Progression in acute stroke: value of the initial NIH stroke scale score on patient stratification in future trials

BACKGROUND AND PURPOSE

The objective was to determine the occurrence of neurological changes during the first 48 hours after acute stroke as it relates to initial stroke severity.

METHODS

The National Institutes of Health Stroke Scale (NIHSS) was performed serially for the first 48 hours on 127 consecutive ischemic stroke patients (129 strokes) admitted to the neuroscience intensive care unit. Incidence of stroke progression (a >/=3-point increase on the NIHSS) was recorded and analysis performed to determine its association with initial stroke severity and other demographic and physiological variables. Deficit resolution by 48 hours, defined as an NIHSS score of 0 or 1, measured the frequency of functional recovery predicted by the initial deficit.

RESULTS

Overall progression was noted in 31% of events (40/129). Applying Bayes' solution to the observed frequency of worsening, the greatest likelihood of predicting future patient progression occurs with stratification at NIHSS scores of </=7 and >7. Patients with an initial NIHSS of </=7 experienced a 14.8% (13/88) worsening rate versus a those with a score of >7 with a 65.9% (27/41) worsening rate (P<0.000005). Forty-five percent (40/88) of those with an initial score of </=7 were functionally normal at 48 hours, whereas only 2.4% (1/41) of those with scores of >7 returned to a normal examination within this period (chi2, P<0.000005).

CONCLUSIONS

This study suggests that the early clinical course of the neurological deficit after acute stroke is dependent on the initial stroke severity and that a dichotomy in early outcome exists surrounding an initial NIHSS score of 7. These findings may have significant implications for the design and patient stratification in treatment protocols with respect to primary clinical outcome.

Acute carpal tunnel syndrome in a diver: evidence of peripheral nervous system involvement in decompression illness

Conclusive evidence for involvement of the peripheral nervous system in decompression illness is lacking. We report a case of decompression illness associated with shoulder pain and the clinical features of median nerve injury at the wrist. Initial recompression and hyperbaric oxygen treatment produced prompt relief of all symptoms and signs, but carpal tunnel syndrome subsequently recurred. Nerve conduction studies confirmed median nerve conduction delay at the wrist. Repeat measurements after treatment with hyperbaric oxygen showed electrophysiologic improvement that was consistent with improvement in symptoms. We believe this is the first objectively substantiated case of injury to the peripheral nervous system caused by decompression illness.

Hyperbaric oxygen after global cerebral ischemia in rabbits reduces brain vascular permeability and blood flow

BACKGROUND AND PURPOSE

Hyperbaric oxygen (HBO) has been advocated as a therapy to improve neurological recovery after ischemia, since HBO may improve tissue oxygen delivery. We examined the effect of HBO treatment after global cerebral ischemia on early brain injury.

METHODS

Rabbits were subjected to 10 minutes of global cerebral ischemia by cerebrospinal fluid compression. After 30 minutes of reperfusion, rabbits either were subjected to HBO for 125 minutes and then breathed 100% O2 at ambient pressure for 90 minutes or breathed 100% O2 for 215 minutes. At the end of reperfusion and 90 minutes after exposure, brain vascular permeability and cerebral blood flow were measured. Somatosensory evoked potentials were monitored throughout the experiment.

RESULTS

HBO treatment reduced (P < .05) brain vascular permeability by 16% in gray matter and by 20% in white matter. Cerebral blood flow was lower (P < .05) in the HBO group (40.9 +/- 1.9 mL/min per 100 g, mean +/- SEM) compared with controls (50.8 +/- 2.0 mL/min per 100 g). Somatosensory evoked potential recovery was similar in the two groups (P > .05).

CONCLUSIONS

HBO administered after global cerebral ischemia promoted blood-brain barrier integrity. HBO treatment also reduced cerebral blood flow; this effect was not associated with a reduction in evoked potential recovery. Since neurological outcome after global cerebral ischemia is generally poor and treatment options are limited, HBO should be further investigated as a potential therapy.

Hyperbaric oxygen after global cerebral ischemia in rabbits does not promote brain lipid peroxidation

OBJECTIVE

To determine whether hyperbaric oxygen administered immediately after global cerebral ischemia increases free radical generation and lipid peroxidation in the brain or alters neurophysiologic recovery.

DESIGN

Prospective, randomized, controlled trial.

SETTING

Animal research laboratory.

SUBJECTS

Adult male New Zealand white rabbits.

INTERVENTIONS

Anesthetized rabbits were subjected to 10 mins of global cerebral ischemia by infusing a mock cerebrospinal fluid into the subarachnoid space and increasing intracranial pressure equal to mean arterial pressure. Immediately upon reperfusion, one group of rabbits (n = 9) was treated with hyperbaric oxygen at 2.8 atmospheres absolute for 75 mins while the control group (n = 9) breathed room air for an equivalent period of time. At the end of the reperfusion period, oxyradical brain damage was determined by measuring brain levels of oxidized and total glutathione and free malondialdehyde. Neurophysiologic brain injury was assessed with cortical somatosensory evoked potentials.

MEASUREMENTS AND MAIN RESULTS

Both oxidized glutathione and the ratio of oxidized glutathione to reduced glutathione (total minus oxidized) were higher (p < .05) in the hyperbaric oxygen group, indicating that hyperbaric oxygen increased free radical generation. Nonetheless, brain malondialdehyde content, an index of lipid peroxidation, was similar (p > .05) in the two groups. Cortical somatosensory evoked potential recovery at the end of reperfusion was 50% higher (p < .05) in the hyperbaric oxygen-treated animals compared with controls.

CONCLUSIONS

Treatment with hyperbaric oxygen after ischemia increased the amount of oxygen free radicals in the brain. However, this increase in free radical generation was not associated with an increase in lipid peroxidation or a reduction in neurophysiologic recovery when measured after 75 mins of recirculation. These results suggest that hyperbaric oxygen administered immediately after global ischemia does not promote early brain injury.

