Secondary Logo

Journal Logo

Effect of High Inspiratory Oxygen Fraction on Endothelial Function in Healthy Volunteers: A Randomized Controlled Crossover Pilot Study

Larsen, Mikkel Hjordt Holm MD*; Ekeloef, Sarah MD*; Kokotovic, Dunja MD*; Schou-Pedersen, Anne-Marie PhD; Lykkesfeldt, Jens PhD, DMSc; Gögenür, Ismail MD, DMSc*

doi: 10.1213/ANE.0000000000002357
Basic Science

It has been suggested that high inspiratory oxygen concentrations during anesthesia may be associated with higher postoperative mortality due to endothelial dysfunction. A randomized controlled crossover study was conducted with 25 healthy male volunteers. They inhaled an oxygen concentration of 30% and 80%. The endothelial function was assessed using noninvasive digital pulse amplitude tonometry (EndoPAT) supported by endothelial biomarkers. The difference in endothelial function between the 2 treatments was 0.05 (95% confidence interval, −0.36 to 0.27; P = .77). Endothelial biomarkers were unaffected. Inhalation of a high oxygen fraction in healthy volunteers did not result in a significant reduction of endothelial function.

From the *Department of Surgery, Center for Surgical Science, Zealand University Hospital, Koege, Denmark; and Faculty of Health and Medical Sciences, University of Copenhagen, Frederiksberg, Denmark.

Accepted for publication June 8, 2017.

Funding: This study received a grant of $285,943 by The A.P. Møller Foundation for the Advancement of Medical Science.

The authors declare no conflicts of interest.

Reprints will not be available from the authors.

Address correspondence to Mikkel Hjordt Holm Larsen, MD, Department of Surgery, Zealand University Hospital, Lykkebaekvej 1, 4600 Koege, Denmark. Address e-mail to

A recent clinical study has shown a 2.9-fold increase in the risk of myocardial infarction among patients undergoing abdominal surgery with inhalation of a high oxygen fraction during the procedure.1 Similar findings has been shown among intensive care unit (ICU) patients, where a higher mortality was reported among patients receiving hyperoxia treatment compared with conservative oxygen treatment.2 The higher mortality may in part be caused by oxidative stress resulting in endothelial dysfunction.3

However, the putative relationship between high inspiratory oxygen fraction and development of endothelial dysfunction is poorly studied. Nitric oxide (NO) is crucial in the regulation and homeostasis of the endothelium, and inhibition of the pathway by oxidative stress may lead to endothelial dysfunction.4

Here, we conducted a pilot study, which aimed to investigate the effect of high versus low inspiratory oxygen fraction on the endothelial function measured by noninvasive digital pulse amplitude tonometry (EndoPat) and biomarkers reflecting NO bioavailability among healthy male volunteers.

Back to Top | Article Outline


Study approval was obtained from the regional ethical committee (SJ-409) and the Danish Data Protection Agency (REG-109–2014). The study was registered at (NCT-02342405).

We conducted a randomized controlled crossover study on 25 healthy male volunteers between 18 and 30 years. Participants were excluded if they were smokers/ex-smokers, took daily medications, had a history of atopic dermatitis or arteriosclerotic diseases among first-/second-degree relatives, known familiar hypercholesterolemia, or any known or unknown arrhythmias measured by baseline heart rate variability recording.

The randomization scheme was generated at the website (, and each participant was allocated a randomization number and an oxygen treatment sequence, either 80% and then 30% oxygen or 30% and then 80% oxygen on test days 1 and 2, respectively. Each participant did only receive each treatment once. The investigators and participants were blinded to the oxygen treatment. Allocations were unblinded after all data were collected, verified, and analyzed.

Participants were instructed not to perform strenuous exercise for 48 hours before the trial and not to consume any alcohol or caffeine for 24 and 12 hours before the trial, respectively. The study took place in a quiet room after an overnight fast. Participants were given low-fat yogurt breakfast with fruits at the beginning of each trial day. The participants were tested at the same hour on 2 separate days with minimum 14 days between to avoid any potential carryover effect.

At the beginning of each trial day, a cubital intravenous catheter was placed, in the arm opposite the EndoPat measurement, and baseline blood samples were drawn. Afterward, the oxygen treatment was started. The oxygen intervention was delivered via a facemask (High Concentration Oxygen Mask; Intersurgical Ltd, Wokingham, United Kingdom). The oxygen was mixed with ambient air delivered through a flowmeter in flows of 14 L/min O2 + 2 L/min air in the 80% group and 2 L/min O2 + 14 L/min air in the 30% group. The planned administration of the oxygen inhalation was 165 minutes each day. The measurement of endothelial function by EndoPat was performed after a 120-minute preoxygenation period. The oxygen inhalation was terminated 30 minutes after the EndoPat measurement.

