Cardiac CT: the trials at a glance
Trial · the question it asked · what changed afterwards. Red = nothing changed.
| Trial | Question asked | What changed |
|---|---|---|
| Acute chest pain | ||
| CT-STAT2011 | Is CCTA faster and cheaper than SPECT? | Diagnosis in 2.9 vs 6.3 h. Part of the Level A evidence pool. |
| ACRIN-PA2012 | Is a negative CCTA safe for ED discharge? | 0 events in 640 negative scans → Class 1, LOE A. |
| ROMICAT-II2012 | Does CCTA shorten hospital stay? | −7.6 h, but more testing, more radiation, no cost saving. |
| CT-COMPARE2014 | CCTA vs exercise ECG? | Sensitivity 100% vs 83%. Helped retire exercise ECG. |
| CATCH2015 | Does CCTA-guided care improve outcomes? | HR 0.62 — one of few acute trials with a clinical endpoint. |
| BEACON2016 | Any value added to hs-troponin? | Nothing. Stay identical at 6.3 h. No Class I in Europe. |
| RAPID-CTCA2021 | Death or MI at 1 year in higher-risk ACS? | HR 0.91, not significant → ESC 2023 downgraded to Class IIa. |
| TARGET-CTCA2026 | Any benefit once MI is already ruled out? | HR 0.95, not significant. Nothing yet — challenges Class 1. |
| Pooled 22 RCTs2023 | What does CCTA actually deliver in the ED? | Only revascularization rose (RR 1.37). Not fewer events. |
| Stable chest pain | ||
| SCOT-HEART2015 / 2018 / 2025 | Does CCTA change the diagnosis — then the events? | 10-y HR 0.79 with revascularization identical → Class 1 A both sides of the Atlantic. |
| PROMISE2015 / 2025 | Anatomic or functional test first? | Equivalent, not superior. 10.6-y mortality HR 0.98. |
| DISCHARGE2022 | CT or straight to invasive angiography? | Complications 0.5% vs 1.9% → ESC 2024 Class I A gatekeeper. |
| CRESCENT2016 | Does a calcium-gated CT protocol work? | Downstream testing 25% vs 53%. Embedded in ESC 2024 algorithm. |
| CAPP2015 | CT vs exercise ECG for symptoms? | Inconclusive results 2.4% vs 26.9%. Used by NICE 2016. |
| CONSERVE2019 | Selective referral after CCTA — safe? | Angiography 23% vs 89%, events identical. |
| PRECISE / DEFER-CTA2023 | Can we defer testing in minimal-risk patients? | HR 0.35; deferral safe → ESC 2024 defer at ≤5%. |
| ISCHEMIA2020 | Can CCTA gatekeep before randomization? | 97% concordance excluding left main. Now standard trial design. |
| Coronary calcium | ||
| Western Denmark2022 | Does CAC = 0 rule out disease at every age? | 58% of under-40s with obstructive CAD had CAC = 0. No guideline caveat yet. |
| CONFIRM CAC = 02011 | Is there plaque despite a zero score? | 16% had plaque, 3.5% had ≥50% stenosis. Nothing. |
| CT physiology — FFR-CT and perfusion | ||
| DISCOVER-FLOW2011 | Can fluid dynamics reproduce invasive FFR? | Per-vessel accuracy 84% vs 58% for stenosis alone. |
| DeFACTO2012 | Is per-patient accuracy good enough? | Missed its endpoint (73%). Forced an algorithm rewrite. |
| NXT2014 | Is the revised algorithm good enough? | AUC 0.90 vs 0.81 → FDA De Novo grant, Nov 2014. |
| PLATFORM2015 | Can it prevent the empty catheterization? | Normal angiography 12% vs 73% → NICE MTG32. |
| ADVANCE2018 | Is deferral on FFR-CT >0.80 safe? | Zero MACE in 1,592 deferred → supports COR 2a. |
| FORECAST2021 | Does FFR-CT save money? | No (p=0.10), though angiography fell. |
| TARGET2023 | Does on-site machine-learning CT-FFR work? | Composite 28% vs 46%. Adopted in Asia. |
| FUSION2026 | Does it hold up without industry funding? | Unnecessary angiography 18% vs 33%. Likely lever for upgrade. |
| CREDENCE2020 | Plaque or perfusion — which predicts FFR? | AUC 0.81 vs 0.67. Physiology depends on plaque, not just stenosis. |
