An Animal Model for Airway Sensory Deprivation Producing Obstructive Apnea with Postmortem Findings of Sudden Infant Death Syndrome¹ 
 
 

  ABSTRACT. A series of experiments was performed in rabbits to investigate the effects of airway sensory stimuli on upper airway patency. Pharyngeal airway closure was observed in rabbits breathing through a tracheostomy tube; pharyngeal patency was rapidly restored either by closing the tracheostomy tube, which forced the animals to resume nasal breathing, or by creating cyclical pressure changes in the nose and pharynx to stimulate respiratory tidal airflow. This airway opening effect of pressure fluctuations was eliminated by topical anesthesia of the airway mucosa, an observation suggesting that sensory stimulation from pressure change is needed for airway patency. The observation that dead animals have a patent pharyngeal airway that is resistant to collapse from negative intra-luminal pressure, whereas animals breathing via a tracheostomy have a readily collapsible airway that is closed at zero transmural pressure, suggests that airway-constricting muscles close the airway when the animals breathe via the tracheostomy. Loss of electromyographic activity from airway-dilating muscles (genioglossus) was observed during tracheostomal breathing and was restored by cyclical pressure changes applied to the upper airway lumen, an observation further supporting the concept that airway reflexes responding to pressure regulate the activity of airway-dilating and airway constricting muscles. Topical anesthesia of the upper airway mucous membrane, which eliminated these responses to pressure, was associated with an obstructed pharyngeal airway and death from apparent asphyxia in either pentobarbital-anesthetized adult animals or young animals without general anesthetic. Death resulting from airway obstruction in this manner was associated with postmortem findings of sudden infant death syndrome (pulmonary edema and pleural petechiae) in the majority of animals.  
 

  Upper airway closure (obstructive apnea) is a postulated cause of death in sudden infant death syndrome (SIDS)^1,2 Obstructive apnea is known to occur in a variety of other medical conditions, but in none of these is the mechanism of the obstruction well defined. Recent investigations suggest that upper airway patency depends on the coordination of airway-dilating and airway-constricting muscles that presumably have an agonist-antagonist relationship.^3-5 Obstructive apnea during sleep may result from increased relaxation of airway-dilating muscles, leading to passive closure of the pharyngeal airway.^4,6,7 In other conditions, eg, rabies,^8,9 tetanus,^10,11 and hypocalcemic tetany¹², airway obstruction is believed to result from spasm of airway-constricting muscles. The pathophysiology of spasmodic airway closure in these conditions is obscure. Recently, attention has been given to airway protective reflexes. Noxious chemical1³ and mechanical¹⁴ stimuli have been shown to cause laryngo-spasm. In contrast, the observations of Bosma and co-workers ^15,16 have led them to speculate that mechanical stimuli arising from respiratory airflow in the nose and pharynx facilitate airway-maintaining mechanisms and may actually be required for airway patency. The purpose of the experiments described here was to examine ways in which upper airway stimuli might contribute to airway patency. We used the rabbit as a model. In a series of experiments we studied how the effects of stimuli from tidal airflow and from cyclical pressure changes, without airflow, affect airway patency. We examined the response of airway-constricting and airway-dilating muscles to these stimuli. Additionally, we studied the postmortem findings resulting when compromise of airway sensation produces fatal airway closure.

METHODS

  We studied mature New Zealand rabbits that were anesthetized with pentobarbital (20 to 30 mg/ kg). We devised a system utilizing a tight-fitting face mask and a tracheal T-tube in tracheostomized animals, enabling us to alternate between nasal and tracheostomal breathing. Animals were studied in the supine position with the head firmly secured at an angle 900 to the spine. We provided an airway in the nose, inserted to a depth of 1 cm from the nares and secured with a mask made of quick-drying epoxy glue which covered the mouth and nose (Fig 1). We divided the trachea approximately 1 cm below the cricoid cartilage and inserted a T-tube airway. By clamping the nasal airway and the proximal end of the tracheal T -tube, we could made a closed system of the nose, pharynx, and larynx. In this system, airway patency could be tested either by direct observation or indirectly by pressure transmission. We inserted

  Fig 1. Schematic representation of experimental rabbit preparation fitted with face mask and nasal and tracheal airways. Preparation allows diversion of pharyngeal airflow via tracheostomy and/or isolation of upper airway segment. A, Diversion of respiratory airflow via tracheostomy is associated with pharyngeal airway closure. B, Using syringe to produce cyclic pressure changes in isolated segment, mimicking respiratory airflow, is associated with sustained pharyngeal patency. Pressure transducers (PH) were used to detect airway patency or closure. 
 

an endoscope through the proximal end of the trachea facing the inferior surface of the vocal cords. Airway patency was also tested by transmission of low amplitude (less than 0.5 cm H2O) pressure signals. Failure of the pressure signals to transmit from one end of the closed system to the other was interpreted to mean airway closure. We recorded airway pressure, diaphragm, and geniglossus direct and integrated electromyograms (EMGs) polygraphically. In some animals, nasal respiratory airflow was also recorded.

