Abstract
Rationale: Tidal volume and plateau pressure limitation decreases mortality in acute respiratory distress syndrome. Computed tomography demonstrated a small, normally aerated compartment on the top of poorly aerated and nonaerated compartments that may be hyperinflated by tidal inflation.
Objectives: We hypothesized that despite tidal volume and plateau pressure limitation, patients with a larger nonaerated compartment are exposed to tidal hyperinflation of the normally aerated compartment.
Measurements and Main Results: Pulmonary computed tomography at end-expiration and end-inspiration was obtained in 30 patients ventilated with a low tidal volume (6 ml/kg predicted body weight). Cluster analysis identified 20 patients in whom tidal inflation occurred largely in the normally aerated compartment (69.9 ± 6.9%; “more protected”), and 10 patients in whom tidal inflation occurred largely within the hyperinflated compartments (63.0 ± 12.7%; “less protected”). The nonaerated compartment was smaller and the normally aerated compartment was larger in the more protected patients than in the less protected patients (p = 0.01). Pulmonary cytokines were lower in the more protected patients than in the less protected patients (p < 0.05). Ventilator-free days were 7 ± 8 and 1 ± 2 d in the more protected and less protected patients, respectively (p = 0.01). Plateau pressure ranged between 25 and 26 cm H2O in the more protected patients and between 28 and 30 cm H2O in the less protected patients (p = 0.006).
Conclusions: Limiting tidal volume to 6 ml/kg predicted body weight and plateau pressure to 30 cm H2O may not be sufficient in patients characterized by a larger nonaerated compartment.
Limiting tidal volume to 6 ml/kg and plateau pressure to 30 cm H2O protects the lungs of patients with acute respiratory distress syndrome from ventilator-induced lung injury (VILI).
Patients characterized by a larger amount of collapsed lung may be exposed to VILI despite tidal volume and pressure limitation; plateau pressure should be limited to 28 cm H2O to guarantee lung protection.
Analysis of computed tomography (CT) images of patients with ARDS has demonstrated a nonhomogeneous distribution of pulmonary alterations grouped into four patterns: hyperinflated, normally aerated, poorly aerated, and nonaerated compartments interspersed and/or distributed along the ventral–dorsal axis (5–7). The normally aerated compartment is relatively small but receives the largest part of the tidal volume (5, 6) and may therefore be exposed to excessive alveolar wall tension and stress failure (8); the nonaerated compartment can be reaerated during ventilation and the tidal reaeration of alveoli adjacent to fully expanded and consolidated regions may therefore cause shear stress (8). Nieszkowska and coworkers (9) found that in 14 of 32 patients with ARDS, prevention of expiratory derecruitment with 15 cm H2O of positive end-expiratory pressure (PEEP) was obtained at the price of hyperinflation of the normally aerated compartment. More recently, Gattinoni and coworkers provided direct visual evidence that patients with greater nonaerated and smaller normally aerated compartments had a worse outcome than did patients with smaller nonaerated and greater normally aerated compartments (10).
The present study set out to examine the hypothesis that patients characterized by a CT scan distribution of pulmonary lesions with a large dependent nonaerated compartment and a small nondependent normally aerated lung compartment may be exposed to tidal hyperinflation despite the use of the ARDSnet protective ventilatory strategy.
Some of the results of these studies have been previously reported in the form of an abstract (11).
Inclusion criteria were as follows: age, 18 yr or more, and diagnosis of ARDS (3, 12). Exclusion criteria were as follows: more than 3 d elapsed since ARDS criteria were met and mechanical ventilation was initiated; pulmonary artery occlusion pressure exceeding 18 mm Hg, if measured; history of ventricular fibrillation or tachyarrhythmia, unstable angina, or myocardial infarction within the preceding month; preexisting chronic obstructive pulmonary disease (3); major chest wall abnormalities (2); chest tube with persistent air leak; abdominal distension (2); body mass index greater than 30; pregnancy; known intracranial abnormality; and/or enrollment in another interventional study (3, 12). The institutional review board approved the study (10).
