The Incidence of Class “Zero” Airway and the Impact of Mallampati Score, Age, Sex, and Body Mass Index on Prediction of Laryngoscopy Grade

The ability of a specific test (i.e., the Mallampati scoring system) to predict a difficult intubation is decreased by the variability of definitions of difficult intubation/laryngoscopy (1–4) and the inherent inaccuracy of numeric grading systems (5). However, the Mallampati scoring system is still widely used to evaluate airways before surgery. Ezri et al. (6) and Maleck et al. (7) have described a new class of airway view, class zero, and propose to add this class to the four modified Mallampati classes. This study estimates the incidence of class zero airway, determines the ability of a class zero view to predict laryngoscopy grade, and assesses the effect of the airway classes, age, sex, body mass index (BMI), and other factors on the prediction of the laryngoscopy grade. Methods During a 2-mo period, all patients greater than 18 yr of age who spent time in the preoperative holding area between 8:00 am and 4:00 pm were enrolled in this prospective study. With approval of the institutional ethics committee from the University of Texas at Houston, informed consent was obtained before each assessment. We excluded patients receiving regional anesthesia and patients receiving general anesthesia without endotracheal intubation. Also excluded were patients with upper airway pathology (i.e., maxillofacial fractures, tumors, etc.), cervical spine fractures, and increased risk for aspiration of gastric contents. The airway class was assessed according to the Samsoon and Young (8) modification of Mallampati’s classification. We added class zero (6) to the four classes of the Mallampati system. Class zero is defined as the ability to see any part of the epiglottis upon mouth opening and tongue protrusion (Fig. 1). All the airway assessments were done by the same anesthesiologist, in the sitting position, with the patient’s head in neutral position, mouth fully open, tongue fully extended, and without phonation.Figure 1: Class zero airway: the epiglottis is seen at the back of the mouth view.Previous difficult laryngoscopy, protruding upper teeth, loose teeth, thyro-mental distance <6 cm, interincisor gap <3 cm, and limited neck extension were also recorded and correlated with airway classes 3 and 4 and laryngoscopy grade III. Laryngoscopy was performed in “sniffing” position with a Macintosh blade, and stylettes were routinely used in the endotracheal tubes. The laryngoscopy grade was assessed by an attending anesthesiologist by using the Cormack and Lehane grading scale (9). Difficult laryngoscopy was defined as grade III or IV laryngoscopy. After 5 min of preoxygenation, anesthesia was induced with fentanyl (1 μg/kg), thiopental (3 mg/kg), and rocuronium (0.6 mg/kg). Patients with a BMI >35 received succinylcholine (1 mg/kg) for endotracheal intubation. Analysis of variance tests were used to determine whether there were any significant differences in age, BMI, airway classes, and laryngoscopy grades. Student’s t-tests were performed to determine whether there were significant differences in age and BMI for pairwise comparison, and χ2 or Fisher’s exact tests were performed to determine whether there were significant differences in sex among classes and grades. Linear regression analysis was performed to determine whether there was a significant trend in age or BMI as airway class and laryngoscopy grade increased. Cochran-Armitage trend tests were used to determine whether male sex was correlated with an increase in airway class and laryngoscopy grade. Logistic regression analysis was performed to determine the independence of each factor in predicting difficult laryngoscopy. Results A total of 764 patients (52% men, aged 44.4 ± 17 yr, BMI 28 ± 8, mean ± sd) were enrolled in the study. Class zero airway occurred in 1.18% of patients. Thirty-five percent of the patients had class 3 or 4 airways, and 10.6% demonstrated a laryngoscopy grade of III. The distribution of age, sex, and BMI among the five classes and three grades is presented in Table 1. Older mean ages were observed in those patients with airway class 4 and laryngoscopy grade III (57 and 53 yr, respectively).Table 1: Distribution of Age, Body Mass Index, and Sex Among Airway Classes and Laryngoscopy GradesTable 2 shows the statistical significance