Background and Aims:
Normobaric oxygen (NBO) is potentially a readily accessible neuroprotective therapy. We undertook a systematic review to assess the effect of NBO in acute stroke.

Methods:
Medline, EMBASE and CENTRAL databases were searched from inception to November 2018. Randomised controlled trials of NBO administered <7 days after stroke to normoxic patients and no other indication for oxygen were identified by two researchers independently. Data on early recovery within 7 days; functional outcome within 9 months; mortality; oxygen saturation and imaging markers were collected.

Results:
15 publications involving 12 cohorts were identified. Oxygen regimes ranged from 2l/min via nasal cannula nocturnally to 45l/min via facemask for 8 hours from randomisation. The proportion of patients with ischaemic stroke was 82.8% (range 61-100%); 3.5% were stroke mimics (where declared). Median time to treatment starting was 19.3 hours. Meta-analysis demonstrated no significant effect on: modified Rankin scale at follow up (Standardised mean difference (SMD)=-0.015 (-0.060–0.030)); reduction in National Institutes of Health Stroke Scale (NIHSS) at 7 days in all stroke or ischaemic stroke only (Mean difference=-0.16 (-1.11–0.80) and -0.73 (-3.54–2.08) respectively); neurological improvement at week 1 assessed by NIHSS or Scandinavian Stroke Scale (SMD=-0.026 (-0.154–0.103)); improvement in Barthel index (SMD=0.001 (-0.043–0.045)) or mortality (odds ratio=1.03 (0.90–1.19).

Conclusions:
Studies to date of NBO have shown no significant effect on early recovery, functional outcome or mortality in acute stroke. Study protocols and populations have varied considerably. Populations have included significant numbers of non-ischaemic stroke and stroke mimics. Oxygen has been predominantly low dose commenced in the subacute phase.

RANDOMISED CONTROLLED TRIALS OF NORMOBARIC OXYGEN THERAPY IN ACUTE STROKE: A SYSTEMATIC REVIEW AND META ANALYSIS