Patients were ventilated according to the ARDSnet protective ventilatory strategy (3, 12). As soon as targets of the ventilatory protocols were reached and physiologic parameters were stable (10), patients were transferred to the CT scan facility. Lung scanning was performed from apex to base at end-expiratory and end-inspiratory occlusions (8, 10). During transport and the examination, ventilator settings and the ventilator itself were those used for clinical management; particular attention was paid to avoid ventilator disconnection.
The CT scanner was set as previously described (8, 10): nonaerated (between +100 and −100 Hounsfield units [HU]), poorly aerated (between −101 and −500 HU), normally aerated (between −501 and −900 HU), and hyperinflated (between −901 and −1,000 HU) lung compartments were identified (6, 13). The volume of each compartment (i.e., the sum of gas plus tissue volume) for each slice, as well as the volume of the entire lung, was measured at end-expiration and end-inspiration (8, 10).
Protected tidal inflation and tidal hyperinflation were defined as the volume of the normally aerated and hyperinflated compartment at end-inspiration minus the volume of the normally aerated and hyperinflated compartments at end-expiration, respectively. Tidal recruitment of the nonaerated compartment and tidal recruitment of the poorly aerated compartment were defined as the volume of the nonaerated and poorly aerated compartments at end-expiration minus the volume at end-inspiration. All were expressed as a percentage of the total tidal inflation–related change in CT lung volume (8).
Weight of the entire lung and of each compartment at end-inspiration was measured (8, 10).
Five to 10 min after CT measurements, bronchoalveolar lavage fluid was collected in the CT suite and stored as previously described (2). Tumor necrosis factor-α soluble receptor (TNF-αsR)55 and TNF-αsR75, interleukin (IL)-6, IL-8, and IL-1β, and IL-1 receptor antagonist (IL-1Ra) were measured (2).
The number of ventilator-free days and the number of patients alive 28 d immediately after study entry were calculated (3, 12).
Values are given as means ± SD. Cluster analysis and the cubic clustering criterion were used to identify the maximal degree of association between patients and protected tidal inflation, tidal hyperinflation, tidal recruitment of the nonaerated compartment, and tidal recruitment of the poorly aerated compartment. Cluster analysis entails grouping similar objects into distinct, mutually exclusive subsets referred to as clusters; elements within a cluster share a high degree of natural association, whereas the clusters are relatively distinct from one another (14–16). The two-tailed t test, Mann-Whitney U test, χ2 test, and Fisher exact test were used. Multivariate stepwise regression analysis with backward elimination was used to determine whether differences in cytokines concentration and number of ventilator-free days between clusters were related to disease severity and/or to VILI. Dependent variables (clinical [4] and CT [6, 13] markers of disease severity and CT markers of VILI [8]) were entered in the regression if significantly different between clusters (4) (SAS Institute, Cary, NC).
Clinical characteristics of the study subjects are shown in Table 1. Ventilator-free days and mortality rate at 28 d were 5 ± 7 d and 33%, respectively. The time between onset of acute lung injury/ARDS and the study varied from 1 to 3 d.
