Prospective validation of the modified LEMON criteria to predict difficult intubation in the ED

Evidence to predict difficult intubation remains scarce in the emergency department (ED) setting. A previously defined clinical decision rule, the modified LEMON criteria, may provide a reliable and reproducible means of identifying difficult intubations. We aimed to prospectively evaluate the external validity of the modified LEMON criteria in the EDs.

We conducted a 13-center prospective observational study, the second Japanese Emergency Airway Network study. We prospectively collected data on all patients undergoing intubations in the ED from February 2012 through September 2014. The primary outcomes were sensitivity, specificity, and predictive values of the modified LEMON criteria for predicting difficult intubation (≥2 attempts by emergency attending physicians or anesthesiologists).

The database recorded a total of 4034 encounters (capture rate, 96%) in the EDs. Of these, 3313 patients (84%) underwent the intubation attempt with a direct laryngoscope and 610 patients (16%) with a video laryngoscope. The proportion of difficult intubation was 5.4% (95% confidence interval [CI], 4.7%-6.2%) in the direct laryngoscope group and 7.4% (95% CI, 5.6%-9.7%) in the video laryngoscope group. The sensitivity was 85.7% (95% CI, 79.3%-90.4%) with direct laryngoscope and 94.9% (95% CI, 83.5%-98.6%) with video laryngoscope. The specificity was 47.6% (95% CI, 47.2%-47.9%) and 40.3% (95% CI, 39.4%-40.6%), respectively. The negative predictive value was 98.2% (95% CI, 97.5%-98.8%) and 99.0% (95% CI, 96.6%-99.7%), respectively.

In this multicenter prospective study, we found a high sensitivity and a negative predictive value of the modified LEMON criteria for predicting difficult intubation. The modified LEMON might assist ED providers in better identifying difficult intubations.