Kappa-opioid receptor changes and neurophysiological alterations during cerebral ischemia in rabbits

Dynorphins in Development and Disease: Implications for Cardiovascular Disease

It is well-established that cardiovascular disease continues to represent a growing health problem and significant effort has been made to elucidate the underlying mechanisms. In this review, we report on past and recent high impact publications in the field of intracrine network signaling, focusing specifically on opioids and their interrelation with key modulators of the cardiovascular system and the onset of related disease. We present an overview of studies outlining the scope of cardiovascular and cerebrovascular processes that are affected by opioids, including heart function, ischemia, reperfusion, and blood flow. Specific emphasis is placed on the importance of dynorphin molecules in cerebrovascular and cardiovascular regulation. Evidence suggests that excessive or insufficient dynorphin could make an important contribution to cardiovascular physiology, yet numerous paradoxical observations frequently impede a clear understanding of the role of dynorphin. Thus, we argue that dynorphin-mediated signaling events for which an immediate regulatory effect is disputed should not be dismissed as unimportant, as they may play a role in cross-talk with other signaling networks. Finally, we consider the most recent evidence on the role of dynorphin during cardiovascular-related inflammation and on the potential value of endogenous and exogenous inhibitors of kappa-opioid receptor, a major dynorphin A receptor, to limit or prevent cardiovascular disease and its related sequelae.

Effects of intracerebroventricular application of the delta opioid receptor agonist [D-Ala2, D-Leu5] enkephalin on neurological recovery following asphyxial cardiac arrest in rats

The delta opioid receptor (DOR) agonist [D-Ala2, D-Leu5] enkephalin (DADLE) has been implicated as a novel neuroprotective agent in the CNS. The current study was designed to evaluate the effects of intracerebroventricular (ICV) application of DADLE on neurological outcomes following asphyxial cardiac arrest (CA) in rats. Male Sprague-Dawley rats were randomly assigned to four groups: Sham group, CA group, DADLE group (DADLE+CA), and Naltrindole group (Naltrindole and DADLE+CA). All drugs were administered into the left cerebroventricle 30 min before CA. CA was induced by 8-min asphyxiation and the animals were resuscitated with a standardized method. DOR protein expression in the hippocampus was significantly increased in the CA group at 1 h after restoration of spontaneous circulation (ROSC) compared with the Sham group. As time progressed, expression of DOR proteins decreased gradually in the CA group. Treatment with DADLE alone or co-administration with Naltrindole reversed the down-regulation of DOR proteins in the hippocampus induced by CA at 24 h after ROSC. Compared with the CA group, the DADLE group had persistently better neurological functional recovery, as assessed by neurological deficit score (NDS) and Morris water maze trials. The number of surviving hippocampal CA1 neurons in the DADLE group was significantly higher than those in the CA group. However, administration of Naltrindole abolished most of the neuroprotective effects of DADLE. We conclude that ICV administration of DADLE 30 min before asphyxial CA has significant protective effects in attenuating hippocampal CA1 neuronal damage and neurological impairments, and that DADLE executes its effects mainly through DOR.

Randomized trial of Cervene, a kappa receptor-selective opioid antagonist, in acute ischemic stroke

The purpose of this randomized trial was to confirm drug safety and to obtain preliminary efficacy data on Cervene (nalmefene), an opioid antagonist with relative kappa receptor selectivity, for the treatment of acute ischemic stroke. Patients were treated for 24 hours with either intravenous Cervene (0.05 mg/kg as an initial infusion over 15 minutes and 0.01 mg/kg/h maintenance) or placebo within 6 hours of an ischemic stroke. Efficacy was assessed by comparing the change from baseline to day 7 in the National Institutes of Health stroke scale score (NIHSSS) and the Glasgow Outcome Scale and Barthel Index at 3 months. Forty-four evaluable patients were randomized (3:1) to Cervene (n = 34; treated at 5.0 +/- 0.9 hours after onset) and placebo (n = 10; treated at 4.6 +/- 1.5 hours). No deaths or serious adverse events reasonably attributable to Cervene have been reported. A "major improvement" (NHSSS > 4) was seen at day 7: placebo, 33% (three of nine patients) and Cervene, 66% (19 of 29 patients). Only patients with initial NIHSSS >/= 4 were considered evaluable for this primary endpoint. "Good recovery" at 3 months (Glasgow = 5) was as follows: placebo, 50% (5 of 10 patients) and Cervene, 73% (24 of 33 patients). The death rate at 3 months was placebo, 20% (2 of 10 patients) and Cervene, 9.1% (3 of 33 patients). One patient was lost to follow-up. In conclusion, results from this randomized trial suggest that Cervene is safe, tolerable, and may be beneficial in the treatment of acute stroke patients.