| Dan-NICAD 1 & 22018 / 2023 | Does stress testing after an abnormal CCTA work? | Sensitivity only 36–64%. A negative stress test does not exclude ischemia. |
| CORE3202014 | Does CT perfusion add to CCTA? | AUC 0.87 vs 0.84. Nothing — CTP never entered guidelines. |
| Plaque, inflammation and AI | ||
| CONFIRM2011 | Does non-obstructive disease carry risk? | Predicts mortality. Broke the stenosis-only model. |
| ICONIC2018 | What did the culprit look like beforehand? | 75% were <50% stenosis at baseline. Nothing. |
| PROMISE high-risk plaque2018 | Does high-risk plaque predict events? | aHR 1.72 overall, 4.31 if non-obstructive → CAD-RADS 2.0 modifier only. |
| SCOT-HEART LAP2020 | Does low-attenuation plaque predict MI? | >4%: HR 4.65 — beats calcium score and stenosis. No treatment trigger. |
| PARADIGM2018 | Do statins change plaque? | Calcified up, non-calcified down. No serial-CT recommendation. |
| EVAPORATE2020 | Does icosapent ethyl change plaque? | LAP −17% vs +109%. Labeling rests on REDUCE-IT, not this. |
| CRISP-CT / ORFAN2018 / 2024 | Does perivascular inflammation predict death? | HR up to 29.8 → FDA De Novo, July 2026. No CPT code, no guideline. |
| CLARIFY / CERTAIN / CONFIRM22021–2026 | Does AI match — or beat — the reader? | AUC 0.81 vs 0.79 → CPT 75577, Jan 2026. ACC still declines routine use. |
| TRANSFORMongoing | Does plaque-directed therapy improve outcomes? | Has not reported. The trial that settles the field. |
| Structural heart CT | ||
| Jilaihawi / Binder2012 / 2013 | CT or TEE for TAVR annular sizing? | Paravalvular leak 21.9% → 7.5%. Practice changed worldwide — still no Class or Level. |
| VIVID registry2018 | Can CT predict coronary obstruction? | VTC ≤4 mm, AUC 0.943. Consensus-level only. |
| BASILICA2019 / 2021 | Can leaflet laceration prevent obstruction? | 100% freedom at 30 days. CT selects who gets it. |
| Redo-TAVR CT2020 | Will the coronaries stay reachable? | Access impossible in 27% vs 10% by valve type. Drives lifetime planning. |
| Yoon neo-LVOT2019 | Can CT predict LVOT obstruction in TMVR? | <170 mm²: sensitivity 96%. Expert recommendation. |
| LAMPOON2019 | Can laceration prevent LVOT obstruction? | 100% technical success, 93% 30-day survival. |
| SCAI / HRS LAAO2025 | CT or TEE before appendage occlusion? | Either — conditional, very low certainty. CT not preferred. |
| Prosthetic valves | ||
| Makkar2015 | Does 4D CT reveal leaflet abnormality? | Reduced motion in 13–40%. Created a new clinical entity. |
| RESOLVE / SAVORY2017 | How common is leaflet thrombosis? | 12% overall; TIA link but not stroke. |
| GALILEO-4D2020 | Does rivaroxaban fix it? | Yes on CT — but the parent trial stopped for harm. |
| ADAPT-TAVR2022 | Does leaflet thrombosis cause emboli? | No association found. |
| NOTION-4 / ACASA-TAVI2026 | Is the benefit durable? | Gone once the drug stops; safety favoured aspirin. |
| Gündüz2015 | Pannus or thrombus? | ≥145 HU: sensitivity 87.5%, specificity 95.5%. Predicts thrombolysis response. |
| Heart Team, cath lab, operating room | ||
| SYNTAX III2018 | Same revascularization decision from CT? | ~93% agreement, κ 0.82. |
| FASTTRACK CABG2024 | Can surgeons operate from CT alone? | 99.1% feasible without angiography. No guideline change. |
| P32022 | Can CT predict the result of PCI? | Post-PCI FFR predicted to 0.02 ± 0.07. |
| P4ongoing | Can CT-guided PCI compete with angiography? | Has not reported. Stent sizing still unvalidated. |
| CorMicA2018 | What does a normal CCTA miss? | 89% had abnormal coronary function → ESC 2024 Class I B for function testing. |
Full detail, with every reference linked, is in the companion manuscript.
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.