Higher Levels of Spontaneous Breathing Reduce Lung Injury in Experimental Moderate Acute Respiratory Distress Syndrome* Nadja C. Carvalho, PhD1; Andreas Güldner, MD1; Alessandro Beda, PhD1,2; Ines Rentzsch, PhD1; Christopher Uhlig, MD1; Susanne Dittrich, MD1; Peter M. Spieth, MD1; Bärbel Wiedemann, PhD3; Michael Kasper, PhD4; Thea Koch, MD1; Torsten Richter, MD1; Patricia R. Rocco, MD, PhD5; Paolo Pelosi, MD6; Marcelo Gama de Abreu, MD, PhD1

Objectives: To assess the effects of different levels of spontaneous breathing during biphasic positive airway pressure/airway pressure release ventilation on lung function and injury in an experimental model of moderate acute respiratory distress syndrome. *See also p. 2459. 1 Department of Anesthesiology and Intensive Care Medicine, Pulmonary Engineering Group, University Hospital Dresden, Technische Universität Dresden, Dresden, Germany. 2 Department of Electronic Engineering, Federal University of Minas Gerais, Belo Horizonte, Brazil. 3 Institute for Medical Informatics and Biometry, Technische Universität Dresden, Dresden, Germany. 4 Institute of Anatomy, Medical Faculty Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany. 5 Laboratory of Pulmonary Investigation, Carlos Chagas Filho Biophysics Institute, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil.. 6 Department of Surgical Sciences and Integrated Diagnostics, IRCCS San Martino Hospital, University of Genoa, Genoa, Italy. This work was performed by the Pulmonary Engineering Group at the Department of Anesthesiology and Intensive Care Medicine, University Hospital Dresden, Technische Universität Dresden, Dresden, Germany. Drs. Carvalho and Güldner equally contributed. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/ccmjournal). Supported, in part, by grant GA 1256/6-1 from the German Research Council (DFG), Bonn, Germany. Dr. Gama de Abreu consulted for, received grant support from, lectured for, received support for educational presentations from, and received support for travel from Dräger Medical AG (Lübeck, Germany) and Novalung GmbH (Heilbronn, Germany). He and his institution have patents with Dräger Medical AG (Lübeck, Germany). The remaining authors have disclosed that they do not have any potential conflicts of interest. Address requests for reprints to: Marcelo Gama de Abreu, MD, PhD, Department of Anesthesiology and Intensive Care Medicine, Pulmonary Engineering Group, University Hospital Carl Dresden, Technische Universität Dresden, Fetscherstr 74, 01307 Dresden, Germany. E-mail: [email protected] Copyright © 2014 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/CCM.0000000000000605

e702

www.ccmjournal.org

Design: Multiple-arm randomized experimental study. Setting: University hospital research facility. Subjects: Thirty-six juvenile pigs. Interventions: Pigs were anesthetized, intubated, and mechanically ventilated. Moderate acute respiratory distress syndrome was induced by repetitive saline lung lavage. Biphasic positive airway pressure/airway pressure release ventilation was conducted using the airway pressure release ventilation mode with an inspiratory/expiratory ratio of 1:1. Animals were randomly assigned to one of four levels of spontaneous breath in total minute ventilation (n = 9 per group, 6 hr each): 1) biphasic positive airway pressure/airway pressure release ventilation, 0%; 2) biphasic positive airway pressure/airway pressure release ventilation, > 0–30%; 3) biphasic positive airway pressure/airway pressure release ventilation, > 30–60%, and 4) biphasic positive airway pressure/airway pressure release ventilation, > 60%. Measurements and Main Results: The inspiratory effort measured by the esophageal pressure time product increased proportionally to the amount of spontaneous breath and was accompanied by improvements in oxygenation and respiratory system elastance. Compared with biphasic positive airway pressure/airway pressure release ventilation of 0%, biphasic positive airway pressure/airway pressure release ventilation more than 60% resulted in lowest venous admixture, as well as peak and mean airway and transpulmonary pressures, redistributed ventilation to dependent lung regions, reduced the cumulative diffuse alveolar damage score across lungs (median [interquartile range], 11 [3–40] vs 18 [2–69]; p < 0.05), and decreased the level of tumor necrosis factor-α in ventral lung tissue (median [interquartile range], 17.7 pg/mg [8.4–19.8] vs 34.5 pg/mg [29.9–42.7]; p < 0.05). Biphasic positive airway pressure/ airway pressure release ventilation more than 0–30% and more than 30–60% showed a less consistent pattern of improvement in lung function, inflammation, and damage compared with biphasic positive airway pressure/airway pressure release ventilation more than 60%. Conclusions: In this model of moderate acute respiratory distress syndrome in pigs, biphasic positive airway pressure/airway November 2014 • Volume 42 • Number 11

Online Laboratory Investigations ­ ressure release ventilation with levels of spontaneous breath p higher than usually seen in clinical practice, that is, more than 30% of total minute ventilation, reduced lung injury with improved respiratory function, as compared with protective controlled mechanical ventilation. (Crit Care Med 2014; 42:e702–e715) Key Words: acute respiratory distress syndrome; airway pressure release ventilation; biphasic positive airway pressure; spontaneous breathing; ventilator-associated lung injury

P

atients with the acute respiratory distress syndrome (ARDS) require protective mechanical ventilation (MV) to support the respiratory function (1). Usually, MV is delivered by controlled or assist-controlled modes, with no or only minimal patient inspiratory effort, mandating the need for deeper sedation, increased circulatory drug support, and eventually muscle paralysis (2–4). However, controlled MV with muscle paralysis may result in diaphragmatic dysfunction (5–7). Inspiratory effort during MV may counteract the possible negative effects of controlled MV on hemodynamics, pulmonary function, and respiratory muscle dysfunction. Spontaneous breathing (SB) might decrease alveolar collapse in the most dependent lung zones, leading to less ventilatorinduced lung injury (VILI) (8–10). On the other hand, SB results in unpredictable stress/strain of lung tissue, and excessive respiratory effort may favor VILI.

Biphasic positive airway pressure (BIPAP) and airway pressure release ventilation (APRV) are time-cycled modes where the ventilator switches between two levels of continuous airway pressure (Paw), and SB is allowed throughout the respiratory cycle. In APRV, the cycling of the ventilator is completely independent from the respiratory effort. Conversely, in BIPAP, triggers can be used for cycling on and off, and SB can be supported by pressure at the lower Paw. An important advantage of BIPAP and APRV in ARDS is the capacity to maintain higher mean Paw compared with other modes of assisted ventilation (11, 12). Normally, BIPAP is used with inspiratory/expiratory (I:E) ratios of 1:1 or less, whereas APRV is used with inverse I:E. The contribution of SB to total minute ventilation can be modulated by setting different cycling times in both modes. Throughout the text, BIPAP/ APRV denotes time cycling of the ventilator independent of SB (as APRV), I:E = 1:1 (as BIPAP), and no pressure support. It has been suggested that SB should be in the range of 20–30% of total minute ventilation (13). To our knowledge, however, the optimal range of SB and inspiratory effort for lung protection has not been investigated. Thus, in this study, we evaluated the midterm effects of different levels of unsupported SB during BIPAP/APRV on morphofunctional and inflammatory variables in an experimental model of moderate ARDS. We hypothesized that lower to moderate but not higher levels of SB improve lung function and reduce inflammation and histological damage compared with controlled MV.

MATERIALS AND METHODS The Animal Care Committee of the State of Saxony, Germany, approved the study protocol.

Figure 1. Illustrative time courses of airway pressure (Paw) at different levels of inspiratory effort during biphasic positive airway pressure/airway pressure release ventilation (BIPAP/APRV). The values 0% (BIPAP/APRV, 0%), > 0–30% (BIPAP/APRV, > 0–30%), > 30–60% (BIPAP/APRV, > 30–60%), and > 60% (BIPAP/APRV, > 60%) represent the percentage of spontaneous breathing in total minute ventilation. During BIPAP/APRV, the ventilator cycles between two levels of Paw whereby spontaneous breathing without pressure support can occur throughout the respiratory cycle. The rate of changes between the lower and higher Paw is determined by inspiratory and expiratory time settings at the ventilator. Black lines represent Paw as set in the ventilator, whereby the inspiratory to expiratory ratio is kept at 1:1 at all levels of inspiratory effort. Red lines represent fluctuations in Paw resulting from spontaneous breathing activity, which are possible anytime during the respiratory cycle, but occurred mostly at the lower Paw level.

