate concentrations increased to above normal. In the recovery period, the ATP concentration increased but the adenine nucleotide pool remained reduced, even if the ADP and AMP contents were close to normal. Phosphocreatine was restored to normal concentration with reciprocal changes in creatine content. During recovery there was a rise in glutamate and glutamine concentrations, gamma-aminobutyrate content returning to normal value. Ammonia and aspartate decreased below normal, while alanine increased above normal. The effect of some pharmacological agents on the posthypoglycemic recovery was tested: (a) Ergot alkaloids (dihydroergocristine, dihydroergocriptine, dihydroergocornine); (b) Vinca minor alkaloids (vincamine TPS, (-) eburnamonine); (c) Rauwolfia serpentina alkaloids (reserpine, raubasine); (d) synthetic agent (piracetam). During the posthypoglycemic recovery, these different agents exhibited different, or even contrasting, interferences on glycolytic metabolites, amino acids and energy-rich phosphates. The metabolic alterations in the cerebellar cortex were qualitatively of the same character of those in neocortex. However, the metabolic alterations were less extensive and more sensitive to drug action.

Cerebral extracellular calcium activity in severe hypoglycemia: relation to extracellular potassium and energy state

The changes in extracellular Ca2+ (Cae) and K+ (Ke) activities were studied in the rat brain during insulin-induced hypoglycemia. At about the time of onset of isoelectric EEG in severe insulin-induced hypoglycemia (300-g male Wistar rats under 70% N2O anaesthesia), there was an increase in Ke which, at approximately 13 mM, was associated with a fall in Cae. Ke peaked at 48 +/- 12 mM, and Cae at 0.18 +/- 0.28 mM. This ion change began to normalise, but before recovery was complete a second ion change, of magnitude similar to that of the first, occurred from which the cells did not recover. The Cae recovered to only 66% of normal in the time available before the second depolarisation. Measurements on brains frozen at different stages during the sequence of ion changes revealed that ATP and phosphocreatine (PCr) concentrations and energy charge (EC) were not reduced before the first depolarisation. During the first depolarisation there was a 72% decrease in PCr and a 37% fall in ATP level, leading to a 23% drop in EC. These levels decreased further by the 10th minute of isoelectricity , but only the fall in ATP concentration was significant. The results indicate that the first ion change was a spreading depression and that cellular energy state was not the only factor in determining the response of tissue in the early stages of the comatose state.

Cerebral endogenous substrate utilization during the recovery period after profound hypoglycemia

Markedly decreased levels of energy-rich phosphates were seen in cerebral cortex after severe hypoglycemia, followed by their partial restitution during the recovery period. During hypoglycemia the nonglucose endogenous substrates were provided by glycolytic intermediates, by Krebs cycle intermediates, and by related amino acids. Other potential substrates for brain oxidation were provided by the breakdown of phospholipids and fatty acids. After a 20-min period of posthypoglycemic recovery, partial restoration of carbohydrates and amino acids occurred, although the amino acid pool size was still reduced. The alterations in phospholipids and fatty acids persisted, while there was a tendency toward normalization of the free fatty acid content. During the posthypoglycemic recovery, treatment with some specific metabolic modulators (6-aminonicotinamide, hopantenate, uridine, L-acetylcarnitine) suggested the possibility of an alternative cerebral substrate utilization owing to modulation of the cerebral biochemical machinery. Thus, increased carbohydrate utilization by hopantenate was consistent with decreased lipid breakdown, while increased carbohydrate utilization by uridine was concomitant with decreased amino acid degradation. In this way, decreased cerebral carbohydrate utilization by 6-amino-nicotinamide was associated with increased lipid and amino acid breakdown. Furthermore, the increased loss of cerebral phospholipids and phospholipid-bound fatty acids by L-acetylcarnitine occurred in the presence of a large glucose availability and was associated with an extensive reduction of cerebral glycolytic flux.

