Perioperative management of the patient with laryngeal stenosis presents serious challenges. A jet cannula placed in the trachea by the translaryngeal or percutaneous route gives an effective tidal volume (VT) even without entrainment [1,2], but this is dangerous in critical laryngeal stenosis because outflow from the lungs may be inadequate. Positive end-expiratory pressure that increases with each breath can then rapidly occur [3,4]. Hypotension and barotrauma to the lungs will result unless jetting is stopped and an outflow pathway is immediately established [5,6]. The safest method of ventilation in patients with critical laryngeal stenosis is by tracheotomy performed under local anesthesia [3,7]. Sedation may precipitate respiratory arrest in these patients .
We constructed a total laryngeal bypass device (TLBD)  to allow safe and effective jet ventilation in critical laryngeal stenosis, thereby eliminating the need for tracheotomy. The aim of this study was to test its ability to ventilate an artificial lung with a totally occluded proximal airway and, at the same time, provide acceptable volume and airway pressure measurements.
The TLBD consists of a 16- or 14-gauge stainless steel cannula within another similar cannula 3 mm in diameter (Figure 1). Jet ventilation is provided through the inner cannula and suction is provided through the outer cannula. The distal end of the inner cannula has a compliant wire basket welded externally. The compliant wire allows the basket to resume its shape on emerging from the distal end of the outer cannula. The basket keeps the suction tip from abutting the tracheal wall and allows better direction of gas flow. The terminal 1.5 cm of the outer cannula has multiple perforations to achieve safer multiorifice jetting .
Regulated suction is applied to the outer cannula throughout both inspiration and expiration. During inspiration, jet pressure is regulated to provide a flow into the lungs that is greater than the suction outflow. The resultant tidal volume (VT) is the integrated instantaneous "in-out" flow difference over time. Flow during expiration depends on suction strength and available area within the outer cannula.
Airway pressures were measured through the inner cannula between the end of one inspiration and the beginning of the next. A Wright respirometer placed in the suction line allowed an estimation of the tidal volume by measuring the volume suctioned during the expiratory period.
The experiments were performed using a mechanical lung (Model VT-1B; MI Instruments Inc., Grand Rapids, MI) that featured a 2.2-L bellows with adjustable compliance and resistance (Figure 2). Interchangeable resistors with parabolic flow pressure responses RP10, RP20, and RP50 offering 8, 22, and 90 cm H2 O [center dot] L-1 [center dot] s-1 resistance, respectively, were used for this experiment. VT readings were standard volume measurements based on displacement of the top plate of the bellows and verified for accuracy using calibrated ventilators.
A jet ventilating system capable of delivering oxygen at pressures up to 50 lb/in2 (PSI), incorporating an electrical valve and timing device, was assembled (Figure 2). The timing device allowed regulation of inspiratory and expiratory duration through a valve on the jet hose. The distal end of the jet hose was securely attached to the inner cannula by a three-way stopcock. Suction was attached to the outer cannula. A regulator allowed suction adjustment. A Wright respirometer was placed in the suction line.
The TLBD was advanced through the customized CO2 monitoring port of a right-angled elbow adapter (Figure 2p). One end of a straight adapter (Figure 2s) was attached to the elbow adapter, and the other end was sealed (Figure 2m). The free end of the elbow adapter (Figure 2e) was attached to the "trachea" (Figure 2d) so that the TLBD was centered within the airway. A resistor (Figure 2t) was then inserted.
Pressure transducers were connected to the preparation at three locations: the side port of the stopcock on the jet cannula (Figure 2h); the side-port on the straight adapter to allow pressure measurement in the trachea proximal to the jet insertion (Figure 2g); and a port on the distal trachea just proximal to the parabolic resistor for tracheal pressure measurements (Figure 2f). The transducers were then connected to a monitor and printer that allowed observation of three synchronous pressure traces.