Opioids modulate post-ischemic progression in a rat model of stroke

Alterations in the opioidergic system have been found in cerebral ischemia. Neuroprotection studies have demonstrated the involvement of the opioidergic system in cerebral ischemia/reperfusion (I/R). However, the neuroprotective mechanisms remain largely unclear. This study was conducted to investigate whether intracerebroventricular administration of opioidergic agonists has a neuroprotective effect against cerebral ischemia in rats and, if this proved to be the case, to determine the potential neuroprotective mechanisms. Using a focal cerebral I/R rat model, we demonstrated that the opioidergic agents, BW373U86 (delta agonist) and Dynorphin A 1-13 (kappa agonist), but not TAPP (mu agonist), attenuated cerebral ischemic injury as manifested in the reduction of cerebral infarction and preservation of neurons. The antagonism assay showed that the neuroprotective effect of Dynorphin A was attenuated by nor-Binaltorphimine (kappa antagonist). Surprisingly, BW373U86-induced neuroprotection was not changed by Naltrindole (delta antagonist). These findings indicate that BW373U86 and Dynorphin A exerted distinct neuroprotection against ischemia via opioid-independent and -dependent mechanisms, respectively. The post-ischemic protection in beneficial treatments was accompanied by alleviations in brain edema, inflammatory cell infiltration, and pro-inflammatory cytokine interleukin 6 (IL-6) expression. In vitro cell study further demonstrated that the opioidergic agonists, delta and kappa, but not mu, attenuated IL-6 production from stimulated glial cells. Our findings indicate that opioidergic agents have a role in post-ischemic progression through both opioid-dependent and -independent mechanisms. In spite of the distinct-involved action mechanism, the potential neuroprotective effect of opioidergic compounds was associated with immune suppression. Taken together, these findings suggest a potential role for opioidergic agents in the therapeutic consideration of neuroinflammatory diseases. However, a better understanding of the mechanisms involved is necessary before this therapeutic potential can be realized.

Long-term alterations in mu, delta and kappa opioidergic receptors following middle cerebral artery occlusion in mice

Alterations in the opioidergic system may play a role in the molecular mechanisms underlying neurochemical responses to cerebral ischaemia. The present study aimed to determine the delayed expression of mu, delta and kappa opioid receptors, following 1, 2, 7, and 30 days of middle cerebral artery occlusion (MCAO) in mice. Using quantitative autoradiography, we highlighted significant decreases in mu, delta and kappa opioid receptor expression in ipsilateral cortices from day 1 post-MCAO. Moreover, in contralateral nucleus lateralis thalami pars posterior, ipsi- and contralateral nucleus medialis dorsalis thalami, and ipsilateral substantia nigra, pars reticulata (SNr), kappa receptors were increased; mu receptor densities were decreased in nucleus ventralis thalami, pars posterior (VThP), and SNr. delta-Binding sites were increased in the striatum on day 30 post-MCAO. The alterations in opioid receptors in cortical infarcts were correlated with strong histological damage. Further reductions in opioid receptor densities in cortical infarcts were observed at later time points. In subcortical brain regions, opioid receptor densities were also altered but no histological damage was seen, except in the VThP, in which cell density was increased on day 30. Delayed reductions in opioid receptor densities in the infarct appeared as the continuation of the early processes previously demonstrated. However, changes in subcortical opioid receptor expression may correlate with neuronal alterations in remote brain regions. Changes in opioidergic receptor expression in these regions may be involved in the long-term consequences of stroke and could be used as biomarker of neuronal alteration through the use of imaging techniques in the clinic.

Outcome prediction in acute monohemispheric stroke via magnetoencephalography

BACKGROUND

Following an ischemic stroke a highly variable clinical outcome is commonly evident despite similar onset symptoms as well as lesion characteristics. The aim of this study was to identify indexes providing early prediction of functional recovery, in addition to clinical severity and lesion dimension at onset of stroke.

