Unit UPRES "tissue trauma and inflammation, Faculty of Bicêtre, University Paris XI and Intensive Care, Hospital Cochin-Saint Vincent de Paul-La Roche Guyon, 27 rue du Faubourg Saint-Jacques, 75679 Paris cedex 14 , France
Introduction
Starting from the observation that oxygen is essential for the proper functioning of our body, one would think that the toxicity of oxygen is more myth than reality. However, the specific physico-chemical oxygen mean that this element is quite capable in certain circumstances to cause toxic effects
Molecular oxygen or oxygen (O2) is a free radical. A free radical is a neutral or charged chemical species that has an unpaired electron in its orbit said single external. This gives the free electron free radical high chemical reactivity with a half-life very short (of the order of nanoseconds) since the free radical seeks réappareiller its unpaired electron. Free radicals are involved in redox reactions by capturing an electron. The exchange of an electron can either stop the spread of radical reaction or amplify it by forming free radicals more unstable than before.
If oxygen is itself a free radical of low reactivity, it is not the same for the activated derivatives of oxygen are the superoxide anion (O2 ° -), hydrogen peroxide ( H2O2) and hydroxyl radical (· OH). Initially known for their cytotoxic effects, it is now established that they are involved in many signaling pathways. These radical species are continuously formed at low levels in cells, either at the plasma membrane (NADPH oxidase, myeloperoxidase, cyclooxygenase), or in the cytoplasm (heme oxygenase), either at the mitochondria.
As with all active particles in the body, the formation of oxygen radical species (ROS) is tightly controlled by reactions of detoxification. These reactions may be enzymatic or nonenzymatic. Reactions of detoxification enzyme will succeed resulting in the formation of a molecule inert H2O (Figure 1). The non-enzymatic defenses are primarily vitamins A, C, E and glutathione. They capture the free electron of the radical species by redox making it inactive.
Figure 1. Mechanism of enzymatic detoxification of ROS.
It is only in certain circumstances pathophysiological that the toxicity of ROS will be deleterious. These circumstances may be related to either a decrease or lack of defenses antioxidant, or too prolonged exposure to concentrations or high pressure oxygen.
The cellular toxicity of oxygen free radicals occurs by degradation of lipid membranes, by reactions of lipid peroxidation, protein damage by formation of derivatives with the carbonyls and damage nucleic acids with a fragmentation DNA. ROS are also involved in the mechanisms of cell death by activating caspase-dependent apoptotic pathway.
The pathophysiology of oxygen toxicity is a question debated for many years. While some mechanisms have been described for the direct toxicity of oxygen at the cellular level, the importance of the involvement of oxygen in many pathological situations is not yet formally demonstrated.
In clinical practice, it seems difficult to respond to this question, this is mainly due to the fact that the direct detection and measurement of oxygen radical species in vivo is impossible. The only measurement means at our disposal are indirect techniques. The evaluation of oxidative stress is most frequently performed by the detection of degradation products lipids, proteins and DNA generated by the mass production of radical species. Increased levels of these degradation products is interpreted as an increase in pro-oxidant activity. The mechanisms of detoxification of ROS can also be evaluated. This allows for an indirect reflection of the antioxidant activity of the organization and approach the redox balance prooxidant / antioxidant. Thus, an increase of metabolites derived from oxidation of molecules in the body and reduced antioxidant defenses reflect a state of "prooxidant" of the organization. Schematically this state " prooxidant "is interpreted in the clinic as a reflection of a pathological increase in oxidative activity.
must however be realized that few studies explore comprehensively the status of antioxidant defenses due to technical difficulties this represents. Conclude that variation in antioxidant defenses on the measurement of a single variable (enzyme activity or plasma) is probably insufficient.
must also corroborate the results obtained with the clinic. Free radicals are continually formed in the body, variations of certain parameters do not necessarily want that there is a toxicity of ROS in the body.
