Peer Reviewed Article Lack of Oxygen to Brain

• Loading metrics

Open Access

Peer-reviewed

Enquiry Commodity

The impact of hyperoxia on brain activity: A resting-state and task-evoked electroencephalography (EEG) study

• Min Sheng,
• Peiying Liu,
• Deng Mao,
• Yulin Ge,
• Hanzhang Lu

10

• Published: May ii, 2017
• https://doi.org/10.1371/journal.pone.0176610

Figures

Abstract

A better understanding of the effect of oxygen on brain electrophysiological action may provide a more than mechanistic insight into clinical studies that utilise oxygen treatment in pathological weather condition, equally well equally in studies that employ oxygen to calibrate functional magnetic resonance imaging (fMRI) signals. This study applied electroencephalography (EEG) in good for you subjects and investigated how loftier a concentration of oxygen in inhaled air (i.east., normobaric hyperoxia) alters brain activeness nether resting-country and task-evoked conditions. Study 1 investigated its bear upon on resting EEG and revealed that hyperoxia suppressed α (8-13Hz) and β (14-35Hz) ring power (by 15.6±ii.three% and xiv.1±3.1%, respectively), but did not change the δ (1-3Hz), θ (iv-7Hz), and γ (36-75Hz) bands. Sham control experiments did non issue in such changes. Study 2 reproduced these findings, and, furthermore, examined the effect of hyperoxia on visual stimulation upshot-related potentials (ERP). Information technology was found that the main peaks of visual ERP, specifically N1 and P2, were both delayed during hyperoxia compared to normoxia (P = 0.04 and 0.02, respectively). In dissimilarity, the aamplitude of the peaks did not show a change. Our results advise that hyperoxia has a pronounced upshot on encephalon neural activity, for both resting-state and task-evoked potentials.

Citation: Sheng M, Liu P, Mao D, Ge Y, Lu H (2017) The impact of hyperoxia on brain activity: A resting-state and task-evoked electroencephalography (EEG) study. PLoS One 12(v): e0176610. https://doi.org/10.1371/journal.pone.0176610

Editor: Nanyin Zhang, Penn Country Academy, U.s.

Received: January 12, 2017; Accepted: April 13, 2017; Published: May 2, 2017

Copyright: © 2017 Sheng et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in whatever medium, provided the original author and source are credited.

Information Availability: All relevant data are inside the paper and its Supporting Data files.

Funding: This piece of work was partly supported by National Institutes of Wellness grants R01 NS067015, (to H.L.), R01 MH084021 (to H.L.), R01 AG042753 (to H.50.), R01 AG047972 (to H.L.), R21 NS095342 (to H.L.). The funder had no role in study design, information collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Oxygen is critical to human life. Due to the relative convenience and depression-cost to modulate O₂ levels, manipulation of its concentration in the body via control of inspired air has generated opportunities for clinical evaluation and interventions. Oxygen therapy has been widely used in wound-healing from radiation injury and diabetes [ane,2], and every mean solar day, thousands of people receive oxygen handling in major medical centers and private practice facilities. Nevertheless, despite the increasing utilise of oxygen as a treatment, the effect of O₂ gas modulation on the brain is poorly understood. Furthermore, certain applications of oxygen therapy specifically target brain diseases, such every bit cerebral ischemia [3–six], traumatic brain injury [7,8], and cerebral palsy [ix]. While the premise of many of these efforts is based on the metabolic need for hyperoxic gas, i.e., delivering more oxygen to the brain tissue, there could be other effects of hyperoxia on the brain. The goal of the present study was, therefore, to characterize the effect of hyperoxia on neural activity in healthy subjects, in whom oxygen delivery is sufficient under normoxic atmospheric condition.

In addition to therapeutic applications, hyperoxia has as well been used to calibrate the Claret-Oxygen-Level Dependent (BOLD) signal in functional magnetic resonance imaging (fMRI) to better gauge baseline and evoked neural metabolism in the human brain [10–12]. However, in this arroyo, an important assumption is that hyperoxia itself does not modify neural activity. Therefore, a better understanding of the result of hyperoxia on neural activeness may ameliorate fMRI technologies in terms of providing more accurate brain mapping.

Changes in neural activity induced past hyperoxia cannot be examined by indirect approaches, such every bit fMRI, Positron Emission Tomography (PET), or functional optical imaging, considering those methods cannot differentiate the neural effects from the vascular furnishings of hyperoxia. Consequently, several reports have focused on measures that are more directly related to neural activeness [thirteen,14]. Xu et al. [13] measured the cognitive metabolic rate of oxygen (CMRO₂) and observed a subtract in this rate under hyperoxia. Prior studies have also measured neural activity during hyperoxia, with mixed results. Kaskinoro et al. measured EEG under hyperoxia and reported an absence of signal change compared to normoxia [fifteen]. In contrast, Croal et al. used magnetoencephalography (1000000) and observed a moderate reduction (3–5%) in occipital lobe signal ability [16]. Withal, none of the prior studies differentiated the effects of hyperoxia during resting-land versus task-evoked activities.

In the present work, nosotros used EEG to mensurate resting-state and chore-evoked neural activity in a group of healthy subjects and examined EEG signal differences between hyperoxia and room-air conditions.

