Edaravone dexborneol versus placebo on functional outcomes in patients with acute ischaemic stroke undergoing endovascular thrombectomy (TASTE-2): randomised controlled trial.

9 major claims checked against the paper's own evidence: all adequately supported.

Recomputed 11 tests: 11 consistent, 0 inconsistent, 11 via agent-written checks.

2 integrity concerns flagged (0 high).

5 copyedit issues flagged: mostly typo, consistency, other.

Checked 31 references: 31 verified.

4 data/code links checked; 3 live.

Ready after minor edits. The paper is methodologically sound, but the internal contradiction in Table 2 (adverse events: 354/690 is not 33.0%) must be corrected before submission — this is a data integrity issue that editors and reviewers would catch. Additionally, adding a regulatory-compliance statement, a CONSORT checklist mention, verifying statistical assumptions, and depositing code in a public repository would strengthen the submission and pre-empt reviewer requests.

1. 1.
   HIGHcopyedit

   Correct the adverse events count/percentage in Table 2 for the edaravone dexborneol group: '354/690' is arithmetically 51.3%, not 33.0%. Verify the correct numerator (e.g., 228/690 ≈ 33.0%) and denominator, and update the table accordingly.

   This internal contradiction is a data integrity error that will be caught by reviewers or editors and undermines trust in the reporting.
2. 2.
   HIGHethics

   Add an explicit sentence to the Ethics section stating that the study was conducted in accordance with the Declaration of Helsinki and/or ICH Good Clinical Practice guidelines.

   This is a standard requirement for human clinical trial reporting; its absence is a key gap flagged by reviewers.
3. 3.
   HIGHreporting

   Add a statement to the Methods section (e.g., in Study design or Statistical analysis) that the trial is reported in accordance with the CONSORT 2010 statement, and provide the completed CONSORT checklist as supplementary material.

   Many journals require explicit reporting guideline adherence; including it pre-empts a common reviewer request.
4. 4.
   HIGHstatistics

   Add a sentence in the Statistical analysis section stating that the assumptions of logistic regression (e.g., linearity of logit, absence of multicollinearity, no influential outliers) were checked and satisfied, or note that the large sample size makes the model robust.

   The assumptions sub-criterion was not reported; adding this shows analytical thoroughness.
5. 5.
   HIGHdata code

   Deposit the analysis code (currently in appendix S15) in a version-controlled public repository (e.g., GitHub, Zenodo) with a persistent DOI and update the data availability statement to reference that repository.

   Supplementary appendices are not version-controlled and lack a persistent identifier, which is the expected best practice for computational reproducibility.
6. 6.
   MEDIUMstatistics

   Provide exact p-values for all outcomes in Table 2 instead of inequalities such as 'P>0.99' — report the precise value (e.g., 'P=0.99' or 'P=0.999') in the table or a footnote.

   Exact p-values are more informative and allow readers to assess results precisely; the current reporting is inconsistent.
7. 7.
   MEDIUMreporting

   Add a note in the Statistical analysis section clarifying that all secondary outcome p-values are two-sided and unadjusted for multiplicity, and briefly discuss this in the limitations.

   Transparency about multiplicity adjustment (or lack thereof) is important for interpreting the secondary analyses.
8. 8.
   MEDIUMreporting

   In the Discussion, add a brief sentence addressing whether the borderline primary p-value (0.05) remains robust after adjustment for the reported subgroup analyses or secondary comparisons.

   The borderline p-value and exploratory nature of the subgroup finding should be contextualised to avoid over-interpretation.
9. 9.
   MEDIUMreporting

   Add a summary of the random-centre-effect convergence issue to the Methods (Statistical analysis) so readers understand why centre was not included as a random effect.

   The limitations section mentions convergence issues; providing the methodological detail upfront improves transparency.
10. 10.
    LOWother

    Provide a plain-language summary of the subgroup findings and their exploratory nature in the abstract to minimise risk of over-interpretation in news coverage.

    Subgroup analyses are often over-emphasised; explicitly labeling them as exploratory in the abstract is a safeguard.