Ultra-protective mechanical ventilation without extra-corporeal carbon dioxide removal for acute respiratory distress syndrome

Abstract

Background

Tidal hyperinflation can still occur with mechanical ventilation using low tidal volume (LVT) (6 mL/kg predicted body weight (PBW)) in acute respiratory distress syndrome (ARDS), despite a well-demonstrated reduction in mortality.

Methods

Retrospective chart review from August 2012 to October 2014. Inclusion: Age >18years, PaO₂/FiO₂<200 with bilateral pulmonary infiltrates, absent heart failure, and ultra-protective mechanical ventilation (UPMV) defined as tidal volume (VT) <6 mL/kg PBW. Exclusion: UPMV use for <24 h. Demographics, admission Acute Physiology and Chronic Health Evaluation II (APACHE II) scores, arterial blood gas, serum bicarbonate, ventilator parameters for pre-, during, and post-UPMV periods including modes, VT, peak inspiratory pressure (PIP), plateau pressure (Pplat), driving pressure, etc. were gathered. We compared lab and ventilator data for pre-, during, and post-UPMV periods.

Results

Fifteen patients (male:female = 7:8, age 42.13 ± 11.29 years) satisfied criteria, APACHEII 20.6 ± 7.1, mean days in intensive care unit and hospitalization were 18.5 ± 8.85 and 20.81 ± 9.78 days, 9 (60%) received paralysis and 7 (46.67%) required inotropes. Eleven patients had echocardiogram, 7 (63.64%) demonstrated right ventricular volume or pressure overload. Eleven patients (73.33%) survived. During-UPMV, VT ranged 2–5 mL/kg PBW(3.99 ± 0.73), the arterial partial pressure of carbon dioxide (PaCO₂) was higher than pre-UPMV values (84.81 ± 18.95 cmH₂O vs. 69.16 ± 33.09 cmH₂O), but pH was comparable and none received extracorporeal carbon dioxide removal (ECCO₂-R). The positive end-expiratory pressure (14.18 ± 7.56 vs. 12.31 ± 6.84 cmH2O), PIP (38.21 ± 12.89 vs. 32.59 ± 9.88), and mean airway pressures (19.98 ± 7.61 vs. 17.48 ± 6.7 cm H₂O) were higher during UPMV, but Pplat and PaO₂/FiO₂ were comparable during- and pre-UPMV. Driving pressure was observed to be higher in those who died than who survived (24.18 ± 12.36 vs. 13.42 ± 3.25).

Conclusion

UPMV alone may be a safe alternative option for ARDS patients in centers without ECCO₂-R.

Keywords

Mechanical ventilation low tidal volume acute respiratory distress syndrome hypercapnia

Introduction

Mechanical ventilation in acute respiratory distress syndrome (ARDS) can induce ventilator-induced lung injury (VILI) from barotrauma, hyperinflation, and cyclical recruitment and de-recruitment of diseased lung.¹ Lung protective ventilation adopting low tidal volume (LV_T) using 6 mL/kg of predicted body weight (PBW) while limiting plateau pressures (P_plat) to <30 cm H₂O, demonstrated a 22% mortality reduction when compared to using tidal volumes 12 mL/kg PBW.^2,3 However, tidal hyperinflation may still occur leading to hypothetical considerations to use tidal volume <6 mL/kg PBW.^4,5 A nonrandomized trial in 2009 used tidal volumes that ranged from 3.7 to 4.6 mL/kg PBW in conjunction with extra-corporeal carbon dioxide removal (ECCO₂-R) noted improved inflammatory markers and lung morphology.⁶ Bein et al.⁷ compared LV_T to ultra-low tidal volumes <3 mL/kg PBW in combination with ECCO₂-R and reported that survivors with greater hypoxemia in the ECCO₂-R group had less ventilator days. A physiologic study using tidal volumes 4 mL/kg PBW and compensatory tachypnea to maintain minute ventilation (MV) revealed a reduction in cyclic atelectasis and hyperinflation.⁸ In a systematic review, limited by significant technological and practice variations in the use of ECCO₂-R across the few studies (2 randomized controlled trials and 12 observational studies), a mortality benefit favoring ECCO₂-R was not observed.⁹ We performed a chart review to examine the outcomes of ARDS patients ventilated with tidal volumes <6 mL/kg of PBW without the use of ECCO₂-R (ultra-protective mechanical ventilation (UPMV)). The intent of this review is to ascertain whether tidal volume can be safely lowered in hypoxemic patients with P_plat >30 cmH₂O.

