Lung Opening and Closing during Ventilation of Acute

Respiratory Distress Syndrome

Pietro Caironi1,2, Massimo Cressoni1, Davide Chiumello2, Marco Ranieri3, Michael Quintel4, Sebastiano G. Russo4,
Rodrigo Cornejo5, Guillermo Bugedo5, Eleonora Carlesso1, Riccarda Russo2, Luisa Caspani2, and
Luciano Gattinoni1,2
1
Dipartimento di Anestesiologia, Terapia Intensiva, e Scienze Dermatologiche, Fondazione Istituto Di Ricovero e Cura a Carattere Scientifico–
Ospedale Maggiore Policlinico, Mangiagalli, Regina Elena di Milano, Università degli Studi di Milano, Milan, 2Dipartimento di Anestesia,
Rianimazione e Terapia del Dolore, Fondazione Istituto Di Ricovero e Cura a Carattere Scientifico–Ospedale Maggiore Policlinico, Mangiagalli,
Regina Elena di Milano, Milan, and 3Dipartimento di Anestesia, Azienda Ospedaliera San Giovanni Battista-Molinette, Università degli Studi di
Torino, Turin, Italy; 4Anaesthesiologie II, Operative Intensivmedizin, Universitatsklinikum Göttingen, Göttigen, Germany; and 5Departementos de
Anestesiologia y Medicina Intensiva, Facultad de Medicina, Pontificia Universidad Catolica de Chile, Santiago, Chile

Rationale: The effects of high positive end-expiratory pressure (PEEP)

strictly depend on lung recruitability, which varies widely during AT A GLANCE COMMENTARY
acute respiratory distress syndrome (ARDS). Unfortunately, increas-
ing PEEP may lead to opposing effects on two main factors poten- Scientific Knowledge on the Subject
tially worsening the lung injury, that is, alveolar strain and intratidal The response to high levels of positive end-expiratory
opening and closing, being detrimental (increasing the former) or pressure (PEEP) depends on lung recruitability, which
beneficial (decreasing the latter). varies among patients with acute respiratory distress syn-
Objectives: To investigate how lung recruitability influences alveolar
drome. Its application appears to be appropriate only in
strain and intratidal opening and closing after the application of high
patients with a large amount of potentially recruitable lung.
PEEP.
Unfortunately, the increase in PEEP may lead to opposing
Methods: We analyzed data from a database of 68 patients with acute
lung injury or ARDS who underwent whole-lung computed tomo-
effects on the two major determinants of the development
graphy at 5, 15, and 45 cm H2O airway pressure. of ventilator-induced lung injury, that is, alveolar strain and
Measurements and Main Results: End-inspiratory nonaerated lung intratidal lung opening and closing.
tissue was estimated from computed tomography pressure–volume
curves. Alveolar strain and opening and closing lung tissue were
What This Study Adds to the Field
computed at 5 and 15 cm H2O PEEP. In patients with a higher per-
centage of potentially recruitable lung, the increase in PEEP markedly In patients with greater lung recruitability, the beneficial
reduced opening and closing lung tissue (P , 0.001), whereas no impact of reducing the amount of opening and closing lung
differences were observed in patients with a lower percentage of tissue by increasing PEEP appears to prevail over the
potentially recruitable lung. In contrast, alveolar strain similarly potentially harmful effects of increasing alveolar strain.
increased in the two groups (P 5 0.89). Opening and closing lung
tissue was distributed mainly in the dependent and hilar lung regions,
and it appeared to be an independent risk factor for death (odds ratio,
1.10 for each 10-g increase). main pathological features of the ARDS lung. Consequently, the
Conclusions: In ARDS, especially in patients with higher lung recruit- ventilatory approach to this syndrome has included, since the
ability, the beneficial impact of reducing intratidal alveolar opening beginning, the application of positive pressure at end expiration
and closing by increasing PEEP prevails over the effects of increasing to avoid lung collapse (1, 3, 4). We have observed that the
alveolar strain. response to high-level positive end-expiratory pressure (PEEP),
Keywords: acute respiratory distress syndrome; acute lung injury;
that is, the amount of lung collapse prevented by its application, is
ventilator-induced lung injury; mechanical ventilation strictly dependent on the total amount of lung recruitability (as
determined at airway pressures close to total lung capacity) and is
Since the first investigations on patients affected by acute re- therefore greatly variable among patients with ARDS (5).
spiratory distress syndrome (ARDS), lung recruitment has been Therefore, clinical evaluation of the potentially recruitable lung
shown to be critical when approaching the respiratory bundle of seems to be essential when approaching how to set the optimal
its clinical treatment (1, 2), as pulmonary atelectasis is one of the level of PEEP.
During ARDS or acute lung injury (ALI), the use of positive-
pressure ventilation may further increase the lung injury un-
derlying the respiratory insufficiency (6–8). There are two main
(Received in original form May 26, 2009; accepted in final form November 11, 2009) factors widely recognized as mechanical determinants of venti-
Supported in part by an Italian grant provided by the Fondazione Fiera di Milano lator-induced lung injury (VILI): alveolar strain, defined as the
for Translational and Competitive Research (2007, L.G.). ratio between the amount of gas volume delivered during tidal
Correspondence and requests for reprints should be addressed to Luciano Gattinoni, breath and the amount of aerated lung receiving it (9), and
M.D., Ph.D., Dipartimento di Anestesiologia, Terapia Intensiva, e Scienze Dermatolo- intratidal alveolar opening and closing, quantitatively defined as
giche, Fondazione Istituto Di Ricovero e Cura a Carattere Scientifico - Ospedale the amount of collapsed lung tissue reopening during inspiration
Maggiore Policlinico, Mangiagalli, Regina Elena di Milano, Università degli Studi di
and recollapsing during expiration (10–12). When applying a high
Milano, Via F. Sforza 35, 20122 Milan, Italy. E-mail: [email protected]
level of PEEP to ALI/ARDS lungs, two opposing phenomena
This article has an online supplement, which is accessible from this issue’s table of
theoretically occur: on the one hand, the amount of opening
contents at www.atsjournals.org
and closing lung tissue will decrease; on the other hand, as a
Am J Respir Crit Care Med Vol 181. pp 578–586, 2010
Originally Published in Press as DOI: 10.1164/rccm.200905-0787OC on November 12, 2009 consequence of the enhanced distending pressure, the degree
Internet address: www.atsjournals.org of alveolar strain applied to the open lung will increase. It is
Caironi, Cressoni, Chiumello, et al.: Lung Opening and Closing in ARDS 579