Experimental
From Cell ... If cell toxicity of oxygen is now undisputed, its mechanisms are not yet fully elucidated. High concentrations of oxygen (95%) will result in the formation of ROS. This release of ROS will then activate different signaling pathways mediating cellular cytotoxicity and [1]. These findings are reinforced by the fact that administration of antioxidants inhibits some of these phenomena. The fact that the toxicity induced by oxygen is mediated by ROS has prompted various authors to pay special attention to the mitochondria which, through the release of cytochrome C and activation of the caspases, is the ideal culprit involved in cell death. However it seems that the NAD (P) H oxidase plays a crucial role in the formation of ROS intracelllulaire [2].
activation by ROS, including MAPkinases of ERK1 / 2 (Extracellular Signal-Regulated Kinase 1 / 2) will initiate the process of apoptosis [3]. Although the timing of activation of different actors and their roles are not clearly understood, it is demonstrated that cell death induced by oxygen involves modulation pathways of Fas, p53, p21, Bcl, Bax [4]. This signaling pathway results in cleavage of PARP (poly (ADP-ribose) polymerase) and pathway activation of caspases [3], resulting in translocation to the nuclei of NFK-B, leading to cell death.
The oxygen toxicity will also lead to a direct activation of the inflammatory response as evidenced by the activation of neutrophils, the expression of cell adhesion molecules and release of proinflammatory cytokines such as TNF - alpha and IL-1.
If the involvement of TNF-aplpha in oxygen toxicity is not yet clear, it was shown that its inhibition was protective and that the receptor TNFR-I plays a role in cell signaling pathway this toxicity [5].
Oxygen module thus its toxicity via direct signaling pathways of cell death and by direct activation of inflammation.
Another phenomenon of oxygen toxicity is inhibition of cell growth. Oxygen may act directly on DNA due to fragmentation through reactions redox, causing mutations in genes encoding proteins of cell growth [6] [7]. But oxygen also acts directly on the modulation of growth factors such as vascular endothelial growth factor (VEGF). The expression of VEGF was diminished lung during exposure to hyperoxia [8].
If the direct toxicity of oxygen seems to be responsible for apoptotic phenomena, the toxicity of ROS seems more frustrated. An excess formation of ROS leads to cell death [9]. ROS lead to degradation of lipid membranes, proteins and DNA fragmentation leading to cell lysis. Cell death observed was not then an adaptive programmed death but a "simple" phenomenon of necrosis [10].
the animal ... is probably on the lung that direct toxicity of oxygen has been the most studied. Body boundary between air and blood, he is constantly in contact with oxygen concentrations that can vary depending on clinical situations.
Exposure to high concentrations of oxygen is responsible for increased mortality in rats [11]. Morphological changes of the lung, when exposed to high concentrations of oxygen, are similar to those seen during ALI (Acute Lung Injury) with initial exudative phase characterized by inflammation , atelectasis and edema and, later, fibrotic phase resulting in loss of functional parenchyma.
Recently, it has been shown in situ on a model of ex vivo rat lungs, an increase in FIO2 resulted in increased formation of ROS in the pulmonary endothelium and that this increase was dependent on concentration oxygen. The source of these ROS was mitochondrial initially and then mixed with a participation of NAD (P) H oxidase [12]. In addition, there is always a mouse model, production of ROS by lung epithelial cells in vitro and in situ after exposure to hyperoxia. This production is accompanied by an increase in cell death [3].
A remarkable fact is that this toxicity appears to increase in pathological situations. Tateda et al. [13] showed in a mouse model of pneumonia (Legionella) that mortality was significantly higher in rats placed in chambers of hyperoxia (FiO2 = 1) than in those placed in ambient air. The excess mortality associated with an increase of apoptotic phenomena in the hyperoxic group. Hyperoxia appears to increase the pre-existing inflammation.
In summary, experimentally, oxygen alone has a strong cytotoxicity responsible for an increase of the phenomena of cell death, either necrotic or apoptotic. This toxicity is found throughout the body, since mortality is increased in the groups of animals exposed to hyperoxia.
Clinical
If experimental data are in favor of oxygen toxicity, the identification of this toxicity in clinical practice is more difficult. The clinical consequences of oxygen toxicity are mainly studied in neonatology, hyperbaric medicine, anesthesia and intensive care.
In neonatal paediatrics The debate on the toxicity of oxygen has been open since the beginning of the resuscitation of preterm infants in the forties. The two major side effects have been reported are pulmonary toxicity, with the occurrence of bronchopulmonary dysplasia [14], and ocular toxicity with retinopathy of prematurity [15].
Preemies do not have the same sensitivity to oxygen than an adult. Indeed, the conditions of fetal oxygenation are significantly lower than that encountered in a term infant or adult, because the PaO2 of the fetus in utero is about 22 mmHg in the ascending aorta [16]. If one is interested in its antioxidant capacity, they are generally lower with, for example, levels of vitamin E that are lower than 80% of the normal adult. Antioxidant defenses of prematurity are therefore reduced and, moreover, continue to decline during the first days of life. During resuscitation, preterm infants are subjected to non-physiological conditions of hyperoxia with an inability to regulate oxidative stress.
