Superoxide dismutase potentiates platelet-activating factor-induced injury in perfused lung

Platelet-activating factor (PAF) causes pulmonary hypertension and lung edema in animals and isolated perfused lungs by poorly understood mechanisms. Because oxidative mechanisms have been implicated in PAF-mediated cellular injury, we tested the hypothesis that superoxide anion (O2-.) contributes to PAF-induced lung injury by determining whether superoxide dismutase (SOD) could prevent the lung injury. Isolated rabbit lungs were perfused with PAF (100 nM) at a dose that caused transient hypertension and mild edema. Lungs pretreated with Cu,Zn SOD (100 U/ml) for 10 min developed persistent pulmonary hypertension and more lung edema formation in response to PAF. Enhanced responses to PAF also were observed in lungs perfused with 200 U/ml Cu,Zn SOD, but not with 10 or 40 U/ml Cu,Zn SOD. The higher doses of SOD also decreased thromboxane B2 levels in the perfusate. Potentiation of the PAF effect by Cu,Zn SOD was eliminated if the enzyme was inactivated or if the lung was treated with an anion channel blocker. The augmented PAF response in the presence of SOD was not altered by catalase (200 U/ml) or by nitric oxide synthase inhibitor. The data suggest that excessive Cu,Zn SOD enzyme activity potentiates PAF-induced injury in perfused rabbit lung presumably by overscavenging extracellular O2.- generated from intercellular sources. The augmented responses to PAF are not directly attributable to increased hydrogen peroxide, nitric oxide-related products, or thromboxane A2 production. These results suggest the new hypothesis that a balance between O2-. production and its metabolism determines vascular and endothelial responses to PAF.

Natural surfactant and hyperoxic lung injury in primates. II. Morphometric analyses

Diffuse lung injury is accompanied by low compliance and hypoxemia with histological evidence of endothelial and alveolar epithelial cell disruption. The histological effects of treatment of an acute diffuse lung injury with a natural surfactant product were evaluated in a primate model because surfactant function and content have been shown to be abnormal in diffuse lung injury in both animals and humans. Ten baboons were ventilated with 100% O2 for 96 h, and 5 were given an aerosol of natural porcine surfactant. Physiological and biochemical measurements of the effects of hyperoxia and surfactant treatment are presented in a companion paper. After O2 exposure, lungs were fixed and processed for quantitative electron microscopy. The responses to O2 included epithelial and endothelial cell injuries, interstitial edema, and inflammation. The hyperoxic animals treated with surfactant were compared with the untreated animals; the treatments altered neutrophil distribution, fibroblast proliferation, and changes in the volumes of type I epithelial cells and endothelial cells. Surfactant-treated animals also had decreased lamellar body volume density in type II epithelial cells and preservation of endothelial cell integrity. These changes suggest complex effects of natural surfactant on the pulmonary response to hyperoxia, including protection against epithelial and endothelial cell destruction as well as significant interstitial inflammation and fibroblast proliferation. We conclude that natural surfactant treatment of hyperoxic lung injury in primates resulted in partial protection of epithelial and endothelial cells but also increased the accumulation of fibroblasts in the lung.

Bactericidal antibiotics increase tumor necrosis factor-alpha and cardiac output in rats after cecal ligation and puncture

We hypothesized that treatment of experimental sepsis with bactericidal antibiotics, known to enhance microbial toxin release, would alter tumor necrosis factor-alpha production and the hemodynamic response to the syndrome. In the rat, after cecal ligation and puncture (CLP), elevated serum TNF levels and cardiac output were observed following antibiotic treatment. TNF and cardiac output were elevated to a greater extent in bactericidal-treated than bacteriostatic-treated or antibiotic-untreated rats. Animals treated with bactericidal antibiotics also had significantly greater cardiac outputs than untreated rats. Despite increases in circulating TNF with antibiotic administration, the mortality rate at 96 hr decreased after either bactericidal or bacteriostatic antibiotics. We conclude that elevated TNF after CLP in rats treated with antibiotics is associated with enhanced hemodynamic responses to CLP, but does not increase early mortality. In this model of polymicrobial sepsis, bactericidal and bacteriostatic antibiotics led to different hemodynamic effects without compromising survival.

