WSO - January 2024 - 125

Poli et al.
125
oxygen, maintaining physiological levels during middle
cerebral artery occlusion when inspiratory oxygen fraction
was at least 0.95.11 Accordingly, in a randomized pilot trial,
oxygen at 45 L per minute via a simple face mask led to
significant arterial hyperoxygenation, temporary NIHSS
improvement, and stabilization of DWI lesions during
NBO.12 Compared to hyperbaric oxygen therapy, which
may provide superior brain protection according to animal
studies, NBO is inexpensive, widely available and easy to
administer during acute stroke workup including MT.5
Second, we aimed to initiate NBO early. Brain tissue
oxygen is depleted within seconds after blood flow interruption,1
and animal studies have shown no neuroprotection
with delayed NBO.13 Considering results from
extended time window MT trials, the high ASPECTS
required for enrollment into PROOF combined with the
time window of 6 h after symptom onset/notice would sufficiently
well-indicate relevant volumes of potentially salvageable
penumbra.14
Ethical concerns required brain imaging to exclude
hemorrhagic stroke patients who are less likely to benefit
from NBO and might even be harmed.15 To start NBO as
soon as possible after exclusion of hemorrhage and confirmation
of qualifying LVO, PROOF settled for a likely but
not-yet-decided MT. In addition, evaluation of CT perfusion
was not required before enrollment. Consequential
inclusion of fast progressors with malignant perfusion profiles
may help to differentiate their response to NBO from
that of well-collateralized slow progressors.16 To accelerate
enrollment procedure, we opted for deferred consent in
countries where permitted.
Third, as a proof of concept, we focused on MT candidates
as our study population. In animal models, NBO only
consistently led to infarct volume reduction when ischemia
was temporary and lasted up to 3 h.4 Given that " nothing
can hold its breath forever, " beneficial effects of NBO vanished
in case of no recanalization.12 MT provides high rates
of successful reperfusion and clearly defined ischemia
duration and revascularization status.3,9
Fourth, we aimed to stop NBO at the end of MT procedure.
Few animal studies evaluated post-reperfusion NBO
and reported mixed results.11,17,18 While Liu et al.11 warned
that over-oxygenation during reperfusion could potentially
lead to free radical toxicity, they also pointed out that
hyperoxia seems more tolerable to the brain than hypoxia.
This is why we did not stop NBO immediately after successful
reperfusion but allowed NBO to also cover treatment
of accidental new ischemia in case of thrombus
dislocation during clot retrieval.
An early brain imaging primary outcome, heterogenous
baseline imaging and intermodal comparison are potential
limitations of the trial. Early imaging outcomes are predictive
of clinical outcomes at 90 days19 and magnetic
resonance imaging (MRI) can effectively quantify neuroprotection.12
To study the primary outcome, comparison of
DWI at 24 h and at baseline would be optimal. However, to
not distort the standard workflow, PROOF allowed for multimodal
CT at baseline.20 Using baseline imaging modality
as stratum for randomization mitigates the risk of
confounding. To address interindividual ischemia size
variability due to LVO location, infarct growth was chosen
over infarct volume. Primary outcome assessment at
24 h on follow-up imaging avoids drop-outs due to withdrawal
of care and reduces variability due to early clinical
complications such as delayed extubation and aspiration
pneumonia.
Summary and conclusion
Available data indicate
protective effects of NBO on
ischemic but still viable brain tissue if (1) near 100% oxygen
is (2) initiated early in (3) short temporary ischemia
and (4) stopped soon after reperfusion. PROOF is a proof
of concept for this assumption and focuses on a select population
mimicking the animal models in which NBO was
shown beneficial.
We believe that most clinical studies15 failed to show
positive or long-lasting NBO-effects in ischemic stroke
patients because evidence from successful preclinical
application of NBO was not sufficiently considered in the
design of these trials. Two recent Chinese single-center
studies showed a strong benefit of NBO. One study was
similar to PROOF,21 the other, however, started NBO only
after successful MT.22
Considering its low cost, availability, and ease of use,
the prospect of NBO to extend the time window for successful
reperfusion by " freezing the penumbra " may impact
acute stroke care worldwide.
Acknowledgements
The authors thank DSMB (Werner Hacke, Valeria Caso, Christine
Roffe, Gerhard Schroth, and David Petroff), IEAB (Peter
Rosenberger, Matthew Schrag, and Gary Randall), and PROOF
investigators for their diligence, guidance, and support of our trial.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of this
article.
Funding
The author(s) disclosed receipt of the following financial support
for the research, authorship, and/or publication of this article:
European Commission within the Horizon 2020 program (grant
no. SC1-PM-09-2016).
ORCID iDs
Sven Poli
https://orcid.org/0000-0002-0286-8781
Guillaume Turc
https://orcid.org/0000-0001-5059-4095
International Journal of Stroke, 19(1)
https://www.orcid.org/0000-0002-0286-8781 https://www.orcid.org/0000-0001-5059-4095
WSO - January 2024

Table of Contents for the Digital Edition of WSO - January 2024
