Apneic Oxygenation- is it worth doing?

Apneic Oxygenation- Is it worth doing?

Review of Apneic Oxygenation during endotracheal intubation.

I first came across the concept of apneic oxygenation as an observer on a TEAM (training in emergency airway management) course several years ago. An emergency medicine registrar mentioned it as they were asked to justify their approach to a rapid sequence induction (RSI). This was rapidly poo-pooed by an anaesthetic consultant as complete rubbish. This was before I had discovered social media. However, this comment sparked my interest during the development of my advanced airway skills and came upon Weingart’s (2011) paper on pre oxygenation. Of course, I then discovered emcrit.org and Richard Levitan, and off I flew into the world of FOAMED.

As part of my MSc and critical care practitioner training, I produced a BestBets review on the subject in 2013. Since then, a number of papers produced from the emergency medicine and critical care fields have expanded upon the initial anaesthesia based studies. So here is my update.


Three part question.

In patients requiring endotracheal intubation, does the administration of nasal oxygen during the apneic period following induction increase the time to de-saturation?

Author, date and countryPatient groupStudy type (level of evidence)OutcomesKey resultsStudy Weaknesses
Baraka et al.
ASA 1-2 patients undergoing general anaesthesia for gastric band.

Patients received identical pre-oxygenation and induction.

A catheter was inserted nasally to the oral pharynx in both arms and 5L oxygen administered to treatment group. N=34
Randomised controlled trial.

Computer randomisation. Power analysis applied for minimum number of subjects (N=34).

Investigator blinded to nasal oxygen.
Desaturation to less than 95% or 4 minutes without desaturation.All control group de-saturated to less than 95%, mean time 145 seconds. 1 de-saturation in treatment group, remaining patients achieved 4 minutes without de-saturation.

No p-values applied.
Small study despite power calculation and single centre.

No application of statistical significance to oxygenation data. Study limitations not discussed.

Only 5L/min of O2 used for apneic oxygenation.

This population was elective and without significant physiologic pulmonary shunt.

Validity in application of anpnoeic O2 to patient groups with significant VQ mis-match can not be asserted.
South Korea
ASA 1-2 patients receiving general anaesthesia for ENT surgery.

Fibre optic endotracheal intubation. With and without 5L nasal oxygen via nasal cannula during apnea.

All intubations occurred within 4 minutes in both treatment arms.
Randomised controlled trial.


No description of randomisation.

No power analysis.
Comparison of heart rate, mean arterial pressure, PCO2 and PO2.Mean PaO2 between the control and treatment arms was 256(+/- 81) vs 345 (+/-78) after three minutes of apnea.Randomisation method not presented.

Outcome not clearly defined. SpO2 not used, makes comparison difficult.

No power analysis for minimum number of subjects.

No blinding.

No demographics provided.

No data on the number of patients randomised, but excluded from the study.

No discussion about ethical approval.

Study limitations not discussed.

This population was elective and without significant physiologic pulmonary shunt.

Validity in application of anpnoeic O2 to patient groups with significant VQ mis-match can not be asserted.
Mao and Qin, 2015.
N/AN/AN/AN/APaper available but retracted, possible research fraud.
Miguel-Montanes et al, 2014. France.Critical care unit.

RSI for multiple aetiologies. Non-rebreath mask pre oxygenation and 6L nasal apneic O2 O2 vs high flow nasal O2 pre oxygenation and apneic oxygenation.

Before and after study over an 18 month period comparing changes in pre oxygenation and apneic oxygenation standard operating procedure change.De-saturation of less than 80%.2% de-saturation less than 80% in high flow nasal oxygen vs 14% in the NRBM and 6L nasal O2. p.value of 0.03Biases of a before and after study.

Higher Cormack Lehane score and lower junior resident success rate in the non-rebreathe mask group may have confounded the results.

Severely hypoxaemic patients were excluded from the study.

Non-Elective patient group which may more closely represent the critically ill population group.
Patel and Nouraei, 2015. UK.Patients with known difficult airways undergoing elective ENT procedures. THRIVE technique (Transnasal Humidi- fied Rapid-Insufflation Ventilatory Exchange). Patient received a general anaesthetic but not intubated until the procedure was finished. N=25.Case series. Using the THRIVE technique (Transnasal Humidi- fied Rapid-Insufflation Ventilatory Exchange).No stated outcome.No patients experienced SpO2 less than 90%.

The median interquartile range for apnea was 14 minutes.
The range of apnea was 5-65 minutes.

2 patients did not require definitive airway. 4 patients received an LMA and 4 received tracheal intubation.1 tracheostomy and 14 received jet ventilation.
No comparator, but provides proof of principle that apneic oxygenation can significantly extend apnea time.

