Not yet independently audited

This page was written from verified sources but has not yet passed the independent citation audit. Treat its claims as provisional.

2026-10-28 — Periodic literature sweep (Grok; independent audit pending)¶

Searches run. Read CONVENTIONS.md, CLAUDE.md, SOURCES.md, INDEX.md, OPEN-QUESTIONS.md and prior LOG entries first. PubMed E-utilities esearch/efetch: the twelve topic queries of the 2026-10-21 manifest, reused verbatim, plus a Cochrane journal filter per topic, a broad CDSR query ("Cochrane Database Syst Rev"[jour] AND (aort* OR Marfan OR aneurysm*)), a systematic-review/meta-analysis filter and an RCT publication-type filter. Each topic used mindate=2026/10/14, maxdate=2026/10/28, both datetype=pdat and datetype=edat. Result: 182 unique PubMed records. Cochrane: 0 records in every PubMed CDSR query; this is a PubMed/window result, not an assertion that no relevant review exists. Europe PMC REST search (FIRST_PDATE 2026-10-14 to 2026-10-28) per topic plus a non-MEDLINE query: 235 unique source/record pairs; Cochrane hitCount 0; non-MEDLINE hits were case reports, methods preprints or off-topic PMC records — none retained as DOI-only. Crossref queried for every retained DOI (retraction check) without a mailto. ClinicalTrials.gov v2 API: all 54 tracked NCT records fetched individually. ISRCTN API ("aortic aneurysm" OR "aortic dissection" OR Marfan, limit=1000): 121 records. Guideline hubs (ACC, ESC, EACTS, ESVS), NVR reports, GBD listing, and patient-organisation homepages (Marfan Foundation news, Aortic Dissection Awareness UK & Ireland, John Ritter Foundation, Marfan Trust, VASCERN, THINK AORTA/ADCT) were fetched. Not queried / blocked: WHO ICTRP portal attempted and not used as a search engine (JS-driven); ChiCTR/CTIS/CTRI/JPRN not independently retrieved; OpenAlex, DOAJ and Unpaywall skipped (LIFEBANK_CONTACT_EMAIL unset). Cochrane Library/CENTRAL and SciELO/LILACS are HTTP 403 and were not searched. Every query string and returned ID is in the sweep manifest.

Trial sweep. One status change among 54 NCT records: ARISE III (NCT06827990) Recruiting → Active, not recruiting (last update 2026-10-23); enrolment 112 and primary completion 2030-09-01 unchanged; no results. NCT07166133 (TBE post-approval) remains recruiting with a 2026-10-23 registry update and no other field change. The other 52 records match the 2026-10-21 snapshot on status, last-update date, completion estimates, enrolment and results flags. EXTRA (NCT04197648) is still recruiting, estimated primary completion 2026-10, no results. SUNDAY (NCT05215587) still recruiting, primary completion 2026-12-31, no results. EARNEST remains ISRCTN version 28, last updated 2026-10-06. The sole ISRCTN update since 2026-10-14 (ISRCTN12054536) is unrelated to aortic disease.

Retraction screen. All eight retained PMIDs: no Retracted Publication/Expression of Concern publication type and empty Crossref updated-by. The window contained a published erratum (PMID 42923748) for an unrelated type A left-subclavian paper; that PMID is cited nowhere in this condition.

Triage. Of 182 PubMed records, 8 retained. Most of the rest were case reports, TAVI/valve or AAA work, single-centre perioperative series, mechanism/bioinformatics papers, or records already screened in the 2026-10-21 unique set. Excluded but noted: VQI n=50 single-branched TEVAR (PMID 42946304); B-SAFER distal-reintervention report (42918407); FET/AMDS proportional meta-analysis (42936324); German biobank design paper with no genomic results (42931949); STS borderline-left-heart consensus (42901411, congenital, not TAA). No landmark deep note.

What changed. - wiki/clinical-trials-landscape.md (already draft): ARISE III status update; Nordic TBE consecutive series (Karelis 2026, PMID 42913781) added beside the Brinkman pivotal readout. - wiki/surgical-and-endovascular-repair.md (already draft): one sentence on the Nordic TBE series. - wiki/medical-therapy.md (already draft): TAA-specific statin-growth MD from the Rodríguez-Palomares SRMA full text (PMC13706600). - wiki/genetics-of-taa.md: demoted curated → draft. Yale 3,113-patient family-history rates and WES VUS caveat (Harling 2026, PMID 42938321); abstract only. - wiki/epidemiology-and-natural-history.md (already draft): adiposity–dissection meta-analysis (Dillman 2026, PMID 42902107) and Yale family-history update. - wiki/guidelines.md (already draft): AHA 2026 aortic-surgery scientific statement, scope-only (DeAnda 2026, PMID 42910416). - OPEN-QUESTIONS.md: OQ-5 and OQ-14 sharpened; count stays 41. - literature/statistics/STATISTICS.md: adiposity RRs; Johannesburg in-hospital mortality 21.9% (Mmabatswa 2026, PMID 42926805); Yale family-history/WES row. - literature/guidelines/REGISTRY.md: #32 AHA scientific statement. Count 31 → 32. - literature/BIBLIOGRAPHY.md: 8 new entries. Unique PMIDs 378 → 386. - INDEX.md: counts and genetics status. Every edited file has last-curated: 2026-10-28. No page was promoted.

Regional evidence. Johannesburg outcomes are PubMed-indexed African data, not a Latin American literature search. SciELO/LILACS could not be searched (HTTP 403), so nothing here supports an absence claim about Latin American literature.

