Search

The dysautonomic (DC) or neurovegetative crisis remains an imperfectly known entity; it associates in a paroxysmal manner a reaction of sympathetic hyperreactivity that can lead to the prognosis. Our objective is to specify the etiological circumstances (DC) and their modality of treatment in pediatric intensive care unit., Materials-Methods: Descriptive study on files of children admitted in the intensive care unit of 2010-2015 who presented a DC acquired during their hospitalization., Results: In total, 41 patients included with an average age of 56.92 months presented DC. Among the etiological circumstances Guillain-Barré syndrome and head trauma are noted. Observed symptoms occur on average at one week of admission; they are related to the consequences of DC. The manifestations are polymorphic: a systolic hypertension is present in all cases with an average PAS of 141.24±13.48mmHg, an average PAD of 86.80±11.01mmHg, a vasomotor disorder, a hyperthermia are noted. Cerebral anoxia post cardiac arrest in 4 patients preceded the onset of DC. Apart from the etiologic treatment, 39 patients were intubated with mechanical ventilation, sedated with morphinomimetic and benzodiazepine±lioresal (baclofen). Treatment of hypertension resulted in the administration of a central antihypertensive. Evolution is good in addition to 5 deaths related to neurovegetative disorders., Conclusion: DC is a poorly understood situation in pediatric intensive care unit, and the circumstances of the disease are variable. The diagnosis must be made with careful consideration because the prognosis may be fatal., (Copyright © 2017 Elsevier Masson SAS. All rights reserved.)

Each year, in France, the number of cardiac arrests is evaluated between 30,000 to 50,000. When a patient survives, he undergoes a post-resuscitation syndrome which can aggravate the injuries and for which nowadays, no medication is available. In some kinds of cardiac arrest, a hypothermia protocol can be applied with a need for monitoring because of the appearance of side effects. In this context, hydrogen sulfide, which is a gasotransmitter with numerous physiological and pharmacological properties, may be interesting. Indeed, its use could protect against oxidative, inflammatory and apoptotic troubles induced by the post-resuscitation syndrome. The implied biochemical mechanisms are adenosine triphosphate potassium channels activation and cytochrome c oxidase inhibition. This molecule can also induce a suspended animation state characterized by a metabolism decrease, which could give a delay for physicians to start a therapeutic monitoring. Thus, in spite of a modest and sometimes contradictory literature, this compound could become the first neuroprotective molecule in cardiac arrest., (Copyright © 2015 Académie Nationale de Pharmacie. Published by Elsevier Masson SAS. All rights reserved.)

The occurrence of post-cardiac arrest syndrome may lead to death in some children who have recovered from a cardiac arrest. The post-cardiac arrest syndrome includes systemic ischaemia/reperfusion response, brain injury, myocardial dysfunction, and persistence of the precipitating pathology. The main cause of death is brain injury. Management includes strictly control of ventilation, oxygen therapy and haemodynamics associated with protection of the brain against any secondary injury: management of seizures, control of glycaemia and central temperature. Mild hypothermia should be considered in comatose children after cardiac arrest., (Copyright © 2012 Société française d’anesthésie et de réanimation (Sfar). Published by Elsevier SAS. All rights reserved.)

In spite of recent advances, cardiac arrest remains a serious illness for which survival rate stays very low. If prehospital death remains the major problem, in-hospital death is also important. Two thirds of deaths in intensive care are the result of initial brain damage. After discharge, risk of recurrence for victims of sudden death is important and requires specific care and sometimes the implant of a cardiac defibrillator. In survivors, medium and long-term survival is satisfactory and close to that of patients with similar age and comorbidities that have not suffered cardiac arrest. The << minor >> cerebral sequels remain unknown and their impact on quality of life needs further attention., (Copyright 2010 Elsevier Masson SAS. All rights reserved.)

Paraplegia by medullar ischaemia after cardiac arrest is rarely reported, especially without an oxo-ischaemic cerebral lesions. We report the case of a 17-year-old woman who underwent a 70 minute cardiac arrest with recovery of consciousness, but a symptomatic medullar ischaemia (paraplegia under T4)., (Copyright (c) 2010 Elsevier Masson SAS. All rights reserved.)

