To the Editor:—We applaud the recent landmark investigation by Langeron et al. 1regarding difficult mask ventilation (DMV). We find three of the established independent DMV risk factors (age > 55 yr, body mass index > 26 kg/m2, and history of snoring) particularly interesting and believe they merit further discussion.The current study by Langeron et al. 1has shown that DMV is correlated significantly with both obstructive sleep apnea (OSA) and difficult intubation (DI). Previous literature 2,3also indicated a significant correlation between DI and OSA. Therefore, although we recognize that DMV, DI, and OSA are complex clinical entities, each with their own multiple contributing anatomic or pathologic factors, DMV, DI, and OSA apparently are correlated with one another. It is then quite reasonable to postulate that they may share a common anatomic abnormality.In 1983, the cephalometric analysis by Riley et al. 4showed that the distance from the mandibular plane to the hyoid (MP-H) tended to be longer in patients with OSA. They suggested the presence of a relatively inferior hyoid. In 1993, we conducted a radiographic study 5of DI patients and found that the rostrocaudal position (in relation to the cervical vertebrae) and vertical distance between the mandible and hyoid (MHD) varied widely among the adult population. However, DI patients tended to have a longer MHD—a greater rostrocaudal separation of the mandible and hyoid—which indicated the presence of a relatively shorter mandibular ramus or more caudally positioned hyoid. We then proposed that the longer MHD and MP-H most likely signified the presence of a similar anatomical abnormality in both the DI and OSA patients. 5The fact that the laryngeal structure is positioned relatively more rostrally in children than in adults is well-known. Schwartz and Keller 6have documented that maturational descent of the epiglottis and larynx (in relation to the cervical vertebrae) occurs in a predictable pattern from infancy into adulthood. Maltais et al. 7have found that although the MP-H is consistently longer in patients with OSA, the length of MP-H also correlates with age: the older adults tend to have a longer MP-H, indicating a more caudally positioned hyoid in the older population.The clinical implication of these observations is that because the tongue muscle is hinged to the hyoid, the base of the tongue also tends to descend caudally with increase in age and to bring down a portion of the tongue from the oral cavity into the hypopharynx. The more caudally the hyoid is positioned, the larger the tongue mass that is collected in the hypopharyngeal space. This large hypopharyngeal tongue then tends to proportionally compromise the posterior airway space and create supraglottic soft tissue obstruction during sleep or anesthesia, thus resulting in various degrees of snoring, OSA, DMV, and DI. The effect of a large hypopharyngeal tongue on airway obstruction can be aggravated when combined with other unfavorable anatomic or pathologic factors, such as a small atlanto-occipital gap, temporomandibular joint problems, obesity, or lingual tonsil.We have studied lateral cervical soft tissue radiographs of normal individuals and heavy snoring, OSA, DMV, and DI patients, which illustrate the effect of a relatively longer MHD and its associated large hypopharyngeal tongue. Here, because of space limitations, we will show only one patient (fig. 1). Interested readers are welcome to contact us privately to share more of our radiographs and for details about our radiographic method.The theoretical explanation and clinical evidence regarding the large hypopharyngeal tongue in OSA, DMV, and DI are in accordance with the findings by Langeron et al. 1that older, obese, or snoring individuals tend to have DMV and DI. However, a relatively longer MHD, with its associated large hypopharyngeal tongue, usually is not visible as an obvious external physical feature, 5such as a receding jaw, bucked teeth, or a long maxilla. Thus far, practitioners have not paid much attention to this anatomic characteristic. Langeron et al. 1also did not consider it a possible DMV risk factor in their study. Perhaps the unawareness of this important anatomic abnormality may account for the low prediction rate of DMV and DI. We have found that because the mandible and hyoid are identifiable anatomic landmarks, with experience, patients with a relatively longer MHD can be identified in a routine physical examination without cervical radiography. We believe that the role of a large hypopharyngeal tongue in the understanding and management of difficult airway should be duly assessed.