Vestibulo-ocular reflex gain as a measure of vestibular function in guinea pigs while in a recompression chamber: apparatus design and effects of nitrogen narcosis

There are several mechanisms whereby alteration of barometric pressure can produce vertigo in divers or aviators. Development of a reliable measure of vestibular function in an animal model is the first requirement for further study of these mechanisms. This report presents the development of a rotatory table device capable of evoking the vestibulo-ocular reflex (VOR) of a guinea pig while in a hyperbaric chamber. To assess the reproducibility of this response, eight animals were monitored by electronystagmography during rotations at three table velocities (62.4, 83.3, and 100 degrees/s). Two test sessions were performed on each animal with a 6-hour interval between sessions. The VOR gain was calculated by dividing the average peak velocity of the slow phase component of the nystagmus by the peak stimulus velocity. At least eight observations per test speed were averaged; calibration of eye movement was performed prior to each session by forced ocular abduction. Multifactorial analysis of variance revealed no significant differences (p > .05) between the differing rotation speeds nor between test sessions for individual animals. However, there was a significant difference in VOR gain between animals (p < .002). The VOR gain was then measured, using the same techniques, in another group of seven animals before, during, and after an air dive to the equivalent of 200 feet of seawater (7.06 atmospheres absolute) to assess the effects of nitrogen narcosis. Pre- and post-dive VOR gains were significantly greater than those measured at depth (p < .05). These results are consistent with the slow processing model of nitrogen narcosis and the controversial theory that central nervous system depressants decrease the VOR gain. The results also demonstrate the ability of this inexpensive apparatus to provide a sensitive measure of pressure-induced changes of vestibular function in guinea pigs.

Exercise conditioning reduces the risk of neurologic decompression illness in swine

During development of a pig model of neurologic decompression illness (DCI) we noted that treadmill-trained pigs seemed less likely to develop DCI than sedentary pigs. The phenomenon was formally investigated. Twenty-four immature, male, castrated, pure-bred Yorkshire swine were conditioned by treadmill running, while 34 control pigs remained sedentary. All pigs (weight 18.75-21.90 kg) were dived on air to 200 feet of seawater (fsw) in a dry chamber. Bottom time was 24 min. Decompression rate was 60 fsw/min. Pigs that developed neurologic DCI were treated by recompression. Pigs without neurologic signs were considered neurologically normal if they ran on the treadmill without gait disturbance at 1 and 24 h postdive. Of the 24 exercise-conditioned pigs, only 10 (41.7%) developed neurologic DCI, compared to 25 of 34 (73.5%) sedentary pigs (X2 = 5.97; P = < 0.015). Neither mean carcass density (adiposity) nor mean age were significantly different between groups. No patent foramen ovale was detected at necropsy. An additional control group of 24 pigs was dived to clarify the influence of weight. The results suggest that the risk of neurologic DCI is reduced by physical conditioning, and the effect is independent of differences in age, adiposity, and weight.

Failure to prevent decompression illness in rats by pretreatment with a soluble complement receptor

Controversy exists over the role of complement activation in the natural history of decompression illness (DCI), and whether an individual's predisposition to DCI might be influenced by susceptibility to activation of complement by intravascular gas bubbles. Treatment with a soluble complement receptor (sCR-1), which neutralizes activated complement components, is known to be beneficial in other complement-dependent disease processes. This study investigated the effect of treating rats with sCR-1 or saline before decompression from a dive profile known to produce a high incidence of DCI. No statistical difference in the incidence of DCI was observed between the 27 rats treated with sCR-1 and 26 control rats treated with saline. The study was unable to confirm the previously reported observation in rats of a positive correlation between DCI incidence and increasing weight.

Complement activation during saturation diving

In this study, the levels of activated complement fragments C3a and C5a were measured on 11 U.S. Navy divers as they performed a 28-day saturation dive to a pressure equivalent of 1,000 feet of seawater (fsw, 31.3 atm abs). Two subjects developed symptoms consistent with the high pressure nervous syndrome (HPNS) and three were treated for type I DCS (joint pain only). These events allowed us to test two hypotheses: a) alterations in C3a or C5a levels during compression are related to the occurrence of HPNS and b) increases in complement fragments are an indicator of decompression stress associated with type I DCS. There was no correlation between changes in C3a and C5a levels during compression and the diagnosis of HPNS. Our results suggest that an increase in C3a and C5a levels during saturation diving correlates with decompression stress and the clinical diagnosis of type I DCS.

Effects of an increased PO2 during recompression therapy for the treatment of experimental cerebral arterial gas embolism

In this study we investigated the efficacy of an initial compression to 6 atm abs on a 53% nitrogen:47% oxygen mixture (PO2 = 2.8 atm abs) before breathing oxygen at 2.8 and 1.9 atm abs for the treatment of feline cerebral arterial gas embolism. Neurophysiologic function was determined by measuring the cortical somatosensory evoked potential (SEP) amplitude in anesthetized ventilated cats. Air was infused into the carotid artery until the SEP amplitude was reduced to less than 10% of baseline values. The animals were randomly separated into 3 groups. The first group (CONTROL) (n = 7) served as control and remained at the surface, breathing air. The second group (NITROX) (n = 10) was compressed to 6 atm abs breathing a 53:47% nitrox mixture for 30 min followed by breathing 100% oxygen at 2.8 and 1.9 atm abs. The third group (HBO) (n = 10) was compressed to 2.8 and 1.9 atm abs breathing 100% oxygen. Air infusion suppressed the SEP amplitude to the same level in all groups. The CONTROL group recovered 27.6 +/- 31.2% (mean +/- standard deviation) of the baseline SEP amplitude, whereas the NITROX group recovered 63.2 +/- 28.2%, and the HBO group recovered 66.0 +/- 19.3%. An analysis of variance with repeated measures revealed that both treatment profiles promote significant (P = 0.03) recovery of the SEP amplitude compared to no treatment. We find no additional benefit, however, by initiating treatment at 6 atm abs, even when additional oxygen is provided.

Effect of lidocaine on somatosensory evoked response and cerebral blood flow after canine cerebral air embolism

BACKGROUND AND PURPOSE

Victims of air embolism often recover rapidly on hyperbaric treatment then deteriorate again, even if hyperbaric treatment is continued. In previous animal experiments, lidocaine has been shown to improve recovery of somatosensory evoked response amplitude after air embolism. However, animals in these experiments rarely deteriorated. We have shown that the induction of air embolism and transient hypertension in canines produces deterioration despite hyperbaric treatment, and we decided to test the effect of lidocaine on somatosensory evoked potential recovery and cerebral blood flow in this model.

METHODS

Dogs were treated with repeated doses of lidocaine or equivalent volumes of saline during hyperbaric therapy after internal carotid air embolism and transient hypertension. The investigators were unaware of treatment group assignment during the experiments. The amplitude of the median nerve somatosensory evoked potential and cerebral blood flow measured with carbon-14-labeled iodoantipyrine autoradiography were used to assess effect of therapy.