The usefulness of therapeutic hypothermia is highly debated after traumatic brain injury. A neuroprotective effect has been demonstrated only in experimental studies: decrease in cerebral metabolism, restoration of ATP level, better control of cerebral edema and cellular effects. Despite negative multicenter clinical studies, therapeutic hypothermia is still used to a better control of intracranial pressure. However, important issues need to be clarified, particularly the level and duration of hypothermia, the depth and modalities of sedation. A clear understanding of blood gases variations induced by hypothermia is needed to understand the cerebral perfusion and oxygenation changes. It is essential to recognize and to use hypothermia-induced physiological hypocapnia and alkalosis under strict control of cerebral oxygen balance (jugular venous saturation or tissue PO(2)) and also to take into account the increased affinity of hemoglobin for oxygen. Management of post-traumatic intracranial hypertension using hypothermia, directed by intracranial pressure level, and consequently for long duration, is potentially beneficial but needs further clarification.

There is a large body of experimental evidence showing benefits of deliberate mild hypothermia (33-35 degrees C) on the injured brain as well as an improvement of neurological outcome after cardiac arrest in humans. However, the clinical evidence of any benefit of hypothermia following stroke, brain trauma and neonatal asphyxia is still lacking. Controversial results have been published in patients with brain trauma or neonatal asphyxia. Hypothermia can reduce the elevation of intracranial pressure, through mechanisms not completely understood. Hypothermia-induced hypocapnia should have a role on the reduction of intracranial pressure. The temperature target is unknown but no additional benefit was found below 34 degrees C. The duration of deliberate hypothermia for the treatment of elevated intracranial pressure might be at least 48 hours, and the subsequent rewarming period must be very slow to prevent adverse effects.

Fever is a secondary brain injury and may worsen neurological prognosis of neurological intensive care unit (NICU) patients. In response to an immunological threat, fever associates various physiological reactions, including hyperthermia. Its definition may vary but the most commonly used threshold is 37.5 degrees C. In animal studies, hyperthermia applied before, during or after cerebral ischemia may increase the volume of ischemic lesions. The mechanism of this effect may include increase in blood brain barrier permeability, increase in excitatory amino acid release and increase in free radical production. In NICU patients, fever is frequent, occurring in up to 20-30% of patients. Moreover, after haemorrhagic stroke, fever has been reported in 40-50% of patients. In half of the patients, fever may be related to an infectious cause but in more than 25% of patients, hyperthermia may be of central origin. After ischemic stroke, hyperthermia during the first 72 hours is associated with an increase in infarct size and increase in morbidity and mortality. This holds true also after subarachnoid haemorrhage. After traumatic brain injury, fever is not related to mortality but may increase morbidity. Whereas no causal link has been established between fever and unfavourable outcome, it seems reasonable to treat hyperthermia in patients suffering from brain injuries. In such patients, antipyretics have a moderate efficacy. In case of failure, they should be replaced by physical cooling techniques.

Therapeutic hypothermia (less than 35 degrees C) is a promising strategy to improve neuroprotection after severe brain injury. Except in patients resuscitated from cardiac arrest, its effectiveness has not yet been demonstrated. Therapeutic hypothermia results in various side effects, including cardiovascular, hydroelectrolytic and infectious disorders, which could explain, in part, the lack of conclusive clinical studies. These hazards are associated with practical difficulties to induce and maintain targeted hypothermia and with rewarming management. An improvement in the techniques for achieving targeted hypothermia, more knowledge about side effects and further randomized clinical trials are needed before recommending the use of therapeutic hypothermia for patients with severe traumatic brain injury.