PO2-dependent hydroxyl radical production during ischemia-reperfusion lung injury

Pulmonary ischemia-reperfusion results in transient hypertension and edema formation. Implicated in this injury are partially reduced oxygen species including the highly reactive hydroxyl radical. We measured ischemia-reperfusion injury and hydroxyl radical production following 90 min of either air-ventilated, N2-ventilated, or nonventilated ischemia in an isolated rabbit lung preparation. We found that edema formation was independent of alveolar oxygen tension (PO2); all ischemic groups had similar edema formation, regardless of the type of ventilation. Weight gain was 37-50 g of fluid during 40 min of reperfusion. Production of hydroxyl radical, measured by nonenzymatic hydroxylation of salicylate, was influenced by PO2 with a significant increase after air-ventilated ischemia (P < 0.05) but not after N2-ventilated ischemia. Treatment with dimethylthiourea or superoxide dismutase reduced edema formation 60-80% after air (P < 0.05)- and N2 (P < 0.05)-ventilated ischemia, whereas treatment with catalase protected only N2-ventilated ischemia (P < 0.05). Our results implicate two distinct mechanisms by which partially reduced oxygen species may contribute to pulmonary ischemia-reperfusion injury. One is by a mechanism capable of generating hydroxyl radical at normal PO2; the second is from reactions active at low PO2, the products of which are metabolized readily by extracellular enzymatic scavengers. The precise mechanisms of oxidant generation are not clear, but the findings suggest that a complex oxidative injury occurs during ischemia-reperfusion.

Myocardial oxygenation in dogs during partial and complete coronary artery occlusion

Regional myocardial oxygenation was assessed during partial and complete coronary artery occlusion using near infrared spectroscopy. In eight open-chest dogs, partial occlusions resulting in an approximately 42% decrease in left anterior descending coronary artery (LAD) blood flow produced an approximately 21% decrease in tissue O2 stores (tissue oxyhemoglobin plus oxymyoglobin) and no change in the oxidation level of mitochondrial cytochrome aa3. An approximately 81% reduction in LAD blood flow produced nadir levels of tissue oxyhemoglobin plus oxymyoglobin, maximal levels of deoxyhemoglobin plus deoxymyoglobin, a decline in tissue blood volume, and an approximately 39% decrease in cytochrome aa3 oxidation level. These changes were associated with an approximately 52% decrease from the preischemic baseline in mean transmural myocardial blood flow, measured by radiolabeled microspheres, and an approximately 41% decrease in myocardial O2 consumption. Complete occlusion resulted in further decreases in myocardial blood flow, O2 consumption, tissue blood volume, and cytochrome aa3 oxidation state but also produced increases in tissue O2 stores to above the nadir levels noted during partial occlusion. These results indicate that decreases in O2 delivery during partial coronary occlusion increase O2 extraction to sustain mitochondrial O2 availability, but as little as a 52% reduction in myocardial blood flow produces maximal O2 extraction and depletion of tissue O2 stores. Mitochondrial O2 availability is restricted further during complete occlusion because of limited O2 delivery and, possibly, decreases in tissue blood volume and O2 extraction.