All connections were closed before activating the system except that to the trachea, which was left open at the sliding adapter (Figure 2n) used to insert a resistor. Inspiratory and expiratory times were chosen, and the jetting system was activated. The jet pressure was adjusted to a desired level. The suction was activated, and the trachea was closed at the beginning of a jet inspiration. The suction strength was regulated to allow the top plate of the bellows to fall, during expiration, to a point just barely above the support plate (Figure 2b) on which it rests when the lung is open to the atmosphere and observing each tracheal pressure trace approach zero. This procedure was repeated to allow recording of VT, suction and airway pressures at preset inspiratory and expiratory times, jet pressure, airway resistance, and lung compliance.
(Figure 3) depicts representative samples of the range of VT generated at safe airway pressures using different jet and suction pressures and various compliance and resistance combinations. VT was increased by greater jet pressure, longer inspiratory time, or lower suction pressure. Suction was applied during both inspiration and expiration. Therefore, increasing suction to shorten expiratory time decreased the VT. This was apparent when observing the smaller VT values generated at inspiratory and expiratory times of 1.5 and 3.5 s, compared with those at 1.5 and 6 s at any particular driving pressure (Figure 3 and Figure 4). Compliance and airway resistance changes had almost no effect on the VT generated (Figure 3). The jet ventilator therefore acted as a constant flow generator in the closed lung.
Compliance and resistance changes determined both the magnitude and shape of the pressure curves generated in the airway (Figure 3 and Figure 5). The expiratory pressure curve did not return to baseline precipitously, but did so gradually at a rate dependent on the suction strength and area available within the outer cannula.
The only airway pressure measurement available when using the TLBD to ventilate a patient is obtained through the side port of the jet cannula. Therefore, it was important to compare this measurement with those directly measured in the trachea. At resistance values up to RP20, the peak airway pressure (PAP) measured through the jet closely approximated the PAP measured independently in the distal trachea (Figure 3 and Figure 5). As the compliance decreased, this close relationship remained apparent. At RP20 settings, the maximal divergence between the two PAPs was 6 mm Hg at the highest driving pressures. At the RP50 resistance, a larger divergence appeared.
Both airway pressure measurements during expiration were otherwise identical and returned to baseline in parallel (Figure 5). The pressure trace measured from the jet cannula therefore accurately represents the true airway pressure at the time and allows optimal timing of the next jet inspiration as the airway pressure reaches baseline. The accuracy of this pressure measurement through the cannula greatly enhances the safety of TLBD use.
The Wright respirometer significantly underread the VT at suction pressures <15 mm Hg, gave clinically acceptable accuracy between 25 and 55 mm Hg, and significantly overread VT at suction pressures >55 mm Hg (Figure 6). Higher VT values were more accurately represented when the smaller (16-gauge) jet catheter, lower lung compliance, and longer expiratory times diminished the suction force requirement (Figure 3 and Figure 6).
The safe use of the TLBD depends on the reliability of the PAP, as measured through the jet cannula, to accurately reflect PAP in the trachea, and on the reliability of the expiratory pressure slope to accurately reflect tracheal pressure. Both measurements proved reliable. The Wright respirometer was less reliable than anticipated.
Under general anesthesia, adult airway resistance varies between 4 and 8 cm H2 O [center dot] L-1 [center dot] s-1. Values of 3-18 cm H2 O [center dot] L-1 [center dot] s-1 are found in emphysema and asthma . The airway resistance of patients ventilated for acute exacerbation of obstructive lung disease varied between 18 and 26 cm H2 O [center dot] L (-1) [center dot] s-1 in two studies [13,14]. Based on this information and review of RP flow-pressure curves, RP5-RP10 represents the equivalent of normal airway resistance, and RP20 is the equivalent of lung disease severe enough to require ventilatory support. Consequently, the greater PAP differences noted with the RP50 resistor are not clinically relevant. The maximal PAP difference that occurred using RP20 was 6 mm Hg. Therefore, with the exception of the PAP, all expiratory pressures measured through the TLBD are completely accurate in normal and diseased lung states (Figure 3 and Figure 5). The PAP measurement is acceptably accurate.
