Skip to main content

Foundational Trials

COACT

Image: Shutterstock / ChaNaWiT

Publication

  • Title: Coronary Angiography after Cardiac Arrest without ST-Segment Elevation
  • Acronym: COACT
  • Year: 2019
  • Journal published in: The New England Journal of Medicine
  • Citation: Lemkes JS, Janssens GN, van der Hoeven NW, et al. Coronary angiography after cardiac arrest without ST-segment elevation. N Engl J Med. 2019;380:1397-1407.

Context & Rationale

  • Background
    • Out-of-hospital cardiac arrest (OHCA) is frequently related to coronary disease, but post-ROSC ECG lacks ST-segment elevation in many patients.
    • Acute coronary occlusion and unstable culprit lesions can exist without ST elevation, creating clinical uncertainty about “routine cath lab” activation.
    • Observational studies suggested benefit from early coronary angiography/revascularisation, but were vulnerable to selection bias, confounding by indication, and “survivor bias” (only those surviving long enough receive angiography).
    • Immediate angiography could expedite culprit treatment but might also delay other time-critical post-arrest care (particularly targeted temperature management and ICU stabilisation) and exposes patients to procedural risks.
  • Research Question/Hypothesis
    • Among comatose survivors of OHCA with an initial shockable rhythm and no ST-segment elevation, does an immediate coronary angiography strategy improve 90-day survival compared with a delayed strategy (angiography after neurological recovery, with urgent angiography allowed for clinical deterioration)?
  • Why This Matters
    • Determines whether routine emergent cath lab activation should be standard for a common OHCA phenotype (shockable rhythm, no ST elevation).
    • Has major implications for regional systems of care, cath lab utilisation, ICU workflow, and risks/benefits of invasive management in comatose patients.
    • Provides randomised evidence to address a long-standing practice gap driven largely by observational data and pathophysiological plausibility.

Design & Methods

  • Research Question: In comatose OHCA survivors with initial VF/VT and no ST-segment elevation, does immediate coronary angiography (with PCI if indicated) improve 90-day survival versus delayed angiography after neurological recovery (with urgent angiography permitted for instability)?
  • Study Type: Investigator-initiated, multicentre, randomised, open-label, parallel-group strategy trial in 19 hospitals (the Netherlands); ICU-based post-resuscitation care with catheterisation laboratory access; deferred consent process.
  • Population:
    • Setting: post-ROSC hospital care after OHCA; patients managed in ICU with targeted temperature management as standard care.
    • Key inclusion: adult OHCA; initial shockable rhythm (ventricular fibrillation or pulseless ventricular tachycardia); sustained ROSC; comatose (GCS <8); no ST-segment elevation on post-ROSC ECG; randomised within 4 hours after ROSC.
    • Key exclusion (as operationalised in the trial): ST-segment elevation myocardial infarction; haemodynamic instability/shock; obvious non-coronary cause of arrest; severe renal dysfunction; pregnancy.
  • Intervention:
    • Immediate coronary angiography strategy: angiography initiated as soon as possible and intended within 2 hours after randomisation; PCI and adjunctive antithrombotic therapy at the discretion of the treating interventional cardiologist.
  • Comparison:
    • Delayed coronary angiography strategy: angiography performed after neurological recovery (typically after ICU discharge); urgent angiography allowed for cardiogenic shock, recurrent life-threatening arrhythmias, or recurrent ischaemia during hospitalisation.
  • Blinding: Unblinded (strategy intervention); primary endpoint (90-day survival) is objective, but functional outcomes (e.g., CPC) may be susceptible to unblinded assessment.
  • Statistics: A total of 552 patients were required to detect a 13% absolute increase in survival (from 32% to 45%) with 85% power at the 5% significance level (two-sided); primary analysis was intention-to-treat after deferred consent (patients without obtained consent were excluded from analyses); effect measures reported as odds ratios or ratios of geometric means with 95% confidence intervals.
  • Follow-Up Period: 90 days (primary endpoint); in-hospital outcomes (including ICU discharge neurological status) were also assessed.

Key Results

This trial was not stopped early. One interim analysis after enrolment of the first 400 patients was performed (data and safety monitoring board review); recruitment continued to the target sample size.

