⾮ARDS患者における Lower PEEP VS Higher PEEP

Journal Club

⾮ARDS患者における

Lower PEEP VS Higher PEEP

-RELAx trial-

2021/3/23

練⾺光が丘病院 総合救急診療科 集中治療部⾨

島⽥ 侑祐 / ⽚岡 惇

本⽇の論⽂

RELAx trial

JAMA. 2020;324(24):2509-2520.

⼈⼯呼吸器関連肺傷害を防ぐ

The Scientific World Journal 5

T

able 2: Main characteristics of RCTs for ARDS patients included in this study.

Author Year Patients Low PEEP level High PEEP level

Amato et al. [38]

1998

ARDS

≥5

cm H

₂

O

16

cm H

₂

O or Pflex +

2

Ranieri et al. [39]

1999

ARDS

3–15

cm H

₂

O

15

cm H

₂

O or Pflex +

3

The NHLBI Institute ARDS

Clinical Trial Network [40]

2004

ALI/ARDS

5

cm H

₂

O

5–24

cm H

₂

O according FiO

₂

Villar et al. [50]

2006

ARDS

≥5

cm H

₂

O

15

cm H

₂

O or Pflex +

3

Mercat et al. [41]

2008

ALI/ARDS

5–9

cm H

₂

O PEEP according to Plateau

28–30

cm H

₂

O

Meade et al. [42]

2008

ALI/ARDS

5

cm H

₂

O

5–24

according FiO

₂

Talmor et al. [43]

2008

ALI/ARDS

10

cm H

₂

O

17

cm H

₂

O

Pressure

Volume

Lower inflection point

Upper inflection point

Alveolar collapse

Alveolar overdistension

Alveolar recruitmen

t

Figure 2: Pressure-volume curve with lower and upper inflection points. According to PEEP level, the recruitment of collapsed alveoli could be set between the lower and the upper inflection points.

were lower in high PEEP group [38, 39]. In 2004 the ARDS network performed a clinical trial with the aim to investigate the role of high PEEP levels on clinical outcome in ARDS patients receiving mechanical ventilation [40]. PEEP levels were set at 8 and 14 cm H

₂

O during the days. As results, there were no significant differences in mortality, in ventilator free- days, or organ failure between low and high PEEP groups [40]. ARDS network failed to show the best degree of PEEP to be applied during mechanical ventilation for mild to severe ARDS. General consensus exists about the use of PEEP in ARDS to keep open alveoli and small airway. After the ARDS network, Ranieri et al. compared the effect of high PEEP with low PEEP as protective and standard ventilation [39]. In this study the authors found a reduction in plateau pressure and mortality in patients ventilated with high PEEP in a contest of protective ventilation [38]. The role of PEEP in ARDS was also evaluated in association with a fixed tidal volume [41, 42]. In LOVS trial, there was no significant difference in mortality but the incidence of refractory hypoxemia was significantly lower in high PEEP group [40]. In EXPRESS trial, the authors found no difference in mortality, but there was a significant increase in ventilator and organ failure free- days [42]. In a RCT by Talmor et al., PEEP was set at 13 cm

H

₂

O for three days and then changed to 17 or 10 cm H

₂

O [43]. As results, from the third day oxygenation, respiratory compliance and plateau pressure significantly improved in the high PEEP group [43]. The role of higher PEEP in severe ARDS seems to be established by several RCTs to improve survival or respiratory function even if it was associated with fixed or differ from tidal volume.

In 2010, a meta-analysis evaluating the effect of higher versus lower PEEP in ARDS patients suggested that treat- ments with different PEEP levels were not associated with an improvement in hospital survival, even if high PEEP level was associated with an improvement of survival in the subgroup of ARDS patients [44]. Recently, the ARDS definition task force proposed a new definition for ARDS, the Berlin definition, categorizing this pathology in three mutual exclusive degrees as mild, moderate, and severe [45].

According to this task force, high PEEP level should be reserved in severe ARDS patients [45].