Materials and methods

Participants

The Institutional Review Board of the Academy of Texas Southwestern Medical Center canonical the study protocol. The studies were conducted between January 28, 2013 and December 20, 2013. Thirty-ix good for you subjects, xix–46 years of age (30.0±vii.three years) were enrolled. Twenty-six subjects were enrolled in the resting country EEG study, 13 of whom (30.0±half-dozen.7 years, 7 male, six female) were assigned to the O₂-breathing group and the other 13 (30.iv±viii.two years, vii male, 6 female) were assigned to the sham control group. Thirteen subjects (29.7±7.7 years, six male, seven female person) were enrolled in the task-evoked EEG study. The social club of the room-air and O₂-breathing weather were counterbalanced across participants in the chore-evoked study, thus a sham study was not performed. The participants were recruited from the University campus through flyers. None of the participants reported a history of neurological or psychiatric disorders, or pulmonary conditions, such as asthma. Participants were asked not to consume whatsoever caffeine-containing beverages for one hr before the experiment. All participants gave informed, written consent before participating in the study.

We conducted a power calculation based on the typical sensitivity of our EEG system in which a one-minute EEG indicate has a coefficient-of-variation (CoV) of approximately xx%. Based on our previous study of brain metabolic changes during hyperoxia, we assume that the EEG signal will change by 15% [xiii]. We estimated that, with a sample size of N = thirteen, we will take 0.95 ability to detect an EEG signal change in our written report when comparing O₂-breathing to room-air status.

Full general experimental procedures

Hyperoxia was accomplished with a breathing apparatus used extensively in our laboratory (http://world wide web.mri.jhmi.edu/hlulab/index_files/Page363.html) [17–20]. The hyperoxic gas consisted of 98% oxygen and 2% CO_two contained in a Douglas bag (Fig 1A). The reason the hyperoxic gas contained a small amount of CO₂ is that hyperoxia tends to cause hyperventilation, which could result in a decrease in EtCO₂, if non accounted for. Thus, by adding a pocket-size amount of CO_two in the inspired gas, the EtCO₂ value can be maintained relatively constant if subjects hyperventilate. The proper amount of CO₂ that can commencement the hyperventilation effect just non crusade an increase in EtCO₂ was previously calibrated. EtCO₂ results shown in Tabular array 1 confirmed that there were minimal changes in EtCO₂ during the oxygen claiming.

Fig 1. Report process and paradigm.

(a) EEG experimental setup. Components of the setup are labeled in the diagram and point the following: 1. EEG cap; two. Mock curl to back up the breathing valve; three. Gas delivery tube; four. Ii-mode non-rebreathing valve; 5. U-shaped tube; 6. Gas sampling tubes (ane for CO₂ and the other for O_two); 7. EEG amplifier; eight. Stimulus tracker for accurate recording of stimulus onset fourth dimension; ix. Stimulus markers; and ten. EEG recording laptop. (b) Timing image of the resting EEG study. (c) Timing paradigm of the task-evoked EEG study.

https://doi.org/10.1371/journal.pone.0176610.g001

The experiment was conducted with the participant in a supine position in a quiet mock-MRI room, in which visual stimulation and breathing comfort-level rating questions could be delivered via a back-projection display, and subject responses could be recorded through a push box. The participant breathed through a mouthpiece that was connected to a two-way valve (Hans Rudolph, 2600 Serial, Shawnee, KS). A nose clip was used to block nasal breathing to ensure the command of inhaled air. The participant was instructed to avoid biting on the mouthpiece, which could result in artifacts in the EEG signal due to an electric signal from the facial muscles. EtCO₂ and EtO₂ were recorded continuously during the entire session using a capnograph device (Capnogard, Model 1265, Novametrix Medical Systems, CT) and an O₂ sensor (Biopac, Module O_ii 100C, Biopac Systems, CA). One researcher remained inside the recording room throughout the experiment to monitor the physiology of the participant and to switch the valve to change the inspired gas. The participant was not aware of the timing of the gas switches, as the gas valve was exterior the bore.

During the experiment, the participant was asked (via the dorsum-project brandish) to rate their comfort level of breathing every one infinitesimal, on a calibration of i to 5, with 1 indicating very comfortable and five indicating very uncomfortable.

EEG recording

A 32-channel EEG system (Encephalon Products, Gilching, Germany) was used to perform the EEG experiments. Later the EEG cap was placed on the head of the participant, high-chloride annoying electrolyte gel (Brain Products, Gilching, Deutschland) was applied on each electrode such that the impedance values of all electrodes, except for the electrocardiogram (ECG) electrode, were beneath 10kΩ. The participant was then instructed to lie down on the table, and a mouthpiece and nose clip were placed. The EEG recording was then initialized and continued through the entire session beyond all respiratory states.

Resting-country study (Report 1)

In the resting state EEG study, each participant underwent a 25-minute session, as shown in Fig 1B, in which the participant breathed room-air during the first v minutes, followed past 10 minutes of animate a gas mixture contained in a Douglas purse, and and then returned to room-air breathing during the final ten minutes. For participants in the O_ii-breathing grouping, the Douglas pocketbook contained a hyperoxic gas mixture of 98% O₂ and ii% CO₂. For those in the sham-control group, the bag contained 21% O₂ and 79% Northward₂, equivalent to the composition of room-air. The participants of both groups were given the aforementioned instructions (both were told they would inhale oxygen).

