Skip to main content

Foundational Trials

CURASMUR

Image: Shutterstock / ChaNaWiT

Publication

  • Title: Effect of Rocuronium vs Succinylcholine on Endotracheal Intubation Success Rate Among Patients Undergoing Out-of-Hospital Rapid Sequence Intubation: A Randomized Clinical Trial
  • Acronym: CURASMUR
  • Year: 2019
  • Journal published in: JAMA
  • Citation: Guihard B, Chollet-Xémard C, Lakhnati P, et al. Effect of rocuronium vs succinylcholine on endotracheal intubation success rate among patients undergoing out-of-hospital rapid sequence intubation: a randomized clinical trial. JAMA. 2019;322(23):2303-2312.

Context & Rationale

  • Background
    • Rapid sequence intubation (RSI) aims to optimise first-pass tracheal intubation and minimise aspiration and physiological deterioration during emergency airway management.
    • Succinylcholine offers rapid onset and short duration, but has clinically important contraindications and adverse effects (notably hyperkalaemia and malignant hyperthermia susceptibility).
    • High-dose rocuronium (≥1 mg/kg) can achieve a rapid onset comparable to succinylcholine in controlled settings, with a different contraindication profile, but produces longer paralysis (a potential safety concern if airway control fails).
    • Before CURASMUR, comparative evidence for neuromuscular blocking agent choice in out-of-hospital RSI was limited, and largely extrapolated from peri-operative studies and heterogeneous emergency department/ICU data.
  • Research Question/Hypothesis
    • In adult out-of-hospital patients undergoing physician-led RSI, is rocuronium (1.2 mg/kg) noninferior to succinylcholine (1 mg/kg) for successful first-attempt endotracheal intubation (noninferiority margin -7% absolute difference)?
    • Secondary objectives included comparison of intubation conditions, need for alternative airway techniques, and early intubation-related complications.
  • Why This Matters
    • First-pass success is strongly associated with fewer airway complications; the prehospital environment adds complexity (limited positioning, variable lighting/space, unstable physiology).
    • Neuromuscular blocker choice is a system-level decision that affects protocols, training, stock (including reversal agents), and rescue planning for “can’t intubate/can’t ventilate” scenarios.
    • Many services increasingly favour rocuronium; robust, pragmatic comparative data are required to inform practice when both agents are considered clinically permissible.

Design & Methods

  • Research Question: Among adults undergoing out-of-hospital RSI, is rocuronium noninferior to succinylcholine for successful first-attempt endotracheal intubation (noninferiority margin -7%)?
  • Study Type: Multicentre, randomised, parallel-group, single-blind, noninferiority trial in physician-staffed out-of-hospital emergency medical units (France); January 9, 2014 to August 31, 2016.
  • Population:
    • Setting: Out-of-hospital emergency tracheal intubation with RSI performed by physician-led mobile emergency units (17 sites).
    • Inclusion: Adults (≥18 years) requiring out-of-hospital emergency RSI and tracheal intubation as assessed by the treating emergency physician.
    • Key exclusions: Cardiac arrest; pregnancy; no health insurance; adults under guardianship.
    • Drug-related exclusions: Any contraindication to rocuronium, succinylcholine, or sugammadex (eg, malignant hyperthermia history, myopathy/neuromuscular disease, pseudocholinesterase deficiency, allergy; hyperkalaemia and other standard succinylcholine contraindications).
    • Consent model: Emergency deferred consent with next-of-kin/patient consent pathways, including post-event opt-out when no relative was present.
  • Intervention:
    • Neuromuscular blocker: Rocuronium 1.2 mg/kg IV (after hypnotic induction).
    • RSI approach: Direct laryngoscopy with metallic Macintosh blades (sizes 3 and 4); standard procedure did not include a stylet; Sellick manoeuvre recommended.
    • Timing: Intubation recommended 60 seconds after neuromuscular blocker injection.
    • Induction/maintenance: Hypnotic choice and dosing per clinician; maintenance sedation recommended as benzodiazepine or propofol plus opioid; hypotension treated with crystalloids/ephedrine and, if prolonged, continuous catecholamine infusion.
    • Rescue planning: Alternative airway procedures per French difficult airway guidance (including bougie, intubating laryngeal mask airway [ILMA], cricothyrotomy); sugammadex permitted as part of rescue strategy for rocuronium at clinician discretion.
  • Comparison:
    • Neuromuscular blocker: Succinylcholine 1 mg/kg IV (after hypnotic induction).
    • Other care: Same laryngoscopy approach, timing recommendation, sedation/analgesia approach, and rescue algorithms (without rocuronium reversal).
  • Blinding: Single-blind (patients); operators were not blinded.
  • Statistics:
    • Power calculation: Assuming 75% first-attempt success in the succinylcholine group and a noninferiority margin of 7% (absolute difference), 602 patients per group were required with type I error 5% and type II error 20% (power 80%); target enrolment was 1300 to account for protocol deviations.
    • Analysis: Primary endpoint analysed per protocol using a generalised estimating equation model (centre as clustering factor); randomised-group analysis reported as sensitivity; noninferiority assessed using a one-sided 97.5% confidence interval.
  • Follow-Up Period: Procedural outcomes and complications assessed during RSI and within 15 minutes after intubation; additional reporting during prehospital care until hospital arrival.