RESULTS

Lidocaine-treated dogs recovered 60 +/- 10% (mean +/- 95% confidence limits) of the baseline somatosensory evoked potential amplitude 220 minutes after air embolism; saline-treated dogs recovered 32 +/- 10% (a significant difference at p less than 0.01). Lidocaine-treated dogs also had higher cerebral blood flow values than saline-treated dogs 220 minutes after air embolism.

CONCLUSIONS

Lidocaine ameliorated the delayed deterioration of evoked potential associated with air embolism and hypertension in this canine model. The improved cerebral blood flow may be a mechanism of action of lidocaine or an associated effect of improved neuronal survival.

Management of herniated intervertebral disks during saturation dives: a case report

During research saturation dives at 5.0 and 5.5 atm abs, 2 divers developed an acute herniation of the nucleus pulposus of the L5-S1 intervertebral disk. In both cases the pain was severe enough to require intravenous morphine or intramuscular meperidine. Although the symptoms presented by these divers are frequently considered to be an indication for immediate surgical consultation, we decided that emergency decompression posed an unacceptable risk that decompression sickness (DCS) would develop in the region of acute inflammation. In both cases strict bedrest and medical therapy were performed at depth. In the first case, 12 h was spent at depth before initiating a standard U.S. Navy saturation decompression schedule with the chamber partial pressure of oxygen elevated to 0.50 atm abs. In the second case, a conservative He-N2-O2 trimix decompression schedule was followed to the surface. In both cases, no initial upward excursion was performed. The required decompression time was 57 h 24 min from 5.5 atm abs and 55 h 38 min from 5.0 atm abs. During the course of decompression, the first diver's neurologic exam improved and he required decreasing amounts of intravenous narcotic; we considered both to be evidence against DCS. The second diver continued to have pain and muscle spasm throughout decompression, however he did not develop motor, reflex, or sphincter abnormalities. Both divers have responded well to nonsurgical therapy.

Comparison of two recompression profiles in treating experimental cerebral air embolism

The standard treatment for cerebral arterial gas embolism (CAGE) is an initial recompression to 6 atm abs on air for 30 min followed by oxygen breathing at 2.8 and 1.9 atm abs. It has been suggested that initial recompression to 2.8 atm abs on O2 may be as beneficial, thus avoiding potential treatment complications associated with the deeper depth. To test this hypothesis, we measured the recovery of the somatosensory evoked potential (SEP) following air embolism in anesthetized, ventilated cats. Air was infused into the carotid artery in increments of 0.08 ml until the SEP amplitude was reduced to less than 10% of the baseline value for 15 min. Three groups were studied. A control group (n = 10) received no further treatment after SEP suppression. The second group (6 atm abs/HBO] (n = 8) was compressed to 6 atm abs on air for 30 min followed by O2 breathing at 2.8 atm abs for 100 min. The third group (HBO) (n = 8) was compressed to 2.8 atm abs on O2 for 130 min. The control group recovered 28.8 +/- 18.2% (mean +/- SD) of the baseline amplitude, whereas the 6 atm abs/HBO group recovered 48.6 +/- 22.6%, and the HBO group recovered 62.0 +/- 20.3%. An analysis of variance revealed that only the HBO group had significantly (P less than 0.01) better recovery than the control group. There was no significant difference in SEP recovery between the 2 treatment groups. These results suggest that treating CAGE at 2.8 atm abs with O2 is a viable alternative to the current therapy.

Blood flow and vascular permeability during motor dysfunction in a rabbit model of spinal cord ischemia

BACKGROUND AND PURPOSE

Delayed deterioration of neurological function after central nervous system ischemia is a well-documented clinical problem. The purpose of our study was to elucidate the role of spinal cord blood flow and spinal cord-blood barrier integrity in the evolution of delayed neurological deterioration after transient spinal cord ischemia in rabbits.

METHODS

Anesthetized rabbits were subjected to lumbar spinal cord ischemia (25 minutes) and variable periods of reperfusion (30 minutes to 48 hours after ischemia). Regional spinal cord blood flow was monitored by carbon-14-labeled iodoantipyrine autoradiography; vascular permeability was assessed by quantitative microhistofluorescence of Evans blue-albumin in frozen sections of spinal cord. Hindlimb motor function was assessed by standard scoring system and tissue edema by wet/dry weight method.

RESULTS

Hindlimb motor function indicated complete paralysis during ischemia and partial gradual recovery upon reperfusion (up to 8 hours), followed by progressive deterioration to severe deficits over 48 hours. Severe vascular permeability disruption was noticed early (30 minutes) after reperfusion, but almost complete recovery reestablished at 8 hours was followed by a secondary progressive increase in vascular permeability. Blood flow was reduced by 20-30% (p less than 0.01) 4 hours after ischemia in the gray matter, but hyperemia (200-300%, p less than 0.01) was observed 12-24 hours after ischemia. Spinal cord water content increased by 5.7% (p less than 0.05) 24 hours after ischemia.

CONCLUSIONS

This study demonstrates that delayed neurological and motor deterioration after spinal cord ischemia is associated with severe progressive breakdown of spinal cord-blood barrier integrity that develops late (hours) after the injury. Our data suggest that no ischemic insult in early or late reperfusion is associated with delayed motor deterioration.

Effects of treatment with dexamethasone on recovery from experimental cerebral arterial gas embolism

Dexamethasone is often recommended as an adjunct to recompression in the treatment of serious central nervous system decompression accidents. We studied the effects of prophylactic and therapeutic administration of dexamethasone combined with hyperbaric treatment in anesthetized dogs that were subjected to carotid air embolism and a brief episode of arterial hypertension. To assess recovery we measured somatosensory evoked potential (SSEP) amplitude, intracranial pressure, brain water, and cerebral blood flow. Three groups were studied: pre-air treatment (dexamethasone 1 mg/kg 3-4 h before carotid air embolism, and 1 mg/kg immediately after air embolism); post-air treatment (2 mg/kg immediately after air embolism); and control (equivalent volumes of saline pre- and post-air). There was a slight improvement in SSEP early in the course of hyperbaric therapy in the pre-air treated group; the post-air group never differed from control. No differences in intracranial pressure or brain water were found among groups. No blood flows below those lethal to neurons occurred in treated animals but 4 of 7 control animals had low flows. Although prophylactic treatment with dexamethasone produces some improvement in recovery, we cannot confirm that dexamethasone is an effective adjunct to recompression when administered therapeutically.

