Error detection during VMAT delivery using EPID-based 3D transit dosimetry.

PURPOSE: To investigate the effectiveness of an EPID-based 3D transit dosimetry system in detecting deliberately introduced errors during VMAT delivery. METHODS: An Alderson phantom was irradiated using four VMAT treatment plans (one prostate, two head-and-neck and one lung case) in which delivery, thickness and setup errors were introduced. EPID measurements were performed to reconstruct 3D dose distributions of "error" plans, which were compared with "no-error" plans using the mean gamma (γmean), near-maximum gamma (γ1%) and the difference in isocenter dose (ΔDisoc) as metrics. RESULTS: Out of a total of 42 serious errors, the number of errors detected was 33 (79%), and 27 out of 30 (90%) if setup errors are not included. The system was able to pick up errors of 5 mm movement of a leaf bank, a wrong collimator rotation angle and a wrong photon beam energy. A change in phantom thickness of 1 cm was detected for all cases, while only for the head-and-neck plans a 2 cm horizontal and vertical shift of the phantom were alerted. A single leaf error of 5 mm could be detected for the lung plan only. CONCLUSION: Although performed for a limited number of cases and error types, this study shows that EPID-based 3D transit dosimetry is able to detect a number of serious errors in dose delivery, leaf bank position and patient thickness during VMAT delivery. Errors in patient setup and single leaf position can only be detected in specific cases.

Characterization of the a-Si EPID in the unity MR-linac for dosimetric applications.

Electronic portal imaging devices (EPIDs) are frequently used in external beam radiation therapy for dose verification purposes. The aim of this study was to investigate the dose-response characteristics of the EPID in the Unity MR-linac (Elekta AB, Stockholm, Sweden) relevant for dosimetric applications under clinical conditions. EPID images and ionization chamber (IC) measurements were used to study the effects of the magnetic field, the scatter generated in the MR housing reaching the EPID, and inhomogeneous attenuation from the MR housing. Dose linearity and dose rate dependencies were also determined. The magnetic field strength at EPID level did not exceed 10 mT, and dose linearity and dose rate dependencies proved to be comparable to that on a conventional linac. Profiles of fields, delivered with and without the magnetic field, were indistinguishable. The EPID center had an offset of 5.6 cm in the longitudinal direction, compared to the beam central axis, meaning that large fields in this direction will partially fall outside the detector area and not be suitable for verification. Beam attenuation by the MRI scanner and the table is gantry angle dependent, presenting a minimum attenuation of 67% relative to the 90° measurement. Repeatability, observed over two months, was within 0.5% (1 SD). In order to use the EPID for dosimetric applications in the MR-linac, challenges related to the EPID position, scatter from the MR housing, and the inhomogeneous, gantry angle-dependent attenuation of the beam will need to be solved.

A back-projection algorithm in the presence of an extra attenuating medium: towards EPID dosimetry for the MR-Linac.

In external beam radiotherapy, electronic portal imaging devices (EPIDs) are frequently used for pre-treatment and for in vivo dose verification. Currently, various MR-guided radiotherapy systems are being developed and clinically implemented. Independent dosimetric verification is highly desirable. For this purpose we adapted our EPID-based dose verification system for use with the MR-Linac combination developed by Elekta in cooperation with UMC Utrecht and Philips. In this study we extended our back-projection method to cope with the presence of an extra attenuating medium between the patient and the EPID. Experiments were performed at a conventional linac, using an aluminum mock-up of the MRI scanner housing between the phantom and the EPID. For a 10 cm square field, the attenuation by the mock-up was 72%, while 16% of the remaining EPID signal resulted from scattered radiation. 58 IMRT fields were delivered to a 20 cm slab phantom with and without the mock-up. EPID reconstructed dose distributions were compared to planned dose distributions using the [Formula: see text]-evaluation method (global, 3%, 3 mm). In our adapted back-projection algorithm the averaged [Formula: see text] was [Formula: see text], while in the conventional it was [Formula: see text]. Dose profiles of several square fields reconstructed with our adapted algorithm showed excellent agreement when compared to TPS.

Automatic in vivo portal dosimetry of all treatments.

