© 2003 American Thoracic Society
Right level of positive end-expiratory pressure in acute respiratory distress syndromeTo the Editor:In the April 2002 issue of AJRCCM, Rouby and coworkers (1) reported their opinion concerning the selection of the right positive end-expiratory pressure (PEEP) in patients with acute respiratory distress syndrome (ARDS). They tried to demonstrate that ventilation of ARDS requires high PEEP to optimize lung recruitment, to keep the lung fully open during tidal ventilation, and to limit FIO2. They reported that "the right PEEP level is the PEEP allowing the highest PaO2 and SaO2 at the lowest FIO2" which should be maintained below 0.6. In our opinion, this assertion is debatable. First of all, nothing in the literature indicates that an FIO2 above 0.6 could be deleterious in ARDS patients. Whereas oxygen toxicity has been clearly proved in healthy humans, high FIO2 seems to be well tolerated in ARDS patients, and it has been proved that mechanical ventilation with a high FIO2 for 96 hours at least was not deleterious in these patients (2). Use of high PEEP only to limit FIO2 therefore appears unjustified. Moreover, recent studies in ARDS have reported that circulatory failure was significantly associated with mortality, whereas hypoxemia was not (3). Second, to determine the ability of high PEEP to recruit without inducing lung overinflation, the authors propose first to evaluate the functional or morphological properties of lungs. When the presence of a clear lower inflexion point (LIP) on the inspiratory limb of the pressure–volume curve is demonstrated, or when diffuse loss of aeration in computed tomography scan is visualized, high PEEP could be applied without a risk of overinflation and hemodynamic deterioration. However, we have recently demonstrated that the presence of LIP on the inspiratory limb of pressure–volume curves only resulted from localized airflow limitation, and has nothing to do with recruitment (4). Moreover, diffuse loss of aeration is expected in patients in whom computed tomography scans were performed under FIO2 1, a procedure well known to induce atelectasis in previously aerated areas (5). Another surprising point of this report is the explanation given by the authors for loss of aeration of the lower lobes caused by cardiac dilatation. Whereas they excluded any increase in right ventricular size in a recent letter (6), they now admit that our explanation of cardiac dilatation was correct (7). Whereas acute cor pulmonale has become infrequent with protective ventilation (8), one may thus wonder about the responsibility of their respiratory strategy in the abnormal frequency of acute cor pulmonale observed in their patients. What exactly is the causal agent? Excessive PEEP? Excessive respiratory rate producing intrinsic PEEP? Or use of almitrine? But the most surprising thing in this critical care perspective is the decisional tree proposed in Figure 2, which finally leads to extra-corporeal membrane oxygenation. Really a perspective? Or a retrospective?
University Hospital Ambroise Paré Boulogne Cedex, France REFERENCES
From the Authors:Antoine Vieillard-Baron and François Jardin raise important questions that require nonpolemic responses.The main message of our critical care perspective (1) is to suggest that in many patients with acute respiratory distress syndrome (ARDS), keeping the lung fully open during tidal ventilation by using high positive end-expiratory pressure (PEEP) is potentially dangerous because it is frequently associated with overinflation of lung regions already open at zero end-expiratory pressure. When the loss of aeration is focally distributed, that is in more than 75% of patients fulfilling the ARDS criteria (2), we recommend performing a PEEP trial between 5 and 12 cm H2O. In such patients, the "right" PEEP can be considered as the PEEP between 5 and 12 cm H2O, allowing the highest PaO2 and SaO2 at the lowest FIO2. Only in the small minority of patients with diffuse loss of aeration, can higher PEEP levels corresponding to the concept of keeping the lung fully open be administered without introducing a risk of overinflation and ventilator-induced lung injury. In these patients, we recommend a PEEP trial between 10 and 25 cm H2O, the "right" PEEP level being the one allowing the highest PaO2 and SaO2 at the lowest FIO2. Short-term administration of FIO2 greater than 0.6 does not worsen pulmonary shunt and VA/Q ratio in patients with ARDS and cannot be the cause of a diffuse loss of aeration (3). In contrast, several articles concerning ARDS patients suggest that long-term administration of FIO2 greater than 0.6 can be deleterious to injured lungs. The article quoted by Vieillard-Baron and Jardin (4) shows that an exposure to an FIO2 greater than 0.9 for 4 days is significantly associated with death. Due to the retrospective nature of the study, the authors could not conclude whether high FIO2 was a contributing factor to death or a marker of severity. Three pathologic studies performed in ARDS patients have shown extensive lung lesions possibly related to oxygen toxicity: nonspecific alveolar damage (5), emphysema-like lesions (6), and lung fibrosis (7). The majority of critically ill patients fulfilling the ARDS criteria have 30 to 50% of their lungs fully open at zero end-expiratory pressure (2, 8) and are thereby, potentially exposed to oxygen toxicity. Therefore, the principle of precaution strongly suggests avoiding the use of FIO2 greater than 0.6 for prolonged periods of time. The selection of a PEEP level based on a good compromise between lung reopening and overinflation may prevent oxygen-induced worsening of lung damage by allowing the reduction of FIO2 below 0.6. The presence of a lower inflexion point on the inspiratory limb of the total respiratory pressure-volume curve cannot be reduced to a single mechanism. In addition to the presence of a slow compartment producing intrinsic PEEP (9), the role of lung morphology (10), chest wall (11), and opening pressures of edematous and collapsed lung regions has been clearly evidenced. It seems quite hazardous to claim that the presence of a lower inflexion point has nothing to do with lung morphology and alveolar recruitment by referring to a study in which neither lung morphology nor alveolar recruitment was measured (9). Similarly, the increased cardiac weight and transversal dimensions observed in ARDS patients cannot be reduced to a single mechanism (12). Right ventricular dilation can be present (13) or absent (14) depending on the respective importance of PEEP-induced increase in right ventricle afterload and decrease in right ventricle preload. In our response to one of Dr. Jardin's letters (15), we did not find significant correlations between the increase in cardiac mass and the increase in pulmonary artery pressure or pulmonary vasular resistance. However, right ventricle dilation could not be excluded simply because right ventricle volume was not directly measured. Several other mecanisms such as myocardial edema and elevated cardiac index are likely involved in the increase in cardiac mass observed in ARDS patients with septic shock (12). In ARDS patients with very stiff lungs, it has been shown that respiratory frequency can be increased to rates as high as 30 bpm without generating intrinsic PEEP (PEEPi) (16). As a consequence, if respiratory rate is increased without producing PEEPi—the highest respiratory rate preceding the occurrence of PEEPi can be easily determined by monitoring the expiratory flow on the screen of the ventilator—there is no reason why hemodynamics should deteriorate. Along these lines, it is well known that intravenous almitrine at doses of 2–4 mcg·kg-1·min-1 does not increase pulmonary artery pressure and pulmonary vascular resistance in ARDS patients with pulmonary hypertension (17). Therefore, almitrine-induced acute cor pulmonale is unlikely to occur if adequate doses are administered. A very small number of patients remain severely hypoxemic despite PEEP, inhaled NO, almitrine, prone position, recruitment maneuver, and fluid depletion. Extracorporeal membane oxygenation is an invasive rescue therapy that may keep them alive. When all previous therapies have failed, it offers the last avenue for survival in ARDS patients with refractory and life-threatening hypoxemia.
Hospital Pitié-Salpétrière, Université Paris VI Paris, France REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