Why does PAP measured through the jet cannula increasingly diverge from PAP separately measured in the trachea as airway resistance and driving pressures increase? Airway pressure recordings from the jet cannula appear a fraction of a second after the PAP recordings from the trachea (Figure 5). This small time lag after jet deactivation allows pressure in the cannula to equilibrate with airway pressure and predictably increases with increasing driving pressures and smaller jet cannulae. It ranges from 0.10 s at 10 PSI to 0.18 s at 45 PSI when using a 16-gauge jet cannula. The time lag is shorter with a 14-gauge jet cannula, varying between 0.07 and 0.10 s, because pressure equilibrates faster through a larger cannula (Figure 5 d, and Figure 5 e). The behavior of the separately measured airway pressure traces during the time lag explains why the two PAP measurements differ. Pressure gradients generated by resistance are flow-dependent. Therefore, when flow stops, the pressure gradient immediately disappears. This manifests as a sudden pressure decrease from PAP, the size of which is resistance-dependent, when the jet is deactivated (Figure 5). Unfortunately, this pressure decrease occurs in the brief time lag described and therefore is not seen on the pressure trace recorded through the jet cannula. The PAP measured through the jet cannula therefore underestimates the real PAP. The underestimation, as noted, is not significant when the resistance is <25 cm H2 O [center dot] L-1 [center dot] s-1.
The airway pressure decreases due to compliance only as gas escapes from the lung. The suction of gas results in the gradual slope of expiratory airway pressure toward baseline. The decrease in pressure secondary to lung compliance is therefore negligible during the lag time. This explains why lung compliance changes do not cause a significant difference between PAP measurements.
The TLBD acts as a constant flow generator in the closed lung model. When used with an open larynx it is, by contrast, a constant pressure generator. Factors that influence airway pressures generated in the latter situation have been described [3,15-18].
The proximal airway pressure traces have a significant negative deflection on initiation of jetting at normal compliance and resistance levels (Figure 5). Lower compliance and higher resistance cause less negative deflection. The magnitude of the negative deflection was, as expected, greater with the larger cannula and higher driving pressures.
The TLBD also allows intermittent measurement of PETCO2 as follows. A stopcock is placed between the outer TLBD cannula and the suction line, and its side port is connected to a capnometer. Near the end of an expiration, the capnometer is opened to the trachea by switching the stopcock closed to the suction line. PETCO (2) read in this manner closely approximated the PaCO2 in blood samples drawn at the same time in experimental animals .
The Wright respirometer underestimated VT at low flow rates and overestimated VT at high flow rates. The reasons for this have been described . Even if it is inaccurate with strong suction, the Wright respirometer is a useful part of the TLBD system. Slowing of the volume hand warns of possible impending obstruction of the suction cannula. The SaO2 and PETCO2, in association with the estimated tidal volume, complement clinical judgment.
Transtracheal jet ventilation was first described in 1971 . Placing a second transtracheal catheter for airway pressure measurement to prevent barotrauma has been recommended . The TLBD allows ventilation, airway pressure and PETCO (2) measurement, and VT estimation through one catheter in all circumstances, including total upper airway obstruction. We regularly use transtracheal jet ventilation in our practice and have found that upper airway obstruction during laryngeal surgery is common. It may occur because of instrumentation, laryngoscope movement, or even laryngospasm [5,15]. Although more extensive animal and clinical testing of the TLBD is necessary, its potential to allow ventilation in patients with totally occluded upper airways may reduce or eliminate the need for many intraoperative tracheotomies.
We thank Daniel B. Carr, MD, for helpful comments during preparation of this manuscript; John Murray, BS, and Abigail Bentch, BA, for preparation of Figure 3, Figure 4, and Figure 6; Christina Demur, BS, clinical engineer, for help with equipment setup; and Deirdre Garry, BS, for help with collection and coordination of data.