At our institution EPID (electronic portal imaging device) dosimetry is routinely applied to perform in vivo dose verification of all patient treatments with curative intent since January 2008. The major impediment of the method has been the amount of work required to produce and inspect the in vivo dosimetry reports (a time-consuming and labor-intensive process). In this paper we present an overview of the actions performed to implement an automated in vivo dosimetry solution clinically. We reimplemented the EPID dosimetry software and modified the acquisition software. Furthermore, we introduced new tools to periodically inspect the record-and-verify database and automatically run the EPID dosimetry software when needed. In 2012, 95% of our 3839 treatments scheduled for in vivo dosimetry were analyzed automatically (27,633 portal images of intensity-modulated radiotherapy (IMRT) fields, 5551 portal image data of VMAT arcs, and 2003 portal images of non-IMRT fields). The in vivo dosimetry verification results are available a few minutes after delivery and alerts are immediately raised when deviations outside tolerance levels are detected. After the clinical introduction of this automated solution, inspection of the detected deviations is the only remaining work. These newly developed tools are a major step forward towards full integration of in vivo EPID dosimetry in radiation oncology practice.

Catching errors with in vivo EPID dosimetry.

The potential for detrimental incidents and the ever increasing complexity of patient treatments emphasize the need for accurate dosimetric verification in radiotherapy. For this reason, all curative treatments are verified, either pretreatment or in vivo, by electronic portal imaging device (EPID) dosimetry in the Radiation Oncology Department of The Netherlands Cancer Institute-Antoni van Leeuwenhoek hospital, Amsterdam, The Netherlands. Since the clinical introduction of the method in January 2005 until August 2009, treatment plans of 4337 patients have been verified. Among these plans, 17 serious errors were detected that led to intervention. Due to their origin, nine of these errors would not have been detected with pretreatment verification. The method is illustrated in detail by the case of a plan transfer error detected in a 5 x 5 Gy intensity-modulated radiotherapy (IMRT) rectum treatment. The EPID reconstructed dose at the isocenter was 6.3% below the planned value. Investigation of the plan transfer chain revealed that due to a network transfer error, the plan was corrupted. 3D analysis of the acquired EPID data revealed serious underdosage of the planning target volume: On average 11.6%, locally up to 20%. This report shows the importance of in vivo (EPID) dosimetry for all treatment plans as well as the ability of the method to assess the dosimetric impact of deviations found.

A systematic review of the clinical effectiveness and cost-effectiveness of enzyme replacement therapies for Fabry's disease and mucopolysaccharidosis type 1.

OBJECTIVES: To determine the clinical effectiveness and cost-effectiveness of the administration of intravenous enzyme replacement therapy (ERT) to symptomatic patients for the prevention of long-term damage and symptoms in Fabry's disease and in mucopolysaccharidosis type 1 (MPS1). DATA SOURCES: Electronic databases from inception up to mid-2004. Contact with clinical experts. REVIEW METHODS: Relevant studies were identified and assessed using recommended quality criteria. RESULTS: The results suggested beneficial effects of ERT for Fabry's disease on measures of pain, cardiovascular function and some end-points reflecting neurosensory function. Renal function appeared to be stabilised by ERT. At present there are no utility-related health-related quality of life data on which to assess the relative health gain of ERT in MPS1. In order to be able to demonstrate the full extent of health gain from treatment, it was necessary to review the natural history of untreated patients in each disease in order to try to estimate the health loss prevented. The published information for Fabry's disease tallied with descriptions of a multi-system, life-threatening disorder particularly involving kidney, heart and brain with individual patients exhibiting many manifestations. The fragmentary information reviewed in 16 studies relevant to the natural history of MPS1 did not generate a coherent picture of disease progression and could provide little added value to published narrative reviews. For Fabry's disease, the mean cost per patient (50 kg) treated is around pounds sterling 85,000 per annum in England and Wales. The cost per patient varies considerably by dose. No published evidence reporting an economic evaluation of ERT for Fabry's disease was identified by this review. A dynamic decision model was constructed based on a birth cohort of male patients who are followed up until death. Owing to lack of information reported in the literature, many assumptions had to be applied. The key assumptions were that ERT returns patients to full health and a normal life expectancy. As far as possible, all assumptions favoured rather than detracted from the value of ERT. ERT was assumed to restore patients to full health in the base case. The estimated incremental cost-effectiveness ratio (ICER) in the base case was pounds sterling 252,000 per QALY (agalsidase beta). Univariate sensitivity analysis around the key assumptions produced ICERs ranging from pounds sterling 602,000 to pounds sterling 241,000. The base case unit cost of ERT was taken as pounds sterling 65.1/mg based on the cost of agalsidase beta. The unit cost would have had to be reduced to pounds sterling 9 to obtain an ICER of pounds sterling 30,000 per QALY. For MPS1, the mean cost per child patient (20 kg) treated is approximately pounds sterling 95,000 and an adult (70 kg) around pounds sterling 335,000 per annum in England and Wales. The cost per patient varies considerably by dose. There is no published evidence reporting an economic evaluation of ERT for MPS1 and no study was identified that reported the quality of life of MPS1 patients within a utility format. Furthermore, no or minimal information of the severity and rate of change of clinical manifestations of disease or the impact of ERT on these factors was identified. Information on the effect of ERT on mortality is also lacking owing to the relatively short time that the treatment has been available. Given this lack of data, it was not possible to develop a cost-effectiveness model of ERT treatment for MPS1 as the model would consist almost completely of assumptions based on no published evidence, leading to an incremental cost per QALY result that would be meaningless. CONCLUSIONS: Although ERT for treating the 'average' patient with Fabry's disease exceeds the normal upper threshold for cost-effectiveness seen in NHS policy decisions by over sixfold, and the value for MPS1 is likely to be of a similar order of magnitude, clinicians and the manufacturers argue that, as the disease is classified as an orphan disease under European Union legislation, it has special status, and the NHS has no option but to provide ERT. More information is required before the generalisability of the findings can be determined. Although data from the UK have been used wherever possible, this was very thin indeed. Nonetheless, even large errors in assumptions made will not reduce the ICER to anywhere near the upper level of treatments usually considered cost-effective. In order to overcome limited evidence on the natural history of the disease and the clinical effectiveness of the intervention, the establishment of disease-specific data registries is suggested to facilitate the process of technology assessment and improving patient care. These registries should attempt to include all affected patients in the UK, and collect longitudinal patient level data on clinically relevant problems, interventions received and quality of life in a utility format.