Although experimental studies in animal models tend to be in favor of direct toxicity of oxygen in prematurity, it is less clear for clinical studies.
Indeed, despite an abundant literature on the subject, no study has now definitely concluded debate. The clinician's main concern is to leave a newborn in conditions of hypoxia may be responsible for neurological damage. That is why the objectives of arterial oxygen saturation (SaO2) are recommended above 95% [17]. This recommendation is based on the results of studies on the oxygenation of the newborn at term and not in preterm infants. In older preterm various published studies have concentrated mainly to demonstrate that SaO2 between 91 and 94% were not harmful [18]. Only a meta-analysis could show results on mortality in favor of resuscitation in air [19]. Indeed, the authors conclude that 20 children would be saved if he was resuscitated in room air resuscitation versus FiO2 1. As another study suggests, and if one considers that the fetus lives and grows in conditions of hypoxia, it might be to set goals saturation well below those currently recommended for neonatal resuscitation [20] . But such a strategy needs to be confirmed.
In hyperbaric medicine pathophysiological knowledge of the direct toxicity of oxygen have made great progress with the development of certain techniques of scuba diving (including those using oxygen-enriched gas) and better knowledge of side effects of treatment of decompression sickness with hyperbaric oxygen.
Clinically there are two main syndromes:
- the effect Paul Bert: the effects of hyperbaric oxygen on the brain. During acute exposure to hyperbaric hyperoxia, there are clinical manifestations of the 1.8 atmosphere (ATA) of pressure inspired O2. Neurological toxicity is manifested by an increased risk of generalized seizures, grand mal, driven by weakness, hypercapnia and any increase in metabolism (effort in cold water);
- Lorraine Smith effect: the effects of oxygen on lung function that appears beyond 0 , 5 ATA. If the oxygen toxicity is manifested clinically by cough, heartburn and shortness of breath, accompanied by pathophysiological events with the appearance of a macroscopic and microscopic inflammatory state in the bronchi as evidenced by a significant increase the percentage of neutrophils in BAL. These events are accompanied by inflammatory disturbance of EFR including the appearance of a restrictive pattern and impaired diffusion (decreased DLCO) [21].
These abnormalities disappear after cessation of exposure if it is not extended, with a gradual return to normal.
Adults The direct toxicity of oxygen on the normal lung was studied in anesthesia. It was shown during minor surgery, that ventilation with 100% FiO2 favored the occurrence of actélectasie and this despite an optimized ventilation with recruitment maneuvers [22] [23]. These are attributed atelectasis by the authors of gas resorption.
In ICU, the problem is not the same as in neonatology since the organization in adulthood is accustomed to higher PO2 with antioxidant defenses that are more developed. So under certain stress conditions that oxygen can become toxic. During hemorrhagic shock, the phenomena of ischemia-reperfusion are responsible for the formation of ROS. During acute infectious conditions, increased ROS production will quickly surpass the antioxidant capacity of the organism, which then become detrimental. Neutrophils are the main source of ROS. Once activated, they will release a large amount of ROS produced by NAD (P) H oxidase membrane. This "oxygen burst" is the first line of host defense against pathogens. But if the release of ROS is prolonged, the body's antioxidant defenses are outdated. These phenomena have been studied especially in the respiratory distress syndrome (ARDS). Analyses of BAL performed in patients with ARDS showed increased protein degradation products related to oxidative stress and a reduced antioxidant defenses [24].
Similarly, in states of shock, it was shown that there was an imbalance prooxidant / antioxidant status in favor of a prooxidant [25] [26]. The relevance of the results of these studies is sometimes difficult to identify. This is mainly due to the fact that the majority of clinical studies are only interested in one part of the reaction of ROS detoxification and it seems difficult to conclude a reasonably imbalance prooxidant / antioxidant if one or two only elements that constitute are analyzed.
Nevertheless, all studies seem to go in the direction of an imbalance prooxidant / antioxidant status in favor of a prooxidant during shock states. However, keep in mind that some of this oxidative stress may contribute to induce adaptive parallel phenomena of toxicity.
Implications for clinical practice?
The experimental data are in favor of oxygen toxicity, either in experimental models of cellular or animal models. The lesions caused by oxygen free radicals cause an increase in cell death and thus tissue necrosis. That we can not yet conclude on the predominant mechanism involved in cell death, whether apoptotic or necrotic, the debate is more specialist than the actual clinical relevance.