  In order to assess activity of airway-constricting muscles, we determined the transmural airway pressure required to collapse the upper airway.^3,17 Briefly, the airway-closing pressure was determined by making a closed system of the upper airway as depicted in Fig 1, B. The intraluminal pressure in this closed system was then increased until the airway was patent and then slowly lowered until the airway closed. This was done in both tracheostomal breathing and in freshly killed animals.

  The effect of topical anesthesia applied to the upper airway mucous membrane was observed in adult animals that were under general anesthesia, surgically prepared as described above, and also in a group of young, unanesthetized animals varying in age from 3 to 6 weeks. In this group of young animals no surgical procedures or artificial airways were used. These young animals were lightly restrained in the supine position and respiratory movements were monitored by a thoracic respiration belt. 
 

RESULTS AND DISCUSSION

Respiratory Tidal Airflow Diversion

  The relation of airway patency and airway muscle activity to respiratory airflow was assessed in 17 animals. We found that shifting from nasal breathing to tracheostomal breathing had the effect of producing upper airway closure in two to 20 seconds (Fig 1, A). The airway opened in one to three seconds when the animal was returned to nasal breathing. Consistent with previous observations,18,19 during these maneuvers we noted that genioglossus muscle (airway dilator) activity was immediately increased by nasal breathing and was reduced on shifting to tracheostomal breathing, suggesting that airflow or air pressure is a stimulus to this muscle. During tracheostomal breathing, the site of airway obstruction was observed in the pharynx 0.5 to 1.5 cm above the vocal cords.

Airway-Closing Pressures

  In tracheostomal breathing animals, under light or moderate anesthesia, the airway closed at pressures above atmospheric (closing pressure = +1.1 + 1.9 (SD) cm H2O). In several such animals in which the depth of anesthesia was gradually increased by giving incremental doses of pentobarbital, we noted a progressive decrease in airway-closing pressure. In fact, in deeply anesthetized animals, the airway remained patent during tracheostomal breathing. Similarly, in dead animals the airway remained patent unless exposed to negative transmural pressures (closing pressure = -4.9 + 1.4 cm H2O). In all cases, the site of closure was observed to be in the pharynx. Thus, the pharyngeal airway was more resistant to collapse by negative pressure in dead or deeply anesthetized animals than it was in lightly anesthetized animals. The requirement for an airway-dilating pressure (ie, positive closing pressure) to maintain pharyngeal patency in lightly anesthetized, tracheostomal breathing animals indicates a pharyngeal-constricting force in this state. We attribute this force to pharyngeal-constricting muscles, inasmuch as death and deep anesthesia, both of which produce muscle relaxation, reopen the airway. The rapidity and reversibility of the closure when the animals were changed from tracheostomal to nasal breathing rule out other potential causes of airway obstruction such as tissue edema.

Influence of Cyclical Pressure Changes

  In order to determine whether lack of sensation of respiratory airflow was the cause of the pharyngeal closure in the tracheostomal breathing animals, we simulated respiratory airflow in the isolated upper airway system by creating cyclic pressure changes applied in a manner so as to approximate the animal's breathing frequency and transpharyngeal pressure amplitudes encountered in breathing (-1.5 cm H2O during inspiration, + 1.5 cm H2O during expiration). We observed that cyclic pressure opened the airway in nine of nine animals tested. The airway remained open so long as the maneuver continued (Fig 1, B). The negative pressure that was applied (-1.5 cm H2O) was less than the mechanical closing pressure of the airway (+ 1.1 cm H2O) and yet the airway remained patent; this observation supports the concept that sensory stimuli from airflow, rather than mechanical effects, were responsible for airway patency. In these experiments it was clear that the cyclic pressure oscillations elicited muscle tone-regulating reflexes, inasmuch as phasic, inspiratory genioglossus muscle (airway dilator) activity promptly increased in response to air pressure oscillations and returned to base line with removal of the stimulus.