Experimental proof of a structural origin for the shadow fermi surface of Bi2Sr2CaCu2O8+delta.

By combining surprising new results from a full polarization analysis of nodal angle-resolved photoemission data from pristine and modulation-free Bi(2)Sr(2)CaCu(2)O(8+delta) with structural information from LEED and ab initio one-step photoemission simulations, we prove that the shadow Fermi surface in these systems is of structural origin, being due to orthorhombic distortions from tetragonal symmetry present both in surface and bulk. Consequently, one of the longest standing open issues in the investigation of the Fermi surface of these widely studied systems finally meets its resolution.

Metabolic abnormalities and grade of encephalopathy in acute hepatic failure.

Acute hepatic failure is associated with many biochemical abnormalities in plasma and brain. Changes that correlate well with the degree of behavioral impairment may be important factors in the development of encephalopathy. We measured the concentrations of intermediary metabolites, ammonia, and amino acids in brain and plasma and the rate of whole-brain glucose utilization in rats with an acutely devascularized liver. In all rats an estimate of the grade of encephalopathy (reflected by behavioral impairment) was made. Rats underwent portacaval shunting and hepatic artery ligation (or sham operation) and were kept normoglycemic and normothermic thereafter. We sampled blood and whole brain (by near-instantaneous freeze-blowing) 2, 4, or 6 h later. There were no alterations in levels of high-energy phosphate metabolites in the brain or in metabolites associated with the glycolytic pathway and Krebs cycle, except lactate and pyruvate. Brain glucose use was decreased similarly at all times after surgery. Levels of ammonia and many amino acids were increased in brain and plasma; brain aspartate, glutamate, and arginine levels were decreased. The increases in content of plasma ammonia and brain glutamine, proline, alanine, and aromatic amino acids and the decreases in brain aspartate and glutamate were most strongly correlated with behavioral impairment.

Optimizing the measurement of regional cerebral glucose consumption with [6-14C]glucose.