Premedication and Initial Mechanical Ventilator Settings Thity-six pigs (28.5–40.1  kg) were intramuscularly premedicated with midazolam (1 mg/ kg) and ketamine (10 mg/kg). An ear vein was punctured and IV anesthesia induced and maintained with midazolam (bolus = 0.5–1 mg/kg, followed by 1–2 mg/kg/hr) and ketamine (bolus = 3–4 mg/kg, followed by 10–18  mg/kg/hr); paralysis was achieved with atracurium (bolus = 3–4 mg/kg, followed by 1–2 mg/kg/hr). Animals had the trachea intubated with a cuffed endotracheal tube (8.0mm internal diameter) and lungs ventilated with the ventilator EVITA XL (Dräger Medical AG, Lübeck, Germany),

Critical Care Medicine

www.ccmjournal.org

e703

Carvalho et al

Figure 2. Tracings of flow, airway pressure (Paw), and esophageal pressure (Pes) at different levels of inspiratory effort during biphasic positive airway pressure/airway pressure release ventilation (BIPAP/APRV). The values 0% (BIPAP/APRV0%, A), > 0–30% (BIPAP/APRV>0–30%, B), > 30–60% (BIPAP/APRV>30–60%, C), and > 60% (BIPAP/APRV>60%, D) represent the percentage of spontaneous breathing in total minute ventilation. The inspiratory effort is characterized by negative spikes in Pes. Gray zones in the Paw and Pes tracings during BIPAP/APRV>0–30% (B) evidence mixed cycles, which are characterized by a drop in Pes, slight negative swings in Paw, and ventilator cycling from lower to higher Paw. Controlled and spontaneous cycles are also present, mainly during BIPAP/APRV0% (A) and BIPAP/APRV>60% (D), respectively (see text).

volume-controlled mode with a FIo2 1.0, tidal volume (VT) 10 mL/kg, positive end-expiratory pressure (PEEP) 5 cm H2O, I:E ratio 1:1, and respiratory rate to Paco2 35–45 mm Hg. Preparation The right anterior surface of the neck was infiltrated with 5 mL lidocaine 1%, followed by skin incision. An indwelling catheter was inserted through the external carotid artery for mean arterial pressure monitoring (Cardiac and Monitoring Systems - CMS Monitor, Philips, Böblingen, Germany). A pulmonary artery catheter (Opticath; Abbott, Abbott Park, IL) was advanced through the external jugular vein, and the mean pulmonary artery pressure continuously measured e704

www.ccmjournal.org

(CMS monitor). Urine was collected with a bladder catheter inserted during a minilaparotomy. An esophageal catheter (Erich Jaeger, Höchberg, Germany) was placed in the mid chest. A silicon belt with 16 electrodes was placed on the midchest circumference for electrical impedance tomography (EIT) (Evaluation Kit 2, Dräger Medical AG). The relative distribution of ventilation was estimated using images recorded during 5 minutes preceding each measurement time point. Intravascular volume was maintained with 10–15 mg/kg/hr E153 (Serumwerk Bernburg AG, Bernburg, Germany) and 6% hydroxyethyl starch as necessary (Fresenius kabi Deutschland GmbH, Bad Homburg, Germany). November 2014 • Volume 42 • Number 11

Online Laboratory Investigations

Table 1.

Respiratory Variables at Different Levels of Inspiratory Effort BIPAP/ APRV0%

BIPAP/ APRV>0–30%

5.73 ± 1.35

5.44 ± 1.53a

5.64 ± 1.22

5.59 ± 1.12a

 MinVentSB (% of total MinVent)

21.8 ± 16.3

43.3 ± 9.5b

66.3 ± 8.9b,c

 MinVentspont. cycles (%)

5.6 ± 4.9

37.9 ± 7.9b

59.7 ± 6.9b,c

Variable

Baseline 1

Injury

Baseline 2

MinVent (L/min)

5.62 ± 0.80 5.42 ± 0.82 6.43 ± 1.57

BIPAP/ APRV>30–60%

BIPAP/ APRV>60%

 MinVentcontr. cycles (%)

25.0 ± 23.4a,b

15.5 ± 14.3a,b

12.4 ± 10.2a,b

 MinVentmixed cycles (%)

70.6 ± 25.5

47.9 ± 16.6b

28.9 ± 14.2b

6.0 ± 0.1

5.6 ± 0.1

5.7 ± 0.1

 VTspont cycles (mL/kg)

3.0 ± 0.8

4.4 ± 0.2b

5.0 ± 0.2b

 VTcontr. cycles (mL/kg)

5.9 ± 0.2

4.9 ± 0.2

4.7 ± 0.2a,b

 VTmixed cycles (mL/kg)

6.5 ± 0.3

7.4 ± 0.4

8.1 ± 1.1b

20.0 ± 3.3a

16.6 ± 1.9a,b

15.8 ± 1.5ab

 Paw,peak;spont cycles (cm H2O)

14.3 ± 1.2

14.0 ± 1.0

14.2 ± 0.8

 Paw,peak;contr. cycles (cm H2O)

14.3 ± 1.2

20.0 ± 2.4

19.5 ± 1.3a,b

 Paw,peak;mixed cycles (cm H2O)

22.5 ± 3.6

21.3 ± 1.8

20.7 ± 1.5

15.5 ± 1.6

13.3 ± 0.9

12.7 ± 0.6a,b

 Paw,mean;spont cycles (cm H2O)

10.8 ± 0.4

10.6 ± 0.5

10.5 ± 0.1

 Paw,mean;contr. cycles (cm H2O)

15.8 ± 1.7

16.4 ± 1.6

16.8 ± 1.1b

 Paw,mean;mixed cycles (cm H2O)

15.6 ± 1.6

16.1 ± 1.2

16.3 ± 0.8

11.8 ± 3.6

9.1 ± 2.7a

8.7 ± 2.1a

 PL,peak;spont cycles (cm H2O)

4.7 ± 2.0

6.7 ± 2.5

7.5 ± 2.2

 PL,peak;contr. cycles (cm H2O)

11.3 ± 4.0

9.6 ± 3.9

8.0 ± 1.4a,b

 PL,peak;mixed cycles (cm H2O)

12.0 ± 3.5

13.1 ± 4.9

11.1 ± 2.4

5.6 ± 1.7

5.0 ± 2.6

3.9 ± 1.0a

 PL,mean;spont cycles (cm H2O)

3.0 ± 1.7

3.4 ± 2.6

2.8 ± 1.0

 PL,mean;contr. cycles (cm H2O)

5.3 ± 2.0a

6.3 ± 3.0

5.3 ± 1.2a

 PL,mean;mixed cycles (cm H2O)

5.6 ± 1.7

6.8 ± 2.7

5.8 ± 1.1

VT (mL/kg)

Paw,peak (cm H2O)

Paw,mean (cm H2O)

PL,peak (cm H2O)

PL,mean (cm H2O)

