How to achieve the best outcomes for acute lung injury patients receiving ventilation is under debate.

By Graeme A’Court, BSc, ASCT, RRT

The optimal ventilation strategy in patients with acute lung injury is currently a subject of much debate. Participants in a recent American College of Chest Physicians Consensus Conference1 suggested that some of the primary clinical objectives of mechanical ventilation are to reverse hypoxemia, reverse respiratory acidosis, relieve respiratory distress, prevent or reverse atelectasis, and reverse respiratory muscle fatigue. For many years, the conventional strategy was to achieve these goals and normalize blood gases using a volume-oriented approach to ventilation.

Today, we know from Gattinoni et al2 that the reduced compliance in patients with adult respiratory distress syndrome (ARDS) occurs because much of the lung is collapsed and not accessible to ventilation; the tidal volume is delivered to the small amount of aerated lung. The specific compliance of the aerated lung (compliance per gram of aerated lung tissue) may be relatively normal. If a volume-oriented strategy is used, it is necessary to set lower tidal volumes to prevent alveolar overdistension in the aerated lung. Evidence now suggests that such overdistension may result in progressive lung injury.


Extensive evidence from animal studies3-6 has shown that mechanical ventilation can result in acute parenchymal lung injury that is histologically similar to ARDS (in addition to conventional barotrauma). The lung injury can be progressive and cause death from respiratory failure. It is also associated with an inflammatory response, and greatly increased concentrations of cytokines have been demonstrated both in lung lavage fluid7 and in blood in animals ventilated using injurious strategies. Mechanical ventilation has also been shown to potentiate the development of bacteremia after instillation of bacteria into the trachea in animal models.8,9 These findings have led to concerns that mechanical ventilation may augment lung injury and possibly induce a systemic inflammatory response that could contribute to the development of the multiple organ dysfunction syndrome (MODS) in patients with ARDS.6 The detailed mechanisms of lung injury and the associated inflammatory response are not yet known.

Two characteristics of the mechanical ventilation pattern appear to be important in determining ventilator-induced lung injury in animal models. First, end-inspiratory over-distension causes injury, and greater degrees of overdistension result in more rapid injury. It is clearly a high end-inspiratory lung volume that is important, rather than a high end-inspiratory pressure, so the lung injury resulting from this mechanism has been called volutrauma. Second, in surfactant-depleted animals, ventilation with low positive end-expiratory pressure (PEEP)-and, thus, a low end-expiratory volume-also results in injury even without end-inspiratory overdistension.10,11 This may be caused by repeated end-expiratory collapse and tidal reinflation in some lung regions. It is thought that this results in very high shear forces at the interfaces between collapsed and aerated lung areas during reinflation. Sufficient PEEP to prevent end-expiratory collapse largely prevents ventilator-induced lung injury in animal models10,11 when end-inspiratory overdistension is avoided.


To avoid ventilator-induced lung injury, mechanical ventilation should be conducted with sufficient PEEP to prevent end-expiratory collapse and tidal recruitment, and with a small tidal volume to avoid end-inspiratory overdistension. The dependent lung regions in ARDS are subject to a superimposed pressure from the weight of the overlying lung, and this tends to result in collapse of these regions at the end of expiration.12 Providing sufficient PEEP to avoid end-expiratory collapse of alveoli or small airways in these dependent lung regions (optimum PEEP) probably results in higher than normal end-expiratory volume in alveoli in nondependent regions because they have less superimposed pressure (and, therefore, higher transpulmonary pressure, since static end-expiratory airway and alveolar pressures are equal in all regions). Thus, the nondependent regions are at risk of end-inspiratory overdistension if a normal tidal volume is used. In addition, the amount of aerated lung in ARDS is often quite small, even with optimum PEEP.2 For both reasons, it is necessary to use a low tidal volume with optimum PEEP to avoid end-inspiratory overdistension, and a tidal volume of 6 mL/kg or less is often required.