The technical difficulty is the evaluation of the toxicity of oxygen on the body is probably largely responsible for the disappointing results of all clinical studies. However, experimental and clinical data give rise to some reflections. There is no doubt that oxygen toxicity is well demonstrated by the experimental data. The fact that an animal exposed to high concentrations of oxygen dies in an array of respiratory distress is the proof. In our practice, anesthesia as intensive care, we are frequently confronted with pathological (ARDS, shock, neonatology) where the treatment of hypoxia plays a role in the management of the patient. An inflammatory condition or prolonged ischemia may be the bed of an increased toxicity of oxygen through the formation of ROS Excess body in a previous assault with antioxidant defenses in jeopardy. Correct hypoxia for normoxia is indisputable. But in view of literature data, it seems reasonable that clinicians pay particular attention not to exceed the thresholds normoxia (90 mmHg) may cause additional training of ROS, thus aggravating preexisting lesions. It seems more logical to correct the factors that decrease the affinity of hemoglobin for oxygen (hypercapnia, fever, acidosis) rather than unnecessarily increase the concentrations of oxygen. To insist
Prices obtain figures PO2 "physiological" can be particularly difficult, especially in ARDS, requiring the clinician to employ techniques of mechanical ventilation often iatrogenic (volotraumatisme, barotrauma). Cases of acute deep hypoxia (PO2 \u0026lt;50 mmHg) are not situations encountered daily in the ICU. The appearance of stigmata of peripheral hypoxia, detected by a rise of lactate reflecting a formulation of anaerobic metabolism, should alert the clinician. However, when lactate is normal and clinical examination did not reveal the signs of hypoxia (cyanosis), it does not seem necessary to increase the PO2, assuming a low PO2 alone without signs of poor tolerance need not be treated but monitored.
Oxygen free radical is a radical whose derivatives have high cytotoxicity by participating in redox reactions.
Highlights
· The oxygen free radicals play a physiological role and are trained continuously in the body.
· on antioxidant mechanisms regulate enzymatic and nonenzymatic continuous training.
· In some pathological situations, there is an excessive formation of free radicals exceeds the regulatory mechanisms and thereby becoming detrimental.
· The cellular toxicity of oxygen leads to the phenomena of necrosis and apoptosis.
· on experimental animal models show that exposure to high concentrations of oxygen (> 95%) is lethal.
· In neonatology, oxygen appears to be involved in the development of lung injury and eye.
· In hyperbaric medicine, the increase partial pressure of oxygen is responsible for the appearance of neurological disorders and impaired lung function.
· In Anesthesiology, in certain pathological conditions, free radicals of oxygen can increase tissue damage preexisting.
· The oxygen is a treatment that requires full, like any therapy, monitoring of the efficacy and tolerance.
Conclusion
The oxygen toxicity seems more closer to reality than myth. This toxicity is mainly related the use made by the clinician. Thus, consider oxygen as a treatment for complete with its own side effects. Its use should not be reduced to the mere opening and closing of a manometer but must be a full prescription with monitoring of the effectiveness and tolerance of treatment.
We expected a difficult game Saturday against a team comprising CAPBRETON many quality players that we know the value but also because our week was marred by numerous absences during training. This was true. Victory is at the end in an encounter generally mastered by good defense and despite some difficult moments. 
After equalities early encounter between two teams that fear and dare not wholly break free, we take the score to do more release. Throughout the first act of our advance will be limited to four goals up. Our biggest challenge has been put out of position goalkeeper from the visiting team that has too often thwarted because we were not lucid enough in the last act. Our attack lacked a bit of sharpness and motion do not allow easy solutions. We dug a small gap in the first half thanks to a few digs important activity caused by our defense. The pause is reached on the score of 15-12 in our favor. 
The beginning of the second half is consistent the first, ie players CAPBRETON very motivated, do not let go and fighting for every ball, sometimes with actions that deserved stronger sanctions. Face with some difficulty we control the score and gradually dig even the gap. Just 10 minutes from the end, our lead climbs to 8 goals. Unlike the first phase of game where players from the coast had squandered the score, they clung to the end to minimize their deficits in anticipation of the return match (indeed in a tie at the end of the championship the teams are in first place by goal average goal particular). While taking them because the final whistle is given a score of 30 to 24 for the Stadium. 