Key Results

This trial was not stopped early. Recruitment and follow-up were completed as planned (final follow-up August 31, 2016).

Outcome Rocuronium (1.2 mg/kg) Succinylcholine (1 mg/kg) Effect p value / 95% CI Notes
Successful first-attempt intubation (primary; per protocol) 455/610 (74.6%) 489/616 (79.4%) Absolute difference -4.8% 1-sided 97.5% CI -9.0% to ∞ Noninferiority margin -7%; noninferiority not demonstrated
Successful first-attempt intubation (sensitivity; randomised-group analysis) 456/613 (74.4%) 489/617 (79.2%) Absolute difference -4.8% 1-sided 97.5% CI -9.1% to ∞ Consistent with per-protocol primary analysis
Cormack–Lehane grade II view 125/609 (20.5%) 173/615 (28.1%) Absolute difference -7.6% 95% CI -11.5% to -3.7%; P<0.001 Lower grade II proportion with rocuronium (grade I proportion 61.6% vs 56.2%; P=0.06)
ILMA use (alternative airway procedure) 10/610 (1.6%) 2/616 (0.3%) Absolute difference 1.3% 95% CI 0.5% to 2.4%; P=0.003 ILMA = intubating laryngeal mask airway
Intubation failure under direct laryngoscopy 11/610 (1.8%) 4/616 (0.7%) Absolute difference 1.1% 95% CI 0.3% to 2.3%; P=0.01 Among 15 failures: blind ILMA intubation (n=6), ILMA ventilation then hospital intubation (n=6), cricothyrotomy (n=1), face-mask ventilation to hospital (n=2)
At least 1 intubation-related complication within 15 minutes 111/610 (18.2%) 143/616 (23.2%) Absolute difference -5.0% 95% CI -9.8% to -0.03%; P=0.04 Composite of prespecified complications
Severe arrhythmia within 15 minutes 12/610 (2.0%) 26/616 (4.2%) Absolute difference -2.2% 95% CI -3.8% to -0.7%; P=Not reported Defined as ventricular tachycardia or ventricular fibrillation
Hypotension episode within 15 minutes 39/610 (6.4%) 62/616 (10.1%) Absolute difference -3.7% 95% CI -6.8% to -0.3%; P=Not reported Defined as a new episode of systolic blood pressure <90 mmHg
Cardiac arrest within 15 minutes 22/610 (3.6%) 13/616 (2.1%) Absolute difference 1.5% 95% CI -0.1% to 3.3%; P=Not reported Prespecified complication component
Mean number of intubation attempts (exploratory) 1.4 (SD 0.8) 1.3 (SD 0.6) Mean difference 0.1 95% CI 0.05 to 0.2; P<0.001 Exploratory post hoc analysis
At least 1 severe complication (exploratory) 115/610 (18.9%) 150/616 (24.4%) Absolute difference -5.5% 95% CI -10.8% to -0.05%; P=0.04 Exploratory post hoc definition (as per trial report)
Death during prehospital care 6/606 (1.0%) 3/613 (0.5%) Absolute difference 0.5% 95% CI -0.4% to 1.3%; P=0.26 Follow-up limited to prehospital phase
  • Rocuronium did not meet the prespecified noninferiority criterion for first-pass success (absolute difference -4.8%; 1-sided 97.5% CI lower bound -9.0%, crossing the -7% margin).
  • Despite a lower proportion of Cormack–Lehane grade II views with rocuronium, escalation to ILMA and direct-laryngoscopy failure were more frequent in the rocuronium group (ILMA 1.6% vs 0.3%; failure 1.8% vs 0.7%).
  • Early complications were lower with rocuronium (18.2% vs 23.2%), including fewer severe arrhythmias and hypotension episodes; these were secondary/exploratory outcomes with potential sensitivity to post-intubation co-interventions and multiple comparisons.

