Antithrombotic Therapy after Successful Catheter Ablation for Atrial Fibrillation.
Verma A, Birnie DH, Jiang C, Heidbüchel H, Hindricks G, Kirchhof P, Healey JS, Wang Y, Dagres N, Deyell MW, Sanders P, Pathak RK, Koopman P, Nuyens D, Novak P, Amit G, Dussault C, Makanjee B, Quinn FR, Jolly U, Iden L, Kuniss M, Sharma M, Ha A, Essebag V, Champagne J, Hill MD, Smith EE, Wells GA, OCEAN Investigators
- DOI
- 10.1056/NEJMoa2509688
- Record issued
- 2026-08-15
- Engine
- 7.39.0
- Exported
- 2026-09-22
Prepared by Alpha1. This document is confidential: it is intended for the recipient it was shared with and must not be redistributed. The live record at alpha1science.com/verify/4f996029-3bf8-41e7-8267-473bd772ea27 is authoritative.
How this rating was calculated
- CitationsCitations & links (capped) ×5−1★
- ClaimsEfficacy rests on an unvalidated surrogate endpoint−0.5★
- ClaimsTreatment effect not shown to be clinically meaningful−0.5★
- ReportingData & code availability partially met−0.25★
Citations & links are capped at −1★ combined, however many are flagged.
- The numeric-impossibility checks (GRIM/GRIMMER/DEBIT/SPRITE) did not run: 10 reported means were read, and their group size is not stated where the values are printed (this source has no machine-readable table structure). These checks need the count the mean was averaged over, so none was performed.
- No data or code availability links were detected to verify.
- 01Efficacy rests on an unvalidated surrogate endpoint
The primary efficacy outcome is a composite that includes 'new covert embolic stroke' detected by MRI, which is a surrogate marker for clinical stroke. The trial does not demonstrate target engagement for the tested dose of rivaroxaban (15 mg) specifically, nor does it cite validated evidence linking covert embolic stroke to clinical outcomes in this context. The low event rates and lack of significant difference further undermine the surrogate's adequacy.
“The primary outcome was a composite of stroke, systemic embolism, or new covert embolic stroke (defined by ≥1 new infarct measuring ≥15 mm on MRI) at 3 years.”
- 02Treatment effect not shown to be clinically meaningful
The primary outcome event rates are very low (0.31 vs 0.66 events per 100 patient-years), and the relative risk reduction is not statistically significant (P=0.28). The absolute risk difference is -0.6 percentage points, which is small and not clinically meaningful. The trial was stopped early due to futility, and the effect size is not anchored to a minimal clinically important difference.
“A primary-outcome event occurred in 5 patients (0.31 events per 100 patient-years) in the rivaroxaban group and in 9 patients (0.66 events per 100 patient-years) in the aspirin group (relative risk, 0.56; 95% confidence interval [CI], 0.19 to 1.65; absolute…”
This Kaimen Rigor review uses Kaimen Rigor reviewers trained on a curated corpus of high-fidelity and retracted papers, with expert supervision and curation. It can still make mistakes; verify each finding against the source before relying on it.
This is a well-conducted and well-reported randomized clinical trial with rigorous design, clear ethics approvals, and appropriate statistical methods. The main weakness is the vague data availability statement, which does not specify the access mechanism or repository. Minor reporting gaps include lack of explicit CONSORT adherence and a typo in the author affiliations.
Both reviewers independently scored all eight dimensions and agreed on all statuses; no divergence required reconciliation. The statistics component recomputed only 3 tests (those with test statistics/df or effect+CI); the remaining analyses were not machine-verified and should not be assumed correct. The citation check flagged 5 references as not found in registries, which may indicate fabrication or incorrect DOIs.
Numerical inconsistencies
None foundValues that contradict each other or are impossible for the stated sample: recomputed p-values and test statistics, GRIM/GRIMMER checks on summary numbers, percentages against their own counts, totals against their parts, and estimates against their own confidence intervals.
Checked — nothing surfaced.
Recomputed 3 tests: 3 consistent, 0 inconsistent; 3 via agent-written checks.