The EndoPat system (Itamar Medical, Caesarea, Israel) is a noninvasive system. The EndoPat measurement consisted of a 5-minute baseline measurement followed by 5 minutes of forearm ischemia before the final 5-minute reperfusion measurement. The endothelial function is expressed as the reactive hyperemia index (RHI) and has been proven to be compatible with endothelial dysfunction.5

NO bioavailability was measured with plasma l-arginine, plasma asymmetric dimethylarginine (ADMA), plasma tetrahydrobiopterin (BH4), plasma dihydrobiopterin (BH2), and total plasma biopterin. All blood samples were drawn in K2-EDTA tubes.

BH4, BH2, and total plasma biopterin samples were drawn after 120 minutes of preoxygenation and 30, 60, 120, and 240 minutes after forearm reperfusion. l-Arginine and ADMA samples were drawn at baseline and 240 minutes after forearm reperfusion.

Back to Top | Article Outline

Blood Samples


One milliliter of blood was mixed with 25 μL dithioerythritol and centrifuged at 4°C, 2000 RPM for 5 minutes. Plasma was snap-frozen after centrifugation and stored at −80°C until analyses. Plasma was analyzed using high-performance liquid chromatography.6

Back to Top | Article Outline

l-Arginine and ADMA.

The blood sample was centrifuged at 4°C, 3410 RPM for 10 minutes and plasma was snap-frozen and stored at −80°C until analysis. High-performance liquid chromatography with fluorescence was used to determine the level of ADMA and l-arginine,7 but with the following alterations: The plasma sample was deposited of protein with cold trichloroacetic acid and afterward neutralized in advance of isocratic elution of l-arginine and ADMA on a Phenomenex Gemini C18 column (150 × 4.6 mm; 5 μm; Phenomenex, Torrance, CA). The mobile phase consisted of 50 mM potassium phosphate (pH 6.5) with 14% (v/v) acetonitrile and was pumped with a flow rate of 1.2 mL/min in the mobile phase. The levels of l-arginine and ADMA were quantified at excitation and emission wavelengths of 340 and 455 nm, respectively.

Back to Top | Article Outline


All analyses were performed using SAS University Edition v.3.1. After visual inspection of histograms and QQ-plots, all continuous variables except the biopterin metabolites were log-transformed. Tetrahydrobiopterin, dihydrobiopterin, their ratio, as well as total plasma biopterin measurements were baseline adjusted before mixed model analyses. Differences in RHI, arginine, and ADMA levels were analyzed using paired t test. All analyses were conducted with the assumption of no carryover effect, adjusted for test sequence, and analyzed for carryover effect. The statistical analyses were considered significant when P < .05.

Back to Top | Article Outline

Sample Size

We suspected a difference of 15% between the low and high oxygen treatment could be clinically relevant corresponding to an RHI of 1.55 in the high oxygen treatment. With a 5% risk of a type I error and a power of 80%, this resulted in a necessary n value of 22 in this crossover design.8 We therefore chose to include 25 subjects.

Sample size was based on data showing a mean RHI of 1.82 with a within subject SD of 0.39 and an assumption of intraclass correlation of 0.74.8

Back to Top | Article Outline


Mean age of the participants was 24 years (SD ±2.26), and mean body mass index was 25 (SD ±3.14). Mean systolic/diastolic blood pressure at the 30% oxygen day and the 80% oxygen day were 129 (SD ±12.02)/71 (SD ±8.96) and 127 (SD ±8.70)/71 (SD ±5.91), respectively. All participants had all blood samples taken.

RHI in the 80% oxygen group was 1.83 (SD 0.47) and 1.88 (SD 0.60) in the 30% oxygen group (P = .85, Figure 1). The difference in endothelial function between the 2 treatments was 0.05, but statistically insignificant (95% confidence interval, −0.36 to 0.27; P = .77). The change in ADMA and l-arginine from baseline to 240 minutes after tourniquet deflation was 0.004 (SD 0.06) and 2.05 (SD 16.13), respectively. This was an insignificant difference between the 30% and 80% oxygen treatments (Figure 1).

Figure 1.

Figure 1.

Figure 2.

Figure 2.

BH4 and total plasma biopterin decreased significantly during inhalation of 80% oxygen (Figure 2), but inhalation with 80% oxygen did not significantly change the level of BH4 and total plasma biopterin over time when compared with 30% oxygen inhalation, P = .75 and P = .88. BH2 and the ratio of BH2/BH4 was unaffected by both 30% oxygen and 80% oxygen inhalation.

Back to Top | Article Outline


In this randomized crossover study on healthy volunteers, endothelial function did not differ significantly after inhalation of 80% vs 30% oxygen.