A sustained-attention mensurate was collected throughout the session to verify that any EEG alteration was non due to a change in attending. The sustained-attention measure was relatively low-demand and comprised 26 alphabetical letters actualization in the center of the screen sequentially, in a randomized order. The participant was instructed to printing a push in their right paw every bit apace as possible when they saw a target letter of the alphabet, such as 'm.' The duration each letter of the alphabet appeared on the screen was 500ms. The size of the font was relatively small and was non expected to induce whatever visual evoked potentials. Response fourth dimension, which is the time gap between the display onset of the letter and the button printing, was recorded and sorted by respiratory atmospheric condition. Thus, it should be pointed out that the resting-country study was not performed nether completely "resting" atmospheric condition. Even so, at least the mental state of the participants was under controlled conditions.

The EEG information were recorded at a sampling charge per unit of 5000 Hz. Pre-processing of the EEG data was performed using a vendor-provided software Analyzer v2.0 (Brain Products, Gilching, Deutschland), which included the following steps. 1) Band-laissez passer filtering of 0.iii to 75Hz was applied with additional 60Hz notch filtering. two) The bespeak of each electrode was re-referenced to the average potential of all electrodes. 3) Independent component analysis was applied to remove center-blinking artifacts, and (4) cardioballistic correction was used to eliminate pulse artifacts. Visual inspection was conducted to manually exclude epochs with biting signal artifacts.

Mail-processing of the EEG information was performed using Matlab R2008b (Mathworks, Natick, MA). Specifically, afterwards linear detrending, the entire EEG recording was divided into one-second segments and Fast Fourier transform (FFT) was applied to each segment and for each electrode. The power spectrum from ane to 75Hz was computed in units of μV² at each Hz. The power spectra were then divided into standard frequency bands: ane to 3Hz for the δ band; 4 to 7Hz for the θ band; viii to 13Hz for the α band; 14 to 35Hz for the β ring; and 36-75Hz for the γ band [21]. Adjacent, the EEG power was averaged across the 31 channels. Finally, EEG ability nether each physiological state was quantified. The power under room-air was calculated as the boilerplate of the first room-air flow (v minutes) and the steady-state portion of the second room-air period (the last seven minutes of the second room-air period, excluding the transition period). The first and 2d room-air periods were averaged to account for potential drift of electrode impedance with time. The power under gas-mixture breathing used the steady-country portion of the gas challenge period (the concluding 7 minutes of the gas-challenge, excluding the transition period). The EEG powers were compared beyond physiological states using paired t tests. The O₂-breathing and sham command data were too analyzed together in a two-way replication ANOVA assay to determine whether the observed effects were specific to the O₂ grouping.

Job-evoked report (Study two)

In the task-evoked EEG study, the subject breathed four periods of room air (v-minute per catamenia) and iv periods of hyperoxic gas (for five minutes) in an interleaved fashion (Fig 1C). The order of room-air and hyperoxia periods was counterbalanced across subjects, thus no sham arm was used in the job-evoked study. During each cake, the first 2 minutes contained no task (but a fixation crosshair was shown on the screen) as it represents a transition period between physiological states. During the last iii minutes of the block, 96 visual stimulation trials were presented, in each of which a blue-yellow checkerboard (E-Prime software, Psychology Software Tools) was displayed for 500 ms, followed by a crosshair with a randomized duration between 900 and 1400 ms. To maintain the attending of the participant, the centre of the checkerboard (Fig 1A) independent different geometric shapes and the participant was instructed to press the right-hand button if it was a circle and the left-hand push button if it was any other shape. The reason we used different shapes rather than different letters, as in the resting-state study, is that the large number of letters would result in a long interval between matches, which is not ideal for maintaining a subject's attention. The full number of button presses between the two easily was balanced.

The digital output of the video display often causes an uncertainty in the onset time of the stimulus, which is related to the refreshing frequency of the display. To obtain a better accuracy with regard to stimulus onset, we applied a photo sensor (Stim tracker, Cedrus Corporation, San Pedro, CA) on the corner of the screen, the timing of which was recorded as an event marker on the EEG system. This improved the accurateness of the EEG timing to 0.two ms.

Pre-processing of the task-evoked EEG information followed steps similar to those used for the resting-state EEG. Post-processing was performed using ii Matlab-based toolboxes, EEGLab (UC San Diego, The states) and ERPLab (UC Davis, U.s.). For quantification of aamplitude and temporal features of the effect-related potentials (ERP), the continuous EEG data were segmented into trials based on the stimulus onset. The data for each trial consisted of EEG recordings from 100 ms before stimulus onset to 500 ms after stimulus onset. Trials from the entire session were sorted into room-air and hyperoxic conditions. ERP waveforms under the same respiratory conditions were averaged over trials (approximately 384) to yield a mean waveform. Amplitude and latency of the 2 most prominent peaks on the ERP waveform, N1 and P2, were quantified using ERPLab. For amplitude, the maximum bespeak intensity was determined. For latency, the time at which the signal reached 50% peak amplitude was determined [22,23].

The ability spectrum of the ERP signal was also examined. Information segments corresponding to the first 500 ms later the stimulus onset were subject to the FFT analysis and point powers in the δ, θ, α, β, and γ frequency bands were computed. For comparison, a similar analysis was carried out for the signal at 500ms preceding the stimulus onset. Paired t tests were used to compare the results betwixt respiratory conditions.