METHODS

In 32 patients, magnetoencephalographic (MEG) parameters collected in the acute phase (<10 days from symptoms onset, T0) from affected (AH) and unaffected (UH) hemispheres at rest and evoked by sensory stimuli were evaluated in association with the clinical outcome in a stabilized phase (T1, median 7.8 months) classified with three levels: worsening, partial and full recovery.

RESULTS

Multiple multinomial logistic regression indicated AH gamma and UH delta band powers able to prognosticate clinical outcome at T1. After inclusion in this analysis, lesion volume had the strongest predictive ability, and UH delta band power remained as a predictive factor with a measurable cut-off, maximizing both sensitivity and specificity of the prediction: a patient with UH delta below cut-off would recover to some extent; a patient with UH delta above cut-off would have a probability of about 70% to worsen.

CONCLUSIONS

MEG UH delta and AH gamma band powers were found to provide useful information about long-term outcome prognosis. Only the increase of delta band activity in the unaffected hemisphere contains information about the outcome in addition to the lesion volume.

Delta dipole density and strength in acute monohemispheric stroke

Previous electroencephalographic and magnetoencephalographic (MEG) studies demonstrate that the activity in the delta band can increase not only in the perilesional area, but also in the contralesional hemisphere in patients affected by a monolateral stroke. The increase of delta activity in the unaffected hemisphere (UH) in the acute phase after an unilateral stroke seems to add prognostic information about clinical recovery. Delta activity in perirolandic regions was investigated via MEG in a group of 27 patients affected by stroke in the territory of middle cerebral artery in the first week following the symptom onset. Clinical evaluation was performed in the acute and the post-acute (median 9 months) phase. Delta band power, delta dipole density (DDD, the number of accepted single dipole fits per second in parieto-frontal region) and delta dipole strength (DDS, the average of the accepted dipoles strengths) were evaluated in both hemispheres, separated on the basis of the lesion level (cortical and subcortical involvement) and correlated with lesion volume, clinical status in acute phase and recovery level. Although in our patient cohort DDD did not differ in both hemispheres with respect to control values, DDS showed higher level than in controls both in affected hemisphere (AH) and UH, was dependent on the lesion level and positively correlated with the lesion volume. Moreover, while AH and UH DDSs were not associated with clinical status in acute phase, they correlated with clinical recovery in post-acute phase. These properties confirmed findings obtained by spectral power analysis and provided a localized delta activity amplitude estimate, independent of measuring system and allowing inter-laboratory standardization.

Quantified EEG analysis monitoring in a novel model of general anaesthesia in rats

The aim of this research was to evaluate the safety and reliability of an anaesthetic mixture (Equitensine: pentobarbital, chloral hydrate, dihydroxypropane, ethanol) which, unlike other 'classic' anaesthetics, such as ketamine [The Electroencephalogram in Anaesthesia, Springer, Berlin, 1984], has been demonstrated not to induce alterations in the extracellular concentrations of cerebral excitatory amino acids. Quantified EEG analysis monitoring and behavioural observation were used to quantify the degree and the time course of the changes in cerebral electrical activity, analgesia and sedation induced, in rats, by the compound under investigation. Equitensine (0.33 ml/100 g), administered intraperitoneally, induced analgesia (monitored by the tail flick method) for 60-70 min and a pattern of behavioural sedation (loss of the righting reflex) lasting, on average, 130-150 min. The EEG monitoring revealed a pattern typical of burst suppression which lasted 15-20 min, followed by another, lasting 270-300 min, characterized by slow waves of high amplitude. The quantified EEG analysis demonstrated that the changes in cerebral electrical activity lasted longer than behavioural observation suggested. The compound under examination was found to be safe, reliable and non-invasive to administer and sustain in all the animals, and quantified EEG analysis proved to be a very sensitive method for highlighting the functional changes in the central nervous system.

Cervene in acute ischemic stroke: Results of a double-blind, placebo-controlled, dose-comparison study

Cervene (nalmefene), an opioid antagonist with relative kappa receptor selectivity, has shown neuroprotective effects in multiple experimental central nervous system injury and ischemic models. The agent already has a well-established safety profile in various clinical indications. Results from an earlier pilot study in 44 acute stroke patients suggest that Cervene administered by 24-hour maintenance infusion was safe and tolerable. The primary and secondary objectives of the current study were to assess the dose-related safety and preliminary efficacy of Cervene in patients with acute ischemic stroke.

