LEMON SCORE: A TOOL TO PREDICT DIFFICULT AIRWAY IN ED IN INDIAN SETTING

Background: Airway management is a critical need in many acutely ill and injured patients . Inadequate delivery of oxygen to brain and other vital structures is the quickest killer. Emergency airway management is the major key for successful resuscitation in ED. Emergency medicine is an emerging branch in India at present and no research study has been conducted to evaluate difficult airway prediction score. Hence this study aimed to find the use of LEMON score as a tool to predict difficult airway in our ED. Methods: All the patients requiring invasive mechanical ventilation with age >12 years, not admitted outside and admitted to the emergency medicine department from July 2017 to September 2019 were included in our study. A total of 67 patients required invasive ventilation and were assessed by LEMON score for difficult intubation. This score is related to number of attempts required and Cormack lehane class of laryngoscopic view while intubation. Results: In the “LOOK EXTERNALLY” the most common finding was edentulous mouth with occurrence of 26.87%. In ‘EVALUATE’ component the most common difficulty was 2-3-2. In MALLAMPATI CLASS 56.71% were class I, 39.39% were class II. In OBSTRUCTION component of LEMON, 98.51% had no obstruction. In NECK MOBILITY component we found that 85.1% subjects had mobile neck and only 15.15% subjects had restricted neck mobility. We observed that the LEMON score is 60% sensitive and 96.15% specific to predict difficult airway. The positive predictive value was 83.33%. Conclusion: This tool can reduce the chance of unexpectedly encountering difficult airway.


Introduction
Airway management is a critical need in many acutely ill and injured patients . Inadequate delivery of oxygen to brain and other vital structures is the quickest killer. An unobstructed, protected airway and adequate ventilation are critical to prevent hypoxemia.
Emergency airway management is the major key for successful resuscitation in ED. Emergency department deals with most of the cases with threatened airway and impending respiratory failure requiring mechanical ventilation. This early period of care has significant impact on the outcomes of these patients. Emergency medicine is an emerging branch in India at present and no research study has been conducted to evaluate baseline demographic features, indications, difficult airway assessment, modes of ventilation etc for patients requiring mechanical ventilation in the ED setting. Therefore by this observational study we try to study the baseline characteristics of patients requiring mechanical ventilation, difficult airway assessment and difficulties in airway management in ED setting. Objective signs of airway obstruction are: 1. Agitation (hypoxia) or obtundation (hypercarbia). 2. Use of accessory muscles of ventilation that predicts the airway compromise. 3. Abnormal sounds i.e. Snoring, gurgling and crowing sounds can be associated with partial occlusion of pharynx or larynx. 4. Abusive behaviour of the patient may be due to hypoxemia. 5. Involuntary motor movements. Definitive airway is defined as tube placed in the trachea with cuff inflated below the vocal cords secured in place with tape.

INDICATIONS FOR INTUBATION
a. Inability to maintain a patent airway by other means, with impending or potential airway compromise( e.g. following, facial fractures, retropharyngeal hematoma or inhalational injury) b. Inability to maintain adequate oxygenation with the help of facemask oxygen supplementation, or the presence of apnea. c. Obtundation resulting from cerebral hypoperfusion d. Head injury with GCS less than or equal to 8, sustained seizure activity, and need to protect the airway from aspiration of blood or vomitus. An intubation is called difficult if a trained physician needs 2 or more attempts or more than 10 min for a successful endotracheal intubation. In this study we used LEMON score to assess the difficult airway.

'LEMON' ASSESSMENT FOR DIFFICULT INTUBATION L: LOOK EXTERNALLY
Look for signs of inhalational burn, large overbite, facial trauma, edentulous mouth,short neck or obesity etc.

E: EVALUATE 3-3-2 RULE:
To allow for the alignment of laryngeal, pharyngeal and oral axes, observe the following: Distance between the patient's incisor teeth-atleast 3 finger breadths Distance between the chin and hyoid bone -3 finger breadths Distance between thyroid notch and floor of the mouth -2 finger breadths. M: Mallampati score: Class I: soft palate, fauces pillars, uvula entirely visible Class II: soft palate, fauces, uvula partially visible. Class III: soft palate, base of uvula visible Class IV: only hard palate visible O: Obstruction Any condition that causes obstruction of the airway will make laryngoscopy difficult. Eg foreign body. N: Neck mobility Patients who require cervical spinal motion restriction are more difficult to intubate This is usually seen in cases of trauma, cervical stenosis, etc.

Figure 1: Distribution of 'Look externally' component among study subjects (ii) Evaluate 3-3-2 rule:
It was observed that 46 out of 67 subjects intubated maintained 3-3-2 finger breadth with respective anatomy. The most common variation was 2-3-2 , with frequency of 7 (10.45%). (iii) Mallampati score: most of the patients (56.71%) were class I, followed by class II (39.39) and class III (4.55). None of the patient had class IV score.     The cut off for lemon score is taken as equal to or more than 3. After comparing the 'lemon' score with difficult intubations, it was observed that: The overall sensitivity of 'LEMON SCORE' to predict difficult intubation (i.e. more than or equal to 2 attempts) is 60% and specificity is 96.15%. The positive predictive value that means the probability of patients with positive test will truly have the disease is 83.33%.