Next sweep should chase. - EXTRA (NCT04197648) results after the October 2026 estimated primary completion; SUNDAY's December 2026 estimate; EARNEST via ISRCTN. - Peer-review status of the Horjus CCB MR preprint and the Murdock rare-variant/syndromic-gene preprints. - Extract recommendations from the EACTS valve-sparing consensus (PMID 42839658) if a full text becomes reachable; Brinkman 77-versus-79 enrolment reconciliation. - ChiCTR2600123230 (CoQ10 type A feasibility protocol, PMID 42941619) if the ChiCTR portal can be retrieved. - Carry forward: ACC/AHA gene-specific table, historic untreated-dissection mortality sources, rehabilitation-denominator full text, unsampled surgical statistics (partially addressed by Johannesburg; high-volume elective series still thin). - A different engine must audit these edits before genetics-of-taa.md returns to curated.

Audit status. Authored by Grok. A different engine must audit, particularly the SRMA TAA-subgroup numbers, the ARISE III status change, and the Yale WES VUS caveat.

2026-10-14 — Periodic literature sweep (Claude Opus 5.5; independent audit pending)¶

Searches run. Read CONVENTIONS.md, CLAUDE.md, SOURCES.md, INDEX.md and the previous LOG entries first. PubMed E-utilities esearch/efetch: the twelve topic queries of the 2026-09-30 manifest, reused verbatim, plus a Cochrane query ("Cochrane Database Syst Rev"[jour] AND (aort* OR Marfan OR aneurysm*)), a systematic-review/meta-analysis filter and an RCT publication-type filter. Each was run in three windows: Entrez date 2026/10/07–2026/10/14 (primary, the requested window), publication date over the same range, and Entrez date 2026/10/01–2026/10/06. The third was a gap catch-up: the last sweep ended at publication date 2026/09/30 and today's window started 2026-10-07, so without it six days would have gone unswept. Result: 117 unique PubMed records. Cochrane: 0 records in every window; no new or updated Cochrane review touches this condition. Europe PMC REST search (first-publication 2026-10-01 to 10-14; core, dissection, syndromic and non-MEDLINE-only queries) surfaced 3 topically relevant records PubMed lacks, all preprints (CACNA1C MR, TNXB, biaxial mechanics); one was retained. Crossref was queried for every retained DOI (retraction check, and resolution of the DOI-only preprint), without a mailto. ClinicalTrials.gov v2 API: all 52 tracked NCT records fetched individually, plus a Thoracic Branch Endoprosthesis term search. ISRCTN API ("aortic aneurysm" OR "aortic dissection" OR Marfan, 121 records, 5 updated since 2026-09-30). Guideline and patient-organisation sites were fetched directly (list in the manifest). Not queried: WHO ICTRP portal (JS-driven; ISRCTN queried directly instead), EudraCT/CTIS, ChiCTR, CTRI, JPRN, OpenAlex, DOAJ, and Unpaywall (skipped because LIFEBANK_CONTACT_EMAIL is unset). Cochrane Library/CENTRAL and SciELO/LILACS are HTTP 403 and were not searched. Every query string, translation and returned ID is in the sweep manifest.

Trial sweep. No change in any of the 52 tracked registrations since the 2026-09-30 baseline: status, last-update date, completion dates, results flags and enrollment are all identical. EXTRA (NCT04197648) is still recruiting with estimated primary completion 2026-10 and no results. SUNDAY (NCT05215587) is unchanged at 2026-12-31. RADAR (NCT06044259) still shows the 2027-09-30 primary completion recorded last time. Newly cited: NCT02777528 (GORE TBE Zone 0/1 pivotal, results posted 2025-06-25), mapped from the Brinkman 2026 paper; NCT07166133 (its post-approval study, recruiting); and ISRCTN30838068 (EARNEST), a UK 470-patient RCT of early TEVAR versus best medical therapy in uncomplicated type B dissection. EARNEST has been recruiting since 2025-04 and is the second randomised trial on the question SUNDAY addresses. Earlier sweeps never saw it because they searched ClinicalTrials.gov only.

Retraction screen. All 11 retained PMIDs: no Retracted Publication/Expression of Concern publication type and an empty Crossref updated-by. The DOI-only preprint is Crossref posted-content with no update relations. The window contained a retraction notice (PMID 42866932) for PMID 36132983, an aortic-dissection VSMC paper. That PMID is cited nowhere in this condition.

Triage. Of 117 PubMed records plus 3 topically relevant Europe PMC preprint records, 12 records retained (11 PMIDs + 1 DOI). Most of the rest were case reports, TAVI/valve or AAA work picked up by broad terms, single-centre perioperative series, or preliminary mechanism/bioinformatics papers. Excluded but noted: Jämsen histopathology meta-analysis (PMID 42861241), Turner CHD cohort (42838142), Chinese type B WGS/HLA study (42857991), AAORTA arch risk score (42869089), TNXB and biaxial-mechanics preprints. No landmark deep note: none changes practice.