Effects of muscle contraction on cytochrome a,a3 redox state

The relationships among mitochondrial O2 availability, O2 delivery, and lactate formation in exercising skeletal muscle remain unclear. Some data suggest that muscle O2 provision is sufficient at maximal O2 consumption (VO2max) to challenge the concept of a mitochondrial O2 limitation at VO2max. The relationships among VO2, mitochondrial O2 availability, and net lactate production were studied over a wide range of exercise intensities. Using near-infrared spectroscopy, the oxidation-reduction state of cytochrome a,a3 was monitored in the canine gracilis in vivo. Twenty adult dogs were anesthetized with alpha-chloralose, intubated, and mechanically ventilated on room air. Five-minute stimulation periods at rates of 2, 3, 4, 5, 7, 8, 10, or 12 stimuli/s were performed. VO2max generally was achieved at a stimulation rate of 8 stimuli/s; mean VO2max was 0.12 +/- 0.09 (SE) ml.min-1 x g-1. The concentration of oxidized mitochondrial cytochrome a,a3 decreased at all work loads relative to resting state and demonstrated a near-linear relationship with muscle VO2 (r2 = 0.99). Muscle lactate efflux and the lactate-pyruvate ratio also were correlated positively with cytochrome a,a3 reduction, suggesting a common regulatory mechanism coupling the processes of aerobic glycolysis and oxidative phosphorylation. At VO2max, the corresponding cytochrome oxidation was not significantly different from that observed at death. Thus, in the gracilis maximal exercise leads to near-complete reduction of cytochrome a,a3 secondary to deficient O2 provision. We conclude that VO2max is limited primarily by O2 delivery to this muscle and not by other factors limiting mitochondrial ATP production or substrate oxidation.

Cold-induced brain edema in mice. Involvement of extracellular superoxide dismutase and nitric oxide

The role of extracellular superoxide in the pathogenesis of vasogenic edema was studied using transgenic mice expressing a 5-fold increase in extracellular superoxide dismutase (EC-SOD) activity in their brains. Increased EC-SOD expression offered significant protection against edema development after cold-induced injury (44% less edema than nontransgenic littermates, p < 0.05). Since iron may contribute to vasogenic edema by catalyzing the production of hydroxyl radical from superoxide and hydrogen peroxide, the effects of the chelator deferoxamine were studied. Deferoxamine reduced edema formation after cold-induced injury (43% less edema than controls, p < 0.05); however, treatment with iron-saturated deferoxamine also reduced edema development in mice (32-48% less edema, p < 0.05). This suggested that the protection offered by deferoxamine was independent of its ability to chelate iron. An iron-independent mechanism by which superoxide can contribute to vasogenic edema is via reaction with nitric oxide to produce the potentially toxic peroxynitrite anion, which is also scavenged by deferoxamine. Mice treated with an inhibitor of nitric oxide synthase were protected against cold-induced edema (37% less edema, p < 0.05). EC-SOD transgenic mice received no additional protection by inhibition of nitric oxide synthesis, supporting this novel alternative mechanism of edema formation.

Inhibition of nitric oxide synthase increases extracellular cerebral glutamate concentration after global ischemia

The effects of nitric oxide synthase (NOS) inhibition on extracellular glutamate release were investigated in rats during global brain ischemia and reperfusion (IR) using cerebral microdialysis. A dialysis probe was inserted into the hippocampus of anesthetized rats. Forebrain ischemia was produced by hypotension and occlusion of both carotid arteries. After 15 min, brain flow was restored for 60 min. Time-dependent changes in the dialysate glutamate concentration were analyzed with HPLC in both control rats and those treated with N omega-nitro-L-arginine methyl ester 30 min prior to ischemia. The data show that the NOS inhibitor did not prevent glutamate release from hippocampus during ischemia. Inhibition of NOS also enhanced glutamate release during reperfusion resulting in dialysate concentrations up to 10 times higher than control values.

Pulmonary drug toxicity in patients with primary breast cancer treated with high-dose combination chemotherapy and autologous bone marrow transplantation

