Marco GonzalezAlejandro C. ArroligaFernando Frutos-VivarKonstantinos RaymondosAndres EstebanChristian PutensenCarlos ApezteguıaJavier HurtadoPablo DesmeryVinko TomicicJose ElizaldeFekri AbrougYaseen ArabiRui MorenoAntonio AnzuetoNiall D. Ferguson

Airway pressure release ventilationversus assist-control ventilation: a comparativepropensity score and international cohortstudy

Received: 29 April 2009Accepted: 15 November 2009Published online: 13 March 2010� Copyright jointly held by Springer andESICM 2010

M. GonzalezClınica Medellın y Universidad PontificiaBolivariana, Medellın, Colombia

A. C. ArroligaScott and White, and Texas A&M HealthScience Center, College of Medicine,Temple, TX, USA

F. Frutos-Vivar � A. Esteban ())Intensive Care Unit, Hospital Universitariode Getafe, Carretera de Toledo, Km 12.500,28905 Madrid, Spaine-mail: aesteban@ucigetafe.comTel.: ?34-916834982Fax: ?34-916832095

F. Frutos-Vivar � A. EstebanCIBER Enfermedades Respiratorias,Madrid, Spain

K. RaymondosMedizinische Hochschule, Hannover,Germany

C. PutensenDepartment of Anesthesiology andIntensive Care Medicine,University of Bonn, Bonn, Germany

C. ApezteguıaHospital Profesor A. Posadas, El Palomar,Buenos Aires, Argentina

J. HurtadoHospital de Clınicas, Montevideo, Uruguay

P. DesmerySanatorio Mitre, Buenos Aires, Argentina

V. TomicicClınica Alemana de Santiago,Santiago, Chile

J. ElizaldeHospital ABC, Mexico DF, Mexico

F. AbrougFattouma Bourguiba Monastir,Monastir, Tunisia

Y. ArabiKing Abdulaziz Medical City,Riyadh, Saudi Arabia

R. MorenoHospital de Santo Antonio dos Capuchos,Lisbon, Portugal

A. AnzuetoSouth Texas Veterans Health Care Systemand University of Texas Health ScienceCenter, San Antonio, TX, USA

N. D. FergusonInterdepartmental Division of Critical CareMedicine, Division of Respirology,Department of Medicine, University HealthNetwork and Mount Sinai Hospital,University of Toronto, Toronto, Canada

Abstract Purpose: To comparecharacteristics and clinical outcomesof patients receiving airway pressurerelease ventilation (APRV) or bipha-sic positive airway pressure (BIPAP)to assist-control ventilation (A/C) astheir primary mode of ventilatorysupport. The objective was to esti-mate if patients ventilated withAPRV/BIPAP have a lower mortality.Methods: Secondary analysis of anobservational study in 349 intensivecare units from 23 countries. A totalof 234 patients were included whowere ventilated only with APRV/BI-PAP and 1,228 patients who wereventilated only with A/C. A case-matched analysis according to a pro-pensity score was used to makecomparisons between groups.Results: In logistic regression anal-ysis, the most important factorassociated with the use of APRV/BI-PAP was the country (196 of 234patients were from German units).Patients with coma or congestiveheart failure as the reason to startmechanical ventilation, pH \7.15prior to mechanical ventilation, andpatients who developed respiratoryfailure (SOFA score [2) after intu-bation with or without criteria ofacute respiratory distress syndromewere less likely to be ventilated withAPRV/BIPAP. In the case-matched

Intensive Care Med (2010) 36:817–827DOI 10.1007/s00134-010-1837-1 ORIGINAL

analysis there were no differences inoutcomes, including mortality in theintensive care unit, days of mechani-cal ventilation or weaning, rate ofreintubation, length of stay in theintensive care unit or hospital, andmortality in the hospital.

Conclusions: In this study, theAPRV/BIPAP ventilation mode isbeing used widely across many cau-ses of respiratory failure, but only inselected geographic areas. In ourpatient population we could notdemonstrate any improvement in

outcomes with APRV/BIPAP com-pared with assist-control ventilation.

Keywords Mechanical ventilation �Airway pressure release ventilation �Assist-control ventilation �Mortality �Epidemiology � Propensity score

Introduction

Airway pressure release ventilation (APRV) was originallyproposed in 1987 as a strategy to treat lung-injured patientswho required continuous positive airway pressure (CPAP)and mechanical ventilatory support, with the purportedbenefits of not depressing cardiac output or increasingairway pressure (Paw) excessively [1]. APRV is a time-triggered, pressure-limited, and time-cycled mode of ven-tilation, in which spontaneous breathing is allowed at anypoint during the ventilatory cycle. Conceptually it can beconsidered as two levels of CPAP—the majority of time isspent at high CPAP (Phigh), with intermittent releases to thelow CPAP level (Plow) being used to facilitate ventilation.The theoretical advantages of APRV are the ability tomaximize and maintain alveolar recruitment throughoutthe ventilatory cycle and the use of lower peak inflationpressure but a higher mean airway pressure, resulting in abetter oxygenation. In addition, with the reduction inpleural pressure due to the spontaneous diaphragmaticmovements, improvements in the cardiac index may beseen. Biphasic positive airway pressure ventilation(BIPAP) is similar to APRV in allowing spontaneousbreathing, but there are no restrictions on the timing of thepressure release, i.e., the time at Plow is not always shorterthan that at Phigh as it is in APRV [2, 3].

Despite the availability of this mode on many moderncritical care ventilators, data on the clinical outcome inpatients receiving APRV are limited. To our knowledge,there is only one small randomized controlled trial thathas examined the effect of APRV on ventilator-free daysin patients with acute respiratory distress syndrome [4].To date, the rest of the literature regarding the use ofAPRV/BIPAP consists of small case series or cross-overstudies with surrogate outcomes or physiological end-points such as oxygenation [5]. Because no goodinformation exists regarding to what extent APRV/BIPAPis used in patients admitted to an intensive care unit, weanalyzed a database from an international, multicenterstudy on mechanical ventilation to address this issue [6].Patients who were ventilated with APRV/BIPAP duringthe whole period of mechanical ventilation were matchedusing a propensity score with patients ventilated withvolume-cycled assist-control (A/C) ventilation. Our mainobjective was to estimate if patients ventilated with

APRV/BIPAP had a lower mortality in the intensive careunit compared with those ventilated with A/C.

Materials and methods

Patients

We analyzed data from a prospective, multicenter, inter-national cohort of 4,968 adult patients who receivedmechanical ventilation for more than 12 h during a1-month period beginning 1 April 2004 in 349 intensivecare units in 23 countries (see Appendix 1 for list ofcenters) [6]. The study protocol was approved by the localInstitutional Review Board of each participating centerwith a waiver for consent. For the purpose of this analysis,we selected only the patients who were ventilated withAPRV/BIPAP or with assist-control ventilation during thetotal time of ventilatory support (prior to initiation ofweaning). We excluded 29 patients ventilated with APRVwho received neuromuscular blockers.