Results

The effects of hyperoxia on resting-land and task-evoked EEG were examined in carve up groups of young, good for you participants. At that place were no differences in historic period across the groups (ANOVA, P = 0.eight). All participants were able to complete the study and none reported whatsoever discomfort during the experiments.

Resting-state EEG study (Study 1)

Twenty-six participants were randomly assigned to an O_two-animate grouping (Northward = 13) or a sham control group (Due north = 13). End-tidal O₂ (EtO_two) and end-tidal CO2 (EtCO_two) levels of the participants are summarized in Tabular array 1. As tin can be seen, breathing of hyperoxic gas, just not medical air (21% O_ii and 79% N_two), increased EtO₂. EtCO₂ was not altered during the hyperoxic condition because of the small corporeality of CO_ii added to the hyperoxic gas, which offset the effect of hyperventilation. This ensured that any neural activity alteration that nosotros observed could be attributed to O_ii changes rather than to a CO₂ effect. The condolement level of breathing recorded during the session also showed no differences across physiological states (Tabular array 1), suggesting that any neural changes observed during the experiment could not be attributed to an increased attention to breathing.

Fig two displays resting-state EEG signal ability past frequency band for both the O₂-animate and the sham control groups (run across S1 File for individual bailiwick data). Paired comparison tests betwixt the breathing states revealed that, in the O₂-breathing group (Fig 2A), the indicate power was significantly decreased in the α ring (8-13Hz) by 15.half-dozen±2.iii% (P = 0.003) and in the β band (fourteen-35Hz) by fourteen.1±three.1% (P = 0.005) (mean ± SEM) when breathing oxygen. The δ (one-3Hz), θ (four-7Hz), and γ (36-75Hz) bands did not prove a meaning deviation (P = 0.iii, P = 0.five, and P = 0.3, respectively). The δ band plot (Fig 2A) showed a slight increase in hyperoxia although this did not accomplish statistical significance. Upon further investigation of the information, it was noted that this increment was due to a unmarried subject area who showed a large increment in δ band power. When this information point was omitted from the averaging, the plot showed no change in δ power (run across S1 Fig). In the sham control group (Fig 2B), none of the bands showed a meaning change in signal power.

Fig ii. Bar plots of EEG spectral power in (a) O₂-breathing (N = 13) and (b) sham command (North = thirteen) experiments.

NO = normoxia, HO = hyperoxia, MA = medical air. 1 asterisk = P<0.05. Two asterisks = P<0.01. Error bar = standard error across participants.

https://doi.org/10.1371/periodical.pone.0176610.g002

Next, the ratio of EEG ability between the gas mixture and the room-air periods was computed for both the O_ii-animate group and the sham control group (Fig 3), which factored out intersubject variations in EEG power. The ratio values were then compared using a two-way replication ANOVA analysis, in which one cistron was the frequency bands and the other factor was the written report grouping (O_ii-breathing versus sham control). The group-by-ring interaction effect was significant (F = 3.ii, P = 0.03), suggesting that the alteration in neural activity was specific to the O₂-breathing task and was frequency-dependent. Mail-hoc t-tests showed that the EEG power ratio in the α (P = 0.002) and β (P = 0.04) bands was significantly dissimilar betwixt the O_ii-animate and the sham command groups (Fig 3).

Fig 3. Ratio of EEG signal power for special gas mixture compared to room-air breathing.

For the O₂-breathing experiment, the special gas mixture used hyperoxic gas (98% O₂ + 2% CO₂). For the sham control experiments, the special gas mixture used medical air (21% O₂ + 79% North₂). Ane asterisk = P<0.05 for the comparison between red and blue symbols. Two asterisks = P<0.01 for the comparing between cerise and bluish symbols. Error bar = standard mistake.

https://doi.org/10.1371/journal.pone.0176610.g003

Fig 4 shows the group-averaged topographic maps of EEG power, too every bit the ratio (gas mixture divided past room-air) of power. It tin be seen that the hyperoxic effect on EEG ability was widespread across the brain. The higher frequency ability was suppressed by O₂-breathing and the lower frequency components appear to have been enhanced. In dissimilarity, no obvious modify was observed in the sham control case.

Fig 4. Topographic maps of EEG signals nether normoxia and gas-claiming conditions.

In the O_two-animate experiment, the gas-challenge condition used hyperoxic gas. In the sham command experiment, the gas challenge status used medical air. The EEG bespeak power was calculated for each frequency band and the maps shown are averages of all participants. In the topographic maps, the triangle on the top represents the subject's olfactory organ, while ii rectangles on each sides represent the ears. Each black dot indicates one electrode. The images on the correct-manus side evidence the ratio between two topographic maps.

https://doi.org/10.1371/journal.pone.0176610.g004

The level of attending was maintained throughout the experimental session, every bit there was no difference in response time (P = 0.17) for the attending measure.

Task-evoked EEG study (Study 2)

Fig 5 shows the ERP waveforms in response to visual stimulation in the EEG electrode that corresponds to the visual cortex, the Oz channel. Several ERP peaks, most prominently N1 and P2, can be clearly identified in every participant. Results of the quantitative analysis are shown in Tabular array two. There was not a significant difference between room air and hyperoxic air in the ERP response aamplitude. However, a difference in response latency was observed (Tabular array 2). Specifically, the ERP response during hyperoxia appears to exist slower than that during normoxia. This difference was noted in both the N1 (paired Student t test, P = 0.04) and P2 (P = 0.02) peaks.

