Conflict of Interest Declaration
The ILCOR Continuous Evidence Evaluation process is guided by a rigorous ILCOR Conflict of Interest policy. The following Task Force members and other authors were recused from the discussion as they declared a conflict of interest: none applicable
The following Task Force members and other authors declared an intellectual conflict of interest and this was acknowledged and managed by the Task Force Chairs and Conflict of Interest committees: none applicable
CoSTR Citation
Chang WT, Sakamoto T, Lee CC, Singletary EM, Zideman D, Schmölzer G, Bendall J, Berry DC, Borra V, Carlson JN, Cassan P, Charlton NP, Djarv T, Douma M, Epstein J, Hood NA, Markenson D, Meyran D, Orkin A, Swain J, Woodin JA, Morley P: First Aid Supplementary Oxygen in Acute Stroke Consensus on Science with Treatment Recommendations [Internet] Brussels, Belgium: International Liaison Committee on Resuscitation (ILCOR) First Aid Task Force, 1 Jan 2020. Available from http://ilcor.org
Methodological Preamble and Link to Published Systematic Review
The continuous evidence evaluation process for the production of this Consensus on Science with Treatment Recommendations (CoSTR) started with a systematic review of first aid supplementary oxygen for acute stroke (Chang, 2019 – PROSPERO registration ID 162958, submitted 16 December 2019)conducted by First Aid Task Force systematic reviewers (WT Chang and T Sakamoto) with assistance from clinical content experts and ILCOR information specialists. Evidence for adult literature was sought and considered by the First Aid Task Force.
Systematic Review
Webmaster to insert the Systematic Review citation and link to PubMed using this format when it is available.
PICOST
The PICOST (Population, Intervention, Comparator, Outcome, Study Designs and Timeframe):
Population: Adults with suspected acute stroke
Intervention: Use of supplementary oxygen
Comparators: No use of supplementary oxygen
Outcomes: Clinical outcomes such as survival, neurological outcomes (e.g. NIHSS, Scandinavian stroke scale, modified Rankin scale score, etc.), and neurological recovery in the acute phase were ranked as critical outcomes. Quality of life (e.g. Barthel index, EuroQol, Nottingham ADL score, etc.) and hospital length of stay were ranked as important outcomes. Adverse effects and complications (pneumonia, pulmonary edema, necessity of non-invasive positive pressure ventilation, intubation with mechanical ventilation, etc.) were listed as important outcomes. Imaging outcomes such as magnetic resonance imaging (MRI) indicators (diffusion-weighted imaging, lesion volume, diffusion/perfusion mismatch, magnetic resonance spectroscopic indicators, etc.) and reperfusion rate were ranked as important outcomes. Laboratory outcomes such as oxygen saturation (highest, lowest, incidence or duration of oxygen saturation < 90% or 95%, etc.) were listed as good-to-know outcomes.
Study Designs: Randomized controlled trials (RCTs) and non-randomized studies (non-randomized controlled trials, interrupted time series, controlled before-and-after studies, cohort studies) are eligible for inclusion. Unpublished studies (e.g., conference abstracts, trial protocols) were excluded.
Timeframe: All years and all languages were included; unpublished studies (e.g., conference abstracts, trial protocols) were excluded. Literature search updated to Oct 16, 2019.
PROSPERO Registration 162958, submitted on 16 December 2019
NOTE FOR RISK OF BIAS: In most cases bias was assessed per comparison rather than per outcome, since there were no meaningful differences in bias across outcomes. In cases where differences in risk of bias existed between outcomes this was noted.”
Consensus on Science
For the critical outcome of survival at one week, we identified moderate certainty evidence (downgraded for indirectness) from one randomized controlled trial {Roffe 2017 1206} recruiting 8003 adults with acute stroke showing no difference between use of continuous supplementary oxygen at 2-3 L/min via nasal cannula for 72h and use of room air (oxygen only if clinically indicated) (risk ratio 1.00, 95% CI 0.99-1.01).
For the critical outcome of survival at 3 months, we identified moderate certainty evidence (downgraded for indirectness) from one randomized controlled trial {Roffe 2017 1206} recruiting 8003 adult patients with acute stroke showing no difference between use of continuous supplementary oxygen at 2-3 L/min via nasal cannula for 72h (n=2668) and use of room air (n=2668) (risk ratio 1.0, 95% CI 0.98-1.01).
For the critical outcome of survival at 6 months, we identified moderate certainty evidence (downgraded for indirectness) from one randomized controlled trial {Ali 2014 937} recruiting 289 adult patients with acute stroke showing no difference between use of supplementary oxygen at 2-3 L/min via nasal cannula for 72h and use of room air (risk ratio 1.0, 95% CI 0.91-1.10).