It has been increasingly recognized that chest-wall compliance is low in many ARDS patients13 because of chest-wall edema and abdominal distension. In such patients, a greater than normal proportion of airway pressure is dissipated in distending the chest wall rather than the lung, and the transpulmonary pressure is lower (and the pleural pressure is higher) for any level of end-inspiratory plateau pressure. In patients with reduced chest-wall compliance, therefore, a higher plateau pressure can be accepted without risk of lung overdistension. Chest-wall compliance and transpulmonary pressure can be estimated using an esophageal balloon, or reduced chest-wall compliance can be assumed in patients with abdominal distension and chest-wall edema. The plateau pressure is usually measured to estimate lung distension rather than peak inspiratory pressure (PIP) because the latter is affected by changes in airway and endotracheal-tube resistance; thus, a high PIP may not necessarily reflect excessive lung distension. The plateau pressure is not affected by resistance, as it is measured during no-flow conditions.

The limitation of plateau pressure with optimum PEEP in severe ARDS usually requires a low tidal volume and frequently results in hypercapnia, but it is usually possible to maintain satisfactory oxygenation. It is generally necessary to use a higher level of PEEP when a low tidal volume is used.14 Hypercapnia is usually remarkably well tolerated in patients with ARDS, especially if it develops gradually, although a number of potential adverse effects may occur. It is usually considered to be contraindicated in patients with raised intracranial pressure (although, in one recent study,15 hypercapnia was allowed to develop gradually in ARDS patients with head injuries, provided intracranial pressure was monitored and remained normal).


Many approaches to selecting PEEP have been suggested. In the past, the emphasis was on optimizing oxygen saturation or oxygen delivery. More recently, the concept of using a lung-protective strategy by preventing end-expiratory collapse has been proposed.16-18 The best method of determining the level of PEEP required to achieve this remains unclear. It has been suggested that the lower inflection point of the static pressure-volume curve represents the pressure range over which recruitment of previously collapsed lung occurs; setting PEEP above the lower inflection point should prevent end-expiratory collapse.19 It has also been suggested that the upper inflection point represents the pressure range above which lung overdistension occurs, and that the plateau pressure should be limited to this pressure.19 These concepts may, however, be overly simplistic. A mathematical model of ARDS lungs developed by one of the authors20 suggests that the lower inflection point may represent only the beginning of recruitment, and recruitment may continue on the linear portion of the pressure-volume plot. It also suggests that the slope of the linear portion may be greater than the compliance of all aerated alveoli because of recruitment, and that an upper inflection point may occur at low pressures as recruitment diminishes, even in the absence of lung overdistension. An upper inflection point that is related to overdistension may be greatly modified or even obliterated by continuing recruitment. The model suggests that the PEEP level giving best compliance may also be misleading, and may overestimate or underestimate optimum PEEP. Thus, the pressure-volume curve measured at zero PEEP may not predict optimum ventilator settings; further studies are needed to clarify these issues. Pressure-volume curves at incremental PEEP levels may be more helpful, but measurement of absolute lung volume is probably the most reliable approach to detecting the optimum PEEP to prevent end-expiratory collapse. Using this approach, optimum PEEP is the level giving the maximum absolute lung volume at a pressure of 15 or 20 cm H2O (or the pressure of the highest PEEP level evaluated) during inspiration. Unfortunately, it is not easy to measure absolute lung volume at the bedside. The PEEP giving the maximum Pao2 may not differ greatly from that producing minimum end-expiratory collapse, but further studies will be required to determine whether this approach is satisfactory. Current evidence suggests that higher PEEP levels than those used in the past may be beneficial provided the plateau pressure is limited to a safe level by reducing tidal volume. However, the best method of selecting “optimum PEEP” remains unclear.


To avoid ventilator-induced lung injury, a recruitment/pressure-limitation strategy should be implemented at the initiation of mechanical ventilation. Hypercapnia then usually develops gradually, limiting the associated physiological effects by allowing time for compensation of the intracellular and extracellular pH. Nevertheless, some patients develop high respiratory drive and rapid respiratory rates. In conscious patients, this is often associated with dyspnea and discomfort, and heavy sedation may be required. If the tachypnea is allowed to persist, some patients’ respiratory drive gradually adapts to hypercapnia, and the respiratory rate falls over 24 to 48 hours. In other patients, the tachypnea and high respiratory workload are more persistent. Opiates or sedation may be used to moderate the respiratory muscle workload. Alternatively, muscle relaxants may be used, and if the ventilation is gradually reduced over 24 hours (allowing progressive hypercapnia), many patients then adapt better to hypercapnia. Frequently, the relaxants can be stopped after 48 hours, and hypercapnia is then tolerated with less tachypnea.21 It is important to realize that high respiratory drive and the associated development of highly negative pleural pressure during inspiration greatly increase transpulmonary pressure during pressure-control ventilation (PCV) or pressure-support ventilation, or even during continuous positive airway pressure. It may, therefore, be important to depress respiratory drive in such patients in order to limit lung distension, as well as for the patient’s comfort-although this is often difficult to achieve. It may also be important to limit the inspiratory pressure to much less than 35 cm H2O so that transpulmonary pressure and lung distension are not further increased.

