Secondary Logo

Journal Logo

Original Papers

Value of oropharyngeal Mallampati classification in predicting difficult laryngoscopy among obese patients

Voyagis, G. S.; Kyriakis, K. P.; Dimitriou, V.; Vrettou, I.

Author Information
European Journal of Anaesthesiology: May 1998 - Volume 15 - Issue 3 - p 330-334



A short fat neck is widely believed to be associated with difficult intubation because of limited volume for displacement of tongue during laryngoscopy [1]. Wilson et al.[2] quantified the problem by simply equating it with obesity. The latter has been disputed [3] and the role of obesity in complicating airway management remains controversial. Previous studies have suggested that the degree of visibility of the oropharyngeal structures (OP class) as a means of predicting the degree of difficulty in exposure of larynx by direct laryngoscopy is determined by the ratio of tongue to pharyngeal size [4]. The original Mallampati test and all its modifications evaluate the OP class with or without phonation [5-10], with the body in sitting or supine position [4-11], with the head in neutral or sniffing position [4,6,10,11], but always with the tongue being maximally protruded by the patient. The test has considerable observer variability and has been criticized with regard to its important false positive and false negative rates [5,9,12,13].

The purpose of this study was to assess the usefulness of OP class estimation in predicting difficult laryngoscopy for obese and non obese patients. The OP class estimation methodology was carried out using the original and a modified technique of tongue protrusion.

Materials and methods

The Medical Ethics Committee of our institution approved the study protochol. Because no therapeutic intervention was made only the routine pre-operative consent was required. The study was conducted on 1833 consecutive adult patients undergoing endotracheal anaesthesia for any type of non-emergency surgical procedures except obstetric and cardiac surgery, which are not performed in our hospital. Patients with other known risk factors for difficult intubation such as traumatic facial abnormalities, limitation of mobility of head and neck (<90°) or airway pathology [1,2] were excluded from the study as these might make interpretation of the results more difficult.

Data were collected prospectively by means of a standardized record sheet, which was designed in such a way that it was easy to complete. The results were entered at the time of the pre-operative interview and during administration of the anaesthetic. Two senior anaesthetists carried out the patients' airway assessment, the induction of anaesthesia and the airway control according to the protocol. Body mass index (BMI=body weight × height−2) was used as a measure of obesity [14]. Ranges of BMI in kg m−2 were defined as follows: morbid obesity >40, moderate obesity=30-40, absence of obesity <30 [15]. The OP class was evaluated with the patient sitting upright, the head being in neutral position and the tongue protruded out by the patient. Inability to see any of the posterior pharyngeal wall was accepted as a positive prediction of a difficult laryngoscopy [5,13]. Thereafter the initial method was compared with the result obtained when the tongue was actively pulled out by the anaesthetist. Still in the sitting position the patient was instructed to open the mouth as wide as possible. The patient's tongue covered with a gauze was maximally but gently pulled out and downwards with a two-fingers manoeuvre. The operator held the tongue between the thumb and index finger, with the thumb placed deep to the floor of the tongue up to the frenulum linguae. The OP class estimation was determined by the original procedure when the tongue was protruded by the patient and modified when the tongue was actively pulled out. A standard schedule for induction and orotracheal intubation was used as follows: intravenous (i.v.) anaesthesia, suxamethonium 1.5 mg kg−1, head in the optimum sniffing position, laryngoscopy following cessation of fasciculations in the face and neck using a Macintosh laryngoscope, blade 3 or 4. Difficulty in intubation was estimated according to the view obtained at direct laryngoscopy (all or part of vocal cords = easy, no part of glottis or epiglottis = difficult) [16].

Statistical terms used in the data analysis are defined as follows: Sensitivity = number of difficult laryngoscopies correctly predicted vs. total number of difficult laryngoscopies; positive predictive value = number of difficult laryngoscopies correctly predicted vs. total number of laryngoscopies predicted to be difficult. Different frequencies were analysed by the Fisher's exact test, χ2 test Yates corrected and the respective 95% confidence limits for relative risk (RR) and odds ratio (OR) were calculated.