This population was elective and without significant physiologic pulmonary shunt.

Validity in application of apneic O2 to patent groups with significant VQ miss-match can not be asserted.
Ramachandran et al.
Deliberate prolonged laryngoscopy in obese patients.

Elective surgical patients receiving general anaesthesia. Exclusion numbers and reason for exclusion included.

5L/minute oxygen via nasal cannula used in treatment group.

All patients kept in deliberate grade 4 view during laryngoscopy N=30.
Randomised controlled trial.

Computer randomisation.

No blinding.

Power calculation applied (N=30).
Time to de-saturation <95% during apnea following rapid sequence induction and pre-oxygenation.Time to SpO2 <95% 3.49 minutes (no nasal O2) vs. 5.29 minutes (nasalO2) P = 0.001.Non-blinded.

Study limitations discussed in the paper.

This population was elective and without significant physiologic pulmonary shunt.

Validity in application of apneic O2 to patent groups with significant VQ miss-match can not be asserted.

Only 5L/min of O2 used for apneic oxygenation.
Sakles et al, 2016. USA.Adults with intracranial haemorrhage (trauma and medical causes) who required RSI. N= 127 (apneic O2 n=72/no apneic O2 n=55).Single-center observational cohort study of RSI over the 2-year period from 2013 to 2015 when AP OX was in use the emergency department.
Prospective data collection as part of ongoing quality improvement. Level 1 trauma centre.
Primary outcome of any Spo2 <90%. Secondary outomes of SpO2 <80% and <70%.In the first pass group 6 % de-saturated <90% in the apneic O2 group vs 20% in the no apneic O2 group.
In the moderate hypoxia group, 4% de-saturated to <80% in the apneic O2 group vs 18%.
In the severe hypoxia group, 3% desaturated to <70% vs 9%.
Multivariant analysis of adjusted odds ratio for de-saturation for patients receiving apneic O2 was 0.13 ( 95 % CI 0.03–0.53).
Higher first pass success in the apneic O2 group (93% vs 84%) despite more junior operators.
Inherent biases of an observational non-randomised study.

Variable intubation method (direct and video laryngoscopy) may confound results. More video laryngoscopy used in the apneic O2 group.

Differences between traumatic and non-traumatic causes may have influenced tendency for hypoxaemia.

Differences in seniority of operator between groups may confound results.

Variable flow rates of apneic O2 use may confound results.

It’s not clear if hypoxia caused first pass abandonment in the no apneic O2 group.

No apneic O2 group had lower lowest SpO2, this may reflect less effective pre oxygenation with NRBM alone rather than the effect of apneic O2.

Data is self reported by the operator and may over state SpO2 or CL view.
Sakles et al, 2016.
Emergency department of a level 1 trauma centre.

Adults requiring RSI for a variety of medical and trauma related reasons. N=635.
Single centre observational study using data from quality improvement database of emergency department intubations.

Comparison of patients that received apneic O2 vs those that did not.

80% power calculation to detect a 10% difference in de-saturation rate between groups, n=250. =380 (59%) in apneic O2 group vs 255 (49%).
First pass success without desaturation below 90%.In the apneic O2 group 8.8% de-saturated below 90% vs 16.2% in the no apnoeic O2 group (absolute difference7.4%; 95% CI -13.2% to - 1.6%). Inherent biases of an observational non-randomised study.

Data is self reported by the operator and may over state SpO2 or CL view.

Differences between traumatic and non-traumatic causes may have influenced tendency for hypoxaemia.

Variable flow rates of apneic O2 use may confound results.

A higher proportion of patients in the apneic O2 group were intubated using video laryngoscopy, this increased the first pass success rate and may be a confounder between the two groups.

Population representative of the critically ill.

Semler, Janz and Lentz, 2016.
N=150 critical care patients who required RSI for a number of reasons.Single centre, randomised, open-label, parallel-group, pragmatic trial.
Participants were also enrolled in to a second study comparing video vs direct laryngoscopy for intubation.
Comparing apneic oxygenation with usual care during endotracheal intubation of critically ill adults.No difference between apneic oxygenation (92%) versus standard care (90%) P = 0.16. Position of the patient for intubation was variable, semi-recumbent or supine.

Different intubation techniques were used, it was a duel video laryngoscope study, this may also have influenced results.

50% of patients were ventilated through the apneic period with either BVM or BIPAP.

Only 56 patients in total were not ventilated through apnea although there was roughly equal distribution manual ventilation between groups.

Some of the crucial confounders in this study are only mentioned in the online supplemental data.
Taha et al.
Elective surgery requiring general anaesthesia.