What changed. - wiki/clinical-trials-landscape.md (already draft): added a TBE Zone 0/1 row and 1-year readout (Brinkman 2026, PMID 42863733), with the registry's posted 12-month primary-endpoint success (72.7% aneurysm / 75.0% dissection). The paper reports 77 patients but the registry lists 79 enrolled, and the abstract prints no NCT number, so the paper-to-registry match is stated as resting on identical device, zone and design. Added the EARNEST row and context sentence. - wiki/surgical-and-endovascular-repair.md (already draft): one sentence on the TBE 1-year results in the arch section, single-arm caveat included. - wiki/medical-therapy.md: demoted curated → draft. Added the CACNA1C aortic-eQTL Mendelian-randomisation preprint (Horjus 2026, medRxiv doi:10.64898/2026.10.02.26364570) to the CCB section and sharpened the CCB open-question bullet. Its figures (β=−0.18, P=0.005; β=−0.08, P=0.005) come from the Europe PMC abstract only. No full text was read, so an auditor must check them. - OPEN-QUESTIONS.md: OQ-6 sharpened (five-biobank rare-variant burden preprint, syndromic-gene preprint, ADAMTS6). OQ-30 sharpened (Schwartz 2026: complete cascade screening in 33% of HTAAD families). New OQ-41 (are CCBs harmful to the human thoracic aorta?). - literature/statistics/STATISTICS.md: added the sport-screening FoCUS prevalence row (Macquaire 2026, labelled as a screening-clinic sample, not population prevalence) and the cascade-screening delivery row. Corrected conflict #3 and the Obel row to match the audited 2026-10-07 OQ-27/D9 revision. The text now distinguishes in-hospital from 30-day mortality, adds Bossone 2024, and states that prehospital deaths are excluded from the Danish series. The earlier text said "flat in validated data" without the endpoint qualifier. All three PMIDs (26205591, 36321467, 39145396) were re-fetched live today. wiki/aortic-dissection.md was re-read for the same framing and does not carry it; it reports only IRAD in-hospital trends, so no edit was needed. This closes the 2026-10-07 follow-up. - literature/guidelines/REGISTRY.md: #29 EACTS valve repair/valve-sparing root replacement consensus (PMID 42839658), #30 EACTS iatrogenic aortic dissection (42839659), #31 EACTS civil aviation after cardiothoracic/aortic treatment (42839649). Scope only, no recommendations extracted (full texts not read). Count 28 → 31. - literature/BIBLIOGRAPHY.md: new "October 2026 sweep additions" section with 12 entries. Unique PMIDs counted mechanically: 359 → 370. The pre-sweep file held 359 distinct PMID: strings although INDEX said 360. - literature/patient-voice/sources.md: Marfan Foundation Barcelona meeting/Spanish symposium listings and Aortic Dissection Awareness UK & Ireland October listings (hub summaries only, no private names). - INDEX.md: counts, medical-therapy status, curation-state paragraph. Every edited file has last-curated: 2026-10-14. No page was promoted.

Regional evidence. No question touched this sweep turns on regional transportability. Note anyway that SciELO/LILACS could not be searched, so nothing here supports an absence claim about Latin American literature.

Next sweep should chase. - EXTRA (NCT04197648) results/status after its October 2026 estimated primary completion; SUNDAY's December 2026 estimate; EARNEST via the ISRCTN API, not ClinicalTrials.gov. - Extract recommendations from the EACTS valve-sparing consensus (PMID 42839658), the one new document that bears on root-aneurysm surgery, and read the aviation consensus full text (PMC13642968) for anything aortic-specific. - Watch whether the Murdock rare-variant and syndromic-gene preprints and the Horjus CCB MR preprint reach peer review, and whether the new candidate genes (FNDC3B, ROCK1, ADAMTS6 and others) receive ClinGen curation before any gene-table edit. - Get the Brinkman full text to reconcile 77 vs 79 and confirm the NCT. - Make future sweeps query ISRCTN (and EudraCT/CTIS where reachable) as standard. The EARNEST miss shows a ClinicalTrials.gov-only trial sweep is incomplete here. - Pre-existing gap: Pape 2015, Obel 2022 and Bossone 2024 are cited on wiki/statistics pages but absent from BIBLIOGRAPHY.md. Worth a consistency pass, which was out of scope today. - Carried forward: rehabilitation-denominator full text, ACC/AHA gene-specific table, historic untreated-dissection mortality sources, unsampled surgical statistics.

Audit status. Authored by Claude Opus 5.5. A different engine must audit these edits, particularly the medical-therapy CCB paragraph, the TBE paper–registry mapping and the STATISTICS conflict-#3 rewrite, before medical-therapy.md returns to curated.

2026-10-07 — dot D9 and OQ-27 corrected: the asserted contradiction was an endpoint mismatch (Claude)¶

What prompted this. The novelty gate (tools/verify-dot.sh) assessed dot D9 with three engines. All three returned narrowed and all three rejected the row's premise rather than its novelty. Verdicts: novelty/thoracic-aortic-aneurysm/D9-{claude,codex,grok}.md.

What was wrong¶

D9 asserted a contradiction that does not exist. IRAD reports type A in-hospital mortality (31%→22%, 28 referral centres, 1995–2013; Pape 2015, PMID 26205591). The Danish cohort reports 30-day mortality (22.0%, no significant change 1996–2016; Obel 2022, PMID 36321467). Different clock, different denominator. As codex put it, reading them as a direct contradiction is not supported by these sources.
OQ-27 described the Danish cohort as validated national data in implied contrast to referral bias. Re-read live: Obel validated only cases "registered with codes in the Danish National Patient Registry at time of admission to a hospital with available medical records" — 3,023 of 5,018 registered cases (60.2%), 1,620 type A after exclusions. Prehospital deaths are not in the 22%. Survivorship bias therefore cannot be the explanation.
The dot was out of date. The 2024 IRAD update (Bossone 2024, PMID 39145396; 61 centres, 15 countries, 1996–2022, tertiles of 3,785) reports type A hospital mortality 24.1%→16.7% (p<0.001) with surgical repair rising 89.1%→92.5% — and no change in 3-year survival (p=0.296). Both halves of the supposed conflict now sit inside one registry. It was cited nowhere in this condition.