5.2. PEEP in Traumatic Brain Injured Patients. The use of PEEP in traumatic brain injured (TBI) patient is still contro- versial. In mechanical ventilation for respiratory disease, mild PEEP levels and recruitment maneuver avoided progressive alveolar collapse and possible lung consolidation, improved arterial oxygenation, and reduced elastance of the respiratory system [46]. As discussed above, the application of PEEP in TBI patients could affect the cerebral circulation by a raised of mean intrathoracic pressure resulting in a reduction of cerebral venous return and then in an increase of ICP [47]. Videtta et al. investigated the variation of ICP and CPP at different levels of PEEP in mechanically ventilated brain injured patients raising PEEP from 5 to 15 cm H

₂

O with an increase of ICP about 3 mm Hg but no changes in CPP [48]. Young et al. investigated the ICP response to a gradual increment of PEEP in 3 randomized groups of patients with severe brain injured patients with pulmonary dysfunction [45]. Interestingly, the authors reported a decrease in ICP of 6 mm Hg in the group of patients with PEEP from 0 to 5 cm H

₂

O, of 8 mm Hg in the group with PEEP from 6 to 10 cm H

₂

O, and of 12 mm Hg in the group of PEEP from 11 to 15 cm H

₂

O. This study seemed to suggest a useful and safe application of PEEP for mechanical ventilation in brain injury [49]. The effects of PEEP were also investigated by Caricato et al. in comatose patients with severe TBI and normal or low lung compliance [19]. The rise of PEEP reduced CPP and

「過伸展の防⽌」

低1回換気

「虚脱の防⽌」

PEEP

Review Article

PEEP Role in ICU and Operating Room:

From Pathophysiology to Clinical Practice

M. Vargas,

¹

Y. Sutherasan,

²

C. Gregoretti,

³

and P. Pelosi

⁴

1 Department of Neurosciences and Reproductive and Odontostomatological Sciences, University of Naples “Federico II,” 80100 Naples, Italy

2Ramathibodi Hospital, Mahidol University, Bangkok 10400, Thailand

3Department of Critical Care Medicine, “Citt`a della Salute e della Scienza” Hospital, 10121 Turin, Italy

4Department of Surgical Sciences and Integrated Diagnostics, University of Genoa, 16132 Genoa, Italy

Correspondence should be addressed to P. Pelosi; [email protected]

Received

5

October

2013; Accepted 24

December

2013; Published 14

January

2014

Academic Editors: M. Elbarbary, L. M. Gillman, A. E. Papalois, and A. Shiloh

Copyright ©

2014

M. Vargas et al.

This is an open access article distributed under the Creative Commons Attribution License,

which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Positive end expiratory pressure (PEEP) may prevent cyclic opening and collapsing alveoli in acute respiratory distress syndrome (ARDS) patients, but it may play a role also in general anesthesia.

This review is organized in two sections. The first one reports

the pathophysiological effect of PEEP on thoracic pressure and hemodynamic and cerebral perfusion pressure.

The second section

summarizes the knowledge and evidence of the use of PEEP in general anesthesia and intensive care. More specifically, for intensive care this review refers to ARDS and traumatic brain injured patients.

1. Introduction

Positive end expiratory pressure (PEEP) is applied during the end of expiration to maintain the alveolar pressure above atmospheric pressure. PEEP is different from continuous positive airway pressure (CPAP), because this one refers to a positive pressure maintained during inspiration and expiration phase of spontaneous ventilation. The benefit of PEEP has been demonstrated in terms of preventing cyclic opening and collapsing alveoli in acute respiratory distress syndrome patients (ARDS). Moreover, protective ventilation, even in noninjury lungs, should be considered such as during perioperative period aiming to prevent collapsing of alveoli.

However, applying PEEP may affect cardiac function and vital organ perfusion by complex mechanisms (Figure 1).

To minimize the adverse effects of PEEP in intensive care unit (ICU) and in operating room, better knowledge and understanding of the interaction between heart, lung, and brain during applying PEEP are required.

The aims of this review are

(1) to clarify the pathophysiology of PEEP on thoracic pressure and hemodynamic and cerebral perfusion;

(2) to clarify the role of PEEP during general anesthesia;

(3) to clarify the role of PEEP in intensive care for ARDS, with a special focus on traumatic brain injured patients.

2. Methods

In the first section of this paper, we considered general issues related to pathophysiology of PEEP. In the second and third parts we focused on randomized clinical trials evaluating the role of PEEP during general anesthesia for different types of surgery and for ARDS patients. The specific search for traumatic brain injured patients was conducted with the best available evidence according the aim of this paper. The research was conducted mainly in PUBMED from 1996 to 2013.