b
Figure 1. (A) Nonaerated lung tissue at end-inspiration and end-
expiration and (B) end-inspiratory alveolar strain for patients with either
a lower or a higher percentage of potentially recruitable lung after the
application of 5 and 15 cm H2O positive end-expiratory pressure (PEEP).
For clarity, data are expressed as means 6 standard error. To describe the
relationship between nonaerated lung tissue and the airway pressure
applied, exponential decay functions were used [y 5 y0 1 a 3 exp(–b 3
x)]. Alveolar strain was defined as the ratio between end-inspiratory lung
inflation and lung resting volume. For more details, see the online
supplement. In patients with a higher percentage of potentially recruit-
able lung, the increase in PEEP to 15 cm H2O markedly reduced the
difference between end-expiratory and end-inspiratory nonaerated lung
tissue, that is, the amount of opening and closing lung tissue (P , 0.001,
dotted lines). In contrast, no effect was observed in patients with a lower
percentage of potentially recruitable lung, as the amount of opening and
closing lung tissue already equaled a negligible amount at 5 cm H2O PEEP
(*P , 0.001 vs. patients with a lower percentage of potentially recruitable
lung at the same PEEP, #P , 0.001 vs. 5 cm H2O PEEP within the same
group). At 5 cm H2O PEEP, the alveolar strain of patients with a higher
percentage of potentially recruitable lung was significantly greater than
that observed in patients with a lower percentage of potentially recruit-
able lung. In contrast, the application of 15 cm H2O PEEP similarly
increased alveolar strain both in patients with either a lower or higher
percentage of potentially recruitable lung (P , 0.001).

METHODS
We analyzed data from a database of a multicenter study investigating
lung recruitment during ALI/ARDS (5). The study design is briefly
summarized.

Study Population
Sixty-eight patients were studied at four university hospitals, after
approval by the local institutional review boards had been given.
Informed consent was obtained according to the national regulations of
each institution. Patient enrollment was based on standard criteria for
ALI/ARDS (21), excluding those with age less than 16 years, preg-
nancy, and chronic obstructive pulmonary disease.

Study Design
Under sedation and paralysis, patients underwent whole-lung com-
puted tomography (CT) scanning at three different airway pressures:
45 cm H2O end-inspiratory airway pressure, and 5 and 15 cm H2O
PEEP applied in random order. Baseline mechanical ventilation was
set according to the daily clinical treatment. Before each CT scanning,
a recruitment maneuver was performed (see additional METHODS in the
online supplement) (5).
therefore conceivable that the potential effectiveness of high CT Scan Analysis
levels of PEEP will depend on the balance between deleterious
and beneficial effects (13). Each cross-sectional image was processed and analyzed with a custom-
designed software package (5, 22). Assuming the specific lung weight to
So far, no clinical data are available to support the application
be equal to 1, total lung weight and tissue weights of lung regions with
of higher levels of PEEP to improve survival in unselected different degree of aeration were calculated on the basis of the ‘‘CT
patients with ALI/ARDS (14, 15). After the conclusion of the number’’ of each voxel (5). To investigate lung morphology along the
ALVEOLI (Assessment of Low Tidal Volume and Elevated sterno–vertebral and cephalo–caudal axes, CT images of the whole
End-expiratory Volume to Obviate Lung Injury) study (16), two lung were divided into 4 equal segments from sternum to vertebral
further multicenter randomized clinical trials apparently did not region, and then arbitrarily grouped into 10 equal intervals, from lung
demonstrate any benefit for the application of higher levels of apex to base (23).
PEEP (as compared with lower levels) during low tidal volume
ventilation (17, 18). However, no study has considered the impact CT-derived Variables
of potentially recruitable lung characterizing each patient in 1. Potentially recruitable lung was defined as the difference be-
relation to the effects of different levels of PEEP on the de- tween the weight of nonaerated lung tissue at 5 and 45 cm H2O
airway pressure, expressed as a proportion of the weight of the
terminants of VILI (19). We therefore set out to investigate how
total lung tissue, and it represents the proportion of the total
lung recruitability may influence the development and the effects lung weight accounted for by nonaerated lung tissue in which
of alveolar strain as well as alveolar opening and closing after the aeration was restored from 5 to 45 cm H2O airway pressure. The
application of PEEP during ALI/ARDS. study population was divided into two groups, according to its
Some of the results of these studies have been previously median value (9% of the total lung tissue), as previously
reported in form of an abstract (20). reported (5).
580 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 181 2010

2. On the basis of the strict similarity between pressure–volume strain were analyzed at 5 and 15 cm H2O PEEP for the whole lung.
and pressure–recruitment curves during ALI/ARDS (24), a sig- At 5 cm H2O PEEP, opening and closing lung tissue recorded in
moid equation describing the pressure–volume curve of the patients with a higher percentage of potentially recruitable lung
respiratory system (25, 26) was derived from the four values of was markedly greater than that observed in patients with a lower
gas volume and airway pressure available for each patient (5 cm
percentage of potentially recruitable lung (141 6 85 vs. 20 6 22 g,
H2O PEEP, end-inspiratory plateau pressure from 5 and 15 cm
H2O PEEP, and end-inspiratory 45 cm H2O airway pressure). P , 0.001; Figure 1A). In the latter group of patients, its amount
Gas volume was expressed as a percentage of total lung capacity, was almost negligible, equaling only 2 6 2% of the total lung
defined as the gas volume inflated into the lungs at 45 cm H2O weight (1,266 6 327 g). When PEEP was increased to 15 cm H2O,
airway pressure. From each equation, end-inspiratory nonaer- opening and closing lung tissue significantly decreased only in
ated lung tissue at the corresponding plateau airway pressure patients with a higher percentage of potentially recruitable lung
was estimated as a percentage of the total amount of potentially (down to 63 6 87 g, P , 0.001), resulting unchanged in patients
recruitable lung of each patient (assuming 0 and 100% lung with a lower percentage of potentially recruitable lung (7 6 35 g,
recruitment, respectively, at 5 and 45 cm H2O airway pressure). P 5 0.23 vs. 5 cm H2O PEEP; Figure 1A).
3. The amount of opening and closing lung tissue at 5 and 15 cm After the application of a similar VT at 5 cm H2O PEEP, the
H2O PEEP was calculated as the difference between end- alveolar strain of patients with a higher percentage of potentially
expiratory and end-inspiratory nonaerated lung tissue, that is, recruitable lung was significantly greater than that observed in
the cycling reduction of nonaerated lung tissue caused every patients with a lower percentage of potentially recruitable lung
breath by VT application. (1.29 6 0.58 vs. 0.93 6 0.41, P , 0.001; Figure 1B), likely due to
4. Alveolar strain at 5 and 15 cm H2O PEEP was defined as the a lower estimated FRC recorded in the former group of patients
ratio between end-inspiratory lung inflation, that is, lung volume (P 5 0.005; Table 1). In contrast, after 15 cm H2O PEEP was
variation above the estimated volume at 0 cm H2O airway applied, alveolar strain similarly increased in both groups of
pressure (due to both VT and PEEP application), and the lung patients (up to 1.82 6 0.79 and 1.55 6 0.70, respectively, P 5 0.89;
resting volume, estimated as lung volume at 0 cm H2O airway
Figure 1B).
pressure, that is, FRC (27). The calculation was corrected for
PEEP-induced and intratidal lung recruitment (see additional Regional Distribution of Potentially Recruitable Lung
methods and Figure E1 in the online supplement).
To examine the possibility that inhomogeneity of the ALI/
5. Superimposed pressure was defined as the hydrostatic pressure ARDS lung may affect the distribution of the determinants of
at the dependent portion of the lung resulting from the weight of VILI within the lung parenchyma, we first investigated the
the tissue above.
regional distribution of potentially recruitable lung along the
Statistical Analysis cephalo–caudal axis.
In patients with a lower percentage of potentially recruitable
Statistical significance was defined as a P value less than 0.05. Unless lung, the total amount of nonaerated lung tissue at 5 cm H2O
otherwise indicated, data are expressed as means 6 SD.
PEEP progressively and linearly increased along the entire
Additional details on the methods used in this study are provided in
the online supplement. cephalo–caudal axis, from about 10% to about 50% of the total
lung tissue weight at each lung level (P , 0.001, one-way analysis
of variance [ANOVA]; Figure 2A). In contrast, the amount of
RESULTS nonaerated lung tissue potentially recruitable at 45 cm H2O
airway pressure, that is, the potentially recruitable lung, appeared
Intratidal Opening and Closing Lung Tissue and Alveolar
to be negligible throughout the entire parenchyma. Of note, the
Strain of Whole Lung
application of 45 cm H2O airway pressure determined, at lung
To initially investigate the effect of mechanical ventilation on the base (levels 8 to 10), an alveolar derecruitment, as shown by
determinants of VILI and their relationship with the potentially a negative percentage of potentially recruitable lung (P , 0.05 for
recruitable lung, opening and closing lung tissue and alveolar levels 9 and 10 vs. other lung levels; Figure 2A).

