Tongue size matters: revisiting the Mallampati classification system in patients with obstructive sleep apnea

OBJECTIVE: The Mallampati classification system has been used to predict obstructive sleep apnea (OSA). Upper airway soft tissue structures are prone to fat deposition, and the tongue is the largest of these structures. Given that a higher Mallampati score is associated with a crowded oropharynx, we hypothesized that the Mallampati score is associated with tongue volume and an imbalance between tongue and mandible volumes. METHODS: Adult males underwent clinical evaluation, polysomnography, and upper airway CT scans. Tongue and mandible volumes were calculated and compared by Mallampati class. RESULTS: Eighty patients were included (mean age, 46.8 years). On average, the study participants were overweight (BMI, 29.3 ± 4.0 kg/m(2)) and had moderate OSA (an apnea-hypopnea index of 26.2 ± 26.7 events/h). Mallampati class IV patients were older than Mallampati class II patients (53 ± 9 years vs. 40 ± 12 years; p < 0.01), had a larger neck circumference (43 ± 3 cm vs. 40 ± 3 cm; p < 0.05), had more severe OSA (51 ± 27 events/h vs. 24 ± 23 events/h; p < 0.01), and had a larger tongue volume (152 ± 19 cm(3) vs. 135 ± 18 cm(3); p < 0.01). Mallampati class IV patients also had a larger tongue volume than did Mallampati class III patients (152 ± 19 cm(3) vs. 135 ± 13 cm(3); p < 0.05), as well as having a higher tongue to mandible volume ratio (2.5 ± 0.5 cm(3) vs. 2.1 ± 0.4 cm(3); p < 0.05). The Mallampati score was associated with the apnea-hypopnea index (r = 0.431, p < 0.001), BMI (r = 0.405, p < 0.001), neck and waist circumference (r = 0.393, p < 0.001), tongue volume (r = 0.283, p < 0.001), and tongue/mandible volume (r = 0.280, p = 0.012). CONCLUSIONS: The Mallampati score appears to be influenced by obesity, tongue enlargement, and upper airway crowding.