Overall Population | More Protected Subpopulation | Less Protected Subpopulation | p Value | |
|---|---|---|---|---|
| Demographics | ||||
| Age, yr | 66.2 ± 11.2 | 65.8 ± 10.2 | 67.2 ± 13.5 | NS |
| Male/female | 19/11 | 13/7 | 6/4 | NS |
| SAPS II | 43.6 ± 16.2 | 43.5 ± 15.5 | 44.2 ± 19.4 | NS |
| Arterial blood gases | ||||
| PaO2:FiO2, mm Hg | 134 ± 38 | 149 ± 34 | 102 ± 24 | 0.009 |
| PaO2, mm Hg | 80.5 ± 13.0 | 80.8 ± 13.6 | 80.0 ± 12.4 | NS |
| PaCO2, mm Hg | 42.9 ± 3.1 | 43.1 ± 3.7 | 42.4 ± 0.7 | NS |
| pH | 7.41 ± 0.02 | 7.40 ± 0.03 | 7.41 ± 0.01 | NS |
| Causes of lung injury | ||||
| Pneumonia, no. (%) | 14 (47) | 9 (45) | 5 (50) | NS |
| Sepsis, no. (%) | 14 (47) | 10 (50) | 4 (40) | NS |
| Trauma, no. (%) | 2 (6) | 1 (5) | 1 (10) | NS |
| Ventilatory variables | ||||
| Tidal volume, ml/kg predicted body weight | 6.0 ± 0.3 | 6.0 ± 0.3 | 6.0 ± 0.3 | NS |
| Plateau pressure, cm H2O | 26.6 ± 1.8 | 25.5 ± 0.5 | 28.9 ± 0.9 | 0.006 |
| Static compliance, ml/cm H2O | 26.5 ± 5.9 | 26.2 ± 6.1 | 26.4 ± 6.3 | |
| Respiratory rate, breaths/min | 25 ± 5 | 22 ± 4 | 30 ± 4 | 0.008 |
| Minute ventilation, L/min | 10.5 ± 2.8 | 9.2 ± 2.1 | 12.7 ± 2.7 | 0.001 |
| FiO2 | 0.64 ± 0.16 | 0.56 ± 0.11 | 0.80 ± 0.11 | 0.0008 |
| PEEP, cm H2O | 10.4 ± 2.8 | 9.3 ± 2.3 | 12.6 ± 2.5 | 0.005 |
| Intrinsic PEEP,* cm H2O | 3.7 ± 2.3 | 3.9 ± 2.3 | 3.5 ± 2.0 | NS |
Amount of protected tidal inflation and tidal hyperinflation best discriminated two clusters of patients (R2 = 0.73). In a cluster of 20 patients protected tidal inflation and tidal hyperinflation represented 69.9 ± 6.9 and 8.1 ± 5.4% of the total tidal inflation–associated change in CT lung compartments, respectively (more protected). In a second cluster of 10 patients, protected tidal inflation and tidal hyperinflation represented 23.1 ± 14.4 and 63.0 ± 12.7% of the total tidal inflation–associated change in CT lung compartments, respectively (less protected) (Figure 1). Tidal recruitment of the poorly aerated compartment and tidal recruitment of the nonaerated compartment were 12.6 ± 4.7 and 9.3 ± 5.7% and 6.7 ± 4.3 and 7.1 ± 6.1% of the total tidal change in CT lung compartments in the more protected and less protected patients, respectively.
Figure 1. Amount of tidal hyperinflation and protected tidal inflation in the two clusters of patients. Open circles, more protected patients; solid circles, less protected patients. CT = computed tomography.
[More] [Minimize]Representative CT slices of the lung, obtained 2 cm above the dome of the diaphragm at end-expiration and end-inspiration, are shown for a more protected patient (Figure 2A, left) and a less protected patient (Figure 2B, left). Lung density histograms of tidal inflation–related changes in CT lung compartments in the more protected patient (Figure 2A, right) show an increase in volume in the normally aerated compartment, with a peak at −810 HU. In the less protected patient, tidal inflation reduced volume in the normally aerated compartment, with an increased volume of the hyperinflated compartment, with a peak at −910 HU (Figure 2B, right).
Figure 2. Representative CT slices of the lung obtained 2 cm above the dome of the diaphragm at end-expiration and end-inspiration and average lung density histograms (right) of tidal changes in CT lung compartments in a representative patient of the more protected (A; left and right) and less protected (B; left and right) clusters. The images were digitally adjusted for contrast, using Adobe Photoshop 7.0 (Adobe Systems, Mountain View, CA); red, blue, yellow, and green are associated with Hounsfield units (HU) identifying the hyperinflated (between −901 and −1,000 HU), normally aerated (between −501 and −900 HU), poorly aerated (between −101 and −500 HU), and nonaerated (between +100 and −100 HU) compartments, respectively.