What changed¶

D9 rewritten to state the endpoint/denominator mismatch and the residual that survives it: no study has put a referral registry and a chart-validated national cohort on one estimand and partitioned the decline into operative-rate change, case-mix, transfer selection and risk-adjusted case fatality. OQ-27 rewritten to match, with both corrections recorded in the text so the error is visible rather than quietly removed. Bossone 2024 now cited in both.

Searched. Nothing new: this entry applies verdicts already on disk. The three engines between them recorded their query strings in the verdict files, including a structural citation-intersection test.

Next time. No wiki page was edited, so no status changed. The claim in wiki/aortic-dissection.md and literature/statistics/STATISTICS.md that type A mortality is "flat in validated data" should be re-read against this correction — it may carry the same framing error.

Provenance. Every PMID written in this entry and in the rows above was retrieved from a live PubMed E-utilities query in this session, and every one was checked for retraction against both the PubMed publication types and the Crossref updated-by field (per the new SOURCES.md rule). All were clean.

Audit status. These edits were authored by Claude Opus 5, which also ran the gate that found the defects. A different engine must therefore audit them before any demoted page returns to curated — a model does not mark its own homework (CLAUDE.md, "Audit").

Curation log — thoracic aortic aneurysm¶

Newest entries first. Every content-changing session appends an entry: date, what changed, what was searched, follow-ups for next time. See CLAUDE.md.

2026-09-30 — Periodic literature sweep (Codex; independent audit pending)¶

Scope and searches. Read CONVENTIONS.md, CLAUDE.md, INDEX.md and the previous LOG entries before writing. PubMed E-utilities esearch used mindate=2026/08/31, maxdate=2026/09/30, datetype=pdat, retmax=1000, followed by efetch metadata/abstracts. Twelve topic searches returned 348 unique records (347 journal records and one general StatPearls study guide): all thoracic 91; epidemiology 114; genetics syndromes 130; bicuspid 21; mechanisms models omics 46; imaging biomechanics 141; risk biomarkers 75; medical exercise safety 96; surgery devices 192; dissection 134; guidelines 90; patient experience 86. Exact queries, translations, returned IDs, selected abstracts and record-level trial checks are preserved in the sweep manifest. The date range is inclusive and includes September print issues with earlier online dates; it is not a first-indexed-since filter. An additional live efetch verified the missing 2025 sports statement and retained citation context. A quoted-title search initially missed that statement; direct identifier retrieval resolved it. No PMID or NCT was introduced from memory.

Trial sweep. Fetched all 52 cited NCT records individually through https://clinicaltrials.gov/api/v2/studies/{NCT_ID}; no fetch failed. Existing recruitment-status entries matched. Five records have posted results, all predating the window and with no in-window registry update; no new or updated posted results were identified. Only TITAN:SvS and RADAR have in-window update-posted dates. TITAN remains recruiting, 610 planned, primary completion 2035-05-31; sites increased from 29 to 31 (NCT03536312). RADAR remains recruiting, 72 planned, but estimated primary/study completion moved to 2027-09-30/2028-09-30 from September 2026/2027 (NCT06044259). August raw registry snapshots were unavailable: comparisons use the existing condition text plus current update/results-post dates, not reconstructed version history. August landscape counts were explicitly kept historical.

Triage. Added 14 date-window papers and one older guideline correction to the bibliography; 334 screened records did not merit addition. Most excluded records were case reports, commentary, unrelated broad-query hits, narrow technical series or preliminary biomarker/mechanism studies. Retained clinical synthesis is limited to episodic growth, long-term Marfan root-repair tradeoffs, monogenic–polygenic risk, an observational IRAD comparison, rehabilitation associations and contrasting patient-experience evidence. A few additional registry, surveillance and preclinical papers are bibliography-only. No landmark deep note was justified. The four retained September-issue papers first available online before the window are growth, Marfan root surgery, rehabilitation and CIRSE standards; they are catch-up coverage within the requested publication-date filter, not claimed September discoveries.

Guidelines, statistics and patient organisations. Checked AHA/ACC, ESC, EACTS, ESVS and VASCERN sources, the NVR publication hub and GBD sources; checked John Ritter Foundation, Aortic Dissection Awareness UK & Ireland, Marfan Foundation/Trust and VASCERN material. Added the September tactical-athlete statement and September-issue CIRSE standards. Corrected two earlier omissions: the 2025 sports framework replaces the registry’s 2015-current claim, and the March 2026 VASCERN version-2 pathway updates its earlier pathway. Registry now has 28 documents. Scope-only entries do not invent unextracted thresholds. The new GBD figure still combines thoracic and abdominal aneurysms; OQ-29 remains open. No new September NVR thoracic annual report was identified on the retrieved hub. Added the September awareness-week material and VASCERN youth-guide listing, without interpreting educational activity as improved outcomes. The youth booklet itself and the UK charity’s ESVS-news detail page were inaccessible; no detail-only claims were imported.

