Case file · From a real deliberation · De-identified
An ICU decision: tracheostomy, or antithrombotics?
A real clinical question put before the Consilium panel: four frontier models answered independently,
reviewed each other blind, and the verdict was fact-checked against the live web — with unverifiable
numbers removed.
Medical education — not a substitute for the treating team.
This is a general educational deliberation, not patient-specific medical advice.
— The question —
A 68-year-old male is in the ICU after an out-of-hospital cardiac arrest caused by an acute
myocardial infarction. He underwent successful coronary angiography with coronary artery stent
placement and remains comatose due to suspected hypoxic brain injury. He is intubated but has passed
ventilator weaning assessments and appears capable of breathing spontaneously. The patient is
receiving mandatory antithrombotic therapy because of the recent coronary stent.
Option 1: temporarily interrupt antithrombotic therapy to perform a tracheostomy, allowing
ventilator removal while providing a secure airway.
Option 2: continue antithrombotic therapy without interruption and keep the patient intubated
longer, avoiding the bleeding risk of tracheostomy but accepting the risks of prolonged intubation.
Based on current ICU, cardiology, and critical care guidelines, which approach is generally
preferred? Discuss bleeding risk, stent thrombosis, prolonged ventilation, airway protection,
aspiration, and extubation failure — and cite relevant evidence.
The deliberation below is shown in the panel's original English.
— How this verdict was produced —
①Four models answer independently
②Blind peer review — no names
③The lead drafts one verdict
④Every claim fact-checked on the live web
⑤Corrected verdict — unverified numbers removed
— Where each panelist stood (Phase ①) —
deepseek-v4-proLead
Continue antithrombotics and tolerate longer intubation: the immediate mortality risk of stent
thrombosis outweighs the slower-accumulating harms of a delayed tracheostomy in the early weeks.
gpt-5-mini
Same default; if tracheostomy is forced early, continue aspirin and hold the P2Y₁₂ inhibitor
only with the cardiologist's agreement — never a blanket full stop.
qwen3.6-plus
Option 2 for the first 30 days after stenting; any unavoidable procedure should be done with
aspirin continued and the shortest possible P2Y₁₂ interruption.
gemini-3.1-flash-lite
Delay tracheostomy in the first 1–2 weeks unless the airway destabilises; involve interventional
cardiology and keep any interruption as brief as possible.
Four models, answering independently and reviewed blind — all four converged on the same option.
— The final verdict (Phase ⑤, web-checked & corrected) —
Critical safety point: Never interrupt dual antiplatelet therapy (DAPT)
after recent coronary stenting without direct input from the interventional cardiologist who placed
the stent; premature cessation can trigger acute stent thrombosis, a catastrophic event with high
mortality.
Short answer
In the first 30 days after coronary stent placement—especially for an acute myocardial
infarction—the balance of risks strongly favours continuing antithrombotic therapy and maintaining
endotracheal intubation (Option 2). Elective tracheostomy should be deferred until the highest-risk
period for stent thrombosis has passed, unless an immediate, life-threatening airway or
secretion-management crisis forces intervention. If tracheostomy becomes unavoidable before that
window closes, it should be performed with aspirin continued and the P2Y₁₂ inhibitor held for the
shortest guideline-recommended duration, not with a complete interruption of all antithrombotics.
Why continuing antithrombotics and delaying tracheostomy is preferred
1. Stent thrombosis risk dwarfs the early risks of prolonged intubation
- Stent thrombosis is most frequent in the first 30 days after drug-eluting stent (DES)
implantation, with the majority of events in the first 1–2 weeks
[8][15]. Premature
discontinuation of the P2Y₁₂ inhibitor during this period is strongly associated with early stent
thrombosis and is described as carrying an unacceptable risk
[8][15]. Reported mortality
varies across cohorts but is substantial [5][15].
- Current ACC/AHA and ESC guidelines recommend uninterrupted DAPT for at least 1 month after
bare-metal stent placement and 6–12 months after DES for acute coronary syndrome, unless high
bleeding risk mandates a shorter course
[8][10]. If a procedure is
unavoidable, aspirin should be continued and the P2Y₁₂ inhibitor held for the minimum washout
(clopidogrel 5 days, ticagrelor 3–5 days, prasugrel 7 days), restarted within 24–48 hours
post-procedure [30][31][41][43].
- The immediate, life-threatening danger of stent thrombosis therefore outweighs the harms of an
extra 1–3 weeks of translaryngeal intubation, which are real but generally manageable with
meticulous bundle-based care.