Methods

At this institution, we routinely use LV_T for mechanical ventilation in ARDS patients targeting P_plat <30 cmH₂O per ARDS-net protocol; however, changes can occur per physician discretion. In patients with persistent P_plat >30 cmH₂O, in an attempt to mitigate lung injury, we reduced the tidal volume to <4 mL/kg if oxygenation, acceptable acid base, and hemodynamic status could be maintained in the presence of hypercapnia. Our medical intensive care unit is not equipped with ECCO₂-R capabilities. We selected the interval between August 2012 and October 2014 for chart review, because of the abundance of patients with ARDS during this period.¹⁰ The University of Missouri Institutional Review Board (IRB) approved this study (IRB#1213901HS). The local instance of Informatics for Integrating Biology and Bedside (i2b2) identified patients aged 18–64 years in the electronic medical records (EMR) database, diagnosed with ARDS according to Berlin definition, who were placed on invasive mechanical ventilation for the study period. Data on 1310 patients were loaded into the Research Electronic Data Capture (REDCap) application hosted at the University of Missouri, a secure web-based application designed to support data capture for research studies, providing (1) an intuitive interface for validated data entry; (2) audit trails for tracking data manipulation and export procedures; (3) automated export procedures for seamless data downloads to common statistical packages; and (4) procedures for importing data from the external sources.¹¹

The following filters were then applied to the EMR to identify the patients. Inclusion: Age >18 years, ratio of arterial partial pressure of oxygen to fraction of inspired oxygen (PaO₂/FiO₂) <200 with bilateral pulmonary infiltrates and no evidence of heart failure, and use of an UPMV strategy defined by tidal volume <6 mL/kg PBW. Of note, all patients had a PEEP level ≥5 cmH₂0. The patients in whom UPMV was used for <24 h were excluded. Demographics, Acute Physiology and Chronic Health Evaluation II (APACHE II) scores at admission and serial data of arterial blood gas, serum bicarbonate, and ventilator parameters for pre- and post-implementation of UPMV including mode, set tidal volume, peak inspiratory pressure (PIP), P_plat, etc. were gathered. Data were compiled on an excel spreadsheet and we performed descriptive statistics. Continuous variables were expressed as mean ± standard deviation (SD) or median and range as appropriate. As we did not have a control group and because of the relatively low numbers and retrospective nature of our study, we did not perform further comparative statistical analysis.

Results

Fifteen patients satisfied criteria with complete data. Demographic data are summarized in Table 1. Mean age was 42.13 ± 11.29 years and body mass index (BMI) was 29.24 ± 8.9 kg/m². Nine (60%) received paralysis and 7 (46.67%) required inotropes. Eleven patients (73.33%) survived and four (26.67%) died. Four patients (36.36%) were discharged home, six (54.55%) were discharged to long-term acute care (LTAC) facilities, and one to a rehabilitation facility. Of the 11 patients (8 pre-UPMV period and 3 during-UPMV) who had echocardiography at admission or within a few days, 7 (63.64%) had evidence for right ventricular volume or pressure overload (5 pre-UPMV and 2 during-UPMV). Two patients (13.33%) developed pneumothorax and one of them also had pneumo-mediastinum.

Table 1.

Demographic and baseline characteristics at admission.

Total number of patients (Male: Female)	15 (7:8)
Age (years)^a	42.13 ± 11.29
Height (m)^a	1.71 ± 0.12
Weight (kg)^a	85.14 ± 24.43
Body mass index (kg/m²)^a	29.24 ± 8.9
APACHE II score^a	20.6 ± 7.1
Number of cases requiring inotropes	7 (46.67%)
Number of cases requiring paralysis	9 (60%)
Number survived	11 (73.33%)
ICU days^a	18.5 ± 8.85
Hospital days^a	20.81 ± 9.78

Data expressed as mean ± standard deviation (SD).

APACHE II: Acute Physiology and Chronic Health Evaluation II; ICU: intensive care unit.

