Special Feature
Update on Experimental Ventilation Approaches in ARDS
by Robert M. Kacmarek, PhD, RRT, FCCM, FCCP
Introduction
Over the last 40 years, numerous approaches to ventilatory support during acute respiratory distress syndrome (ARDS) have been debated. The debate has centered on volume ventilation vs. pressure ventilation, high frequency vs. normal frequency, inverse inspiration-to-expiration ratio vs. normal ratio, and minimal positive end-expiratory pressure (PEEP) vs. super PEEP vs. PEEP based on lung mechanics, among others. Paralleling these debates has been the development of experimental or adjunctive techniques to support gas exchange during ARDS. Currently, much attention has been directed toward prone positioning, tracheal gas insufflation (TGI), inhaled nitric oxide (INO), and partial liquid ventilation (PLV).
Prone Positioning
Position has a dramatic effect on functional residual capacity (FRC), especially during anesthesia. In anesthesized/paralyzed normal adults, the supine position decreases FRC more than the prone position does.1 (See Table 1.) These data, coupled with the fact that lateral positioning in unilateral lung disease (good lung down) improves oxygenation, have prompted many to consider prone positioning in ARDS patients.
Table 1
Effects of body position on FRC in anesthetised and/or paralyzed patients
· Supine FRC decrease 44%
· Prone FRC decrease 12%
Mechanism of Action. The precise mechanism responsible for improved oxygenation in ARDS with the prone position is not well understood. However, a few physiologic facts contribute to the success of prone positioning. As a result of normal chest wall/lung compliance in the supine position and the weight of the heart and great vessels on the lungs, there is a greater transpulmonary pressure gradient in nondependent lung than in dependent lung. As a result, in ARDS, there is a greater tendency for dependent lung (supine position) to collapse. In the prone position, the distribution of transpulmonary pressures is more equal, dependent to non-dependent, ensuring a more uniform distribution of ventilation.2 In addition, a greater percentage of the lung is dependent (about 60%) in the supine position as compared to the prone position (about 40%). In addition, chest wall compliance is reduced in the prone position, and improvement in PaO2 with prone positioning is significantly correlated to the decrease in chest wall compliance when patients are placed prone.2 This would lead one to predict that patients with normal chest wall compliance and dependent consolidation in ARDS would be likely to show a positive response to prone positioning.
Controlled Trials. A number of controlled trials have shown that on being moved into the prone position, more than 75% of ARDS patients demonstrate a positive response (> 20% increase in PaO2/FIO2 ratio).3-5 This type of response outweighs the potential adverse effects of prone positioning, such as dislodgement of the endotracheal tube, loss of vascular access, and hemodynamic instability. In my experience with this technique, loss of the airway or vascular lines can be avoided by involving at least four individuals in the change-over from supine to prone position. In fact, Chatte and colleagues reported only one loss of the airway and one loss of vascular access in 294 positionings.3
Recommendations. Considering the low cost and relative ease of use of this adjunct and its low morbidity, in spite of the lack of prospective randomized controlled trials, I would recommend the use of prone positioning in ARDS patients when chest wall compliance is normal, provided PEEP is appropriately set (that is, above the lower inflection point on the pressure-volume curve of the respiratory system) and FIO2 is more than 0.6 with a PaO2 less than 60. How long to keep a patient prone and when to return them to the supine position is dependent upon response. As long as the improvement in oxygenation in the prone position persists, the patient does not require turning supine. If the PaO2 starts to decrease, repositioning to the supine position for 1-2 hours is indicated.
