The underestimated effect of normobaric hyperoxia on cerebral blood flow and its relationship to neuroprotection

* These authors contributed equally to this work. Sir, Neurons in the ischaemic penumbra are at risk from ischaemia-induced constrained oxygen delivery. Therefore, increasing arterial oxygen content by normobaric oxygen therapy (NBO) has been thought a logical strategy, and as expected further shown to increase markedly and rapidly the very low tissue oxygen pressure that prevails in the ischaemic penumbra ( Shin et al. , 2007 ; Baskerville et al. , 2011 ). However, due to controversial results in rodent stroke models, there is still no clear consensus on the benefits of NBO. By demonstrating in a recent issue of Brain that NBO prevents behavioural deficits and almost suppresses neuronal loss in a rat model of transient ischaemic attack, the work of Ejaz et al. (2016) brings renewed interest as for the benefits of NBO in brain ischaemia. One caveat when critically assessing the effects and mechanisms by which NBO, alone or in combination with tissue plasminogen activator (tPA), provides benefits on brain ischaemia is that little attention is paid to the facilitating action of NBO on cerebral blood flow ( Shin et al. , 2007 ; Baskerville et al. , 2011 ; David et al. , 2012 ) and to the use of transient rather than permanent, and thromboembolic rather than mechanical models of ischaemic stroke. Doubtlessly, this is due to the fact that the ability of NBO to improve the very low tissue oxygen pressure that prevails in the ischaemic penumbra is generally believed to occur through passive-mediated oxygen transport (i.e. oxygen diffusion). However, contrary to this straightforward reasoning, and to hyperbaric oxygen therapy, NBO has been shown to fail reducing cell injury in acute brain slices exposed to oxygen and glucose deprivation ( Chazalviel et al. , 2016 ). Also, while NBO and hyperbaric oxygen therapy have both been reported to reduce infarct size in rats subjected to transient thromboembolic brain ischaemia when administered 1 h before tPA, it is noteworthy that only hyperbaric oxygen therapy has been shown to reduce infarct volume in similar conditions in rats subjected to permanent thromboembolic ischaemia ( Sun et al. , 2010 ). Consistent with these data, administration of NBO for 3 h, started 30 min after ischaemia onset, has been further reported to reduce infarct size in rats subjected to transient, but not to permanent, ischaemia (Figure 3 in Henninger et al. , 2007 ). Taken together, these data provide evidence that NBO has no ability to induce passive oxygen transport within the brain parenchyma, and that the mechanisms by which NBO provides benefits require vascular support. Consistently, NBO has been reported not only to increase tissue oxygen pressure in the ischaemic penumbra but also to increase perfusion in an oxygen concentration-dependent manner ( Shin et al. , 2007 ; Baskerville et al. , 2011 ), an effect whose mechanisms are still poorly known and whose critical role in neuroprotection is to our opinion largely underestimated. Indeed, NBO started 45 min after ischaemia onset has been shown to achieve clot lysis until reperfusion through direct interaction with tPA ( David et al. , 2012 ), which endogenous activity contrary to what is generally thought is significantly increased within the ischaemic brain ( Wang et al. , 1998 ), and to lead to subsequent reduction of infarct size, brain swelling, brain haemorrhages, and blood–brain barrier leakage in a rat model of thromboembolic ischaemia ( David et al. , 2012 ). Consistent with such a prothrombolytic effect of NBO through facilitation of endogenous tPA, NBO administration of 1 h duration, started 1 h after ischaemia onset until tPA injection, has been further reported to reduce infarct volume, brain haemorrhages, and blood–brain barrier damage compared to tPA alone in a rat model of transient, but not of permanent, thromboembolic ischaemia ( Sun et al. , 2010 ). Likewise, administration of NBO of 3 h 30 min duration, from 30 min before ischaemia onset to tPA injection 3 h later, has been shown to reduce infarct size in rats subjected to thromboembolic ischaemia compared to NBO/saline-treated rats, which, in contrast with air/tPA-treated rats, had decreased infarct volume compared to untreated air/saline controls ( Henninger et al. , 2009 ). Therefore, based on these data it is likely that oxygen transport through the collateral microvasculature, downstream of the blood clot and the ischaemic core, could lead to a facilitation of the thrombolytic efficiency of endogenous tPA, a mechanism that might reduce thrombin generation and blood platelet aggregation and coagulation, and explain the increase in perfusion ( Shin et al. , 2007 ; Baskerville et al. , 2011 ) and reduction of infarct volume provided by NBO in the ischaemic penumbra of rats subjected to brain ischaemia, even of mechanical type ( Shin et al. , 2007 ; Henninger et al. , 2007 ; Liu et al. , 2009 ). In contrast with the benefits of NBO reported in the above mentioned thromboembolic studies, in which halogenated anaesthesia was performed either in air ( Henninger et al. , 2009 ; David et al. , 2012 ) or in nitrous oxide, whose administration was stopped at least 60 min before NBO administration ( Sun et al. , 2010 ), another study has reported no thrombolytic or neuroprotective effect in rats shifted to NBO immediately after nitrous oxide anaesthesia ( Fujiwara et al. , 2009 ). This latter protocol from nitrous oxide to NBO, without room air in-between, is likely to have blocked the facilitating action of NBO on the thrombolytic properties of endogenous tPA, as nitrous oxide has been shown to inhibit the catalytic and thrombolytic efficiency of tPA through a similar mechanism than the selective bis-benzamidine tPA inhibitor tPA-stop and to suppress the benefit of tPA injection in a rat model thromboembolic brain ischaemia ( Haelewyn et al. , 2011 ). Indeed, once bound, inert gases with higher affinity than oxygen, such as xenon and nitrous oxide, are not removed from the brain as soon as it is generally thought when administration is stopped, so that wash-out time for anaesthesia of 1–3 h duration with 75% nitrous oxide has been estimated to be ∼60–90 min ( David et al. , 2012 ). In line with such effects and mechanisms, clinical studies in patients anaesthetized with or without nitrous oxide for non-cardiac surgery have shown that the use of nitrous oxide is associated with an increased risk of postoperative myocardial ischaemia ( Badner et al. , 2000 ). Finally, four studies investigated possible interactions between NBO and tPA in models of thromboembolic brain ischaemia; two studies in which NBO was administered 1 h or 3 h before tPA have reported reduced infarct size compared to air/tPA-treated rats ( Sun et al. , 2010 ) and air/saline- and NBO/saline-treated rats ( Henninger et al. , 2009 ), respectively, suggesting that early NBO might have induced recanalyzation or partial recanalyzation before tPA treatment. In contrast, another study in which NBO was administered simultaneously with tPA has reported increased infarct volume and brain swelling compared to NBO/saline- and air/tPA-treated rats, but not to air/saline-treated rats, and increased brain haemorrhages and blood–brain barrier damage compared to NBO/saline-treated rats ( David et al. , 2012 ) In line with these findings, other data that have shown on one hand that tPA and NBO, respectively, increases and decreases serum concentrations of MMP9 ( Michalski et al. , 2012 ), whose expression is critically associated with ischaemia-induced disruption of the blood–brain barrier, and that combining NBO and tPA suppresses the beneficial effect of NBO at decreasing MMP9 ( Michalski et al. , 2012 ). Finally, coadministration of NBO and tPA has been shown to induce neither adverse nor beneficial effect compared to air/tPA in rats shifted to NBO immediately after nitrous oxide, a condition that could have blocked adverse interactions between NBO and tPA as explained above ( Fujiwara et al. , 2009 ). Taken together, these data support that sequential administration, but not co-administration, of NBO and tPA could be a safe therapeutic strategy ( David et al. , 2012 ; Michalski et al. , 2012 ). In conclusion, the data above support that the main mechanism by which early NBO provides benefits in ischaemic stroke models is increasing cerebral blood flow in the ischaemic penumbra, a condition known to be truly beneficial only if it occurs early enough ( Fisher, 2011 ). This work was funded by the Service de Santé des Armées (Institut de Recherches Biomédicales) and the Université de Caen Normandie.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.