TABLE 1. VENTILATORY SETTINGS AND RESPIRATORY VARIABLES DURING THE STUDY PROTOCOL
Patients with Lower Patients with Higher
Percentage of Potentially Percentage of Potentially
Recruitable Lung (n 5 34) Recruitable Lung (n 5 34) P Value*

VT at 5 PEEP, ml 570 6 106 523 6 109 0.08

VT at 15 PEEP, ml 568 6 105 520 6 110 0.07
Plateau pressure from 5 PEEP, cm H2O† 18 6 3 21 6 4 ,0.001
Plateau pressure from 15 PEEP, cm H2O† 29 6 4 30 6 4 0.32
Respiratory rate at 5 PEEP, breaths/min 18 6 6 19 6 7 0.28
Respiratory rate at 15 PEEP, breaths/min 18 6 6 19 6 7 0.29
Minute ventilation at 5 PEEP, L/min 9.7 6 2.6 9.8 6 3.1 0.89
Minute ventilation at 15 PEEP, L/min 9.7 6 2.7 9.7 6 2.9 0.94
Intrinsic PEEP at 5 PEEP, cm H2O 0.9 6 1.6 0.7 6 1.2 0.44
Intrinsic PEEP at 15 PEEP, cm H2O 0.4 6 0.7 0.4 6 0.7 0.82
Total lung tissue weight, g† 1,266 6 327 1,735 6 547 ,0.001
Estimated FRC, ml 1,119 6 523 708 6 635 0.005

Definition of abbreviation: PEEP 5 positive end-expiratory pressure.

Patients with a lower percentage of potentially recruitable lung had values at or below 9% of the total lung tissue weight (the
median value for the study population), whereas patients with a higher percentage of potentially recruitable lung had values
greater than 9%. Plus-minus values represent means 6 SD.
* P values were obtained by Student t test or Wilcoxon test, as appropriate.
†
Data regarding these variables, which have been previously reported (5), are presented here for completeness.
Caironi, Cressoni, Chiumello, et al.: Lung Opening and Closing in ARDS 581

b
Figure 2. Regional distribution of consolidated and potentially recruit-
able lung for patients with either (A) a lower or (B) a higher percentage of
potentially recruitable lung, and regional distribution of superimposed
pressure at the most dependent lung regions (C ) for both groups of
patients at 5 cm H2O positive end-expiratory pressure (PEEP). For regional
analysis, lung parenchyma was equally divided into 10 intervals along the
cephalo–caudal axis. Consolidated lung tissue denotes the lung tissue that
remained nonaerated even after the application of 45 cm H2O airway
pressure, whereas the potentially recruitable lung denotes the amount of
nonaerated lung tissue at 5 cm H2O PEEP that has been recruited at 45 cm
H2O airway pressure. Therefore, the sum of both the consolidated and the
potentially recruitable lung denotes the total amount of nonaerated lung
tissue at 5 cm H2O PEEP for each lung interval. Both the consolidated and
the potentially recruitable lung were expressed as a percentage of the total
lung tissue weight of each cephalo–caudal lung interval. For clarity, data
are expressed as means 6 standard error. For statistical analysis, see the
text. In patients with a lower percentage of potentially recruitable lung, the
amount of consolidated lung tissue and the total amount of nonaerated
lung tissue progressively and linearly increased along the entire cephalo–
caudal axis, whereas the percentage of potentially recruitable lung
appeared to be negligible throughout the entire parenchyma. Conversely,
in patients with a higher percentage of potentially recruitable lung, the
total amount of potentially recruitable lung was homogeneously distrib-
uted along the cephalo–caudal axis, and the amount of nonaerated lung
tissue at 5 cm H2O PEEP equaled about 40 to 50% of the tissue weight of
each lung level, with the exception of a slight reduction observed in the
first levels. Finally, superimposed pressure at the most dependent lung
regions detected in patients with a higher percentage of potentially
recruitable lung was markedly greater than that observed in patients with
a lower percentage of potentially recruitable lung along the entire lung
parenchyma, with the exception of levels at lung base (*P , 0.005 vs.
patients with a lower percentage of potentially recruitable lung).