Changes. Updated six wiki pages: clinical trials, omics, natural history, surgery, guidelines and patient experience; updated BIBLIOGRAPHY.md, REGISTRY.md, STATISTICS.md, patient-voice/sources.md, OPEN-QUESTIONS.md and INDEX.md. Corrected the inventory to 360 unique bibliography PMIDs (the pre-sweep file contained 345, although INDEX said 344). Sharpened OQ-15/18 (genetic modifiers versus clinical prediction), OQ-29 (pooled burden), OQ-34 (guidance now exists), OQ-36 (population/trajectory differences) and OQ-38 (exercise trial pending). Corrected D8’s zero-PROM-trials assertion: TITAN registers SF-36; EXTRA is a recruiting exercise trial with no posted results. Added the requested cross-link from guidelines to the previously audited pre-pregnancy table instead of duplicating it. All edited Markdown has last-curated: 2026-09-30 and status: draft; no page was promoted. Author: Codex. The separate auditor has not yet run.

Verification and limits. New/changed-line PMIDs were checked against this session’s live E-utilities records, NCTs against live API responses, and local links/frontmatter/reference placement were checked mechanically. Claims use the retrieved abstracts unless an official-source excerpt/PDF is explicitly identified. The rehabilitation abstract’s cohort counts do not reconcile (1,710 stated versus group counts totalling 1,612), so group denominators were deliberately not imported; clarify with full text before reusing them. No git commands, repository metadata, publication or frontend build were run; publishing is handled after this sweep as instructed.

Next sweep. Recheck EXTRA’s October 2026 estimated primary completion and results posting (NCT04197648), SUNDAY’s December 2026 estimate (NCT05215587), and RADAR’s revised dates. Seek full text for the rehabilitation denominator discrepancy, CIRSE/tactical-athlete recommendations and genetic-risk calibration beyond diameter; track external replication and prospective outcomes, not just associations. Retrieve the VASCERN youth booklet and UK charity’s ESVS patient-research detail page if accessible. Existing full-text/audit follow-ups remain: ACC/AHA gene-specific table extraction, historic untreated-dissection mortality sources, and unsampled surgical statistics. Preserve the distinction between September print catch-ups and genuinely new online records. Use this sweep’s manifest as the trial-status baseline next time.

2026-08-31 — Audit pass: red-flags-and-safety-concerns.md → curated¶

Scope: the one remaining draft page in this condition. No other TAA page touched.

Searches run. PubMed esummary + efetch for all 44 unique PMIDs cited on the page (existence, journal, year, title vs. the reference list); efetch abstracts for all of them; full text of the 2022 ACC/AHA aortic guideline via PMC (PMC9876736 ← PMID 36322642) to settle the two [unverified] flags.

Result: 44/44 records exist and match their reference-list entries; 0 fabricated or wrong-paper citations. Every numeric claim on the page was checked token-by-token against the source abstract, then the high-leverage ones (Hagan sign frequencies, Lovatt misdiagnosis rate, Rogers ADD-RS sensitivity, Nazerian failure/efficiency, Tsai false-lumen mortality, Chaddha surveillance HR, Mariscalco relative yield, Kamel IRR, Hibino meta-analysis, Smith diameter failures, Hatzaras case series) re-read in context for direction as well as value. All correct.

Changes made. - Both [unverified] flags resolved against the guideline's own text, not a secondary source. (1) Symptomatic aneurysm: §"Aneurysms of the Aortic Root and Ascending Aorta" recommendation 1 is "In patients with aneurysms of the aortic root and ascending aorta who have symptoms attributable to the aneurysm, surgery is indicated" (COR 1, LOE C-LD), with the ≥5.5 cm / ≥5.0 cm / growth-rate thresholds all scoped to asymptomatic patients — the page now states this with grades instead of flagging it. (2) Pre-pregnancy thresholds: §8.3 transcribed into a condition-by-condition table (Marfan >4.5 cm COR 1 C-LD, 4.0–4.5 cm COR 2b; LDS by gene; nsHTAD ≥4.5 cm; Turner ASI ≥2.5 cm/m²; BAV ≥5.0 cm; sporadic ≥5.0 cm), with the observation that every threshold is C-LD or C-EO. - Exercise section strengthened with verified guideline text: the "no consensus on intermediate intensities" claim is now quoted exactly ("No uniform consensus exists about the safety of intermediate-level static and aerobic exercise"), paired with the two graded recommendations (avoidance COR 1 C-EO; 30–60 min mild-to-moderate aerobic 3–4×/week COR 2a C-LD) and the >300 mm Hg Valsalva figure the guideline gives as rationale. - Over-precision corrected (Anagnostopoulos, PMID 34140225): the 86% thoracic-graft sensitivity holds only under the analysis that assigns MAGIC-"suspected" cases to not diseased; the page stated it unconditionally. Now carries the condition, the CI (73–95), and the cohort denominators. - Greve cocaine series age SD corrected 8.7 → 8.67 (as published). - Reference 2 (PMID 36334952) was an orphan — never cited in the body. Annotated as the simultaneous JACC co-publication of reference 1, with a note that guideline text is cited to the Circulation version. - Frontmatter: status: draft → curated, last-curated: 2026-08-31.

Follow-ups. The §8.3 table duplicates material that belongs on guidelines — check that page carries the same thresholds with the same grades, or cross-link rather than restate. PMC9876736 is the full guideline text and is now known to be retrievable; other TAA pages citing PMID 36322642 for specific COR/LOE claims could be re-verified against it cheaply.