Outcome Immediate angiography Delayed angiography Effect p value / 95% CI Notes
Survival at 90 days (primary) 176/273 (64.5%) 178/265 (67.2%) OR 0.89 95% CI 0.62 to 1.27; P=0.51 No strategy benefit on primary endpoint
Survival at 90 days with good neurological outcome (CPC 1–2) 171/272 (62.9%) 170/264 (64.4%) OR 0.94 95% CI 0.66 to 1.31; P=0.71 Functional outcome broadly concordant with mortality
Major bleeding 7/272 (2.6%) 10/264 (4.0%) OR 0.63 95% CI 0.22 to 1.80; P=0.38 Low absolute event rates
Need for renal replacement therapy 8/272 (2.9%) 11/264 (4.2%) OR 0.68 95% CI 0.27 to 1.72; P=0.41 No signal of excess dialysis with early contrast exposure
Time to target temperature (hours) Median 5.4 (IQR 4.0–6.7) Median 4.7 (IQR 3.6–5.8) Ratio 1.19 95% CI 1.04 to 1.36; P=0.01 Immediate strategy modestly delayed temperature target attainment
Peak troponin level Median 1.2 (IQR 0.3–3.1) Median 1.3 (IQR 0.4–3.4) Ratio 0.76 95% CI 0.63 to 0.92; P=0.005 Lower geometric mean peak troponin with immediate strategy
Duration of mechanical ventilation (days) Median 4.0 (IQR 3.0–7.0) Median 4.0 (IQR 3.0–6.0) Ratio 0.96 95% CI 0.80 to 1.14; P=0.62 No clear difference in ventilator duration
  • No improvement in 90-day survival or 90-day survival with favourable neurological outcome with an immediate angiography strategy versus delayed angiography.
  • There was strong “strategy separation” (median time randomisation-to-angiography 0.8 hours vs 119.9 hours; angiography performed 97.1% vs 64.9%), yet outcome neutrality persisted.
  • Pre-specified subgroup analyses showed statistically significant interactions for age and known coronary artery disease (hypothesis-generating): age <70 years OR 0.57 (0.34–0.97) vs age ≥70 years OR 1.62 (0.95–2.78); interaction P=0.007; known coronary artery disease OR 1.30 (0.73–2.32) vs none OR 0.69 (0.44–1.10); interaction P=0.009.

Internal Validity

  • Randomisation and allocation: Web-based randomisation (variable block sizes), stratified by centre; allocation concealment occurred up to assignment.
  • Post-randomisation exclusions/withdrawal: 552 patients randomised; 538 included in the primary analysis (273 immediate vs 265 delayed) after deferred consent exclusions (14 patients, evenly distributed, withdrew/declined consent).
  • Performance/detection bias: Unblinded strategy; objective primary endpoint mitigates detection bias, but neurological outcomes (CPC, ICU discharge assessments) are potentially susceptible to unblinded assessment.
  • Protocol adherence and contamination: Angiography performed in 266/274 (97.1%) immediate vs 173/267 (64.9%) delayed; urgent angiography occurred in 37/265 (14.0%) of delayed; crossover was limited (immediate: 12/273 [4.4%] delayed or no angiography; delayed: 3/265 [1.1%] immediate angiography).
  • Baseline comparability: Groups were closely matched (median age 65 vs 65 years; male 79.9% vs 80.8%; bystander CPR 72.5% vs 74.3%; median time to ROSC 15 vs 16 minutes; shock on admission 2.9% vs 4.2%).
  • Timing and dose of intervention (strategy intensity): Median randomisation-to-angiography 0.8 hours (IQR 0.5–1.1) vs 119.9 hours (IQR 23.2–214.2); target temperature attainment was later with the immediate strategy (median 5.4 vs 4.7 hours; ratio 1.19).
  • Separation of the variable of interest: Marked separation in coronary procedures (PCI in 89/270 [33.0%] vs 42/173 [24.2%] among those undergoing angiography); acute thrombotic occlusion was uncommon (9/266 [3.4%] vs 13/172 [7.6%] among those undergoing angiography).
  • Outcome assessment: Survival at 90 days is objective; functional outcomes at 90 days were recorded using CPC categories (pre-specified) and ICU discharge neurological measures (GCS, CPC).
  • Statistical rigour: Sample size achieved as planned; effect estimates reported with 95% CIs; multiple secondary outcomes increase risk of multiplicity-driven findings (subgroup interactions should be treated as exploratory).

Conclusion on Internal Validity: Overall, internal validity appears moderate-to-strong given central randomisation, balanced baseline characteristics, and strong strategy separation; the main limitations are open-label delivery and deferred-consent exclusions, with subgroup signals best viewed as hypothesis-generating.

External Validity

  • Population representativeness: Applies primarily to OHCA survivors with an initial shockable rhythm who are comatose and haemodynamically stable without ST elevation; it does not address non-shockable rhythms, STEMI, or shock—groups that comprise a large fraction of real-world OHCA.
  • System context: Conducted in a high-resource national system with ready catheterisation laboratory access and standardised ICU post-arrest care (including targeted temperature management at 33°C), which may not translate directly to all health systems.
  • Clinical applicability: Most applicable to “stable, no-STEMI” pathways where clinicians face the decision of routine immediate cath lab activation versus delayed/selective angiography after initial ICU management.

Conclusion on External Validity: Generalisability is moderate for stable, shockable-rhythm OHCA survivors without ST elevation in systems with catheterisation access and robust post-arrest ICU care, but limited for unstable patients, non-shockable arrests, and settings where immediate cath lab transfer competes with essential ICU therapies.