The ventilator and blood gas parameters for during-, pre-, and post-UPMV values are given in Table 2. Pre-UPMV tidal volume ranged 5–8 mL/kg of PBW, whereas during UPMV it was 2–5 mL/kg of PBW (12(80%) had tidal volume ≤4 mL/kg PBW). Of the 7 (46.67%) patients who had the lowest tidal volumes, 3.5 mL/kg of PBW was utilized in two patients (13.3%), 3 mL/kg of PBW in 4 (26.67%), and 2 mL/kg of PBW in one (6.67%). Only one patient received extra-corporeal membrane oxygenation for severe hypoxic respiratory failure for approximately 72 h prior to demise. The arterial partial pressure of carbon dioxide (PaCO₂) during-UPMV was higher than pre-UPMV values, but pH was similar for all periods. Intravenous bicarbonate was administered in one patient and initiated 24 h before UPMV for a pH 7.0 and continued for 24 h after initiation of UPMV (5000 mL of 0.45% sodium chloride with each liter containing 75 mEq sodium bicarbonate). Positive end-expiratory pressure (PEEP) and mean airway pressures (MAP) were higher during-UPMV than pre-UPMV. Peak inspiratory pressures were higher during the-UPMV period than the pre-UPMV timeframe, but P_plat was similar during the UPMV and pre-UPMV periods. Likewise, FiO₂ and PaO₂/FiO₂ were comparable between pre- and during-UPMV intervals (Table 2). All four patients who died received paralysis and their driving pressures (determined by the difference between P_plat and the corresponding PEEP) were higher (24.18 ± 12.36) when compared to the five patients who received paralysis and survived (13.42 ± 3.25).

Table 2.

Comparison of ventilatory and laboratory parameters prior to, during, and after ultra-protective ventilation.

Parameters	Pre-UPMV	During-UPMV	Post-UPMV
Ventilator days	3.5 (0–17)	5 (1–25)	2.5 (0–27)
Tidal volume (mL/kg IBW)	6.48 ± 0.83	3.99 ± 0.73	6.86 ± 0.13
Mean airway pressure (cm H₂O)	17.48 ± 6.7	19.98 ± 7.61	15.07 ± 5.13
PIP (cm H₂O)	32.59 ± 9.88	38.21 ± 12.89	28.22 ± 10.67
P_plat (cm H₂O)	37.33 ± 7.52	36.75 ± 8.54	31.27 ± 3.04
PEEP (cm H₂O)	12.31 ± 6.84	14.18 ± 7.56	12.41 ± 4.96
Driving pressure (cm H₂O)	No data	18.2 ± 9.73^a	No data
MV (L/min)	9.40 ± 2.32	8.06 ± 2.23	9.07 ± 2.31
pH	7.26 ± 0.12	7.24 ± 0.10	7.27 ± 0.10
PaCO₂ (cm H₂O)	69.16 ± 33.09	84.81 ± 18.95	No data
PaO₂/FiO₂	135.14 ± 100.11	131.49 ± 63.53	No data
FiO₂	0.55 ± 0.23	0.54 ± 0.20	0.48–0.24
Respiratory rate	22.0 ± 7.57	26 + 9.02	22.0 ± 8.22

All data are expressed as mean ± SD or median (range) as appropriate.

All data from paralyzed patients. Driving pressures were lower in those who survived (N = 5, 13.42 ± 3.25) than who died (N = 4, 24.18 ± 12.36). FiO₂: fraction of inspired oxygen; MV: minute ventilation; P_plat: plateau pressure; PaO₂: arterial partial pressure of oxygen; PEEP: positive end-expiratory pressure; PIP: peak inspiratory pressure; UPMV: ultra-protective mechanical ventilation.