Around 50,000 cardiac arrests (CA) occur each year in France and survival remains as low as 3 to 5%. Cardiopulmonary resuscitation (CPR) includes several treatment techniques for CA that are regularly updated in French, European, and international guidelines. Extracorporeal life support (ECLS) has been suggested as a therapeutic option in refractory CA since 1976. However, the use of this technique has remained limited to hypothermic CA and to CA occurring during the perioperative period of cardiothoracic surgery, mainly because the results of the initial trials were deceptive. The ease of use of more recent miniaturized ECLS devices has permitted a wider use of the technique in cardiac surgery departments and intensive care units (ICU). Encouraging results have been published recently by several teams in France and Taiwan, in single centre retrospective and prospective cohorts. In these studies, most CA were from toxic or cardiac causes and occurred in the hospital. In these highly selected cohorts, survival with good neurological outcome has been observed in up to 20 to 30% of cases. Nevertheless, the preliminary results of the use of ECLS in out-of-hospital CA in France are very poor, with less than 1% survival being observed. It should be emphasized that the time delay to commencing ECLS in out-of-hospital CA was far greater than that previously reported in in-hospital CA. These contrasting results lead physicians who perform CPR to question the indications and contra-indications of ECLS in these conditions and the French health authorities to question the value of such costly techniques (real cost as well as use of important and highly specialized human resources). The authors shared the following concerns that require emphasis: that an uncontrolled development of ECLS in out-of-hospital CA may lead to its abandonment because of very poor favourable outcome; that ECLS may lead to the survival of patients with poor neurological recovery and the associated considerable suffering for the patient and its relatives (although further evolution to brain death has been observed in most of these surviving patients with poor neurological outcome); that nonhomogeneous criteria may be applied in France for the use of ECLS in case of refractory CA because of the lack of any published data on its indications and contra-indications.Therefore, French medical scientific societies, under the auspices of the French Ministry of Health, selected a group of experts to propose guidelines that could help physicians performing CPR for refractory CA in deciding if ECLS should be used or not. The following text reflects a consensus obtained by these experts coming from different scientific and medical background at the present time. It should be noted that the views expressed are very likely to be modified in the near future because this topic is evolving rapidly.

We report 4 cases of lactic acidosis in diabetic patients usually treated with metformin. For the first 3 patients, the clinical history was similar because lactic acidosis was precipitated by gastro-intestinal disorders whereas all of them were simultaneously treated with several nephrotoxic drugs. These 3 patients presented with acute renal failure on arrival at hospital. Their issue was fatal whereas any obvious cause of overproduction of lactate was found. The fourth case, which was due to a voluntary intoxication, was the only one presenting with a favourable evolution. The metformin plasma and red blood cell levels were performed for 2 of 4 patients and confirmed the overdose. These observations remind that metformin-associated lactic acidosis remains a serious complication, and that medical doctors must respect strictly contra-indications and guidelines for withdrawing metformin.

The aim of this study was to assess management of patients resuscitated after pre-hospital cardiac arrest, initially indicated to preserve neurological status, the aetiological investigation only being undertaken when the outcome is favourable. Eighty-nine pre-hospital cardiac arrests were analysed retrospectively. The hospital survival was 16%, death being due to neurological lesions (55%), uncontrollable haemodynamic instability -39%) or other causes (7%). One year after the initial episode, none of the survivors had died, all living autonomously without (8 patients) or with minimal neurological sequellae (5 patients). These results are concordant with reports in the literature. The 11 cases of cardiac arrest with a favourable outcome of presumed cardiac origin underwent coronary angiography (6 cases) or endocavitary electrophysiological investigation (8 cases). These investigations showed or suggested an ischaemic process in 4 cases, an arrhythmia in 6 cases and severe valvular heart disease in 1 case. The independent predictive factors of survival were a Glasgow score of 6 or more on admission, the persistence of a light reflex and benign EEG appearances according to Synek's classification. The authors conclude that these results are comparable to those reported in the literature with aetiological investigations reserved for cases of favourable neurological outcome. The investigations including coronary angiography and electrophysiological investigation are essential as shown by the diversity of the cardiac pathologies identified.

Purpose: The goal of this study was to follow over time the MR imaging features of cortical laminar necrosis., Patients and Methods: Six patients with cortical laminar necrosis were included. There were two women and four men aged 54-84 years, with a mean age of 68 years. In four patients, cortical laminar necrosis was caused by ischemic stroke, one case occurred after a cardiac arrest and the last patient had a meningoencephalitis. The time delay from insult to the first MR study varied between one week and 3 months., Results: The MRI showed hyperintense lesions in the cerebral cortex on T1W and T2W images. The high intensity signal was still observed a few months after the insult. Cortical laminar necrosis lesions did not demonstrate hemorrhage on CT and MRI studies., Conclusion: MRI allowed detection of cortical laminar necrosis and could differentiate it from hemorrhage.

Neuromuscular complications acquired in the intensive care unit form a new clinical entity. Three neuromuscular deficits are described: so-called critical illness polyneuropathy where neuromyopathic changes are associated with corticosteroid and/or neuromuscular blocking agents, and catabolic myopathy. We report four new cases: three of them concerning polyneuropathy and one neuromyopathic change.

We report the case of a 51-year-old patient admitted after a transient cardiorespiratory arrest. The abdominal CT scan revealed the presence of hepatic portal venous gas. At laparotomy, a diffuse mesenteric ischaemia was diagnosed. The patient died from multiple organ failure in the subsequent hours. Necrotic bowel is associated with hepatic portal venous gas in 50% of the cases and the current mortality rate is 85%. Gas originates either through intestinal transmucosal passage, either by intraportal bacterial gas production, or through both mechanisms.