The following information was collected on eachenrolled patient at baseline: demographic data [sex, age,weight, height, Simplified Acute Physiology Score (SAPSII)] at the time of ICU admission; day of initiatingmechanical ventilation; primary indication for mechanicalventilation: acute or chronic pulmonary disease [chronicobstructive pulmonary disease (COPD), asthma, chronicpulmonary disease other than COPD]; coma; neuromus-cular disease; acute respiratory failure [acute respiratorydistress syndrome (ARDS), postoperative, congestiveheart failure, aspiration, pneumonia, sepsis, trauma, andcardiac arrest]. The following variables were collecteddaily until ICU discharge or to day 28, whichever camefirst: ventilator settings [tidal volume, positive end-expi-ratory pressure (PEEP), peak pressure, plateau pressure],sedative use, and complications arising [acute respiratorydistress syndrome (ARDS), barotrauma, ventilator-asso-ciated pneumonia, sepsis and organ failures(cardiovascular failure, respiratory failure, renal failure,hepatic failure, coagulopathy) defined as a sequentialorgan failure assessment (SOFA) score higher than 2points]. The patients were prospectively followed foroutcomes until hospital discharge.

818

Statistical analysis

Data are expressed as mean (SD), median (interquartilerange) and proportions as appropriate. Student’s t test andMann-Whitney U test were used to compare continuousvariables, and the chi-square test or Fisher’s exact testwas used to compare proportions. To compare the venti-lator settings and arterial blood gases between bothgroups over three time points: days 1, 3, and 7, we used ageneralized estimating equation (GEE), including treat-ment, time (coded as a dummy variable), and theinteraction between treatment and time as variables in theequation.

To estimate the simultaneous effects of multiple vari-ables on the decision to initiate and continue use of APRV/BIPAP, a multivariate analysis was performed using alogistic regression model and a backward stepwise selec-tion method. We used a significance threshold of p \ 0.10for entering tested variables into the model. Goodness offit of the model was evaluated with Hosmer-Lemeshow’stest as well as by visual inspection of contingency tables,and model discrimination was assessed by evaluating thearea under the receiver-operator curve (ROC).

Since the utilization of APRV/BIPAP was not ran-domly assigned, treatment-indication bias and potentialconfounding variables were accounted for by developinga propensity score using variables predictive for the use ofAPRV/BIPAP determined by the previously describedlogistic regression model. These variables were: geo-graphical area, reasons for initiation of mechanicalventilation (coma or congestive heart failure), pH \7.15prior to starting mechanical ventilation, and complica-tions arising over the course of mechanical ventilation(ARDS or respiratory failure). We matched patients whowere ventilated with APRV/BIPAP during their ICU stayto patients ventilated with A/C on the basis of the pro-pensity score. We sought to match each ventilated patientwith APRV/BIPAP with a control patient ventilated withA/C who had the closest propensity score (within 0.05 ona scale of 0–1).

Analyses were performed with SPSS (version 17.0)and Stata (version 10.0) statistical programs.

Results

Patients

Five hundred sixty-three patients (11.3%) were ventilatedfor at least 1 day with APRV/BIPAP. Of these, 263(5.2%) were ventilated continually with APRV/BIPAP, ofwhom we excluded 29 patients who received neuromus-cular blockers. In the analysis we therefore compared 234patients ventilated with APRV/BIPAP to 1,228 patientswho were ventilated continually with A/C. Table 1 showsthe comparison of baseline characteristics.

Factors associated with the use of APRV/BIPAP

The most important factor associated with the use ofAPRV/BIPAP was the country from which the patientswere included. Ninety-six percent of patients ventilatedwith this mode were from European units, especially fromGermany (196 of 234 patients). Other factors associatedwith the use of APRV/BIPAP are shown in Table 2.Patients with coma or congestive heart failure as thereason to start mechanical ventilation, acidosis (pH\7.15) prior to beginning mechanical ventilation, andpatients who developed respiratory failure (respiratorySOFA score [2), with or without acute respiratory dis-tress syndrome, were less likely to be ventilated withAPRV/BIPAP.

Neither visual inspection of the contingency tables northe Hosmer–Lemeshow goodness of fit test for the modelshowed evidence of inadequate fit (v2 = 10.96; p =0.09), and the model showed good discrimination (areaunder ROC curve = 0.90; 95% confidence interval: 0.88–0.92; p \ 0.001).

Case-matched study: characteristics

Two hundred thirty-four patients ventilated with APRV/BIPAP were matched to 234 controls ventilated with A/C.Table 3 shows the characteristics of patients included inthis analysis; there were no clinically or statistically sig-nificant differences in the characteristics of patientsmatched based on propensity score. Table 4 shows thecomparison of ventilator settings and arterial blood gaseson day 1, 3 and 7. Patients ventilated with APRV/BIPAPhad a lower inspiratory pressure for an equal tidal volume,and they were ventilated with a higher expiratory pressure(Fig. 1). These differences in ventilatory parameters wereassociated with significantly better oxygenation in thegroup of patients ventilated with APRV (Table 4).

There were no significant differences in the proportionof patients that received sedatives (70% in the APRV/BIPAP group vs. 75% in the A/C group; p = 0.21) nor inthe number of days of sedative use: median time 2 days(interquartile range 1, 4) in the APRV/BIPAP groupversus 2 days (interquartile range 1, 5) in the A/C group(p = 0.12).

In patients with cardiovascular failure, there were nodifferences in the duration of this organ failure: mediantime 3 days (interquartile range 2, 5) in the APRV/BIPAPgroup versus 3 days (interquartile range 2, 7) in the A/Cgroup (p = 0.22).

Case-matched study: outcomes

Table 5 shows outcomes of the cases and controls. Therewere no differences in the majority of the major clinical

819

outcomes including days of mechanical ventilation orweaning, rate of reintubation, length of stay in theintensive care unit, or mortality in the ICU or hospital.There was a significant (p = 0.05) trend toward a longerstay in hospital in the APRV/BIPAP group by 1 day. Wealso found a higher rate of tracheostomy in the APRVgroup (20 vs. 11%; p = 0.007) without significant dif-ferences (p = 0.21) in the timing of tracheostomy: in theAPRV median time to tracheostomy was 6 days (inter-quartile range 3, 11), and in the control group it was8 days (interquartile range 4.5, 13.5).

Discussion

The main finding of our study is that major clinical out-comes in a heterogeneous population of mechanicallyventilated patients are similar when they are ventilated withairway pressure release ventilation/biphasic positive air-way pressure ventilation or with assist-control ventilation.

APRV is a mode of mechanical ventilation that isbased on the open lung concept, maintaining adequatelung volumes and recruiting alveoli. It allows patients tobreathe spontaneously while receiving high continuous

Table 1 Characteristics of patients included in the analysis—unmatched comparisons

A/CN = 1,228

APRV/BIPAPN = 234

p Value

Geographical area, n (%) \0.001Europe 373 (30) 229 (98)Latin America 488 (40) 3 (1)USA and Canada 331 (27) 2 (1)Other 36 (3) –

Age, mean (SD), years 57 (18) 59 (17) 0.37Female sex, n (%) 486 (40) 89 (38) 0.65Simplified Acute Physiology Score II, mean (SD), points 46 (18) 44 (19) 0.13Main reason for mechanical ventilation, N (%)Chronic obstructive pulmonary disease 52 (4) 10 (4) 0.98Asthma 21 (2) 1 (0.4) 0.14Other chronic lung disease 13 (1) 2 (1) 0.77Coma 359 (29) 29 (12) \0.001Neuromuscular disease 17 (1) 1 (0.4) 0.22