2026-08-28 (later) — Red-flags page; trials stage/safety enrichment¶

Added wiki/red-flags-and-safety-concerns.md (draft, 44 verified PMIDs): dissection symptom frequencies, misdiagnosis quantification (33.8%), screening triggers, graded cautions (hypertension RR 3.07 dose-dependent; pregnancy IRR 4.0; weightlifting case-series-only; cocaine; fluoroquinolone cross-link), surveillance and post-repair warning signs. Needs its own audit pass; two [unverified] flags (symptomatic-aneurysm COR text; pre-pregnancy numeric thresholds) deferred to a guidelines full-text pass.
Enriched wiki/clinical-trials-landscape.md (stays curated): trial-stage primer; per-trial phases; published efficacy + safety readouts (PERSEVERE 30-day/1-year, Thoraflex IDE + FDA approval, PHN atenolol/losartan AE profile, Marfan Sartan, N-TA³CT null, INSTEAD-XL 5-year added as SUNDAY precedent); 16 ongoing trials re-verified (no status changes; ARISE III acronym landed; several completion dates extended).
BIBLIOGRAPHY.md grew to 344 unique PMIDs (red-flags + trials-enrichment merges).

2026-08-28 — Full citation audit; ask capability; weekly sweep armed¶

Audit pass (5 parallel auditors, all 19 pages): every inline citation re-verified against live PubMed metadata/abstracts (full texts via PMC/publisher where claims demanded it); all 49 NCT registrations re-verified against ClinicalTrials.gov (no status changes since 2026-08-27 — first trial sweep on record). Results: 0 fabricated PMIDs, 0 wrong-paper citations across ~600 claim-citation pairs; ~15 substantive fixes applied, the biggest: Paruchuri relative-risk conflation (346.8 vs-distribution ≠ vs-<3.5 cm, which is ~6,305×; fixed on 2 pages + statistics sheet), 2 wrong LOE cells + 1 missed LOE-A item in the 2022 ACC/AHA tables (58 COR/LOE rows re-verified against the guideline full text), Lacro 2014 blinding mischaracterized, FDG-aortitis claim cited to an unrelated retroperitoneal-fibrosis study (replaced), Quint reproducibility mis-attributed inter- vs intra-observer, Yacoub 4.6% denominator. All 19 pages promoted to curated.

[unverified] flags 15 → 2: resolved with real sources — Renard 2018 per-gene ClinGen tiers (from PMC full text; gene table + paper note updated), 2024 EACTS/STS numeric thresholds (full EJCTS PDF retrieved; new verified section in guidelines.md incl. the 52.5-mm hinge revision and gene-specific tables), acute/chronic dissection boundary (2022 guideline verbatim + SVS/STS 2020), large-animal models (3 porcine + 1 ovine primary papers; key negative: no root/ascending large-animal model exists), sST2 (a documented replication failure: AUC 0.97 → 0.63 → uninformative), desmosine, lumican, MR blood-pressure and lipids (remnant cholesterol causal for AAA, null for TAA — sharp negative test of the atherosclerotic hypothesis). Converted to explicit evidence-gap statements: FDG threshold, prospective proteomics, gene-editing rescue. Remaining 2 notes on guidelines.md document an extraction limitation (ACC/AHA per-gene table: PMC strips gene symbols; publisher sites 403/CAPTCHA — all routes recorded in-page).

Missing SMAD3 discovery paper found (van de Laar 2011, Nat Genet, PMID 21217753) and added to the gene table.

Bibliography: +27 audit-addition entries (all re-verified this session), 1 orphaned entry removed (Thuermel), 2 print-year fixes → 312 unique PMIDs. Statistics sheet: 36-source spot-check, 1 error fixed (Bickerstaff "presented with" vs "occurred"). Guidelines registry: 24/24 documents verified clean.

System changes: /ask project skill added (.claude/skills/ask/); OPEN-QUESTIONS.md gained OQ-35–40 (patient-centered) and the "Dots not yet connected" table (12 cross-domain junctions); frontend gained the Ask panel (client-side passage retrieval + comment-to-Claude path for synthesized answers) and an Unconnected-dots nav entry; weekly literature sweep scheduled (Mondays ~09:00 local, task taa-weekly-literature-sweep — includes artifact-comment answering and auto-republish).

Follow-ups: ACC/AHA per-gene thresholds still need a full-text route (print copy or authenticated access); ePub-vs-print years could be normalized bibliography-wide in one pass (list in this entry's audit reports); statistics §4 surgical-outcome block not yet sampled (Krüger, Rylski, Farag, Benedetto, Dobaria, Catalano, Cefarelli, Harky, Bottle, Brown, Grant); Davies 2006 vs Zafar ASI band cross-check; duplicate-PMID topic lines (by design, but confirm cited-by lists stay consistent); blocked web sources from 08-27 list still pending.

2026-08-27 — Initial build¶

Created repository scaffold (README, CONVENTIONS, CLAUDE.md protocol, templates) and the TAA condition tree.
Built all 18 canonical wiki pages via five domain research passes (foundations/natural history, genetics, mechanism, clinical management, research frontier), each grounded in live PubMed searches; trials page grounded in live ClinicalTrials.gov queries.
Created literature layer: per-domain bibliography fragments, master BIBLIOGRAPHY.md, landmark paper notes.
Synthesized OPEN-QUESTIONS.md from cross-domain gaps surfaced during the build.
All pages start at status: draft — no page is curated until it passes a verification audit (see CLAUDE.md "Audit" workflow).