9.9 ± 0.1

19.4 ± 1.3

10.8 ± 0.8

10.1 ± 2.8

3.5 ± 1.9

9.8 ± 0.1

34.3 ± 3.2

16.9 ± 2.1

24.7 ± 5.6

9.3 ± 2.1

6.0 ± 0.2

27.0 ± 2.9

17.6 ± 1.6

15.0 ± 3.7

6.6 ± 1.8

6.0 ± 0.1

23.6 ± 2.9

17.5 ± 2.6

15.8 ± 5.9

7.0 ± 2.5

a

a

a

a

b

a,b

ab

a

a

BIPAP/APRV = biphasic positive airway pressure/airway pressure release ventilation, MinVent = minute ventilation, MinVentSB = measured fraction of MinVent from spontaneous breathing activity, MinVentspont. cycles = fraction of MinVent from spontaneous cycles, MinVentcontr. cycles = fraction of MinVent from controlled cycles, MinVentmixed cycles = fraction of MinVent from mixed cycles, VT = mean tidal volume, VTspont cycles = VT of spontaneous cycles, VTcontr. cycles = VT of controlled cycles, VTmixed cycles = VT of mixed cycles, Paw,peak = peak airway pressure, Paw,peak;spont cycles = peak airway pressures of spontaneous cycles, Paw,peak;contr. cycles = peak airway pressures of controlled cycles, Paw,peak;mixed cycles = peak airway pressures of mixed cycles, Paw,mean = mean airway pressure, Paw,mean;spont cycles = mean airway pressures of spontaneous cycles, Paw,mean;contr. cycles = mean airway pressures of controlled cycles, Paw,mean;mixed cycles = mean airway pressures of mixed cycles, PL,peak = peak transpulmonary pressure, PL,peak;spont cycles = peak transpulmonary pressures of spontaneous cycles, PL,peak;contr. cycles = peak transpulmonary pressures of controlled cycles, PL,peak;mixed cycles = peak transpulmonary pressures of mixed cycles, PL,mean = mean transpulmonary pressure, PL,mean;spont cycles = mean transpulmonary pressures of spontaneous cycles, PL,mean;contr. cycles = mean transpulmonary pressures of controlled cycles, PL,mean;mixed cycles = mean transpulmonary pressures of mixed cycles. a p < 0.05 vs BIPAP/APRV0%. b p < 0.05 vs BIPAP/APRV>0–30%. c p < 0.05 vs BIPAP/APRV>30–60%. Data are shown as mean ± sd. Variables were measured during BIPAP/APRV with different levels of spontaneous breathing activity in total minute ventilation: 0% (BIPAP/APRV0%), > 0–30% (BIPAP/APRV>0–30%), > 30–60% (BIPAP/APRV>30–60%), and > 60% (BIPAP/APRV>60%). Group effects were assessed with: 1) mixed linear model with the Tukey-Kramer procedure for adjustments (factors: time, group, and cycle's type); or 2) general linear model with the Sidak procedure for adjustments (factors: time and group). Statistical significance was accepted at p < 0.05.

Acquisition of Respiratory Signals, Respiratory Variables, and Classification of Cycles Airflow signal was obtained from the mechanical ventilator. Paw was monitored by a pressure transducer (163PC01D48-PCB; Critical Care Medicine

Sensortechnics GmbH, Puchheim, Germany) placed next the tracheal tube. The esophageal balloon catheter was connected to a second pressure transducer (163PC01D48-PCB; Sensortechnics GmbH) for esophageal pressure (Pes) monitoring. www.ccmjournal.org

e705

Carvalho et al

points before randomization and 60 minutes for time points thereafter. VT was computed through numerical integration of airflow. Inspiratory occlusion maneuvers were performed, and the drop in Paw 100 ms thereafter (P0.1) was determined. The esophageal pressure time product (PTP) was determined as the area above Pes. Inspiratory peak and mean Paw were computed (Paw,peak and Paw,mean, respectively). The transpulmonary pressure (PL) was determined as PL = Paw – Pes, and peak and mean PL (PL,peak and PL,mean, respectively) computed cycle-by-cycle.

Figure 3. Respiratory system mechanics (elastance [A] and resistance [B]), inspiratory drive (P0.1, [C]), and effort (pressure time product, PTP [D]) at different levels of inspiratory effort during biphasic positive airway pressure/ airway pressure release ventilation (BIPAP/APRV). The values 0% (BIPAP/APRV0%, A), > 0–30% (BIPAP/ APRV>0–30%, B), > 30–60% (BIPAP/APRV>30–60%, C), and > 60% (BIPAP/APRV>60%, D) represent the percentage of spontaneous breathing in total minute ventilation. Values are given as mean and sd. Differences between groups and time-group effects were tested with general linear model statistics and adjusted for repeated measurements according to the Sidak procedure. Interactions between type of respiratory cycle and groups were tested with mixed linear model and adjusted for repeated measurements with the Tukey-Kramer procedure. ns = not significant.

The position of the esophageal catheter was optimized as described elsewhere (14). Signals were acquired continuously with sampling frequencies of 180–200 Hz using a data acquisition card (NI USB-6210; National Instruments, Austin, TX) connected to a computer. Calculation of respiratory variables was performed off-line from respiratory signals. Recordings lasted 5 minutes for time e706

www.ccmjournal.org

Gas Exchange and Hemodynamics Arterial and mixed venous blood samples were analyzed for respiratory gases and pH using the ABL 505 (Radiometer, Copenhagen, Denmark) and for oxygen saturation and hemoglobin concentration with the OSM3 Hemoximeter (Radiometer). Measurements were done at 37°C and corrected for actual body temperature measured with the pulmonary artery catheter. Hemodynamic variables were assessed with the CMS monitor. Cardiac output was measured with the pulmonary artery catheter through the conventional thermodilution method, and oxygen-derived variables, including venous admixture, obtained using standard formulae.

Classification of Respiratory Cycles During BIPAP/APRV, two basic types of respiratory cycles can occur, namely controlled and spontaneous cycles. A third type of respiratory cycle may also exist if the inspiratory effort, that is, negative swings in Pes, occurs simultaneously with cycling of the ventilator to the higher Paw, so-called mixed cycles. Classification was performed automatically and checked visually by one of the investigators (N.C.C.). November 2014 • Volume 42 • Number 11

Online Laboratory Investigations

Respiratory Mechanics Variables The mechanical properties of the respiratory system were calculated in controlled respiratory cycles, as shown in Equation 1: . Paw (t ) = Rrs .V (t ) + Ers .V (t ) + P0 . where V is volume, V is airflow, t is time, and P0 is the total Paw at end-expiration. Rrs and Ers represent the respiratory system resistance and elastance, respectively. Values were computed off-line and averaged. Protocol for Measurements Following instrumentation, lungs were recruited with an Paw of 30 cm H2O during 30 seconds, followed by 15-minute stabilization, when baseline measurements were taken (baseline 1). Lung injury was induced by repetitive lung lavage with warmed (38°C) 0.9% saline solution (15), until Pao2/FIo2 less than 200 mm Hg for at least 30 minutes, when measurements were performed (injury). After injury, the ventilator was switched to the APRV mode, without SB activity (BIPAP/APRV0%), with FIo2 = 0.5, PEEP = 10 cm H2O, driving pressure to VT = 6 mL/kg, inspiratory time and expiratory time to I:E = 1:1, while resulting in Paco2 in the range of 45–60 mm Hg. The ventilator switched between levels of Paw, in a time-cycled manner, without use of inspiratory or expiratory flow triggers. These settings were kept for 30 minutes, when new measurements followed (baseline 2). After that, animals were randomly assigned to one of four levels of SB in total minute ventilation (n = 9 per group, 6 hr each), which are illustrated also in Figure 1: 1) 0% (BIPAP/APRV0%); 2) > 0–30% (BIPAP/APRV>0–30%); 3) > 30– 60% (BIPAP/APRV>30–60%) and; 4) > 60% (BIPAP/APRV>60%). Simultaneously, anesthesia depth was adapted according to the assigned levels of SB and inspiratory effort (midazolam = 1–2 mg/kg/hr, ketamine = 5–15 mg/kg/hr). In BIPAP/APRV0%, 30 minutes was allowed to compensate for the time needed to resume SB in others groups. By end of the therapy time, animals were anticoagulated with heparin (1,000 IU/kg) and killed with 2 g thiopental (Altana, Konstanz, Germany) and 50 mL KCl 1 M (Serumwerke Bernburg) intravenously. Postmortem Analysis Lungs were extracted at continuous Paw of 10 cm H2O. Bronchoalveolar lavage fluid (BALF) samples were obtained from the middle lobe. Diffuse alveolar damage (DAD) was evaluated in dependent and nondependent zones of the left lung by an expert blinded to groups as described previously by our group (16). Gene expression of tumor necrosis factor-(TNF)-α, interleukin (IL)-6, IL-8, amphiregulin, and tenascin-C (TNC) was analyzed using quantitative real-time polymerase chain reaction. Plasma, BALF, and lung tissue cytokine levels were measured by commercially available enzyme-linked immunosorbent assay kits according manufacturer’s instructions. Details of postmortem analysis are given in the supplemental data (Supplemental Digital Content 1, http://links.lww.com/ CCM/B58). Critical Care Medicine