Acute respiratory failurePostoperative 182 (15) 77 (33) \0.001Community-acquired pneumonia 72 (6) 15 (6) 0.75Hospital-acquired pneumonia 40 (3) 6 (2) 0.58Sepsis 132 (11) 26 (11) 0.87Acute respiratory distress syndrome 27 (2) 2 (1) 0.17Congestive heart failure 67 (5.5) 6 (3) 0.06Cardiac arrest 61 (5) 16 (7) 0.24Trauma 36 (3) 14 (6) 0.02Aspiration 34 (3) 10 (4) 0.21Other cause of acute respiratory failure 115 (9) 19 (8) 0.54

Complications during mechanical ventilation, n (%)Ventilator-associated pneumonia 124(10) 13 (6) 0.03Sepsis 98 (8) 10 (4) 0.05Acute respiratory distress syndrome 58 (5) 2 (1) 0.006Barotrauma 32 (3) 5 (2) 0.67

Organ dysfunction during mechanical ventilation, n (%)Cardiovascular failure 386 (31) 83 (35.5) 0.22Acute renal failure 223 (18) 47 (20) 0.49Respiratory failure 429 (35) 63 (27) 0.02Hepatic failure 138 (11) 26 (11) 0.95Coagulopathy 198 (16) 39 (17) 0.84

Arterial blood gases previous to start mechanical ventilationpH, mean (SD) 7.31 (0.13) 7.34 (0.11) 0.007PaCO2, mmHg, mean (SD) 45 (21) 40 (12) 0.18Ratio PaO2 to FiO2, mean (SD) 220 (118) 215 (127) 0.81

OutcomesDuration of ventilatory support, days, median (interquartile range) 3 (2, 7) 3 (2, 5) 0.03Tracheostomy 105 (9) 46 (20) \0.001Mortality in the intensive care unit, n (%) 514 (42) 65 (28) \0.001Mortality in the hospital, n (%) 546 (44.5) 81 (37) 0.01

820

positive airway pressure above the lower inflection pointof the pressure volume curve [5, 7]. In several clinicalstudies [4, 8], patients with ALI/ARDS or at risk ofdeveloping these conditions had lower levels of peakairway pressure with APRV/BIPAP compared with con-ventional ventilation. When used appropriately, APRV/

BIPAP may prevent alveolar collapse and overdistentionduring tidal breathing while allowing spontaneousbreathing [3].

In our population of mechanically ventilated patients,APRV/BIPAP was used in a minority of patients (5.2%)and was used almost exclusively in Germany. It is

Table 2 Univariable and multivariable logistic-regression analysis: factors associated with ventilation with APRV/BIPAP

Univariable analysis Multivariable analysis

Odds ratio (CI 95%) p Odds ratio (CI 95%) p

Reason to start mechanical ventilationComa 0.39 (0.27 to 0.57) \0.001 0.20 (0.13 to 0.33) \0.001Postoperative acute respiratory failure 2.27 (1.79 to 2.88) \0.001Trauma 1.79 (1.13 to 2.84) 0.02Congestive heart failure 0.50 (0.23 to 1.08) 0.06 0.27 (0.11 to 0.69) 0.005

Arterial blood gases prior to start mechanical ventilationpH \7.15 0.49 (0.28 to 0.86) 0.008 0.35 (0.18 to 0.70) 0.003

Complications over the course of mechanical ventilationRespiratory failure 0.73 (0.56 to 0.97) 0.02 0.59 (0.40 to 0.86) 0.007Ventilator-associated pneumonia 0.58 (0.36 to 0.95) 0.03Sepsis 0.56 (0.31 to 1.02) 0.05Acute respiratory distress syndrome 0.20 (0.05 to 0.79) 0.006 0.19 (0.04 to 0.85) 0.03

Table 3 Characteristics of patients matched on propensity score

CasesN = 234

ControlsN = 234

p Value

Geographical area, n (%) 0.93Europe 229 (98) 228 (98)Latin America 3 (1) 4 (1)USA and Canada 2 (1) 2 (1)

Age, mean (SD), years 59 (17) 61 (17) 0.09Female sex, n (%) 89 (38) 80 (34) 0.39Simplified Acute Physiology Score II, mean (SD), points 44 (19) 46 (20) 0.32Main reason for mechanical ventilation, n (%)Chronic obstructive pulmonary disease 10 (4) 13 (6) 0.52Asthma 1 (0.4) 1 (0.4) 1.00Other chronic lung disease 2 (1) 4 (2) 0.68Coma 29 (12) 29 (12) 1.00Neuromuscular disease 1 (0.4) 2 (1) 0.62

Acute respiratory failurePostoperative 77 (33) 65 (28) 0.23Pneumonia 21 (9) 20 (8.5) 0.75Sepsis 26 (11) 25 (11) 0.88Acute respiratory distress syndrome 2 (1) 7 (3) 0.17Congestive heart failure 6 (3) 6 (3) 1.00Cardiac arrest 16 (7) 21 (9) 0.39Trauma 14 (6) 13 (6) 0.84Aspiration 10 (4) 5 (2) 0.19Other cause of acute respiratory failure 19 (8) 23 (10) 0.52

Complications over the course of the mechanical ventilation, n (%)Cardiovascular failure 83 (35.5) 72 (31) 0.28Respiratory failure 63 (27) 79 (35) 0.11Acute renal failure 47 (20) 40 (17) 0.41Hepatic failure 26 (11) 23 (10) 0.65Coagulopathy 39 (17) 38 (16) 0.90Sepsis 10 (4) 15 (6) 0.30Acute respiratory distress syndrome 2 (1) 2 (1) 1.00Ventilator-associated pneumonia 13 (6) 21 (9) 0.15Barotrauma 5 (2) 6 (3) 0.76

821

important to note that patients receiving APRV/BIPAP inour study represented a broad range of causes of acuterespiratory failure. Thus, our results may not be gener-alizable to specific subgroups of patients such as thosewith ALI/ARDS. Based on the logistic regression resultspredicting the use of APRV/BIPAP, clinicians in thisstudy appeared to favor this mode in the setting of post-operative respiratory failure and trauma, but avoided it forcoma, congestive heart failure, and when severe acidosiswas present.

Potential advantages of APRV/BIPAP could be arelatively lower peak airway pressure, a lower intratho-racic pressure, and better matching of ventilation andperfusion. In our cohort, the peak or high airway pressurein patients ventilated with APRV/BIPAP was lower thanthe peak pressure on patients ventilated with A/C. How-ever, we must interpret these results cautiously as we areessentially comparing a plateau pressure (APRV) with apeak pressure (A/C) and because we do not have infor-mation about patient efforts and true transpulmonarypressures. The low pressure/positive end-expiratorypressure was higher in the group ventilated with APRV/BIPAP when compared with the group of patients venti-lated with A/C, similar to previous reports [4, 8]. The useof APRV in most but not all studies has been associatedwith better oxygenation and better cardiac function [4,8–11]. In this study after adjustment for time and treat-ment, we observed a better PaO2 to FiO2 ratio in thegroup of patients ventilated with APRV.