Frontend: published as a private artifact at https://claude.ai/code/artifact/571412b5-22f6-436b-b6cc-03f7e51b45a7 (build: python3 tools/build_frontend.py → frontend/taa-wiki.html; republish to the same URL after every sweep).

Build inventory: 19 wiki pages (all draft); BIBLIOGRAPHY.md 286 unique PMIDs; 16 landmark notes; guidelines registry 24 documents (5 superseded chains); statistics sheet ~116 figures + 9 documented conflicts; patient-voice layer (17 orgs verified by direct fetch, 10 themes, 4 files); OPEN-QUESTIONS.md 40 questions (8 Tier-1). Frontend build tooling in tools/.

Follow-ups for next session(s): - Audit pass: spot-check citations on 2–3 pages (claims vs abstracts), promote clean pages to curated. Overview and guidelines pages already passed an informal read-through. - [unverified] items to resolve (15 flags across 6 pages): per-gene ClinGen tier assignments (needs Renard 2018 full-text table); 2024 EACTS/STS numeric thresholds (no retrievable full text yet); gene-level HTAD thresholds in 2022 ACC/AHA Table (PMC rendering dropped gene symbols); 14-day acute/chronic dissection boundary source; desmosine/sST2/lumican primary TAA cohorts (may not exist — if confirmed absent, convert to open-question evidence); FDG threshold for degenerative TAA; porcine/ovine model protocols. - Full-text pulls needed: Hirst 1958 (PMID 13577293) and Anagnostopoulos 1972 (PMID 4557973) internals behind the "1–2%/hour" claim (OQ-31). - Blocked sources to retry: vascern.eu (403), healthdata.org GBD factsheet (403), ctdna.org.au (refused), Marfan Network Japan; patient-org coverage gaps: Spain, Italy, Nordics, Latin America. - Literature sweep with date_from: 2026/08; trial status re-check for TITAN:SvS (NCT03536312), ARISE II (NCT05800743), and the 50 catalogued NCTs; guideline watch list in literature/guidelines/REGISTRY.md (ESVS 2026 just published — watch for ACC/AHA focused updates). - Missing SMAD3 discovery paper (van de Laar 2011 Nat Genet) — did not surface in searches; find and add. - After any sweep: rebuild + republish the artifact frontend (CLAUDE.md workflow step 6).

2026-10-07 — independent audit of Claude Opus 5's D9 and OQ-27 edits (Codex, auditor)¶

Records and checks. Independently re-fetched Pape 2015 (PMID 26205591), Obel 2022 (PMID 36321467) and Bossone 2024 (PMID 39145396) through live PubMed E-utilities. Checked each publication type and each DOI's Crossref updated-by field; no retraction or expression of concern was flagged. A live PubMed search for IRAD/Denmark or population-based dissection with standardisation, harmonisation or decomposition terms returned zero indexed abstracts; this is a search result, not proof of global absence.

Figures and correction. None of the specified figures failed: Pape's 31%→22% is in-hospital mortality among 2,952 type A patients at 28 centres; Bossone's 24.1%→16.7% is hospital mortality, with no significant difference in 3-year postdischarge survival (p=0.296); Obel's 22.0% is 30-day mortality in 1,620 validated type A cases from admissions with retrievable records. Claude's categorical "not a contradiction" overcorrected. Different endpoints and inclusion rules block direct comparison but do not resolve an improving-versus-flat trend difference; Bossone's postdischarge analysis also conditions on surviving the hospital episode. Revised D9 and OQ-27 to retain that open question and make the proposed common estimand explicit. Neither row was closed. No wiki status changed.

Judgement. The numerical correction is reliable after this audit; the original claim that the trend disagreement had been explained was not. D9 and OQ-27 can now be used as an unresolved, qualified research question. A harmonised type A admission cohort remains a follow-up.

2026-10-21 — Periodic literature sweep (Codex; independent audit pending)¶

Scope and searches. Read CONVENTIONS.md, CLAUDE.md, SOURCES.md, this log, INDEX.md and OPEN-QUESTIONS.md before edits. PubMed E-utilities: reused the twelve major-topic queries (all thoracic, epidemiology, genetics/syndromes, BAV, mechanisms/models/omics, imaging/biomechanics, risk/biomarkers, medical/exercise/safety, repair/devices, dissection, guidelines and patient experience), plus broad Cochrane, systematic-review and trial-publication screens. Each used mindate=2026/10/14, maxdate=2026/10/21, both datetype=pdat and datetype=edat; added a CDSR journal filter separately for each topic/screen. 44 searches, 87 unique PubMed records fetched with metadata, abstracts and publication types. No CDSR record returned in these PubMed searches; this is a database/window result, not an assertion that no relevant review exists. Entry dates catch late indexing: three retained papers first published before the requested window (beta blockers October 2; CMR October 12; SUNDAY survey September 30), explicitly treated as catch-ups.

Beyond PubMed. Queried Europe PMC REST with a corresponding first-publication window per topic, plus a focused guideline query; 1,184 unique source/record pairs returned. The broad guideline translation was noisy (1,132 results), fully paginated, and supplemented with explicit TITLE_ABS grouping. Additional aorta-relevant non-MEDLINE records were case reports, preliminary methods/model work or preprints; none warranted a DOI-only addition. Re-fetched the 130 Europe PMC MED identifiers absent from the topic-search set through E-utilities so even the manifest's MED identifiers have live PubMed provenance. Queried Crossref for all eight retained papers and the aviation follow-up: no updated-by retraction/expression-of-concern relation, and no corresponding PubMed publication-type warning. Initial Crossref rate limits were resolved by sequential retries. No personal email was read or sent; Unpaywall was not used. Queries, translations, IDs, retraction checks, full-text evidence and trial snapshots are in the manifest.