Regional differences in vascular autoregulation in the rat brain in severe insulin-induced hypoglycemia

The present experiments were undertaken to determine if loss of vascular autoregulation during severe hypoglycemia shows regional differences that could help to explain the localization of hypoglycemic cell damage. Artificially ventilated rats (70% N2O) were subjected to a 30-min insulin-induced hypoglycemic coma (with cessation of EEG activity), with mean arterial blood pressure being maintained at 140, 120, 100, and 80 mm Hg. After 30 min of hypoglycemia, local cerebral blood flow (CBF) in 25 brain structures was measured autoradiographically with a [14C]iodoantipyrine technique. Since local CBF values did not differ between the 120 and the 100 mm Hg groups, the animals of these groups were pooled (110 mm Hg group). The results showed that at a blood pressure of 140 mm Hg, CBF was increased in 22 of 25 structures analyzed, the maximal values approximating 300% of control. At 110 mm Hg, cerebral cortical structures had CBF values that were either decreased, normal, or slightly increased; however, many subcortical structures (and cerebellum) showed markedly increased flow rates. Although a lowering of blood pressure to 80 mm Hg usually further reduced flow rates, some of these latter structures also had well-maintained CBF values at that pressure. Thus, there were large interstructural variations of local CBF at any of the pressures examined. Analysis of the pressure-flow relationship showed loss of autoregulation in some structures, whereas others had remarkably well-preserved CBF values at low pressures. The results indicate that during severe hypoglycemia, even relatively moderate arterial hypotension may add a circulatory insult to the primary one, and they strongly suggest that any such insult affects some brain structures more than others.

Feeding elicited by 2-deoxyglucose occurs in the absence of reduced glucose oxidation

We have previously reported that feeding in response to either systemic or intracerebroventricular administration of 2-deoxy-D-glucose (2-DG) persists for at least 6--8 h post-2-DG injection when the sympathoadrenal hyperglycemic response to glucoprivation has abated. This delayed feeding response to 2-DG suggests either that ongoing glucoprivation is not necessary for feeding or that responses such as hyperglycemia abate while glucoprivation is still extant and able to stimulate feeding. In order to determine whether glucoprivation is still present after the sympathoadrenal hyperglycemic response to 2-DG has abated, we measured systemic 14CO2 evolution and glucose oxidation in Sprague-Dawley rats treated with 2-DG (200 mg/kg) 0.5 and 6 h earlier. In addition, we measured cerebral 14C-2-DG accumulation in rats treated with unlabelled 2-DG 6 h prior to tracer administration We found that 14CO2 evolution and glucose oxidation were reduced by 47 and 667%, respectively, during the first 3.5 h post-2-DG but were normal by 6 h post-2-DG. Furthermore, we found that the uptake of 14C-2-DG into the brain was not diminished between 6 and 7 h after administration of unlabelled 2-DG. These results suggest that ongoing reduction of systemic glucose oxidation and ongoing impairment of hexose availability to the brain need not occur during 2-DG-induced feeding.

Changes in hexokinase isoenzymes in regions of rat brain during insulin-induced hypoglycemia

Activities of hexokinase isoenzymes were determined during insulin-induced hypoglycemia in soluble and total particulate fractions from three regions of rat brain. Type I hexokinase isoenzyme activity showed a small decrease in both soluble and particulate fractions from the cerebral hemispheres. In cerebellum and brain stem, however, Type I isoenzyme showed a decrease only in the soluble fraction. A significant increase was observed in hexokinase Type II isoenzyme from both the fractions, in all the three brain regions 1 h after insulin administration.

Cerebral blood flow, brain pH, and oxidative metabolism in the cat during severe insulin-induced hypoglycemia

The effects of severe hypoglycemia on brain pH, cerebral blood flow (CBF), and other physiologic and metabolic parameters were studied in 26 cats subjected to insulin hypoglycemia. Two groups were utilized to compare the effects of anesthesia. The halothane group was composed of 14 animals and the barbiturate group contained 12 animals. Insulin was administered by both the intravenous and intramuscular routes until there was a severe electroencephalographic (EEG) change or until 6 h had elapsed. The cerebral responses to hypoglycemia demonstrated the following: CBF was unaffected by severe hypoglycemia in normotensive animals with or without EEG changes; brain pH was essentially constant in all groups regardless of glucose levels, CBF, or EEG; and the EEG abnormalities corresponded closely to brain glucose levels. Cerebral adenosine triphosphate and phosphocreatine levels were lowest in the animals with isoelectric EEGs. We conclude that CBF and brain pH in the normotensive cat under general anesthesia are relatively unaffected by insulin hypoglycemia despite the presence of severe EEG changes and diminished cerebral energy reserves. The study suggests tha the PaCO2-CBF response curve is ot dependent upon the metabolic integrity of cerebral tissue and is mediated by pathways separate from those of autoregulation.