Strengths & Limitations

  • Strengths:
    • Large, pragmatic, investigator-initiated multicentre randomised trial addressing a high-impact systems-of-care question.
    • Objective primary outcome (90-day survival) with near-complete ascertainment.
    • High adherence in the immediate group and clear separation in timing to angiography (median 0.8 vs 119.9 hours).
    • Contemporary co-interventions standardised across arms (high uptake of targeted temperature management).
  • Limitations:
    • Open-label strategy with potential for unblinded co-interventions and functional outcome assessment bias.
    • Deferred consent led to post-randomisation exclusions (14 patients), which can introduce bias if not balanced (here, exclusions were evenly distributed).
    • Delayed strategy included urgent angiography in a substantive minority (14%), and many delayed-group patients never underwent angiography (64.9% received angiography), reflecting strategy reality but complicating “pure” mechanistic inference.
    • Excluded key higher-risk phenotypes (STEMI, shock), limiting inference to haemodynamically stable arrests and potentially under-representing patients most likely to have an acute culprit.
    • Acute thrombotic occlusion rates were low, potentially limiting the theoretical benefit of routine immediate revascularisation in the enrolled population.

Interpretation & Why It Matters

  • Clinical implication
    • Routine immediate coronary angiography is not supported for haemodynamically stable, comatose OHCA survivors with an initial shockable rhythm and no ST-segment elevation.
    • Selective/triggered urgent angiography remains essential for patients with STEMI, cardiogenic shock, recurrent malignant arrhythmias, or recurrent ischaemia.
  • Mechanistic insight
    • Low frequency of acute thrombotic occlusion in this phenotype suggests that a “culprit amenable to immediate PCI” is uncommon once STEMI and shock are excluded.
    • Immediate cath lab transfer modestly delayed target temperature achievement, emphasising that pathway choices can affect other pillars of post-arrest care.
  • Research implication
    • Future work should prioritise improved risk stratification (clinical, ECG, biomarkers, imaging, and potentially intracoronary imaging) to identify subgroups more likely to benefit.
    • Subgroup interactions by age and known coronary disease should be tested prospectively rather than acted on clinically in isolation.

Controversies & Subsequent Evidence

  • The accompanying editorial emphasised that the key question may be “the right patients” rather than “the right timing”, noting the low prevalence of acute coronary occlusion in the enrolled cohort and the importance of avoiding pathway decisions that compromise ICU stabilisation and targeted temperature management.1
  • Published correspondence highlighted that angiography may underestimate plaque disruption without intracoronary imaging (e.g., OCT/IVUS), raising the possibility that some acute coronary pathology was not detectable with angiography alone; this critique challenges mechanistic inference but does not negate the strategy-level neutrality on survival.2
  • Correspondence also argued that early “rescue” angiography in the delayed group (urgent angiography for deterioration) could dilute contrasts between strategies, while also reflecting real-world clinical safety-netting expected in any delayed approach.2
  • Subsequent randomised trials and meta-analyses in similar populations have largely aligned with COACT’s central message: a routine immediate angiography strategy does not improve survival in haemodynamically stable OHCA survivors without ST elevation; contemporary guideline updates generally support selective (trigger-based) angiography rather than default immediate cath lab activation.

Summary

  • In 552 comatose OHCA survivors with an initial shockable rhythm and no ST elevation, an immediate angiography strategy did not improve 90-day survival versus delayed angiography (64.5% vs 67.2%; OR 0.89; 95% CI 0.62 to 1.27).
  • Neurological outcomes at 90 days were similarly neutral (CPC 1–2: 62.9% vs 64.4%; OR 0.94; 95% CI 0.66 to 1.31).
  • Strategy separation was large (angiography performed 97.1% vs 64.9%; median randomisation-to-angiography 0.8 vs 119.9 hours), yet clinical outcomes remained unchanged.
  • Acute thrombotic coronary occlusion was uncommon, consistent with limited potential for routine immediate PCI benefit once STEMI and shock are excluded.
  • Immediate angiography modestly delayed time to target temperature (median 5.4 vs 4.7 hours; ratio 1.19), underscoring pathway trade-offs in post-arrest care.

Further Reading

Other Trials

Systematic Review & Meta Analysis

Observational Studies

Guidelines

Notes

  • COACT’s statistically significant subgroup interactions (age; known coronary artery disease) were not definitive and should be treated as hypothesis-generating.

Overall Takeaway

COACT is a landmark strategy trial showing that, in haemodynamically stable comatose OHCA survivors with a shockable rhythm and no ST-segment elevation, routine immediate coronary angiography does not improve 90-day survival or neurological outcomes compared with a delayed/selective approach. The trial reframed post-arrest pathways towards selective cath lab activation (based on STEMI or clinical deterioration), while highlighting the need for better phenotyping to identify the minority with actionable acute coronary pathology.

Overall Summary

  • In stable, shockable-rhythm OHCA survivors without ST elevation, COACT found no survival or neurological benefit from routine immediate coronary angiography versus delayed angiography, despite strong separation in angiography timing and exposure.

Bibliography