Fig 5. Averaged ERP waveforms (N = thirteen) in response to a visual stimulus under room-air and hyperoxic conditions.

The ERP of the Oz channel is shown. N1 and P2 peaks of the ERP are conspicuously visible.

https://doi.org/x.1371/journal.pone.0176610.g005

Nosotros further examined the power spectrum of the ERP data. Information technology was institute that α and β signal power was significantly decreased past seven.7% ± iii.v% (P = 0.04) and 10.0% ± two.v% (P = 0.003), respectively, in the hyperoxic menses relative to normoxia. To further examine whether this effect was due to a power change in the evoked response itself or reflected a alter in the underlying spontaneous activity, we conducted an analysis similar to that used for the EEG data during the pre-task catamenia (500ms before the onset of the visual stimulus). Interestingly, during the pre-chore flow, α and β signal ability as well decreased (by 13.half dozen% ±iii.five%, P = 0.03, and xi.six% ± 3.6%, P = 0.03, respectively), suggesting that hyperoxia acquired a reduction in α and β power regardless of the presence of task simulation. The δ, θ, and γ bands did non testify any significant difference in either the pre-task or the job periods.

Response time did not testify a significance difference between normoxic and hyperoxic states.

Discussion

This study investigated the influence of hyperoxia on neural electrophysiological activity in the brain nether resting-state and during the operation of a visual job. Our data suggested that hyperoxia, via inhalation of O_ii-enriched gas, has a pronounced effect on neural activity. Specifically, hyperoxia decreased the α and β ring power of spontaneous neural activity. For task-evoked neural action, hyperoxia resulted in a slower appearance of ERP signal peaks, when compared to normoxia.

Oxygen therapy is widely used for several clinical conditions. Nevertheless, most of the weather condition in which oxygen treatment has demonstrated a definitive benefit are related to wound-healing in diabetes and cancer patients [2], but were non associated with brain function. Indeed, the proposed use of oxygen in brain diseases, such as autism, cerebral ischemia, brain injury, and cerebral palsy, is nevertheless controversial [3–seven,9]. Similarly, the issue of oxygen inhalation on cognitive performance is likewise a thing of debate [24–27]. Consequently, a neurobiological understanding of the impact (positive or negative) of O₂ on the brain is urgently needed.

Due to the intimate human relationship between oxygen and hemodynamic signals, conventional functional imaging techniques, such as BOLD fMRI, are contaminated by the directly effect of oxygen, and thus, are not suitable for the examination of neural activity changes under hyperoxia. Therefore, the number of human studies of the effects of hyperoxia on neural action is express. Several early studies have reported that hyperoxia does not alter resting-country EEG or evoked potentials [15,28]. With advances in neuroimaging technologies, recent evidence has suggested a potential effect of hyperoxia on neural activity. Xu et al. [13] utilized MRI-based methods to measure out metabolic responses to hyperoxia, and plant that hyperoxia with 98% FiO2 decreased the cognitive metabolic rate of oxygen (CMRO₂) by 10.3±1.5%. Richards et al. [29] measured oxygen metabolism using a ¹³C NMR technique and revealed that oxygen treatment in a canine model of ischemia reduced O₂ metabolism. Using MEG, Croal et al. plant that α, β, and lower γ bands in the occipital lobe showed a ability reduction of 3–v% [sixteen]. The observations of the present study are in full general understanding with these prior reports and showed that EEG rhythms associated with cerebral processing, such as the α and β bands, showed a diminishment under hyperoxic conditions. Our findings involve a larger portion of the brain compared to that studied by Croal et al., which was more localized to the occipital lobe [16]. This discrepancy could be due to the poorer spatial localization of the EEG technique, relative to 1000000. The changes observed in the present report appear to be reproducible and were noted in iii carve up experimental settings: 1) the resting land; 2) the interval between chore operation; and iii) during chore performance. Furthermore, the present study also examined evoked-related potentials and revealed that the response was delayed during hyperoxia. Collectively, these data provide farther support for a neural modulation issue of hyperoxia on the brain.

Our data suggest that the visual ERP is delayed past 1–ii ms. This divergence may be as well small to find in the response time, which is influenced past other factors, such as musculus response time. Thus, objective markers, such as EEG, may be more sensitive than behavioral measures, e.chiliad., response time, in terms of indexing the negative touch of hyperoxia on the brain.

The cellular and molecular underpinnings for the observed EEG changes need to be studied in future mechanistic investigations. Yet, i possible machinery is oxygen toxicity. It is known that up to 2% of the oxygen consumed by our body is only partially reduced, which produces reactive oxygen species (ROS) [30]. Excessive amounts of ROS in the body can cause oxidative stress [31,32]. A loftier O₂ content, i.due east., hyperoxia, presumably could increase the production of ROS [33,34]. In brain slices and animal models, hyperoxia has been shown to reduce neurotransmission [35] and suppress neural sensitivity to sensory stimulus [36]. However, most prior studies showing oxidative stress have used a longer period, e.yard., hours to days, of hyperoxia. Thus, it is unclear whether short-term hyperoxia on the order of minutes is sufficient to cause a detectable oxidative stress consequence. If not, then other mechanisms would demand to be identified to explicate the findings of this and other previous studies [13,16].