For the critical outcome of survival at one year, we identified low certainty evidence (downgraded for risk of bias and indirectness) from one randomized controlled trial {Ronning 1999 408} recruiting 550 adult patients with acute stroke showing no difference between use of supplementary oxygen via nasal cannula at 3 L/min for 24h and use of room air (risk ratio 0.95, 95% CI 0.85-1.05).
For the critical neurological outcome of neurological outcome of National Institute of Health Stroke Scale (NIHSS) at one week, we identified low certainty evidence (downgraded for indirectness) from 5 randomized controlled trials {Ali 2014 937; Padma 2010 840; Singhal 2005 1035; Roffe 2017 1206; Roffe 2011 1297} recruiting 5969 adult patients with acute stroke showing no difference between use of supplementary oxygen at 2-4 L/min via nasal cannula, the use of facemask for 8-72h and the use of room air (absolute difference 0 points, 95% CI, −0.01 to 0.01).
For the critical neurological outcome of National Institute of Health Stroke Scale (NIHSS) at 3 months, we identified very low certainty evidence (downgraded for risk of bias, indirectness and imprecision) from 2 randomized controlled trial {Padma 2010 840; Singhal 2005 1035} recruiting 54 adult patients with acute stroke showing no difference between use of supplementary oxygen at 10-45 L/min via facemask for 8-12h and use of room air (oxygen only if clinically indicated) (absolute difference 0.62 points lower, 95% CI, −2.79 to 1.56).
For the critical neurological outcome of National Institute of Health Stroke Scale (NIHSS) difference between baseline and one week, we identified moderate certainty evidence (downgraded for indirectness) from one randomized controlled trial {Roffe 2011 1297} recruiting 289 adult patients with acute stroke showing no difference between use of continuous supplementary oxygen via nasal cannula at 2-3 L/min for 72h and use of room air (absolute difference 1.00 point lower, 95% CI −2.83 to 0.83).
For the critical neurological outcome of improvement of National Institute of Health Stroke Scale (NIHSS) score of more than 4 at one week, we identified moderate certainty evidence (downgraded for indirectness) from one randomized controlled trial {Roffe 2011 1297} recruiting 289 adult patients with acute stroke showing that the patients receiving supplementary oxygen at 2-3 L/min via nasal cannula for 72h had higher chance of NIHSS improvement of more than 4 at one week as compared to those breathing room air (risk ratio 2.19, 95% CI 1.37 to 3.51).
For the critical neurological outcome of favorable modified Rankin score (mRS) at hospital discharge, we identified very low certainty of evidence (downgraded for risk of bias) from one retrospective observational study {Dylla 2019} recruiting 1352 patients with acute stroke showing no difference between the patients without hypoxia at baseline receiving prehospital supplementary oxygen and those breathing room air (relative risk 1.06, 95% CI 0.84-1.33). The dosage of supplementary oxygen was not provided in this study.
For the critical neurological outcome of modified Rankin sore (mRS) at 3 months, we identified moderate certainty evidence (downgraded for indirectness) from three RCTs {Roffe 2017 1206; Singhal 2015 1035; Padma 2010 284}. The large RCT by Roffe et al. {Roffe 2017 1206} recruiting 8003 individuals showed no difference in mRS score for the group receiving supplementary oxygen at 2-3 L/min via nasal cannula for 72h and the group receiving room air (odds ratio 0.97, 95% C.I. 0.89 to 1.05). The RCT by Singhal (Singhal 2015 1035) recruiting 16 patients with acute stroke, found no difference in mRS score in the group receiving supplementary oxygen at 45 L/min by facemask for 8h compared with the group receiving room air (absolute difference 0.90 points higher, 95% CI −2.84 to 1.04). Oxygen was delivered if clinically indicated in the study by Singhal et al.
For the critical neurological outcome of modified Rankin sore (mRS) at 6 months, we identified low certainty evidence (downgraded for risk of bias and indirectness) from 2 randomized controlled trials {Ali 2014 937; Mazdeh 2015 1069} recruiting 340 adult patients with acute stroke showing no difference in mRS score with use of supplementary oxygen via nasal cannula or Venturi mask for 12-72h and room air (oxygen only if clinically indicated) (absolute difference 0.22 points lower, 95% CI −0.01 to 0.45).