- CONSISTENTreported p = .280 · recomputed p = .293Reviewer 1Primary outcome relative risk p-value from CI
“relative risk, 0.56; 95% confidence interval [CI], 0.19 to 1.65; ... P = 0.28”
Taken as given: The relative risk is a ratio (log scale).; The CI is a 95% confidence interval.Method: Two-tailed p-value derived from the log-relative-risk and its 95% CI using a normal approximation.How we recomputed it: pCI(0.56, 0.19, 1.65, 1) - CONSISTENTreported p = .130 · recomputed p = .118Reviewer 1Primary safety outcome hazard ratio p-value from CI
“hazard ratio, 2.51; 95% CI, 0.79 to 7.95”
Taken as given: The hazard ratio is a ratio (log scale).; The CI is a 95% confidence interval.Method: Two-tailed p-value derived from the log-hazard-ratio and its 95% CI using a normal approximation.How we recomputed it: pCI(2.51, 0.79, 7.95, 1) - CONSISTENTreported p = .280 · recomputed p = .285Reviewer 2Primary efficacy outcome: chi-square test for 5/641 vs 9/643
“A primary-outcome event occurred in 5 patients (0.8%) in the rivaroxaban group and in 9 patients (1.4%) in the aspirin group (relative risk, 0.56; 95% confidence interval [CI], 0.19 to 1.65; absolute risk difference at 3 years, −0.6 percentage points; 95% CI, −1.8 to 0.5; P = 0.28)”
Taken as given: The numbers 5 and 9 are the event counts in the rivaroxaban and aspirin groups, respectively.; The denominators are 641 and 643, the total randomized in each group.; The test used is a chi-square test without continuity correction.Method: Pearson's chi-square test on the 2x2 table (5, 636, 9, 634).How we recomputed it: pChi2x2(5, 636, 9, 634)
Overstated conclusions
2 findings · worst highConclusions that reach past what the paper's own results support — including a significance claim that no longer holds when the statistic is recomputed, and efficacy resting on an unvalidated surrogate endpoint.
- Efficacy rests on an unvalidated surrogate endpointAssessed
- Treatment effect not shown to be clinically meaningfulAssessed
4 major claims checked against the paper's own evidence: all adequately supported.
- supportedReviewers 1, 2Rivaroxaban did not result in a significantly lower incidence of the primary composite outcome than aspirin.The primary outcome result (RR 0.56, 95% CI 0.19-1.65, P=0.28) directly supports the claim of no significant difference.Evidence: Primary outcome: 5 events (0.8%) vs 9 events (1.4%), RR 0.56, 95% CI 0.19-1.65, P=0.28.
“treatment with rivaroxaban did not result in a significantly lower incidence of a composite of stroke, systemic embolism, or new covert embolic stroke than treatment with aspirin.”
AbstractFind in source - supportedReviewers 1, 2The incidence of major bleeding appeared similar between groups.The safety outcome shows no statistically significant difference in major bleeding (HR 2.51, 95% CI 0.79-7.95), supporting the claim of similarity.Evidence: Fatal or major bleeding occurred in 10 (1.6%) vs 4 (0.6%), HR 2.51, 95% CI 0.79-7.95.
“Fatal or major bleeding (the composite primary safety outcome) had occurred in 10 patients (1.6%) with rivaroxaban and in 4 patients (0.6%) with aspirin (hazard ratio, 2.51; 95% CI, 0.79 to 7.95) at 3 years.”
AbstractFind in source - supportedReviewer 1The incidence of minor and clinically relevant nonmajor bleeding was higher with rivaroxaban.The reported rates and hazard ratios for minor and clinically relevant nonmajor bleeding are higher in the rivaroxaban group, supporting the claim.Evidence: Clinically relevant nonmajor bleeding: 5.5% vs 1.6%, HR 3.51 (95% CI 1.75-7.03); minor bleeding: 11.5% vs 3.1%, HR 3.71 (95% CI 2.29-6.01).
Clinically relevant nonmajor bleeding occurred in 5.5% of the patients in the rivaroxaban group, as compared with 1.6% of those in the aspirin group (hazard ratio, 3.51; 95% CI, 1.75 to 7.03).