The findings regarding altered neural activity also have implications in hyperoxia-calibrated fMRI. Bold calibration using gas challenges, such as hyperoxia, has been used to judge job-evoked neurometabolic responses [37,38] and, more recently, to measure out baseline oxygen extraction fraction and cerebral metabolic charge per unit of oxygen [10–12]. If the resting-state EEG changes observed in the present written report are also accompanied by a metabolic suppression [13], this result should be incorporated in the calibrated fMRI model,s or, more than interestingly, used as an boosted variable in the parameter estimation [39]. If unaccounted for, the M factor and CMRO2 in hyperoxia-calibrated fMRI could both be overestimated.

The knowledge of how hyperoxia alters neural activity is likewise of significance in another surface area of public health. Millions of divers, also every bit cancer and diabetes patients, receive oxygen therapy for wound-healing [1,ii]. Information technology is, therefore, important to empathize and narrate the neurobiological effects of this hyperoxia-induced oxidative stress on brain function, even in the absence of clinical symptoms (eastward.thou., seizure). Tonic alpha power is generally thought to be positively correlated with cognitive abilities, such as retrieval of long-term retentiveness [40]. Therefore, if further verified, a reduction in alpha ability due to hyperoxia may suggest a potential detrimental issue of high concentration oxygen on memory. However, more spatially-specific techniques are needed to examine such changes in relevant brain regions, such equally the thalamo-cortical networks.

This study has a few limitations. Beginning, although EEG tin provide an assessment of brain neural action, it lacks the spatial resolution that MRI or MEG can provide. Thus, our findings in this written report showed minimal spatial heterogeneity in terms of the O₂ effect on neural action. It is possible that there is some spatial heterogeneity, simply our EEG technique could not finer observe it. A 2nd limitation is that our information provided limited mechanistic insight into the reasons for the observed changes. For instance, a measurement of the concentration of reactive oxygen species in the brain, e.g., by obtaining a lumbar puncture to draw cerebrospinal fluid (CSF) and use CSF every bit a surrogate for the encephalon tissue, before and during hyperoxia, would have been useful to ascertain the presence of oxidative stress during the short-term hyperoxia challenge applied in this study. A third limitation is that we studied only a unmarried hyperoxic country, but did not perform graded hyperoxia or hypoxia studies, unlike some prior studies [13,41,42].

In conclusion, this study showed that normobaric hyperoxia has a pregnant issue on human brain activity. Spontaneous neural activeness is suppressed in the α and β bands while event-related potentials are delayed in fourth dimension. These effects may exist relevant for calibrated fMRI, too as for the affect of oxygen therapy on the brain and cognition.

Supporting data

Acknowledgments

The authors are grateful to Dr. Jun Wang for providing some of the MATLAB scripts for the EEG data analysis and to Ms. Mary McAllister for copyediting the manuscript.

Author Contributions

1. Conceptualization: HL.
2. Data curation: MS PL DM.
3. Formal analysis: MS YG HL.
4. Funding acquisition: HL.
5. Investigation: MS PL DM.
6. Methodology: MS PL HL.
7. Project administration: MS PL DM YG HL.
8. Resources: MS PL DM.
9. Software: MS DM.
10. Supervision: HL.
11. Validation: HL.
12. Visualization: MS.
13. Writing – original draft: MS.
14. Writing – review & editing: HL.