Unlabelled: In order to evaluate the determinants of malignant vasovagal syncope with asystole revealed by the tilting test and to determine the possible therapeutic implications, 179 patients (91 women and 88 men, mean age 36.6 +/- 20.1 years) referred for the assessment of unexplained a were studied. The test was performed with a tilt of 60 degrees for 45 minutes. A bolus of isoprenaline (0.02 to 0.08 microgram/kg.min) was injected in the case of a negative passive test. Asystole was defined as a ventricular pause lasting > or = 5 seconds., Results: Ten (13%) of the 77 patients with a positive tilting test experienced a cardio-inhibitory reaction with prolonged asystole lasting an average of 11.9 +/- 4.9 seconds. Compared to the other 67 patients with a positive test, those with asystole were younger (23/9 +/- 14.8 years vs 32.9 +/- 18.5 years, NS) and had a more frequent history of convulsions (6/10 vs 9/67, p = 0.05) during spontaneous episodes and trauma (9/10 vs 27/67, p = 0.005). Implantation of a pacemaker was chosen first-line treatment for the first 6 patients. Their follow-up tilting tests remained positive (pre S = 4, S = 1) despite DDD stimulation of 45 bpm. Five of these patients and the following 4 patients were retested under beta-blockers. In six patients treated with beta-blockers, the clinical symptoms resolved completely (n = 3) or improved (n = 3), in contrast with 3 other patients in whom the tilting test remained positive with recurrence of asystole. The mean follow-up for the 169 patients is 22.7 +/- 11 months and the ten patients with asystole remained totally asymptomatic., Conclusion: An asystolic response during the tilting test is characteristic of vasovagal syncope described as malignant. The syndrome essentially affects young patients, with a more frequent history of trauma and convulsions. Beta-blockers appear to be at least as effective as permanent pacemaker to prevent symptoms in this specific subgroup.

We report a case of a permanent flaccid paraplegia, with a sensory loss at T12 level, not associated with cerebral damage, subsequent to a cardiac arrest of 15 minutes duration, in a 67-year-old patient, undergoing haemorragic surgery for gangrenous purulent cholecystitis. Besides cardiovascular collapse and subsequent circulatory arrest, possible favouring factors include anatomical anomalies in the territory of anterior spinal artery, surgical posture with hyperlordosis generating venous stasis, emergency haemostatic maneuvers with compression of the arterial territory providing spinal blood supply. Spinal cord lesions are probably more frequent than expected, as the often associated cerebral anoxic encephalopathy impedes their recognition. Only a systematic anatomopathological examination of the spinal cord, in patients who died after a cardiac arrest, would provide the accurate incidence of spinal complications.

Authors report a case of focal action myoclonus due to a localized hemispheric lesion. The woman had cardiac arrest complicated by ischemic stroke involving the left parietal region. Action myoclonus of the inferior right limb was the sole residual complaint, provoked by movement and maintenance of posture. Patient noted that myoclonus was worsened by carbamazepine and vigabatrin. EEG-EMG polygraphic study demonstrated action myoclonus and showed a particular EEG activity consisting in a volley of spikes on the vertex region during the right inferior limb movements. Contralateral movements did not elicit this activity. Pharmacological response to clonazepam, piracetam and alcohol was studied. A marked improvement of myoclonus was obtained with alcohol. Pathophysiology of this exceptional form of cortical action myoclonus is discussed.

The case of a diabetic 62-year-old man with a past history of myocardial infarction, developing a cardial arrest followed by successful cardiopulmonary resuscitation, is reported. In the late clinical course, the patient displayed abdominal signs related to mesenteric ischaemia. The pathophysiology of non-occlusive mesenteric ischaemia is discussed. Risk factors such as diabetes, cardiovascular disease, hemodialysis, the use of digoxine or alpha-adrenergic drugs are listed. Non-occlusive mesenteric ischaemia is not an infrequent complication of cardiac failure in high risk patients.

The study of 33 cerebrospinal fluids of infants victims of sudden death shows a very significant increase of the metabolites of dopamine and serotonin. These determinations, compared to a control group, indicate a failure, concerning these two neurotransmitters, which could induce a cardiorespiratory seizure. This failure has likely a multifactorial origin.