For the critical neurological outcome of modified Rankin sore (mRS) less than 3 at 6 months, we identified low certainty evidence (downgraded for risk of bias and indirectness) from 2 randomized controlled trials {Ali 2014 937; Mazdeh 2015 1069} recruiting 340 adult patients with acute stroke showing no difference between supplementary oxygen via nasal cannula at 2-3 L/min for 72 hours or Venturi mask for 12-72h compared with room air (oxygen only if clinically indicated) (risk ratio 1.06, 95% CI 0.84 to 1.34).
For the critical neurological outcome of Scandinavian stroke scale (SSS) at 3 months, we identified low certainty evidence (downgraded for indirectness and imprecision) from one randomized controlled trial {Singhal 2015 1035} recruiting 16 adult patients with acute stroke showing no difference with use of supplementary oxygen at 45 L/min via simple facemask for 8h compared with room air (oxygen used only if clinically indicated) (absolute difference 5.00 points higher, 95% CI 5.65 points lower to 15.65 points higher).
For the critical neurological outcome of Scandinavian stroke scale (SSS) at 7 months, we identified low certainty evidence (downgraded for risk of bias and indirectness) from one randomized controlled trial {Ronning 1999 408} recruiting 550 adult patients with acute stroke showing benefit with use of supplementary oxygen at 3 L/min via nasal cannula for 24h compared with room air (SSS at 7 months: absolute difference 0.50 points lower, 95% CI 0.98 lower to 0.02 points lower).
For the important quality of life outcome of Barthel index at 3 months, we identified moderate certainty evidence (downgraded for indirectness) from one randomized controlled trial {Roffe 2017 1206} recruiting 8003 patients with acute stroke showing no difference with use of supplementary oxygen at 2-3 L/min via nasal cannula for 72h compared with room air (absolute difference 0.70 points lower, 95% CI 1.49 points lower to 2.89 points higher).
For the important quality of life outcome of Barthel index at 6 months, we identified very low certainty evidence (downgraded for risk of bias, indirectness and imprecision) from one randomized controlled trial {Mazdeh 2015 1069} recruiting 51 adult patients with acute stroke showing no difference with use of supplementary oxygen via Venturi mask for 12h compared with room air (absolute difference 7.70 points higher, 95% CI 11.01 points lower to 26.41 points higher).
For the important quality of life outcome of Barthel index at 7 months, we identified low certainty evidence (downgraded for risk of bias and indirectness) from one randomized controlled trial {Ronning 1999 1069} recruiting 550 adult patients with acute stroke showing that the patients receiving supplementary oxygen 3 L/min via nasal cannula for 24h had a lower Barthel index as compared with those breathing room air (absolute difference 5.00 points lower, 95% CI 6.24 points lower to 3.76 points lower).
For the important quality of life outcome of Nottingham Extended ADL score at 3 months, we identified moderate certainty evidence (downgraded for indirectness) from one randomized controlled trial {Roffe 2017 1206} recruiting 8003 patients with acute stroke showing no difference with use of supplementary oxygen at 2-3 L/min via nasal cannula for 72h compared with room air (absolute difference 0.11 lower, 95% CI 0.28 points lower to 0.50 points higher).
For the important quality of life outcome of (EuroQOL [EQ5D-3L]) score at 3 months, we identified moderate certainty evidence (downgraded for indirectness) from one randomized controlled trial {Roffe 2017 1206} recruiting 8003 patients with acute stroke showing no difference with use of supplementary oxygen at 2-3 L/min via nasal cannula for 72h compared with room air (absolute difference 0.01 points higher, 95% C.I. 0.03 points lower to 0.01 points higher).
For the important quality of life outcome of VAS for at 3 months, we identified moderate certainty evidence (downgraded for indirectness) from one randomized controlled trial {Roffe 2017 1206} recruiting 8003 patients with acute stroke showing no difference with use of supplementary oxygen at 2-3 L/min via nasal cannula for 72h compared with room air (absolute difference 0.10 points lower, 95% C.I. 1.67 points lower to 1.57 points higher).
For the important imaging outcome of lesion volume change at 6 hours, we identified low certainty evidence (downgraded for indirectness and imprecision) from one randomized controlled trial {Wu 2012 894} recruiting 16 adult patients with acute stroke showing no difference with use of high-flow supplementary oxygen via facemask for 8 h compared with room air (absolute difference 63% higher, 95% CI 16% lower to 142% higher).
For the important imaging outcome of lesion volume change at 24 hours, we identified low certainty evidence (downgraded for indirectness and imprecision) from one randomized controlled trial {Wu 2012 894} recruiting 16 adult patients with acute stroke showing no difference with use of high-flow supplementary oxygen via facemask for 8 h compared with room air (absolute difference 57% higher, 95% CI 60% lower to 174% higher).