[6-14C]Glucose is used to trace the cerebral metabolic rate of glucose (CMRGlc) in vivo in experiments lasting 5-10 min. Initially 14C is trapped in intermediary metabolite pools. Subsequently 14C is lost as a function of time and metabolic rate, primarily as 14CO2. Experiments were designed to evaluate the rate of 14C lost as 14CO2 or as [14C]lactate from brain labeled with [6-14C]glucose during times up to 15 min. CMRGlc was measured during 5, 7.5, 10 and 15 min in 60 brain areas. At longer times the loss of 14C was reflected by lower apparent values of brain CMRGlc. Arteriovenous measurements across brain revealed no significant loss of [14C]lactate in normal rats or rats with bicuculline-induced seizures. It was concluded that the primary form in which 14C was lost was as 14CO2. As expected, the rate of 14CO2 loss was greater in structures with high metabolic rates. The data were analyzed to determine the parameters necessary to rectify the data so that uniform values of CMRGlc were obtained up to 15 min. Tables were made to predict the degree of 14C loss as well as the 14C-metabolites/[6-14C]glucose ratio as a function of time and metabolic rate. These tables can be used to plan the maximum and minimum experimental times for optimal results.

Neomycin reduces the intestinal production of ammonia from glutamine.

The mechanism by which neomycin treatment reduces circulating ammonia concentrations was studied in normal and portacaval shunted rats. Rats were given neomycin for 3 days and then fasted for 24 hours to eliminate feces. Neomycin decreased arteriovenous differences of ammonia across the intestine even when the intestines were empty. Neomycin treatment lowered the activity of glutaminase in the intestinal mucosa and the rate of ammonia production from glutamine by isolated intestinal segments. The intestines from portacaval shunted rats had higher glutaminase activity (by 57%), and produced ammonia from glutamine at a greater rate (by 31%), than intestines from controls. Neomycin treatment lowered glutaminase activity and ammonia production in shunted rats, but glutaminase activity still remained higher than in controls (by 23%). The data indicate that the mechanism by which neomycin lower plasma ammonia is owing, at least in part, to a direct effect on the intestines. Specifically, neomycin causes a reduction in mucosal glutaminase activity and thereby decreases the ability of the mucosa to consume glutamine and produce ammonia.

Effect of reducing brain glutamine synthesis on metabolic symptoms of hepatic encephalopathy.

Liver failure, or shunting of intestinal blood around the liver, results in hyperammonemia and cerebral dysfunction. Recently it was shown that ammonia caused some of the metabolic signs of hepatic encephalopathy only after it was metabolized by glutamine synthetase in the brain. In the present study, small doses of methionine sulfoximine, an inhibitor of cerebral glutamine synthetase, were given to rats either at the time of portacaval shunting or 3-4 weeks later. The effects on several characteristic cerebral metabolic abnormalities produced by portacaval shunting were measured 1-3 days after injection of the inhibitor. All untreated portacaval-shunted rats had elevated plasma and brain ammonia concentrations, increased brain glutamine and tryptophan content, decreased brain glucose consumption, and increased permeability of the blood-brain barrier to tryptophan. All treated rats had high ammonia concentrations, but the brain glutamine content was normal, indicating inhibition of glutamine synthesis. One day after shunting and methionine sulfoximine administration, glucose consumption, tryptophan transport, and tryptophan brain content remained near control values. In the 3-4-week-shunted rats, which were studied 1-3 days after methionine sulfoximine administration, the effect was less pronounced. Brain glucose consumption and tryptophan content were partially normalized, but tryptophan transport was unaffected. The results agree with our earlier conclusion that glutamine synthesis is an essential step in the development of cerebral metabolic abnormalities in hyperammonemic states.

Pentobarbital-enhanced [3H]flunitrazepam binding throughout the rat brain: an autoradiographic study.

The gamma-aminobutyric acidA/benzodiazepine receptor contains distinct ligand binding sites for hypnotic barbiturates and benzodiazepines. It is thought that barbiturate-induced sedation is produced, in part, by enhancing agonist binding to this receptor. The present study tested the hypothesis that pentobarbital would enhance benzodiazepine binding in a site-specific manner across the rat brain. In vitro receptor autoradiography was used to localize and quantitatively map [3H]flunitrazepam ([3H]FLU) binding in the absence and presence of pentobarbital in 133 brain areas. Each area demonstrated a statistically significant increase in [3H]FLU binding in the presence of in vitro pentobarbital (P < or = 0.05). Hindbrain nuclei dominated the top 20% of brain areas demonstrating the greatest pentobarbital-induced increases in [3H]FLU binding. The greatest mean percent increase in [3H]FLU binding occurred in the medulla, including areas known to be important for cardiovascular control, breathing, motor tone and regulating levels of arousal. These findings show that differential enhancement of benzodiazepine binding in the presence of pentobarbital occurred in brain areas controlling physiological functions known to be impaired by systemically administered pentobarbital.