An open prospective study of brain energy metabolism was carried out in 20 consecutive cases of coma occurring after cardiopulmonary arrest (CPA) of various causes. Mean age was 54 years and mean duration of CPA 2.75 min, with a mean of 11 min for resuscitation. Brain energy metabolism was compared with clinical course, electroencephalogram (EEG), and neurological outcome. All the patients were given the usual intensive care: intermittent positive pressure ventilation, haemodynamic control, sedation with benzodiazepines, neuromuscular paralysis, anticonvulsants, mannitol. A clinical examination and an EEG were carried out daily, 4 h after all drugs which might interfere with these investigations or brain energy metabolism had been stopped (except for anticonvulsants). Successive measurements of the cerebral arteriovenous oxygen (C (a - v) O2) and glucose (C (a - v)gluc) contents were made. The oxygen glucose index (OGI) was calculated according to Cohen's formula. The first set of measurements were carried out within the first 30 h after CPA. The last measurements were made a few hours before death or recovery, or stabilization of the coma. Patients were assigned to 2 groups according to their neurological outcome: group R (n = 8), patients who recovered consciousness; group D (n = 12), patients who developed brain death or a vegetative state - Safar's cerebral performance category 4. During the first 30 h, there was no relationship between clinical course, EEG and cerebral outcome. Half of the patients died or recovered within 72 h.(ABSTRACT TRUNCATED AT 250 WORDS)

Ischaemic brain damage can follow global cerebral hypoxia, localized cerebral hypoxia and global cerebral anoxia as it occurs after circulatory arrest. The calcium-ion-mediated mechanism is one of the main routes to cerebral deterioration. Barbiturates are restricted for treatment of increased intracranial pressure and seizures. Calcium channel blockers cannot yet be recommended. Therapy remains mainly symptomatic. Hyperglycaemia should be avoided.

This study reports on a group of eight patients having experienced a cardiac arrest while in the hospital. After a quick recovery from the accident they are screened for neuropsychic disorders during and up to six months after their hospitalization. These symptoms are aspecific and are similar to those encountered after any hospitalization in an intensive care unit. Only patients with coronary artery disease exposed to the risk of recurrence, remain with a feeling of insecurity partially supported by the family.

One of the crucial factors affecting mortality and morbidity after circulatory arrest the ischemic neuronal damage following complete cessation of cerebral blood-flow. To date, no accepted pharmacologic neuroprotective therapy has emerged. Cerebral ischemia causes a rapid shift of Ca++ from the extracellular spaces into cells and it is assumed that this excessive entry of Ca++ is the final pathway of cell death. In addition, Ca++ is involved in the diffuse vasospasm which occurs after global cerebral ischemia. Therefore, calcium entry blockers such as dihydropyridines derivatives have sparked considerable interest especially because of their preferential cerebrovasodilating effects. In vivo studies have demonstrated protection from brain ischemia with calcium entry blockers. However no direct protective effect of these drugs has been shown on neurons. More recent results have underscored the importance of excitatory amino acid neurotransmitters and receptors (particularly N-Methyl-D-Aspartate receptors) in causing intracellular calcium overload and neuronal death after ischemia. Blockade of these receptors or their associated channels may be an interesting way to protect the brain against ischemic damage.

The outcome of anoxic coma following cardiac arrest depends on the aetiological circumstances, on pre-existing visceral deficiencies and on the duration of inefficient circulation. Outside the extreme cases of prompt return to consciousness or early death, in many patients this course is marked by neurological sequelae of varying severity which may result in a persistent vegetative state. Initially, there is nothing that can predict the quality of survival, but within 72 hours the neurological examination usually makes it possible to foresee irreversible situations with permanent loss of consciousness. The decisional problems that ensue are discussed.