Morphologic and electrophysiologic effects of cochlear implantation and electrical stimulation

The nondeafened guinea pig model was utilized in this study to assess the functional and morphologic effects of cochlear implantation and electrical stimulation. Auditory brainstem responses (ABRs) were recorded prior to and following intrascalar implantation of a 3M-House cochlear electrode (n = 41 ears), as well as after electrical stimulation (n = 23 ears). The experimental population was divided into the following groups according to implantation and stimulation parameters: 200 microA for 3 hours (group I); 200 microA for 24 hours (group II); 400 microA for 3 hours (group III); implanted, but not stimulated (group IV); and nonimplanted, not stimulated ears (group V). Of those cochleae that sustained the trauma of implantation, 32 percent had no detectable ABR to 110 dB SPL clicks, while only 7 percent additionally failed to respond to 130 dB SPL clicks. No significant difference (one-way ANOVA with repeated measures at the 95 percent confidence limit) could be detected when comparing those ears that retained ABRs according to experimental grouping. Morphologic analysis was performed on 29 cochleae. Spiral ganglion "packing densities" were not found to be significantly different among the groups (ANOVA). The status of the organ of Corti was significantly better in groups II and V in comparison to the other groups (Kruskal-Wallis test with pairwise comparisons, p less than 0.05); there was no discernible dose-response relationship. Morphologic and electrophysiologic changes correlated with insertion trauma and infection rather than with electrical stimulation at the levels tested in this study.(ABSTRACT TRUNCATED AT 250 WORDS)

Allopurinol pretreatment improves evoked response recovery following global cerebral ischemia in dogs

The reperfusion of previously ischemic tissue may lead to the formation of highly reactive free radicals that promote tissue injury. Xanthine oxidase has been implicated as one source of these free radicals. We examined the role of xanthine oxidase in brain injury using a cerebrospinal fluid compression model of global cerebral ischemia with 15 minutes of ischemia and 4 hours of reperfusion. Seven dogs were pretreated with the xanthine oxidase inhibitor allopurinol (50 mg/kg for 5 days). Neurophysiological recovery was monitored with cortical somatosensory evoked potentials. As an attempt to correlate brain recovery with the mechanism of protection, free brain malondialdehyde was measured at the end of reperfusion by high-performance liquid chromatography. Brain water content was measured by wet-dry weights. Compared with seven untreated control dogs, allopurinol pretreatment significantly improved recovery of somatosensory evoked potentials after 4 hours of reperfusion. However, the amount of free malondialdehyde in the allopurinol-treated dogs was 32% greater than that in the controls. Brain water content was similar in the two groups. These results suggest that xanthine oxidase contributes to brain injury after ischemia and reperfusion. However, tissue damage caused by xanthine oxidase may be mediated through mechanisms other than free radical production.

Lipopolysaccharide-induced production of tumor necrosis factor activity in rats with and without risk factors for stroke

Rats produced more TNF activity in cerebrospinal fluid (CSF) than in blood after intracerebroventricular (i.c.v.) injection of lipopolysaccharide (LPS). After intravenous (i.v.) LPS, blood TNF levels exceeded CSF levels. Thus, brain cells appear to produce TNF in response to LPS. Rats with the stroke-risk factors hypertension or combined hypertension and genetic stroke-proneness produce more TNF in response to a provocative dose of LPS i.v. than control animals free of these risk factors. The possible relevance to stroke vulnerability is discussed.

Cerebral air embolism treated by pressure and hyperbaric oxygen

We used pressure and hyperbaric oxygen to treat 2 patients with cerebral air embolism, occurring as the result of invasive medical procedures, and neither suffered any permanent damage detectable by clinical examination and MRI. This outcome contrasts with reports of infarct and disability among untreated victims of air embolism.

No conversion of xanthine dehydrogenase to oxidase in canine cerebral ischemia

Xanthine oxidase (XO) has been implicated as a source of free radicals mediating ischemia-reperfusion injury. Conversion of the non-free radical generating xanthine dehydrogenase (XD) to the free radical producing XO during ischemia has been demonstrated in several tissues. We examined the irreversible conversion of XD to XO in the dog brain after ischemia and after ischemia and reperfusion. Under pentobarbital sodium anesthesia and by use of a cerebrospinal fluid compression model of global cerebral ischemia, dogs were subjected to 30 min of ischemia (n = 8) or 30 min of ischemia and 60 min of reperfusion (n = 8). A cerebral perfusion pressure of 60 mmHg was maintained during reperfusion. Eight control dogs were not subjected to ischemia. After the dogs were killed their brains were rapidly removed and frozen in liquid nitrogen. XO and XD + XO activities were measured with a radioassay utilizing 8-[14C]hypoxanthine and separating substrate and products by thin-layer chromatography. Total XD + XO activity was significantly (P less than 0.05) decreased after ischemia and reperfusion (35.6 +/- 8.0 vs. 60.8 +/- 20.8 nmol.min-1.g protein-1 in controls, means +/- SD) but not after ischemia alone (48.2 +/- 20.4). XO/(XD + XO) was approximately 20% in all three groups. Irreversible XD to XO conversion is not an important mechanism leading to early tissue injury in global cerebral ischemia.

Background review and current concepts of reperfusion injury

We define the concept of reperfusion injury, and we present a background chronology of experimental work supporting and questioning this concept. We identify several new influences, such as current clinical interest in thrombolytic therapy for acute ischemia of heart and brain and the growing recognition of endothelium as a regulator of homeostasis. We propose that these influences will encourage a reexamination of reperfusion injury as a factor in the ultimate outcome of tissue exposed to reversible ischemia. We briefly discuss the major mechanisms presently implicated in reperfusion injury--loss of calcium homeostasis, free radical generation, leukocyte-mediated injury, and acute hypercholesterolemia.

Treatment of experimental cerebral air embolism with lidocaine and hyperbaric oxygen

Experiments were performed to assess the combined therapeutic effects of hyperbaric oxygen (HBO) and i.v. lidocaine on neural function after ischemia induced by cerebral air embolism in anesthetized cats. Neural function was determined by measuring the somatosensory evoked potential (SEP) amplitude. Air was infused into the carotid artery in increments of 0.08 ml to maintain the SEP amplitude at 10% or less of baseline values for 15 min. Three groups were studied. A control group (n = 9) received no further treatment after SEP suppression. An HBO group (n = 8) was treated with oxygen at 2.8 atm abs for 130 min. A third group (n = 8) received an i.v. lidocaine infusion in addition to HBO. Air infusion suppressed the SEP amplitude to the same level in all groups. The control group recovered 27.4 +/- 5.5% (mean +/- SEM) of the baseline SEP amplitude, whereas the HBO group recovered 62.0% +/- 7.2%, and the HBO plus lidocaine group recovered 75.3 +/- 5.7%. The results show that both HBO and the combination of HBO and lidocaine promote a significant recovery of the SEP amplitude compared to no treatment. However, lidocaine therapy adds no benefit to HBO therapy alone.