Relationship between brain mitochondrial hexokinase and neuronal function: comparable effects of 2-deoxy-D-glucose and thiopental

Mitochondrially bound brain hexokinase is solubilized by anesthetics and this effect has been suggested to contribute to anesthesia. In the present investigation the influence of the metabolic inhibitor 2-deoxy-D-glucose (2-DOG) was studied. An isolated rat brain preparation was used to avoid the contribution of peripheral reactions. Isolated rat brains were perfused for 45 min with media containing 4 mmol/l glucose, 10 mmol/l 2-DOG and/or 0.4 mmol/l thiopental. The EEG was monitored and acetylcholine, 2-DOG and its 6-phosphate, as well as the intracellular distribution of hexokinase activity were determined in brain tissue. Soluble hexokinase activity in brain cortex was enhanced by 2-DOG, as also by thiopental, and even more pronounced by both drugs used together, Results from in vitro experiments suggest that solubilization of mitochondrial hexokinase after 2-DOG is mediated by intracellularly accumulated 2-DOG-6-phosphate. 2-DOG produced a significant impairment of neuronal activity, revealing EEG patterns similar to those caused by thiopental anesthesia. Cortical acetylcholine levels were elevated by 2-DOG, as well as by thiopental, and again both drugs showed an additive effect when used in combination. This effect which may be the result of an inhibition of acetylcholine release, was also detectable in mice in vivo after 5 g 2-DOG/kg i.p., whereas the same dose of 3-O-methylglucose had no effect. The results provide further evidence that mitochondrial hexokinase may be involved in the relationship between cerebral metabolism and brain function.

Delayed auditory brainstem responses in diabetes mellitus

Diabetic patients have longer interpeak latencies in the brainstem auditory evoked responses than age-matched controls. The delay is not related to clinical hearing loss or blood glucose level at time of testing. Since waves I and II are normal in latency, the conduction velocity of the eighth nerve is not involved. The delay occurs between waves II and V, which would reflect altered transmission times in auditory brainstem and midbrain structures, and suggests the presence of a central neuropathy in patients with diabetes.

Insulin-induced hypoglycemic coma and regional cerebral energy metabolism

Swiss-Albino female mice weighing 20 g were rendered hypoglycemic by injecting insulin (2 units/kg). Animals were sacrificed at 40 min (pre-coma), 2 h (coma) and 4.5 h (recovery) after insulin injection by rapid submersion in liquid N2. Following sectioning at 20 micrometer, samples from the ascending reticular activating system and the inferior colliculus were freeze-dried and assayed for glucose, lactate, ATP and phosphocreatine (PCr). There was a preferential effect of hypoglycemia on ATP and PCr in cells of the ascending reticular activating system. ATP was depleted 30%, and PCr 55% in the pre-coma stage. ATP and PCr in cells from the inferior colliculus were not decreased. This selective effect on cells of the ascending reticular activating system followed by coma suggests that the coma per se may not represent total failure of the organism, but rather a compensatory mechanism designed to permit the animal to correct its compromised energy status.

The effect of severe hypoglycemia upon cerebrospinal fluid formation, ventricular iodide clearance, and brain electrolytes in rabbits

Severe insulin-induced hypoglycemia in rabbits reduces cerebrospinal fluid (CSF) formation, but not ventricular iodide clearance as measured by ventriculocisternal perfusion. This indicates that CSF production is ultimately glucose-dependent but that ventricular iodide clearance is not. The data suggest that severe hypoglycemia results in intracellular potassium loss within the brain and show that extracellular sodium replaces lost intracellular potassium. Hypoglycemia probably results in cellular adenosine triphosphate (ATP) reduction which affects membrane Na/K ATPase and the ability of the brain cell to maintain a potassium gradient. Potassium levels in the CSF also rise consequent to hypoglycemia. Homeostatic mechanisms that maintain a constant CSF potassium, therefore, are also affected by hypoglycemia.