[More] [Minimize]With the exception of IL-1Ra, bronchoalveolar lavage fluid concentrations of IL-6, IL-1β, IL-8, and both TNF-α receptors were lower in more protected than in less protected patients (p < 0.05) (Figure 3). The number of ventilator-free days in more protected patients was higher (p = 0.01) than in less protected patients (7 ± 8 vs. 1 ± 2, respectively). Mortality rates 28 d from admission were 30 and 40% in more protected and less protected patients, respectively (p = 0.21). The amount of tidal hyperinflation correlated with the pulmonary concentration of all inflammatory cytokines (p < 0.01).
Figure 3. Box plot of bronchoalveolar lavage concentration of inflammatory cytokines in more protected (open boxes) and less protected (gray boxes). Horizontal lines indicate medians. The two ends of the rectangles represent the 25th to 75th percentiles, bars indicate the maximum and minimum values of the data set, and dots indicate outliers. TNF-αsR55 and TNF-αsR75 = tumor necrosis factor-α soluble receptors; IL-6 = interleukin 6; IL-1β = interleukin 1β; IL-1ra = IL-1 receptor antagonist.
[More] [Minimize]Clinical characteristics of the more protected and less protected subpopulations are shown in Table 1. Age, sex, Simplified Acute Physiological Score II, and underlying diseases responsible for ARDS did not differ between the two groups of patients; the PaO2:FiO2 ratio was higher in more protected than in less protected patients (p = 0.009). Plateau pressure (Pplat) in more protected patients ranged between 25 and 26 cm H2O and in less protected patients between 28 and 30 cm H2O (p = 0.006).
End-inspiratory weight and volume of the total lung and of the various CT lung compartments in the overall population and in the more protected and less protected subpopulations are shown in Table 2. Lungs were heavier in less protected than in more protected patients (p = 0.008); weight and volume of the hyperinflated and nonaerated CT lung compartment were higher, and those of the normally aerated compartment were lower in less protected than in more protected patients (p < 0.05).
Overall Population | More Protected Subpopulation | Less Protected Subpopulation | p Value | |
|---|---|---|---|---|
| End-inspiratory total lung weight, g | 1,788 ± 325 | 1,541 ± 386 | 1,912 ± 206 | 0.008 |
| End-inspiratory total lung volume, ml | 3,795 ± 1,184 | 3,776 ± 1,225 | 3,836 ± 1,159 | NS |
| CT lung compartments | ||||
| Percent total lung weight | ||||
| Nonaerated (between +100 and −100 HU) | 33.9 ± 15.9 | 26.2 ± 12.0 | 49.3 ± 10.9 | 0.01 |
| Poorly aerated (between −101 and −500 HU) | 20.3 ± 7.7 | 20.7 ± 8.3 | 19.4 ± 6.6 | NS |
| Normally aerated (between −501 and −900 HU) | 43.8 ± 15.6 | 52.5 ± 10.3 | 26.4 ± 7.3 | 0.01 |
| Hyperinflated (between −901 and −1,000 HU) | 2.1 ± 2.8 | 0.6 ± 1.3 | 5.0 ± 2.9 | 0.01 |
| Percent total lung volume | ||||
| Nonaerated (between +100 and −100 HU) | 17.7 ± 8.6 | 16.1 ± 7.7 | 27.1 ± 14.3 | 0.002 |
| Poorly aerated (between −101 and −500 HU) | 20.6 ± 7.6 | 12.7 ± 6.1 | 10.5 ± 4.9 | NS |
| Normally aerated (between −501 and −900 HU) | 50.9 ± 10.2 | 68.2 ± 11.3 | 39.1 ± 19.8 | 0.003 |
| Hyperinflated (between −901 and −1,000 HU) | 10.8 ± 10.6 | 3.0 ± 2.2 | 23.3 ± 10.1 | 0.01 |
An amount of tidal hyperinflation exceeding 40% of the tidal inflation–associated change in CT lung compartment identified the less protected patients and corresponded to a Pplat value of at least 28 cm H2O (Figure 4).