Resultsreviewer’s wording - supportedReviewer 2The low event rates suggest that a much larger sample would be needed to detect a benefit.The paper calculates that over 11,000 patients would be needed, which is consistent with the observed low event rates.Evidence: Discussion states: 'Given such low event rates, a sample size of more than 11,000 patients would have been needed to detect a relative reduction of 40% in the risk of stroke in the rivaroxaban group.'
“Given such low event rates, a sample size of more than 11,000 patients would have been needed to detect a relative reduction of 40% in the risk of stroke in the rivaroxaban group.”
Discussion ¶3Find in source
Premise concern: surrogate not validated for clinical benefit; effect size not shown to be clinically meaningful.
- INADEQUATESurrogate endpointThe primary efficacy outcome is a composite that includes 'new covert embolic stroke' detected by MRI, which is a surrogate marker for clinical stroke. The trial does not demonstrate target engagement for the tested dose of rivaroxaban (15 mg) specifically, nor does it cite validated evidence linking covert embolic stroke to clinical outcomes in this context. The low event rates and lack of significant difference further undermine the surrogate's adequacy.
“The primary outcome was a composite of stroke, systemic embolism, or new covert embolic stroke (defined by ≥1 new infarct measuring ≥15 mm on MRI) at 3 years.”
- INADEQUATEEffect sizeThe primary outcome event rates are very low (0.31 vs 0.66 events per 100 patient-years), and the relative risk reduction is not statistically significant (P=0.28). The absolute risk difference is -0.6 percentage points, which is small and not clinically meaningful. The trial was stopped early due to futility, and the effect size is not anchored to a minimal clinically important difference.
“A primary-outcome event occurred in 5 patients (0.31 events per 100 patient-years) in the rivaroxaban group and in 9 patients (0.66 events per 100 patient-years) in the aspirin group (relative risk, 0.56; 95% confidence interval [CI], 0.19 to 1.65; absolute risk difference at 3 years, −0.6 percentage points; 95% CI, −1.8 to 0.5; P = 0.28).”
Data authenticity concerns
None foundAn adversarial read for patterns associated with data that may not be genuine: results that look too clean, implausibly large effects, duplicated data or images, and methods that do not match the results reported.
Checked — nothing surfaced.
Reporting gaps
1 finding · worst mediumRequired detail the manuscript never states — study design, biological variables, ethics approval and consent, key resources, statistical reporting, data and code availability, and overall transparency.
- Data/code availability incompleteAssessed
The introduction cites prior studies and guidelines, acknowledges that current recommendations are based on small nonrandomized studies, and states the trial's purpose. The rationale linking the premise to the objective is explicit. Limitations of prior research are addressed by noting the lack of randomized data.
“These recommendations are based on evidence from small, nonrandomized studies.”
“These recommendations are based on evidence from small, nonrandomized studies.”
Randomization method (web-based, blocks, stratified by site) and unit (patient) are reported. Blinding of outcome assessment is described. A sample size calculation with assumptions is provided. Inclusion/exclusion criteria are detailed. The analysis population (ITT and per-protocol) is defined, addressing outlier handling. Controls are inherent in the comparator arm. Independent replication is not applicable for a single pivotal trial.
“Data for all clinical outcomes were adjudicated by a committee whose members were unaware of the treatment assignments.”
“Data for all clinical outcomes were adjudicated by a committee whose members were unaware of the treatment assignments.”
“We calculated a sample size of 1572 patients.”
Sex is reported (28.6% women overall, and per-group in Table 1). Age and health status (comorbidities, CHA2DS2-VASc) are reported. Demographics are adequate. Species/strain and housing are not applicable for a human trial.
“The mean (±SD) age of the patients was 66.3±7.3 years, and 28.6% were women.”
“Demographics Age — yr 66.3±7.1 66.3±7.6 Male sex — no. (%) 458 (71.5) 459 (71.4)”
“The mean (±SD) age of the patients was 66.3±7.3 years, and 28.6% were women.”
