The genuinely unsettled questions in clinical cardiology, ranked by debate-worthiness. Scanned continuously from the major journals, FDA actions, new guidelines, and the conversation among cardiologists on X.
2 live debatesLast scanned Aug 9, 2026, 8:38 AM UTC
Heat — how live this debate is
A 0–10 estimate of how live and consequential this disagreement is in cardiology right now. It drives the ranking.
This debate’s factors
How split the field is6.3 · 40%
How much practice changes8.0 · 30%
How much is genuinely new7.7 · 20%
Clinician attention7.6 · 10%
Weighted (40 / 30 / 20 / 10) into the 7.5 heat score.
1.A continuous scan (Grok, with live web access) reads the cardiology literature, FDA actions, new guidelines, and the clinician conversation on X to find genuinely contested questions and gather the evidence behind them.
2.Two AI reviewers from different model families — Claude and GPT-5 — then independently rate four factors, led by how genuinely the field disagrees, plus how much practice changes, how much genuinely new evidence there is, and how much real clinician (not lay) attention it draws. Neither sees the other’s ratings.
3.Key factors are grounded in real signals we collected — named experts quoted on opposing sides, and actual new trials, guidelines, or FDA actions — so heat isn’t just the models’ opinion.
4.Fixed editorial weights combine the factors — models never set the weights.
5.Scores are smoothed across scans, so one viral moment can’t whipsaw the board.
On this topic, the two reviewers’ ratings agreed 97%.
Heat is an AI estimate, not a vote — no cardiologist has voted on it yet. As cardiologists vote on Synapse, measured expert disagreement replaces this AI estimate as the headline number.
Movement — vs the previous scan
The board re-ranks after every scan. This chip is measured bookkeeping, not a model’s opinion:
▲ / ▼ — places climbed or dropped since the last scan.
NEW — first appearance on the board.
— — held its position.
Positions only swap when a debate’s heat moves decisively; small wobbles never reshuffle the board.
Evidence gap — what this means
The answer is genuinely unknown and a trial or readout is still pending. New data we don't have yet will settle it.
The type sets how the debate is framed below — an evidence gap leads with the pending trial, a coordination gap with the two camps, a values & turf split with the trade-off.
Patient reach — how many this affects
A coarse sense of how big the affected patient population is — from a small subspecialty group (“Few patients”) up to millions in routine practice (“Millions of patients”).
It’s a magnitude band the model sets from the clinical context, not a counted total — we never show a fabricated patient number. This is a display signal and doesn’t change the heat score.
Expert split — who’s on record
The tally of named experts we found publicly taking each side (4 vs 0 here). The bar shows how that lean divides between the two camps.
This is who we found on the record — a snapshot, not a full poll of the field. The measured split will come from cardiologists voting on Synapse.
Sources — the evidence behind it
How many distinct sources we linked for this debate (6 sources here) — journals, guidelines, FDA actions, preprints, news, and posts on X. More bars mean deeper coverage.
Open “The evidence” below to see each one. Counts and quotes are scan-reported and link to their primary source — not yet independently verified.
Two clocks: active since vs. on the board
These measure different things, so the card keeps them apart:
Active sincewhen the real-world debate became active, from the earliest dated source we found (Nov 2025).
On the boardhow long the question has been live on Synapse (3 weeks).
A debate can be contested for years but new to the board — or brand-new science that’s only days old.
01★ Top debateField-wide debate
Drop to blood thinner alone one month after a stent?
One month or 12 months dual antithrombotic therapy after PCI in AF patients?
How quickly you drop the P2Y12 inhibitor decides whether AF stent patients bleed less or clot more — and the answer sets everyday practice for a huge, growing group.
Heat — how live this debate is
A 0–10 estimate of how live and consequential this disagreement is in cardiology right now. It drives the ranking.
This debate’s factors
How split the field is9.3 · 40%
Debates are surfaced by a continuous scan of the cardiology literature and the conversation among clinicians. The heat score (0-10) reflects how live and consequential the disagreement is right now and drives the ranking; it becomes the measured expert split as cardiologists weigh in. Quotes and engagement counts are scan-reported and link to their primary source.
In AF patients undergoing PCI, should dual antithrombotic therapy be stopped at 1 month and converted to blood thinner alone across broader real-world populations, or extended toward 12 months in higher-risk subgroups?
De-escalate at one month
One month of dual therapy prevented ischemic events just as well as a year while causing far less bleeding, so keeping patients on two agents longer just wounds them without protecting them.
“Patients with atrial fibrillation who receive a stent may only need 1 month of dual antithrombotic therapy, instead of 1 year.”
“The marked reduction in the primary bleeding endpoint supports earlier de-escalation in carefully selected patients, particularly those with chronic coronary syndrome, treated with contemporary drug-eluting stents, undergoing intravascular imaging-guided PCI.”
Raffaele Piccolo · Interventional cardiologist · PCRonline ↗
“OPTIMA-AF trial: In AF patients undergoing PCI with EES, 1-month therapy with DOAC + P2Y12 inhibitor was non-inferior for preventing ischemia and superior for reducing bleeding.”
“OPTIMA-AF provides the first randomised evidence directly addressing conversion to DOAC monotherapy after only 1 month of dual antithrombotic therapy in patients with atrial fibrillation undergoing PCI.”
