Table 1 Studies evaluating the use of oxygen therapy in acute ischaemic stroke patients

Rønning et al. 1999 [30]

Quasi-randomised trial

To test the hypothesis that breathing 100% O₂ for the first 24 h after an AS would not reduce mortality, impairment, or disability

AS

555 (66/555 ICH)

O₂ therapy (100% atm, 3 L/min for 24 h [n = 292]) versus controls (no additional O₂)

Supplemental O₂ should not routinely be given to non-hypoxic stroke victims with minor or moderate strokes.

Ali et al. 2005 [35]

Prospective study

To assess the effects of different doses and routes of O₂ administration on SatO₂ in patients with stroke

AS

21 (15 AIS; 6 ICH)

Steps:

-Room air for 30 min; − 2-3 L/min of O₂ (nasal cannula)

-FiO₂ 0.24 (VM)

-FiO₂ 0.3535% (VM)

-Room air

There was a dose-response relationship between the amount of O₂ given and the resultant changes in SatO₂.

Singhal et al. 2005 [36]

Randomised trial

To investigate the effects of high-flow oxygen (HFO) in AIS

AIS

16

HFO (humidified O2 at 45 L/min via facemask for 8 h [n = 9]) versus controls (room air or nasal O₂ 1–3 L/min to maintain SaO₂ 95% [n = 7])

HFO is associated with a transient improvement of clinical deficits and MRI abnormalities in select patients with AIS.

Chiu et al. 2006 [37]

Prospective study

To investigate the feasibility of eubaric hyperoxia therapy by VM in a group of patients who experienced a severe AIS

AIS

46

O₂ therapy (FiO₂ 0.4via VM [n = 17]) versus controls (nasal cannula [n = 29])

By using VM therapy with a FiO₂ of 0.4, there might be less mortality and comorbidities in treated patients who experienced a severe AIS.

Singhal et al. 2007 [38]

RCT

To investigate the metabolic effects of normobaric oxygen (NBO) on AIS brain tissues using MRSI and diffusion-perfusion MRI

AIS

6

NBO (45 L/min O₂ via face mask for 8 h [n = 4]) versus controls (room air [n = 2])

NBO improves aerobic metabolism and preserves neuronal integrity in AIS brain.

Padma et al. 2010 [39]

Randomised trial

To study the role of NBO in AIS in Indian patients

AIS

40

NBO (10 L/min O₂ for 12 h [n = 20]) versus controls (room air/2 L/min O₂ via face mask to maintain SaO2 ≥ 95% [n = 20])

NBO did not improve the clinical scores of stroke outcome in Indian patients with AIS.

Roffe et al. 2011 [31]

RCT

To report the effects of routine use of O₂ supplementation for 72 h on SatO₂ and neurological outcomes at 1 week after an AS

AS (clinical diagnosis)

289 (257 AIS, 24 ICH, 8 undetermined)

O₂ therapy (O₂ via nasal cannula for 72 h [n = 148]) or controls (room air/O₂ if clinically indicated [n = 141])

Routine O₂ supplementation started within 24 h of hospital admission with AS led to a small, but statistically significant, improvement in neurological recovery at 1 week. The difference in NIHSS improvement may be due to baseline imbalance in stroke severity between the two groups.

Ali et al. 2013 [40]

RCT

To report the effects of routine O₂ supplementation for 72 h on SatO₂ and neurological outcomes at 6 months after AS

AS (clinical diagnosis)

289 (257 AIS, 24 ICH, 8 undetermined)

O₂ therapy (O₂ via nasal cannula for 72 h [n = 148]) or controls (room air/O₂ if clinically indicated [n = 141])

None of the key outcomes differed at 6 months between the groups. Although not statistically significant and generally of small magnitude, the effects were predominantly in favour of the O₂ group.

Jeon et al., 2014 [28]

Prospective, observational cohort study

To determine the association between exposure to hyperoxia and the risk of DCI after SAH

SAH

252

Hyperoxia (the highest quartile of an area under the curve of PaO₂, until the development of DCI [PaO₂ ≥ 173 mmHg])

Exposure to excess O₂ after SAH may represent a modifiable factor for morbidity and mortality in this population.

Rincon et al. 2014 [41]

Retrospective multicentre cohort study

To test the hypothesis that hyperoxia was associated with higher in-hospital mortality in ventilated AS patients admitted to the ICU

AS

2894 (554 AIS, 936 SAH, 1404 ICH)

O₂ therapy to obtain PaO₂ ≥ 300 mmHg

Hyperoxia was an independent predictor of in-hospital death.

Mazdeh et al. 2015 [42]

RCT

To evaluate the effects of normobaric hyperoxia on clinical outcomes of patients with severe AS

AS

52

O₂ therapy (VM for 12 h) versus controls (no O₂)

NBO therapy in the first 12 h of AS could improve long-time outcome of the patients with either ischaemic or haemorrhagic stroke.

Roffe et al., 2017 [32]

Multicentre single-blind RCT

To assess whether routine prophylactic use of low-dose O₂ therapy was more effective than control O₂ administration at reducing death and disability at 90 days, and if so, whether O₂ given at night only, when hypoxia is most frequent, and O₂ administration is least likely to interfere with rehabilitation, was more effective than continuous supplementation

AS (clinical diagnosis)

8003 (6555 AIS, 588 ICH, 294 AS without CT diagnosis, 168 TIA, 292 non-stroke diagnoses, 106 missing data)

Continuous O₂ (2–3 L/min via nasal cannula for 72 h [n = 2668]) versus nocturnal O₂ (2–3 L/min via nasal cannula for 3 nights [n = 2667]) versus controls (O₂ if clinically indicated [n = 2668])

Among non-hypoxic patients with AS, the prophylactic use of low-dose O₂ supplementation did not reduce death or disability at 3 months.

Ding et al. 2018 [43]

Meta-analysis

To analyse the current data of NBO on brain protection as used in the clinical settings

AS

6366

NBO group (… [n = 3207]) versus controls [n = 3159]

The existing trends toward benefits revealed in this meta-analysis help us appreciate the promising value of NBO, although current evidence of NBO on improving clinical outcomes of stroke is insufficient.

Roffe et al. 2018 [44]

Multicentre, prospective, randomised, open, blinded-end point trial

(1) To assess whether or not routine low-dose of O₂ supplementation in patients with AS improves outcome compared with no O₂ and (2) to assess whether or not O₂ given at night only, when SatO₂ is most likely to be low, is more effective than continuous supplementation

AS (clinical diagnosis)

8003 (6555 AIS, 588 ICH, 168 TIA, 292 non-stroke diagnoses, 106 missing data)

Continuous O₂ (2–3 L/min via nasal cannula for 72 h [n =  668]) versus nocturnal O₂ (2–3 L/min via nasal cannula for 3 nights [n = 2667]) versus controls (O₂ if clinically indicated [n = 2668])

Routine use of low-dose O₂ supplementation in stroke patients who are not severely hypoxic is safe but does not improve outcome after AS.