References

1. ane. Levy SD, Alladina JW, Hibbert KA, Harris RS, Bajwa EK, Hess DR (2016) Loftier-menstruum oxygen therapy and other inhaled therapies in intensive care units. Lancet 387: 1867–1878. pmid:27203510
   • View Article
   • PubMed/NCBI
   • Google Scholar
2. 2. Elraiyah T, Tsapas A, Prutsky G, Domecq JP, Hasan R, Firwana B, et al. (2016) A systematic review and meta-analysis of adjunctive therapies in diabetic pes ulcers. J Vasc Surg 63: 46S–58S e41-42. pmid:26804368
   • View Article
   • PubMed/NCBI
   • Google Scholar
3. three. Calvert JW, Yin W, Patel M, Badr A, Mychaskiw Thousand, Parent AD, et al. (2002) Hyperbaric oxygenation prevented encephalon injury induced by hypoxia-ischemia in a neonatal rat model. Brain Res 951: 1–8. pmid:12231450
   • View Article
   • PubMed/NCBI
   • Google Scholar
4. 4. Singhal AB (2006) Oxygen therapy in stroke: past, present, and future. Int J Stroke 1: 191–200. pmid:18706016
   • View Article
   • PubMed/NCBI
   • Google Scholar
5. v. Singhal AB, Ratai East, Benner T, Vangel 1000, Lee V, Koroshetz WJ, et al. (2007) Magnetic resonance spectroscopy written report of oxygen therapy in ischemic stroke. Stroke 38: 2851–2854. pmid:17761914
   • View Article
   • PubMed/NCBI
   • Google Scholar
6. 6. Shoja MM, Tubbs RS, Shokouhi G, Loukas M, Ghabili K, Ansarin Yard (2008) The potential role of carbon dioxide in the neuroimmunoendocrine changes following cerebral ischemia. Life Sci 83: 381–387. pmid:18706431
   • View Article
   • PubMed/NCBI
   • Google Scholar
7. 7. Magnoni S, Ghisoni L, Locatelli 1000, Caimi M, Colombo A, Valeriani V, et al. (2003) Lack of improvement in cognitive metabolism after hyperoxia in astringent head injury: a microdialysis report. J Neurosurg 98: 952–958. pmid:12744353
   • View Article
   • PubMed/NCBI
   • Google Scholar
8. 8. Rockswold SB, Rockswold GL, Zaun DA, Zhang X, Cerra CE, Bergman TA, et al. (2010) A prospective, randomized clinical trial to compare the effect of hyperbaric to normobaric hyperoxia on cerebral metabolism, intracranial pressure, and oxygen toxicity in severe traumatic brain injury. J Neurosurg 112: 1080–1094. pmid:19852540
   • View Commodity
   • PubMed/NCBI
   • Google Scholar
9. ix. Rosenbaum P (2003) Controversial treatment of spasticity: exploring alternative therapies for motor function in children with cerebral palsy. J Child Neurol 18 Suppl 1: S89–94.
   • View Article
   • Google Scholar
10. 10. Wise RG, Harris AD, Stone AJ, Spud 1000 (2013) Measurement of OEF and absolute CMRO2: MRI-based methods using interleaved and combined hypercapnia and hyperoxia. Neuroimage 83: 135–147. pmid:23769703
    • View Article
    • PubMed/NCBI
    • Google Scholar
11. 11. Gauthier CJ, Hoge RD (2012) Magnetic resonance imaging of resting OEF and CMRO(2) using a generalized calibration model for hypercapnia and hyperoxia. Neuroimage 60: 1212–1225. pmid:22227047
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
12. 12. Bulte DP, Kelly Grand, Germuska M, Xie J, Chappell MA, Okell TW, et al. (2012) Quantitative measurement of cerebral physiology using respiratory-calibrated MRI. Neuroimage 60: 582–591. pmid:22209811
    • View Article
    • PubMed/NCBI
    • Google Scholar
13. 13. Xu F, Liu P, Pascual JM, Xiao G, Lu H (2012) Issue of hypoxia and hyperoxia on cerebral blood menses, blood oxygenation, and oxidative metabolism. J Cereb Blood Flow Metab 32: 1909–1918. pmid:22739621
    • View Article
    • PubMed/NCBI
    • Google Scholar
14. xiv. Sicard KM, Duong TQ (2005) Effects of hypoxia, hyperoxia, and hypercapnia on baseline and stimulus-evoked Bold, CBF, and CMRO2 in spontaneously breathing animals. Neuroimage 25: 850–858. pmid:15808985
    • View Article
    • PubMed/NCBI
    • Google Scholar
15. 15. Kaskinoro One thousand, Maksimow A, Laitio R, Scheinin H, Jääskeläinen South (2010) Normobaric hyperoxia does not induce pregnant electroencephalogram changes in healthy male subjects. European Periodical of Anaesthesiology 27: 121–122.
    • View Article
    • Google Scholar
16. 16. Croal PL, Hall EL, Commuter ID, Brookes MJ, Gowland PA, Francis ST (2015) The upshot of isocapnic hyperoxia on neurophysiology as measured with MRI and Million. Neuroimage 105: 323–331. pmid:25462687
    • View Article
    • PubMed/NCBI
    • Google Scholar
17. 17. Xu F, Uh J, Bramble MR, Hart J Jr, Yezhuvath U.s., Gu H, et al. (2011) The influence of carbon dioxide on brain activity and metabolism in conscious humans. J Cereb Blood Flow Metab 31: 58–67. pmid:20842164
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
18. 18. Lu H, Xu F, Rodrigue KM, Kennedy KM, Cheng Y, Flicker B, et al. (2011) Alterations in Cognitive Metabolic Rate and Claret Supply across the Adult Lifespan. Cereb Cortex 21: 1426–1434. pmid:21051551
    • View Article
    • PubMed/NCBI
    • Google Scholar
19. 19. Yezhuvath United states, Lewis-Amezcua Yard, Varghese R, Xiao G, Lu H (2009) On the assessment of cerebrovascular reactivity using hypercapnia BOLD MRI. NMR Biomed 22: 779–786. pmid:19388006
    • View Article
    • PubMed/NCBI
    • Google Scholar
20. twenty. Lu H, Liu P, Yezhuvath U, Cheng Y, Marshall O, Ge Y (2014) MRI mapping of cerebrovascular reactivity via gas inhalation challenges. J Vis Exp.
    • View Article
    • Google Scholar