3. Pathophysiology of PEEP

3.1. PEEP and Thoracic Pressure. The intrathoracic pressure (ITP) should be categorized in airway pressure (Paw), pleural

Hindawi Publishing Corporation e Scientific World Journal

Volume 2014, Article ID 852356, 8 pages http://dx.doi.org/10.1155/2014/852356

虚脱

過伸展

呼気

吸気

ARDS患者における 最適なPEEPとは︖

含気のない 背側肺

⽐較的含気のある

腹側肺

^{ならないように 過伸展に}

しないように虚脱

ARDS患者における

Lower PEEP VS Higher PEEP

ARDSNet Mechanical Ventilation Protocol

ARDS患者におけるLower PEEP VS Higher PEEP 主な3つの⼤規模RCT

ALVEOLI

NEJM 2004 LOVS

JAMA 2008 ExPress

JAMA 2008

対象患者 P/F≦300

273 vs 276 P/F≦250

508 vs 475 P/F≦300

382 vs 385 介⼊ FiO2によって決められる

high PEEP

FiO2によって決められるhigh PEEPプラトー圧≦40

プラトー圧が28〜30になるよ うな最⼤PEEP

⽐較 FiO2によって決められるlow PEEP

FiO2によって決められるlow

PEEP 酸素化の⽬標が達成される5〜

9のPEEP 介⼊初⽇の

PEEP 8.9 vs 14.7 (p<0.01) 10.1 vs 15.6 (p<0.001) 7.1 vs 14.6 (p<0.001) 死亡率院内 24.9% vs 27.5% (p=0.48) 40.4% vs 36.4% (p=0.19) 39.0% vs 35.4% (p=0.31)

⾮⼈⼯呼吸器 使⽤⽇数（28

⽇間）

14.5⽇ vs 13.8⽇ (p=0.5) 10⽇ vs 10⽇ (p=0.92) 3⽇ vs 7⽇ (p=0.04)

ARDS患者における

Lower vs Higher PEEP メタ解析

ference, 1.6%) with higher PEEP, but dif- ferences in fatal consequences from such barotrauma are unlikely (absolute risk difference, 0.6%; RR, 1.20; 95% CI, 0.79- 1.81). Otherwise, we found no evi- dence suggesting serious adverse ef- fects associated with higher PEEP in patients with ARDS.

The strengths of this review include an explicit study protocol and analysis plan; access to trial protocols, case re- port forms, and complete, unedited data sets; standardized outcome definitions across trials (except for rescue thera- pies); and analyses based on the inten- tion-to-treat principle. To minimize the risk of overfitting and data-driven asso- ciations, we prespecified a limited num- ber of prognostic factors and potential effect modifiers for our statistical mod- els.²³We calculated RRs adjusted for im-

portant prognostic factors using log- binomial models¹⁶ and allowed for potential clustering effects by using ran- dom effects for recruiting hospitals.²⁴Our results proved robust in sensitivity analy- ses applying alternate statistical ap- proaches. We followed current recom- mendations for subgroup analyses in meta-analysis of individual-patient data,²⁵ thereby overcoming limitations of meta- analyses using aggregated data.^26-29All in- cluded trials met high methodological quality standards (concealed random- ization, explicit study protocols, and complete follow-up) and systematically collected data on important, potential ad- verse effects of high PEEP administra- tion by routinely documenting deaths, pneumothorax, use of vasopressors (he- modynamic instability) and rescue thera- pies (refractory hypoxemia), and dura-

tion of mechanical ventilation and intensive care. An independent data and safety monitoring committee was estab- lished to monitor and protect the safety of participants in each trial. The 3 ma- jor trials included 90 multidisciplinary intensive care units with international representation; these features enhance the generalizability of our findings.

The subgroup effect for ARDS at baseline meets all criteria for a cred- ible subgroup analysis.³⁰ We found a large and statistically significant (P=.02 for interaction) difference in RRs that was consistent across individual trials and efficacy outcomes. The hypoth- esis was generated a priori and was one of a small number tested. Exploring the effect of higher vs lower PEEP across quintiles suggests a threshold effect, rather than a progressive increase in

Table 4.Clinical Outcomes in All Patients and Stratified by Presence of ARDS at Baseline

Outcomes

All Patients With ARDS Without ARDS

No. (%)

Adjusted RR (95% CI)^a P

Value

No. (%)

Adjusted RR (95% CI)^a P

Value

No. (%)