ASA grade 1 and 2 patients. Patients received identical pre-oxygenation and induction.

5L nasal oxygen administered in treatment arm. N=30.
Randomised controlled trial.

Computer randomisation.


Power analysis applied for minimum number of patients (N=30)..
Time to SpO2 <95% during apnea over a maximum 6 minute period.No O2 group , mean de-saturation 3.65 minutes .
In the apneic O2 group, no de-saturations over 6 minute period.
No blinding.

No application of statistical significance to oxygenation.

No p values applied to desaturation.

No discussion of study limitations.
Teller et al.
ASA 1-2 patients receiving general anaesthesia for elective surgery.

3L O2 via nasal cannula. Half of the patients studied post-induction, cross over following manual ventilation. N=12.
Cross-over of treatment arms. Patients acted as their own controls.

No power analysis.

Nasal cannula applied, but flow blinded to investigator.
SpO2 <92% or no de-saturation at 10 minutes of apnea.All patients achieved 10 minutes of apnea with SpO2 >96% with apneic O2 (p =0.001).

6.8 minutes was the mean time to de-saturation in the no apneic O2 group (p 0.001).
Re-oxygenation and hyperventilation to allow cross-over.

Only 3L O2 used.

Different pre-oxygenation method between cross over may have confounded results.

Not repeatable, unlikely to get ethical approval today.

No power analysis applied for minimum number of subjects.
Volteau et al, 2015. France.
N=124, adults requiring RSI for hypoxic respiratory failure, from 6 French intensive care units.

A mixture of surgical and medical intensive care units.

Comparison between high flow nasal oxygen and high flow face mask oxygen for pre oxygenation and apneic O2.

High flow face mask patients did not receive apneic oxygenation.
Multicenter, randomized, controlled, parallel, open-label trial.
Power calculation for 122 subjects to detect 6% difference in oxygen saturation.
Primary outcome was the difference in the lowest oxygen saturation level at the end of RSI.No statistical difference between the two groups. Interquartile range of oxygen saturation of 91.5 % in high flow nasal O2 versus the high flow face mask.group, 89.5 % p = 0.44. A well designed randomised controlled trial.

The high flow face mask patients actually received peep during their pre oxygenation period which may have provided greater de-nitorgenation and FRC recruitment.

The high flow nasal O2 group did not have their airway maintained during apnea which could obliterate any possible benefit from apneic oxygenation if the airway was closed.

Details regarding absence of airway manoeuvres and delivery of peep not published in the paper.
Wimalasena, Burns, Reid et al, 2015. Australia.Patients who received RSI in the pre-hospital environment by Sydney helicopter emergency medical service. N=728.A before and after observational study based on the introduction of apnoeic oxygenation.Oxygen saturation below 93% at any point during induction and intubation.In the preapneic oxygenation period 22% de-saturated below 93% vs 16% in the apneic oxygenation period.

6% absolute reduction in de-saturation 95% CI 0.2% to 11.2%.

Odds ratio for de-saturation 0.68 with 95% CI 0.47 to 0.98 comparing before and after apnoeic oxygenation.
Convenience sample.

No propensity matching.

Larger post apneic oxygenation population may skew results.

Potential improvements in practice over the time course of the study.

No description of the RSI process.

Very broad confidence interval.


 Apneic oxygenation has been used for decades in the assessment of brain stem death, maintaining oxygenation, but allowing carbon dioxide to rise, maximising the brain stem’s stimulus to breathe. Frumin et al (1959) showed that patients can remain oxygenated for greater than thirty minutes. However, this is usually achieved with insufflation of oxygen via a tracheal cannula.

Basic Physiology

The basic physiology behind apneic gas flow centres on the continuing consumption of oxygen during apnea, at resting and unstressed physiological states, this is around 250ml per minute. The gradient that this produces between the capillary and the alveolus ensures that alveolar oxygen continuously flows across the concentration gradient. As a result of this continuing gas flow from the alveolar space to the capillary, a negative pressure is generated in the alveolus. Where the airway is kept open, this negative alveolar pressure will continuously draw gas from the upper airway along this negative pressure gradient, apneic gas flow. Apneic oxygenation utilises this physiology to provide oxygenation during apnea.

Early versus later studies.