A protocol consisting of standard-dose adjuvant chemotherapy, high-dose combination alkylating agent chemotherapy, and autologous bone marrow transplant (ABMT) used at our institution for patients with primary breast cancer and extensive axillary lymph node involvement has been associated with a clinical syndrome of pulmonary drug toxicity in 23 of 59 patients (39%). In 10 patients in whom open-lung biopsies or transbronchial lung biopsies were obtained, we correlated the pulmonary pathology with the clinical features of the syndrome. These 10 patients presented with dyspnea, cough, fever, and hypoxemia at a mean time of 48 +/- 14 days after initiation of high-dose chemotherapy. Chest radiographs and CT scans showed interstitial and alveolar opacities. Pulmonary function tests revealed restrictive lung disease and reduced diffusing capacities. Open-lung and transbronchial lung biopsies showed alveolar septal thickening with fibrosis, atypical Type II pneumocytes, and pulmonary endothelial cell injury characteristic of drug toxicity. Corticosteroid therapy resulted in clinical improvement in 7 of 10 patients, but significant pulmonary function abnormalities remained. Local radiation therapy to the chest wall and regional lymph nodes appeared to exacerbate preexisting pulmonary drug toxicity in 4 patients. Two agents in the protocol, cyclophosphamide and carmustine (BCNU), can be implicated in the pathogenesis of this syndrome, and these agents most likely act synergistically to deplete reduced glutathione and impair antioxidant defenses. Since these drugs appear to contribute to the protocol in prolonging disease-free survival, prophylactic therapy of the lung should be investigated to reduce the high incidence of pulmonary toxicity.

Cerebral amino acid, norepinephrine and nitric oxide metabolism in CNS oxygen toxicity

CNS oxygen (O2) toxicity is complex, and the etiology of its most severe manifestation, O2 convulsions, is yet to be determined. A role for depletion of the brain GABA pool has been proposed, although recent data have implicated production of reactive O2 species, e.g. H2O2, in this process. We hypothesized that the production of H2O2 and NH3 produced by monoamine oxidase (MAO) would lead to depletion of GABA and production of nitric oxide (NO.) respectively, and thereby enhance CNS O2 toxicity. In this study, rats treated with an MAO inhibitor (pargyline) or a nitric oxide synthase inhibitor (LNNA) were protected against O2-induced convulsions. Selected cerebral amino acids including arginine were measured in control and O2 treated rats (6 ATA, 20 min) with or without drug pretreatment. After O2 exposure, the cerebral pools of glutamate, aspartate, and GABA decreased significantly while glutamine content increased relative to control (P < 0.05). After treatment with either enzyme inhibitor, glutamine, glutamate and aspartate concentrations were maintained near control levels. Remarkably, GABA depletion by O2 was not prevented despite protection from seizures by both pargyline and LNNA. The NO. precursor, arginine, was increased significantly in the brain by toxic O2 exposure, but both pargyline and LNNA inhibited this effect. Simultaneous norepinephrine measurements indicated that its storage substantially decreased during hyperoxia (P < 0.05), but this effect too was blocked by either pargyline or LNNA. These data indicate that protection against O2 by these inhibitors is not related to preservation of the GABA pool. More importantly, O2 dependent norepinephrine metabolism and NO. synthesis appear to be interactive during CNS O2 toxicity.

Hydrogen peroxide production by monoamine oxidase during ischemia-reperfusion in the rat brain

Monoamine oxidase (MAO) as a source of hydrogen peroxide (H2O2) was evaluated during ischemia-reperfusion in vivo in the rat brain. H2O2 production was assessed with and without inhibition of MAO during and after 15 min of ischemia. Metabolism of H2O2 by catalase during ischemia and reperfusion was measured in forebrain homogenates using aminotriazole (ATZ), an irreversible H2O2-dependent inhibitor of catalase. Catecholamine and glutathione concentrations in forebrain were measured with and without MAO inhibitors. During ischemia, forebrain blood flow was reduced to 8% of baseline and H2O2 production decreased as measured at the microperoxisome. During reperfusion, a rapid increase in H2O2 generation occurred within 5 min as measured by a threefold increase in oxidized glutathione (GSSG). The H2O2-dependent rates of ATZ inactivation of catalase between control and ischemia-reperfusion were similar, indicating that H2O2 was more available to glutathione peroxidase than to catalase in this model. MAO inhibitors eliminated the biochemical indications of increased H2O2 production and increased the catecholamine concentrations. Mortality was 67% at 48 h after ischemia-reperfusion, and there was no improvement in survival after inhibition of MAO. We conclude that MAO is an important source of H2O2 generation early in brain reperfusion, but inhibition of the enzyme does not improve survival in this model despite ablating H2O2 production.

