| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
We administered high-frequency jet ventilation (HFJV) to a tracheal-lung model with connectors of internal diameter 2.58.5 mm to simulate ventilation through varying degrees of laryngotracheal stenosis. With reductions in diameter, end-expiratory pressure (EEP) and peak inspiratory pressure (PIP) increased. During supraglottic, translaryngeal, and transtracheal HFJV, respectively, EEP was 10 mm Hg at diameters narrower than 5.5, 4.0, and 3.5 cm, and PIP was >20 mm Hg at diameters narrower than 5.5, 3.5, and 3.0 cm. EEP and PIP were greater during supraglottic HFJV than during translaryngeal and transtracheal HFJV (P < 0.01). At diameters of <3.5 and 4.0 cm, respectively, PIP and EEP increased and were significantly greater (P < 0.01) during translaryngeal HFJV than during transtracheal HFJV. In a second experiment, the degree of ventilation and air entrainment was assessed by administering nitrous oxide 4 L/min to the model. Nitrous oxide concentrations were significantly (P < 0.01) smaller and nitrogen concentrations were significantly (P < 0.01) larger during supraglottic HFJV than either translaryngeal or transtracheal HFJV. The larger EEP and PIP associated with supraglottic HFJV may be attributable to increased ventilation and air entrainment compared with translaryngeal and transtracheal HFJV. IMPLICATIONS: Ventilatory driving pressure during supraglottic high-frequency jet ventilation may be reduced to minimize high airway pressures and hence the potential for pulmonary barotrauma in patients with laryngotracheal stenosis.
The choice of method of oxygenation and ventilation is a major consideration when providing anesthesia for surgery in patients with laryngotracheal stenosis. Airway management using a conventional tracheal tube may not be possible and may obscure the surgical field. Alternatively, the technique of ventilation without a tracheal tube and total IV anesthesia may be preferable (13). High-frequency jet ventilation (HFJV) in these circumstances can be delivered by three routes. The supraglottic route involves delivery of HFJV proximal to the stenosis, whereas the translaryngeal route involves placement of a catheter through the stenotic glottis and laryngeal inlet, with delivery of HFJV distal to the stenosis. The transtracheal route involves subglottic administration of HFJV via a cannula placed percutaneously through the cricothyroid membrane or anterior wall of the trachea, below the stenosis. Inadequate oxygenation and ventilation during anesthesia can lead to complications such as hypoxemia, myocardial ischemia, and arrhythmias. During HFJV in patients with laryngotracheal stenosis, there is great potential for respiratory outflow tract obstruction leading to increased intrathoracic pressure and pneumothorax (3,4). With a critical diameter of <4.04.5 mm, tracheal stenosis has been associated with air trapping in a mechanical lung model ventilated with HFJV (5). End-expiratory pressure (EEP) correlates with end-expiratory pulmonary volume above apneic functional residual capacity (6) and is associated with gas trapping and upper airway obstruction during HFJV (7). By using EEP and peak inspiratory pressure (PIP), the primary aim of this study was to assess the potential for pulmonary barotrauma, during simulated supraglottic, translaryngeal, and transtracheal HFJV in a bench top model. In a second experiment, the degree of ventilation and air entrainment was assessed during these three methods of HFJV.
We used a tracheal lung model comprising a hollow tube of 22-mm internal diameter connected to an inflatable 2-L bag. To simulate laryngotracheal stenosis, connectors of fixed internal diameter were connected to the proximal end of the hollow tube as shown in Figure 1. The compliance of the 2-L bag was measured by using a Drager Evita II ventilator (Drager, UK), over the range of airway pressure developed.
A Monsoon Universal Jet Ventilator, set at an inspiratory time of 30%, delivering 100% oxygen at 2 bar and a rate of 150 min-1 was connected to the tracheal lung model. HFJV was administered in one of three positions:
Ten sizes of connectors of internal diameter 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, 5.5, 6.5, 7.5, and 8.5 mm were connected to the proximal end of the hollow tube to simulate varying degrees of tracheal stenosis during supraglottic and translaryngeal HFJV. Because of the size of the catheter, only connectors of sizes larger than 2.5 mm were used during transtracheal HFJV. During each of the 3 routes of HFJV, EEP and PIP were measured on 12 occasions for each stenosis, with the a pressure transducer sampling the intratracheal pressure 4 cm proximal to the 2-L bag. Before each reading, ventilation was discontinued to allow the bag to deflate. In a second experiment, we simulated the degree of ventilation and air entrainment during the three routes of HFJV delivered to connectors of internal diameters 3.5, 4.0, 5.0, 6.5, and 7.5 mm. Nitrous oxide (N2O) 4 L/min was administered to the system through the hollow tube during HFJV (Fig. 2). HFJV was commenced, and N2O and O2 concentrations were measured by sampling gas from within the 2-L bag. When the readings of these two gases reached a steady state, three measurements were taken at each degree of stenosis during each method of HFJV.
