Correction: Trials 24, 20 (2023)

https://doi.org/10.1186/s13063-022-07061-7


Following the publication of the original article [1], the authors noticed two errors in the Methods section, which became relevant during data analyses. In the sub-section “Statistical Analysis”, it was stated that subgroup analyses will be performed according to Glasgow Coma Scale (GCS) at the moment of randomization (3-5; 6-8; 9-11). However, this is in contrast with the “Sequence generation” sub-section, in which all eligible acute brain injured patients underwent randomization which was also stratified by the GCS at the moment of randomization (3-5; 6-9; 10-13). Moreover, the initial proposal would exclude some patients (e.g. GCS 12 or 13) from the subgroup analysis. This should read: “Subgroup analyses will be performed according to … b) GCS at the moment of randomization (3-5; 6-9; 10-13); …”. The second error is the lack of cerebral ischemia among serious adverse events, which was present in the protocol submitted for Ethical Approval. The definition of cerebral ischemia was “New (i.e. after randomization) ischemic lesion visible on brain imaging (i.e. either CT-scan or MRI).”