Regional Distribution of Alveolar Opening and Closing Tissue

and Alveolar Strain
As shown in Figure 3A, the amount of opening and closing lung
tissue in patients with a lower percentage of potentially recruit-
able lung was quite small along the entire cephalo–caudal axis,
both at 5 and 15 cm H2O PEEP. In contrast, in patients with a
higher percentage of potentially recruitable lung, the amount of
opening and closing lung tissue at 5 cm H2O PEEP was relatively
high, particularly at lung apex and hilum (from level 2 to level 7;
Figure 3B). The application of 15 cm H2O PEEP significantly
reduced opening and closing lung tissue (P , 0.05 vs. 5 cm H2O
Conversely, in patients with a higher percentage of potentially PEEP), with the exception of levels 8 to 10 (P 5 1.00).
recruitable lung, the amount of nonaerated lung tissue at 5 cm Alveolar strain at 5 cm H2O PEEP significantly decreased
H2O PEEP appeared to equal about 40–50% of the tissue weight toward lung bases in patients with a lower percentage of
of each lung level, with the exception of the first levels at lung potentially recruitable lung (P 5 0.002, two-way ANOVA for
apex, in which a slight reduction was observed (P 5 0.03, one-way repeated measurements; Figure E3 in the online supplement).
ANOVA). Similarly, the amount of the potentially recruitable In contrast, in patients with a higher percentage of potentially
lung was homogeneously distributed along the cephalo–caudal recruitable lung alveolar strain was homogeneously distributed
axis, equaling about 20–30% of the lung tissue weight of each lung along the entire cephalo–caudal axis (P 5 0.40, two-way ANOVA
level (P 5 0.14, one-way ANOVA; Figure 2B). for repeated measurements). After the application of 15 cm
As lung collapse may be related to the gravitational forces H2O PEEP, alveolar strain significantly increased in both groups
exerted by the increased tissue weight of the lung parenchyma (P , 0.001, two-way ANOVA for repeated measurements,
(28), we analyzed the distribution of superimposed pressure at for both). However, in patients with a higher percentage of
the most dependent lung regions along the cephalo–caudal axis potentially recruitable lung the increased strain appeared to be
in the two groups. Both in patients with either a lower or mainly localized only in specific apex and hilar lung regions (see
a higher percentage of potentially recruitable lung, superim- Figure E3).
posed pressure was relatively high along the entire cephalo–
caudal axis (Figure 2C). Moreover, in patients with a higher
percentage of potentially recruitable lung, superimposed pres- Impact on Survival
sure markedly exceeded that recorded in patients with a lower To assess the impact of potentially recruitable lung on the
percentage of potentially recruitable lung within the entire lung determinants of VILI in relation to the clinical outcome, we
parenchyma, with the exception of the last two levels at lung analyzed the amount of opening and closing lung tissue as well as
base (P 5 1.00, for both). alveolar strain based on the ventilatory parameters set before the
582 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 181 2010

potentially recruitable lung (254 6 78 g, P , 0.001, one-way

ANOVA; Figure 4). The progressive increase in opening and
closing lung tissue along the increased percentage of potentially
recruitable lung was paralleled by a linear increase in mortality at
intensive care unit discharge (P 5 0.34 by the Hosmer-Lemeshow
goodness-of-fit test; c 5 0.72), as previously reported (5).
To further examine a possible association between the de-
terminants of VILI and mortality, a multivariate analysis for
independent predictors of mortality was performed. The Simpli-
fied Acute Physiology Score (SAPS) II score, as well as opening
and closing lung tissue resulting from the ventilatory setting
clinically used, appeared to be independently associated with an
increased risk of death (P 5 0.86 by the Hosmer-Lemeshow
goodness-of-fit test; c 5 0.81); the odds ratios for each one-point
increase in SAPS II score and for each 10-g increase in opening
and closing lung tissue were, respectively, 1.10 (95% confidence
interval, 1.01–1.20) and 1.10 (95% confidence interval, 1.02–1.18).
Of note, alveolar strain did not differ between survivors and
nonsurvivors (Table 2). Finally, although the level of PEEP did
not appear to be independently associated with an increased risk
of death, patients with a very high percentage of potentially
recruitable lung who survived were clinically ventilated with a
PEEP level significantly higher than that applied in patients who
did not survive (14 6 4 vs. 10 6 2, P , 0.05; Figure 5).

DISCUSSION
During ALI/ARDS, when PEEP is increased, two different
phenomena may simultaneously occur: on the one hand, the
increase in pressure applied at end-expiration will determine
a greater inflation of the aerated portion of the lung, leading to
hyperinflation and an increase in alveolar strain; on the other
hand, the increase in PEEP will prevent a greater portion of the
lung from collapsing, thereby reducing the amount of lung tissue
undergoing intratidal and cycling opening and closing (27, 29).
We found that in patients with a lower percentage of potentially
recruitable lung the application of higher levels of PEEP did not
Figure 3. Regional distribution of opening and closing lung tissue for significantly affect the amount of opening and closing lung tissue,
patients with either (A) a lower or (B) a higher percentage of potentially as it already equaled, at lower PEEP, a negligible fraction of lung
recruitable lung at 5 and 15 cm H2O positive end-expiratory pressure tissue weight. In contrast, in patients with a higher percentage of
(PEEP). For regional analysis, lung parenchyma was equally divided into
potentially recruitable lung, opening and closing lung tissue was
10 intervals along the cephalo–caudal axis. Opening and closing lung
almost halved by the increase in PEEP from 5 to 15 cm H2O.
tissue was computed for each PEEP level as the difference between end-
expiratory and end-inspiratory nonaerated lung tissue. In patients with
Surprisingly, the increment of PEEP led to an identical increase in
a lower percentage of potentially recruitable lung, the amount of alveolar strain in the two groups of patients, despite the lower
opening and closing lung tissue was similar and almost negligible both estimated FRC detected in patients with a higher percentage of
at 5 and 15 cm H2O PEEP along the entire parenchyma. In contrast, in potentially recruitable lung. These findings suggest a great impact
patients with a higher percentage of potentially recruitable lung, the of the amount of intratidal lung recruitment on the effective end-
amount of opening and closing lung tissue was particularly high at lung inspiratory alveolar strain resulting from the ventilatory setting
apex and hilum and was significantly reduced, although not to a neg- applied (30). In fact, in patients with a higher percentage of
ligible amount, by the application of 15 cm H2O PEEP (*P , 0.05 vs. potentially recruitable lung, the ‘‘open’’ lung area potentially
5 cm H2O PEEP). receiving VT at the beginning of inspiration (i.e., FRC) is
relatively smaller. In contrast, at end-inspiration, such an area
will consist of the sum of the aerated regions at the beginning of
beginning of the study. For this purpose, the study population was inspiration and the nonaerated lung regions that have been
divided into quartiles according to the percentage of potentially recruited during the tidal breath (31) (i.e., the amount of opening
recruitable lung (5). As previously reported (5), VT and PEEP and closing lung tissue), which will increase the size of the initial
appeared to be identical between the four groups of patients ‘‘baby lung’’ receiving the delivered gas volume. Therefore, it is
(see Table E1 in the online supplement), equaling, respectively, conceivable that the greater the potentially recruitable lung, the
about 9 ml/kg predicted body weight and 11 cm H2O. Alveolar lower the actual increment of end-inspiratory alveolar strain
strain appeared to be constant in the first three quartiles of derived from the application of specific VT and PEEP (see the
potentially recruitable lung, whereas it significantly increased online supplement for details of modeling), thereby ‘‘protecting’’
only in patients with a very high percentage of potentially the lung parenchyma from PEEP-induced overdistention.
recruitable lung (Figure 4 and Table E1, P 5 0.03, one-way It has been previously observed that the amount of lung
ANOVA). In contrast, the amount of opening and closing lung tissue that remained nonaerated after the application of 45 cm
tissue linearly increased from patients with a very low percent- H2O airway pressure is quite constant in the ALI/ARDS pop-
age (21 6 12 g) to patients with a very high percentage of ulation (5) (z25% of the total lung tissue). We have therefore
Caironi, Cressoni, Chiumello, et al.: Lung Opening and Closing in ARDS 583