Adjusted RR (95% CI)^a P

Value Higher

(n = 1136)PEEP

Lower (n = 1163)PEEP

Higher (n = 951)PEEP

Lower (n = 941)PEEP

Higher (n = 184)PEEP

Lower (n = 220)PEEP

Death in hospital 374 (32.9) 409 (35.2) 0.94

(0.86 to 1.04) .25 324 (34.1) 368 (39.1) 0.90

(0.81 to 1.00) .049 50 (27.2) 44 (19.4) 1.37

(0.98 to 1.92) .07 Death in ICU^b 324 (28.5) 381 (32.8) 0.87

(0.78 to 0.97) .01 288 (30.3) 344 (36.6) 0.85

(0.76 to 0.95) .001 36 (19.6) 37 (16.8) 1.07

(0.74 to 1.55) .71 Pneumothorax

between day 1 and day 28^c

87 (7.7) 75 (6.5) 1.19

(0.89 to 1.60) .24 80 (8.4) 64 (6.8) 1.25

(0.94 to 1.68) .13 7 (3.8) 11 (5.0) 0.72

(0.37 to 1.39) .33 Death after

pneumothorax^c 43 (3.8) 40 (3.5) 1.11

(0.73 to 1.69) .63 41 (4.3) 35 (3.7) 1.20

(0.79 to 1.81) .39 2 (1.1) 5 (2.3) 0.44

(0.08 to 2.35)^g .34 Days with

unassisted breathing between day 1 and day 28, median (IQR)^d

13 (0 to 22) 11 (0 to 21) 0.64

(−0.12 to 1.39)^e .10 12 (0-21) 7 (0-20) 1.22

(0.39 to 2.05)^e .004 17 (0-23) 19 (5.5-24) −1.74

(−3.60 to 0.11)^e .07

Total use of rescue

therapies^f 138 (12.2) 216 (18.6) 0.64

(0.54 to 0.75) !.001 130 (13.7) 200 (21.3) 0.63

(0.53 to 0.75) !.001 8 (4.4) 16 (7.3) 0.60

(0.25 to 1.43)^g .25 Death after rescue

therapy^f 85 (7.5) 132 (11.3) 0.65

(0.52 to 0.80) !.001 82 (8.6) 124 (13.2) 0.66

(0.52 to 0.82) !.001 3 (1.6) 8 (3.6) 0.37

(0.10 to 1.46)^g .15 Use of

vasopressors 722 (63.6) 759 (65.3) 0.93

(0.75 to 1.14)^g .49 627 (65.9) 647 (68.8) 0.90

(0.72 to 1.13)^g .37 95 (51.6) 111 (50.5) 0.92

(0.56 to 1.50)^g .72 Abbreviations: ARDS, acute respiratory distress syndrome; CI, confidence interval; ICU, intensive care unit; IQR, interquartile range; PEEP, positive end-expiratory pressure; RR,

relative risk.

aMultivariable regression with the outcome of interest as dependent variable; PEEP group, age, probability of dying in hospital derived from prognostic scores at baseline, severe sepsis at baseline, and trial as independent variables; and hospital as a random effect.

bPatients who died before being discharged from the intensive care unit for the first time up to day 60.

cDefined as the need for chest tube drainage.

dMedian number of days of unassisted breathing to day 28 after randomization, assuming a patient survives and remains free of assisted breathing for at least 2 consecutive calendar days after initiation of unassisted breathing.

eCoefficient from a corresponding linear regression model using the same independent variables and random effect as the above-described log-binomial model; for example, a coefficient of 1.22 means that patients in the group treated with higher PEEP have, on average, 1.22 days more of unassisted breathing during the first 28 days compared with patients in the group treated with lower PEEP.

fAs defined in each trial; rescue therapies included in the Assessment of Low Tidal Volume and Elevated End-Expiratory Pressure to Obviate Lung Injury and the Lung Open Ven- tilation to Decrease Mortality in the Acute Respiratory Distress Syndrome studies: inhaled nitric oxide, prone ventilation, high-frequency oscillation, high-frequency jet ventilation, extracorporeal membrane oxygenation, partial liquid ventilation, and surfactant therapy. Rescue therapies included in the Expiratory Pressure Study: prone ventilation, inhaled nitric oxide, and almitrine bismesylate.

gAdjusted odds ratio substitutes for relative risk, because the corresponding log-binomial model did not converge.

POSITIVE END-EXPIRATORY PRESSURE IN ACUTE LUNG INJURY AND ARDS

870 JAMA,March 3, 2010—Vol 303, No. 9(Reprinted) ©2010 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by TOKYO BAY MEDICAL CTR, jun kataoka on 04/23/2016

Higher PEEP群で、ICU死亡率、レスキュー治療を⾏った割合が低い P/F≦200に限れば、Higher PEEP群で病院死亡率が低い

CARING FOR THE

CRITICALLY ILL PATIENT

Higher vs Lower Positive End-Expiratory Pressure in Patients With Acute Lung Injury