Of the 1833 patients there were 152 (8.3%) who proved to be difficult to intubate. The modified and original techniques were found to be equally sensitive, but the modified technique had a significantly increased positive predictive value (Table 1).

Table 1
Table 1:
Sensitivity and positive predictive value of the oropharyngeal classification methodologies (original vs. modified procedure)

The prevalence of obesity was 5.4% (morbid obesity: 0.7%, 12 of 1833; moderate obesity: 4.7%, 87 of 1833). Table 2 shows the relation between the visibility of pharyngeal structures and the presence of obesity. There was no significant difference with results obtained using the two techniques for morbidly obese and moderately obese patients. Thereafter, the nonobese group was compared with the overall obese population and was found to differ significantly in respect of the OP findings (Table 2, non-visible posterior pharynx: 13.7%, 237 of 1734 vs. 27.3%, 27 of 99; χ2 test Yates corrected, P<0.001, OR=2.4, 95% confidence intervals: 1.5<OR<3.9).

Table 2
Table 2:
Distribution of modified oropharyngeal classification according to the presence of obesity

There was no difference in the predicted difficulty of intubation in the morbidly obese patients (16.7%, 2 of 12) and those whose obesity was characterized as moderate (20.7%, 18 of 87). As both obese populations were found to be statistically homogeneous with regard to the probability of difficulty, they were analysed as a single population. Statistical analysis revealed an increased risk of difficult intubation in obese patients compared with non-obese patients (probability of difficulty: 20.2%, 20 of 99 vs. 7.6%, 132 of 1734; P<0.001, RR = 2.7, 95% confidence intervals: 1.7<RR<4.1).

The modified technique of OP estimation showed a trend towards a higher positive predictive value in obese patients, although the difference between obese and non-obese patients was not statistically significant. The sensitivity of the test was similar in obese and non-obese patients (Table 3).

Table 3
Table 3:
Distribution of results of direct laryngoscopy using the pre-operative assessment of the modified oropharyngeal classification and the presence of obesity

Obesity alone, though a sensitive predictor, may yield false-positive results, as 80% of all intubations were proved to be easy. However, when obese patients are considered according to OP class the positive predictive value is greatly improved (66.7%, 18 of 27 vs. 20.2%, 20 of 99; χ2 Yates corrected, P<10−5, OR = 7.9, 95% confidence intervals: 2.8<OR<22.6).


The data presented here suggest that any increase in BMI above 30 kg m−2 contributes to difficult intubation. This result is in agreement with previously reported data [2,17] and also theoretical conclusions [1,18,19]. Nonetheless, there exist some different conclusions, resulting from a limited series of data, have been drawn [3].

In the multifactorial scoring system used by Wilson et al.[2] a body weight of 95-110 kg rates 1 risk factor for difficult intubation and a weight in excess of 110 kg rates 2. This scoring system presumes that the greater the degree of obesity the greater the probability and degree of difficulty. On the contrary our findings show that morbid obesity should not be considered to be a more serious risk factor than moderate obesity.

Though the Mallampati test has been widely used it is not considered reliable as the degree of prediction ranges widely [9]. Lewis et al.[8] found that the test results are more reproducible when the tongue is out during phonation. On the other hand, many previous studies have reported that gagging or phonation resulted in unpredictable motion of the pharynx and should therefore be avoided as this may obstruct the view [5,9,20,21]. Provided that the patient opens his or her mouth and protrudes the tongue maximally the view should be constant [5]. However, many individuals are unable to perform correctly both the above instructions at the same time giving rise to many false positive results. The active pulling of the tongue by the anaesthetist succeeds in achieving maximal tongue protrusion without any motion of the posterior pharyngeal wall. This modification of the OP estimation leads to a more objective evaluation of the real dimensions of the base of the tongue, especially in individuals with low compliance and improves the positive predictive value.

The tongue, the largest single structure in the mouth, affects the accessibility of the laryngeal inlet to direct laryngoscopy. Obesity contributes to difficult laryngoscopy whenever the weight gain is accompanied by an increase in the size of the tongue such that the relative proportions of the tongue vs. the pharyngeal capacity are altered. The change in pharyngeal capacity may not affect all obese subjects, as obesity was not invariably associated with an inability to see the posterior pharyngeal wall. However, it should be noted that in the case of obese subjects inability to see the posterior pharyngeal wall occurred twice as often as in non-obese subjects and a positive correlation was found between the BMI and the OP class estimation.