of the differences in age, BMI, and sex between the classes and grades. All patients with class zero airways were women. Laryngoscopy grade III occurred twice as frequently in males as in females. There was a significant correlation between increased class with increased age and BMI. An increase in age but not BMI was associated with high laryngoscopy grade.Table 2: Statistical Significance of the Correlation Between Classes and Grades Versus Age, Body Mass Index, and SexTable 3 depicts the correlation between classes and grades. All patients with class zero airways had a grade I laryngoscopy. Class 1 airway was associated with 10.9% grade II and 3.2% grade III laryngoscopy. There is a stepwise increase in the incidence of laryngoscopy grade III as the airway class changes from 2 to 3 and from 3 to 4. Classes 3 and 4 had a sensitivity, specificity and positive and negative predictive values of 84%, 71%, 97%, and 26%, respectively, for a grade III laryngoscopy view. With stepwise logistic regression analysis, positive associations of grade III laryngoscopy were found with increased age, male sex, protruding upper teeth, loose teeth, and increased airway class (Table 4). The incidence of failed intubation or ventilation, as well as grade IV laryngoscopy view, was zero.Table 3: Distribution of the Four Laryngoscopy Grades Among the Five Airway ClassesTable 4: Stepwise Logistic Regression: Laryngoscopy Grade III Versus Grade I and IIDiscussion In this study the incidence of class zero airway was 1.18%. All patients with class zero airways were women and had grade I laryngoscopy. The fact that all patients with class zero airways were women may be explained by differences in neck fat deposition between the sexes, as demonstrated in a magnetic resonance imaging study by Whittle et al. (10). This may also explain the larger percentage of difficult laryngoscopies found in our male patients. As the airway class increased, so did the laryngoscopy grade (Table 3). The 10- to 30-fold increase in the incidence of grade III from classes 0–2 to classes 3 and 4 is in agreement with the results of Mallampati et al. (11), which showed that of the 15 patients with class 3 (15 of 210 patients, 7.14%) airways, 60% had grade III laryngoscopy, compared with none in class 1. Some studies have shown obesity to be a risk factor for difficult intubation (12,13), yet others (14,15) have found that the incidence of difficult intubation in morbidly obese patients is not more frequent than in normal subjects. In our patients, an increased BMI was not correlated with a high laryngoscopy grade. By using magnetic resonance imaging measurements in obese patients with and without obstructive sleep apnea, Horner et al. (16) found more fatty tissues in areas surrounding the collapsible segments of the pharynx in patients with sleep apnea. This may explain why some obese patients are easy to intubate or ventilate and others are not. Age between 40 and 59 years may carry a risk for difficult intubation (13). In our study, an increase in age was consistent with increase in both airway classes and laryngoscopy grades. Osteoarthritic changes and poor dentition may explain the age-related difficult laryngoscopy. Logistic regression analysis revealed that grade III laryngoscopy had a positive correlation with advanced age, male sex, protruding upper teeth, loose teeth, and increased airway class, but not with BMI, interincisor distance, or thyro-mental distance. The 71% specificity of the Mallampati test demonstrates that 29% of our patients who were not Mallampati 3 and 4 still experienced a difficult laryngoscopy. Combined with the low positive predictive value, this suggests that for a better prediction of difficult intubation, the Mallampati scoring should be combined with other predictors. Class zero airway had an incidence of 1.18% and proved to be an excellent predictor of grade I laryngoscopy. All patients with class zero airways were women. Class 1 airway was not as good as class zero for predicting an easy intubation. An airway class >2 was a good predictor of difficult laryngoscopy (grade III). An increased laryngoscopy grade had a positive correlation with increased age, male sex, protruding upper teeth, loose teeth, and increased airway class, but not with increased BMI. We thank Professor Jacques Chelly and Dr. Sam Lurie for their support in preparing this manuscript.