A cooperative study involving 23 centres enabled review of 69 cases of sudden death occurring less than one hour after onset of symptoms recorded by the Holter method and not related to recent, clinically documented myocardial infarction or to class IV cardiac failure. The 15 cases of asystole (22 p. cent) were observed in elderly patients (73.3 +/- 2.7 years) whose known ischaemic heart disease (12/15) was confirmed in 10 cases as the direct cause by the preceding acute ST changes. In 2 cases, death resulted from AV block presumed to be iatrogenic. The 13 episodes of torsades de point (19 p. cent) occurred mainly in younger women (58.8 +/- 6 years) without apparent cardiac disease (8 cases) and were provoked by a Group IA antiarrhythmic drug (7 cases) or by hypokalemia (3 cases). Apart from 1 case of congenital long QT syndrome, slowing of the sinus rhythm was observed (78.3 +/- 2.6 to 60.2 +/- 2.7 bpm, p less than 0.001) in the 3 hours preceding these episodes, and ventricular bigeminy with a long coupling interval was recorded in the lasts seconds before the torsades. The 41 (59 p. cent) cases of ventricular fibrillation (VF) were observed in men aged 64.9 +/- 2 years with coronary artery disease (39/41). However signs of acute ischaemia were only found in 5 cases. The VF was primary in 8 cases and secondary to ventricular tachycardia (VT in 33 cases). An acceleration of the cardiac rhythm (83.3 +/- 3.4 to 90 +/- 4.1 bpm, p less than 0.01) was recorded in the hour preceding VF and other arrhythmias were common: atrial tachycardia (4 cases), atrial extrasystoles (4 cases), a new type of ventricular extrasystoles (VES). The VF and VT were preceded by a long cycle in 17 cases. The first complex was different from previous VES in 10 cases and identical to the previous VES in 16 cases; in 4 cases this feature could not be identified and in 11 cases there were no premonitory VES. The coupling interval of the initial VES was shorter than that of the most premature preceding VES (368 +/- 13 ms vs 442 +/- 19 ms, p less than 0.001), especially in primary VF (335 +/- 9 ms, N = 8) compared to polymorphic VT (360 +/- 12 ms, N = 11) or monomorphic VT (384 +/- 18 ms N = 22).(ABSTRACT TRUNCATED AT 250 WORDS)

Since the advent of intensive care units the number of cases of posthypoxic action or intention myoclonus has greatly increased. This has given rise to renewed research into the physiopathology and therapy of the disease. The latent effective treatment is based on derivatives of benzodiazepine, and especially L-5-hydroxytryptophan, administered with or without an inhibitor of peripheral decarboxylase. The disadvantages of these compounds, associated with their secondary effects, make our fortuitous discovery of the effectively curative action of piracetam doubly valuable.

Two cases of post-anoxic occipital blindness, one in an adult and the other in a child, are presented. A quantitative and qualitative semiological analysis was performed using tests described in a previous paper. Mechanisms underlying restitution of visual function were chronologically followed up over three years of rehabilitation. In both cases there was a noteworthy improvement in the visual function, related to the beginning of rehabilitation therapy. Five years of an almost complete cerebral blindness had preceded the improvement in case 2. Spatial construction acquisition and ontogenetic data are compared with the present data. It is suggested that there is a similar loss of activation in Broca's aphasia and cerebral visual disturbances of occipital origin.

Sixteen cases of anoxic encephalopathy in adult patients admitted to the intensive Care Department of Raymond Poincaré Hospital (Garches) have been selected on the following criteria: resuscitation by external cardiac massage for cardio-respiratory failure, without any previous history of anoxia and with detailed data on the subsequent evolution of neurological signs and symptoms. Main clinical features were compared to the results of the neuropathological examination. The pattern and the distribution of changes are detailed. The cerebral cortex was always involved by diffuse ischemic cell changes or by laminar necrosis. The Sommer's sector of Ammon's horn was affected in 15 cases, the basal ganglia in 14, the cerebellum in 11, the brain stem in 6. The very rarely mentionned necrotic brain stem changes, which involved inferior colliculi, periaqueductal gray matter and substantia nigra were sometimes associated with lesions of the floor of the IVth Ventricle and of the mamillary bodies. They were different from Wernicke's encephalopathy with could be dismissed on clinical and pathological grounds. No clinical signs or symptoms specific for this distribution of changes could be found by the retrospective study of the clinical files.

Twenty patients suffering from post-ischaemic encephalopathy were treated with high doses of barbiturates during the period immediately following resuscitation. The duration of cerebral ischaemia was assessed retrospectively. The degree of ischaemic damage was evaluated on the one hand by the pupillary signs seen 10 minutes after the reestablishment of the circulation and secondly by enzyme levels in the CSF. This barbiturate load was not associated with major complications and the excretion of barbiturate continued for several days. The clinical signs seen 12 hours after ischaemia and continuous observation of the tracing of the cerebral function monitor made it possible to give an early favourable prognosis from a neurological standpoint. In all the patients (apart from one) in whom there was total cerebral ischaemia for less than 10 minutes, neurological recovery was complete.