Figure 4. Alveolar strain and opening and

closing lung tissue as derived from the venti-
latory parameters clinically set before the
beginning of the study, as well as mortality
at intensive care unit (ICU) discharge in re-
lation to the potentially recruitable lung. The
study population was divided into quartiles of
17 patients each according to the percentage
of the potentially recruitable lung. Alveolar
strain was defined as the ratio between end-
inspiratory lung inflation and lung resting
volume, whereas opening and closing lung
tissue was computed as the difference be-
tween end-expiratory and end-inspiratory
nonaerated lung tissue. For more details, see
the online supplement. Alveolar strain result-
ing from the ventilatory setting clinically set
did not differ between the four groups of
patients, with the exception of a significant
increase in patients with a very high percent-
age of potentially recruitable lung. In contrast,
opening and closing lung tissue progressively
and linearly increased with the increase in the
percentage of the potentially recruitable lung,
and was paralleled by a linear increase in mor-
tality at ICU discharge (*P , 0.05 vs. other
quartiles of potentially recruitable lung).

hypothesized that the potentially recruitable lung may reflect and colleagues [32]), it is conceivable that the visual inspection
the extent of the inflammatory reaction surrounding the initial of lung imaging morphology may initially single out patients less
‘‘core disease,’’ that is, the unrecruitable or ‘‘consolidated’’ lung responsive to high levels of PEEP and at a greater risk of lung
tissue. The regional analysis here presented may provide an hyperinflation (37).
important confirmation of this hypothesis. In fact, whereas in When we regionally investigated alveolar strain at the two
patients with a higher percentage of potentially recruitable lung different levels of PEEP, a similar distribution was observed
the consolidated lung tissue was homogeneously distributed between patients with either a lower or higher percentage of
along the entire parenchyma, in patients with a lower percent- potentially recruitable lung. In contrast, patients with a higher
age of potentially recruitable lung it appeared to be located percentage of potentially recruitable lung showed a greater
mainly in the lung basal section, suggesting a ‘‘lobar’’ pattern of prevalence of opening and closing lung tissue at the dependent
its distribution (32, 33). Moreover, the derecruitment observed regions of lung apex and hilum, as compared with that observed at
in the latter group at the highest airway pressure applied (levels the lung base (see Figure E5 in the online supplement). These
8–10; Figure 2A), as compared with the moderate lung re- data suggest that, as a consequence of intratidal lung opening and
cruitment in patients with a higher percentage of potentially closing, the dependent areas of the hilar regions may represent
recruitable lung at similar values of superimposed pressure, may pulmonary areas more susceptible to the injury induced by
indicate different causes for the development of atelectasis in mechanical ventilation. In support of this hypothesis, CT scan
those lung regions (23, 34, 35). In fact, we may speculate that analysis of lung morphology demonstrated a higher prevalence of
the lobar ‘‘nature’’ of the collapsed lung regions in patients with bullae per lung in patients with late ARDS after long-term
a lower percentage of potentially recruitable lung determines mechanical ventilation, with predominance in the dependent
high values of threshold opening pressure in those lung regions, portion of lung hilum and base (38). Moreover, the histological
thereby leading to local alveolar compression when insufficient analysis at autopsy of a series of patients with ARDS revealed
opening pressures are applied (even 45 cm H2O), as previously a higher incidence of ‘‘bronchopneumonia’’ (defined as a general
observed (24, 36). Taken together, these findings support the inflammation) in the lung segments of posterior lobes (39).
hypothesis that patients with a lower amount of potentially Therefore, we may speculate that the intratidal lung opening
recruitable lung are those in whom the initial pathologic lesion and closing may create alveolar shear stresses located predomi-
remains anatomically and functionally compartmentalized. In nantly in the dependent portion of the lung, which may lead to
contrast, patients with a higher percentage of potentially pathological lesions (such as bullae and pseudocysts) that are not
recruitable lung are likely those in whom a more diffuse insult simply related to the development of alveolar overdistention,
affects the lung parenchyma, leading to the loss of compart- located predominantly in the nondependent lung regions (40).
mentalization, a generalized inflammation with edema forma- Because we have previously observed a linear increase in
tion, a greater increment of lung weight, and a greater and mortality rate at intensive care unit discharge associated with the
widespread alveolar collapse (24, 36). Moreover, these results increment of potentially recruitable lung (5), we analyzed in
could further suggest how the possible findings detectable both the study population the determinants of VILI derived from the
by CT scanning and chest X-ray may help to elucidate lung ventilatory settings clinically used before the beginning of the
functional morphology of patients with ALI/ARDS. In fact, on study. Despite similar values of VT and PEEP, the amount of
the basis of the more frequent recurrence of a ‘‘lobar’’ pattern opening and closing lung tissue linearly increased with the pro-
of lung atelectasis in patients with a lower percentage of po- gressive increment of potentially recruitable lung, as previously
tentially recruitable lung (as previously observed by Puybasset suggested by Grasso and colleagues (41). No major differences
584 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 181 2010

TABLE 2. BASELINE CLINICAL AND RESPIRATORY VARIABLES, VENTILATORY SETTINGS, AND LUNG
MORPHOLOGICAL CHARACTERISTICS OF SURVIVORS AND NONSURVIVORS
Survivors (n 5 49) Nonsurvivors (n 5 19) P Value*

Age, yr 53 6 17 59 6 16 0.26
SAPS II†‡ 34 6 10 42 6 11 0.008
VT, ml/kg predicted body weight 8.9 6 1.9 8.9 6 1.5 0.89
PEEP, cm H2O 12 6 3 10 6 2 0.08
Plateau airway pressure, cm H2Ox 25 6 4 25 6 4 0.86
Respiratory rate, breaths/min 18 6 6 19 6 8 0.37
Minute ventilation, L/min 9.7 6 3.0 10.0 6 3.0 0.76
PaO2/FIO2k 172 6 66 147 6 75 0.18
PaCO2, mm Hgk 40 6 8 45 6 10 0.03
Respiratory system compliance, ml/cm H2Ok{ 46 6 16 41 6 24 0.30
Dead space, % VT†k** 53 6 12 67 6 12 ,0.001
Shunt, % cardiac outputk†† 37 6 14 45 6 17 0.06
Total lung tissue weight, gk 1,446 6 489 1,640 6 534 0.16
Nonaerated lung tissue, % total lung weightk 34 6 16 45 6 15 0.02
Potentially recruitable lung, % total lung weight 10 6 10 19 6 13 0.005
Opening and closing lung tissue, gx 96 6 105 184 6 116 0.006
Alveolar strainx 1.52 6 0.80 1.74 6 0.95 0.35