The anaesthetized, muscle relaxed, obese patient can be difficult to oxygenate by using positive pressure ventilation through a facemask and oxygen haemoglobin desaturation occurs quickly, especially when multiple attempts at intubation are required [3,22]. Whenever obesity is accompanied by an inability to see the posterior pharyngeal wall, an elective awake intubation should be considered.


1 Murrin KR. Intubation procedure and causes of difficult intubation. In: Latto IP, Rosen M, eds. Difficulties in Tracheal Intubation. London: Baillière Tindall, 1985: 75-89.
2 Wilson ME, Spiegelhalter D, Robertson J. Predicting difficult intubation. Br J Anaesth 1988; 61: 211-216.
3 Bond A. Obesity and difficult intubation. Anaesth Intens Care 1993; 21: 828-830.
4 Mallampati SR, Gatt SP, Glugino LD, Desai SP, Waraksa B, Freiberger D. A clinical sign to predict difficult tracheal intubation: a prospectivey study. Can Anaesth Soc J 1985; 32: 429-434.
5 Frerk CM. Predicting difficult intubation. Anaesthesia 1991; 46: 1005-1008.
6 Roucke DA, Murray WB, Rout CC, Gouws E. Relative risk analysis of factors associated with difficult intubation in obstetric anesthesia. Anesthesiology 1992; 77: 67-73.
7 Cohen SM, Laurito CE, Segil LJ. Examination of the hypopharynx predicts ease to laryngoscopic visualization and subsequent intubation: A prospective study of 664 patients. J Clin Anesth 1992; 4: 310-314.
8 Lewis M, Keramati S, Benumof JL, Berry CC. What is the best way to determine oropharyngeal classification and mandibular space length to predict difficult laryngoscopy? Anesthesiology 1994; 81: 69-75.
9 Jacobsen J, Jensen E, Waldau T, Poulsen TD. Preoperative evaluation of intubation conditions in patients scheduled for elective surgery. Acta Anaesth Scand 1996; 40: 421-424.
10 Thani EJ, Gildersleve CD, Sanders LD, Mapleson WW, Vaughan RS. Effects of posture, phonation and observer on Mallampati clasification. Br J Anaesth 1992; 68: 32-38.
11 Samsoon GT, Young JB. Difficult tracheal intubation: A retrospective study. Anaesthesia 1987; 42: 487-490.
12 Oates JDL, McIeod AD, Oates PD, Pearsall FJ, Howie JC, Murray GD. Comparison of two methods for predicting difficult intubation. Br J Anaesth 1991; 66: 305-309.
13 Wilson ME. Predicting difficult intubation (Editorial). Br J Anaesth 1993; 71: 333-334.
14 Frankel HM. Determination of body mass index (Letter). J Am Med Ass 1986; 255: 1292.
15 Kral JG. Morbid obesity and related healthy risks. Ann Intern Med 1985; 103: 1043-1047.
16 Cormack RS, Lehane J. Difficult tracheal intubation in obstetrics. Anaesthesia 1984, 39, 1105-1111.
17 Rose DK, Cohen MM. The airway: problems and predictions in 18 500 patients. Can J Anaesth 1994; 41: 372-383.
18 Yentis SM. A laryngoscope adaptor for difficult tracheal intubation. Anaesthesia 1987; 42: 764-766.
19 Fahy L, Horton WA, Charters P. Factor analysis in patients with a history of failed tracheal intubation during pregnancy. Br J Anaesth 1990; 65: 813-815.
20 Oates JDL, Oates PD, Pearsall FJ, McIeod AD, Howie JC. Phonation affects Mallampati class (Letter). Anasthesia 1990; 45: 984.
21 Wilson ME, John R. Problems with the Mallampati sign. Anaesthesia 1990; 45: 486-487.
22 Benumof JL. Management of the difficult adult airway. Anesthesiology 1991; 75: 1087-1110.


© 1998 European Society of Anaesthesiology