Ischemic brain damage is not ameliorated by 1,3-butanediol in hyperglycemic rats.

BACKGROUND AND PURPOSE: Treatment with the ketone body precursor 1,3-butanediol has been suggested to ameliorate hypoxic/ischemic brain damage. Butanediol could provide an alternative energy substrate for the brain, thereby decreasing the amount of glycolytically produced lactate. Hyperglycemia aggravates brain damage after brain ischemia and causes fatal postischemic seizures, probably by increasing the production of lactate and decreasing the pH. We studied whether butanediol treatment altered the adverse consequences following ischemia complicated by hyperglycemia. METHODS: Hyperglycemic adult male rats were given 25 or 50 mmol.kg-1 body wt butanediol intravenously 30 minutes before 10 minutes of transient forebrain ischemia. Morphological evaluation was performed following perfusion-fixation after 15 hours of recovery. Blood concentrations of beta-hydroxybutyrate, acetoacetate, glucose, and lactate and brain tissue concentrations of energy metabolites were measured before and after ischemia. RESULTS: Blood levels of ketone bodies increased in the butanediol-treated rats. Ischemia decreased the blood levels of acetoacetate but increased the levels of beta-hydroxybutyrate by a similar amount, resulting in unchanged high levels of total ketone bodies in the animals that received butanediol. Brain tissue levels of glucose, energy metabolites, and lactate showed no difference between butanediol- and saline-treated rats. Furthermore, compared with saline-treated animals butanediol-treated rats showed no decrease in brain damage and no attenuation in the development of postischemic seizures. CONCLUSIONS: The ketone body precursor 1,3-butanediol offers no protective effect in transient forebrain ischemia complicated by hyperglycemia.

Regional distribution and kinetics of three sites on the GABAA receptor: lack of effect of portacaval shunting.

The regional distribution of binding sites on the GABAA receptor and their kinetic parameters were measured by quantitative autoradiography in brains from normal rats and rats with a portacaval shunt, a model of portal systemic encephalopathy in which GABA neurotransmission may be altered. The ligands used were [3H]flunitrazepam (a benzodiazepine-site agonist), [3H]-Ro15-1788 (a benzodiazepine-site antagonist), [3H]muscimol (a GABA-site agonist), and [35S]t-butylbicyclophosphorothionate (35S-TBPS, a convulsant that binds to a site near the chloride channel). Some brains were analyzed by computerized image analysis and three-dimensional reconstruction. The regional distribution of binding of the benzodiazepines was very similar, but the patterns obtained with [3H]muscimol and [35S]TBPS were different in many areas, suggesting a heterogeneous distribution of several subtypes of the GABAA receptor. The kinetic parameters were determined in brain regions for [3H]flunitrazepam, [3H]Ro15-1788, and [3H]muscimol. For each ligand, the Kd showed a significant heterogeneity among brain regions (at least threefold), contrary to conclusions drawn from earlier studies. In portacaval shunted rats, binding of all four ligands was essentially unchanged from that in control rats, indicating that, if there was an abnormality in GABA neurotransmission during portal systemic shunting, it was not reflected by altered binding to the main sites on the GABAA receptor.

Regional blood-brain barrier transport of ketone bodies in portacaval-shunted rats.

The permeability of the blood-brain barrier to ketone bodies, substrates of the monocarboxylic acid carrier, was measured in individual brain structures of control and portacaval-shunted rats. The measurements were made 5-7 wk after the shunt or sham operation by quantitative autoradiography. Portacaval shunting caused the permeability to ketone bodies to decrease throughout the brain by approximately 70%. There was a striking change in the transport pattern in the cerebral cortex; deeper cortical layers were affected more than superficial layers. Ketone body consumption by brain is limited by the transport capacity of the monocarboxylic acid system. Therefore, in portacaval-shunted rats the very low activity of this system makes it unlikely that ketone bodies can make a substantial contribution during situations such as fasting. Likewise, other substrates of the monocarboxylic acid system, e.g., lactate and pyruvate, will have greatly restricted access to the brain after portacaval shunting. If the carrier is symmetrical, another consequence will be that exit of endogenously produced lactate will be retarded.

Hyperammonaemia causes many of the changes found after portacaval shunting.