Tracheal Gas Insufflation
Tracheal gas insufflation is the injection of a secondary gas flow into the trachea that flushes the trachea, endotracheal tube, and ventilator circuit dead space of end-expiratory CO2, thereby decreasing PaCO2. A number of studies in both animals and humans have clearly demonstrated that TGI is effective in reducing PaCO2, and that the more hypercapnic the patient, the more effective the TGI in this respect.6-8
Technical Issues. A number of groups have compared the effect of various approaches to TGI, including continuous flow (during both inspiration and exhalation), intermittent flow (during expiration only), or volume-adjusted continuous flow. The direction of injection of the TGI flow (that is, toward the carina or toward the mouth) also may be varied. Continuous-flow TGI is the most problematic, since injection during inspiration and exhalation results in increases in peak airway and end-inspiratory plateau pressures with both volume- and pressure-targeted ventilation.9 This occurs because, even during pressure ventilation, few mechanical ventilator exhalation valves are active during inspiration (exceptions include the Dräger Evita 4 and Nelcor Puritan Bennett 840 ventilators). As a result, the TGI flow is additive to the ventilator flow, resulting in elevated system pressure. Adjusting delivered tidal volume based on the TGI flow avoids this problem but can only be considered when using square wave flow during volume ventilation.10
Expiratory phase-only TGI with pressure ventilation would appear to be the ideal approach to the use of this modality.10 In this approach, the benefits of both pressure ventilation and TGI can be appreciated without affecting inspiratory airway pressures. Whether TGI flow should be injected toward the mouth or toward the carina is still controversial. Some indicate an advantage to direct TGI (toward the carina),11 while others find no difference between direct and reverse (toward the mouth) flow.12 However, with direct TGI, total PEEP can be markedly increased by the TGI flow, whereas with reverse TGI, total PEEP is only mildly reduced.12
Recommendation. Expiratory phase-only TGI is ready for use in patients with markedly elevated PaCO2 (> 60 mmHg). However, it should not be attempted without the availability of appropriate manufacturer-designed devices, which are not currently available. Expiratory phase TGI requires electrical coordination with ventilator triggering. (See Table 2.) In addition, safeguards against excessive airway pressure as a result of system obstruction must be developed as well as appropriate humidification devices capable of tolerating the high TGI pressure. Although TGI is clinically useful, routine clinical use must await systems appropriately designed by manufactures.
Table 2
Ideal approach to TGI
· Expiratory phase only
· Pressure control ventilation
· Either direct or reverse direction
Inhaled Nitric Oxide (INO)
Over the last 10 years, nitric oxide (NO) has been one of the most extensively studied molecules in history. In 1992, Science considered NO the "molecule of the year".13 In ARDS, INO has been extensively studied because it is a selective pulmonary vasodilator, affecting only blood vessels in direct communication with well-ventilated lung units, and not affecting systemic vasculature because of the rapid reaction of NO with hemoglobin.
Clinical Data. Not all ARDS patients inhaling NO respond with a more than 20% increase in PaO2 or a more than 20% decrease in pulmonary arterial pressure (PAP). Although it is not clear why, it is clear that patients who are septic respond poorly to INO,14 and that response requires maximal recruitment of the lung.15 Dose-response studies clearly indicate that the optimal dose of INO in ARDS is less than 10 ppm.16 Regardless of NO dose, a post-discontinuation rebound increase in PAP and decrease in PaO2 are common and most pronounced the more compromised the patient's oxygenation status is at the time of discontinuation.17
Randomized Controlled Trials. A number of randomized comparisons of INO vs. placebo in ARDS have been conducted,18-21 all of which have failed to show improvement in outcome. In addition, each has shown that the oxygenation benefit of INO is lost by about 48 hours. Two of these trials-by Dellinger et al18 and by Landin et al21-were large prospective multicenter trials. In post hoc analysis, Dellinger and associates did identify a subgroup receiving 5 ppm INO, in which the percentage of patients alive and off the ventilator by day (28) was greater (62%) with NO than with placebo (44%; P < 0.05). Currently, a phase III multicenter trial is comparing a double-blinded approach 5 ppm NO to placebo.
Recommendation. The likelihood of INO to show an effect on outcome in ARDS seems low. The results of the current phase III trial, if similar to those of previous trials, could signal the end of INO as a potential adjunctive therapy in ARDS. However, many clinicians are counting on the anti-inflammatory affects of INO, identified in the basic science literature, to improve outcome in this new study that uses INO early in the course of ARDS.
Partial Liquid Ventilation
Partial liquid ventilation is the provision of gas ventilation to a lung filled to FRC with a perflurocarbon liquid. The perflurocarbon used in clinical trials is perflubron (C8F17Br, Alliance Pharmaceuticals). It is colorless, odorless, insoluble in water, biologically inert, and chemically stable. The single bromide molecule on the carbon chain makes the substance radiographically opaque. Perflubron has a density of 1.9 gm/mL, a surface tension of 18 dynes/cm, and a spreading coefficient of 2.7 dynes/cm. These properties afford the substance surfactant-like abilities, and the high density of perflubron has resulted in the fluid being referred to as "liquid PEEP." The solubility of oxygen in perflubron is 53 mL/100 mL (as compared to the normal arterial blood O2 content of 20 mL/100 mL), and the solubility of CO2 is 210 mL/100 mL.