Internal Validity

  • Randomisation and allocation:
    • Permuted block randomisation (block sizes 4, 6, and 8), stratified by centre.
    • Numbered, sealed opaque envelopes used for assignment; emergency physicians were blinded to envelope contents until allocation.
  • Dropout/exclusions and analysis sets:
    • Randomised: 1248 patients (624 per group).
    • Primary analysis (per protocol): 1226 patients (610 rocuronium; 616 succinylcholine).
    • Did not receive allocated paralytic: 11/624 (rocuronium group) vs 7/624 (succinylcholine group); received the alternative paralytic: 1 in each group.
  • Performance/detection bias:
    • Single-blind design: operators were not blinded; first-pass success is clinically objective but operator-dependent.
    • Time from paralytic injection to first laryngoscopy attempt was not reported, limiting assessment of whether onset time differed by group.
  • Protocol adherence, timing, and dose:
    • Paralytic dosing was protocolised (rocuronium 1.2 mg/kg vs succinylcholine 1 mg/kg) with a recommended 60-second wait before intubation.
    • Induction hypnotic use was similar (etomidate used in 539/613 [87.9%] vs 544/617 [88.1%] in the randomised groups; median dose 30 mg in both); hypnotic/analgesic and vasopressor use after intubation differed in post hoc analyses.
  • Separation of the variable of interest and co-interventions:
    • Post-intubation sedation/analgesia and vasopressor use (post hoc): midazolam median 5 mg (IQR 4–10) vs 8 mg (IQR 5–13); propofol median 133 mg (IQR 50–200) vs 200 mg (IQR 100–250); norepinephrine use 9.9% vs 16.5%.
    • Rescue/reversal: sugammadex administered to 2 rocuronium-assigned patients; direct-laryngoscopy failure occurred in 15 total patients (11 rocuronium vs 4 succinylcholine), managed with ILMA strategies, cricothyrotomy (n=1), or mask ventilation to hospital.
  • Baseline characteristics and heterogeneity:
    • Groups were broadly comparable (age mean 57.2 vs 54.6 years; median GCS 6 in both; mean SpO2 95.4% in both), with broadly similar indications and patient positioning.
    • Centre effects were accounted for using a clustered model; prespecified subgroup analyses did not demonstrate heterogeneity by indication or patient position (interaction P values reported in supplement).
  • Outcome assessment and statistical rigor:
    • Primary endpoint (first-attempt success) is clinically concrete; complication definitions were prespecified (eg, SpO2 <90%, systolic blood pressure <90 mmHg, ventricular tachycardia/fibrillation).
    • Noninferiority margin prespecified at -7%; primary analysis was per protocol with a sensitivity randomised-group analysis; enrolment (n=1248) was below planned 1300 but above the calculated minimum (602 per group).
    • Secondary/exploratory outcomes were not adjusted for multiplicity, increasing the risk of chance findings.

Conclusion on Internal Validity: Overall, internal validity is moderate: randomisation and follow-up were robust and the primary endpoint is clinically objective, but operator unblinding and unreported timing from paralytic administration to laryngoscopy introduce plausible performance bias; secondary safety outcomes are additionally sensitive to post-intubation co-interventions.

External Validity

  • Population representativeness:
    • Represents adult out-of-hospital RSI candidates in a French physician-staffed emergency medical system, across common indications (coma, self-poisoning, acute respiratory failure, trauma).
    • By design, excludes cardiac arrest and patients with contraindications to either paralytic (and to sugammadex), limiting applicability to those higher-risk subgroups where drug choice is most constrained.
  • Applicability to other systems:
    • Most applicable to services using direct laryngoscopy (Macintosh blades) without routine stylet use and with access to bougie/ILMA/cricothyrotomy pathways.
    • Generalisation may be limited in paramedic-only systems, settings with routine video laryngoscopy, different sedation practices, or limited access to advanced rescue devices and reversal agents.
    • Follow-up was confined to the prehospital phase; longer-term ICU/hospital outcomes were not assessed, which limits policy translation focused on downstream outcomes.

Conclusion on External Validity: External validity is moderate: findings are most generalisable to adult, non-arrest, physician-led prehospital RSI using similar technique and dosing, but apply less directly to other EMS configurations, modern video-laryngoscopy–dominant practice, and patients with paralytic contraindications.

Strengths & Limitations

  • Strengths:
    • Large, pragmatic, multicentre randomised trial in a high-risk out-of-hospital RSI context.
    • Clinically meaningful primary endpoint (first-pass success) with high procedural relevance.
    • Prespecified noninferiority framework with clustered modelling to account for centre effects and sensitivity analysis in the randomised groups.
    • Detailed reporting of rescue airway strategies and early complications within a defined time window.
  • Limitations:
    • Single-blind design with operator unblinding; procedural decisions (timing of attempt, declaring failure) may be influenced by knowledge of paralytic.
    • Time from neuromuscular blocker injection to first laryngoscopy attempt was not reported, limiting mechanistic inference about onset-related differences.
    • Primary per-protocol analysis is appropriate for noninferiority but may introduce bias if protocol deviations are related to prognosis (though sensitivity ITT analysis was consistent).
    • Multiple secondary/exploratory comparisons without multiplicity correction.
    • Follow-up limited to the prehospital phase; no in-hospital outcomes, ventilator-free days, ICU/hospital length of stay, or mortality beyond prehospital care.
    • Exclusion of patients with contraindications to either agent restricts inference to settings where both drugs are considered acceptable options.