Bubble-induced dysfunction in acute spinal cord decompression sickness

Five anesthetized dogs undertook a chamber dive, on air, to 300 feet of seawater for 15 min. After the dive, spinal cord decompression sickness was detected by recording a reduced amplitude of the somatosensory evoked potential compared with predive base-line values. After the diagnosis of decompression sickness and rapid perfusion fixation of the animal, the spinal cord was removed and examined histologically. Numerous space-occupying lesions (SOL) that disrupted the tissue architecture were found in each cord, mainly in the white matter. The size and distribution of the SOL were determined using computerized morphometry. Although SOL occupied less than 0.5% of the white matter volume, we tested a number of algorithms to assess whether the SOL may have been directly involved in the loss of spinal cord function that followed the dive. We determined that the loss of somatosensory evoked potential amplitude may be attributed to the SOL if 30-100% of the spinal cord fibers that they displaced were rendered nonconducting. A number of possible mechanisms by which SOL may interfere with spinal nerve conduction are discussed.

Influence of cerebral ischemia and post-ischemic reperfusion on mitochondrial oxidative phosphorylation

Unilateral ischemia in the right cerebral hemisphere of the rat was induced by ligation of the right common carotid artery coupled with controlled hemorrhage to produce hypotension (25 +/- 8 mm/Hg). Where indicated after 30 min of ischemia, the withdrawn blood was reinfused to restore arterial pressure to normal. Mitochondria isolated from the ipsilateral hemisphere after 30 min of ischemia showed significantly lower respiratory rates than the organelles isolated from the contralateral side. Oxidation of NAD(+)-linked substrates was more sensitive to inhibition in ischemia (30%) than was of ferrocytochrome c (12%), succinate oxidation being intermediate. The activities of membrane-bound dehydrogenases (both NADH and succinate-linked) were also significantly lowered. Ischemia did not affect the cytochrome content of mitochondria. Respiratory activity (NAD(+)-linked) of mitochondria isolated from the ipsilateral hemisphere was twice as sensitive to inhibition by fatty acid as was of preparations from the contralateral side. Mitochondria isolated from cerebral cortex after 90 min of post-ischemic reperfusion showed no significant improvement in the rate of substrate oxidation. Adenine nucleotide translocase activity and energy-dependent Ca2+ uptake, both of which decreased significantly in mitochondria isolated from the ischemic brain, showed little recovery, on reperfusion. These observations suggested the strong possibility that the deleterious effects of ischemia on mitochondrial respiratory function might be mediated by free fatty acids that are known to accumulate in large amounts in ischemic tissues. The pattern of inhibition of ATPase activity was consistent with this view.

Pharmacologic therapy for ischemic cerebrovascular disease

This article considers pharmacologic therapy for the prevention and treatment of ischemic cerebrovascular disease. The emphasis is on prevention of stroke, since current pharmacologic treatments do not reliably reduce morbidity or mortality once a stroke has occurred. Antihypertensive treatment, anticoagulants, and antiplatelet agents are discussed along with general guidelines for poststroke therapy.

Alterations in brain monoamine neurotransmitter release at high pressure

High pressure exposure produces neurological changes which manifest as tremors, EEG changes and convulsions. Since previous studies have implicated the involvement of the monoaminergic system in these symptoms, it was of interest to study monoamine release at high pressure. Synaptosomes isolated from guinea pig brain were used to follow monoamine efflux at 68 ATA. The major observation was a decrease in the initial calcium dependent release of all three monoamines in response to K+ induced depolarization. This response is similar to that previously observed for GABA, glycine and glutamate. This generalized pressure induced depression of initial transmitter release suggests a mechanism common to the release process for both excitatory and inhibitory neurotransmission.

Arterial gas embolism as a pathophysiologic mechanism for spinal cord decompression sickness

A continuous infusion of air (1.0 ml.min-1) was delivered via a fine aortic cannula into the arterial circulation of 7 anesthetized dogs until no spinal cord function could be elicited by somatosensory evoked potentials. The animals were then rapidly perfusion-fixed and the spinal cords removed for histological examination. The appearance of the embolized cords differed substantially from eight spinal cords injured by fulminant decompression sickness (DCS). The embolized cords appeared essentially normal whereas the DCS cords featured extravascular, nonstaining, space-occupying lesions (SOLs) scattered throughout the cord, mainly in the white matter. Two spinal cords injured by DCS with a delayed onset (30 min from surfacing) appeared similar to the embolized cords. These findings are compatible with the hypothesis that two mechanisms are involved in the onset of spinal cord DCS. Fulminant disease is associated with SOLs, which are probably caused by the in situ evolution of a gas phase. Disease with a delayed onset is more likely to be caused by an ischemic mechanism, which in the acute phase is histologically indistinguishable from gas embolism.

Pressure-dependent changes in the release of GABA by cerebrocortical synaptosomes

Previous studies describe a depression in potassium-evoked, calcium-dependent transmitter release from guinea pig cerebrocortical synaptosomes in response to compression to 68 ATA with heliox. The study described in this paper was designed to evaluate whether a similar effect could be detected at lower pressures and to determine whether the effect had an abrupt onset with a fixed response or a progressive response varying over a range of pressures. A typical response to potassium-evoked depolarization was observed at all pressures studied. This consisted of an initial rising phase lasting 2 to 3 min followed by a falling phase. There was a pressure-dependent depression in the absolute amount of transmitter released as well as a depression in the rate of release in the first minute following stimulation. The mean depression in gamma-aminobutyric acid release during the first minute was an average of 15% at 19 ATA, 28% at 37 ATA, 38% at 50 ATA, and 54% at 62 ATA when compared to a 1 ATA control.