Another potential advantage of APRV/BIPAP couldbe a lower requirement of sedation and/or neuromuscularblockade [5]. However, in our analysis, the proportion ofpatients who received sedatives and the number of daysthat they received sedatives were similar in the twogroups. Previous studies have reported contradictoryfindings in term of the doses and duration of sedatives andanalgesics [4, 8, 12]. Putensen et al. [8] used a sedationregimen with less midazolam and sufentanyl in patientsventilated with APRV to allow spontaneous breathingwhen compared with patients receiving pressure-limited,time-cycled, controlled mechanical ventilation [8]. In

Table 4 Comparison of ventilator settings and arterial blood gases of matched patients

Day 1 Day 3 Day 7 Effect size (CI 95%) pValue

CasesN = 234

ControlsN = 234

CasesN = 128

ControlsN = 128

CasesN = 40

ControlsN = 53

Ventilatory settingsTidal volume, ml/kg of

predicted bodyweight, mean (SD)

9.3 (2.2) 9.2 (2.4) 9.3 (2.2) 9.0 (1.7) 9.8 (2.3) 9.1 (1.8) 0.26 (-0.06 to 0.59) 0.11

High pressure (cases)/peak pressure(controls), cmH2O,mean (SD)

25 (9) 31 (9) 24 (6) 31 (9) 25 (5) 31 (8) -5.98 (-7.90 to -4.06) \0.001

Low pressure (cases)/positive end-expiratory pressure(controls), cmH2O,mean (SD)

7 (4) 3 (3) 8 (4) 4 (4) 7 (5) 5 (4) 3.75 (3.22 to 4.28) \0.001

Arterial blood gasespH, mean (SD) 7.36 (0.10) 7.36 (0.12) 7.41 (0.08) 7.39 (0.08) 7.41 (0.11) 7.41 (0.07) 0.008 (-0.006 to 0.023) 0.24PaCO2, mmHg, mean

(SD)42 (12) 40 (12) 41 (9) 39 (8) 42 (11) 41 (11) 1.56 (0.025 to 3.09) 0.05

Ratio PaO2 to FiO2,mean (SD)

263 (134) 232 (120) 254 (99) 210 (99) 250 (102) 208 (84) 36.10 (19.92 to 52.28) \0.001

Effect size estimated by generalized estimating equation models. In all models, the interaction term was eliminated because it was notsignificant

0

5

10

15

20

25

30

35

1 2 3 4 5 6 7Days

Air

way

pres

sure

s, c

mH

2O

0

50

100

150

200

250

300

Rat

io P

aO2

toF

iO2

Fig. 1 High and low airway pressures for first week of mechanicalventilation in cases ventilated with APRV/BIPAP (white squares)and in controls ventilated with assist-control ventilation (blackcircles). Values are expressed as mean (SEM). There weresignificant differences (p \ 0.05) in every measurement. Thecolumns show the daily ratio of PaO2 to FiO2 (grey columnscorrespond to cases and white columns to controls)

822

contrast, Varpula et al. [4] reported no differences in theuse of propofol and fentanyl in patients with the acuterespiratory distress syndrome ventilated with APRVversus patients ventilated with pressure-controlled syn-chronized intermittent ventilation mode with pressuresupport. It is unclear why we did not find a reduction inthe use of sedatives agents in this study, but we cannotexclude the possibility that the doses of these medicationswere different even though the proportions of patientsreceiving them were the same.

Data on the clinical outcome in patients receivingAPRV are limited. The outcomes in the cases and controlsin our study were similar except for tracheostomy (morecommon in the APRV/BIPAP group) and a small differ-ence in hospital length of stay (1 day longer in the APRV/BIBPAP group). However, the rate of tracheostomy isinfluenced by local practice that may or may not berelated to the mode of mechanical ventilation. The lack ofsignificant differences in the outcomes such as days ofmechanical ventilation and days of weaning from venti-lation, length of stay in the ICU, and in ICU and hospitalmortality are at odds with previous reports [10, 11].

In the past few years, the propensity score method hasbecome a common method used for confounder adjust-ment in observational studies. The propensity score is apredicted probability resulting from the developed model.Its main purpose is to control multiple confounderssimultaneously and to account for treatment indicationbias, and it is useful to maximize the balance between thetreatment groups. The key purpose of this method is togenerate probabilities of treatment assignment conditionalon a set of variables that are both related to the treatmentand the outcome. The distribution of these confoundersshould be fairly equal between the treatment groups whenestimating the effect of treatment on the outcome [13].For the purposes of constructing a propensity score, weconsidered that patients receiving mechanical ventilationwere indeed receiving a dynamic and potentially chang-ing therapy. As such, and in accordance with existingrecommendations for propensity scoring [13], we inclu-ded variables that were present at baseline and those thatarose after the initial choice of ventilator mode. We didthis because in order to be included as an APRV/BIBAP

patient in our study, patients needed to start on this mode,and continue with it throughout their period of ventilatorysupport; we reasoned that variables that arose afterbaseline (such as complications arising during ventilation)could have influenced clinicians’ decisions to stay with orswitch away from APRV.

A limitation of our study is that, although we assumethat in most instances physicians chose APRV to allowpatients to breath spontaneously, we cannot guarantee thatall the patients ventilated with APRV/BIPAP had spon-taneous breathing, as these data were not collected.Regardless of the amount of spontaneous breathing thatoccurred in the APRV/BIPAP group, we believe that ouranalysis is valid in the sense that it compares these twomodes of ventilation as they are used in usual clinicalpractice.

From this study we conclude that the APRV/BIPAPventilation mode is being used in almost all pathologiesthat lead to acute respiratory failure, but only in selectedgeographic areas. Moreover, we did not find any evidenceof its superiority in clinical outcomes compared to assist-control ventilation. More studies with strict methodolog-ical designs are required to compare this ventilation modewith others, in order to determine whether or not it isactually favorable in terms of clinical outcomes.

Acknowledgment Dr. Ferguson is supported by a New Investi-gator Award from the Canadian Institutes of Health Research(Ottawa, Canada).

Appendix 1: the ventilation group members includethe following collaborators

Argentina: Coordinators: Carlos Apezteguia (HospitalProf. A. Posadas, El Palomar, Buenos Aires) and PabloDesmery (Sanatorio Mitre, Buenos Aires).

A. Sarasino and D. Ceraso (Hospital Dr. Juan A.Fernandez, Buenos Aires), D. Pezzola and F.Villarejo(Hospital Prof. A. Posadas, El Palomar), C. Cozzani andM. Torres Boden (Hospital Dr. C. Argerich, BuenosAires), C. Santos and E. Capparelli (Hospital Eva Peron,

Table 5 Outcomes of patients included in the matched-case study

CasesN = 234

ControlsN = 234

p Value

Days of mechanical ventilation, median (interquartile range) 3 (2, 5) 3 (2, 6) 0.61Days of weaning, median (interquartile range) 1 (1, 2) 1 (1, 2) 0.28Reintubation, n (%) 10/140 (7) 10/139 (7) 0.99Tracheostomy, n (%) 46 (20) 25 (11) 0.007Length of stay in the intensive care unit, median (interquartile range) 6 (3, 14) 6 (3, 12) 0.11Mortality in the intensive care unit, n (%) 65 (28) 78 (33) 0.19Length of stay in the hospital, median (interquartile range) 18 (11, 38) 17 (8, 31) 0.05Mortality in the hospital, n (%) 81 (35) 90 (38) 0.50