21. 21. Kiloh LG, McComas AJ, osselton JW (1972) Clinical Electroencephalography: APPLETON-CENTURY-CROFTS.
22. 22. Lopez-Calderon J, Luck SJ (2014) ERPLAB: an open-source toolbox for the analysis of event-related potentials. Forepart Hum Neurosci 8: 213. pmid:24782741
    • View Article
    • PubMed/NCBI
    • Google Scholar
23. 23. Kiesel A, Miller J, Jolicoeur P, Brisson B (2008) Measurement of ERP latency differences: a comparison of single-participant and jackknife-based scoring methods. Psychophysiology 45: 250–274. pmid:17995913
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
24. 24. Malle C, Bourrilhon C, Quinette P, Laisney M, Eustache F, Pierard C (2016) Physiological and Cognitive Furnishings of Acute Normobaric Hypoxia and Modulations from Oxygen Breathing. Aerosp Med Hum Perform 87: three–12. pmid:26735227
    • View Article
    • PubMed/NCBI
    • Google Scholar
25. 25. Gill M, Natoli MJ, Vacchiano C, MacLeod DB, Ikeda 1000, Qin M, et al. (2014) Effects of elevated oxygen and carbon dioxide fractional pressures on respiratory role and cognitive operation. J Appl Physiol (1985) 117: 406–412.
    • View Article
    • Google Scholar
26. 26. Bloch Y, Applebaum J, Osher Y, Amar S, Azab AN, Agam 1000, et al. (2012) Normobaric hyperoxia handling of schizophrenia. J Clin Psychopharmacol 32: 525–530. pmid:22722511
    • View Article
    • PubMed/NCBI
    • Google Scholar
27. 27. Gao B, Long Z, Zhao L, He Yard (2011) Effect of normobaric hyperoxia on behavioral deficits and neuropathology in Alzheimer's disease mouse model. J Alzheimers Dis 27: 317–326. pmid:21799251
    • View Article
    • PubMed/NCBI
    • Google Scholar
28. 28. Smith DB, Strawbridge PJ (1974) Auditory and visual evoked potentials during hyperoxia. Electroencephalogr Clin Neurophysiol 37: 393–398. pmid:4136548
    • View Article
    • PubMed/NCBI
    • Google Scholar
29. 29. Richards EM, Fiskum G, Rosenthal RE, Hopkins I, McKenna MC (2007) Hyperoxic reperfusion afterward global ischemia decreases hippocampal energy metabolism. Stroke 38: 1578–1584. pmid:17413048
    • View Article
    • PubMed/NCBI
    • Google Scholar
30. thirty. Shadel GS, Horvath TL (2015) Mitochondrial ROS Signaling in Organismal Homeostasis. Prison cell 163: 560–569. pmid:26496603
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
31. 31. Zuo Fifty, Zhou T, Pannell BK, Ziegler Air-conditioning, All-time TM (2015) Biological and physiological function of reactive oxygen species—the good, the bad and the ugly. Acta Physiol (Oxf) 214: 329–348.
    • View Article
    • Google Scholar
32. 32. Federico A, Cardaioli E, Da Pozzo P, Formichi P, Gallus GN, Radi East (2012) Mitochondria, oxidative stress and neurodegeneration. J Neurol Sci 322: 254–262. pmid:22669122
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
33. 33. Dean JB, Mulkey DK, Henderson RA third, Potter SJ, Putnam RW (2004) Hyperoxia, reactive oxygen species, and hyperventilation: oxygen sensitivity of brain stem neurons. J Appl Physiol (1985) 96: 784–791.
    • View Commodity
    • Google Scholar
34. 34. D'Agostino DP, Putnam RW, Dean JB (2007) Superoxide (*O2-) production in CA1 neurons of rat hippocampal slices exposed to graded levels of oxygen. J Neurophysiol 98: 1030–1041. pmid:17553943
    • View Article
    • PubMed/NCBI
    • Google Scholar
35. 35. Bickford PC, Chadman One thousand, Williams B, Shukitt-Unhurt B, Holmes D, Taglialatela G, et al. (1999) Effect of normobaric hyperoxia on ii indexes of synaptic role in Fisher 344 rats. Free Radic Biol Med 26: 817–824. pmid:10232824
    • View Article
    • PubMed/NCBI
    • Google Scholar
36. 36. Shibata M, Blatteis CM (1991) High perfusate PO2 impairs thermosensitivity of hypothalamic thermosensitive neurons in piece preparations. Brain Res Bull 26: 467–471. pmid:1904788
    • View Article
    • PubMed/NCBI
    • Google Scholar
37. 37. Bulte DP, Chiarelli PA, Wise RG, Jezzard P (2007) Cerebral perfusion response to hyperoxia. J Cereb Blood Period Metab 27: 69–75. pmid:16670698
    • View Article
    • PubMed/NCBI
    • Google Scholar
38. 38. Chiarelli PA, Bulte DP, Wise R, Gallichan D, Jezzard P (2007) A calibration method for quantitative Bold fMRI based on hyperoxia. Neuroimage 37: 808–820. pmid:17632016
    • View Article
    • PubMed/NCBI
    • Google Scholar
39. 39. Driver ID, Wise RG, Murphy Thou (2016) Graded hypercapnia-calibrated Bold: Beyond the iso-metabolic hypercapnia assumption. In: Proceedings of ISMRM 24th Annual Meeting. Singapore. pp. 762.
40. 40. Klimesch W (1999) EEG alpha and theta oscillations reflect cognitive and memory performance: a review and analysis. Brain Res Brain Res Rev 29: 169–195. pmid:10209231
    • View Article
    • PubMed/NCBI
    • Google Scholar
41. 41. Smith ZM, Krizay E, Guo J, Shin DD, Scadeng M, Dubowitz DJ (2013) Sustained loftier-altitude hypoxia increases cerebral oxygen metabolism. J Appl Physiol (1985) 114: xi–xviii.
    • View Article
    • Google Scholar
42. 42. Driver ID, Whittaker JR, Vivid MG, Muthukumaraswamy SD, Murphy K (2016) Arterial CO2 Fluctuations Modulate Neuronal Rhythmicity: Implications for MEG and fMRI Studies of Resting-State Networks. J Neurosci 36: 8541–8550. pmid:27535903
    • View Commodity
    • PubMed/NCBI
    • Google Scholar

carmonaagall1988.blogspot.com

Source: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0176610

Share this post