Data were analyzed by using general linear model analysis of variance for repeated measures with Bonferroni correction, using SPSS version 9.5.
The change in volume of the bag was linear over the range of airway pressure developed (Table 1).
With reductions in the diameter of the stenotic airway, EEP and PIP increased in a similar manner, during each of the three methods of HFJV (Fig. 3). EEP and PIP were significantly greater during supraglottic HFJV than during either translaryngeal or transtracheal HFJV (P < 0.01).
EEP and PIP were larger at wider diameters during supraglottic HFJV than during either translaryngeal or transtracheal HFJV. During these routes of HFJV, respectively, mean EEP was 10 mm Hg at diameters narrower than 5.5, 4.0, and 3.5 cm, and mean PIP was >20 mm Hg at diameters narrower than 5.5, 3.5, and 3.0 cm.
At narrow diameters of The difference between PIP and EEP (PIP - EEP) at stenoses larger than 4.5 cm was significantly (P < 0.01) different between the three methods of HFJV (Fig. 4). The range of mean PIP - EEP for diameters larger than 4.5 cm was 8.211.3, 3.54.6, and 1.82.0 mm Hg, for supraglottic, transtracheal, and translaryngeal HFJV, respectively.
N2O concentrations were significantly (P < 0.01) smaller and calculated nitrogen concentrations were significantly (P < 0.01) larger during supraglottic HFJV than either translaryngeal HFJV or transtracheal HFJV throughout the range of stenoses measured (Fig. 5).
In this tracheal lung model, supraglottic HFJV was associated with significantly larger EEP and PIP than either translaryngeal or transtracheal HFJV, at all diameters. This difference may in part be explained by the possibility of increased gas trapping during supraglottic HFJV than during either translaryngeal or transtracheal HFJV. During supraglottic HFJV with these ventilatory settings, 30% of the respiratory cycle was spent in inspiration, allowing 70% of the time for expiratory flow of gas. However, during either translaryngeal or transtracheal HFJV, gas trapping distal to the stenosis was less likely because ventilation was subglottic, allowing expiratory flow of gas through the stenotic larynx throughout respiration.
During supraglottic, translaryngeal, and transtracheal HFJV, mean EEP was During HFJV, pressures oscillate between PIP and EEP, and the value of EEP will affect the value of the next PIP, and vice versa. At large diameters, PIP and EEP were significantly (P < 0.01) larger during supraglottic HFJV than during either translaryngeal or transtracheal HFJV (Fig. 3). This difference cannot be explained alone by gas trapping and higher EEP that occur at smaller diameters. At stenosis >4.5 cm, the difference between PIP and EEP (PIP - EEP) was significantly higher during supraglottic HFJV than during either transtracheal or translaryngeal HFJV (Fig. 4). Thus, it may be inferred that the significantly higher difference between PIP and EEP at diameters >4.5 cm is caused predominately by increased ventilation and air entrainment during supraglottic HFJV than during either translaryngeal or transtracheal HFJV. At smaller diameters, however, this difference between PIP and EEP was not significantly different among all three routes of HFJV (Fig. 4). It was observed that at diameters of 3.5 cm and smaller, there was a plateau in the increase of EEP and PIP during supraglottic HFJV (Fig. 3). It is likely that at these narrow diameters, delivery of HFJV and air entrainment become restricted, thus limiting the increase in airway pressure. At a diameter of 3 cm, EEP and PIP were significantly larger during translaryngeal HFJV than during transtracheal HFJV. This difference is likely attributable to increased gas trapping during translaryngeal HFJV than during transtracheal HFJV. At small diameters, the effective orifice area of the stenosis available for expiratory flow of gas becomes critically reduced by the presence of the catheter during translaryngeal HFJV, but not during transtracheal HFJV. The results of our second experiment show that the difference in airway pressure among the three routes of HFJV may be attributable not only to differences in gas trapping but also to differences in ventilation and air entrainment. Because of the smaller N2O concentrations and larger calculated nitrogen concentrations developed during supraglottic HFJV, the larger EEP and PIP associated with supraglottic HFJV may be also attributable to increased ventilation and increased air entrainment, compared with the two subglottic methods. Our results differ from that of a clinical trial in which supraglottic HFJV was associated with higher end-tidal CO2 and hence reduced ventilation compared with subglottic HFJV (8). The measured end-tidal CO2 during supraglottic HFJV increased and decreased on several occasions in response to frequent readjustment of the position of the rigid supraglottic ventilation tube during surgery. In a clinical setting, therefore, there may be inevitable reductions in the delivery