823

San Martın), M. Tavella and C. Irrazabal (Hospital deClınicas Jose de San Martın, Buenos Aires), L. Cardonnetand A. Diez (Hospital Provincial del Centenario,Rosario), A. Giannelli and L. Vargas (Policlınico deNeuquen), M. Bustamante (Hospital Heroes de Malvinas,Merlo), E. Turchetto (Hospital Privado de la Comunidad,Mar del Plata), J. Teves and O. Elefante (Hospital OscarAlende, Mar del Plata), C. Sola and J. Mele (HospitalDr. Jose Penna, Bahıa Blanca), V. Sciuto and P. Grana(Hospital Provincial de Neuquen), G. Jannello and R.Valentini (CEMIC, Buenos Aires), S. Ilutovich (SanatorioMitre, Buenos Aires), L. Huespe Gardel (Hospital EscuelaJose F. de San Martın, Corrientes), J. Scapellato andE. Orsini (Hospital F. Santojanni, Buenos Aires),G. Aguero and A. Sanchez (Policlınico Regional J. Peron,Mercedes), R. Fernandez and L. Villalobos Castaneda(Hospital Italiano, Buenos Aires), F. Gonzalez andE.Estenssoro (Hospital General San Martın, La Plata),S. Lasdica (Hospital Privado del Sur, Bahıa Blanca),A. Gomez and J. Scapellato (Clınica de la Esperanza,Buenos Aires), P. Pratesi (Hospital Universitario Austral,Pilar), M. Blasco and F. Villarejo (Clınica Olivos, Oli-vos), G. Olarte and C. Bevilacqua (Clınica Modelo deMoron/Hospital San Juan de Dios, R.Mejıa), M. Quint-eros (Sanatorio San Lucas, San Isidro).

P. Ripoll (Clınica La Sagrada Familia, Buenos Aires),S. Filippus (Clınica del Valle, Comodoro Rivadavia),F. Guzman Dıaz and M.Deheza (Hospital B. Rivadavia,Buenos Aires), E. Garcıa and J. Arrieta (Hospital Regionalde Comodoro Rivadavia), P. Pardo and J. Neira (Sanatoriode la Trinidad de Palermo, Buenos Aires), J. Nunez andF. Palizas (Clınica Bazterrica, Buenos Aires), A. Ciccoliniand G. Murias (Sanatorio Santa Isabel, Buenos Aires),W. Vazquez and M. Grilli (Hospital Espanol de Mendoza,Godoy Cruz), F. Chertcoff and E. Soloaga (HospitalBritanico, Buenos Aires), D. Vargas and J. Beron (Hos-pital Pablo Soria, San Salvador de Jujuy), A. Maceira andP. Schoon (Hospital Prof. Luis Guemes, Haedo), D. Pina(Sanatorio Franchın, Buenos Aires), E. Sobrino andA. Raimondi (Sanatorio Mater Dei, Buenos Aires), E. DeVito (IIM Alfredo Lanari, Buenos Aires).

Belgium: M. Malbrain (Ziekenhuis Netwerk,Antwerpen).

Bolivia: Coordinator: Freddy Sandi Lora (Hospitalobrero N� 1, La Paz).

A. Lavandez and C. Alfaro (Complejo HospitalarioViedma, La Paz), J. Guerra (Instituto gastroenterologicoboliviano japones, Santa Cruz).

Canada: Coordinators Niall D.Ferguson (TorontoHospital Western Division) and Maureen O. Meade(McMaster University).

J. T. Granton (Toronto General Hospital), S. E. Lapinsky(Mount Sinai, Toronto), J. Meyer (St. Joseph0s Hospital,Toronto), D. C. Scales (St. Michael0s Hospital, Toronto),R. A. Fowler (Sunnybrook Health Sciences Centre,

Toronto), B. Kashin (William Osler Health Centre,Brampton, Ontario), D. J. Cook (St. Joseph’s Healthcare).

Colombia: Coordinator Marco A. Gonzalez (ClınicaMedellın y Universidad Pontificia Bolivariana, Medellın).

A. Guerra (Hospital General de Medellın and ClınicaSOMA, Medellın), C. Cadavid (Hospital Pablo TobonUribe, Medellın), R. Panesso (Clınica Las Americas,Medellın), M.Granados (Clınica Valle del Lilli, Cali),C. Duenas (Hospital Bocagrande, Cartagena), F. Molina(Clınica Bolivariana, Medellın), R. Camargo (ClınicaGeneral del Norte de Barranquilla), G. Ortiz (Hospital deSanta Clara, Bogota), M. Gomez (Hospital de San Jose).

Chile: Coordinator: Vinko Tomicic (Clınica Alemanade Santiago).

L. Soto (Instituto Nacional del Torax, Santiago),C. Romero (Hospital Clınico Pontificia UniversidadCatolica, Santiago), M. Teresa Caballero and L. Chiang(Hospital naval almirante NEF), E. Poch (Instituto deNeurocirugıa), J. Canteros Gatica (Hospital Curico), H.Ugarte (Hospital de Coquimbo), M. Calvo, C. Vargas andM. Yacsich. (Hospital Regional de Valdivia), E. Tobar(Hospital Clınico de la Universidad de Chile, Santiago),J. G. Urra (Clınica Alemana de Temuco).

England: Coordinator: Peter Nightingale (WythenshaweHospital, Manchester).

J. Hunter (Macclesfield District General Hospital,Macclesfield), J. Hunter (Rotherdam District GeneralHospital, Rotherdam), S. Mousdale (Blackburn RoyalInfimary, Blackburn), J. Harper (Royal Liverpool Uni-versity Hospital, Liverpool), A. Conn (Wansbeck GeneralHospital, Ashington), D. Higgins (Southend Hospital,Westcliffe-on-Sea), D. Jayson (Southport & FormbyDistrict General Hospital, Southport), D. Hawkins (NorthStaffordshire Hospital, Stoke on Trent).

Ecuador: Coordinator: Manuel Jibaja (Hospital Militarde Quito).

G. Paredes and E. Bazantes (Hospital Enrique Garces,Quito), P. Jimenez (Hospital Carlos Andrade Martın,Quito), J. Vergara and L. Gonzalez (Hospital Luis Ver-naza Valdez, Guayaquil).

France: Coordinators Laurent Brochard (Hopital HenriMondor, Creteil) and Arnaud Thille (Hopital HenriMondor, Creteil).

L. Mallet (Centre Hospitalier D’Auch), P. Andrivet(Centre Medico-Chirurgical de Bligny, Bris-sous-Forges),O. Peyrouset (Hopital Ambroise Pare, BoulogneBillancourt), I. Mohammedi (Hopital Edouard Herriot,Lyon), E. Guerot (Hopital Europeen Georges Pompidou,Paris), N. Deye (Hopital Lariboisiere, Paris), S. Monseland F. Bouvet (Hopital Pitie Salpetriere, Paris), M. Dar-mon (Hopital Saint Louis, Paris), M. Fartoukh andA. Harb (Hopital Tenon, Paris), N. Anguel (Hopital deBicetre, Kremlin-Bicetre).

Germany: Coordinator: Konstantinos Raymondos(Medizinische Hochschule Hannover).