Figure 4. Relationship between amount of tidal hyperinflation and plateau pressure (Pplat) in more protected (open circles) and less protected (solid circles) patients.
[More] [Minimize]Dependent variables entered into the multivariate stepwise regression analysis included weight of the entire lung (6, 13), PaO2:FiO2 ratio and Pplat (4), and amount of protected tidal inflation and tidal hyperinflation (8). Tidal recruitment of the nonaerated compartment and that of the poorly aerated compartment were not included in the model because they were not selected as differentiating characteristics between clusters. Because total lung weight and Pplat correlate consistently, along with disease severity, with weights of nonaerated, normally aerated, and hyperinflated compartments (6, 13) and to minute ventilation (4), respectively, the latter were not included in the regression analysis although they differ in the two groups of patients.
Tidal hyperinflation was the only variable independently associated with concentration of IL-6 (p = 0.001), IL-1β (p = 0.0025), IL-8 (p = 0.001), TNF-αsR55 (p = 0.007), and TNF-αsR75 (p = 0.001), and number of ventilator-free days (p = 0.005).
The present study demonstrates that the ARDSnet strategy may not be protective of all patients with ARDS because (1) one-third of patients experienced substantial tidal hyperinflation with tidal volumes of 6 ml/kg PBW and Pplat lower than 30 cm H2O; in these patients the concentration of inflammatory mediators was higher and the number of ventilator-free days was lower than in the two-thirds of patients who experienced less (although not zero) tidal hyperinflation; and (2) values of Pplat lower than 28 cm H2O were associated with less tidal hyperinflation than values of Pplat ranging between 28 and 30 cm H2O (3). Although our data do not indicate that a “safe” limit of Pplat exists, values less than 28 cm H2O seem to be associated with the more protective ventilatory settings.
To interpret these results it is crucial to clarify whether the higher concentration of inflammatory mediators and the lower number of ventilator-free days seen in less protected patients are due simply to more severe underlying lung injury. Multivariate stepwise linear regression analysis with backward elimination showed that the amount of tidal hyperinflation was the only variable associated with cytokine concentration and number of ventilator-free days. We may therefore speculate that in patients with heavier lungs, a larger dependent nonaerated compartment, and a smaller nondependent normally aerated compartment (i.e., lungs characterized by a high “potential for recruitment” [10] and a small “baby lung” [17]) the ARDSnet protective ventilatory strategy does not fully protect the lungs from VILI because hyperinflation of the small “normal” lung may occur despite lowering Vt to 6 ml/kg PBW and limiting Pplat to 30 cm H2O.
Before discussion of these results, some considerations are required. First, although our analysis separated a first cluster of patients characterized by predominant protected tidal inflation from a second cluster of patients characterized by predominant tidal hyperinflation (Figure 1), these two clusters represent different ranges of a continuum because most patients with ARDS experience tidal overdistension in some regions whereas tidal recruitment and increased normal aeration occur simultaneously in other regions (13, 18). Second, the lack of a CT scan at zero end-expiratory pressure does not allow the identification of patients at risk of tidal hyperinflation because at the PEEP levels used in the present study, lung morphology is influenced by factors not related to the kind of lung injury, the most important being the potential for recruitment (10). Third, the CT scan thickness used in the present study was 5 mm. Such spatial resolution may result in significant underestimation of tidal hyperinflation because Vieira and coworkers demonstrated that higher spatial resolution (2 mm) may have provided more accurate measurement (19). Fourth, patient age ranged between 49 and 82 yr. As a consequence it is likely that some degree of hyperinflation was present in some patients because it is well known that lung emphysema is related to age (20). Fifth, tidal recruitment of nonaerated and poorly aerated compartments was relatively small in both clusters. This might indicate that with tidal volumes of 6 ml/kg PBW, pressure and volume excursions are small enough that tidal recruitment/derecruitment is not significant, that is, that low tidal volume ventilation, intended to reduce tidal hyperinflation, may also minimize tidal recruitment.