Methyl prednisolone in the treatment of acute spinal cord decompression sickness

Sixteen anesthetized dogs undertook a chamber dive that was designed to induce decompression sickness. Somatosensory evoked potentials (SEP) were used to diagnose and quantify the outcome of spinal cord involvement in the disease. Following diagnosis, 8 animals were treated with methyl prednisolone (MP), 20 mg.kg-1 ("megadose"), as an adjuvant to recompression on an abbreviated U.S. Navy Treatment Table 6. Eight control animals were recompressed in a similar manner, but received the MP diluent only as an adjuvant. Analysis of the SEP at the conclusion of treatment showed that there was no significant difference in outcome for the 2 groups of animals. However, if all the SEP recorded during the treatment period are compared, the MP-treated animals experienced a significantly worse outcome than the diluent-treated controls. The risks and benefit of using corticosteroids in the treatment of human spinal cord DCS are discussed.

Influence of granulocytopenia on canine cerebral ischemia induced by air embolism

We subjected nine dogs with severe granulocytopenia 4 days after the administration of mechlorethamine to 1 hour of cerebral ischemia induced by the controlled, incremental injection of air into the internal carotid artery. Cortical somatosensory evoked responses and cerebral blood flow determined by [14C]iodoantipyrine autoradiography were compared with those of six control dogs that had received mechlorethamine 1 day previously and were not yet granulocytopenic. Eleven additional control dogs received no mechlorethamine but had identical ischemic insults and were followed for 4 hours after ischemia. Both control groups had identical evoked response outcomes after 1 hour of recovery from ischemia. Granulocytopenic dogs had improved evoked response recoveries compared with either control group after 1 hour of recovery. No areas of very low blood flow were observed 1 hour after ischemia in the granulocytopenic dogs, but three of five dogs in the control group receiving mechlorethamine had such areas.

Experimental determination of latency, severity, and outcome in CNS decompression sickness

Twenty-eight dogs underwent a 300 fsw chamber dive designed to generate spinal cord decompression sickness (DCS), which was detected by observing a reduction in the amplitude of the spinal somatosensory evoked potential (SEP). After an interval of 15 min on the surface following diagnosis, the animals received a therapeutic recompression. The latency was defined as the time between surfacing from the dive and the diagnosis of DCS, the severity as the minimum SEP amplitude, and the outcome as the amplitude of the SEP after 2 h of treatment. Significant correlations between latency and severity (P less than 0.05), latency and outcome (P less than 0.01), and severity and outcome (P less than 0.05) were found. Canine spinal cord latency is shown to be very similar to that found in man up to a surface interval of 30 min. The association between latency, severity, and outcome of spinal cord DCS is discussed with reference to the possible mechanisms involved in this disease.

Central nervous system decompression sickness: latency of 1070 human cases

Many aspects of central nervous system (CNS) decompression sickness (DCS) are poorly understood, including the temporal pattern of its presentation and the pathogenic mechanisms involved in the development of the disease. Using case histories and clinical series published in the literature and retrieved from treatment center records, this study is an attempt to define the interval between surfacing from a hyperbaric exposure and the onset of symptoms of CNS DCS. The results of 1070 cases of human CNS DCS were included in the study. The results show that the disease generally occurs rapidly: over 50% became symptomatic within 10 min of returning to 1 ATA, and in only 15% of cases was the onset of symptoms delayed for more than 1 h. Cerebral DCS had a more rapid onset than spinal cord disease: 50% of cerebral cases became apparent within about 3 min and a similar proportion of spinal cord cases within about 9 min from surfacing. The influence of these results on the diagnosis and treatment of dysbaric illness, on the safety of certain diving practices, and on possible pathogenic mechanisms is discussed.

Is there a role for the autochthonous bubble in the pathogenesis of spinal cord decompression sickness?

Histological examination by light and electron microscopy of the spinal cords of four dogs rapidly perfusion-fixed after the onset of decompression sickness revealed the presence of numerous non-staining, space-occupying lesions that were absent in similarly prepared sections of control or ischemic spinal cords. We propose the hypothesis that these lesions are caused by the liberation of a gas phase. The possible significance of these lesions in the evolution of spinal cord dysfunction is discussed with reference to the principal theories of the pathogenesis of spinal cord decompression sickness.

Stroke risk factors prepare rat brainstem tissues for modified local Shwartzman reaction

Stroke risk factors such as hypertension, diabetes, advanced age, and genetic predisposition to stroke were demonstrated to prepare rat brainstem tissues for a modified local Shwartzman reaction. A single intracisternal injection of endotoxin provoked the reaction, and affected rats manifested neurologic deficits accompanied by pathologic lesions. Brainstem infarcts developed in only a small proportion of rats without recognized risk factors after intracisternal injection of endotoxin. Thus, stroke risk factors, which are ordinarily regarded as operating through acceleration of atherosclerosis, may predispose to brain ischemia by local effects on brain microcirculation such as those thought to underlie preparation of a tissue for the local Shwartzman reaction.

Effects of prostacyclin, indomethacin, and heparin on cerebral blood flow and platelet adhesion after multifocal ischemia of canine brain

Seven anesthetized dogs treated with prostaglandin I2, indomethacin, and heparin were compared with 12 controls to test the hypothesis that the salutary effect of treatment on recovery of neuronal function and cerebral blood flow (CBF) after ischemia is coupled to the inhibition of platelet accumulation. In this model of right hemisphere multifocal ischemia, cortical somatosensory evoked response (CSER) amplitude, 14C autoradiographic blood flow, and 111In-labeled platelet accumulation were measured. The ratio of injured to noninjured hemispheric 111In activity (cpm/g) provided an index of platelet accumulation. Treatment improved CBF of the injured hemisphere compared with control after 4 hours of reperfusion (74 +/- 17 versus 53 +/- 13 ml/100 g/min, p less than 0.05), and it enhanced recovery of CSER amplitude (percent of baseline) after 1 hour of reperfusion compared with control (27.1 +/- 4.7% [treatment] versus 15.5 +/- 2.8% [control], p less than 0.05). However, the effect on CSER was not sustained after 4 hours of recovery. Despite these effects on CSER and CBF, treatment failed to inhibit 111In-labeled platelet accumulation in the injured hemisphere (1.7 +/- 0.3% [treatment] versus 1.5 +/- 0.1% [control], p greater than 0.05). Platelets may adhere to damaged endothelium despite aggressive platelet antiaggregant therapy.