824

A. Nowak, T. Pahlitzsch and K. F. Rothe (Kranken-haus Dresden-Friedrichstadt), M. Ragaller and T. Koch(Universitaetsklinikum Carl Gustav Carus Dresden), G.Sterzel (Kreiskrankenhaus Loebau, Ebersbach), R. Wit-tich (Carl-Thiem-Klinikum Cottbus gGmbH), K. Rudolphand J. Raumanns (St. Elisabeth gGmbH Leipzig), U.Grueneisen and F. Stupacher (BundeswehrkrankenhausLeipzig), H. Bromber, G. Leonhardt and J. Soukup(Universitaetsklinikum der Martin-Luther-UniversitaetHalle-Wittenberg), C. Wuttke (Krankenhaus St. Elisabethund St. Barbara Halle, Saale), M. Holler (StaedtischesKrankenhaus Martha-Maria Halle-Doelau gGmbH), J.Haberkorn (Georgius-Agricola-Klinikum Zeitz), P. Jehle(Paul-Gerhard-Stiftung, Lutherstadt Wittenberg),B. Albrecht (Zeisigwaldkliniken Bethanien Chemnitz),D. M. Klut (Kreiskrankenhaus Rochlitz), H. J. Hartung(Vivantes Krankenhaus am Urban, Berlin-Kreuzberg),H. Gerlach (Vivantes-Klinikum Neukoelln, Berlin),T. Henneberg, S. Weber-Carstens, K. Haid, and C. Melzer-Gartzke, M. Oppert (Charite Universitaetsklinikum,Campus Virchow, Berlin), M. Reffenberg (LungenklinikHeckeshorn, Berlin), Ch. Werel and A. Kopietz (Klini-kum Barnim GmbH, Werner Forßmann Krankenhaus,Eberswalde), T. Nippraschk and D. Hoffmeister (Rupp-iner Klinikum GmbH, Neuruppin), M. Schneider(Dietrich-Bonhoeffer-Klinikum-Neubrandenburg), D. A.Vagts and G. Noeldge-Schomburg (Medizinische Fakul-taet der Universitaet Rostock), G. Savinski and T. Kloess(Allgemeines Krankenhaus Harburg, Hamburg), C.Frenkel, D. Yakisan, H. Schroeder and C. Daniels(Staedtisches Klinikum Lueneburg), B. Sedemund-Adib(Universitaetsklinikum Schleswig Holstein-Campus Lue-beck), S. Krueper (Klinikum Hannover Nordstadt), J.Ahrens, U. Molitoris and K. Johanning (MedizinischeHochschule Hannover), D. Korth and W. Seitz (Kreisk-rankenhaus Hameln), J. Kleideiter and P. Palomino(Staedtische Kliniken Bielefeld gGmbH), A. Lunkeit andJ. Schlechtweg (Klinikum Bad Salzungen gGmbH), M.Quintel (Universitaetsklinikum der Georg-August-Uni-versitaet Goettingen), Schild and C. P. Criee(Evangelisches Krankenhaus Goettingen-Weende e.V.,Bovenden-Lenglern), M. Bund (Albert-Schweitzer-Kran-kenhaus Northeim), M. Hundt, U. Schulze and J. Kolle(Kreiskrankenhaus Charlottenstift, Stadtoldendorf), J.Offensand, S. Youssef, and J.P. Juvana (Klinikum Salz-gitter GMBH), W. Seyde (Staedtisches KlinikumWolfenbuettel), T. Luecke and A. Gruener (Universita-etsklinikum Mannheim), E. Calzia (Universitaetsklinikumfur Anasthesiologie, Ulm), J. Heine, M. Borth, U. vonLeitner and M. Hoffmann (Dr. Herbert-Nieper-Kranken-haus-Goslar), W. Brandt (UniversitaetsklinikumMagdeburg), A. Keller and S. Scieszka (KrankenhausNeuwerk, Moenchengladbach), E. Schroeder and F. L.Deres (Kreiskrankenhaus Dormagen), M. Burrichter, T.Bernhardt and W. Wilhelm (St.-Marien-Hospital, Lue-nen), M. Beiderlinden (Universitaetklinikum Essen),

H. Steiniger and V. Weißkopf (Ruhrlandklinik, Essen), H.Militzer (Evangelisches und Johanniter Klinikum, Din-slaken), K. Eicker and F. Hinder (UniversitaetsklinikumMuenster), C. Weilbach and M. Raab (St. Josefs-StiftCloppenburg), F. Ragalmuto (Kliniken der Stadt KoelnKrankenhaus Holweide), T. Moellhoff and K. Tsompa-nidis (Katholische Stiftung Marienhospital Aachen), D.Henzler and R. Kuhlen (Universitaetsklinikum Aachen),H. Wrigge, C. Putensen and F. L. Dumoulin (Universi-taetsklinikum Bonn), M. Foedisch and J. Busch(Evangelisches Waldkrankenhaus Bad GodesberggGmbH, Bonn), W. Theelen (St. Johannes-KrankenhausTroisdorf), A. Deller (Krankenhaus der BarmherzigenBrueder, Trier), W. Baier (St. Nikolaus-StiftshospitalGmbH, Andernach), B. Eller (Staedt. Hellmig-Kranken-haus, Kamen), K. Schwarke (Evang. KrankenhausSchwerte GmbH), J. Buttner (Evangelisches KrankenhausElisabethenstift gGmbH, Darmstadt), K. P. Wresch andK. Steidel (St.-Vincentius-Krankenhaus Speyer), J. F.Meyer (Universitaetsklinikum der Ruprecht-Karls-Uni-versitaet Heidelberg), M. Layer (Thoraxklinik HeidelberggGmbH), G. Meinhardt (Robert-Bosch-Krankenhaus,Stuttgart), J. Fritschi and P. Zaar (Ermstalklinik Staed-tisches Krankenhaus Sindelfingen), H. P. Stegbauer(Kreiskrankenhaus Leonberg), V. Tumbass and S. Hahn(Ermstalklinik Bad Urach), H. Mende, M. Fischer, J.Martin and A. Assmann (Klinik am Eichert Goeppingen),V. Schoeffel, K. van Deyk and S. Seyboth (StadtklinikBaden-Baden), H. Kerger and J. Ernst (EvangelischesDiakoniekrankenhaus, Freiburg), H. F. Ginz (Kreiskran-kenhaus Loerrach), F. Brettner (Krankenhaus derBarmherzigen Brueder, Muenchen), O. Karg (ASKLE-PIOS Fachkliniken Muenchen-Gauting), M. Glaser and T.P. Zucker (Klinikum Traunstein), J. Jahn and A. Schnei-der (Fachkliniken Wangen), M. Burkert(Bundeswehrkrankenhaus Ulm), H. Kuenzig and T. Bein(Klinikum der Universitaet Regensburg), A. Speicher(Krankenhaus der Barmherzigen Brueder, Regensburg), J.Brederlau, E. Kaufmann, F. Schuster and C. Soellmann(Universitaetsklinik Wuerzburg), S. Frenzel and L.Pfeiffer (Unstrut-Hainich Kreiskrankenhaus Muehlhau-sen), S. Weber-Carstens, K. Haid, C. Melzer-Gartzke, C.von Heymann and B. Temmesfeld (Charite Universita-etsklinikum, Campus Mitte, Berlin).

Greece: Coordinator: Dimitrios Matamis (PapageorgiouGeneral Hospital, Thessaloniki).