ARDS is morphologically characterized by the distribution of the loss of lung aeration along the vertical axis, with a small number of normal alveoli located in the nondependent lung and a large consolidated, nonaerated region located in the dependent lung (6, 10, 21–23). Analysis of pulmonary CT images of patients (10, 18, 24) and animals (8, 25, 26) with ARDS during mechanical ventilation has demonstrated that the normally aerated compartment may receive the largest part of each breath and may therefore be hyperinflated and exposed to excessive alveolar wall tension and stress failure. Insufficient levels of PEEP may cause tidal recruitment/derecruitment of parts of the consolidated region and may therefore expose these regions to shear stress (8). These events may lead to worsening of the pulmonary and systemic inflammatory response, distal organ dysfunction, and ultimately organ failure (27). The ARDSnet study demonstrated that a 22% reduction in mortality could be obtained by using a Vt of 6 ml/kg PBW instead of 12 ml/kg PBW. In that study, the mean Pplat on the first day was 25 ± 7 cm H2O in the 6-ml/kg group versus 33 ± 9 cm H2O in the 12-ml/kg group (3).
Controversy exists regarding the extent to which Vt and inspiratory airway pressures should be reduced to protect the lungs from VILI (28–31). Some investigators have recommended that inspiratory plateau pressures lower than 30 to 35 cm H2O may be considered safe and that further reductions in Vt and Pplat are without benefit (28, 29, 31). Hager and coworkers examined the data collected in the ARDSnet Study and found that mortality decreases as Pplat declined from high to low levels at all levels of Pplat (32). These data suggest that patients in the higher Vt group would have benefited from Vt reduction even if they already had Pplat < 30 cm H2O (32). In the present study all patients were ventilated according to the low Vt arm of the ARDSnet Study; pulmonary concentration of inflammatory cytokines was higher and number of ventilator-free days was smaller in patients who had Pplat ⩾ 28 cm H2O than in patients who had Pplat ⩽ 26 cm H2O. Prospective clinical trials are required to prove that ventilation at lower Pplat values would improve outcomes in those patients ventilated with Pplat ranging between 28 and 30 cm H2O.
Lung hyperinflation has been previously reported as resulting from mechanical ventilation with PEEP (7, 9, 19, 21). Nieszkowska and coworkers (9) found that in 32 patients with ARDS, expiratory derecruitment was prevented by maintaining a PEEP of 15 cm H2O. However, the “price” of this beneficial effect of PEEP in one-third of the patients was hyperinflation of the nondependent lung regions. These patients had a CT scan distribution of pulmonary lesions characterized by a large amount of nonaerated and poorly aerated lung distributed in the dependent regions and a small amount of normally aerated lung distributed in the nondependent regions (9), similar to those observed in our less protected patients. Under these circumstances it is likely that what has been repeatedly reported for PEEP-induced hyperinflation is also true for tidal inflation–related hyperinflation: patients with a focal loss of lung aeration at zero end-expiratory pressure are at higher risk of hyperinflation than are patients with a diffuse loss of lung aeration (7, 9, 19, 21).
Information regarding the effects of tidal inflation on hyperinflation of lung regions in patients with ARDS is limited. Crotti and coworkers (33) quantified the amount of hyperinflated lung in five patients with ARDS ventilated in pressure control mode at a Pplat of 30 cm H2O and at various levels of PEEP and found that hyperinflated lung tissue ranged between 1 and 5% of the whole lung tissue. In our study, the end-inspiratory weight of hyperinflated lung tissue ranged between 0.1 and 3.9% of total lung weight in the more protected pattern and between 1.5 and 8.7% of total lung weight in less protected pattern (p = 0.01).