“Hypertension — (%) 442 (69.0) 434 (67.5)”
The protocol was approved by the ethics committee at each site and national authorities. Written informed consent was obtained from all patients. Regulatory compliance is implied by the trial being conducted under standard oversight, though not explicitly named.
Aspirin and rivaroxaban are named with doses and rationale. The statistical software (SAS 9.4) is identified. No antibodies, cell lines, or other bench reagents are used, so those are not applicable.
“All the analyses were conducted with the use of SAS software, version 9.4 (SAS Institute).”
“All the analyses were conducted with the use of SAS software, version 9.4 (SAS Institute).”
Tests are named (chi-square, log-rank, Cox, Fine-Gray). Assumptions are handled by design (e.g., nonparametric tests). Exact p-values are reported for the primary outcome. Effect sizes with CIs are reported. Software is identified. Data presentation includes per-group n and event rates. Mathematical plausibility checks are not applicable due to large N and continuous outcomes.
“relative risk, 0.56; 95% confidence interval [CI], 0.19 to 1.65”
The paper states 'A data sharing statement provided by the authors is available with the full text of this article at NEJM.org.' This is vague; the actual statement is not included in the text provided. No repository or accession numbers are given. Code sharing is not applicable.
“A data sharing statement provided by the authors is available with the full text of this article at NEJM.org.”
“A data sharing statement provided by the authors is available with the full text of this article at NEJM.org.”
Trial registration number is provided (NCT02168829). Methods are detailed. Limitations are explicitly discussed. Conclusions are appropriately cautious. Funding and COI are disclosed. Reporting guideline adherence is not explicitly mentioned but the paper follows CONSORT-like structure.
“OCEAN ClinicalTrials.gov number, NCT02168829.”
“Our trial has limitations.”
“OCEAN ClinicalTrials.gov number, NCT02168829.”
“Our trial has limitations.”
Registered (1 ID: ClinicalTrials.gov). No reporting guideline cited.
Broken references and links
1 finding · worst lowReferences checked against Crossref, OpenAlex and Retraction Watch for retractions and resolvability, plus declared data and code links probed for whether they resolve to content matching the paper.
- References not resolvable to a published paperRecomputed
Checked 36 references by DOI: 31 verified — 5 DOI unresolved.
- UNRESOLVED10.1161/strokeaha.122.040499Covert brain infarction as a risk factor for stroke recurrence in patients with atrial fibrillationCited DOI does not resolve to any Crossref record.
- UNRESOLVED10.1177/17474930211062323Rivaroxaban versus aspirin for prevention of covert brain infarcts in patients with embolic stroke of undetermined source: NAVIGATE ESUS MRI substudyCited DOI does not resolve to any Crossref record.
- UNRESOLVED10.1001/jamaneurol.2024.4550Apixaban to prevent covert infarcts after cryptogenic stroke in patients with atrial cardiopathy: a secondary analysis of the ARCADIA randomized clinical trialCited DOI does not resolve to any Crossref record.
- UNRESOLVED10.1001/jama.2025.12345Long-term anticoagulation discontinuation after catheter ablation for atrial fibrillation: the ALONE-AF randomized clinical trialCited DOI does not resolve to any Crossref record.
- UNRESOLVED10.1056/nejmoa2412987Left atrial appendage closure after ablation for atrial fibrillationCited DOI does not resolve to any Crossref record.
Copyediting
2 minorWording, consistency and formatting errors that need correcting before submission.
No major wording or formatting errors. 2 minor suggestions below.
2 copyedit issues flagged: mostly typo, consistency.
- MINORtypoAuthor affiliations“Deutches Herzzentrum der Charité”→ Deutsches Herzzentrum der CharitéSpelling of 'Deutches' should be 'Deutsches'.
- MINORconsistencyTable 1 footnote“Percentages may not total 100 because of rounding.”→ Consider adding a similar note to other tables if applicable.Consistency in footnotes across tables.
The published work is methodologically robust and the reported statistics are internally consistent for the subset checked. An informed reader should weigh the vague data availability statement and the 5 references that could not be located in registries; these warrant verification or correction via an erratum, but do not undermine the trial's core findings.