For the important imaging outcome of lesion volume change at hospital discharge, we identified low certainty evidence (downgraded for indirectness and imprecision) from one randomized controlled trial {Wu 2012 894} recruiting 16 adult patients with acute stroke showing no difference with use of high-flow supplementary oxygen via facemask for 8 h compared with room air (absolute difference 31% higher, 95% CI 58% lower to 120% higher).
For the important adverse effects and complications outcome of hospital-acquired pneumonia, we identified very low certainty evidence (downgraded for risk of bias) from one retrospective observational study {Dylla 2019 30742} recruiting 1352 adult patients with acute stroke showing that the patients without hypoxia at baseline who received prehospital supplementary oxygen had a lower rate of hospital-acquired pneumonia compared with those breathing room air (risk ratio 0.50, 95% CI 0.26-0.98).
For the important adverse effects and complications outcome of any documentation of pneumonia at hospital discharge, we identified very low certainty evidence (downgraded for risk of bias) from one retrospective observational study {Dylla 2019 30742} recruiting 1352 adult patients with acute stroke showing no difference between patients without hypoxia at baseline who received prehospital supplementary oxygen compared with those breathing room air (risk ratio 1.77, 95% CI 0.97-3.21).
For the important adverse effects and complications outcome of pulmonary edema, we identified very low certainty evidence (downgraded for risk of bias) from one retrospective observational study {Dylla 2019 30742} recruiting 1352 adult patients with acute stroke showing no difference between patients without hypoxia at baseline who received prehospital supplementary oxygen compared with those breathing room air (risk ratio 1.41, 95% CI 0.52-3.86).
For the important adverse effects and complications outcome of use of non-invasive positive pressure ventilation, we identified very low certainty evidence (downgraded for risk of bias) from one retrospective observational study {Dylla 2019 30742} recruiting 1352 adult patients with acute stroke showing no difference between the patients without hypoxia at baseline who received prehospital supplementary oxygen and those breathing room air (risk ratio 1.57, 95% CI 0.56-4.38).
For the important adverse effects and complications outcome of intubation with mechanical ventilation, we identified very low certainty evidence (downgraded for risk of bias) from one retrospective observational study {Dylla 2019 30742} recruiting 1352 adult patients with acute stroke showing that the patients without hypoxia at baseline who received prehospital supplementary oxygen had a higher rate of intubation with mechanical ventilation in comparison with those breathing room air (risk ratio 2.80, 95% CI 2.1-3.70).
For the adverse effects and complications outcome of any respiratory complications during hospitalization, we identified very low certainty evidence (downgraded for risk of bias) from one retrospective observational study {Dylla 2019 30742} recruiting 1352 adult patients with acute stroke showing that the patients without hypoxia at baseline who received prehospital supplementary oxygen had a higher rate of respiratory complications in comparison with those breathing room air (risk ratio 1.92, 95% CI 1.54-2.39).
Treatment Recommendations
For adults with suspected acute stroke, we suggest against the routine use of supplementary oxygen in the first aid setting compared with no use of supplementary oxygen (weak recommendation, low to moderate certainty of evidence)
Justification and Evidence to Decision Framework Highlights
This topic was prioritized by the FA task force because of ongoing controversy in the published literature regarding the effects of supplementary oxygen administration for acute stroke. This is a new systematic review of the literature on this topic by the FA task force. The use of oxygen by first aid providers or first responders is not common, but first aid courses for the administration of oxygen are a means of training and certification in this skill, making oxygen use within the scope of care for some first aid providers.
A single observational study was identified and considered as direct evidence from the prehospital setting to inform this review, supported by 8 RCTs from the in-hospital setting. They all compare the use of supplementary oxygen with varying flow rates and delivery methods with no use of supplementary oxygen (i.e., room air) in individuals with acute stroke. With few exceptions, the results of these studies consistently failed to find a benefit for critical outcomes such as survival and neurological outcomes including NIHSS score, nor for the important outcomes related to quality of life. A limitation of some of the included RCTs {Padma 2010 284; Singhal 2005 297} was the use of low dose oxygen in the comparison group when clinically indicated, and the failure to separate this group from the remainder of the comparison group. Although the intervention group received high flow oxygen in these studies, the inclusion of patients clinically in need of oxygen is a confounder.