H. Mouloudi (Ippokration General Hospital, Athens).Italy: Coordinator: Paolo Pelosi (Ospedale di Circolo

di Varese)A. Pesenti and N. Rossi (Ospedale San Gerardo, Mon-

za), D. Chiumello and L. Gattinoni (Ospedale MaggiorePoliclinico, Milano), P. Severgnini (Ospedale di Circolo diVarese), R. Fumagalli and A. Nikiforov (Ospedali Riunitidi Bergamo), S.Grasso (Ospedale di Venere, Bari).

Mexico: Coordinator: Jose Elizalde (Hospital ABC,Mexico DF).

825

P. Cerda (Centro Medico de las Americas, Merida), R.Mercado (Hospital Universitario de Monterrey), J.AlbeCastanon (Instituto mexicano del seguro social HECMNSXXI, Mexico DF).

The Netherlands: Michael Kuiper, P.H.M. Egbers andM. Koopmans (Medical Center Leeuwarden).

Peru: Coordinator: Ana Marıa Montanez.M. Contardo, J. Cerna and R.Roldan (Hospital

Edgardo Rebagliati Martins, Jesus Marıa), J. Zevallos andS. Alcabes (Hospital Guillermo Almenara Irigoyen, LaVictoria), C. Salcedo and D. Bruzone (Hospital NacionalDaniel Alcides Carrion, Callao), J. Quinones (Hospital deEmergencias Grau, Lima), M. Suarez Lazo (HospitalNacional Hipolito Unanue, El Agustino), A. Cifuentes(Hospital de Emergencias Jose Casimiro Ulloa, Miraflo-res), M. Mayorga (Clınica San Pablo, Lima).

Portugal: Coordinator: Rui Moreno (Hospital de SantoAntonio dos Capuchos, Lisboa).

P. Casanova (Hospitais da Universidade de Coimbra),R. Matos and A. L. Jardim (Hospital de Santo Antoniodos Capuchos, UCIP, Lisboa), A. Godinho (Hospital dosSAMS, UCI, Lisboa), P. Povoa (Hospital Sao FranciscoXavier, UCIM, Lisboa), P. Coutinho (Centro Hospitalarde Coimbra), L. Reis (Hospital de Sao Jose, Unidade deUrgencia Medica, Lisboa).

Saudi Arabia: Coordinator: Yaseen Arabi (King FahadNational Guard Hospital).

N. Abouchala (King Faisal Hospital), F. Hameed(King Khalid National Guard Hospital).

Spain: Coordinators: Nicolas Nin and Eva Tejerina(Hospital Universitario de Getafe).

F. Gordo (Fundacion Hospital de Alcorcon), R. Fer-nandez (Complejo Hospitalario Parc Taulı, Sabadell), R.de Pablo (Hospital Universitario Prıncipe de Asturias,Alcala de Henares), J. Ibanez (Hospital Son Dureta,Palma de Mallorca), E. Fernandez Mondejar (HospitalVirgen de las Nieves, Granada), F. del Nogal (HospitalSevero Ochoa, Leganes), F. Taboada (Hospital Central deAsturias, Oviedo), A. Garcıa Jimenez (Hospital Arqui-tecto Marcide, El Ferrol), Ll. Cabre and J. Morillas(Hospital de Barcelona-SCIAS), S. Macias (HospitalGeneral de Segovia), R. de Celis (Hospital de Galdakao),J. M. Anon (Hospital Virgen de la Luz, Cuenca),P. Ugarte(Hospital Marques de Valdecilla, Santander), T. Mut(Hospital de la Plana, Vila-Real), J. Diarte (ComplejoHospitalario de Ciudad Real), V. Sagredo (Hospital Clı-nico de Salamanca), M. Valledor (Hospital San Agustın,Aviles), G. Gonzalez and L. Rodrıguez (Hospital MoralesMeseguer, Murcia), V. Parra and E. Gomez (Hospital deSagunto), F. Jara (Hospital Mutua de Terrassa), J. M.Quiroga (Hospital de Cabuenes, Gijon), L. Arnaiz (Hos-pital Clınico Universitario de San Carlos, Madrid), A.Ayensa (Hospital Virgen de la Salud, Toledo), F. SuarezSippman (Fundacion Jimenez Dıaz), F. Charizosa (Hos-pital General de Jerez de la Frontera), J. A. RodrıguezSarrıa (Hospital de Elda), C. Homs (Hospital San Jorge,

Huesca), A. Dıaz Lamas (Hospital Cristal Pinor, Ou-rense), M. Leon (Hospital Arnau de Vilanova, Lleida), J.Allegue (Hospital Nuestra Senora del Rosell, Cartagena),M.Ruano (Hospital La Fe, Valencia).

Tunicia: Coordinator: Fekri Abroug (FattoumaBourguiba Monastir).

M. Besbes, J. Ben Khelil, K. Belkhouja and K. Ben-Romdhane (Hospital Abderrahmane Mami, Ariana), S. BenLakhal, S. Abdellatif and K. Bousselmi (La Rabta Tunis),M. Amamou and H. Thabet (CAMUR), L. Besbes andN. Nciri (Fattouma Bourguiba Monastir), M. Bouaziz, H.Kallel and M. Bahloul (Habib Bourguiba Sfax), S. ElA-trous, S. Merghli and M. Feki Hassen (Tahar Sfar Mahdia).

Turkey: Coordinator: Nahit Cakar (Istanbul MedicalFaculty, Istanbul).

R. Iscimen (Uludag University School of Medicine,Bursa), M. Kyzylkaya (College of medicine, AtaturkUniversity, Erzurum), B. Yelken (Osmangazi University,Eskisemir), I. Kati (Medical Faculty of Yuzuncu YilUniversity, Van), T. Guldem (Haydarpasa NumuneTeaching and Research Hospital, Istambul), U. Koca(Dokuz Eylun University, Istanbul), M. Cicek (InonuUniversity of Medical Faculty, Malatya), H. Sungurtekin(Pamukkale University Medical Faculty).

United States: Coordinator: Antonio Anzueto (Univer-sity of Texas Health Science Center, San Antonio, Texas).

A. C. Arroliga (Cleveland Clinic, Cleveland), O. Gajicand M. Ali (Mayo Clinic, Rochester), D. Ost, A. Fein,A. Kyprianou, L. Shulman and S. Chang (North ShoreUniversity Hospital, New York), J. S. Steingrub, M. A.Tidswell and K. Kozikowski (Baystate Medical Center,Springfield), C. A. Piquette and L. Morrow (CreightonUniversity Medical Center, Nebraska), P. Scheinberg and J.Green (Saint Joseph’s Hospital, Atlanta), L. Penogreen andK. Kannady (Georgia State University Kennestone), M.Moss, M. Mealer, and R. D. Restrepo (Grady HospitalGeorgia, Atlanta), H. E. Fessler, R. Brower, D. Hager andA. Scully (John Hopkins University Hospital, Baltimore), J.Beamis, D. E. Craven and W. Miner (Lahey Clinic MedicalCenter, Burlington), S. Blosser, K. Miller, L. Cornman andJ. Breidinger (Penn State Hershey Medical Center, Her-shey), J. T. Huggins and Ch. Strange (Medical University ofSouth Carolina, Charleston), N. S. Hill and L. Lawler(Tufts-New England Medical Center, Boston), M. Rembert(Newark Beth Israel Medical Center), H. K. Donnelly, J. D.D’Amico, R. G. Wunderink, N. Queseda and J. Topin(Northwestern Memorial Home Health University, Chi-cago), G. T. Kinasewitz and G. L. Lee (University ofOklahoma Health Sciences Center, Oklahoma City), J.Walls and V. Zimmer (Presbyterian Healthcare, Charlotte),A. X. Freire (Regional Medical Center, Memphis), C.Steven and L. Caskey (Louisiana State University HealthSciences Center, Shreveport), R. Dhand and L. A. Despins(University Hospital and Clinics MU Healthcare, Colum-bia), R. Hyzy, R. E. Dechert, C. Haas and D. Fickle(University of Michigan Medical Center), Ch. Burger and