Discussion

Mechanical ventilation using LV_T, higher PEEP, and lower P_plat improved the survival in ARDS patients in randomized trials, the benefit primarily attributed to less VILI from reduced tidal stretch (with lower V_T) and atelectrauma (by using higher PEEP).^2,3,12–14 Utilizing LV_T invariably results in hypercapnia, which is often feared to worsen acidosis and cause deleterious systemic effects, especially in patients with myocardial and central nervous system affection, leading to a therapeutic dilemma. In our study, during-UPMV, the lowest tidal volume was 2 mL/kg (approaching dead space ventilation), but acidosis (pH) did not worsen despite a rise in the arterial PaCO₂, mitigating the need for ECCO₂-R. Although our center has limited capabilities (number of trained staff and equipment) for extra-corporeal membrane oxygenation strictly within the division of cardiothoracic surgery and mainly for cardio-pulmonary bypass surgery purposes, it is unavailable for MICU due to nonavailability of a dedicated perfusionist. All prior studies, where UPMV was employed, have used ECCO₂-R to ameliorate worsening hypercapnic acidosis, but from our data, it is evident that UPMV may be carefully and safely employed in ARDS patients without the use of ECCO₂-R.^6,7 With a corresponding rise in the respiratory rate to maintain reasonable minute ventilation, most patients tolerated UPMV without the need for a buffer or ECCO₂-R. Interestingly, there is further evidence from animal studies for direct beneficial physiological effects of hypercapnia on the immune system, and possibly lung compliance and ventilation-perfusion ratio (V/Q).^15–19 This has not been confirmed in the clinical trials involving human subjects and the direct effect of hypercapnia in patients with ARDS remains unclarified.²⁰ Like pH, oxygenation and P_plat were indifferent in our study, which is most likely from the use of higher PEEP during-UPMV that may have mitigated atelectrauma. Current evidence for guiding physicians towards an optimal strategy for balancing hypercapnic acidosis and specific ventilator strategies remains controversial and hence ventilator strategies must be individualized to patient characteristics e.g. the presence or the absence of myocardial or central nervous system co-morbidities, severity of ARDS, observed effects of hypercapnic acidosis and its severity, etc.^21–23

As mentioned, of the 11 patients who had echocardiography (at admission or within the first few days), 7 (63.64%) had evidence of either or both right ventricular volume or pressure overload. A prospective observational study involving 11 adult subjects with severe ARDS, keeping a constant P_plat (within 20–25 cm H₂O), compared various ventilator strategies for an hour each as follows: low or high PEEP with high respiratory rate (compensating for low V_T) or high PEEP and low respiratory rate (compensating for dead space), and evaluated cardiac function using trans-esophageal echocardiography along with alveolar dead space and recruitment. Blood pH was lower and PaCO₂ higher in the groups with high PEEP, who also demonstrated right ventricular dilation, left ventricular deformation, and a decline in the cardiac index.²⁴ In our study, none had echocardiography performed both in the pre-UPMV and during-UPMV periods to ascertain if right ventricular function changed in response to the lower tidal volumes.

Overall mortality for this study population was 26.67% and was comparable to the ARDSnet data, even among those who received UPMV at LV_T ≤ 4 mL/kg (4 deaths in 12 subjects; 33.33%). Driving pressures reflect the tidal increase in the static trans-pulmonary pressure, and have been associated with improved mortality when kept minimal (<15 cm of H₂O or changes <7 cm of H₂O).²⁵ In our cohort, driving pressures were higher in those who died than those who survived and appears to be concordant with the current level of evidence in literature, notably those who survived had lower driving pressures. We are unable to derive any further conclusions from this observation, as we suspect that driving pressure was not used by our providers for clinical management during the study period because the concept of driving pressure was still evolving and not prevalent in common practice at that time.

The feasibility and safety of ECCO₂-R was assessed by Faneilli et al.²⁶ in a multi-center trial involving 15 patients with moderate ARDS and demonstrated that ECCO₂-R is safe and efficient, but noted that 27% had worsening hypoxemia and required prone positioning. There are ongoing multi-center trials to determine if ECCO₂-R will improve outcomes and provide further insight into its safety and efficacy in the treatment of patients with ARDS: pRotective vEntilation with veno-venouS lung assisT in respiratory failure (REST, ClinicalTrials.gov Identifier: NCT02654327)²⁷ and Strategy of UltraProtective Lung Ventilation With Extracorporeal CO₂ Removal for New-Onset Moderate to severe ARDS (SUPERNOVA, ClinicalTrials.gov Identifier: NCT02282657). Although the results of these trials may shed light into the efficiency and safety of ECCO₂-R to enable UPMV, our study, likely the first and largest retrospective series to date to display results on the successful use of UPMV without the use of ECCO₂-R in a series of ARDS patients, demonstrates that an UPMV strategy may be adopted safely in ARDS patients without the use of ECCO₂-R.