Protection against platelet-activating factor-induced injury by interferon inducer in perfused rabbit lung

Platelet-activating factor (PAF) and the interferons (IFN) are released during sepsis and the adult respiratory distress syndrome. The proinflammatory nature of PAF and anti-inflammatory property of IFN led us to investigate interactions between these two mediators in an isolated perfused lung (IPL) preparation. In the IPL, mean pulmonary arterial pressure (Ppa), lung weight gain, and peak airway pressure (Paw) were monitored continuously for 1 h in six groups of rabbits: 1) control, 2) the IFN-alpha/beta inducer polyinosinic:cytidylic acid (polyI:C) alone, 3) PAF alone, 4) polyI:C + PAF, 5) indomethacin + PAF, and 6) AA861 (a 5-lipoxygenase inhibitor) + PAF. At the end of 1 h, microvascular pressure was determined by double-occlusion technique and partition of total pulmonary vascular resistance (RT) was calculated. Serial eicosanoid concentrations in the perfusate also were measured. PAF increased Ppa, Paw, lung weight gain, and RT. These changes were associated with increased thromboxane B2 and decreased leukotriene production. PolyI:C, which induced high levels of serum IFN in rabbits, blocked the PAF-induced increase in Ppa, Paw, lung weight gain, and RT, similar to indomethacin and AA861. PolyI:C suppressed PAF-stimulated release of thromboxane B2 and increased leukotriene levels in the perfusate. The PAF-induced lung responses also were attenuated by pretreatment with human recombinant IFN. These data indicate that polyI:C protects against PAF-induced responses in the rabbit IPL, most likely via its induction of IFN. This effect is related in part to inhibition of thromboxane A2 production stimulated by PAF and leukotrienes.

Generation of hydroxyl radical by crocidolite asbestos is proportional to surface [Fe3+]

Differences among fibrous silicates to effect injury in biological systems have been postulated to reflect oxidant generation by structural iron within the crystal lattice of amphiboles. Iron is also coordinated to the surface of all silicates in concentrations which depend on the density of acidic functional groups. We tested the hypothesis that oxidant generation by crocidolite is proportional to surface-complexed iron rather than variance in the lattice concentrations of this transition metal. Surface iron was quantified after its reduction to Fe2+ and chelation by citrate. Thiobarbituric acid (TBA) reactive products and dihydroxybenzoic acid products of salicylate were employed as indices of nonspecific oxidant and hydroxyl radical generation, respectively. Surface iron, TBA reactive products, and dihydroxybenzoic acid products all diminished after pretreatment of crocidolite with the metal chelator deferoxamine in concentrations varying from 0 to 250 mM. Inclusion of deferoxamine in the reaction mixture provided similar results of diminishing both TBA reactive products and dihydroxybenzoic acid generation. We conclude that oxidant generation by crocidolite is proportional to surface concentrations of iron which can be chelated using deferoxamine. The design of synthetic fibers without health effects after exposure will likely necessitate decreasing the number of surface acidic functional groups to diminish the capacity to complex iron (i.e., minimize the percentage SiO2).

Extracellular superoxide dismutase, nitric oxide, and central nervous system O2 toxicity