Survival was calculated at intensive care unit discharge. Plus-minus values represent means 6 SD.
* P values were obtained by Student t test or Wilcoxon test, as appropriate.
†
Data regarding these variables, which have been previously reported (5), are presented here for completeness.
‡
The Simplified Acute Physiology Score (SAPS II) (48) was used to assess the severity of systemic illness at study entry. Scores
can range from 0 to 163, with higher values indicating more severe illness.
x
Values of plateau airway pressure resulting from the VT clinically applied, and consequently values of opening and closing lung
tissue and alveolar strain are available for 61 patients (43 survivors and 18 nonsurvivors).
k
Values of PaO2/FIO2, PaCO2, respiratory system compliance, dead space, shunt, total lung tissue weight, and nonaerated lung
tissue were obtained at 5 cm H2O PEEP.
{
Respiratory system compliance was calculated as the ratio of VT to the difference between inspiratory plateau pressure and
PEEP.
** Dead space was calculated by a standard formula, as previously reported (5), and was available for 48 patients (33 survivors
and 15 nonsurvivors).
††
The intrapulmonary right-to-left shunt was calculated by a standard formula, as previously reported (5), and was available for
60 patients (42 survivors and 18 nonsurvivors).

were observed regarding alveolar strain, with the exception of ventilatory setting clinically used. Moreover, these findings
patients with a very high percentage of potentially recruitable support the hypothesis that these patients may really benefit
lung. Moreover, opening and closing lung tissue appeared to be an from a level of PEEP greater than 10 cm H2O. In fact, if we
independent risk factor for death. On the basis of these findings, estimate the balance between the increase in alveolar strain and
although a greater severity of lung disease cannot be separated the associated decrease in opening and closing lung tissue for each
from lung recruitability, we may speculate that the higher cm H2O increase in PEEP, two considerations become evident:
mortality rate observed in patients with a higher potentially first, in patients with a lower percentage of potentially recruitable
recruitable lung may be due not only to the greater severity of lung the balance is entirely in disfavor of an increase in PEEP, as it
the underlying lung injury (as previously observed [5]), but also to will cause an excessive increase in alveolar strain and a negligible
the higher amount of opening and closing lung tissue caused by the reduction of opening and closing lung tissue (Figure E7 in the

Figure 5. Values of positive end-expiratory pressure (PEEP) clinically

applied before the beginning of the study, for both survivors and
nonsurvivors, as a function of different percentages of potentially
recruitable lung. The study population was divided into quartiles of
17 patients each according to the percentage of potentially recruit-
able lung. Mortality was computed as the number of deaths at
intensive care unit discharge. As shown, in the first three quartiles of
potentially recruitable lung, no difference was observed between
values of PEEP clinically used for either survivors or nonsurvivors. In
contrast, in patients with a very high percentage of potentially
recruitable lung, the PEEP clinically applied in survivors appears to
be significantly greater than that applied in nonsurvivors (P , 0.05).
Caironi, Cressoni, Chiumello, et al.: Lung Opening and Closing in ARDS 585