We also considered potential harm from use of supplementary oxygen. A single retrospective observational stroke registry study reported on rates of respiratory complications as well as neurological outcomes (NIHSS). Dylla et al. {Dylla 2019 30742} grouped the registry patients by 1) need for oxygen and they received oxygen due to hypoxia, 2) oxygen delivery despite normoxia (‘hyperoxia’ group) and 3) no oxygen given (control group). They evaluated mean prehospital and discharge NIHSS and respiratory complication rates for each of the three groups and concluded that when controlling for confounders there was no significant increase in respiratory complications or difference in neurological outcomes at discharge, suggesting that brief, early supplementary oxygen from stroke may be safe to evaluate prospectively.
In making this recommendation the task force recognizes the research equipoise of the currently available evidence related to the use of supplementary oxygen for acute stroke. Task force deliberations weighed heavily on this recommendation. The balance of effects changes when considering certainty of evidence, resource requirements, cost and potential impact on health equity, acceptability by all stakeholders and feasibility, particularly from an international perspective.
The resources required for oxygen delivery are considerable, including oxygen equipment and supplies, the need for a carrying container, and need for oxygen storage. A specialized course and certification in first aid oxygen use may be required, and some countries may require a prescription or a license to use oxygen. The expense associated with equipment, supplies and training may be considerable compared with the use of room air and may contribute to a potential negative impact on health equity in resource-limited countries. The stocking, storage or transportation of equipment and supplies may not be feasible or acceptable to first aid providers or first responders. Occupational and accidental injuries and mishaps related to the use of oxygen canisters were also considered in TF discussion. Finally, concern was expressed that the process of setting up and administering oxygen may delay other critical immediate care goals, such as calling a designated emergency number or transporting a person to a hospital. These considerations tip the balance between benefits and harms, as shown in the accompanying Evidence to Decision table.
Knowledge Gaps
- There are no randomized controlled trials comparing routine supplementary oxygen with room air in acute stroke patients in first aid settings.
- The effect of short-term use of supplementary oxygen only in the first aid settings remains unknown.
- There are no studies about optimal concentration of supplementary oxygen for adults with suspected acute stroke.
- There are no studies about the best device to administrate supplementary oxygen for adults with suspected acute stroke.
Attachment
EtD table FA-1549-Oxygen-for-stroke-Et D
References
Ali K, Warusevitane A, Lally F, Sim J, Sills S, Pountain S, Nevatte T, Allen M, Roffe C. The SOS pilot study: A randomized controlled trial of the effects of routine oxygen supplementation early after acute stroke - Effect on key outcomes at six months. PLoS One 2013;8(6):e59274
Dylla L, Adler DH, Abar B, Benesch C, Jones C, O’Banion K, Cushman JT. Prehospital supplementary oxygen for acute stroke – a retrospective analysis. Am J Emerg Med 2019 Nov 18. Pii: S0735-6757(19)30742-7. doi: 10.1016/j.ajem.2019.11.002. [Epub ahead of print]
Mazdeh M, Taher A, Torabian S, Seifirad S. Effects of Normobaric Hyperoxia in Severe Acute Stroke: a Randomized Controlled Clinical Trial Study. Acta Med Iran 2015;53(11): 676-680
Padma MV, Bhasin A, Bhatia R, Garg A, Singh MB, Tripathi M, Prasad K. Normobaric oxygen therapy in acute ischemic stroke: A pilot study in Indian patients. Ann Indian Acad Neurology 2010;13(4):284-288
Roffe C, Ali K, Warusevitane A, Sills S, Pountain S, Allen M, Hodsoll J, Lally F, Jones P, Crome P. The SOS pilot study: A RCT on routine oxygen supplementation early after acute stroke - Effect on neurological recovery at one week. PLoS One 2011;6(5):e19113
Roffe C, Nevatte T, Sim J, Bishop J, Ives N, Ferdinand P, Gray R, Stroke Oxygen investigators and the Stroke Oxygen Study Collaborative Group. Effect of routine low-dose oxygen supplementation on death and disability in adults with acute stroke – The Stroke Oxygen Study Randomized Clinical Trial. JAMA 2017 Sep;318(12):1125-1135
Ronning OM, Guldvog B. Should stroke victims routinely receive supplementary oxygen? A quasi-randomized controlled trial. Stroke 1999;30:2033-2037
Singhal A, Benner T, Roccatagliata L, Koroshetz WJ, et al. A pilot study of normobaric oxygen therapy in acute ischemic stroke. Stroke 2005;36:797-802
Wu Q, Benner T, Roccatagliata L, Zhu M, Schaefer PW, Sorensen AG, Singhal AB. Evaluating effects of normobaric oxygen therapy in acute stroke with MRI-based predictive models. Med Gas Res 2012;2:5