2. Prolonged intubation carries meaningful but slower-accumulating risks
- Complications include ventilator-associated pneumonia, laryngeal injury, subglottic stenosis,
sinusitis, increased sedation requirements, and ICU-acquired weakness.
(general standard-of-care; confirm against current guidelines)
Extubation failure is common in comatose patients with impaired airway reflexes, even when they
pass a spontaneous breathing trial. (general standard-of-care)
- Trials and meta-analyses of early tracheostomy (≤7–10 days) show it may reduce ventilator days
and ICU stay but does not improve mortality
[17][24]. The TRACMAN RCT
found no difference in 30-day or longer-term mortality between early and deferred tracheostomy
[24][25]; SETPOINT2 in severe
stroke showed no improvement in survival without severe disability with early timing
[18][19][21].
Here, the "early" tracheostomy window coincides with the period of maximal stent-thrombosis danger,
making delay the safer default.
3. Tracheostomy bleeding risk is manageable on aspirin alone but rises sharply on full DAPT
- Tracheostomy is a high-bleeding-risk procedure. (general standard-of-care)
Observational data indicate acceptable bleeding rates on aspirin monotherapy, but continuing a
P2Y₁₂ inhibitor or therapeutic anticoagulation raises major bleeding rates and the risk of
airway-compromising haematoma [28].
- Many centres therefore continue aspirin, hold the P2Y₁₂ inhibitor for the minimum required
time, and use experienced operators when tracheostomy must be performed early
[29][40][43].
Guidelines strongly caution that stopping all antiplatelet therapy soon after PCI carries a high
risk of catastrophic stent thrombosis
[8][29][38].
Clinical factors that should guide the decision
- Time since stent placement — the first 30 days are the highest-risk period for stent
thrombosis [8][29][38];
after 30 days (BMS) or 3–6 months (DES) the calculus shifts and tracheostomy can be considered
with a brief, controlled P2Y₁₂ interruption.
- Stent and antithrombotic characteristics — DES vs. BMS, lesion complexity, the specific
P2Y₁₂ agent, and any concomitant anticoagulation all modulate risk; validated risk scores help
quantify the trade-off. (general standard-of-care)
- Neurological prognosis — post-arrest neuroprognostication is multimodal and typically
deferred until at least 72 hours after return of spontaneous circulation.
(general standard-of-care)
- Airway and pulmonary status — high secretion burden, recurrent aspiration, or
intolerance of reduced support may force an earlier decision. (general standard-of-care)
- Multidisciplinary coordination — interventional cardiology, critical care, neurology,
and the procedural team must align on antithrombotic management, timing, and goals of care.
Practical synthesis
Within the first 30 days after stenting for acute MI: continue DAPT, maintain intubation, optimise
pulmonary hygiene, minimise sedation, and pursue structured neuroprognostication. If tracheostomy
becomes unavoidable earlier, perform it with aspirin continued, the P2Y₁₂ inhibitor held for the
minimum washout, and restarted as soon as haemostasis is secure — typically within 24–48 hours
[29][38][40][43].
After the highest-risk period, tracheostomy is favoured to support long-term airway care and
rehabilitation.
This framework reflects general guideline-based principles and must be individualised to the
patient's coronary anatomy, antithrombotic regimen, neurological trajectory, and institutional
capabilities. All decisions should be made in close collaboration with the treating cardiologist.
This is general medical education, not patient-specific advice.
— What the fact-check corrected (Phases ④–⑤) —
The panel removed the numbers it could not verify.
A lone chatbot states figures confidently. Here, every number that did not survive the live-web
fact-check was removed from the verdict and replaced with sourced language — this is the actual
change log:
- Removed the specific mortality range "20–45%" for stent thrombosis; replaced with a qualitative
statement consistent with the ranges in the sources.
- Removed the "20- to 30-fold" risk-increase figure for premature P2Y₁₂ discontinuation; replaced
with the guideline language ("strongly elevated, described as unacceptable").
- Removed the specific ">50%" extubation-failure rate; replaced with "common in this population".
- Removed the "10–20%" major-bleeding rate for tracheostomy on DAPT; replaced with "elevated",
per the observational data.
- Generalised the "14-day" and "3–6 month" sub-thresholds as interpretive guidance — the sources
support the first-30-days principle but not those exact cut-offs.
- Replaced the named risk scores with "validated risk scores" — the tools weren't directly
retrieved in the verification set.
- Marked the 72-hour neuroprognostication threshold as standard-of-care rather than a sourced claim.
— Sources —