Although reactive O2 species appear to participate in central nervous system (CNS) O2 toxicity, the exact roles of different reactive O2 species are undetermined. To study the contribution of extracellular superoxide anion (O2-) to CNS O2 toxicity we constructed transgenic mice overexpressing human extracellular superoxide dismutase (ECSOD; superoxide:superoxide oxidoreductase, EC 1.15.1.1) in the brain. Remarkably, when exposed to 6 atm (1 atm = 101.3 kPA) of hyperbaric oxygen for 25 min, transgenic mice demonstrated higher mortality (83%) than nontransgenic litter-mates (33%; P < 0.017). Pretreatment with diethyldithiocarbamate, which inhibits both ECSOD and Cu/Zn superoxide dismutase (Cu/Zn SOD) activity, increased resistance to CNS O2 toxicity, in terms of both survival (100% in transgenics and 93% in nontransgenics) and resistance to seizures (4-fold increase in seizure latency in both transgenic and nontransgenic mice; P < 0.05). Thus, O2- apparently protects against CNS O2 toxicity. We hypothesized that O2- decreased toxicity by inactivating nitric oxide (NO.). To test this, we inhibited NO. synthase (EC 1.14.23) with N omega-nitro-L-arginine to determine whether NO. contributes to enhanced CNS O2 toxicity in transgenic mice. N omega-nitro-L-arginine protected both transgenic and nontransgenic mice against CNS O2 toxicity (100% survival and a 4-fold delay in time to first seizure; P < 0.05), as well as abolishing the difference in sensitivity to CNS O2 toxicity between transgenic and nontransgenic mice. These results implicate NO. as an important mediator in CNS O2 toxicity and suggest that ECSOD increases CNS O2 toxicity by inhibiting O2(-)-mediated inactivation of NO.

Mitochondrial oxidative stress after carbon monoxide hypoxia in the rat brain

To better understand the mechanisms of tissue injury during and after carbon monoxide (CO) hypoxia, we studied the generation of partially reduced oxygen species (PROS) in the brains of rats subjected to 1% CO for 30 min, and then reoxygenated on air for 0-180 min. By determining H2O2-dependent inactivation of catalase in the presence of 3-amino-1,2,4-triazole (ATZ), we found increased H2O2 production in the forebrain after reoxygenation. The localization of catalase to brain microperoxisomes indicated an intracellular site of H2O2 production; subsequent studies of forebrain mitochondria isolated during and after CO hypoxia implicated nearby mitochondria as the source of H2O2. In the mitochondria, two periods of PROS production were indicated by decreases in the ratio of reduced to oxidized glutathione (GSH/GSSG). These periods of oxidative stress occurred immediately after CO exposure and 120 min after reoxygenation, as indicated by 50 and 43% decreases in GSH/GSSG, respectively. The glutathione depletion data were supported by studies of hydroxyl radical generation using a salicylate probe. The salicylate hydroxylation products, 2,3 and 2,5-dihydroxybenzoic acid (DHBA), were detected in mitochondria from CO exposed rats in significantly increased amounts during the same time intervals as decreases in GSH/GSSG. The DHBA products were increased 3.4-fold immediately after CO exposure, and threefold after 120 min reoxygenation. Because these indications of oxidative stress were not prominent in the postmitochondrial fraction, we propose that PROS generated in the brain after CO hypoxia originate primarily from mitochondria. These PROS may contribute to CO-mediated neuronal damage during reoxygenation after severe CO intoxication.

Effects of inhibition and induction of cytochrome P-450 isozymes on hyperoxic lung injury in rats

Pulmonary oxygen toxicity most likely results from excessive production of reactive oxygen species. The role of the cytochromes P-450 in this process is controversial because these enzymes have been reported both to enhance hyperoxic lung injury and to protect from the damaging effects of 100% oxygen. We sought to further determine the role of the cytochromes P-450 in hyperoxic lung injury by inhibiting and inducing pulmonary cytochrome P-450 isozymes in rats. Treatment with the cytochrome P-450 inhibitor cimetidine or 8-methoxypsoralen did not improve survival or reduce lung edema in rats exposed to 100% oxygen. The activity of cytochrome P-450IIB1, the major pulmonary cytochrome P-450 isozyme in rats, was clearly inhibited by 8-methoxypsoralen. beta-Naphthoflavone (beta NF), a selective inducer of cytochrome P-450IA1, was administered in two-dose and five-dose regimens. The two-dose regimen produced significant and sustained induction of cytochrome P-450IA1 activity, but survival in these rats was not improved when exposed to 100% oxygen. In rats treated with five doses of beta NF, a small increase in survival time was found from 71.1 +/- 8.7 to 88.0 +/- 20.2 h; however, there was no difference in the induction of cytochrome P-450IA1 activity between this five-dose regimen and the two-dose regimen. The small improvement in survival after five doses of beta NF is thus unrelated to cytochrome P-450IA1 induction. We conclude that neither inhibition of cytochrome P-450IIB1 activity nor induction of cytochrome P-450IA1 activity protects adult rats against hyperoxic lung injury.