- 1.HIGHotherVerify or correct the 5 references not found in any registry: 'Covert brain infarction as a risk factor for stroke recurrence in patients with atrial fibrillation' (10.1161/strokeaha.122.040499), 'Rivaroxaban versus aspirin for prevention of covert brain infarcts in patients with embolic stroke of undetermined source: NAVIGATE ESUS MRI substudy' (10.1177/17474930211062323), 'Apixaban to prevent covert infarcts after cryptogenic stroke in patients with atrial cardiopathy: a secondary analysis of the ARCADIA randomized clinical trial' (10.1001/jamaneurol.2024.4550), 'Long-term anticoagulation discontinuation after catheter ablation for atrial fibrillation: the ALONE-AF randomized clinical trial' (10.1001/jama.2025.12345), and 'Left atrial appendage closure after ablation for atrial fibrillation' (10.1056/nejmoa2412987).References that cannot be located in any registry may be fabricated or have incorrect DOIs, which is a serious integrity concern that must be resolved.
- 2.HIGHdata codeExpand the data availability statement in the manuscript to specify the access mechanism (e.g., 'de-identified data available on request from the corresponding author with a data-sharing agreement') and any conditions or timeline.The current statement is vague and does not tell readers how to obtain the data, which undermines reproducibility.
- 3.MEDIUMreportingExplicitly state adherence to a reporting guideline (e.g., CONSORT) in the methods or acknowledgments.Explicit guideline adherence improves transparency and is expected for clinical trials.
- 4.MEDIUMethicsClarify the regulatory compliance framework (e.g., Declaration of Helsinki) in the ethics section.Reviewer 1 noted this as inadequate; adding the framework strengthens the ethics reporting.
- 5.MEDIUMdata codeIf applicable, deposit de-identified individual patient data in a recognized repository (e.g., Vivli, YODA) and provide the accession or DOI in the data availability statement.A repository deposit with an accession number would make the data sharing statement concrete and verifiable.
- 6.MEDIUMdata codeConsider adding a statement about the availability of the statistical analysis code, if any custom code was used, in a public repository.Sharing analysis code enhances reproducibility, though it is not required for this trial.
- 7.LOWcopyeditFix the typo in the author affiliations: change 'Deutches Herzzentrum der Charité' to 'Deutsches Herzzentrum der Charité'.Correct spelling is expected in a published manuscript.
- 8.LOWcopyeditAdd a footnote to other tables similar to Table 1's 'Percentages may not total 100 because of rounding' if applicable.Consistent footnotes across tables improve clarity and professionalism.
The star rating is the report’s one-glance summary. Every paper starts at 5★ and loses stars for the concrete problems the review finds — so a rating is never a vague average, it’s a running total you can read line by line under “How this rating was calculated.”
- Reporting — 8 dimensionseach dimension that fully fails−½★
- each dimension partially met−¼★
- Statistics · Integrity · Claimseach serious problem−1★
- each medium problem−½★
- Citationseach retracted or unverifiable reference−¼★
- Copyeditonly when the manuscript needs a full edit−½★
The rating never drops below 1★, and a demonstrable critical failure (an impossible statistic, a proven ethics violation) caps it at 1★ on its own — so the stars can never look healthy when the verdict is CRITICAL.
The rating draws on a panel of agents. Three independent Kaimen Rigor reviewers grade the eight dimensions below across several independent passes (the shown verdict is their majority vote — steadier than any single run), isolate the paper’s major claims and check its own evidence backs them, and flag integrity concerns. Alongside them, a citation agent resolves every reference against Crossref, OpenAlex, and Retraction Watch; a statistics agent recomputes reported tests; and rule-based checks verify that declared data/code links actually resolve. Full text is required — an abstract-only submission is not analyzed.
Graded against NIH, MDAR, ARRIVE 2.0, CONSORT, EQUATOR, and RRID guidelines. A dimension that doesn’t apply to the study type is skipped, never penalized.
This Kaimen Rigor review is model-assisted and is not a substitute for formal expert review. It complements human evaluation by surfacing potential methodological concerns — verify each finding against the source.