Early anaesthesia based studies conducted by Teller et al (1998), Taha et al (2006) and Lee et al (1988), although very small studies, clearly demonstrate that the time before desaturation during apnea can be significantly prolonged or negated altogether. More recently Patel has added to this evidence with an apnea time of up to 66 minutes. The debate from this point has been whether this technique is applicable to the non-anaesthetic environment, the critically ill and injured.  Where patients will likely have ventilation-perfusion miss-match. Miguel-Montanes et al (2014), Sakles et al (2016) and Wimalasena  et al (2015) observational, before and after studies suggested that apneic oxygenation reduced desaturation events across an array of critically ill patients. Despite the size of these cohorts, this is not high-level evidence and the limitations and biases of these kinds of observational studies have to be acknowledged.

What do the RCTs say?

Now we come to the randomised controlled trials (RCT). We now have two RCT’s in critically ill patients, showing no effect. Is this the definitive answer? No. Unfortunately, the significant confounders for these studies are not in the primary papers. Semler’s et al (2016) study was not a true apneic oxygenation study as only 50 patients were not ventilated through the apneic period. There is no subgroup analysis of the “non-ventilated” patients. The on-line supplement for this paper is essential reading to critique this paper. Semler’s investigation tells us that apneic oxygenation does not offer benefit if the patient is ventilated through apnea, and has not answered our question. Vourc’h et al (2015) study has also not answered the question. The “no apneic oxygenation” group received PEEP and this strategy may have accounted for the apneic oxygen benefit by better functional residual capacity recruitment and denitrogenation. More importantly, Vourc’h et al (2015) apneic oxygenation patients did not have their airways maintained open during apnea. Clearly apneic oxygenation delivered to a closed airway prevents any apneic oxygenation. These issues are not evident in the Vourc’h paper, but were high lighted by Scott Weingart of emcrit.org when he interviewed the investigators (https://itunes.apple.com/gb/podcast/emcrit-podcast-critical-care/id314020330?mt=2&i=345966308).

So, we have high-level evidence that apneic oxygenation works in the stable elective patient. Low level evidence (with good patient numbers) suggests we can get an absolute reduction of desaturation of around six percent in the critically ill and injured population. We have high level evidence that significantly muddies the waters.

What are the Harms?

What are the harms? High flow nasal oxygen systems are now becoming commonly employed in critical care and use flows of 40 litres or more. It seems unlikely that there is a significant down side to apneic oxygenation using 15 litres. The only harms I have found in my literature searches are of gastric rupture when nasal oxygen is used with nasal trumpets on a closed airway, there are only two or three reports of this that I have found. Is apneic oxygenation a distraction, from more important interventions. I invariably find that nasal oxygen has often already been used on a patient requiring intubation and is very quickly and easily applied during the preparation for RSI. The technique is low risk, cheap, and in all likelihood effective.

Does apneic oxygenation give the operator a false sense of security? Personally, preparation for RSI is about maximising first-pass success and building redundancy into the procedure. Apneic oxygenation is not a panacea of safe RSI. However, using small percentage gains adds redundancy to a dangerous procedure. Head up, PEEP pre-oxygenation, bougie, and gentle low volume apneic ventilation, end tide carbon-dioxide monitoring, check lists and apneic oxygenation.

Bottom Line.

Apnoeic oxygenation can extend the time to apneic desaturation and reduce desaturation episodes during endotracheal intubation.

Other posts on the web:

St Emlyns- JC: The last breath for apnoeic oxygenation?

LITFL- Apneic oxygenation

PHARM Podcast 132 Apnoeic Oxygenation in Emergency Intubations with Dr John Sakles


EMergucate- FOAM Eye-Catchers 10: Apnoeic Oxygenation – new trial questions value of NODESAT


Baraka, A.S., Taha, S.K., Siddik-Sayyid, S.M., Kanazi, G.E., El-Khatib, M.F., Dagher, C.M., Chehade, J.-M.A., Abdallah, F.W., and Hajj, R.E. (2007) ‘Supplementation of Pre-Oxygenation in Morbidly Obese Patients Using Nasopharyngeal Oxygen Insufflation’. Anaesthesia 62 (8), 769–773

Frumin, M.J., EPSTEIN, R.M., and COHEN, G. (1959) ‘Apneic Oxygenation in Man.’. Anesthesiology 20, 789–798

Lee, S.C. (1998) ‘Improvement of Gas Exchange by Apneic Oxygenation with Nasal Prong During Fiberoptic Intubation in Fully Relaxed Patients.’. Journal of Korean medical science 13 (6), 582–586

Mao, Y. and Qin, Z.-H. (2015) ‘Association of Apneic Oxygenation with Decreased Desaturation Rates During Rapid Sequence Intubation by a Chinese Emergency Medicine Service.’. International journal of clinical and experimental medicine 8 (7), 11428–11434

