INVITED EDITORIAL
Invited Editorial on "Neuromechanical interaction in human snoring and upper airway obstruction"

Anatech, Kennett Square, Pennsylvania 19348

	ARTICLE

TOP ARTICLE REFERENCES

IN THIS ISSUE of the Journal, Huang and Ffowcs Williams (3) report the development of a mathematical model that couples flow mechanics of the upper airway with neurophysiological activity. The model depends on the mechanical principle that luminal collapse occurs when passive wall stiffness is insufficient to withstand falling intraluminal pressure and on the physiological principle that proprioception in the collapsing wall triggers reflex muscle activity in "dilator" muscles that increases the "neural" stiffness of the wall and prevents collapse from continuing. The model shows that, in this situation, reflex latency is a critical factor because, if the latency is larger than an optimal value, part of the neuromuscular force is transformed into a negative damping effect that destabilizes the luminal wall and results in "flutter."

Mathematical modeling is not a new approach to understanding the function of the airway; Isono and colleagues (4, 5) showed that the mechanics of flow through an airway with passive, compliant walls can result directly in luminal collapse; the critical factors are the morphology of the lumen and the rate of flow. As it is generally understood that during rapid-eye-movement (REM) sleep the active components of wall compliance are negated as a result of muscle inactivity, the mechanical conditions of these models are closely approximated during deep sleep. Isono et al. (5) subsequently showed that the morphologies of airways rendered passive in vivo do, indeed, differ significantly between apneic and normal subjects.

In the awake subject, when luminal volume and axial bending, and combined passive and active wall stiffness, are compared, the morphological values for the airway in apneic subjects with low apnea/hypopnea indexes overlap considerably those of subjects with normal airways (unpublished observations). This shows that a functional lumen is maintained during wakefulness and in the pre-REM phases of sleep through the agency of neurological activity that increases the active component of wall compliance, i.e., via dilator muscle activity. Presumably, with the onset of REM sleep and cessation of muscle reflex activity, the overlapping morphologies would be altered sufficiently to become distinct groups. Horner et al. (2) showed that the genioglossus muscle is active during the early phases of sleep and is more active in apneic than in nonapneic airways. This implies that greater neuromuscular effort is required to maintain functional viability in airways that are prone to collapse than in airways that are not prone to collapse. In terms of the Huang and Ffowcs Williams model, the struggle to maintain airway viability can be seen as a progressive destabilization of the lumen wall caused by dilator muscle reflexes unable to match the rapid onset of collapse. However, a neuromechanical coupling model also presents the possibility that the problem may be inherent tardiness in the reflex itself, i.e., within the integrative pathways in the brain stem. These alternatives, and the significance of neuromechanical coupling, can be understood in the context of phylogenetic modifications of the human airway.

The high incidence of snoring and obstructive sleep apnea (3) suggests that the human airway is not optimally adapted to its prime function. However, the airway is a special case and has been extensively modified by adaptive forces not related to its original and prime functions. It is generally accepted that these adaptations relate to a whole body change from a pronograde to a unique orthograde posture. They included extensive bony remodeling of the skull. The facial skeleton rotated beneath the rostral portion of the cranium while, simultaneously, the foramen magnum "advanced" rostrally, constricting the infratemporal fossa (1). The gently curved central axis of the quadrupedal airway was bent and narrowed at the nasopharyngeal junction. The hyoid bone, larynx, and epiglottis were displaced inferiorly, and the soft palate and epiglottis, which through their juxtaposition previously had formed a relatively stiff anterior wall of the airway, were now disengaged. The missing part of the anterior wall was replaced by the far more compliant tissue of the tongue. As the tongue was also "bent" and thrust backward by the general facial remodeling, it may be supposed that this enhanced any inherent tendency it may have had to "slump" into the newly opened airway space.

Shome et al. (6) used airway models based on three-dimensional data obtained in vivo via medical imaging to show that normal flow around the bent airway axis is critically close to becoming turbulent at all times, but the onset of turbulence can be mitigated by repositioning or reshaping the soft palate and dorsal surface of the tongue. This would entail "new" roles for both of these structures. However, the original role of the genioglossus muscle was to protrude the tongue. It did not fall back into the airway because of the complete anterior wall formed by the interdigitation of the larynx and epiglottis. The original function of the soft palate was to close the internal nares during deglutition (although it could be argued that the palatoglossus muscle helped to maintain the palatoepiglottic contact and, thus, indirectly, also maintained the nasopharyngeal lumen). To carry out their new roles, the neural control of both structures had to become more sophisticated. This might well include redefinition (shortening) of the original latency periods required for optimal reflex function. However, the plasticity of the peripheral structures may not have been matched by that of the central structures. Integrative pathways for reflexes that control vital functions are likely to be defined genetically and stored prenatally. As the peripheral structures do not assume their adult morphological relationships for several years postnatally, coinciding with the ability to stand and walk (1), it is likely that a certain amount of postnatal "secondary fine-tuning," through learned behavior, takes place within the integrative pathways. It can be hypothesized that this fine-tuning process cannot match the morphological extremes eventually presented by some airways, or that (in some cases) individual variation in this fine-tuning results in reflex latencies that do not match the altered peripheral morphology they serve, even though it is not excessive. In either case, the end result would be the same hyperactivity in the dilator muscles and progressive destabilization of the luminal wall.

If the model created by Huang and Ffowcs Williams (3) could be developed to the stage where critical values for reflex latencies could be calculated for individual airways, a comparison between the calculated, optimal latencies and actual latencies might represent a means of investigating neural control over the active airway. Considering the complexity of the peripheral structures and the extreme difficulty of measuring some of the parameters involved, this is probably not feasible. Huang and Ffowcs Williams acknowledge that their present contribution is only a very small step in the desired direction, but this does not diminish its value in stimulating further discussion.

	ACKNOWLEDGEMENTS

This work was supported by National Institute of Neurological Disorders and Stroke Small Business Innovation Research Grant Program R43NS34094 and Small Business Technology Transfer Research Program R41NS/HL-34585.

	REFERENCES

TOP ARTICLE REFERENCES

1.   Crelin, E. S. The Human Vocal Tract: Anatomy, Function, Development and Evolution. New York: Vantage, 1987.

2.   Horner, R. L., J. A. Innes, M. J. Morrel, S. A. Shea, and A. Gus. The effect of sleep on reflex genioglossus muscle activation by stimuli of negative airway pressure in humans. J. Physiol. (Lond.) 476: 141-151, 1994[Abstract/Free Full Text].

3.   Huang, L., and J. E. Ffowcs Williams. Neuromechanical interaction in human snoring and upper airway obstruction. J. Appl. Physiol. 86: 1759-1763, 1999[Abstract/Free Full Text].

4.   Isono, S., and J. E. Remmers. Anatomy and physiology of upper airway obstruction. In: Principles and Practice of Sleep Medicine, edited by M. H. Kryger, T. Roth, and W. C. Dement. Philadelphia, PA: Saunders, 1993, p. 642-655.

5.   Isono, S., R. F. Thom, E. A. Hajduk, D. L. Morrison, S. H. Launois, F. G. Issa, W. A. Whitelaw, and J. E. Remmers. Anatomy of the pharyngeal airway in sleep apneics: separating anatomic factors from neuromuscular factors. Sleep 16: 880-884, 1993.

6.   Shome, B., L.-P. Wang, A. K. Prasad, M. H. Santare, A. Z. Szeri, and D. Roberts. Modeling of airflow in the nasopharynx with applications to sleep apnea. J. Biomech. Eng. 120: 416-422, 1998[Medline].