Pressure suppresses serotonin release by guinea pig striatal synaptosomes

Exposure to high pressure produces neurologic changes in humans which manifest as tremor, EEG changes, and convulsions. Since previous studies have implicated the involvement of the serotoninergic system in these symptoms, it was of interest to study serotonin release at high pressure. Synaptosomes isolated from guinea pig striatum were used to follow serotonin efflux at 68 ATA. The major observation was a decrease in [3H]serotonin release from depolarized striatal synaptosomes at 68 ATA. In view of the role of serotonin as an inhibitory neurotransmitter in this area, the observed decrease in synaptic release leads us to conclude that decreased serotoninergic activity in striatal neurons probably is contributing to the hyperexcitability associated with HPNS.

Air embolism may cause unrecognized ischemia of the gray-white junction

The border between the gray and white matter is defined by an abrupt change in average blood flow. This difference allows one to distinguish structure with [14C]iodoantipyrine autoradiography. The angioarchitecture of the cortical gray-white junction suggests that an air embolism might preferentially lodge in this border zone, and thus ischemia of the border might go unrecognized if one depended only on the difference in average blood flow to define the gray-white junction. Accordingly, a computerized image processing technique was applied to compare the area of the cortex measured on an autoradiogram to the area measured on a histologic section after staining for myelin. In dogs that had received air embolism, the autoradiogram underestimated the thickness of the cortical mantle even in sections that did not seem to have an obvious focal zone of low blood flow. This suggests that the deep cortical layers are especially vulnerable to air embolism.

Release of dopamine from striatal synaptosomes: high pressure effects

Adverse neurological manifestations of exposure to high hydrostatic pressure include tremor and convulsions, suggesting an alteration in synaptic transmission, particularly with inhibitory pathways. Because striatal transmission has been implicated in the high pressure neurologic syndrome (HPNS), we investigated the effect of pressure exposure on the release of a major inhibitory neurotransmitter in this region. Synaptosomes isolated from the guinea pig striatum were used to study the effect of compression to 67.7 ATA on [3H]dopamine release. Pressure was found to have a suppressive effect on the initial release of [3H]dopamine by synaptosomes isolated from the striatum of guinea pigs. This finding suggests that decreased inhibitory regulation at the level of the striatum contributes to the hyperexcitability associated with compression to high pressure.

Platelet activating factor receptor blockade enhances recovery after multifocal brain ischemia

We treated four anesthetized dogs (Canis familiaris) with the platelet activating factor (PAF) receptor antagonist kadsurenone prior to 60 min of multifocal ischemia induced by air embolism, and measured neuronal recovery, blood flow and autologous 111In-labeled platelet accumulation for 4 h after ischemia. Four anesthetized animals with identical ischemia served as controls. Kadsurenone (3 mg/kg) administered 5 min prior to ischemia and continuously (1 mg/kg/hr) throughout ischemia and recovery significantly enhanced recovery of cortical somatosensory evoked response (CSER) amplitude (% of baseline) when compared to controls (27-36% vs 9-14%, p less than 0.05). We estimated platelet accumulation as 111In activity (cmp/g tissue) in the injured hemisphere minus that in the non-injured hemisphere. Kadsurenone treated animals did not exhibit significantly altered 111In-labeled platelet accumulation when compared to controls (6158 +/- 2386 vs 9979 +/- 3852, mean +/- SEM). Beneficial effects of PAF receptor blockade other than those on platelet accumulation may be involved.

Effect of pressure on the release of radioactive glycine and gamma-aminobutyric acid from spinal cord synaptosomes

Exposure to high hydrostatic pressure produces neurological changes referred to as the high-pressure nervous syndrome (HPNS). Manifestations of HPNS include tremor, EEG changes, and convulsions. These symptoms suggest an alteration in synaptic transmission, particularly with inhibitory neural pathways. Because spinal cord transmission has been implicated in HPNS, this study investigated inhibitory neurotransmitter function in the cord at high pressure. Guinea pig spinal cord synaptosome preparations were used to study the effect of compression to 67.7 atmospheres absolute on [3H]glycine and [3H]gamma-aminobutyric acid ([3H]GABA) release. Pressure was found to exert a significant suppressive effect on the depolarization-induced calcium-dependent release of glycine and GABA by these spinal cord presynaptic nerve terminals. This study suggests that decreased tonic inhibitory regulation at the level of the spinal cord contributes to the hyperexcitability observed in animals with compression to high pressure.

Indomethacin, prostacyclin, and heparin improve postischemic cerebral blood flow without affecting early postischemic granulocyte accumulation

Six anesthetized dogs treated with indomethacin, prostacyclin (PGI2), and heparin were compared with 7 anesthetized controls (ischemia without treatment) to determine whether cyclooxygenase inhibition would lead to enhanced granulocyte accumulation because of preferential formation of lipoxygenase products. Cortical somatosensory evoked response, [14C]iodoantipyrine autoradiographic blood flow, and 111In-labelled granulocyte accumulation were compared 4 hours after a 60-minute exposure to multifocal brain ischemia. Treatment with indomethacin, PGI2, and heparin eliminated neuron-disabling brain blood flows without altering early postischemic granulocyte accumulation. Granulocyte accumulation after 4 hours of reperfusion was not significantly different in control and treated dogs. The final amplitude of the cortical somatosensory evoked response in the treated group averaged 38.0 +/- 13.6% (mean +/- SEM) of the corresponding baseline value compared with 21.0 +/- 4.6% in the control group, but this difference was not significant.

A brief episode of severe arterial hypertension induces delayed deterioration of brain function and worsens blood flow after transient multifocal cerebral ischemia

Transient arterial hypertension occurs sporadically following cerebral air embolism and may occur during the acute phase of stroke. This study used an animal model of multifocal cerebral ischemia induced by air embolism and reversed by recompression to assess the effect of induced hypertension on the evoked response recovery, local cerebral blood flow, intracranial pressure, and brain water in 19 anesthetized dogs (Canis familiaris). Six received 0.4 ml of air via the internal carotid artery, 8 received intracarotid air and 10 micrograms/kg norepinephrine to produce transient hypertension, and 5 received intracarotid saline and norepinephrine. The average evoked response recovery in the air-only group was 58.3 +/- 7.7% (mean +/- SEM) of control after 4 hours of recompression; the air plus hypertension group recovery was 15.4 +/- 2.7% (p less than 0.01). The final evoked response in the dogs receiving hypertension alone did not differ from control values. Seven of 8 dogs in the air plus hypertension group had very low blood flows; only 1 of 4 in the air-only group had very low flows. The amount of brain water and the intracranial pressure were not detectably different at the end of treatment among all 3 groups. These results support a role for endothelial damage produced by air and hypertension in potentiating the process of postischemic hypoperfusion.