METHODS

The present investigation was a Phase II, placebo-controlled, double-blind, randomized, dose-comparison, parallel-group study of a 24-hour administration of Cervene injection. Patients with acute ischemic stroke, onset of symptoms within 6 hours, and baseline score > or =4 on the National Institute of Health Stroke Scale (NIHSS) were randomized to 1 of 4 treatment groups: Cervene 6 mg, 20 mg, 60 mg or placebo. The primary efficacy outcome was the proportion of patients achieving a score of > or =60 on the Barthel Index and a rating of "moderate disability" or better on the Glasgow Outcome Scale at 12 weeks.

RESULTS

A total of 312 patients were randomized at 28 centers. All doses of Cervene were well tolerated. Overall, there was no significant difference in 3-month functional outcome for any dose of Cervene treatment compared with placebo. However, a prospective secondary analysis showed that both male and female patients less than age 70 years may have had an improved 3-month outcome.

CONCLUSIONS

The results of this study indicate that the competitive kappa receptor opiate antagonist Cervene can be given safely to acute stroke patients at doses up to 60 mg/24 hr. Although overall there was no significant difference in the 3-month outcome, Cervene treatment may be associated with improved outcomes for patients younger than age 70.

The effect of fentanyl on electrophysiologic recovery of CA 1 pyramidal cells from anoxia in the rat hippocampal slice

Fentanyl is widely used in conditions in which the brain is at risk of ischemic or anoxic injury. We evaluated the effect of fentanyl on anoxic injury to CA 1 pyramidal cells in the rat hippocampus. These neurons are extremely sensitive to anoxic injury and are densely populated with opioid receptors. We prepared hippocampal slices from adult Sprague-Dawley rats and evoked a postsynaptic population spike in the CA 1 pyramidal cell region by stimulating the Schaffer collateral pathway. The amplitude of this response was used to evaluate the effect of fentanyl on anoxic injury. Pretreatment with fentanyl (50 or 500 ng/mL) did not alter the amplitude of the CA 1 population spike before anoxia, nor did it alter the recovery of this response after 5,6, or 7 min of anoxia. After 5 min of anoxia, the population spike recovered to 76% of its preanoxic level in the control group and to 87% in the group treated with 500 ng/mL of fentanyl. After 6 min of anoxia, recovery was 45% in the control group, 57% in the group treated with 50 ng/mL of fentanyl, and 58% in the group treated with 500 ng/mL of fentanyl. After 7 min of anoxia, recovery was 5% in the control group and 4% in the group treated with 50 ng/mL of fentanyl. We conclude that fentanyl does not affect the recovery of the electrophysiological response in rat hippocampal neurons subjected to an anoxic insult.

IMPLICATIONS

Because fentanyl is used in large doses during surgical procedures in which the brain is at increased risk of ischemic or anoxic injury, it is important to determine its effect on such injury. Using the rat hippocampal slice model, we found fentanyl to be neither neurotoxic nor protective against anoxic injury to neurons when used in concentrations comparable to those produced in clinical practice.

Maximal densities of mu, delta, and kappa receptors are differentially altered by focal cerebral ischaemia in the mouse

Though opioids are known to have neuroprotective properties, little information is available on the functional state of opioidergic receptors following focal cerebral ischaemia. The present study investigated the evolution of the Bmax and Kd for [3H]DAMGO, [3H]DADLE, and [3H]U69,593, respectively, for the mu, delta, and kappa opioidergic receptors after permanent focal cerebral ischaemia in mice. While the various Kd were unchanged, mu and delta Bmax values were precociously decreased in frontoparietal cortices, earlier than kappa receptors, reflecting infarct extension with time. The Bmax values for mu and delta receptors were also altered in non-infarcted tissues, such as tissues at risk (e.g., temporal auditory cortex) and exofocal (e.g., contralateral and non-infarcted) cortices. These results suggest that, in non-infarcted areas, the observed changes reflect functional modifications to focal ischaemia.

Enhanced brain opioid receptor activity precedes blood-brain barrier disruption

We studied the effects of transient postischemic increased opioid receptors (OPR) binding (mu, delta, kappa) on blood-brain barrier (BBB), brain water content and brain mitochondrial oxidative enzymes system. Cats were exposed to temporary middle cerebral artery occlusion (MCAO). The significant increased OPR bindings observed 10 min after the release of MCAO (ischemic rCBF = 7 +/- 1 to 11 +/- 2 ml/100 g/min) preceded the early and late BBB disruptions, brain edema and postischemic impaired mitochondrial oxidative enzymes functions. Further, the study suggests indirectly that the latter process was irreversible and hence associated with subsequent ischemic cerebral infarction. In addition, the results revealed a possible viable therapeutic window in the early postischemic recirculation period, before the onset of impaired mitochondrial oxidative function.