Interpretation & Why It Matters

  • Pragmatic implications for out-of-hospital RSI
    • When both agents are clinically permissible, succinylcholine achieved higher first-attempt success than rocuronium in this prehospital RSI setting, and rocuronium could not be assumed equivalent under a prespecified noninferiority margin.
    • Rocuronium use should be paired with careful attention to onset time and a robust rescue plan (including advanced airway adjuncts and, where available and appropriate, reversal capability), given higher rates of escalation (ILMA use and direct-laryngoscopy failure).
    • The lower rate of early complications with rocuronium is clinically interesting but should be interpreted cautiously in the context of secondary/exploratory outcomes and differing post-intubation sedation/vasopressor patterns.

Controversies & Subsequent Evidence

  • Noninferiority interpretation and reporting:
    • Emphasised that the appropriate inference is “noninferiority not demonstrated” rather than proof of inferiority, given the confidence interval crossed the noninferiority margin. 1
  • Co-interventions and haemodynamic outcomes:
    • Highlighted that differences in post-intubation sedation/analgesia and vasoactive support could influence hypotension and other physiological complications, complicating causal attribution of safety signals to the paralytic alone. 2
  • Trialist clarifications:
    • Clarified the intended interpretation of the noninferiority analysis and discussed the observed differences in sedative/analgesic use as potential contributors to haemodynamic findings, alongside acknowledgement of procedural constraints. 3
  • Follow-up analyses and external datasets:
    • Secondary analysis of CURASMUR data found first-attempt failure (23.2% overall) was associated with higher complication rates, reinforcing the clinical importance of maximising first-pass success in the out-of-hospital setting. 4
    • A large emergency department airway registry study found no significant difference in first-pass success between succinylcholine and rocuronium (87.0% vs 87.5%), underscoring that observed effects may vary by setting, operator, technique, and case mix. 5
  • Guideline synthesis:
    • Recent ICU and emergency airway guidance generally supports either succinylcholine or rocuronium for RSI, with drug selection guided by contraindications, haemodynamic physiology, and system capability for rescue/reversal; CURASMUR is incorporated as part of the evidence base informing these recommendations. 6789

Summary

  • In this multicentre French out-of-hospital RSI trial, rocuronium (1.2 mg/kg) did not demonstrate noninferiority to succinylcholine (1 mg/kg) for first-attempt intubation success.
  • First-pass success was 74.6% with rocuronium vs 79.4% with succinylcholine (absolute difference -4.8%; 1-sided 97.5% CI -9.0% to ∞; noninferiority margin -7%).
  • Airway escalation occurred more often with rocuronium (ILMA 1.6% vs 0.3%; direct-laryngoscopy failure 1.8% vs 0.7%), although laryngoscopic view distribution did not uniformly favour succinylcholine.
  • Early complications were lower with rocuronium (18.2% vs 23.2%), including fewer severe arrhythmias and hypotension episodes, but these secondary/exploratory findings are potentially sensitive to co-interventions and multiplicity.
  • Findings are most applicable to physician-led prehospital RSI using direct laryngoscopy and similar dosing; robust rescue planning remains essential, particularly when using longer-acting paralysis.

Overall Takeaway

CURASMUR is a landmark pragmatic prehospital RSI trial because it directly tested a widely debated system-level decision—succinylcholine versus high-dose rocuronium—under real out-of-hospital conditions. Rocuronium did not meet a prespecified noninferiority threshold for first-pass success and was associated with more airway escalation, while showing fewer early complications that require cautious interpretation. The trial reinforces that paralytic choice cannot be separated from operational readiness: onset timing discipline, sedation practice, and robust rescue pathways are integral to safe RSI.

Overall Summary

  • Rocuronium 1.2 mg/kg did not demonstrate noninferiority to succinylcholine 1 mg/kg for first-pass intubation in out-of-hospital RSI (74.6% vs 79.4%).
  • Rocuronium was associated with more airway escalation (ILMA use; direct-laryngoscopy failure) but fewer early complications (notably hypotension and severe arrhythmia), with interpretive caveats for secondary/exploratory outcomes.
  • Drug choice should be embedded in a system approach: contraindications, onset timing, sedation strategy, and rescue/reversal capability determine real-world safety.

Bibliography