Statistical Analysis The sample size calculation for testing the primary hypothesis (DAD score is reduced by lower to moderate levels of SB during BIPAP/APRV) was based on effects estimates obtained from pilot studies, as well as previous data of our group (mean value and dispersion, respectively) (17, 18). A sample size of nine animals per group would provide the appropriate power (1 – β = 0.8) to identify significant (α = 0.05) differences in DAD score, taking an effect size d = 2.0, equal number of animals per group, t test family, two-sided test, and multiple comparisons (n = 6) into account (α*= 0.008, Bonferroni adjusted). Data are presented as mean ± sd, unless stated otherwise. One-way analysis of variance (ANOVA) with Bonferroni adjustment, general linear model statistics with Sidak adjustment, Kruskal-Wallis test with Dunn adjustments, and mixed linear model with Tukey-Kramer adjustments were used as appropriate. For functional variables, comparability of groups at baseline 1, injury, and baseline 2 was tested with one-way ANOVA followed by Bonferroni post hoc test or H-test (Kruskal-Wallis) followed by Dunn post hoc test, as appropriate. p values were adjusted for multiple comparisons according to Bonferroni. Differences among and within groups (time effect 1–6 hr) were tested with general linear model statistics using baseline 2 as covariate and adjusted for repeated measurements according to the Bonferroni procedure. The statistical analysis was performed with SPSS (version 15.0 and version 19.0, SPSS, Chicago, IL) and SAS (version 9.3, SAS Institute, Cary, NC). The global significance level for all performed tests was α = 0.05.

RESULTS The number of lung lavages did not differ significantly among groups, whereas the dosage of anesthetics was higher in control animals, but comparable among SB groups (supplemental data, Supplemental Digital Content 2, http://links.lww. com/CCM/B59, and supplemental data, Supplemental Digital Content 3, http://links.lww.com/CCM/B60, respectively). Representative tracing records of airflow, Paw, and Pes are presented in Figure 2. SB occurred only rarely at the higher level of Paw (0.40% ± 0.43% in BIPAP/APRV>0–30%, 0.25% ± 0.36% in BIPAP/APRV>30–60%, and 0.09% ± 0.29% in BIPAP/APRV>60%). Table 1 and supplemental data (Supplemental Digital Content 4, http://links.lww.com/CCM/B61) depict the respiratory variables globally and partitioned according to the classification of cycle, respectively. The mean VT did not differ among groups (Table 1). In controlled cycles, VT was slightly lower during BIPAP/APRV>60% compared with BIPAP/APRV0% and BIPAP/APRV>0–30%, whereas during spontaneous and mixed cycles, VT was higher during BIPAP/APRV30–60% and BIPAP/APRV>60% compared with BIPAP/APRV>0–30% (supplemental data, Supplemental Digital Content 4, http://links. lww.com/CCM/B61). The total minute ventilation was higher during BIPAP/APRV0% compared with BIPAP/APRV>0–30% and BIPAP/APRV>60%. In addition, the contribution of SB cycles to total minute ventilation increased progressively from BIPAP/APRV>0–30% to BIPAP/APRV>60%, while during BIPAP/ www.ccmjournal.org

e707

Carvalho et al

Table 2.

Gas Exchange and Hemodynamic at Different Levels of Inspiratory Effort

Variable

Baseline 1

Injury

Baseline 2

1 Hr

 BIPAP/APRV0%

535.1 ± 24.1

67.4 ± 10.2

147.6 ± 16.7

185.7 ± 45.5

 BIPAP/APRV>0–30%

548.8 ± 23.5

64.4 ± 10.4

132.0 ± 33.5

254.2 ± 125.3

 BIPAP/APRV>30–60%

528.8 ± 29.4

63.6 ± 15.9

148.0 ± 31.5

359.3 ± 36.4

 BIPAP/APRV>60%

567.9 ± 37.1

63.7 ± 14.1

148.6 ± 27.6

346.3 ± 81.1

 BIPAP/APRV0%

15.0 ± 4.2

48.3 ± 12.0

32.8 ± 13.5

28.6 ± 16.0

 BIPAP/APRV>0–30%

11.6 ± 4.3

48.3 ± 9.1

39.3 ± 17.2

25.5 ± 18.0

 BIPAP/APRV>30–60%

14.4 ± 4.1

55.6 ± 18.1

36.2 ± 12.4

15.2 ± 4.3

 BIPAP/APRV>60%

10.3 ± 5.3

49.5 ± 13.0

30.2 ± 9.1

14.2 ± 4.0

 BIPAP/APRV0%

41.3 ± 3.1

67.5 ± 15.4

68.2 ± 9.5

69.3 ± 10.2

 BIPAP/APRV>0–30%

40.3 ± 2.8

62.0 ± 14.2

66.6 ± 9.7

63.4 ± 11.0

 BIPAP/APRV>30–60%

38.9 ± 3.5

52.2 ± 7.6

60.2 ± 4.1

62.6 ± 13.9

 BIPAP/APRV>60%

41.7 ± 1.9

57.4 ± 4.3

63.5 ± 7.1

63.0 ± 10.0

Pao2/FIo2 (mm Hg)

Shunt (%)

Paco2 (mm Hg)

Heart rate (L/min)  BIPAP/APRV0%

92 ± 14.7

87 ± 16.0

90 ± 19.5

88 ± 19.2

 BIPAP/APRV>0–30%

92 ± 7.7

86 ± 16.2

90 ± 19.5

78 ± 16.8

 BIPAP/APRV>30–60%

93 ± 12.2

87 ± 7.7

84 ± 9.2

68 ± 25.5

90 ± 111.4

85 ± 10.0

90 ± 10.7

79 ± 8.9

 BIPAP/APRV>60%

Mean pulmonary arterial pressure (mm Hg)  BIPAP/APRV0%

18.0 ± 2.7

29.2 ± 3.3

30.9 ± 3.1

29.3 ± 3.1

 BIPAP/APRV>0–30%

19.7 ± 2.6

33.7 ± 3.8

33.8 ± 3.9

27.6 ± 5.8

 BIPAP/APRV>30–60%

21.0 ± 3.5

31.2 ± 4.4

31.2 ± 4.0

25.0 ± 2.7

 BIPAP/APRV>60%

17.2 ± 1.6

29.6 ± 4.7

29.2 ± 2.6

22.3 ± 1.4

 BIPAP/APRV0%

72.2 ± 13.0

76.2 ± 10.3

81.2 ± 13.7

84.2 ± 14.1

 BIPAP/APRV>0–30%

76.9 ± 12.8

84.5 ± 13.5

87.1 ± 14.3

83.6 ± 8.1

 BIPAP/APRV>30–60%

73.2 ± 11.2

85.8 ± 6.8

89.8 ± 7.2

85.7 ± 7.3

 BIPAP/APRV>60%

76.8 ± 11.9

82.7 ± 9.0

88.6 ± 10.2

84.2 ± 10.5

 BIPAP/APRV0%

4.3 ± 1.0

3.6 ± 1.0

4.0 ± 1.5

3.7 ± 1.3

 BIPAP/APRV>0–30%

3.8 ± 1.0

3.3 ± 0.4

3.4 ± 0.4

3.2 ± 0.8

 BIPAP/APRV>30–60%

4.1 ± 0.9

3.7 ± 1.0

3.7 ± 0.9

3.3 ± 0.9

BIPAP/APRV>60%

3.9 ± 0.8

3.9 ± 0.8

4.1 ± 0.9

3.5 ± 0.6

Mean arterial blood pressure (mm Hg)