The obvious limitations to our study include the low power, retrospective design, absent controls prohibiting comparative statistics, and the paucity of documentation of specific reasons behind decisions that led to further reducing the V_T to UPMV levels in every case. Prospective randomized controlled trials comparing standard care versus UPMV with and without ECCO₂-R are needed to confirm our results. Given the heterogeneity of disease process in ARDS, stemming from variations in clinical and biochemical variables, future trials should consider measurement of plasma biomarkers of lung injury and inflammation, and serial radiological assessment between the subgroups.²⁸ This will enable identification of various ARDS phenotypes and their comparative response to the different treatment arms of such trials.

Conclusions

Our results demonstrate that UPMV may be well tolerated provided the respiratory mechanics and hemodynamic status allow for permissive hypercapnia, and ECCO₂-R may not be necessary in all cases. This approach may serve as a potential alternative strategy for patients with ARDS, especially in centers where ECCO₂-R is unavailable or when transfer to a qualified facility may be delayed. Such a personalized approach has also been proposed by others.^22,28,29

Footnotes

Authors’ note

One of the cases in this series was presented as a poster in American Thoracic Society Conference 2014, Colorado, USA.¹⁰

Acknowledgements

The authors thank Diane E Johnson for assisting with the use of i2b2 and RedCap for generating the patient list from EMR using the inclusion and exclusion criteria.

Authors’ contribution

Mohammed Alnijoumi, Jonathan Collins, and Troy Whitacre conceived the idea, designed this study, and obtained IRB approval. Nathanial Moulton and Daniel Woolery collected most of the data, performed preliminary descriptive analysis, and first draft of results and discussion. Hariharan Regunath then reviewed the protocol and preliminary data, guided Nathanial Moulton and Daniel Woolery to collect and compile further additional data and descriptive statistics. Hariharan Regunath then revised and compiled the results, created tables with data, and restructured the entire paper. Troy Whitacre, Jonathan Collins, and Alnijoumi reviewed this final paper and added further edits. Hariharan Regunath and Mohammed Alnijoumi will be the guarantors of this paper taking responsibility for the integrity of the work as a whole, from inception to publication.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Slutsky

Ranieri

. Ventilator-induced lung injury. N Engl J Med 2013; 369: 2126–2136.

Acute Respiratory Distress Syndrome Network,

Brower

Matthay

et al.

Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000; 342: 1301–1308.

Needham

Colantuoni

Mendez-Tellez

et al.

Lung protective mechanical ventilation and two year survival in patients with acute lung injury: Prospective cohort study. BMJ 2012; 344: e2124.

Terragni

Rosboch

Tealdi

et al.

Tidal hyperinflation during low tidal volume ventilation in acute respiratory distress syndrome. Am J Respir Crit Care Med 2007; 175: 160–166.

Grasso

Stripoli

De Michele

et al.

ARDSnet ventilatory protocol and alveolar hyperinflation. Am J Respir Crit Care Med 2007; 176: 761–767.

Terragni

Del Sorbo

Mascia

et al.

Tidal volume lower than 6 ml/kg enhances lung protection: Role of extracorporeal carbon dioxide removal. Anesthesiology 2009; 111: 826–835.

Bein

Weber-Carstens

Goldmann

et al.

Lower tidal volume strategy (approximately 3 ml/kg) combined with extracorporeal CO2 removal versus ‘conventional' protective ventilation (6 ml/kg) in severe ARDS: The prospective randomized Xtravent-study. Intensive Care Med 2013; 39: 847–856.

Retamal

Libuy

Jimenez

et al.

Preliminary study of ventilation with 4 ml/kg tidal volume in acute respiratory distress syndrome: Feasibility and effects on cyclic recruitment - derecruitment and hyperinflation. Crit Care 2013; 17: R16.

Fitzgerald

Millar

Blackwood

et al.

Extracorporeal carbon dioxide removal for patients with acute respiratory failure secondary to the acute respiratory distress syndrome: A systematic review. Crit Care 2014; 18: 222–222.

10.