of supraglottic HFJV compared with the more controlled setting of a bench top model. The reported incidence of pulmonary barotrauma during jet ventilation seems to be highly variable. In a 10-year review of 942 patients who received endolaryngeal jet ventilation via a laryngoscope with an adapted Sanders injector, 4 patients developed pneumothoraces (4). In another series of 500 patients ventilated by superimposed HFJV through a laryngoscope, no complications were reported (9). In a prospective audit from 3 centers of 643 patients undergoing upper airway endoscopy with transtracheal HFJV via a catheter placed through the cricothyroid membrane, pneumothorax, subcutaneous emphysema, and pneumomediastinum were reported in 7, 67, and 16 patients, respectively (3). It was shown that subcutaneous emphysema was significantly more frequent after more than two attempts at tracheal puncture. Pulmonary barotrauma after jet ventilation with one or more of the above complications has also been cited in smaller case series, giving the impression of a frequent incidence; for example, 1 of 36 and 3 of 38 cases (10,11) were reported. One of the ways in which barotrauma can be minimized in patients with laryngotracheal stenosis is to prevent gas trapping during HFJV. Measurement of airway pressure during HFJV can give an indication of the potential for gas trapping and the develop-ment of pulmonary barotrauma. During supraglottic HFJV in clinical practice, airway pressure is measured proximal to the stenosis (12), whereas during subglottic HFJV, airway pressure is measured distal to the stenosis (8,12). Because of the possibility of a pressure gradient across a stenotic lesion, airway pressure measured proximally during supraglottic HFJV is likely to be an underestimate of the potential for barotrauma compared with airway pressure measured in patients during subglottic HFJV. An alternative technique that may be used to monitor for barotrauma is respiratory inductive plethysmography (RIP). RIP is noninvasive and is based on the principle that changes in lung and abdominal volumes produce concomitant changes in inductance in the sensors surrounding the chest and abdomen. There is controversy over the accuracy of this technique, with some investigators showing accurate measurements of lung volume (13,14) and others reporting inconsistent results in patients with chronic obstructive pulmonary disease (15). With sufficient expertise, RIP may be used to monitor for barotrauma during HFJV in patients with laryngotracheal stenosis. Another problem for the anesthesiologist is the inability to measure the adequacy of ventilation during HFJV. Although there is good correlation in clinical trials between end-tidal and arterial PCO2 measurements during subglottic normo-frequency jet ventilation (16), and supraglottic normo-frequency jet ventilation (17), there is only one case report of such agreement during subglottic HFJV (18). An alternative method is the measurement of transcutaneous PCO2 that has been used to modify the driving pressure during subglottic HFJV in the setting of a clinical trial (19). In so doing, minimal driving pressures can be applied to prevent possible pulmonary barotrauma. In clinical practice, additional considerations apply. The possible problems of supraglottic HFJV are gastric distension, soiling of the lungs, vocal cord movement, and inability to monitor distal airway pressure and end-tidal CO2 (8). Subcutaneous emphysema is a particular risk with transtracheal HFJV (20,21), and the ability to perform translaryngeal HFJV is dependent on the size of stenotic lesion. Furthermore, other methods of ventilation, such as high-frequency positive pressure ventilation, are reasonable alternatives and would need to be considered. The utility of our model is that it enables a simple comparison of the three routes of HFJV over a range of predetermined stenotic diameters. A study of this type would not be feasible using patients at risk of inadequate oxygenation and ventilation. However, we appreciate that there may be differences between our model and the situation in vivo, in terms of airway anatomy, controlled delivery of HFJV in a shared airway, and specific patient factors. In this tracheal lung model, we have found that supraglottic HFJV was associated with higher EEP and PIP than either translaryngeal or transtracheal HFJV. This difference is attributable not only to increased gas trapping but also to increased ventilation and air entrainment during supraglottic HFJV than during either translaryngeal or transtracheal HFJV. The implication of our results for the practicing anesthesiologist is that ventilatory driving pressure during supraglottic HFJV may be reduced to minimize high airway pressures and hence the potential for pulmonary barotrauma in patients with laryngotracheal stenosis.
Presented in part to the Anaesthetic Research Society, Nottingham, UK, November 22, 2001.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|