Cardiac output (L/min)

BIPAP/APRV = biphasic positive airway pressure/airway pressure release ventilation, NS = not significant. a p < 0.05 vs BIPAP/APRV0%. Data are shown as mean ± sd. Variables were measured during BIPAP/APRV with different levels of spontaneous breathing activity in total minute ventilation: 0% (BIPAP/APRV0%), > 0–30% (BIPAP/APRV>0–30%), > 30–60% (BIPAP/APRV>30–60%), and > 60% (BIPAP/APRV>60%). Group and time-group effects were assessed by general linear model statistics and adjusted for repeated measurements according to Sidak procedure. Statistical significance was accepted at p < 0.05.

e708

www.ccmjournal.org

November 2014 • Volume 42 • Number 11

Online Laboratory Investigations

Time-Group Effect

2 Hr

3 Hr

4 Hr

5 Hr

6 Hr

Group Effect

192.1 ± 44.1

211.7 ± 53.9

231 ± 67.7

241.4 ± 77.4

250.3 ± 82.0

NS

291.2 ± 131.8

319.2 ± 144.0

333.6 ± 142.1

348.7 ± 142.3

346.9 ± 135.1

NS

a

400.9 ± 37.4

442.8 ± 32.7

437.0 ± 40.7

446.1 ± 29.6

455.4 ± 32.6

NS

a

400.9 ± 82.5

438.9 ± 63.0

443.5 ± 73.2

441.4 ± 60.3

442.8 ± 67.2

NS

a

28.1 ± 14.3

24.3 ± 14.0

22.7 ± 14.5

23.0 ± 14.4

22.5 ± 14.8

NS

20.0 ± 15.7

18.5 ± 13.9

16.4 ± 12.6

14.0 ± 8.8

15.3 ± 10.0

NS

11.9 ± 1.9

9.6 ± 2.5

9.8 ± 2.3

10.2 ± 3.4

8.7 ± 1.1

NS

a

11.0 ± 4.1

9.35 ± 2.8

9.1 ± 3.0

9.5 ± 2.6

9.0 ± 2.6

NS

a

67.1 ± 8.5

65.9 ± 8.3

64.7 ± 7.2

63.7 ± 7.2

64.0 ± 5.0

NS

NS

61.0 ± 10.1

59.2 ± 9.9

59.7 ± 7.8

57.8 ± 7.8

58.9 ± 8.5

NS

NS

58.7 ± 12.8

58.9 ± 9.5

58.9 ± 12

57.2 ± 10.9

55.0 ± 11.6

NS

NS

60.1 ± 8.4

59.2 ± 8.3

57.9 ± 6.9

56.8 ± 8.0

56.8 ± 8.1

NS

NS

86 ± 17.8

86 ± 16.8

83 ± 15.6

80 ± 13.6

79 ± 12.3

NS

NS

75.3 ± 15.4

71 ± 11.2

76 ± 12.2

75 ± 10.2

75 ± 10.0

NS

NS

80 ± 11.1

79 ± 7.1

80 ± 9.4

80 ± 10.0

78 ± 10.9

NS

NS

81 ± 8.5

80 ± 8.2

83 ± 8.9

80 ± 4.7

82 ± 8.3

NS

NS

28.7 ± 3.3

28.5 ± 3.4

28.1 ± 4.5

27.5 ± 5.3

27.3 ± 5.8

NS

NS

28.0 ± 7.0

28.0 ± 7.0

27.7 ± 5.8

28.0 ± 6.9

26.4 ± 7.2

NS

NS

24.6 ± 4.7

24.6 ± 4.7

24.6 ± 3.8

23.4 ± 4.5

22.3 ± 3.3

NS

a

21.0 ± 1.8

21.0 ± 1.8

20.5 ± 1.5

20.2 ± 1.9

19.7 ± 1.8

NS

a

83.4 ± 13.4

82.1 ± 12.7

79.7 ± 13.9

76.5 ± 10.0

71.3 ± 9.4

NS

NS

83.1 ± 9.0

82.6 ± 10.1

82.0 ± 8.5

81.00 ± 7.5

80.7 ± 10.1

NS

NS

85.3 ± 5.2

83.0 ± 5.5

82.4 ± 7.8

78.6 ± 6.2

78.8 ± 7.7

NS

NS

82.8 ± 10.3

80.0 ± 11.0

80.4 ± 12.3

78.1 ± 12.0

79.0 ± 10.8

NS

NS

3.7 ± 1.0

3.6 ± 0.7

3.5 ± 0.8

3.3 ± 0.6

3.3 ± 0.6

NS

NS

2.9 ± 0.6

2.9 ± 0.5

2.8 ± 0.6

2.9 ± 0.6

2.9 ± 0.5

NS

NS

3.2 ± 0.8

3.2 ± 0.8

3.1 ± 0.6

3.2 ± 0.6

3.2 ± 0.6

NS

NS

3.5 ± 0.6

3.4 ± 0.5

3.4 ± 0.6

3.4 ± 0.5

3.6 ± 0.5

NS

NS

Critical Care Medicine

www.ccmjournal.org

e709

Carvalho et al

APRV>30–60% compared with BIPAP/APRV0% (Fig. 3). PTP and P0.1 progressively increased from BIPAP/APRV>0–30% to BIPAP/ APRV>60%, mainly due to SB cycles. BIPAP/APRV>0–30%, BIPAP/APRV>30–60%, and BIPAP/APRV>60% showed higher oxygenation compared with BIPAP/APRV0%. BIPAP/APRV30–60% and BIPAP/APRV>60% had lower venous admixture compared with BIPAP/APRV0%. Paco2 did not differ significantly among groups (Table 2). The mean pulmonary arterial pressure was reduced during BIPAP/APRV>30–60% and BIPAP/ APRV>60% compared with BIPAP/APRV0%. Other hemodynamic variables were comparable among groups. As shown in Figure 4, BIPAP/APRV>30–60% and BIPAP/ APRV>60%, but not BIPAP/APRV>0–30%, redistributed ventilation from central to dependent lung regions compared with BIPAP/APRV0%. The overall cumulative DAD score was lower in BIPAP/ APRV>30–60% and BIPAP/APRV>60% compared with BIPAP/ APRV0%, as depicted in Table 3. Only BIPAP/APRV>60% yielded lower overall cumulative DAD score than BIPAP/APRV>0–30%. BIPAP/APRV>30–60% and BIPAP/APRV>60% resulted in lower levels of TNF-α in nondependent lung areas than BIPAP/ APRV0% (Fig. 5). Also, BIPAP/APRV>30–60% yielded lower IL-6, but not IL-8 in those areas, as compared with BIPAP/APRV0%. Levels of IL-6, IL-8, and TNF-α in BALF, as well as of IL-6 and TNF-α in plasma, were comparable among groups (Table 4). Gene expression of IL-6, IL-8, TNF-α, TNC, and amphiregulin in lung tissue did not differ significantly among groups (Table 5).

DISCUSSION

Figure 4. Spatial distribution of lung ventilation by electric impedance tomography at different levels of inspiratory effort during biphasic positive airway pressure/airway pressure release ventilation (BIPAP/APRV). The values 0% (BIPAP/APRV0%), > 0–30% (BIPAP/APRV>0–30%), > 30–60% (BIPAP/APRV>30–60%), and > 60% (BIPAP/APRV>60%) represent the percentage of spontaneous breathing in total minute ventilation. The percentage of ventilation was quantified in gravitational nondependent, central, and dependent lung zones. Data are shown as mean and sd. Group and time-group effects were assessed by general linear model and adjusted for repeated measurements according to Sidak procedure. ns = not significant.

