Routine prophylactic oxygen supplementation may not reduce mortality or disability at 90 days in nonhypoxic patients with acute stroke
EBM Focus - Volume 12, Issue 45
- Early prophylactic supplemental oxygen in patients with acute stroke may be given with the intent to help minimize brain tissue damage.
- The SO2S trial investigated whether routine prophylactic low-dose oxygen supplementation reduces mortality or disability in 8,003 patients with acute stroke and without hypoxemia. Patients were randomized to continuous oxygen for 72 hours vs. nocturnal oxygen for 3 nights vs. control (oxygen given only if clinically indicated).
- No significant differences in mortality or disability were found in comparisons of the combined oxygen groups vs. the control group or between the continuous oxygen vs. the nocturnal oxygen groups. These findings do not support the use of routine prophylactic low-dose oxygen supplementation in nonhypoxic patients with acute stroke.
Stroke leads to disability or death in approximately 40% of patients (Neurology 2011). Early supplemental oxygen for patients with acute stroke is given with the intent to salvage ischemic brain tissue. The American Heart Association (AHA)/American Stroke Association (ASA) guidelines do not recommend supplemental oxygen for patients with acute stroke who do not have hypoxemia (Stroke 2013), but this is based on limited evidence. Three previous trials compared oxygen to room air; two demonstrated improvement in neurologic recovery at 1 week with oxygen therapy (PLoS One 2011, Stroke 2005), while one larger quasi-randomized trial found no differences in 1-year survival or disability at 7 months (Stroke 1999). The recent SO2S trial investigated the effects of routine prophylactic low-dose oxygen supplementation, either continuously or overnight. The purpose of the overnight group was to control for the possible effects of reduced mobility on rehabilitation. The study examined whether supplemental oxygen reduced mortality or disability in patients with a clinical diagnosis of acute stroke and no indication for oxygen therapy. In this trial, 8,003 patients (mean age 72 years) with acute stroke were randomized to continuous oxygen for 72 hours vs. nocturnal oxygen for 3 nights vs. control (oxygen given only if clinically indicated). Oxygen flow rate was 3 L/minute if baseline oxygen saturation ≤ 93% and 2 L/minute if > 93%. The primary outcome at 90 days was odds of improvement by 1 level on the modified Rankin scale (range 0-6 with 0 indicating no disability and 6 indicating death). Comparisons were made between the combined oxygen groups vs. control and the continuous oxygen group vs. the nocturnal oxygen group.
Ischemic stroke was ultimately diagnosed in 82% of patients and intracerebral hemorrhage in 7.3%. Before randomization, mean baseline oxygen saturation was 96.6%, and 20% of the patients in each group received oxygen supplementation. Analyses were performed in 96% of patients. The mean of the lowest oxygen saturation levels recorded during the 72 hours after randomization was 95% in the continuous oxygen group vs. 94.5% in the nocturnal oxygen group vs. 94.1% in the control group. There were no significant differences in the primary outcome between the combined oxygen groups vs. control or between continuous oxygen vs. nocturnal oxygen. Furthermore, there were no significant differences in 90-day mortality, either between the combined oxygen groups (9.6%) and the control group (9.6%) (adjusted odds ratio 1.03, 95% CI 0.81-1.3) or between the continuous oxygen group (10%) and the nocturnal oxygen group (9.2%) (adjusted odds ratio 1.09, 95% CI 0.83-1.43). Also, there were no significant differences in rates of patients with a good functional outcome (modified Rankin Scale score ≤ 2) for both comparisons. The results were consistent across multiple subgroup analyses. Rates of serious adverse events were 13% with continuous oxygen, 11% with nocturnal oxygen, and 12.1% with control (p = 0.02 for continuous vs. nocturnal oxygen).
In patients with acute stroke who did not have hypoxemia at baseline, there were no significant differences in mortality or disability comparing routine use of low-dose (2-3 L/minute) oxygen supplementation versus oxygen given only if clinically indicated. These findings are somewhat tempered by the 4% loss to follow-up, which change the analysis to favor one group or another, depending on the assumptions made to account for the missing data. The findings of the SO2S trial, however, are consistent with a recent publication reporting on the comparison of supplemental oxygen vs. ambient air in over 6,600 patients with suspected myocardial infarction and oxygen saturation ≥ 90% (N Engl J Med 2017), in which no significant differences were found in all-cause mortality or rehospitalization for acute myocardial infarction at 1 year. Overall, the SO2S trial provides strong evidence to support the AHA/ASA guidelines that recommend against the use of routine supplemental oxygen in patients with acute stroke who are not hypoxic.