Decreased affinity of K-receptor binding during reperfusion following ischaemic preconditioning in the rat heart

The effects of ischaemic preconditioning with three cycles of ischaemia of 3 min and reperfusion of 5 min each cycle on ventricular fibrillation threshold (VFT) and ventricular fibrillation (VF), and binding properties of tritiated U69,593, a selective kappa opioid-receptor (k-receptor) agonist, during subsequent ischaemia and/or reperfusion were studied in the rat heart. It was found that ischaemic preconditioning significantly enhanced the VFT values during ischaemic and reperfusion. VF during the subsequent reperfusion period was also significantly reduced. The Kd of the [3H]U69,593 binding sites in the sarcolemma of the heart at 5 min of reperfusion was significantly increased following ischaemic preconditioning. The Bmax was, however, not altered after the preconditioning. The study provides evidence for the first time suggesting that the cardioprotective effects of ischaemic preconditioning may be related to a reduction in affinity of the K-receptor binding.

Neurochemical changes following occlusion of the middle cerebral artery in rats

We have developed a stroke model involving middle cerebral artery occlusion in the rat which elicits changes in cardiac and autonomic variables that are similar to those observed clinically. It is likely that these neurogenic autonomic responses are mediated by changes in neurotransmitter systems subsequent to the stroke. This possibility was investigated by examining changes in immunohistochemical staining for tyrosine hydroxylase, neuropeptide Y, leu-enkephalin, neurotoxins and dynorphin following middle cerebral artery occlusion in the rat. Computerized image analysis was used to provide semi-quantitative measurements of the changes. The ischemic region was centered primarily in the insular cortex. The results indicate that there are significant increases in immunostaining for tyrosine hydroxylase and neuropeptide Y in the insular cortex within the peri-infarct region. Neuropeptide Y staining was also significantly increased in the basolateral nucleus of the amygdala, ipsilateral to the middle cerebral artery occlusion, which did not appear to be included in the infarct. Leu-enkephalin, neurotensin and dynorphin staining was significantly elevated in the central nucleus of the amygdala ipsilateral to the occlusion of the middle cerebral artery. These neurochemical changes are discussed as possible mechanisms mediating the cardiac and autonomic consequences of stroke or as part of a process to provide neuro-protection following focal cerebral ischemia.

The effects of naloxone on retinal ischemia in rats

The efficacy of naloxone (NL), a broad spectrum opioid antagonist, on retinal ischemia, was evaluated in a rat model of retinal ischemia with histopathologic and morphometric criteria. Two intraperitoneal injections of naloxone 3 mg/kg given immediately and 6 hr after reperfusion showed beneficial effects to the retina as evaluated at 2, 7, and 14 days after reperfusion. Morphologically, the naloxone-treated group showed better-preserved ganglion cells, nerve fiber layer, and inner nuclear layer. Morphometrically, in the treated groups, inner retinal thickness at all three time points and ganglion cell counts at 7 days showed higher values than vehicle controls. This beneficial effect of naloxone was dose-dependent with a minimal effective total dose of 6 mg/kg. A possible role of opiate receptors in retinal ischemia is suggested.

The effects of selective opioid antagonists on somatosensory evoked potentials during relative cerebral ischemia in rats

Hemorrhagic hypotension in spontaneously hypertensive rats induces attenuation of somatosensory evoked potentials. In this model of relatively mild cerebral ischemia, our previous studies have shown that naloxone stereospecifically enhances the evoked potentials, without changes in cortical blood flow. The high dose of naloxone needed to enhance the evoked potentials suggests that the attenuation is mediated by low affinity opioid receptors (delta or kappa). In the present study, we used this model to study the effects of naloxone-methobromide (5 mg kg-1, a quaternary derivative of naloxone with selective peripheral action when injected intravenously), MR 2266 (1 mg kg-1, a kappa receptor antagonist), and naloxone (5 mg kg-1) as well as saline injection (as control) in four different groups of rats. Following injection, we examined the changes in somatosensory evoked potentials, cortical blood flow and heart rate for 15 min while mean arterial pressure was held constant by a pressure-regulating reservoir. Only naloxone changed the somatosensory evoked potential amplitude significantly compared with the saline group in which no effect was seen. However, there was a tendency for a delayed effect of naloxone-methobromide on the evoked potentials, possibly indicating that the substance slowly passes the blood-brain barrier. Naloxone and MR 2266 caused a transient decrease in heart rate, while following naloxone-methobromide injection there was a slight increase in heart rate. Our results thus indicate that the beneficial effects of naloxone on somatosensory evoked potentials during relative cerebral ischemia may be centrally mediated by a non-kappa mechanism.