APRV>0–30%, mixed cycles were responsible for almost the total minute ventilation. As shown in Table 1, Paw,peak and Paw,mean were higher in BIPAP/APRV0% compared with other groups. PL,peak was lower during BIPAP/APRV>30–60% and BIPAP/APRV>60% than BIPAP0%. Furthermore, PL,mean was lower during BIPAP/ APRV>60% than BIPAP/APRV0%. Independent of the group, Paw,peak, Paw,mean, PL,peak, and PL,mean were lower during SB compared with controlled cycles. Ers was higher during BIPAP/APRV0% compared with other groups, whereas the respiratory Rrs was lower during BIPAP/ e710

www.ccmjournal.org

In a model of moderate ARDS in pigs, our main findings were that: 1) SB during BIPAP/APRV improved oxygenation and Ers; 2) compared with BIPAP/APRV0%, BIPAP/APRV>30–60% and BIPAP/APRV>60% resulted in lower venous admixture, decreased Paw and PL, redistributed ventilation to dependent lung regions, reduced overall lung damage, and decreased the levels of inflammation markers in ventral lung zones; and 3) BIPAP/APRV>60%, reduced overall lung damage compared with BIPAP/APRV>0–30%. To our knowledge, this is the first study comparing the effects of different levels of SB and inspiratory effort during MV on lung function, damage, and inflammation. We used a saline lung lavage since this model of ARDS has been claimed to provide an ideal way to test the effects of ventilatory strategies on the development of VILI (19, 20). Different strategies of assisted ventilation have been proposed in ARDS. We opted for BIPAP/APRV, because this strategy is able to maintain higher mean Paw (11, 12), which may be useful to stabilize lungs and improve their function (21, 22). Also, it allows modulation of inspiratory effort using cycling times. We maintained a constant I:E ratio in order to minimize differences in mean Paw and PL of controlled breaths among groups. We used SB not supported by pressure due to different reasons: 1) to maximize possible lung recruitment and 2) to avoid a mixture of time and flow-cycled breaths. BIPAP/APRV>0–30% was chosen because even in minimal levels of SB, it may lead to improvement November 2014 • Volume 42 • Number 11

Online Laboratory Investigations

Table 3.

Diffuse Alveolar Damage Score Variables

DAD Score

Alveolar edema

Interstitial edema

Hemorrhage

Alveolar infiltration

Interstitial infiltration

Epithelial destruction

Overdistension

Cumulative DAD score

Region

BIPAP/APRV0%

BIPAP/APRV>0–30%

BIPAP/APRV>30–60% BIPAP/APRV>60%

Nondependent

1 (0–12)

1 (0–15)

1 (0–12)

1 (0–6)

Dependent

0.5 (0–8)

1 (0–9)

1.5 (0–4)

1 (0–8)

Overall

1 (0–12)

1 (0–15)

1 (0–12)

1 (0–8)

Nondependent

2.5 (0–9)

3 (1–6)

2 (0–6)

2 (0–6)

Dependent

3 (0–8)

3 (1–6)

2.5 (0–6)a

2 (1–6)a

Overall

3 (0–9)

3 (1–6)

2 (0–6)a

2 (0–6)a

Nondependent

0 (0–9)

1 (0–12)

1 (0–9)

0 (0–4)

Dependent

0 (0–9)

1.5 (0–8)

1.5 (0–6)

0 (0–4)b

Overall

0 (0–9)

1 (0–12)

1 (0–9)

0 (0–4)a,b,c

Nondependent

0 (0–16)

0 (0–12)

0 (0–15)

0 (0–8)

Dependent

0 (0–20)

1 (0–8)

0 (0–2)

0 (0–6)a

Overall

0 (0–20)

1 (0–12)

0 (0–15)a

0 (0–8)a

Nondependent

4 (0–12)

4 (1–8)

2.5 (0–12)

2 (1–6)a

Dependent

3 (0–8)

3 (1–8)

3 (0–6)

2 (1–6)

Overall

3 (0–12)

3 (1–8)

3 (0–12)

2 (1–6)a,c

Nondependent

4 (0–16)

4 (1–16)

2 (0–9)a,c

2 (0–9)c

Dependent

4 (0–12)

4 (0–9)

3.5 (0–4)a

4 (1–9)a

Overall

4 (0–16)

4 (0–16)

2 (0–9)a,c

3 (0–9)a

Nondependent

3 (0–12)

3 (0–9)

3 (0–9)

3 (0–9)a,b

Dependent

2 (0–8)

3 (0–8)

1 (0–4)c

0 (0–9)

Overall

3 (0–12)

3 (0–9)

2 (0–9)

1 (0–9)a,c

Nondependent

20.0 (3–69)

18.5 (5–66)

14.5 (3–60)

11.5 (3–40)a,c

Dependent

16 (2–52)

18 (4–42)

13 (2–25)a

9.5 (3–33)a,c

Overall

17.5 (2–69)

17.5 (4–66)

14.0 (2–60)a

11.0 (3–40)a,c

a

DAD = diffuse alveolar damage, BIPAP/APRV = biphasic positive airway pressure/airway pressure release ventilation. a p < 0.05 vs BIPAP/APRV0%. b p < 0.05 vs BIPAP/APRV>30–60%. c p < 0.05 vs BIPAP/APRV>0–30%. DAD score variables at different lung’s zones: nondependent (ventral), dependent (dorsal), and overall, after BIPAP with different levels of inspiratory effort. Values are show as median and maximum and minimum. Differences between groups were tested with mixed linear model.

in the patient’s condition (4, 23), whereas the other settings were selected to progressively increase the inspiratory effort. It must be emphasized that BIPAP/APRV was used as a means to modulate SB rather than a particular mode of MV. The improvement of oxygenation with SB could be attributed to different but not mutually exclusive mechanisms, depending on the level of inspiratory effort: 1) redistribution of perfusion and 2) alveolar recruitment. The increase in Pao2/ FIo2 during BIPAP/APRV>0–30% was not associated with a change in the distribution of ventilation. A recent study of our group (24) showed that higher levels of SB during BIPAP/APRV, namely BIPAP/APRV>30–60% and BIPAP/APRV>60%, improve regional and global lung aeration, with minor redistribution Critical Care Medicine

of perfusion to nondependent lung zones. BIPAP/APRV>0–30% and BIPAP/APRV>60% showed lower minute ventilation, but Paco2 did not differ significantly among groups, suggesting a decrease in physiological dead space with SB. The highest PL,peak detected was ≈ 14 cm H2O, which is far below the safety limits of 25–27 cm H2O proposed by different authors (25, 26). In fact, higher levels of SB are able to recruit lungs and reduce the global stress/strain in moderate experimental ARDS (24). Since the injury model used does not affect the chest wall component of elastance, Ers changes are mainly determined by the lung component. Ers decreased in all groups with SB as compared with BIPAP/APRV0%. Possibly, lung recruitment occurred to some degree with all levels of SB, but we cannot www.ccmjournal.org

e711

Carvalho et al

Figure 5. Box plots of the levels of interleukin (IL)-6 (A), tumor necrosis factor (TNF)-α (B), and IL-8 (C) in lung tissue normalized to total protein in lung homogenate. Measurements were obtained at four levels of inspiratory effort during biphasic positive airway pressure/airway pressure release ventilation (BIPAP/APRV), leading to 0% (BIPAP/APRV0%), > 0–30% (BIPAP/APRV>0–30%), > 30–60% (BIPAP/APRV>30–60%), and > 60% (BIPAP/ APRV>60%) of spontaneous breathing in total minute ventilation. Group comparisons were performed with the Kruskal-Wallis test and adjusted for multiple measurements by means of Dunn procedure. *p < 0.05.

discriminate between tidal and permanent recruitment. Also, it is conceivable that variability of VT was responsible, at least in part, for lower lung stiffness, as suggested in previous studies (17, 27). Since BIPAP/APRV>30–60% and BIPAP/APRV>60%, but not BIPAP/APRV>0–30%, were also associated with redistribution of ventilation toward dorsal lung zones, it is conceivable that lower levels of SB, that is, BIPAP/APRV>0–30%, did not result in relevant recruitment of lung tissue. e712

www.ccmjournal.org

The fact that worsening of respiratory function was not observed during the treatment period could be explained by different reasons: 1) the values in the controlled group corresponded already to severe deterioration; 2) time was too short to observe deterioration; and 3) partial resolution of injury might have occurred, but this applies to all groups. Compared with controlled MV, BIPAP/APRV>30–60% and BIPAP/APRV>60%, but not BIPAP/APRV>0–30%, reduced damage November 2014 • Volume 42 • Number 11

Online Laboratory Investigations

Table 4.