We investigated the effect of high inspiratory oxygen fraction on the endothelial function to explore the potential detrimental effect of hyperoxia on an endothelial level. This has not previously been done. However, since the study was executed in healthy volunteers, the results cannot directly be applied to critically ill patients with various physiologic stresses and inflammatory conditions. The clinical effect of hyperoxia has been studied in a variety of critically ill patients including patients in the ICU,2 patients experiencing traumatic brain injury,9 and patients undergoing abdominal surgery.1 Hyperoxia treatment of critically ill ICU patients has in a recent study been associated with an increased mortality.2 Among patients with traumatic brain injury treated with hyperoxia a worse short-term functional outcome and higher fatality was described,9 and lately an increased risk of myocardial infarction has been reported in abdominal surgery patients treated with hyperoxia.1

Hyperoxia and physiological stress could act in a synergistic way in the development of endothelial dysfunction.

To improve our model, we suggest adding an additional external stressor to mimic the pathophysiologic stress. This could be high intense strenuous training where radical oxygen species formation is present.10

The power of the study and sample size was based on a 5% risk of a type I error and a power of 80%, and the effect on RHI might have been to optimistic thereby resulting in a type II error. Also, we chose an AB/BA design, but a stronger design would have been ABBA/BAAB where each subject received the treatment twice.

In conclusion, this study could not confirm that high inspiratory oxygen fraction alone causes endothelial dysfunction in healthy volunteers. Future studies should investigate the potential synergistic effect between high inspiratory oxygen fraction and additional external stressors.

Back to Top | Article Outline


The authors thank Pierre Bouchelouche, MD, Head of Department, Department of Biochemistry, Zealand University Hospital, Lykkebaekvej 1, 4600 Koege, Denmark, for the use of laboratory equipment and sample analyses.

Back to Top | Article Outline


Name: Mikkel Hjordt Holm Larsen, MD.

Contribution: This author was responsible for the trial and helped work the protocol, conduct the trial, analyze the statistics, and write the manuscript.

Name: Sarah Ekeloef, MD.

Contribution: This author was responsible for the trial and helped work the protocol, assist trial conduction, and revise the manuscript.

Name: Dunja Kokotovic, MD.

Contribution: This author helped assist trial conduction and revise the manuscript.

Name: Anne-Marie Schou-Pedersen, PhD.

Contribution: This author helped analyze the endothelial biomarkers and revision of the manuscript.

Name: Jens Lykkesfeldt, PhD, DMSc.

Contribution: This author helped analyze the endothelial biomarkers and revision of the manuscript.

Name: Ismail Gögenür, MD, DMSc.

Contribution: This author was the initiator and responsible for the trial, helped work the protocol, and revise the manuscript.

This manuscript was handled by: Alexander Zarbock, MD.

Back to Top | Article Outline


1. Fonnes S, Gögenur I, Søndergaard ES. Perioperative hyperoxia—long-term impact on cardiovascular complications after abdominal surgery, a post hoc analysis of the PROXI trial. Int J Cardiol. 2016;215:238–243.
2. Girardis M, Busani S, Damiani E, et al. Effect of conservative vs conventional oxygen therapy on mortality among patients in an intensive care unit: the oxygen-ICU randomized clinical trial. JAMA. 2016;316:1583–1589.
3. Meyhoff CS, Jorgensen LN, Wetterslev J, Christensen KB, Rasmussen LS; PROXI Trial Group. Increased long-term mortality after a high perioperative inspiratory oxygen fraction during abdominal surgery: follow-up of a randomized clinical trial. Anesth Analg. 2012;115:849–854.
4. Park KH, Park WJ. Endothelial dysfunction: clinical implications in cardiovascular disease and therapeutic approaches. J Korean Med Sci. 2015;30:1213–1225.
5. Rubinshtein R, Kuvin JT, Soffler M, et al. Assessment of endothelial function by non-invasive peripheral arterial tonometry predicts late cardiovascular adverse events. Eur Heart J. 2010;31:1142–1148.
6. Mortensen A, Hasselholt S, Tveden-Nyborg P, Lykkesfeldt J. Guinea pig ascorbate status predicts tetrahydrobiopterin plasma concentration and oxidation ratio in vivo. Nutr Res. 2013;33:859–867.
7. Teerlink T, Nijveldt RJ, de Jong S, van Leeuwen PA. Determination of arginine, asymmetric dimethylarginine, and symmetric dimethylarginine in human plasma and other biological samples by high-performance liquid chromatography. Anal Biochem. 2002;303:131–137.
8. McCrea CE, Skulas-Ray AC, Chow M, West SG. Test-retest reliability of pulse amplitude tonometry measures of vascular endothelial function: implications for clinical trial design. Vasc Med. 2012;17:29–36.
9. Brenner M, Stein D, Hu P, Kufera J, Wooford M, Scalea T. Association between early hyperoxia and worse outcomes after traumatic brain injury. Arch Surg. 2012;147:1042–1046.
10. Balakrishnan SD, Anuradha CV. Exercise, depletion of antioxidants and antioxidant manipulation. Cell Biochem Funct. 1998;16:269–275.
Copyright © 2017 International Anesthesia Research Society