[Skip to Navigation]
Views 239
Citations 0
Comment & Response
February 6, 2018

Prophylactic Low-Dose Oxygen for Patients With Acute Stroke—Reply

Author Affiliations
  • 1Faculty of Medicine and Health Sciences, Keele University, Staffordshire, United Kingdom
  • 2MRC Population Health Research Unit, University of Oxford, Oxford, United Kingdom
JAMA. 2018;319(5):502. doi:10.1001/jama.2017.20349

In Reply Dr Saber and colleagues note that many patients in SO2S had mild strokes and suggest that this may explain the lack of observed benefit (OR, 0.97 [95% CI, 0.89-1.05]) from low-dose oxygen supplementation for 72 hours after acute stroke. To support this argument, they highlight the nonsignificant improved functional outcome (OR, 1.28 [95% CI, 0.94-1.73] favoring oxygen) in patients with NIHSS scores of 15 to 20. While this might suggest potential benefit, the difference between oxygen and no oxygen in this subgroup was not statistically significant, nor was the test for heterogeneity between this result and that in the 4 other NIHSS subgroups (P = .37), nor was a test for trend of increasing benefit with increasing stroke severity (P = .34). Moreover, outcomes were worse with oxygen in patients with very severe strokes (NIHSS score >20; OR, 0.81 [95% CI, 0.53-1.22]), an opposite finding to that in the NIHSS scores of 15 to 20 group, which is difficult to explain physiologically. These apparent differences between subgroups are therefore most likely to be explained by chance fluctuation.