1. Yealy DM, Stewart RD, Kaplan RM. Myths and pitfalls in emergency translaryngeal ventilation: correcting misimpressions. Ann Emerg Med 1988;17:7:690-2.
2. Gaughan SD, Benumof JL, Ozaki GT. Can an anesthesia machine flush valve provide for effective jet ventilation? Anesth Analg 1993;76:800-8.
3. Dworkin R, Benumof JL, Benumof R, Karagianes TG. The effective tracheal diameter that causes air trapping during jet ventilation. J Cardiothorac Anesth, 1990;4:6:731-6.
4. O'Sullivan TJ, Healy GB. Complications of venturi jet ventilation through the operating laryngoscope. Ann Otol Rhinol Laryngol 1982;91:615-21.
5. Schumacher P, Stotz G, Schneider M, Urwyler A. Laryngospasm during transtracheal high frequency jet ventilation. Anaesthesia 1992;47:855-6.
6. Keohane JD, Forte V, MacPherson B. Pressure, flow and resistance characteristics of the Storz-Hopkins bronchoscopes. J Otolaryngol 1991;20:3:155-7.
7. Benumof JL, Scheller MS. The importance of transtracheal jet ventilation in the management of the difficult airway. Anesthesiology 1989;71:769-78.
8. Crockett DM, Scamman FL, McCabe BF, et al. Venturi jet ventilation for microlaryngoscopy: technique, complications, pitfalls. Larynscope 1987;97:1326-30.
9. Garry B, Woo P, inventors; New England Medical Center Hospitals, Inc., assignee. Laryngeal bypass. US patent 5 669 380. September 23, 1997.
10. Cooper RM, Comen DR. The use of an endotracheal ventilation catheter for jet ventilation during a difficult intubation. Can J Anaesth 1994;41:12:1196-9.
11. Don HF, Robson, JC. The mechanics of the respiratory system during anesthesia: the effect of atropine and carbon dioxide. Anesthesiology 1965;26:168-78.
12. Mushin WW, Rendell-Baker L, Thompson PW, Mapleson WW. Automatic ventilation of the lungs. 3rd ed. Oxford, UK: Blackwell Scientific, 1980:3-5.
13. Broseghini C, Brandoleser, Poggi R, et al. Respiratory mechanics during the first day of mechanical ventilation in patients with pulmonary edema and chronic airway obstruction. Am Rev Respir Dis 1988;138:355-61.
14. Dhand R, Jubran A, Tobin M. Bronchodilator delivery by metered-dose inhaler in ventilator supported patients. Am J Respir Crit Care Med 1995; 151:1827-33.
15. Chakravarty K, Narayanan PS, Spoerel WE. Further studies on transtracheal ventilation: the influence of upper airway obstruction on the patterns of pressure and volume changes. Br J Anaesth 1973;45:733-7.
16. Baum M, Muth N. Physical characteristics of a jet in the airways. Acta Anesthesiol Scand 1989;33:90:46-50.
17. Eckerbom JB, Lindholm CE. Ventilation with rigid bronchoscopes: evaluation of resistance by pressure drop by a proposed international standard. Acta Anaesthesiol Scand 1994;38:462-6.
18. Spoerel WE, Greenway RE. Technique of ventilation during endolaryngeal surgery under general anesthesia. Can Anaesth Soc J 1973;20:369-77.
19. Woo P, Garry B, Perrault D Jr, Shapshay S. Total laryngeal bypass ventilation system for management of the difficult surgical airway [abstract]. Trans Am Laryngol Assoc 1996;170:36.
20. Nunn JF, Ezi-Ashi TI. The accuracy of the respirometer and ventrigator. Br J Anaesth 1962;34:422-32.
21. Spoerel WE, Narayanan PS, Singh NP. Transtracheal ventilation. Br J Anaesth 1971;43:932-9.
22. Craft TM, Chambers, PH, Ward ME, Goat VA. Two cases of barotrauma associated with transtracheal jet ventilation. Br J Anaesth 1990;64:524-7.