1. Portacaval shunting in rats results in several metabolic alterations similar to those seen in patients with hepatic encephalopathy. The characteristic changes include: (a) diminution of cerebral function; (b) raised plasma ammonia and brain glutamine levels; (c) increased neutral amino acid transport across the blood-brain barrier; (d) altered brain and plasma amino acid levels; and (e) changes in brain neurotransmitter content. The aetiology of these abnormalities remains unknown. 2. To study the degree to which ammonia could be responsible, rats were made hyperammonaemic by administering 40 units of urease/kg body weight every 12 h and killing the rats 48 h after the first injection. 3. The changes observed in the urease-treated rats were: (a) whole-brain glucose use was significantly depressed, whereas the levels of high-energy phosphates remained unchanged; (b) the permeability of the blood-brain to barrier to two large neutral amino acids, tryptophan and leucine, was increased; (c) blood-brain barrier integrity was maintained, as indicated by the unchanged permeability-to-surface-area product for acetate; (d) plasma and brain amino acid concentrations were altered; and (e) dopamine, 5-hydroxytryptamine (serotonin) and noradrenaline levels in brain were unchanged, but 5-hydroxyindoleacetic acid (5-HIAA), a metabolite of 5-hydroxytryptamine, was elevated. 4. The depressed brain glucose use, increased tryptophan permeability-to-surface-area product, elevated brain tryptophan content and rise in the level of cerebral 5-HIAA were closely correlated with the observed rise in brain glutamine content. 5. These results suggest that many of the metabolic alterations seen in rats with portacaval shunts could be due to elevated ammonia levels. Furthermore, the synthesis or accumulation of glutamine may be closely linked to cerebral dysfunction in hyperammonaemia.

Early establishment of cerebral dysfunction after portacaval shunting.

Portacaval shunting in rats results in brain dysfunction, as indicated by reduced energy metabolism and behavioral abnormalities, as well as many biochemical changes in plasma and brain. No etiological connections have been made between these findings, which have been studied mainly 2 wk or more after shunting. To determine how soon the various abnormalities occur and which are associated temporally with the decrease in brain glucose use, we studied shunted and sham-operated rats between 6 h and 11 days after surgery. Six hours after portacaval shunting, plasma aromatic amino acids, brain glutamine, aromatic amino acids, 5-hydroxyindoleacetic acid, and tryptophan transport into the brain were all significantly higher than normal. Brain glucose use showed a downward trend and was fully depressed within 1 day. Plasma branched-chain amino acids and threonine were decreased, and brain serotonin and norepinephrine content increased only after 2 days; these changes were therefore dissociated from the other abnormalities that developed in a shorter period. The results showed that the cerebral dysfunction characteristic of portacaval shunting began within hours and was fully established by 2 days.

Ischemia in normoglycemic and hyperglycemic rats: plasma energy substrates and hormones.

Seizures are a documented complication to cerebral ischemia. After 10 min of forebrain ischemia in rats, preischemic hyperglycemia invariably leads to severe, most often fatal epileptic attacks. This outcome is related to the exaggerated lactic acidosis, which has been suggested as a possible contributor to severe membrane changes and widespread edema. To find out if circulating hormones or plasma energy substrates modulate this additive damage caused by the hyperglycemia, plasma concentrations of of corticosterone, epinephrine, norepinephrine, dopamine, glucagon, insulin, glucose, free fatty acids (FFA), 3-hydroxybutyrate, and acetoacetate were measured before and in the early recirculation period after 15 min of forebrain ischemia in the rat. Plasma corticosterone levels did not differ between the normo- and hyperglycemic groups. Although not significantly different from control, the catecholamine levels showed a tendency to be higher in the hyperglycemic groups. Therefore, because catecholamines have been reported to have a protective effect during ischemia the present result cannot explain why hyperglycemia aggravates the ischemic damage. Insulin levels seemed to increase during ischemia but not significantly. Levels quickly returned to normal after 30 min of recirculation. FFA concentrations were reduced after the induction of ischemia and appeared lower in all hyperglycemic groups. The level of one of the ketone bodies, 3-hydroxybutyrate, showed a significant decrease in hyperglycemic ischemia in all groups compared with normoglycemic ischemia. The same tendency was seen for acetoacetate. Results are compatible with a protective role of ketone bodies in ischemia. It is concluded that among the hormones and substrates studied only the ketone body concentrations qualify as a modulator of the exaggerated brain damage after ischemia in hyperglycemic subjects.