Effects of high pressure on the release of excitatory amino acids by brain synaptosomes

Excitatory amino acid antagonists have been shown to protect against the hyperexcitability associated with exposure to high pressure. This suggests that these excitatory neurotransmitter substances may play a role in the development of the symptoms of high pressure nervous syndrome (HPNS). Using a superfusion technique, we investigated the effect of exposure to 67.7 ATA of pressure on the release of aspartate and glutamic acid by isolated presynaptic nerve terminals from the guinea pig cerebral cortex. Pressure exposure was found to significantly increase the depolarization-induced release of aspartate by these synaptosomes. On the other hand, compression to 67.7 ATA had no effect on glutamic acid release. These findings suggest that increased aspartate release may be a contributing factor in the etiology of HPNS.

Pain-only decompression sickness affecting the orbicularis oculi

A case is reported of a diver who experienced the onset of pain in the left orbicularis oculi muscle approximately 20 min after surfacing from an experimental dive. A careful neurologic examination disclosed no abnormality. The response of the orbicularis oculi muscle to stimulation of the facial nerve and the blink reflex were both normal as recorded with a clinical neurophysiologic recording system (Nicolet CA 1000). The pain resolved quickly with recompression, supporting a diagnosis of pain-only decompression sickness in this small facial muscle.

Polymorphonuclear leukocyte accumulation in brain regions with low blood flow during the early postischemic period

In an anesthetized canine model in which ischemia was induced by incremental air embolism, 16 animals were exposed to 1 hr of ischemia and monitored for 10 min (n = 4), 60 min (n = 6), or 240 min (n = 6). Fourteen animals were observed for corresponding periods without being subjected to ischemia 70 min (n = 4), 120 min (n = 4), or 300 min (n = 6). Autologous granulocytes were labeled with 111In and reinfused just before ischemia. At the conclusion of each experiment, a 14C-iodoantipyrine autoradiographic blood flow study was performed. Granulocyte accumulation measured by gamma scintigraphy (cpm/gm) occurred in the injured hemisphere of ischemic animals at 60 min in anterior brain segments and at 240 min in anterior, middle, and posterior segments. By means of a double-label autoradiography technique, clustering of punctate granulocyte images was detected in regions of low flow or heterogeneous flow in half of the animals at both 60 min and 240 min postischemia. Granulocyte clustering did not occur in the autoradiograms of nonischemic animals. The results implicate granulocyte participation in the acute phase of ischemic brain injury and signal a convergence of hemostatic and inflammatory processes during the immediate postischemic period.

A review of the pathophysiology and potential application of experimental therapies for cerebral ischemia to the treatment of cerebral arterial gas embolism

This review considers the pathophysiology of air embolism that occurs as a result of diving and iatrogenic accidents. Various experimental therapies are classified according to their potential mechanism of action. The effectiveness and prospects for application to the clinical situation are assessed for each therapy. Hyperbaric oxygen remains the most effective therapy, and it is one that is simple to apply. The physiologic changes that occur after gas embolism are complex, however, and future improvements in therapy will demand close clinical supervision and multiple individually tailored therapeutic decisions rather than a simple protocol.

The amount of circumscribed brain edema and the degree of post-ischemic neuronal recovery do not correlate well

Fifty-four dogs were exposed to multi-focal ischemia sufficient to maintain suppression of the P1--N1 amplitude of the cortical sensory-evoked response (CSER) for 60 minutes. Subsequently, the P1--N1 amplitude recovery of the CSER was followed for an additional 15, 60, or 120 min while the dogs were treated or left untreated. The combination of PGI2, indomethacin, and heparin promoted a statistically significant augmentation of CSER amplitude return relative to: 1) no treatment; 2) PGI2 alone; 3) indomethacin alone; 4) PGI2 and heparin; 5) indomethacin and heparin; 6) PGI2 and indomethacin. Percentage gray matter water by the wet weight/dry weight technique was significantly elevated in all embolized groups compared to ten non-embolized controls, but the percentage recovery of the CSER did not correlate with the presence or degree of gray matter edema among embolized animals followed for 1 hour. Separation of embolized animals by the presence or absence of "neuron-disabling" flows (defined as 0-15 ml/100 gm/min for gray matter and 0-6 ml/100 gm/min for white matter) did produce significantly different mean CSER percentage recoveries but percentage gray matter water in the two groups was comparable. A proposed explanation of the data is that brain ischemia engenders two parallel processes which may become uncoupled. Ischemia creates metabolic conditions that lead to increased cellular imbibition of water and produces increased vascular leakiness. These perturbations increase brain water content. Concomitantly, there is an occurrence of further metabolic derangements and multifactorial interaction at the blood-endothelial interface which have a direct influence on neuronal function and recovery.

Prostaglandin I2, indomethacin, and heparin promote postischemic neuronal recovery in dogs

Forty-five conditioned male mongrel dogs were exposed to multifocal ischemia sufficient to maintain suppression for 60 minutes of the P1-N1 amplitude of the cortical sensory evoked response (CSER), a quantifiable index of neuronal function. Ischemia was induced and regulated by successive embolization of 20 to 50 microliters increments of air via the right internal carotid artery. Subsequently, the P1-N1 amplitude recovery of the CSER was followed for an additional 15, 60, or 120 minutes while the dogs were treated or left untreated. The combination of prostaglandin I2 (PGI2), indomethacin, and heparin promoted a statistically significant augmentation of return of CSER amplitude relative to no treatment, PGI2 alone, indomethacin alone, PGI2 and heparin, indomethacin and heparin, or PGI2 and indomethacin. After 60 minutes of recovery, animals receiving combined PGI2, indomethacin, and heparin achieved a 57% recovery of P1-N1 amplitude relative to baseline, while the corresponding recoveries in all other groups clustered around 20%. By 120 minutes of postischemic follow-up, the CSER recovery induced by PGI2, indomethacin, and heparin was 80% compared to 17% in untreated animals. By 15 minutes into the recovery period, the combination of the three agents had eliminated very low flows in the "neuron-disabling" range (defined as 0 to 15 ml/100 gm/min for gray matter and 0 to 6 ml/100 gm/min for white matter) in contrast to the relative inefficacy of no treatment or treatment with other than the triple combination of drugs. The study lends some support to a planned clinical trial of PGI2, indomethacin, and heparin in acute occlusive stroke in humans.