Trials. Retrieved every one of the 54 cited NCT registrations individually through the ClinicalTrials.gov v2 API. The 52 with structured snapshots last week have identical status, last-update dates, completion dates, enrollment and results flags. The two newly cited TBE records had no structured baseline in that manifest; current data agree with its prose (pivotal results posted June 2025; post-approval study recruiting). Six registrations currently have posted results; no new results posting identified. EXTRA still shows recruiting, estimated primary completion October 2026, no results (NCT04197648); SUNDAY remains recruiting, estimated primary completion December 31, no results (NCT05215587). ISRCTN API search retrieved all 121 results using limit=1000, rather than relying on its ten-record default. The sole in-window update is unrelated to aortic disease. EARNEST remains version 28, last updated October 6 (ISRCTN30838068). WHO ICTRP query returned 403. Retrieved an EudraCT legacy-register search for aortic dissection; no version-history comparison or CTIS coverage is claimed. CTIS, ChiCTR, CTRI and JPRN were not searched. CENTRAL and SciELO/LILACS could not be searched (documented HTTP 403); no conclusion about absence of Latin American or worldwide trial evidence follows.

Guidelines, burden and patient organisations. Retrieved ACC and ESC guideline hubs, EACTS' October 15 announcement and the existing ESVS DTA listing. The EACTS announcement redistributes the three aorta-related documents catalogued last week; no new guideline document retained, registry stays at 31. Read the aviation consensus full text, §5.3 and Table 4: occupational aircrew recertification is its scope; added this boundary to REGISTRY.md. Valve-sparing consensus publisher still returned 403; Europe PMC lists subscription access. Retrieved the NVR reports hub and checked GBD listings; no new TAA-specific registry/burden estimate retained. Checked Marfan Foundation, Aortic Dissection Awareness UK & Ireland, John Ritter Foundation and Marfan Trust material. The attempted VASCERN disease-group URL redirected to its general homepage and is not counted as an HTAD-page check. Added one patient-voice source annotation for the Foundation/ VEDS Movement's October 17 report (updated October 18), preserving its unpublished/not-peer-reviewed research warning and importing no conference efficacy or safety figures.

Triage and changes. Eight new bibliography records; 79 PubMed topic-search records not retained. Six additions are bibliography-only: genetic-testing implementation, post-type-A care delivery, the SUNDAY investigator survey, exercise narrative review, nonsyndromic beta-blocker serial-CT cohort, and the arch-stroke meta-analysis. The last separates comparative from proportional observational estimates; no superiority claim was imported. No landmark note. Two full-text-grounded syntheses warranted page edits:

Imaging: CMR convention-dependent root reclassification (Zelis 2026, PMID 42890866); retained exclusions for BAV/prior surgery/root dissection and lack of clinical-outcome follow-up. Added the figures to STATISTICS.md with results/Figure 3 provenance (the brief report has no numerical table). Sharpened OQ-32 toward prospective validation of measurement conventions, rather than inferring prognostic superiority from perimeter agreement.
Patient experience: early positive adaptation after type A dissection (Zhang 2026, PMID 42879453); Table 1 supports the 20-patient sample, Tables 2–3 the prompts/themes, and full-text limitations show all interviews occurred before discharge. Added the selection/timing qualifiers and sharpened OQ-36 toward longitudinal recovery and intervention testing.

Also refreshed the trials-page verification paragraph, guideline registry, patient-voice source list and INDEX.md. Bibliography 370 → 378 unique PMIDs, plus its existing DOI-only preprint; open-question count stays 41. Imaging demoted curated → draft for independent audit; other edited pages were already draft. All edited Markdown has last-curated: 2026-10-21; no promotion. Edits are limited to this condition. No git command, repository metadata creation, frontend/global build, publication or deploy was run; publishing is handled after completion as instructed.

Verification limits. Full texts were read for the CMR, qualitative, SUNDAY-survey and aviation records. Other retained records are explicitly metadata/abstract-only; their numerical findings were not imported into wiki/statistics pages. Europe PMC MED availability flags lagged direct fullTextXML retrieval for several retained papers; successful full-text retrieval, not a negative index flag, determined access. Mechanical checks verify newly added identifier provenance, frontmatter, reference numbering, inventory counts and new local links; these checks are not an independent content audit.

Next sweep. Recheck EXTRA results after the estimated October completion, SUNDAY's December estimate, and EARNEST via ISRCTN; keep the 54-record baseline here. Seek the beta-blocker cohort full text before numerical synthesis or changing OQ-5, and the genetic-testing EHR paper before interpreting coded ascertainment as actual testing uptake. Review the arch-stroke synthesis in full before changing technique comparisons. Test whether root-convention effects replicate in BAV and predict events; follow the pre-discharge qualitative cohort with longitudinal work. Carry forward valve-sparing consensus full text, TBE 77-versus-79 enrollment reconciliation, CCB and rare-variant preprint peer-review status, rehabilitation denominator inconsistency, ACC/AHA gene-specific table, historic untreated-dissection mortality, unsampled surgical statistics and the earlier bibliography-consistency gap. A different engine must audit this sweep before imaging is re-promoted.