Full Citation: Freund Y, Viglino D, Cachanado M, Cassard C, Montassier E, Douay B, Guenezan J, Le Borgne P, Yordanov Y, Severin A, Roussel M, Daniel M, Marteau A, Peschanski N, Teissandier D, Macrez R, Morere J, Chouihed T, Roux D, Adnet F, Bloom B, Chauvin A, Simon T. Effect of Non-invasive Airway Management of Comatose Patients with Acute Poisoning: A Randomized Control Trial. JAMA. 2023 Dec 19;330(23):2267–2274.


Abstract Link: https://jamanetwork.com/journals/jama/article-abstract/2812429


Article Type: Therapy


Ratings: Methods—(4/5); Usefulness—(4/5)

Introduction

Background

Poisoned patients with depressed levels of consciousness (LOC) may be at increased risk for aspiration and related complications. Intubation is often recommended for airway protection.

Objectives

To determine the effect of withholding intubation vs routine practice on clinical outcomes of acutely poisoned patients with depressed LOC.

Methods

Design

Multicenter randomized control trial (RCT).

Setting

Twenty emergency departments and one intensive care unit in France.

Subjects

Patients ≥ 18 years of age with suspected acute poisoning and a Glasgow coma score (GCS) < 9.

Intervention

Withholding intubation unless an emergency intubation criterion was met within 4 h of presentation, or up to when the patient had an increase in GCS > 8.

Comparison

Intubation at the discretion of the treating physician.

Outcomes

The primary outcome was a hierarchical composite of in-hospital death, length of ICU stay and length of hospital stay. This was expressed as a win ratio where each patient-to-patient comparison was assigned a win, loss or tie based on a hierarchy of end points ordered by clinical importance. Secondary outcomes of this study included adverse events secondary to intubation and pneumonia within 48 h.

Main results

225 patients were randomized into the intervention (n = 116) and control groups (n = 109). Most patients were male (62%), with a median GCS of 6 [IQR, 3–7]. Associated toxins included alcohol (67%) followed by benzodiazepines (39%). Some patients had multiple exposures. The composite endpoint was significantly better in the intervention group, with a win ratio of 1.85 (95% CI, 1.33–2.58) (Fig. 1). This was driven by a shorter length of ICU stay (RR = 0.39, 95% CI 0.24–0.66) and length of hospitalization (RR = 0.74, 95% CI 0.53–1.03) since there were no deaths in either group. Secondary outcomes were more favourable among the intervention group with fewer ICU admissions (39.7% vs 66.1%; absolute risk difference, – 29.2 percentage points [95% CI – 41.0–17.4]), fewer adverse events during intubation (6. 0% vs 14.7%; absolute risk difference – 8.6 percentage points, 95% CI, – 16.6–0.7) and fewer occurrences of pneumonia (6.9% vs 14.7%; absolute risk difference – 7.8 percentage points, 95% CI, – 15.9–0.3).

Fig. 1
figure 1

Distribution of Wins, Ties and Losses for Patients

Full size image

Appraisal

Strengths

  • Prospective multicenter RCT.

  • First RCT to study this important clinical question.

  • Intention-to-treat primary analysis.

  • Important, patient-oriented outcomes.

  • Equal distribution of substances ingested between groups including benzodiazepines and GHB.

  • Included many patients with profoundly decreased LOC (GCS = 3).

Limitations

  • Unblinded trial; cannot control for potential Hawthorne effect.

  • Primary composite outcome included in-hospital mortality without any patient deaths in either group.

  • Inadequate explanation of adverse events. It is unclear where the n = 113 (intervention group) and n = 107 (control group) were derived. Furthermore, adverse events from intubation should not be reported among patients who were not intubated.

  • Cohort of primarily alcohol intoxicated patients for whom intubation is rarely indicated, with a lack of granularity with respect to coingestants between groups.

  • Excluded patients who overdosed on cardiotoxic medications or substances with a reversal agent.

  • Lack of generalizability in the Canadian context where prehospital assessments are not routinely done by physicians.

  • No measurement of impact on ED workflow, where intervention shifts care from ICU to ED team.

Context

The Glasgow coma scale was initially developed to aid assessment of head injured trauma patients. Over time, a low composite GCS score evolved into a surrogate marker for an unprotected airway, driving the mantra “GCS ≤ 8, then intubate” [1]. Observational studies suggest that a low GCS score does not necessarily predict the need for intubation among poisoned patients, where depth and duration of impairment is a manifestation of drug toxicity and not structural brain injury [2, 3]. Among Toronto emergency medicine and medical toxicology experts, decisions to intubate poisoned patients are based on expected clinical course, informed by ingestion history, and physiologic parameters indicating inadequate ventilation.

Bottom line

This study adds to the evidence that using a conservative airway management strategy among poisoned patients with a depressed LOC may lead to better outcomes including fewer ICU admissions and decreased length of hospitalization.