online supplement); second, in patients with a higher percentage relationship with a commercial entity that has an interest in the subject of this
manuscript. M.R. received $10,001–$50,000 from Eli Lilly in consultancy fees for
of potentially recruitable lung, the increase in PEEP will deter- PI Prowess Shock and $10,001–$50,000 from Maquet and $10,001–$50,000
mine an extensive reduction of opening and closing lung tissue, from Hemodec for serving on an advisory board. M.Q. does not have a financial
associated with a moderate enhancement of alveolar strain. relationship with a commercial entity that has an interest in the subject of this
At this point, we may ask ourselves: is there any clinical manuscript. S.G.R. does not have a financial relationship with a commercial
entity that has an interest in the subject of this manuscript. R.C. does not have
evidence to confirm the critical role of opening and closing lung a financial relationship with a commercial entity that has an interest in the subject
tissue in determining VILI and thereby affecting survival during of this manuscript. G.B. does not have a financial relationship with a commercial
ALI/ARDS? Although not straightforward, the answer may be entity that has an interest in the subject of this manuscript. E.C. does not have
a financial relationship with a commercial entity that has an interest in the subject
considered affirmative. Villar and colleagues (42) showed in of this manuscript. R.R. does not have a financial relationship with a commercial
patients with ALI/ARDS ventilated with low VT and high PEEP entity that has an interest in the subject of this manuscript. L.C. does not have
a lower mortality rate as compared with those ventilated with a financial relationship with a commercial entity that has an interest in the subject
of this manuscript. L.G. does not have a financial relationship with a commercial
a lower PEEP and a relatively higher VT. As the plateau airway entity that has an interest in the subject of this manuscript.
pressure on Day 1 (which may be considered an indirect signal
for alveolar strain) did not differ between the two groups, these Acknowledgment: The authors are indebted to Pietro Biondetti, M.D., Marco
Lazzarini, M.D., Benedetta Finamore, M.D., and Cristian Bonelli, of the Diparti-
results may indirectly suggest an impact of opening and closing mento di Radiologia, Fondazione IRCCS–Ospedale Maggiore Policlinico, Man-
lung tissue on the difference in survival observed. Similarly, giagalli, Regina Elena di Milano, Milan, Italy, for technical assistance with CT scan
Hager and colleagues (43), in a secondary analysis of the ARDS image analyses and to Milena Racagni, M.D., Laura Landi, M.D., Alice D’Adda,
M.D., Serena Azzari, M.D., Sonia Terragni, M.D., Federico Polli, M.D., Paola
Network database, highlighted the critical role of VT reduction Cozzi, M.D., Giuliana Motta, M.D., Federica Tallarini, M.D., Cristian Carsenzola,
for the improvement of survival regardless of the values of M.D., and Monica Chierichetti, M.D., of the Dipartimento di Anestesiologia,
plateau airway pressure derived, further suggesting a greater Terapia Intensiva, e Scienze Dermatologiche, Fondazione IRCCS–Ospedale
Maggiore Policlinico, Mangiagalli, Regina Elena di Milano, Università degli Studi
importance of opening and closing lung tissue in mortality as di Milano, Milan, Italy, for help in data analysis. The authors are also indebted to
compared with the overall alveolar strain. More recently, the two the study patients for their participation and to the physicians and nursing staff of
concluded and published clinical trials, the ExPress (Comparison the participating units for their valuable cooperation.
of Two Strategies for Setting Positive End-expiratory Pressure in
Acute Lung Injury/Acute Respiratory Distress Syndrome) study References
(18) and the LOVS (Lung Open Ventilation to Decrease
1. Ashbaugh DG, Bigelow DB, Petty TL, Levine BE. Acute respiratory
Mortality in the Acute Respiratory Distress Syndrome) study distress in adults. Lancet 1967;2:319–323.
(17), investigating the effects of low versus high levels of PEEP, 2. Fan E, Wilcox ME, Brower RG, Stewart TE, Mehta S, Lapinsky SE,
did actually demonstrate a lower progression of pulmonary Meade MO, Ferguson ND. Recruitment maneuvers for acute lung
damage after the application of higher PEEP (19). Finally, injury: a systematic review. Am J Respir Crit Care Med 2008;178:
Talmor and colleagues (44) have observed a trend toward a lower 1156–1163.
3. Falke KJ, Pontoppidan H, Kumar A, Leith DE, Geffin B, Laver MB.
mortality rate after the application of PEEP adjusted according
Ventilation with end-expiratory pressure in acute lung disease. J Clin
to measurements of esophageal pressure, as compared with Invest 1972;51:2315–2323.
a ventilatory strategy according to the ARDS Network recom- 4. Haitsma JJ, Lachmann B. Lung protective ventilation in ARDS: the
mendations. In these patients, although the end-inspiratory open lung maneuver. Minerva Anestesiol 2006;72:117–132.
transpulmonary pressure was similar between the two strategies, 5. Gattinoni L, Caironi P, Cressoni M, Chiumello D, Ranieri VM, Quintel
the end-expiratory transpulmonary pressure was significantly M, Russo S, Patroniti N, Cornejo R, Bugedo G. Lung recruitment in
patients with the acute respiratory distress syndrome. N Engl J Med
higher in those ventilated according to esophageal pressure, as
2006;354:1775–1786.
compared with the control group, thereby further suggesting 6. Webb HH, Tierney DF. Experimental pulmonary edema due to in-
a possible impact of opening and closing lung tissue on survival termittent positive pressure ventilation with high inflation pressures:
(45, 46). Of note, in our study population, survivors with a very protection by positive end-expiratory pressure. Am Rev Respir Dis
high percentage of potentially recruitable lung were clinically 1974;110:556–565.
ventilated with PEEP levels significantly greater than those of 7. Kolobow T, Moretti MP, Fumagalli R, Mascheroni D, Prato P, Chen V,
Joris M. Severe impairment in lung function induced by high peak
patients who did not survive.
airway pressure during mechanical ventilation: an experimental study.
In conclusion, based on the uncertainty of how much PEEP Am Rev Respir Dis 1987;135:312–315.
would be necessary during ALI/ARDS, the current study 8. Dreyfuss D, Saumon G. Ventilator-induced lung injury: lessons from
suggests evidence in favor of the application of high levels of experimental studies. Am J Respir Crit Care Med 1998;157:294–323.
PEEP, especially in patients with great lung recruitability. In 9. Gattinoni L, Pesenti A. The concept of ‘‘baby lung.’’ Intensive Care Med
fact, although in patients with a lower percentage of potentially 2005;31:776–784.
10. Mead J, Takishima T, Leith D. Stress distribution in lungs: a model of
recruitable lung the use of 15 cm H2O inevitably leads to an pulmonary elasticity. J Appl Physiol 1970;28:596–608.
increase in alveolar strain, such a parameter (at least at the 11. Muscedere JG, Mullen JB, Gan K, Slutsky AS. Tidal ventilation at low
moderate levels observed in this study) does not appear to be airway pressures can augment lung injury. Am J Respir Crit Care Med
associated with an increased risk of death. In contrast, similar 1994;149:1327–1334.
levels of PEEP may cause a robust reduction of opening and 12. Oeckler RA, Hubmayr RD. Cell wounding and repair in ventilator
closing lung tissue in patients with a higher percentage of injured lungs. Respir Physiol Neurobiol 2008;163:44–53.
13. Caironi P, Langer T, Gattinoni L. Acute lung injury/acute respiratory
potentially recruitable lung, possibly ‘‘protecting’’ the lung from distress syndrome pathophysiology: what we have learned from
the harmful effects of mechanical ventilation. If we consider computed tomography scanning. Curr Opin Crit Care 2008;14:
a plateau airway pressure of about 25–28 cm H2O as safer (43, 64–69.
47), in this category of patients a PEEP value between 15 and 14. Villar J. Low vs high positive end-expiratory pressure in the ventilatory
20 cm H2O is likely to reduce the amount of opening and management of acute lung injury. Minerva Anestesiol 2006;72:357–
closing lung tissue to an acceptable and markedly less harmful 362.
15. Phua J, Badia JR, Adhikari NK, Friedrich JO, Fowler RA, Singh JM,
level. Scales DC, Stather DR, Li A, Jones A, et al. Has mortality from acute
Conflict of Interest Statement: P.C. does not have a financial relationship with respiratory distress syndrome decreased over time? A systematic
a commercial entity that has an interest in the subject of this manuscript. M.C. review. Am J Respir Crit Care Med 2009;179:220–227.
does not have a financial relationship with a commercial entity that has an 16. Brower RG, Lanken PN, MacIntyre N, Matthay MA, Morris A,
interest in the subject of this manuscript. D.C. does not have a financial Ancukiewicz M, Schoenfeld D, Thompson BT. Higher versus lower
586 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 181 2010