After the optimum PEEP level has been set, plateau pressure should be measured with an inspiratory hold if volume-control ventilation is being used. Tidal volume should then be set to maintain plateau pressure at less than 35 or 40 cm H2O (or, perhaps higher, in patients with severe abdominal distension). Using PCV, the tidal volume should be limited to the same levels. This ensures that peak transalveolar pressure is limited to a safe level. It is usually necessary to reduce tidal volume to 7 mL/kg or less in severe ARDS.


Severe hypoxemia is still sometimes a major problem in treating a critically ill patient. In ARDS, the cause of hypoxemia is mainly compression atelectasis in the dependent lung regions; this decreases the resting lung volumes and lung compliance and results in ventilation-perfusion mismatching. PEEP can usually prevent some of the end-expiratory collapse, but often other therapies are required.

There is increasing evidence that, in some patients with severe ARDS, maximum lung aeration may not be achieved even at the end of inspiration using a plateau pressure of 35 cm H2O. In a recent CT scan study of patients with ARDS, Gattinoni et al22 showed that end-inspiratory aeration of the most dependent lung regions increased as plateau pressure was increased, up to a mean of 46 cm H2O. Higher levels of plateau pressure were not studied, so it is possible that greater aeration may have occurred at even higher pressures.

A study of saline-lavaged pigs showed that 10 minutes’ ventilation with a plateau pressure of 55 cm H2O was required to achieve full aeration of the lungs on CT scan after previous collapse, whereas aeration could be maintained after this with lower levels of peak and mean airway pressure.23 This is not at all surprising because it is well known that alveoli or small airways require a much higher pressure to reinflate once collapsed than to maintain inflation (especially in surfactant-depleted lungs). Once the alveolus is inflated to a normal or high volume, the radius of curvature is much larger, so the pressure required to overcome surface-tension forces is much less. Some lung units may have high critical opening pressures, and may not become reinflated following collapse until high distending pressures are applied. A period of ventilation with a higher plateau pressure (perhaps 40 to 50 cm H2O) may thus allow progressive recruitment of collapsed lung regions, with gradual improvement in oxygenation. It may then be possible to reduce the plateau pressure while maintaining inflation with sufficient PEEP. Reinflation may also require, at times, that the critical distending pressure be applied for a minimum period of time. In some patients, the application of a sustained inflation for 30 to 45 seconds with a pressure of 40 to 50 cm H2o may be even more effective in improving oxygenation.24

As the inspiratory-to-expiratory ratio is increased above the usual 1:3 or less (extended ratio), the mean airway pressure increases, and this usually results in improved oxygenation. When the ratio becomes greater than 1:1, it is referred to as inverse ratio. As well as an immediate effect, there may be a delayed improvement in oxygenation over several hours. This may result from progressive recruitment of previously collapsed lung regions.24 The longer inspiratory time applied with inverse ratio may allow better recruitment of regions characterized by time-dependent recruitment,24 although clear data to explain the mechanisms of this effect are lacking. A sustained inflation may be even more effective in recruitment.

In patients with unilateral lung consolidation, it has been known for many years that repositioning the patient in the lateral position with the good lung down often leads to significant improvement in oxygenation. This probably occurs because of redistribution of blood flow to the dependent (aerated) areas. In recent years, there has been growing interest in therapeutic prone positioning to improve lung function.

