When life gives you LEMONs- Predicting difficult intubations in the ED


Reed MJ, Dunn MJ, & McKeown DW. Can an Airway Assessment Score Predict Difficulty at Intubation in the Emergency Department? Emerg Med J 2005; 22(2): 99 – 102.

Main Points:

  1. Rapid assessment tools can be helpful in predicting difficult intubations in the emergency department
  1. Use of SOME elements of the LEMON (look, evaluate, mallampati, obstruction, neck mobility) approach to airway assessment MAY be helpful in predicting likely laryngoscopic view (Cormack- Lehane grade) as a proxy for difficulty of intubation. The following are more likely found in patients with high grade views (2-4).
  • large incisors
  • reduced inter-incisor distance
  • reduced thyroid to floor of mouth distance


Predicting difficult intubations is not always straightforward. At the time of publication (2005), little validation of rapid assessment of possible difficult intubations in the ED. The authors test the use of the LEMON approach as a predictor of difficult intubations, and suggest key parts of the assessment that are most helpful.



The study was a prospective, observational trial conducted in the UK at a teaching Emergency Department between June 2002 and September 2003.   156/177 patients intubated over that time were included in the study. Those excluded were done so because no LEMON assessment was completed. Of the remaining included, a modified LEMON assessment was completed including: LOOK- facial trauma, large incisors, large tongue, facial hair; EVALUATE- inter-incisor distance (<3 fingers), hyoid-mental distance (<3 fingers), thyroid to floor of mouth distance (<2 fingers); MALLAMPATI 1/2 versus 3/4 ; OBSTRUCTION; and NECK MOBILITY- cervical collar versus no collar. One point was assigned for each criterion that was found. If a criterion was though unassessable, a score of zero was given. Outcome was determined by laryngoscopic view as outlined by the Cormack-Lehane grading scale; grade 1 was considered an easy intubation, grades 2-4 were considered difficult. ALL intubations were successful, and if multiple attempts were used, the grade of view on the successful attempt was used. Authors used Fischer’s exact test to compare the categorical variables, Student’s t test to compare continuous data. Spearman rank sum test was used to assess correlation between categorical variables.


114 patients were found to be easy intubations, while 42 were in the difficult intubation group. There was no significant difference in proportion of patients in each group by sex or by age (65.8% male in easy vs 76.2% in difficult, p=0.25; 53.8 avg age easy, 55.7 avg age difficult, p=0.86). Taken individually, the LEMON criteria which were more likely present in patients with difficult intubations were “large incisors” (P<0.001), Reduced inter-incisor distance (p=0.05), and reduced thyroid to floor of mouth distance (p=0.05). No significant difference was found among Mallampati score, presence of an airway obstruction or limited neck mobility. Taken together, those with difficult intubations had significantly higher assessment scores (median 1.0 vs 2.0, P<0.05), and there was a positive correlation between higher intubation score and difficulty of intubation ( r=0.38, p<0.001).


Conclusions: The authors propose a modified LEMON which includes the LOOK criteria assessed, EVALUATE criteria, OBSTRUCTION, and NECK MOBILITY. They concluded that Mallampati is not easily assessed (only documented in 57% of patients intubated).


Level of Evidence:

Based on ACEP grading scheme, I would give this a level III class of evidence for possible issues due to exclusion, inter-user reporting of outcome, and the fact that the study only indirectly measured its desired outcome of difficulty of intubation (ie though Cormack Lehane view may represent higher level of technical difficulty, number of attempts, use of assistive devices (glidescope, etc) was not included in the analysis)


Some thoughts:


  • NOT recorded were level of trainee, number of attempts, whether RSI medicaions were used, all of which could contribute to the “difficulty” of the intubation. Several of these things were noted by the authors as limitations, and they certainly contribute to what we as clinicians think of as the “difficult airway.”
  • The use of the Cormack-Lehane is probably not an ideal proxy for the difficult intubation, especially as used in this study. By dichotomizing this classification, rather than using it as is, was likely done more for sample size issues than the fact that each of these views is equally difficult.
  • In a crashing patient, assessing all of these characteristics, even in a modified way as suggested by the authors, is not likely feasible. The study group got around this by giving a score of zero, rather than giving a point. This may have falsely decreased the predicted difficulty.


Relevant Resources:

Life in the Fast Lane- Airway Assessment

Gangadharan L, et al. Prediction of difficult intubations using conventional indicators: Does rapid sequence intubation ease difficult intubations? A prospective randomised study in a tertiary care teaching hospital. Journal of Emergencies, Trauma and Shock. 2011;4(1):42-47.

Levitan, R, et al. Limitations of difficult airway prediction in patients intubated in the emergency department. Annals of Emergency medicine. 2004; 44 (4): 307-313. 

One thought on “When life gives you LEMONs- Predicting difficult intubations in the ED

  1. Pingback: The Cricothyrotomy Part 1: The Procedure | Brown Emergency Medicine

Leave a Reply