Ester Sofia Congedo · Interventional cardiologist · PCRonline ↗
Hold longer in higher-risk patients
The trial's reassurance came in selected patients on modern stents, and the sickest AF-plus-PCI patients — recent ACS, complex anatomy — may still need the longer antiplatelet cover the wider data can't yet rule out.
Named experts we found publicly on record — a sample, not a representative poll of the field.
Some details are pending source verification and are withheld until confirmed.
Who: AF patients undergoing PCI for stable or unstable CAD·6 sources
On the board:since Jul 17, 2026 (3 weeks)— time live on Synapse
Patients affected:Many patients· common overlap of AF and stenting
›The evidence
What we know
✓AF patients after a stent need both a blood thinner and antiplatelet therapy, but combining them raises bleeding risk.
✓OPTIMA-AF found 1-month dual therapy non-inferior for ischemic events versus 12 months.
✓Shorter dual therapy clearly reduced bleeding in the trial.
✓Modern drug-eluting stents lower the baseline risk of stent-related clots.
What's still unknown
?Whether 1 month holds up in higher-risk, real-world patients outside the trial's selected group.
?How patients presenting with acute coronary syndrome versus stable disease fare with early de-escalation.
?
Quotes and engagement counts are scan-reported and link to their primary source — not yet independently verified.
How much practice changes8.0 · 30%
How much is genuinely new7.1 · 20%
Clinician attention8.7 · 10%
Weighted (40 / 30 / 20 / 10) into the 7.3 heat score.
1.A continuous scan (Grok, with live web access) reads the cardiology literature, FDA actions, new guidelines, and the clinician conversation on X to find genuinely contested questions and gather the evidence behind them.
2.Two AI reviewers from different model families — Claude and GPT-5 — then independently rate four factors, led by how genuinely the field disagrees, plus how much practice changes, how much genuinely new evidence there is, and how much real clinician (not lay) attention it draws. Neither sees the other’s ratings.
3.Key factors are grounded in real signals we collected — named experts quoted on opposing sides, and actual new trials, guidelines, or FDA actions — so heat isn’t just the models’ opinion.
4.Fixed editorial weights combine the factors — models never set the weights.
5.Scores are smoothed across scans, so one viral moment can’t whipsaw the board.
On this topic, the two reviewers’ ratings agreed 92%.
Heat is an AI estimate, not a vote — no cardiologist has voted on it yet. As cardiologists vote on Synapse, measured expert disagreement replaces this AI estimate as the headline number.
Movement — vs the previous scan
The board re-ranks after every scan. This chip is measured bookkeeping, not a model’s opinion:
▲ / ▼ — places climbed or dropped since the last scan.
NEW — first appearance on the board.
— — held its position.
Positions only swap when a debate’s heat moves decisively; small wobbles never reshuffle the board.
Evidence gap — what this means
The answer is genuinely unknown and a trial or readout is still pending. New data we don't have yet will settle it.
The type sets how the debate is framed below — an evidence gap leads with the pending trial, a coordination gap with the two camps, a values & turf split with the trade-off.
Patient reach — how many this affects
A coarse sense of how big the affected patient population is — from a small subspecialty group (“Few patients”) up to millions in routine practice (“Millions of patients”).
It’s a magnitude band the model sets from the clinical context, not a counted total — we never show a fabricated patient number. This is a display signal and doesn’t change the heat score.
Expert split — who’s on record
The tally of named experts we found publicly taking each side (2 vs 2 here). The bar shows how that lean divides between the two camps.
This is who we found on the record — a snapshot, not a full poll of the field. The measured split will come from cardiologists voting on Synapse.
Sources — the evidence behind it
How many distinct sources we linked for this debate (6 sources here) — journals, guidelines, FDA actions, preprints, news, and posts on X. More bars mean deeper coverage.
Open “The evidence” below to see each one. Counts and quotes are scan-reported and link to their primary source — not yet independently verified.
Two clocks: active since vs. on the board
These measure different things, so the card keeps them apart:
Active sincewhen the real-world debate became active, from the earliest dated source we found (Sep 2023).
On the boardhow long the question has been live on Synapse (6 weeks).
A debate can be contested for years but new to the board — or brand-new science that’s only days old.
02Field-wide debate
Pumps early in shock, or stabilize first?
Cardiogenic shock: early mechanical support or medical stabilization first?
Guess wrong and you either deny a dying patient a pump that could save them, or you subject the rest to a costly, bleeding-prone device that adds nothing but harm.
In infarct-related cardiogenic shock, should a microaxial flow pump be placed early and routinely, or reserved for selected patients after attempted medical stabilization and revascularization?
Pump early
A microaxial flow pump unloads the failing ventricle before organs die, and there is now trial evidence of a real survival benefit in selected infarct-shock patients.
“Overall the results point to a “lasting effect” of the Impella CP pump that extends beyond the index population.”
“The DanGer Shock trial compared survival using left ventricular (LV) mechanical circulatory support with the Impella CP microaxial flow pump in infarct-related cardiogenic shock.”
Devices bring bleeding, limb loss, and access complications, routine early use hasn't cut mortality, and most patients do as well with drugs and prompt revascularization.
“In patients with acute myocardial infarction complicated by cardiogenic shock with planned early revascularization, the risk of death from any cause at the 30-day follow-up was not lower among the patients who received ECLS therapy than among those who received medical therapy al”
“Although tMCS is being increasingly utilized in the treatment of both AMI-CS and HF-CS, routine use of tMCS in all CS patients is strongly discouraged.”