The goal of this technique is to improve the functional residual capacity (which is reduced because of collapsed alveoli, caused mainly by the superimposed pressure on dependent lung zones). Lung recruitment is improved because of the change of position of the heart and great vessels that takes place in prone positioning and the redistribution of the gravitational gradient of superimposed pressure. Gattinoni et al25 have observed the redistribution of densities in CT scan studies upon transition from supine to prone positions. In the supine position, airway collapse occurs in the paravertebral regions, whereas, in the prone position, it occurs in the parasternal lung regions. This redistribution of the areas of atelectasis is sometimes, but not always, associated with a reversal of the improved oxygenation; thus, in some patients, the improved oxygenation is transient. Other complex changes may also occur, and the mechanisms of improved oxygenation are not yet fully understood. To maximize any benefits of prone positioning, it should be considered early in the care of patients with ARDS who require a high fraction of inspired oxygen.26 The risks of prone positioning are well known and include difficulties in management of the artificial airway, in oral hygiene, and in management of invasive vascular catheters.

Inhaled nitric oxide has been reported to cause selective pulmonary vasodilatation in ventilated lung regions and to improve pulmonary gas exchange in the patient with ARDS. Estimates of its potential therapeutic value have been based on reports of improved oxygenation from a decrease in intrapulmonary shunt, lowered pulmonary arterial pressure, and a potential reduction in morbidity and mortality. Inhaled nitric oxide has been shown to acutely increase Pao2 in patients with ARDS and to improve hypoxemia in a dose-dependent fashion up to a concentration of approximately 20 ppm. Results of the only major clinical trial27 reported in adults confirmed that nitric oxide improves oxygenation in 60 percent of patients (at least temporarily). The mortality rate, however, was not reduced, compared with that of the placebo group. Nitric oxide still needs further investigation to validate its role in the treatment of ARDS. Given current data, the recommendation can be to use nitric oxide only as an adjunct to ventilation to buy time when treating severe hypoxemia and documented pulmonary hypertension.

The role of extracorporeal support in ARDS remains controversial. It can support gas exchange, but a recent randomized trial28 showed no difference in mortality rate between the extracorporeal support and control groups (although survival was better in both groups than in a 1979 trial of extracorporeal membrane oxygenation). The need for extracorporeal support appears to be falling in recent years, perhaps as a result of modified ventilation strategies.


In deciding on a strategy for ventilation management of the injured lung, the prevention of ventilator-associated lung injury should be a major consideration. Trying to put all of the evidence together may not be easy, but one should use a strategy that seeks to improve outcomes by minimizing or preventing additional iatrogenic lung injury, and this should be done early in the course of ARDS. Further studies should gradually clarify optimum management.


Graeme A’Court, BSc, ASCT, RRT, is a clinical education consultant for a respiratory manufacturer in Asia Pacific. Peter Otto, RRT, is a respiratory specialist based in Southeast Asia. Keith Hickling, MD, is the clinical director, intensive care unit, Queen Elizabeth Hospital, Hong Kong.