In conclusion, the present results may confirm the notion that the best ventilatory strategy should be ideally adapted to the size of the aerated lung. The ARDSnet protocol limiting Vt to 6 ml/kg PBW and limiting Pplat to 30 cm H2O may therefore not be sufficient to minimize VILI in patients with ARDS whose disease process is characterized by a distribution of pulmonary lesions with a small, nondependent, normally aerated compartment and a large, dependent, nonaerated compartment.
| 1. | Ware LB, Matthay MA. The acute respiratory distress syndrome. N Engl J Med 2000;342:1334–1349. |
| 2. | Ranieri VM, Suter PM, Tortorella C, De Tullio R, Dayer JM, Brienza A, Bruno F, Slutsky AS. Effect of mechanical ventilation on inflammatory mediators in patients with acute respiratory distress syndrome: a randomized controlled trial. JAMA 1999;282:54–61. |
| 3. | Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000;342:1301–1308. |
| 4. | Sakr Y, Vincent JL, Reinhart K, Groeneveld J, Michalopoulos A, Sprung CL, Artigas A, Ranieri VM. High tidal volume and positive fluid balance are associated with worse outcome in acute lung injury. Chest 2005;128:3098–3108. |
| 5. | Vieira SR, Puybasset L, Lu Q, Richecoeur J, Cluzel P, Coriat P, Rouby JJ. A scanographic assessment of pulmonary morphology in acute lung injury: significance of the lower inflection point detected on the lung pressure–volume curve. Am J Respir Crit Care Med 1999;159:1612–1623. |
| 6. | Gattinoni L, Caironi P, Pelosi P, Goodman LR. What has computed tomography taught us about the acute respiratory distress syndrome? Am J Respir Crit Care Med 2001;164:1701–1711. |
| 7. | Vieira SR, Puybasset L, Richecoeur J, Lu Q, Cluzel P, Gusman PB, Coriat P, Rouby JJ. A lung computed tomographic assessment of positive end-expiratory pressure–induced lung overdistension. Am J Respir Crit Care Med 1998;158:1571–1577. |
| 8. | Grasso S, Terragni P, Mascia L, Fanelli V, Quintel M, Herrmann P, Hedenstierna G, Slutsky AS, Ranieri VM. Airway pressure–time curve profile (stress index) detects tidal recruitment/hyperinflation in experimental acute lung injury. Crit Care Med 2004;32:1018–1027. |
| 9. | Nieszkowska A, Lu Q, Vieira S, Elman M, Fetita C, Rouby JJ. Incidence and regional distribution of lung overinflation during mechanical ventilation with positive end-expiratory pressure. Crit Care Med 2004;32:1496–1503. |
| 10. | Gattinoni L, Caironi P, Cressoni M, Chiumello D, Ranieri VM, Quintel M, Russo S, Patroniti N, Cornejo R, Bugedo G. Lung recruitment in patients with the acute respiratory distress syndrome. N Engl J Med 2006;354:1775–1786. |
| 11. | Terragni P, Rosboch G, Tealdi A, Menaldo E, Corno E, Marchiaro G, Bonetto C, Davini O, Ranieri VM. CT scan evidence of tidal recruitment during NIH protective strategy ventilation in ARDS patients [abstract]. Intensive Care Med 2005;31:S74. |
| 12. | Brower RG, Lanken PN, MacIntyre N, Matthay MA, Morris A, Ancukiewicz M, Schoenfeld D, Thompson BT. Higher versus lower positive end-expiratory pressures in patients with the acute respiratory distress syndrome. N Engl J Med 2004;351:327–336. |
| 13. | Rouby JJ, Puybasset L, Nieszkowska A, Lu Q. Acute respiratory distress syndrome: lessons from computed tomography of the whole lung. Crit Care Med 2003;31:S285–S295. |
| 14. | Toraldo DM, Nicolardi G, De Nuccio F, Lorenzo R, Ambrosino N. Pattern of variables describing desaturator COPD patients, as revealed by cluster analysis. Chest 2005;128:3828–3837. |
| 15. | Clinton D, Button E, Norring C, Palmer R. Cluster analysis of key diagnostic variables from two independent samples of eating-disorder patients: evidence for a consistent pattern. Psychol Med 2004;34:1035–1045. |