Alnijoumi M, Whitacre T and Collins J. Ultraprotective mechanical ventilation without extracorporeal CO₂ removal: Case report. B44 INVASIVE AND NON-INVASIVE MECHANICAL VENTILATION. In: American Thoracic Society 2015 international conference, 2015, pp. A3146–A3146.

11.

Harris

Taylor

Thielke

et al.

Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009; 42: 377–381.

12.

Villar

Kacmarek

Perez-Mendez

et al.

A high positive end-expiratory pressure, low tidal volume ventilatory strategy improves outcome in persistent acute respiratory distress syndrome: A randomized, controlled trial. Crit Care Med 2006; 34: 1311–1318.

13.

Amato

Barbas

Medeiros

et al.

Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome. N Engl J Med 1998; 338: 347–354.

14.

Petrucci

De Feo

. Lung protective ventilation strategy for the acute respiratory distress syndrome. Cochrane Database Syst Rev 2013; 2: Cd003844.

15.

Kregenow

Rubenfeld

Hudson

et al.

Hypercapnic acidosis and mortality in acute lung injury. Crit Care Med 2006; 34: 1–7.

16.

Broccard

Hotchkiss

Vannay

et al.

Protective effects of hypercapnic acidosis on ventilator-induced lung injury. Am J Respir Crit Care Med 2001; 164: 802–806.

17.

Sinclair

Kregenow

Lamm

et al.

Hypercapnic acidosis is protective in an in vivo model of ventilator-induced lung injury. Am J Respir Crit Care Med 2002; 166: 403–408.

18.

Emery

Eveland

Min

J-H

et al.

CO2 relaxation of the rat lung parenchymal strip. Respir Physiol Neurobiol 2013; 186: 33–39.

19.

Laffey

O’Croinin

McLoughlin

et al.

Permissive hypercapnia — role in protective lung ventilatory strategies. In: Pinsky

Brochard

Mancebo

et al. (eds). Applied physiology in intensive care medicine 2: Physiological reviews and editorials, Berlin, Heidelberg: Springer Berlin Heidelberg, 2012, pp. 111–120.

20.

Hodgson

Tuxen

Davies

et al.

A randomised controlled trial of an open lung strategy with staircase recruitment, titrated PEEP and targeted low airway pressures in patients with acute respiratory distress syndrome. Crit Care 2011; 15: R133.

21.

Broccard

. Respiratory acidosis and acute respiratory distress syndrome: Time to trade in a bull market? Crit Care Med 2006; 34: 229–231.

22.

Villar

Slutsky

. GOLDEN anniversary of the acute respiratory distress syndrome: Still much work to do!. Curr Opin Crit Care 2017; 23: 4–9.

23.

Fan

Del Sorbo

Goligher

et al.

An official American Thoracic Society/European Society of Intensive Care Medicine/Society of Critical Care Medicine clinical practice guideline: Mechanical ventilation in adult patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2017; 195: 1253–1263.

24.

Mekontso Dessap

Charron

Devaquet

et al.

Impact of acute hypercapnia and augmented positive end-expiratory pressure on right ventricle function in severe acute respiratory distress syndrome. Intensive Care Med 2009; 35: 1850–1858.

25.

Amato

MBP

Meade

Slutsky

et al.

Driving pressure and survival in the acute respiratory distress syndrome. N Engl J Med 2015; 372: 747–755.

26.

Fanelli

Ranieri

Mancebo

et al.

Feasibility and safety of low-flow extracorporeal carbon dioxide removal to facilitate ultra-protective ventilation in patients with moderate acute respiratory distress syndrome. Crit Care 2016; 20: 36.

27.

McNamee

Gillies

Barrett

et al.

pRotective vEntilation with veno-venouS lung assisT in respiratory failure: A protocol for a multicentre randomised controlled trial of extracorporeal carbon dioxide removal in patients with acute hypoxaemic respiratory failure. J Intensive Care Soc 2017; 18: 159–169.

28.

Jabaudon M, Blondonnet R, Audard J, et al. Recent directions in personalised acute respiratory distress syndrome medicine. Anaesth Crit Care Pain Med 018; 37: 251–258.

29.

Berngard

Beitler

Malhotra

. Personalizing mechanical ventilation for acute respiratory distress syndrome. J Thoracic Dis 2016; 8: E172.