  1. Slutsky A. Mechanical ventilation. Report of the American College of Chest Physicians Consensus Conference. Chest. 1993;104:1833-1859.
  2. Gattinoni L, Pesenti A, Avalli L, Rossi F, Bombino M. Pressure-volume curve of total respiratory system in acute respiratory failure. Computed tomographic study. Am Rev Respir Dis. 1987;136:730-736.
  3. Hickling KG. Ventilatory management of ARDS: can it affect the outcome? Intensive Care Med. 1990;16:219-226.
  4. Hickling KG. Low volume ventilation with permissive hypercapnia in the adult respiratory distress syndrome. Clinical Intensive Care. 1992;3:67-78.
  5. Parker JC, Hernandez LA, Peevy KJ. Mechanisms of ventilator-induced lung injury. Crit Care Med. 1993;21:131-143.
  6. Hickling K. Mechanical ventilation in ARDS: is the lung a promoter of multiple organ failure? J Anesth. 1997;11:50-64.
  7. Tremblay L, Valenza F, Ribero SP, Li J, Slutsky AS. Injurious ventilatory strategies increase cytokines and c-fos mRNA expression in an isolated rat lung model. J Clin Invest. 1997;99:944-952.
  8. Parker JC, Roohparvar S, Foster J, et al. High peak inspiratory pressures (PIP) affect the rate of bacterial clearance from rabbit lungs. Am Rev Respir Dis. 1991;143:A570.
  9. Nahum A, Hoyt J, McKibben A, et al. Effect of mechanical ventilation strategy on E. coli pneumonia in dogs. Am J Respir Crit Care Med. 1996;153:A530.
  10. Sandhar BK, Niblett DJ, Argiras EP, et al. Effects of positive end-expiratory pressure on hyaline membrane formation in a rabbit model of the neonatal respiratory distress syndrome. Intensive Care Medicine. 1988;14:538-546.
  11. Muscadere JG, Mullen JBM, Gan K, et al. Tidal ventilation at low airway pressures can augment lung injury. Am J Respir Crit Care Med. 1994;149:1327-1334.
  12. Gattinoni L, D’Andrea L, Pelosi P, Vitale G, Pesenti A, Fumagalli R. Regional effects and mechanism of positive end-expiratory pressure in early adult respiratory distress syndrome. JAMA. 1993;269:2122-2127.
  13. Mergoni M, Martelli A, Primavera S, Rossi A. Chest wall static PV curves in patients with acute lung injury. Am J Respir Crit Care Med. 1995;153:A765.
  14. Ranieri VM, Mascia L, Fiore T, et al. Cardiorespiratory effects of positive end-expiratory pressure during progressive tidal volume reduction (permissive hypercapnia) in patients with acute respiratory distress syndrome. Anesthesiology. 1995;83:710-720.
  15. Gentilello L, Anardi D, Mock C, Arreola-Risa C, Maier RV. Permissive hypercapnia in trauma patients. J Trauma. 1995;39:846-852.
  16. Lachmann B. Open up the lung and keep the lung open. Intensive Care Medicine. 1992;18:319-321.
  17. Amato M, Barbas C, Medeiros D, et al. Beneficial effects of the “open lung” approach with low distending pressures in acute respiratory distress syndrome. A prospective randomized study on mechanical ventilation. Am J Respir Crit Care Med. 1995;152:1835-1846.
  18. Amato M, Barbas C, Medeiros D, et al. Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med. 1998;338:347-354.
  19. Bentio S, Lemaire F. Pulmonary pressure-volume relationship in acute respiratory distress syndrome in adults: role of positive end expiratory pressure. J Crit Care. 1990;15:27-34.
  20. Hickling KG. The pressure-volume curve is greatly modified by recruitment: a mathematical model of ARDS lungs. Am J Respir Crit Care Med. 1998;158:194-202.
  21. Hickling K. Targets during mechanical ventilation. In: Marini JJ, Slutsky AS, eds. The Physiological Basis of Ventilatory Support. New York: Marcel Dekker; 1997: 655-708.
  22. Gattinoni L, Pelosi P, Crotti S, Valenza F. Effects of positive end-expiratory pressure on regional distribution of tidal volume and recruitment in adult respiratory distress syndrome. Am J Respir Crit Care Med. 1995;151:1807-1814.
  23. Sjostrand UH, Lichtwark-Aschoff M, Neilsen JB, et al. Different ventilatory approaches to keep the lung open. Intensive Care Medicine. 1995;21:310-318.
  24. Marini JJ, Amato MB. Lung recruitment during ARDS. In: Marini JJ, Evans TW, eds. Acute Lung Injury. Berlin, Germany: Springer-Verlag; 1997:236-257.
  25. Gattinoni L, Pelosi P, Vitale G. Body position changes redistribute lung computed-tomographic density in patients with ARDS. Anesthesiology. 1991;74:15-23.
  26. Broccard A, Marini JJ. Effect of position and posture on the respiratory system. In: Vincent JL. 1995 Yearbook of Intensive Care and Emergency Medicine. Berlin, Germany: Springer-Verlag; 1995:165-184.
  27. Dellinger RP, Zimmerman JL, Taylor RW, et al. Effects of inhaled nitric oxide in patients with acute respiratory distress syndrome: results of a phase II trial. Inhaled nitric oxide study group. Crit Care Med. 1998;26:15-23.
  28. Morris AH, Wallace CJ, Menlove RL, et al. Randomized clinical trial of pressure controlled inverse ratio ventilation and extracorporeal co2 removal for adult respiratory distress syndrome. Am J Respir Crit Care Med. 1994;149:295-305.