Pulmonary oxalate deposition associated with Aspergillus niger infection. An oxidant hypothesis of toxicity

Tissue injury by Aspergillus niger infection is associated with the deposition of calcium oxalate crystals. Oxalate is recognized to function as a ligand for numerous metal cations and will react with ferric ion to form a coordination complex. We describe oxalate deposition in the lung of a patient with A. niger infection and quantify surface-complexed Fe3+. Crystals collected from lung tissue demonstrated considerable concentrations of surface iron. In addition, we tested the hypothesis that this surface coordination of Fe3+ by oxalate is associated with increased in vitro oxidant generation. Calcium oxalate crystals (1.0 mg/ml) complexed all available Fe3+ from solutions of ferric chloride to concentrations of as much as 1.0 mM. Oxidant generation in both a chemical and a cellular system, measured as thiobarbituric-acid-reactive products of deoxyribose and chemiluminescence, respectively, increased with coordination of higher concentrations of inorganic iron. We conclude that calcium oxalate associated with A. niger infection complexes iron cations onto the crystalline surfaces and may generate oxidants at the solid-solution interface, which could result in tissue injury.

Recovery of energy metabolism in rat brain after carbon monoxide hypoxia

Carbon monoxide (CO) may inhibit mitochondrial electron transport in the brain and increase the toxic effects of the gas. This hypothesis was investigated in anesthetized rats during CO exposure and recovery at either normobaric or hyperbaric O2 concentrations. During exposure and recovery, we measured the oxidation level of cerebrocortical cytochrome c oxidase by differential spectroscopy and biochemical metabolites known to reflect aerobic energy provision in the brain. CO exposure (HbCO = 71 +/- 1%) significantly decreased blood pressure and cytochrome oxidation level. Cerebral ATP was maintained while lactate/pyruvate, glucose, and succinate rose, and phosphocreatine (PCr) fell, relative to control (P less than 0.05). Intracellular pH (pHi) calculated from the PCr equilibrium also declined during the exposures. During recovery, HbCO fell more rapidly at hyperbaric than at normobaric O2 levels, but returned to 10% or less in both groups by 45 min. Cytochrome oxidation state improved to 80% of control after 90 min at normobaric O2, but recovered completely after hyperbaric O2 (P less than 0.05). In normobaric O2, PCr and pHi continued to fall for 45 min after CO exposure and did not recover completely by 90 min. PCr and pHi in animals after hyperbaric O2 improved within 45 min, but also remained below control at 90 min. These data indicate that intracellular uptake of CO can impair cerebral energy metabolism, despite the elimination of HbCO from the blood.

Comparison of systemic oxygen delivery and uptake with NIR spectroscopy of brain during normovolemic hemodilution in the rabbit

Incremental hyperoxic normovolemic hemodilution was utilized to progressively decrease oxygen delivery (DO2) in anesthetized rabbits. At decreasing DO2, we compared systemic responses related to the adequacy of DO2, i.e. mixed venous oxygen saturation (SvO2), oxygen consumption (VO2), and arterial lactate concentrations, to near infrared spectroscopy (NIRS) of the brain, a regional measure of intracellular oxygen availability. We sought concomitantly to define critical SvO2 and DO2, beyond which whole body VO2 begins to decline and arterial lactate concentrations increase. NIR Spectroscopy provided the means to test the hypothesis that systemic indicators of inadequate DO2 would not accurately reflect the oxygenation of a critical organ such as the brain. In thirteen rabbits anesthetized with fentanyl, paralyzed and artificially ventilated at an FIO2 of 0.60, hemodilution produced an early decrease in mixed venous oxygen saturation. When mixed venous oxygen saturation decreased below approximately 50%, arterial lactate concentrations began to increase significantly. Further decreases in oxygen delivery precipitated a decline in systemic VO2. Finally, NIRS revealed an increase in the reduction level of brain cytochrome a,a3 after systemic parameters of oxygen delivery had been altered. Analysis of the data indicated that falling SvO2 predicted inadequate DO2 to tissue during early hemodilution under narcotic/relaxant anesthesia and that the brain showed evidence of intracellular hypoxia only after systemic parameters such as SvO2 were affected markedly.