Effects of Topical Airway Anesthesia

  To further test the hypothesis that the airwayopening effect of cyclic pressure changes was due to sensory stimuli and not to mechanical  
 
 

  Fig 2. Polygraphic tracings of nasal airflow (upper trace), tracheal pressure (middle trace), and genioglossus (GG) muscle electromyogram (EMG) (lower trace) in an adult rabbit treated with Cetacaine topical anesthetic applied to the nose, pharynx, and larynx. A, Nasal breathing before topical anesthesia is shown. B, Tracheostomy tube is closed (arrow 1). Note absent airflow at nose. Progressively increased tracheal pressure and genioglossus activity occur as asphyxia becomes progressively severe due to airway obstruction. C, Continued airway obstruction and active expiratory efforts are indicated by positive tracheal pressure. The animal's "last gasp" occurs at arrow 2. Note that positive pressure swings during expiratory phase of respiratory cycle produce no nasal airflow, indicating a tightly closed upper airway (C).

forces, in 12 adult animals we applied topical anesthetic spray (Cetacaine) to the mucous membrane of the nose, pharynx, and larynx. The characteristic stimulation of genioglossus activity by airway pressure oscillations was gradually abolished after application of anesthetic spray over a two- to four-minute time course. During this same period sphincteric closure of the pharynx gradually appeared. The disappearance of genioglossus muscle responses to air pressure coinciding with appearance of sphincter-like activity suggests an opposite effect of the airway pressure reflexes on airway-dilating (agonist) and airway-constricting (antagonist) musculature. After administration of topical anesthesia, cyclic pressure changes failed to open the airway of tracheostomal breathing animals. Such airway constriction following topical anesthesia was documented by closing pressure determinations (+5.6 + 1.5 cm H2O) that were significantly higher than preanesthetic control values (P < .001). Direct observation of the upper airways showed complete closure in the pharynx but also considerable narrowing of the glottic aperture, suggesting a generalized upper airway-constricting effect.

  When we shifted the topically anesthetized animals from tracheostomal to nasal breathing, the animals were unable to open their airways and severe asphyxia rapidly occurred. Twelve of 12 animals rapidly died of asphyxia when occlusion of the tracheal T-tube was maintained, even though marked activity of genioglossus muscle occurred in response to severe asphyxia. There was absent nasal airflow even with high airway pressures during the expiratory phase, as is shown in Fig 2. In line with evidence from the preceding experiments, we presume that this airway constriction is produced by the action of airway-constricting muscles. Assuming this to be the case, it is of interest that such activity of an airway-dilating muscle (genioglossus) was counteracted by airway-constricting muscle activity in such fashion that airway closure was sustained until death. Similar pathologic cocontraction of agonist and antagonist muscles occurs in strychnine poisoning, tetanus, and related disorders²° and has been observed in electromyographic recordings of intrinsic laryngeal muscles during experimentally induced laryngospasm.¹⁴

  To rule out contributing effects of general anesthesia in this airway closure phenomenon, we studied nine young unanesthetized animals. In three animals, 0.5 to 1 ml of 2% or 4% lidocaine was instilled into the nasopharynx; in the remaining six animals, Cetacaine was similarly administered. Death resulted in all animals. The mean time from instillation of the drug to the last gasp was 315 seconds (range 180 to 527 seconds) in Cetacaine-treated animals and 746 seconds (range 300 to 1,220 seconds) in lidocaine-treated animals. Although the initial response to application of the anesthetic was diaphragmatic apnea, we observed struggling limb movements and vigorous terminal respiratory activity associated with deep cyanosis similar to that of the pentobarbital-anesthetized adult animals. The time to last gasp in the young unanesthetized animals with the shortest survival times was similar to that of animals killed by sustained upper airway occlusion.²¹ Therefore, we speculate that several factors produced asphyxia. Initially, central apnea occurred, probably as a result of chemoreceptor reflexes.^22,23 Later on, we speculate that terminal asphyxia was caused by airway obstruction resulting from topical anesthesia. In contrast, nasopharyngeal injections of water (six animals) or normal saline (four animals) interrupted respiration briefly and rapid recovery occurred in all but one case.