Inflammation Markers in Plasma Bronchoalveolar Lavage Fluid

Variable

Plasma

TNF-α

IL-6

IL-8

TNF-α

IL-6

BIPAP/APRV0%

33.8 (15.3–74.2)

217.0 (152.0–37.1)

402.2 (222.7–1,323.5)

3.1 (1.5–5.6)

15.7 (2.6–32.5)

BIPAP/APRV>0–30%

18.1 (13.6–29.8)

198.4 (28.5–222.2) 131.4 (50.5–103.1)

7.3 (5.8–9.7)

6.3 (1.6–14.3)

BIPAP/APRV>30–60%

25.3 (19.0–8.1)

112.4 (91.7–245.2) 397.2 (166.6–2,982.5)

2.8 (1.9–7.4)

8.1 (5.1–13.5)

BIPAP/APRV>60%

24.3 (22.6–37.5)

Group effect

NS

97.4 (63.2–144.9) NS

287.1 (181.1–663.2)

7.7 (3.3–12.0)

13.0 (8.1–38.8)

NS

NS

NS

TNF = tumor necrosis factor-α, IL = interleukin, BIPAP/APRV = biphasic positive airway pressure/airway pressure release ventilation, NS = not significant. Cytokines levels in bronchoalveolar lavage fluid and concentration in plasma. Data are show as mean, sd for concentration in plasma, and group effects were assessed by general linear model statistics and adjusted for repeated measurements by means of Sidak procedure. For bronchoalveolar lavage fluid concentration, data are show as median and interquartile range, and statistical tests between groups were performed using Kruskal-Wallis and adjusted for multiple measurements by means of Dunn procedure. Statistical significance was accepted at p < 0.05.

and inflammation in lung tissue. There are three possible explanations for these observations: 1) reduced total minute ventilation during BIPAP/APRV with SB; 2) increased number of breaths with lower VT; and 3) decreased lung stress during moderate to high inspiratory effort. The higher minute ventilation during BIPAP/APRV0% may have increased mechanical stress/strain in the lung parenchyma. In fact, increased minute ventilation with lower VT has been shown to increase levels of IL-6 and activation of metalloproteases in lung tissue, as well as damage of the pulmonary interstitium in spontaneously breathing healthy rats (28). SB during BIPAP/APRV was associated with lower VT than controlled and mixed cycles. A previous study using dynamic CT showed namely that the reduced VT associated with SB is able to reduce tidal reaeration and hyperaeration in a similar model of ARDS in pigs (29), suggesting that higher degrees of SB in BIPAP/APRV could be more protective. The increase in inspiratory effort and drive, as determined by PTP and P0.1, respectively, likely contributed to recruitment of dependent lung zones, with increase in lung volume and the amount of normally aerated lung tissue, as previously suggested (30). The increase in aerated lung volume possibly contributed to lower Ers, which resulted in reduced PL. Recent experimental evidence supports the notion that higher levels of SB during BIPAP/APRV are associated with decreased global stress and strain during experimental ARDS (24). The observation that BIPAP/APRV>0–30% did not reduce lung damage compared with controlled MV is likely explained by the fact that mixed cycles may not generate enough inspiratory effort to recruit lungs. The fact that levels of inflammatory cytokines in BALF and plasma were comparable among groups could be explained by compartmentalization of the inflammatory process in the lung parenchyma, without spill over of those substances into the alveoli and circulation, respectively. Our observation that the gene expression of IL-6, IL-8, and TNF-α in lung tissue did not differ among groups might reflect different dynamics of gene activation and transcription into proteins, as well as of mechanical cell stress in lung tissue. The latter is supported by the fact that gene transcription of TNC and amphiregulin was also comparable among groups. Also, the fact that the concentrations of Critical Care Medicine

some cytokines differed in ventral lung tissue, whereas others did not, could be explained by distinct release dynamics. Possible Clinical Implications of the Findings This study has potentially relevant implications for clinical practice. Since BIPAP/APRV was used only as a means to modulate different degrees of SB, the results might be extrapolated to other modes of MV. Furthermore, the BIPAP/APRV strategy presented in this study can be easily implemented with different devices using the APRV mode with I:E = 1:1. Our findings suggest that to maximize lung protection during BIPAP/APRV with driving pressure targeted at mean VT of 6 mL/kg: 1) inspiratory and expiratory times, with a I:E ratio = 1:1, should be set to allow SB generating more than 30% of total minute ventilation; 2) monitoring of the inspiratory effort could be assessed at bedside either by the amount of SB in total minute ventilation or a P0.1 more than 2 cm H2O, depending on sedation levels; and 3) in contrast to gas exchange and respiratory system mechanics, EIT could be useful to detect a redistribution of ventilation toward dependent zones, working as a surrogate of appropriate degree of SB and inspiratory effort for lung protection. A recent study showed that in severe ARDS volume, assistcontrol ventilation is associated with higher morbidity and mortality when muscle paralysis is not used (31), which likely resulted from repetitive higher inspiratory transpulmonary pressures generated during ventilator cycling (32). The present data suggest that ventilatory modes based on pressure control and time cycling may be more appropriate for assisted MV in patients with ARDS. Our results might be of help for designing clinical trials aimed at investigating the potential of MV with SB for improving outcome in patients with mild to moderate ARDS. Particularly, such trials should consider that SB generating less than 30% of total minute ventilation may not be sufficient to reduce lung injury in patients with ARDS. Limitations This study has several limitations. First, the saline lung lavage model does not reproduce all features of the human ARDS and is highly recruitable. Thus, our findings cannot www.ccmjournal.org

e713

Carvalho et al

Table 5.

Gene Expression of Markers of Inflammation and Cell Stress in Lung Tissue IL-6

Group

IL-8

Ventral

Dorsal

Ventral

BIPAP/APRV0%

0.01 (0.01–0.05)

0.01 (0.01–0.07)

0.12 (0.07–0.16)

0.19 (0.08–0.65)

BIPAP/APRV>0–30%

0.01 (0.01–0.02)

0.01 (0.00–0.01)

0.15 (0.07–0.19)

0.14 (0.09–0.26)

BIPAP/APRV>30–60%

0.01 (0.01–0.03)

0.02 (0.01–0.03)

0.29 (0.15–0.31)

0.16 (0.12–0.32)

BIPAP/APRV>60%

0.01 (0.00–0.01)

0.01 (0.00–0.02)

0.12 (0.07–0.39)

0.19 (0.05–0.65)

NS

NS

NS

Group effect

Dorsal

NS

IL = interleukin, BIPAP/APRV = biphasic positive airway pressure/airway pressure release ventilation, NS = not significant. Gene expression of pulmonary mechanical stress markers (arbitrary units). Values are given as median and interquartile range for amphiregulin and tenascin-C. Statistical tests between groups were performed using Kruskal-Wallis and adjusted for multiple measurements by means of Dunn procedure. Statistical significance was accepted at p < 0.05.

be directly extrapolated to clinical conditions of poor lung recruitability or models of severe ARDS, for example, bleomycin with exposure to high concentrations of oxygen and infusion of TNF-α. Recently, Yoshida et al (32) showed that in presence of severe lung injury, SB may even worsen VILI. Second, we have not confirmed the diagnosis of ARDS by means of chest radiograph. Third, the observation time was limited to 6 hours and long-term effects were not assessed. Particularly, we cannot extrapolate our findings to later phases of ARDS, when fibrosis and remodeling of lungs take place. Fourth, our results might not be valid for modes that support the inspiratory effort, partially or totally, or even more typical settings of APRV, where the pressure release is relatively short. Fifth, the assessment of PL did not deliver any information of regional distribution of stress across the lungs. Sixth, we did not assess lung recruitment directly, but a recent study showed that BIPAP/APRV>30–60% and BIPAP/ APRV>60% are associated with increased lung aeration compared with controlled MV (24).