Warehouse workers' headache. Carbon monoxide poisoning from propane-fueled forklifts

We reviewed over 220 cases of acute carbon monoxide (CO) poisoning and now report on 17 patients whose poisoning occurred from the indoor use of propane-fueled forklifts. All patients in this series presented with neurologic symptoms or persistent headache and were given hyperbaric oxygen to resolve their symptomatology. We investigated the concentration of CO in the exhaust emissions of 12 propane-fueled forklifts used in various workplaces in our location. The average CO concentration in the exhaust during engine idling was 36,000 parts per million (3.6%). This value decreased slightly to 30,000 ppm (3.0%) at working engine speed. Measurements of exhaust flow indicate CO production rates of approximately 60 liters per minute at working engine speed. These quantities of CO constitute a significant occupational exposure risk to workers using propane-fueled forklifts in unventilated indoor environments.

Responses of baboons to prolonged hyperoxia: physiology and qualitative pathology

Cardiopulmonary responses to prolonged hyperoxia and their relationships to the development of lung pathology have not been fully characterized in primates. In this study, circulatory hemodynamics and pulmonary function, vascular permeability, and leukocyte sequestration were measured in male baboons after 100% O2 exposure and related to ultrastructural changes of lung injury by electron microscopy. Three groups of animals were exposed to 100% O2 in an exposure cage for 40, 66, and 80 h, respectively. A fourth group of animals was exposed in a cage for 80 h and then anesthetized and ventilated with 100% O2 for additional time. These animals were exposed for a total duration of 110 h or until death from the injury. Physiological responses to hyperoxia were characterized by decreases in total lung capacity and inspiratory capacity at 80 and 110 h. A significant increase in pulmonary leukocyte accumulation was noted by 80 h. Extravascular lung water and permeability surface-area product increased at 80 and 110 h. Cardiac output and stroke volume also decreased, and systemic vascular resistance increased after 80 and 110 h of hyperoxia. Histopathological changes were present in the lungs of all but the 40-h exposure group. Animals exposed for 66 h showed endothelial injury and neutrophil accumulation. By 80 h, animals showed endothelial cell destruction, interstitial edema, and type I cell injury. At 110 h, animals showed substantial destruction of endothelial and type I epithelial cells, exposure of alveolar basement membrane, congestion of capillaries, and substantial interstitial edema. The data indicate that histological changes by electron microscopy precede physiological responses to hyperoxic pulmonary injury in baboons by as much as 14 h and that the physiological responses to early hyperoxic injury are relatively insensitive to the pathological injury.

MR imaging of microcirculation in rat brain: correlation with carbon dioxide-induced changes in blood flow

Considerable interest has been shown in developing a magnetic resonance (MR) imaging technique with quantitative capability in the evaluation of tissue microcirculation ("perfusion"). In the present study, the flow-dephased/flow-compensated (FD/FC) technique is evaluated for measuring rat cerebral blood flow (CBF) under nearly optimal laboratory conditions. Imaging was performed on a 2.0-T system equipped with shielded gradient coils. Rat CBF was varied by manipulating arterial carbon dioxide pressure (PaCO2). In parallel experiments, optimized MR imaging studies (seven rats) were compared with laser Doppler flowmetry (LDF) studies (nine rats). LDF values showed a high degree of correlation between CBF and PaCO2, agreeing with results in the literature. MR imaging values, while correlating with PaCO2, showed considerable scatter. The most likely explanation is unavoidable rat motion during the requisite long imaging times. Because of this motion sensitivity, the FD/FC technique cannot provide a quantitative measure of CBF. It can, however, provide a qualitative picture.