1. Publication
- Title: Deferring Arterial Catheterization in Critically Ill Patients with Shock (EVERDAC)
- Acronym: EVERDAC
- Year & Journal: New England Journal of Medicine, epublished October 29, 2025 (print: 2025;393(19):1875-1888)
- Citation: Muller G, Contou D, Ehrmann S, et al; CRICS-TRIGGERSEP F-CRIN Network and the EVERDAC Trial Group. Deferring Arterial Catheterization in Critically Ill Patients with Shock. N Engl J Med. 2025;393(19):1875-1888. doi:10.1056/NEJMoa2502136
2. Context & Rationale
Background: Arterial catheterization is recommended for continuous blood-pressure monitoring in shock, but it carries risks (infection, thrombosis, limb ischemia, discomfort) and whether noninvasive monitoring is an effective, safe alternative had not been rigorously tested.
Research Question/Hypothesis: In ICU patients with shock, is a noninvasive blood-pressure monitoring strategy (deferred arterial catheterization) noninferior to early arterial catheter insertion with regard to 28-day all-cause mortality?
Why This Matters: Directly challenges one of intensive care's most entrenched, near-universal practices — routine early arterial catheterization in shock — with implications for patient comfort, resource use, and complication risk if noninferiority is demonstrated.
3. Design & Methods
- Study Type: Multicenter, open-label, investigator-initiated, pragmatic, noninferiority randomized controlled trial
- Setting & Centers: 9 ICUs, France
- Population:
- Inclusion: Patients with shock (hypotension or vasopressor initiation, plus ≥1 sign of tissue hypoperfusion) admitted to ICU within past 24 hours
- Exclusions: No blood-pressure value displayed on noninvasive device or cuff placement impossible
- Intervention: Noninvasive strategy — automated brachial cuff monitoring; later arterial catheter insertion allowed if prespecified safety criteria met
- Comparator: Invasive strategy — early arterial catheter insertion (<4 hours after randomization)
- Randomization: 1:1, 1010 ICU patients
- Blinding: Open-label
- Statistical Power & Follow-Up: Primary outcome: death from any cause at day 28 (noninferiority margin, 5 percentage points). Adverse events of special interest related to the blood-pressure-monitoring device and patient-reported pain also assessed.
4. Key Results
1010 patients randomized.
Outcome | Noninvasive Strategy | Invasive Strategy | Effect Size | 95%/96% CI | p-value | Notes |
28-day mortality (primary) | 173 patients | 185 patients | Adjusted RD −3.2 pp | 96% CI −8.9 to 2.5 | 0.006 for noninferiority | Noninferiority demonstrated |
Device-related adverse events | Recorded | Recorded | Not reported | Not reported | — | Assessed as part of safety profile |
Patient-reported pain | Recorded | Recorded | Not reported | Not reported | — | Assessed as part of safety profile |
5. Internal Validity Assessment
- Randomization & Allocation: Randomized across 9 French ICUs; specific concealment mechanics not detailed in available trial text.
- Protocol Adherence & Separation: Prespecified safety criteria allowed later catheter insertion in the noninvasive arm if needed — a pragmatic, clinically sensible crossover allowance rather than a rigid protocol violation.
- Blinding & Detection Bias: Open-label (inherent to comparing an invasive vs noninvasive monitoring strategy); the primary endpoint (28-day mortality) is objective and resistant to detection bias.
- Missing Data & Sensitivity Analyses: Not detailed in available trial text.
- Overall Internal Validity Conclusion: Strong — well-powered (1010-patient), pragmatic noninferiority trial with a hard, objective primary endpoint and a formally demonstrated noninferiority result (P=0.006 for noninferiority, with the CI entirely within the prespecified 5-percentage-point margin).
6. External Validity Assessment
- Population Representativeness: French ICU shock population (mixed etiology) across 9 centers — broadly representative of general ICU shock populations.
- Practice Context: The pragmatic "deferred, not never" design (allowing later catheterization if clinically indicated) reflects realistic clinical flexibility rather than a rigid noninvasive-only mandate.
- Overall External Validity Conclusion: Good — pragmatic, flexible design supports generalizability to general ICU shock populations in similarly resourced health systems.
7. Strengths & Limitations
Strengths:
- First large randomized trial directly challenging a near-universal, entrenched ICU practice
- Formal, well-powered noninferiority design with appropriate margin
- Pragmatic "deferred, not forbidden" design allows clinically indicated later catheterization
- Comprehensive safety and comfort outcome assessment (device-related AEs, patient-reported pain)
Limitations:
- Open-label design (though objective primary endpoint limits impact)
- Single-country (France) trial
- Noninferiority design cannot establish superiority, only that the noninvasive strategy is not meaningfully worse
8. Interpretation & Practice Impact
- Clinical Implications: Supports deferring routine early arterial catheterization in ICU shock patients in favor of noninvasive blood-pressure monitoring, with catheterization reserved for patients meeting specific escalation criteria — a genuinely practice-changing result given how entrenched routine arterial catheterization has been.
- Mechanistic Coherence: Noninvasive brachial cuff monitoring, while less continuous than arterial waveform monitoring, appears sufficient for safe shock management in this pragmatic trial design with safety-net escalation criteria.
- Systems-Level Takeaway: Could reduce catheter-related complications, cost, and patient discomfort at scale if adopted, while preserving safety through allowed escalation.
9. Controversies & Subsequent Evidence
- Editorial Commentary: An accompanying NEJM editorial (Gershengorn, "A Less Invasive Approach to Intensive Care") frames this trial as part of a broader ICU trend toward reducing unnecessary invasive monitoring where noninferiority can be demonstrated.
- Guideline Integration: Directly relevant to hemodynamic monitoring guidance in shock; subsequent work (e.g., "Developing appropriateness criteria for arterial catheters in adult ICU patients," Australian Critical Care 2026) is already engaging with how to operationalize more selective arterial catheter use following this trial.
10. Summary & Executive Takeaway
Summary: EVERDAC randomized 1010 French ICU shock patients to a noninvasive (deferred arterial catheterization) or invasive (early arterial catheterization) monitoring strategy. 28-day mortality was noninferior with the noninvasive strategy (173 vs 185 deaths; adjusted RD −3.2 percentage points, 96% CI −8.9 to 2.5, P=0.006 for noninferiority).
Overall Takeaway: Deferring routine early arterial catheterization in favor of noninvasive blood-pressure monitoring, with safety-net escalation criteria, is noninferior to routine early catheterization for 28-day mortality in ICU shock — a genuinely practice-changing challenge to one of intensive care's most entrenched invasive monitoring habits.
11. Bibliography
- Muller G, Kamel T, Contou D, et al. Early versus differed arterial catheterisation in critically ill patients with acute circulatory failure: the EVERDAC protocol. BMJ Open. 2021;11(9):e044719.
- Gershengorn HB. A Less Invasive Approach to Intensive Care [editorial]. N Engl J Med. 2025;393(19):1953-1954.