CONCLUSIONS In a saline lung lavage model of experimental moderate ARDS in pigs, SB during BIPAP/APRV improved respiratory function. Furthermore, levels of SB higher than usually seen in clinical practice, that is, more than 30% of total minute ventilation, reduced lung injury, as compared with protective controlled MV.

REFERENCES

1. Young JD, Sykes MK: Assisted ventilation. 1. Artificial ventilation: History, equipment and techniques. Thorax 1990; 45:753–758 2. Putensen C, Muders T, Varelmann D, et al: The impact of spontaneous breathing during mechanical ventilation. Curr Opin Crit Care 2006; 12:13–18 3. Burchardi H: New strategies in mechanical ventilation for acute lung injury. Eur Respir J 1996; 9:1063–1072 4. Putensen C, Hering R, Wrigge H: Controlled versus assisted mechanical ventilation. Curr Opin Crit Care 2002; 8:51–57 5. Vassilakopoulos T: Ventilator-induced diaphragm dysfunction: The clinical relevance of animal models. Intensive Care Med 2008; 34:7–16

e714

www.ccmjournal.org

6. Decramer M, Gayan-Ramirez G: Ventilator-induced diaphragmatic dysfunction: Toward a better treatment? Am J Respir Crit Care Med 2004; 170:1141–1142 7. Levine S, Nguyen T, Taylor N, et al: Rapid disuse atrophy of diaphragm fibers in mechanically ventilated humans. N Engl J Med 2008; 358:1327–1335 8. Putensen C, Wrigge H: Clinical review: Biphasic positive airway pressure and airway pressure release ventilation. Crit Care 2004; 8:492–497 9. Neumann P, Wrigge H, Zinserling J, et al: Spontaneous breathing affects the spatial ventilation and perfusion distribution during mechanical ventilatory support. Crit Care Med 2005; 33: 1090–1095 10. Carvalho AR, Spieth PM, Pelosi P, et al: Pressure support ventilation and biphasic positive airway pressure improve oxygenation by redistribution of pulmonary blood flow. Anesth Analg 2009; 109:856–865 11. Yoshida T, Rinka H, Kaji A, et al: The impact of spontaneous ventilation on distribution of lung aeration in patients with acute respiratory distress syndrome: Airway pressure release ventilation versus pressure support ventilation. Anesth Analg 2009; 109:1892–1900 12. Myers TR, MacIntyre NR: Respiratory controversies in the critical care setting. Does airway pressure release ventilation offer important new advantages in mechanical ventilator support? Respir Care 2007; 52:452–458 13. Putensen C, Räsänen J, López FA: Ventilation-perfusion distributions during mechanical ventilation with superimposed spontaneous breathing in canine lung injury. Am J Respir Crit Care Med 1994; 150:101–108 14. Lanteri CJ, Kano S, Sly PD: Validation of esophageal pressure occlusion test after paralysis. Pediatr Pulmonol 1994; 17:56–62 15. Lachmann B, Robertson B, Vogel J: In vivo lung lavage as an experimental model of the respiratory distress syndrome. Acta Anaesthesiol Scand 1980; 24:231–236 16. Spieth PM, Knels L, Kasper M, et al: Effects of vaporized perfluorohexane and partial liquid ventilation on regional distribution of alveolar damage in experimental lung injury. Intensive Care Med 2007; 33:308–314 17. Spieth PM, Carvalho AR, Pelosi P, et al: Variable tidal volumes improve lung protective ventilation strategies in experimental lung injury. Am J Respir Crit Care Med 2009; 179:684–693 18. Spieth PM, Carvalho AR, Güldner A, et al: Pressure support improves oxygenation and lung protection compared to pressure-controlled ventilation and is further improved by random variation of pressure support. Crit Care Med 2011; 39:746–755 19. Matute-Bello G, Frevert CW, Martin TR: Animal models of acute lung injury. Am J Physiol Lung Cell Mol Physiol 2008; 295:L379–L399 20. Matute-Bello G, Downey G, Moore BB, et al; Acute Lung Injury in Animals Study Group: An official American Thoracic Society workshop report: Features and measurements of experimental November 2014 • Volume 42 • Number 11

Online Laboratory Investigations

Tumor Necrosis Factor-α Ventral

Dorsal

Amphiregulin Ventral

Tenascin-C Dorsal

Ventral

Dorsal

0.28 (0.15–0.50)

0.50 (0.26–0.64)

0.27 (0.25–0.38)

0.42 (0.29–0.61)

0.08 (0.04–0.37)

0.28 (0.15–0.50)

0.28 (0.15–0.34)

0.21 (0.18–0.39)

0.18 (0.07–0.39)

0.18 (0.17–0.35)

0.17 (0.08–0.26)

0.12 (0.10–0.17)

0.09 (0.09–0.23)

0.15 (0.13–0.32)

0.42 (0.16–0.64)

0.21 (0.15–0.73)

0.09 (0.03–0.15)

0.10 (0.09–0.15)

0.20 (0.19–0.33)

0.38 (0.21–0.61)

0.18 (0.09–0.91)

0.14 (0.09–0.93)

0.15 (0.10–0.20)

0.11 (0.08–0.17)

NS

NS

NS

acute lung injury in animals. Am J Respir Cell Mol Biol 2011; 44: 725–738 21. Borges JB, Carvalho CR, Amato MB: Lung recruitment in patients with ARDS. N Engl J Med 2006; 355:319–320; author reply 321 22. Gattinoni L, Caironi P, Cressoni M, et al: Lung recruitment in patients with the acute respiratory distress syndrome. N Engl J Med 2006; 354:1775–1786 23. Putensen C, Zech S, Wrigge H, et al: Long-term effects of spontaneous breathing during ventilatory support in patients with acute lung injury. Am J Respir Crit Care Med 2001; 164:43–49 24. Güldner A, Braune A, Carvalho N, et al: Higher levels of spontaneous breathing induce lung recruitment and reduce global stress/strain in experimental lung injury. Anesthesiology 2014; 120:673–682 25. Chiumello D, Carlesso E, Cadringher P, et al: Lung stress and strain during mechanical ventilation for acute respiratory distress syndrome. Am J Respir Crit Care Med 2008; 178:346–355 26. Grasso S, Terragni P, Birocco A, et al: ECMO criteria for influenza A (H1N1)-associated ARDS: Role of transpulmonary pressure. Intensive Care Med 2012; 38:395–403

Critical Care Medicine

NS

NS

NS

27. Gama de Abreu M, Spieth PM, Pelosi P, et al: Noisy pressure support ventilation: A pilot study on a new assisted ventilation mode in experimental lung injury. Crit Care Med 2008; 36:818–827 28. Moriondo A, Marcozzi C, Bianchin F, et al: Impact of respiratory pattern on lung mechanics and interstitial proteoglycans in spontaneously breathing anaesthetized healthy rats. Acta Physiol (Oxf) 2011; 203:331–341 29. Gama de Abreu M, Cuevas M, Spieth PM, et al: Regional lung aeration and ventilation during pressure support and biphasic positive airway pressure ventilation in experimental lung injury. Crit Care 2010; 14: R34 30. Wrigge H, Zinserling J, Neumann P, et al: Spontaneous breathing improves lung aeration in oleic acid-induced lung injury. Anesthesiology 2003; 99:376–384 31. Papazian L, Forel JM, Gacouin A, et al; ACURASYS Study Investigators: Neuromuscular blockers in early acute respiratory distress syndrome. N Engl J Med 2010; 363:1107–1116 32. Yoshida T, Uchiyama A, Matsuura N, et al: The comparison of spontaneous breathing and muscle paralysis in two different severities of experimental lung injury. Crit Care Med 2013; 41:536–545

www.ccmjournal.org

e715

Higher levels of spontaneous breathing reduce lung injury in experimental moderate acute respiratory distress syndrome.

To assess the effects of different levels of spontaneous breathing during biphasic positive airway pressure/airway pressure release ventilation on lun...
1MB Sizes 0 Downloads 5 Views