826

L. Gambino (Mayo Clinic, Jacksonville), D. Marks and S.Benslimane (University of Texas Health Science Center,San Antonio), V. J. Cardenas Jr. (University of TexasMedical Branch Galveston), M. J. Wing and P. Krumpe(VA Sierra Nevada Health Care System, Reno), J. Truwitand M. Marshall (University of Virginia Health System,Charlottesville), D. L. Herr (Washington Hospital Center,Washington DC), R. D. Hite (Wake Forest Baptist HospitalMedical Center, Winston Salem), P. J. McShane and K. N.Olivier (Wilford Hall Medical Center, Texas), K. W.Presberg (Froedtert & Medical College, Milwaukee).

Uruguay: Coordinator: Javier Hurtado (CudamSanatorio Colon, Sanatorio IMPASA and Hospital deClınicas, Montevideo).

M. Borde, E. Echavarrıa, S. Gomez and M. Beron(Hospital Maciel, Montevideo), F. Villalba (Sanatorio Casade Galicia, Montevideo), I. Porras (Sanatorio CASMU 2,Montevideo), P. Cardinal, C. Surraco and V. Navarrete(Sanatorio CASMU 4, Montevideo), F. Rodrıguez andJ.C. Bagattini (Hospital Britanico, Montevideo), R. Garrido(Hospital Evangelico and Sanatorio IMPASA, Montevideo),S. Infanzon and J. Caraballo (Hospital Militar and CTI-SMI,Montevideo), C. Santos and A. Garcıa (Hospital de Clınicas,

Montevideo), R. Cal (CTI-SMI, Montevideo), G. Pittini andJ. Cabrera (Centro Nacional de Quemados, Montevideo),F. Bazzano and F. Domınguez (Hospital Pasteur, Colonia),P. Alzugaray, D. Gonzalez and M. Machado (SanatorioCAMOC, Carmelo), F. Torres (Sanatorio Mautone andAsistencial Medica de Maldonado, Maldonado), S. Mare-que, M. Korintan, F. Mora, E. Altieri, E. Gianoni, C. Fregosi,A. Crossi, G. Larrarte (Sanatorio CAAMS, Soriano),O. Pereira (Sanatorio COMTA, Tacuarembo), J. Baraibar(Hospital Regional de Tacuarembo), A. Soler (SanatorioCOMEPA, Paysandu), M. Rodrıguez Verde (Hospital Pay-sandu), M. Dıaz (Hospital de Salto and Sanatorio Uruguay,Salto), J. Martınez Ramos (Sanatorio Uruguay, Salto),I. Iturralde, W. Gonzalez and E. Cubas (SanatorioCAMDEL, Minas), A. Cataldo (Sanatorio CAMEDUR,Durazno), O. Rocha (Sanatorio GREMEDA, Artigas),A. Deicas (Sanatorio CASMU 2 and Sanatorio CASMU 4).

Venezuela: Coordinator Gabriel D’Empaire (Hospitalde Clınicas, Caracas).

R. Zerpa (Hospital Militar de Caracas), M. Narvez(Hospital Domingo Luciani, Caracas), F. Perez (Hospitalde Clınicas, Caracas), J. Espana (Hospital Universitariode Caracas).

References

1. Stock MC, Downs JB, Frolicher DA(1987) Airway pressure releaseventilation. Crit Care Med 15:462–466

2. Rose L, Hawkins M (2008) Airwaypressure release ventilation andbiphasic positive airway pressure: asystematic review of definitionalcriteria. Intensive Care Med 34:1766–1773

3. Putensen C, Wrigge H (2004) Clinicalreview: biphasic positive airwaypressure and airway pressure releaseventilation. Crit Care 8:492–497

4. Varpula T, Valta P, Niemi R, TakkunenO, Hynynen M, Pettila VV (2004)Airway pressure release ventilation as aprimary ventilatory mode in acuterespiratory distress syndrome. ActaAnesthesiol Scand 48:722–731

5. Seymour CW, Frazer M, Reilly PM,Fuchs BD (2007) Airway pressurerelease and biphasic intermittentpositive airway pressure ventilation: arethey ready for prime time? J Trauma62:1298–1308

6. Esteban A, Ferguson ND, Meade MO,Frutos-Vivar F, Apezteguia C,Brochard L, Raymondos K, Nin N,Hurtado J, Tomicic V, Gonzalez M,Elizalde J, Nightingale P, Abroug F,Pelosi P, Arabi Y, Moreno R, Jibaja M,D’Empaire G, Sandi F, Matamis D,Montanez AM, Anzueto A (2008)Evolution of mechanical ventilation inresponse to clinical research. Am JRespir Crit Care Med 177:170–177

7. Siau C, Stewart TE (2008) Current roleof high frequency oscillatory ventilationand airway pressure release ventilationin acute lung injury and acuterespiratory distress syndrome. ClinChest Med 29:265–275

8. Putensen C, Zech S, Wrigge H,Zinserling J, Stuber F, Von Spiegel T,Mutz N (2001) Long-term effects ofspontaneous breathing duringventilatory support in patients withacute lung injury. Am J Respir CritCare Med 164:43–49

9. Dart BW IV, Maxwell RA, Richart CM,Brooks DK, Ciraulo DL, Barker DE,Burns RP (2005) Preliminaryexperience with airway pressure releaseventilation in a trauma/surgicalintensive care unit. J Trauma 59:71–76

10. Putensen C, Mutz NJ, Putensen-Himmer G, Zinserling J (1999)Spontaneous breathing duringventilatory support improvesventilation–perfusion distribution inpatients with acute respiratory distresssyndrome. Am J Respir Crit Care Med159:1241–1248

11. Sydow M, Burchardi H, Ephraim E,Zielmann S, Crozier TA (1994) Longterm effects of two different ventilatorymodes on oxygenation in acute lunginjury: comparison of airway pressurerelease ventilation and volume-controlled inverse-ratio ventilation. AmJ Respir Crit Care Med 149:1550–1556

12. Rathgeber J, Schorn B, Falk B,Kazmaier S, Spiegel T, Burchardi H(1997) The influence of controlledmandatory ventilation (CMV),intermittent mandatory ventilation(IMV) and biphasic intermittentpositive airway pressure (BIPAP) onduration of intubation and consumptionof analgesics and sedatives: aprospective analysis in 596 patientsfollowing adult cardiac surgery. Eur JAnesthesiol 14:576–582

13. Weitzen S, Lapane KL, Toledano AY,Hume AL, Mor V (2004) Principles formodelling propensity scores in medicalresearch: a systematic literature review.Pharmacoepidemiol Drug Saf 13:841–853

827