Miguel-Montanes, R., Hajage, D., Messika, J., Bertrand, F., Gaudry, S., Rafat, C., Labbé, V., Dufour, N., Jean-Baptiste, S., Bedet, A., Dreyfuss, D., and Ricard, J.-D. (2014) ‘Use of High-Flow Nasal Cannula Oxygen Therapy to Prevent Desaturation During Tracheal Intubation of Intensive Care Patients with Mild-to-Moderate Hypoxemia’. Critical Care Medicine 1

Patel, A. and Nouraei, S. (2014) ‘Transnasal Humidified Rapid‐Insufflation Ventilatory Exchange (THRIVE): a Physiological Method of Increasing Apnoea Time in Patients with Difficult Airways’. Anaesthesia

Ramachandran, S.K., Cosnowski, A., Shanks, A., and al, E. (2010) ‘Apneic Oxygenation During Prolonged Laryngoscopy in Obese Patients: a Randomized, Controlled Trial of Nasal Oxygen Administration’. Journal of Clinical Anesthesia 22 (3), 164–168

Sakles, J.C., Mosier, J., Patanwala, A.E., Arcaris, B., and Dicken, J. (2016a) ‘First Pass Success Without Hypoxemia Is Increased with the Use of Apneic Oxygenation During RSI in the Emergency Department’. Academic Emergency Medicine n/a–n/a

Sakles, J.C., Mosier, J.M., Patanwala, A.E., and Dicken, J.M. (2016b) ‘Apneic Oxygenation Is Associated with a Reduction in the Incidence of Hypoxemia During the RSI of Patients with Intracranial Hemorrhage in the Emergency Department.’. Internal and emergency medicine

Semler, M.W., Janz, D.R., Lentz, R.J., Matthews, D.T., Norman, B.C., Assad, T.R., Keriwala, R.D., Ferrell, B.A., Noto, M.J., McKown, A.C., Kocurek, E.G., Warren, M.A., Huerta, L.E., Rice, T.W., FELLOW Investigators and the Pragmatic Critical Care Research Group (2016) ‘Randomized Trial of Apneic Oxygenation During Endotracheal Intubation of the Critically Ill.’. American Journal of Respiratory and Critical Care Medicine 193 (3), 273–280

Taha, S.K., Siddik-Sayyid, S.M., El-Khatib, M.F., Dagher, C.M., Hakki, M.A., and Baraka, A.S. (2006) ‘Nasopharyngeal Oxygen Insufflation Following Pre-Oxygenation Using the Four Deep Breath Technique’. Anaesthesia 61 (5), 427–430

Teller, L.E., Alexander, C.M., Frumin, M.J., and Gross, J.B. (1988) ‘Pharyngeal Insufflation of Oxygen Prevents Arterial Desaturation During Apnea’. Anesthesiology 69 (6), 980

Vourc’h, M., Asfar, P., Volteau, C., Bachoumas, K., Clavieras, N., Egreteau, P.-Y., Asehnoune, K., Mercat, A., Reignier, J., Jaber, S., Prat, G., Roquilly, A., Brule, N., Villers, D., Bretonniere, C., and Guitton, C. (2015) ‘High-Flow Nasal Cannula Oxygen During Endotracheal Intubation in Hypoxemic Patients: a Randomized Controlled Clinical Trial’. Intensive Care Medicine 41 (9), 1538–1548

Weingart, S.D. and Levitan, R.M. (2012) ‘Preoxygenation and Prevention of Desaturation During Emergency Airway Management’. Annals of Emergency Medicine 59 (3), 165–75.e1

Wimalasena, Y., Burns, B., Reid, C., Ware, S., and Habig, K. (2015) ‘Apneic Oxygenation Was Associated with Decreased Desaturation Rates During Rapid Sequence Intubation by an Australian Helicopter Emergency Medicine Service’. Annals of Emergency Medicine 65 (4), 371–376


Gavin DentonGavin Denton @dentongavin – Current role: Critical care practitioner, critical care, West Midlands. Roles include; assessment and management of the critically ill patient, insertion of invasive lines, advanced airway management (under supervision), transfer of the critically ill patient, resuscitation (from airway, to team leader to post resus care). Teaching and support of junior doctors of the above.
Graduated from the University of Birmingham with BN(hons). BSc from Birmingham City University. About to complete MSc in health sciences from the University of Warwick.
Working background: 15 years working within various aspects of critical care. 7 years in critical care, 6 years in critical care outreach, 2 years as a critical care practitioner. Adult life support instructor. Independent non-medical prescriber.
Future aims: faculty of critical care medicine affiliation. FEEL course, POCUS training.
Clinical interests: USS, airway management.

Leave a comment

Your email address will not be published. Required fields are marked *