Mechanism of Action. The proposed benefits of partial liquid ventilation with perflubron include improved gas exchange, ability to use a lower FIO2, lower end-inspiratory plateau pressure, and improved pulmonary compliance. The mechanisms by which PLV improves gas exchange are thought to be lung recruitment, prevention of alveolar collapse, redistribution of pulmonary blood flow, and pulmonary lavage.
Clinical Trials. Few data are available on the use of PLV in patients. Hirschl et al22 reported in 1995 on the first 19 patients (adults, children, and neonates) to be treated with PLV. All patients were supported by extra-corporeal membrane oxygenation (ECMO) at the time PLV was administrated. Within 24 hours of beginning ventilation with PLV, oxygenation (as measured by P[A-a]O2) and compliance improved significantly (P < 0.001). More recently, Leach et al23 reported on 13 neonates with severe respiratory distress treated with PLV and conventional ventilation. None of these children were receiving ECMO. Within one hour after dosing with perflubron, PaO2, PaCO2, and dynamic compliance all improved significantly (P < 0.01).
At the 1997 meeting of the American Thoracic Society, Weidemann presented unpublished data on the Alliance Pharmaceutical Company sponsored phase II adult PLV trial. Of particular interest was the fact that post hoc analysis showed that patients younger than 55 had a greater number of ventilator-free days within the 28 days after dosing with PLV than did control patients (9 vs 4 days, P < 0.05) and that there was a non-significant trend toward lower mortality in the PLV group. To date, this is all the data available on the clinical use of perflubron, but, clinical trials will be starting again this winter.
Recommendation. It is too early to tell whether PLV will have any effect on outcome in acute respiratory failure. Laboratory trials are ongoing to identify more appropriate methods for ventilating patients during PLV.
Table 3
Experimental approaches to ventilation in ARDS: Recommendations regarding use
· Prone positioning-use now
· TGI-awaiting manufacturer designed devices
· INO-RCT results do not look promising
· PLV-too little information to evaluate
Summary
All four of the techniques discussed here show theoretical promise as adjuncts to mechanical ventilation in patients with ARDS. I would recommend the careful use of prone positioning in ARDS. TGI is ready for clinical use, but appropriately designed and tested delivery systems are not yet available. Nitric oxide has been extensively studied but has not demonstrated improved outcome. PLV is the most experimental of the four adjuncts, still requiring extensive clinical research before it can be judged as even potentially useful in the management of ARDS. (See Table 3.)
References
1. Coonan TJ, Hope CE. C Anesth Soc J 1983;30:424-437.
2. Pelosi P, et al. Am J Respir Crit Care Med 1998; 157:387-393.
3. Chatte G, et al. Am J Respir Crit Care Med 1997; 155:473-478.
4. Stocker R, et al. Chest 1997;111:1008-1017.
5. Sevillo G, et al. Intensive Care Med 1997;23:1219-1224.
6. Nahum A, et al. Crit Care Med 1995;23:348-356.
7. Ravenscraft SA, et al. Am J Respir Crit Care Med 1996;153:1817-1824.
8. Ravenscraft SA, et al. Am Rev Respir Dis 1993; 148:345-351.
9. Imanaka H, et al. Am J Respir Crit Care Med 1996; 153:1019-1024.
10. Imanaka H, et al. Crit Care Med 1998;26:939-946.
11. Nahum A, et al. J Appl Physiol 1993;75:1238-1246.
12. Imanaka H, et al. Am J Respir Crit Care Med (In Press).
13. Kacmarek RM. Crit Care Alert 1993;1:7-8.
14. Manktelow C, et al. Anesthesiology 1997;87:297-230.
15. Puybasset L, et al. Am J Respir Crit Care Med 1995; 152:318-328.
16. Gerlach H, et al. Eur J Clin Invest 1993;23:499-502.
17. Lavoie A, et al. Am J Respir Crit Care Med 1996; 153:1985-1987.
18. Dellinger RP, et al. Crit Care Med 1998;26:15-23.
19. Toney E, et al. Am J Respir Crit Care Med 1998; 157:1483-1488.
20. Michael JR, et al. Am J Respir Crit Care Med 1998; 157:1372-1380.
21. Landin S, et al. (Abstract) Intensive Care Med 1997; 23:S2.
22. Hirschl RB, et al. JAMA 1996;275:383-389.
23. Leach CL, et al. N Engl J Med 1996;335:761-767.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content