Distribution of Cormack lehane laryngoscopic view among study subjects:
It was observed that maximum 67.16% subjects were of class 1, 20.89% were of class 2a and 11.94% were of class 2b category. Class 3 and 4 views were not observed.

Relationship between Cormack lehane class with number of attempts of intubation:
It was found that 95.5% cases of Cormack Lehane class 1 were intubated in first attempt, 78.57% cases of Cormack Lehane class 2a were intubated in 2 nd attempt and 62.5% cases of Cormack Lehane class 2b were intubated in 3 rd attempt. It was concluded that as the Cormack Lehane class increases number of attempts of intubation also increases.  Out of the total 24 difficult intubations 19 were assisted and successfully intubated with gum elastic bougie. Laryngeal mask airway was not used in any of these subjects and surgical cricothyroidotomy was not required in any of the difficult airway.
There was no incidence of failed intubation in our study.

DISCUSSION:
'LEMON' score as a tool to predict difficult airway: In present study we have used "LEMON" approach to predict difficult airway. In the "LOOK EXTERNALLY" we found that the most popular difficulty was edentulous mouth with occurrence of 26.87%.
In 'EVALUATE' component the most common difficulty was 2-3-2 ie. 2 finger breadth distance between incisors, 3 finger breadth between chin and hyoid bone and 2 finger breadth between thyroid notch and floor of the mouth. The occurrence was 10.45%.
In MALLAMPATI CLASS 56.71% were class I, 39.39% were class II, and 4.55% were class III.
In OBSTRUCTION component of LEMON, 98.51% had no obstruction and only one subject was found to have laryngeal tumor (1.49%).
In NECK MOBILITY component we found that 85.1% subjects had mobile neck and only 15.15% subjects had restricted neck mobility.
Although the mallampati score was assessed in our study as a part of the LEMON approach during emergency intubation, it was observed that it was not feasible as compared to the remaining components.
This conclusion was similar to the study by Reed  easily applied to the population that present to the resuscitation room, and assessment of these was more problematical and prone to inaccuracy.
In present study we have observed that the LEMON score is 60% sensitive and 96.15% specific to predict difficult airway. The positive predictive value i.e. the probability of patients with positive test will truly have the disease was 83.33%.
This is similar to the study conducted by Diane M. Birnbaumer,et al [50] who found that patients in the difficult-intubation group had significantly higher LEMON scores than did those in the easy-intubation group. They also validated that the LEMON method predicts which patients have difficult-to-manage airways.

Distribution of Cormack Lehane Scale to classify laryngoscopic view:
In our study we have used Cormack lehane scale to classify the laryngoscopic view. We found that Maximum 67.16% were of class 1, 20.89% were of class 2a and 11.94% were of class 2b category. Then we tabulated the number of attempts required for Cormack lehane class 1 , 2a and 2b( class 3 and 4 were not found in our subjects) and related it with number of attempts required for intubation. . It was found that 95.5% cases of Cormack Lehane class 1 were intubated in first attempt, 78.57% cases of Cormack Lehane class 2a were intubated in 2 nd attempt and 62.5% cases of Cormack Lehane class 2b were intubated in 3 rd attempt. It was concluded that higher the class of Cormack lehane, more is the number of attempts of intubation.
The sensitivity of Cormack lehane scale, in our study was 33.33% and specificity for difficult intubation was 100% with a positive predictive value of 100%.
Out of the total 24 difficult intubations 19 were assisted and successfully intubated with gum elastic bougie. Laryngeal mask airway was not used in any of these subjects and surgical cricothyroidotomy was not required in any of the difficult airway.
This was in agreement with the study done by M J Reed, M J G Dunn,D W McKeown, et al [49] in which 156 patients were intubated during the study period. There were 114 Cormack and Lehane grade 1 intubations, 29 grade 2 intubations, 11 grade 3 intubations, and 2 grade 4 intubations. they concluded that patients with a poor laryngoscopic view (grades 2, 3, or 4) were more likely to have increased attempts of intubation with difficulty.
It is also in favour with the study done by Shirgoska B , Netkovski J. et al [51] who also concluded that increasing grade of Cormack lehane classification of laryngoscopic view is associated with difficult intubation.

CONCLUSION
For assessing LEMON score, 'look externally', 'evaluate', 'obstruction' and 'neck mobility' were easiest to access while mallampati was not feasible before emergency intubation and was prone to errors. The overall sensitivity and specificity of this tool is 60% and 96.15% respectively to predict difficult airway with a positive predictive value of 83.33%. Use of this tool can reduce the chance of unexpectedly encountering a difficult airway.
The increase in Cormack-lehane classification for laryngoscopic view is related to increased number of attempts for intubation and higher LEMON score.. It is highly specific for difficult intubation.
Gum elastic bougie was used in majority of difficult intubations and has proved to be a useful tool in emergency scenario. As soon as we predict difficult airway by using 'LEMON' score or observe an airway of high Cormack lehane score, next attempt of intubation should be assisted with GEB to save time and maintain patent airway.