K-opioid receptor changes in experimental models of cerebral ischaemia and atherosclerosis in the rabbit

Thromboembolic phenomena and transient ischaemic attacks (TIA) are considered the basis of ischaemic pathologies. The aim of the present research is to investigate the involvement of k-opioid receptors in cerebral blood flow (CBF) impairment which results in experimental stroke or dietary atherosclerosis in rabbits. CBF measurement showed a significant decrease in rabbits submitted to embolization and/or atherosclerosis. Binding studies showed that massive cerebral ischaemia and atherosclerosis produced a significant increase in the number of k-opioid receptors (Bmax), without changing (KD) affinity values. In conclusion, the results obtained seem to indicate that the increase in k-opioid receptors might play a crucial role in a common cerebral biochemical mechanism both in ischaemic and atherosclerotic pathologies.

Effects of NG-nitro-L-arginine and L-arginine on regional cerebral blood flow in the cat

1. We studied the effects of NG-nitro-L-arginine (NOLA), a potent inhibitor of the L-arginine-nitric oxide pathway, and L-arginine, the precursor of nitric oxide, on regional cerebral blood flow, electrocortical activity and ex vivo cerebrovascular reactivity in the cat. Flow was measured via radiolabelled microspheres, and vascular responses were studied by measuring isometric tension of isolated middle cerebral arterial rings. 2. NOLA (30 mg kg-1 bolus followed by 1 mg kg-1 min-1 infusion) caused an approximately 40 mmHg elevation in the mean arterial blood pressure, a regionally heterogenous increase of the regional cerebrovascular resistance and a decrease in the regional cerebral blood flow 15 and 40 min after the start of its administration. In contrast L-arginine (30 mg kg-1 bolus followed by 10 mg kg-1 min-1 infusion) did not alter blood pressure, cerebrovascular resistance nor regional cerebral blood flow 15 min after the start of its administration. The NOLA-induced changes in tissue flow were the most pronounced in the cerebellum, pituitary and medulla oblongata, whereas there was no decrease in the flow of the cortex and white matter. 3. NOLA caused characteristic changes in total fronto-occipital EEG power and in power spectra which were unlikely to have been due to cerebral ischaemia. In addition, the ex vivo reactivity of the middle cerebral arteries showed signs of impaired endothelial nitric oxide synthesis: there were enhanced noradrenaline-induced contractions and N-ethoxycarbonyl-3-morpholino-sydnonimine (SIN-1)-induced relaxations and markedly attenuated acetylcholine- and ATP-induced relaxations after NOLA treatment. 4. The present data indicate that resting cerebral blood flow and cerebrovascular resistance are regulated by nitric oxide derived from L-arginine in a regionally heterogenous way and that exogenous L-arginine availability is not a limiting factor in this nitric oxide generation. Possibly, both the vascular endothelium and the neurons contribute to this basal nitric oxide release.

A double blind randomized pilot trial of naloxone in the treatment of acute ischemic stroke

Attention has focused on naloxone, an opiate receptor antagonist, because of its potential benefit in reversing neurological damage after acute cerebral ischemia. To evaluate the safety and possible efficacy of high-dose naloxone in ischemic stroke patients we planned a double blind pilot study. Between January 1989 and May 1990 24 patients were randomly assigned to the naloxone or placebo group according to age and neurological deficit. Naloxone was given in a loading dose of 5 mg/kg over 10 minutes followed by a 24-hour infusion at the rate of 3.5 mg/kg/h. 10 patients experienced minor side effects but none of them had to discontinue the treatment. 9 patients improved: 6 in the naloxone group and 3 in the placebo group, but no significant difference was found using the non parametric Mann-Whitney test. Our study suggests that naloxone is safe at the dose used, but the results do not support the planning of similar trials on a larger scale.