  In these animals treated with topical anesthetic, we believe that death can be attributed to the local effects of the anesthetic rather than central toxicity, because in several additional young animals, gradually increasing the blood concentration of topical anesthetic (lidocaine) by giving intravenous infusions was associated with a prolonged period of seizures and hyperreflexia prior to death. These symptoms of central toxicity were absent in the animals topically anesthetized with lidocaine or Cetacaine. The blood drug level at the time of death in the animals treated with intravenous anesthetic was substantially greater than that in topically anesthetized animals.

Postmortem Findings

  Postmortem examinations of the upper respiratory tract undertaken in 12 animals dying from obstructive apnea (three with general anesthesia, nine without), were noteworthy with respect to the presence of pulmonary edema foam in the trachea and lung pleural hemorrhages in eight of 12 animals. We observed ecchymoses and petechial hemorrhages scattered over the surface of both lungs, especially the medial, posterior, and inferior surfaces (Fig 3). Hemorrhages were absent in the abdominal viscera. These findings are noteworthy because of the similarity to those seen in SIDS.^24.25 Such findings observed at autopsy in SIDS have long been viewed by many investigators as evidence of acute airway obstruction.^2.26 However, attempts to reproduce these postmortem findings by experimental obstructive apnea in animals (rats) have failed.^27.28 The presence of petechiae in the majority of the animals dying with airway obstruction suggests that the rabbit may be a useful model for study of the intrathoracic pathologic findings that characterize SIDS.

Speculation

  The present data suggest that the upper airway air pressure reflexes playa critical role in regulating the tone of agonist and antagonist muscles that maintain the pharyngeal airway. Elimination of these reflexes leads to an imbalance characterized by increased tone in airway-constricting muscles. There are certain similarities between obstructive apnea in this animal model and obstructive apnea as it occurs in human disease. For instance, tetanus toxin depresses central, inhibitory, spinal path  ways²⁰ and produces obstructive apnea.^10,11 The present agents (topical anesthetics) produce a similar picture of airway muscular cocontraction and airway obstruction through effects on peripheral sensory mechanisms. In each case neural pathways that modulate muscular tone  
 
 

  Fig 3. Posterior surface of lungs in a rabbit dying from asphyxia following application of topical anesthesia to nasopharyngeal airway. Many ecchymotic and petechial hemorrhages (arrows) can be seen. 
 

Are suppressed. Additionally, sudden death from respiratory obstruction can occur when infections such as infectious laryngitis (croup syndrome) or pertussis produce airway inflammation. Although mechanical airway obstruction due to mucosal edema may occur in these diseases, often airway obstruction is clinically attributed to muscular spasm.^29,30 As upper respiratory tract infection could conceivably depress mucosal sensation, it is possible that reflex phenomena similar to those described here are involved in this kind of airway obstruction. Nonspecific upper airway mucosal inflammation is often seen in SIDS.^24,25 In fact, it has often been hypothesized that such inflammation may start a chain of events in which altered mucosal sensation interferes with respiratory muscle reflexes, ultimately causing death from asphyxia.^26,31,32 Assuming that mucosal inflammation could interfere with pressure sensation, the present observations provide a model for this concept. 
 

ACKNOWLEDGMENTS 
 

  This work was supported by National Institutes of Health grant HD-1O993-04 and a training fellowship from the American Lung Association awarded to Dr Mathew. We thank W, E. Dodson for performing drug concentration determinations. Also, we wish to acknowledge P. R. Dodge, L. Hillman, J. Holowach Thurston, M, Mauer, R. E. Marshall, C. Rovainen, W. T, Thach, Jr, and S. Wolf for their critical reviews of the manuscript. 
 