| 16. | Jones DK, Dardis R, Ervine M, Horsfield MA, Jeffree M, Simmons A, Jarosz J, Strong AJ. Cluster analysis of diffusion tensor magnetic resonance images in human head injury. Neurosurgery 2000;47:306–313; discussion 313–314. |
| 17. | Gattinoni L, Pesenti A. The concept of “baby lung.” Intensive Care Med 2005;31:776–784. |
| 18. | Malbouisson LM, Muller JC, Constantin JM, Lu Q, Puybasset L, Rouby JJ. Computed tomography assessment of positive end-expiratory pressure–induced alveolar recruitment in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2001;163:1444–1450. |
| 19. | Vieira SR, Nieszkowska A, Lu Q, Elman M, Sartorius A, Rouby JJ. Low spatial resolution computed tomography underestimates lung overinflation resulting from positive pressure ventilation. Crit Care Med 2005;33:741–749. |
| 20. | Madani A, Zanen J, de Maertelaer V, Gevenois PA. Pulmonary emphysema: objective quantification at multi-detector row CT: comparison with macroscopic and microscopic morphometry. Radiology 2006;238:1036–1043. |
| 21. | Rouby JJ. A lung computed tomographic assessment of positive end-expiratory pressure–induced lung overdistension. Am J Respir Crit Care Med 2000;161:1396–1397. |
| 22. | Rouby JJ, Lu Q, Goldstein I. Selecting the right level of positive end-expiratory pressure in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2002;165:1182–1186. |
| 23. | Rouby JJ, Puybasset L, Cluzel P, Richecoeur J, Lu Q, Grenier P; CT Scan ARDS Study Group. Regional distribution of gas and tissue in acute respiratory distress syndrome. II. Physiological correlations and definition of an ARDS Severity Score. Intensive Care Med 2000;26:1046–1056. |
| 24. | Gattinoni L, Pesenti A, Avalli L, Rossi F, Bombino M. Pressure–volume curve of total respiratory system in acute respiratory failure: Computed Tomographic Scan Study. Am Rev Respir Dis 1987;136:730–736. |
| 25. | Zinserling J, Wrigge H, Neumann P, Muders T, Magnusson A, Hedenstierna G, Putensen C. Methodologic aspects of attenuation distributions from static and dynamic thoracic CT techniques in experimental acute lung injury. Chest 2005;128:2963–2970. |
| 26. | Rylander C, Hogman M, Perchiazzi G, Magnusson A, Hedenstierna G. Oleic acid lung injury: a morphometric analysis using computed tomography. Acta Anaesthesiol Scand 2004;48:1123–1129. |
| 27. | Ranieri VM, Giunta F, Suter PM, Slutsky AS. Mechanical ventilation as a mediator of multisystem organ failure in acute respiratory distress syndrome. JAMA 2000;284:43–44. |
| 28. | Eichacker PQ, Gerstenberger EP, Banks SM, Cui X, Natanson C. Meta-analysis of acute lung injury and acute respiratory distress syndrome trials testing low tidal volumes. Am J Respir Crit Care Med 2002;166:1510–1514. |
| 29. | Tobin MJ. Culmination of an era in research on the acute respiratory distress syndrome. N Engl J Med 2000;342:1360–1361. |
| 30. | Ricard JD. Are we really reducing tidal volume–and should we? Am J Respir Crit Care Med 2003;167:1297–1298. |
| 31. | Dreyfuss D, Saumon G. Evidence-based medicine or fuzzy logic: what is best for ARDS management? Intensive Care Med 2002;28:230–234. |
| 32. | Hager DN, Krishnan JA, Hayden DL, Brower RG. Tidal volume reduction in patients with acute lung injury when plateau pressures are not high. Am J Respir Crit Care Med 2005;172:1241–1245. |
| 33. | Crotti S, Mascheroni D, Caironi P, Pelosi P, Ronzoni G, Mondino M, Marini JJ, Gattinoni L. Recruitment and derecruitment during acute respiratory failure: a clinical study. Am J Respir Crit Care Med 2001;164:131–140. |