positive end-expiratory pressures in patients with the acute respira- acute respiratory distress syndrome. III. Consequences for the effects
tory distress syndrome. N Engl J Med 2004;351:327–336. of positive end-expiratory pressure. Intensive Care Med 2000;26:1215–
17. Meade MO, Cook DJ, Guyatt GH, Slutsky AS, Arabi YM, Cooper DJ, 1227.
Davies AR, Hand LE, Zhou Q, Thabane L, et al. Ventilation strategy 33. Nieszkowska A, Lu Q, Vieira S, Elman M, Fetita C, Rouby JJ. Incidence
using low tidal volumes, recruitment maneuvers, and high positive and regional distribution of lung overinflation during mechanical
end-expiratory pressure for acute lung injury and acute respiratory ventilation with positive end-expiratory pressure. Crit Care Med 2004;
distress syndrome: a randomized controlled trial. JAMA 2008;299: 32:1496–1503.
637–645. 34. Gattinoni L, Pelosi P, Suter PM, Pedoto A, Vercesi P, Lissoni A. Acute
18. Mercat A, Richard JC, Vielle B, Jaber S, Osman D, Diehl JL, Lefrant respiratory distress syndrome caused by pulmonary and extrapulmo-
JY, Prat G, Richecoeur J, Nieszkowska A, et al. Positive end- nary disease: different syndromes? Am J Respir Crit Care Med 1998;
expiratory pressure setting in adults with acute lung injury and acute 158:3–11.
respiratory distress syndrome: a randomized controlled trial. JAMA 35. Malbouisson LM, Busch CJ, Puybasset L, Lu Q, Cluzel P, Rouby JJ; CT
2008;299:646–655. Scan ARDS Study Group. Role of the heart in the loss of aeration
19. Gattinoni L, Caironi P. Refining ventilatory treatment for acute lung characterizing lower lobes in acute respiratory distress syndrome. Am
injury and acute respiratory distress syndrome. JAMA 2008;299:691– J Respir Crit Care Med 2000;161:2005–2012.
693. 36. Gattinoni L, Pelosi P, Crotti S, Valenza F. Effects of positive end-
20. Caironi P, Cressoni M, Chiumello D, Ranieri M, Quintel M, Cornejo R, expiratory pressure on regional distribution of tidal volume and
Bugedo G, Gattinoni L. Key role of the potentially recruitable lung recruitment in adult respiratory distress syndrome. Am J Respir Crit
on the mechanical effects of PEEP during ARDS [abstract]. Intensive Care Med 1995;151:1807–1814.
Care Med 2006;32:S97. 37. Grasso S, Stripoli T, De MM, Bruno F, Moschetta M, Angelelli G,
21. Bernard GR, Artigas A, Brigham KL, Carlet J, Falke K, Hudson L, Munno I, Ruggiero V, Anaclerio R, Cafarelli A, et al. ARDSnet
Lamy M, Legall JR, Morris A, Spragg R. The American–European ventilatory protocol and alveolar hyperinflation: role of positive end-
Consensus Conference on ARDS: definitions, mechanisms, relevant expiratory pressure. Am J Respir Crit Care Med 2007;176:761–767.
outcomes, and clinical trial coordination. Am J Respir Crit Care Med 38. Gattinoni L, Bombino M, Pelosi P, Lissoni A, Pesenti A, Fumagalli R,
1994;149:818–824. Tagliabue M. Lung structure and function in different stages of severe
22. Gattinoni L, Caironi P, Pelosi P, Goodman LR. What has computed adult respiratory distress syndrome. JAMA 1994;271:1772–1779.
tomography taught us about the acute respiratory distress syndrome? 39. Rouby JJ, Lherm T, Martin de Lassale E, Poete E, Bodin L, Finet JF,
Am J Respir Crit Care Med 2001;164:1701–1711. Callard P, Viars P. Histologic aspects of pulmonary barotrauma in
23. Quintel M, Pelosi P, Caironi P, Meinhardt JP, Luecke T, Herrmann P, critically ill patients with acute respiratory failure. Intensive Care Med
Taccone P, Rylander C, Valenza F, Carlesso E, et al. An increase of 1993;19:383–389.
abdominal pressure increases pulmonary edema in oleic acid–induced 40. Treggiari MM, Romand JA, Martin JB, Suter PM. Air cysts and
lung injury. Am J Respir Crit Care Med 2004;169:534–541. bronchiectasis prevail in nondependent areas in severe acute re-
24. Crotti S, Mascheroni D, Caironi P, Pelosi P, Ronzoni G, Mondino M, spiratory distress syndrome: a computed tomographic study of
Marini JJ, Gattinoni L. Recruitment and derecruitment during acute ventilator-associated changes. Crit Care Med 2002;30:1747–1752.
respiratory failure: a clinical study. Am J Respir Crit Care Med 2001; 41. Grasso S, Fanelli V, Cafarelli A, Anaclerio R, Amabile M, Ancona G,
164:131–140. Fiore T. Effects of high versus low positive end-expiratory pressures
25. Venegas JG, Harris RS, Simon BA. A comprehensive equation for the in acute respiratory distress syndrome. Am J Respir Crit Care Med
pulmonary pressure–volume curve. J Appl Physiol 1998;84:389–395. 2005;171:1002–1008.
26. Chiumello D, Carlesso E, Aliverti A, Dellaca RL, Pedotti A, Pelosi PP, 42. Villar J, Kacmarek RM, Perez-Mendez L, Aguirre-Jaime A. A high
Gattinoni L. Effects of volume shift on the pressure–volume curve of positive end-expiratory pressure, low tidal volume ventilatory strat-
the respiratory system in ALI/ARDS patients. Minerva Anestesiol egy improves outcome in persistent acute respiratory distress syn-
2007;73:109–118. drome: a randomized, controlled trial. Crit Care Med 2006;34:
27. Gattinoni L, Carlesso E, Cadringher P, Valenza F, Vagginelli F, 1311–1318.
Chiumello D. Physical and biological triggers of ventilator-induced 43. Hager DN, Krishnan JA, Hayden DL, Brower RG. Tidal volume
lung injury and its prevention. Eur Respir J Suppl 2003;47:15s–25s. reduction in patients with acute lung injury when plateau pressures
28. Pelosi P, D’Andrea L, Vitale G, Pesenti A, Gattinoni L. Vertical are not high. Am J Respir Crit Care Med 2005;172:1241–1245.
gradient of regional lung inflation in adult respiratory distress 44. Talmor D, Sarge T, Malhotra A, O’Donnell CR, Ritz R, Lisbon A,
syndrome. Am J Respir Crit Care Med 1994;149:8–13. Novack V, Loring SH. Mechanical ventilation guided by esophageal
29. Grasso S. Static pressure–volume curves of the respiratory system: worth pressure in acute lung injury. N Engl J Med 2008;359:2095–2104.
keeping to measure them? Minerva Anestesiol 2007;73:107–108. 45. Sarge T, Talmor D. Transpulmonary pressure: its role in preventing
30. Chiumello D, Carlesso E, Cadringher P, Caironi P, Valenza F, Polli F, ventilator-induced lung injury. Minerva Anestesiol 2008;74:335–339.
Tallarini F, Cozzi P, Cressoni M, Colombo A, et al. Lung stress and 46. Sarge T, Talmor D. Targeting transpulmonary pressure to prevent
strain during mechanical ventilation for acute respiratory distress ventilator induced lung injury. Minerva Anestesiol 2009;75:293–299.
syndrome. Am J Respir Crit Care Med 2008;178:346–355. 47. Terragni PP, Rosboch G, Tealdi A, Corno E, Menaldo E, Davini O,
31. Grasso S, Stripoli T, Sacchi M, Trerotoli P, Staffieri F, Franchini D, De Gandini G, Herrmann P, Mascia L, Quintel M, et al. Tidal hyperin-
Monte V, Valentini V, Pugliese P, Crovace A, et al. Inhomogeneity of flation during low tidal volume ventilation in acute respiratory
lung parenchyma during the open lung strategy: a computed tomog- distress syndrome. Am J Respir Crit Care Med 2007;175:160–166.
raphy scan study. Am J Respir Crit Care Med 2009;180:415–423. 48. Le Gall JR, Lemeshow S, Saulnier F. A new simplified acute physiology
32. Puybasset L, Gusman P, Muller JC, Cluzel P, Coriat P, Rouby JJ; CT score (SAPS II) based on a European/North American multicenter
Scan ARDS Study Group. Regional distribution of gas and tissue in study. JAMA 1993;270:2957–2963.