REFERENCES 
 

1. Guilleminault C, Peraita R, Souquet M, et al: Apneas during sleep in infants: Possible relationship with sudden infant death syndrome. Science 190:677, 1975.
2. Beckwith JB: The sudden infant death syndrome. Curr Probl Pediatr 3:1, 1973.
3. Brouillette RT, Thach BT: A neuromuscular mechanism maintaining extrathoracic airway patency. J Appl Physiol 46:772, 1979.
4. Remmers JE, DeGroot WT, Sauerland EK, et al: Pathogenesis of upper airway occlusion during sleep. J Appl Physiol 44:931, 1978.
5. Hairston LE, Sauerland EK, Orr WC: The role of oropha-ryngeal muscles in respiration: An electromyographic study during awakefulness and sleep. Anat Rec 190:411, 1978.
6. Thach BT, Stark AR: Spontaneous neck flexion and airway obstruction during apneic spells in pre term infants. J Pediatr 94:275, 1979.
7. Strohl KP, Saunders NA, Feldman NT, et al: Obstructive sleep apnea in family members. N. Engl J Med 299:969, 1978.
8. Nelson WE (ed): Textbook of Pediatrics, ed 18. Philadelphia, WB Saunders Co, 1964, p 565.
9. Koprowski H: Viral diseases that may involve the central nervous system, in Beeson PB, McDermott WM (eds): Cecil-Loeb Textbook of Medicine. Philadelphia, WB Saunders Co, 1967, p 52.
10. Rothstein RJ, Baker FJ: Tetanus: Prevention and treatment. JAMA 240:675, 1978.
11. Laforce FM, Young LS, Bennett JV: Tetanus in the United States (1965-1966): Epidemiologic and clinical features. N Engl J Med 280:569, 1969.
12. Harrison HE: in Barnett HL, Einhorn AH, (eds): Pediatrics New York, Appleton-Century-Crofts, 1968, p 387.
13. Rex MAE: The production of laryngospasm in the cat by volatile anesthetic agents. Br J Anesth 42:941, 1970.
14. Rex MAE: Laryngospasm and respiratory changes in the cat produced by mechanical stimulation of the pharynx and respiratory tract: Problems of intubation in the cat. Br J Anesth 43:54, 1971.
15. Tagaki Y, Irwin JV, Bosma JF: Effect of electrical stimulation of the pharyngeal wall on respiratory action. J Appl Physiol 21:454, 1966.
16. Bosma JF: Introduction to the symposium, in Bosma JF, Showacre J (eds): Development of Upper Respiratory Anatomy and Function: Implications for Sudden Infant Death Syndrome. US Government Printing Office, 1975, p 18.
17. Wilson SL, Thach BT, Brouillette RT, et al: Upper airway patency in the human infant: Influence of airway pressure and posture. J Appl Physiol 48:500, 1980.
18. Mathew OP, Abu-Osba YK, Thach BT: Response of airway maintaining respiratory muscles (genioglossus) to upper airway pressure changes. Fed Proc 39:1063, 1980.
19. Mathew OP, Roberts JL, Abu-Osba YK, et al: Tidal airflow through the upper airway augments respiratory activity of genioglossus and posterior pharyngeal wall muscles. Fed Proc 40:622, 1981.
20. Esplin DW, Zablocka B: Central nervous system stimulants, in Goodman LS, Gilman A (eds): The Pharmacological Basis of Therapeutics, ed 5. New York, Macmillan Publishing Co, Inc, 1965, p 345.
21. Thach BT, Lawson EE: Death from upper airway obstruction: Decreased survival in 33-day-old-rabbits, abstracted. Pediatr Res 13:542, 1979.
22. Downing SE, Lee JC: Laryngeal chemosensitivity: A possible mechanism for sudden infant death. Pediatrics 55:640, 1975.
23. Bynum TE, Ruoff P A, Richert J, et al: The smoke reflex in rabbits, an example of autonomic control of cardiorespiratory function. Cond Reflex 5:233, 1970.
24. Valdes-Dapena MA: Sudden and unexpected death in infancy: A review of the world literature. Pediatrics 39:123,1967.
25. Adelson L: Specific studies of infant victims of sudden death, in Wedgwood RJ, Benditt EP (eds): Sudden Death in Infants, US Government Printing Office, 1965, pp 14-23.
26. Handforth CP: Sudden unexpected death in infants. Can Med Assoc J 80:872, 1959.
27. Guntheroth WC, Kawabori I, Breazeale DG, et al: The role of respiratory infection in intrathoracic petechiae. Am J Dis Child 134:365, 1980.
28. Guntheroth WC, Breazeale D, McCouth CA: The significance of pulmonary petechiae in crib death. Pediatrics 52:601, 1973.
29. McCracken GH, Eichenwald HF: Acute infections of the larynx and trachea, in Vaughan VC, McKay RJ (eds): Nelson Textbook of Pediatrics, ed 10. Philadelphia, WB Saunders Co, 1975, p 961.
30. Peiper A: Cerebral Function in Infancy and Childhood. New York, Consultants Bureau, 1963, p 377.
31. Steinschneider A: Prolonged apnea and the sudden infant death syndrome: Clinical and laboratory observations. Pediatrics 50:646, 1972.
32. Steinschneider A: Nasopharyngitis and prolonged sleep apnea. Pediatrics 56:967, 1975.