and Acute Respiratory Distress Syndrome

Systematic Review and Meta-analysis

Matthias Briel, MD, MSc Maureen Meade, MD, MSc Alain Mercat, MD

Roy G. Brower, MD

Daniel Talmor, MD, MPH Stephen D. Walter, PhD Arthur S. Slutsky, MD

Eleanor Pullenayegum, PhD Qi Zhou, PhD

Deborah Cook, MD, MSc Laurent Brochard, MD

Jean-Christophe M. Richard, MD Francois Lamontagne, MD

Neera Bhatnagar, MLIS Thomas E. Stewart, MD Gordon Guyatt, MD, MSc

P ROTECTING LUNGS FROM VENTI - lation-induced injury is an im- portant principle in the man- agement of patients with acute lung injury or acute respiratory distress syndrome (ARDS). Although the criti- cal care community has generally en- dorsed lower tidal volumes and inspira- tory pressures, the optimal level of positive end-expiratory pressure (PEEP) remains unestablished. ^1,2 Experimental data suggest that PEEP levels exceed- ing traditional values of 5 to 12 cm H ₂ O can minimize cyclical alveolar collapse and corresponding shearing injury to the lungs in patients with considerable edema and alveolar collapse. ^3-5 For pa- tients with relatively mild acute lung in-

jury, however, potential adverse conse- quences of higher PEEP levels, including circulatory depression ⁶ or lung overdis- tension, ⁷ may outweigh the benefits. Sev- eral multicenter, randomized trials test- ing the incremental effect of higher levels

See also p 883 and Patient Page.

Author Affiliations are listed at the end of this article.

Corresponding Author: Maureen Meade, MD, MSc, Department of Clinical Epidemiology and Biostatis- tics, Room 2C12, 1200 Main St W, Hamilton, ON L8N 3Z5, Canada ([email protected]).

Caring for the Critically Ill Patient Section Editor: Derek C. Angus, MD, MPH, Contributing Editor, JAMA ([email protected]).

Context Trials comparing higher vs lower levels of positive end-expiratory pressure (PEEP) in adults with acute lung injury or acute respiratory distress syndrome (ARDS) have been underpowered to detect small but potentially important effects on mortal- ity or to explore subgroup differences.

Objectives To evaluate the association of higher vs lower PEEP with patient- important outcomes in adults with acute lung injury or ARDS who are receiving ven- tilation with low tidal volumes and to investigate whether these associations differ across prespecified subgroups.

Data Sources Search of MEDLINE, EMBASE, and Cochrane Central Register of Con- trolled Trials (1996-January 2010) plus a hand search of conference proceedings (2004- January 2010).

Study Selection Two reviewers independently screened articles to identify studies randomly assigning adults with acute lung injury or ARDS to treatment with higher vs lower PEEP (with low tidal volume ventilation) and also reporting mortality.

Data Extraction Data from 2299 individual patients in 3 trials were analyzed using uniform outcome definitions. Prespecified effect modifiers were tested using multi- variable hierarchical regression, adjusting for important prognostic factors and clus- tering effects.

Results There were 374 hospital deaths in 1136 patients (32.9%) assigned to treatment with higher PEEP and 409 hospital deaths in 1163 patients (35.2%) assigned to lower PEEP (adjusted relative risk [RR], 0.94; 95% confidence interval [CI], 0.86-1.04; P =.25). Treatment effects varied with the presence or absence of ARDS, defined by a value of 200 mm Hg or less for the ratio of partial pressure of oxygen to fraction of inspired oxygen concentration (P =.02 for interaction). In patients with ARDS (n = 1892), there were 324 hospital deaths (34.1%) in the higher PEEP group and 368 (39.1%) in the lower PEEP group (adjusted RR, 0.90;

95% CI, 0.81-1.00; P =.049); in patients without ARDS (n=404), there were 50 hospital deaths (27.2%) in the higher PEEP group and 44 (19.4%) in the lower PEEP group (adjusted RR, 1.37; 95% CI, 0.98-1.92; P =.07). Rates of pneumotho- rax and vasopressor use were similar.

Conclusions Treatment with higher vs lower levels of PEEP was not associated with improved hospital survival. However, higher levels were associated with improved sur- vival among the subgroup of patients with ARDS.

JAMA. 2010;303(9):865-873 www.jama.com

©2010 American Medical Association. All rights reserved. (Reprinted) JAMA, March 3, 2010—Vol 303, No. 9 865

Downloaded From: http://jama.jamanetwork.com/ by TOKYO BAY MEDICAL CTR, jun kataoka on 04/23/2016

P/F≦200 P/F 201-300

ARDS患者における

Lower vs Higher PEEP メタ解析

ference, 1.6%) with higher PEEP, but dif- ferences in fatal consequences from such barotrauma are unlikely (absolute risk difference, 0.6%; RR, 1.20; 95% CI, 0.79- 1.81). Otherwise, we found no evi- dence suggesting serious adverse ef- fects associated with higher PEEP in patients with ARDS.

The strengths of this review include an explicit study protocol and analysis plan; access to trial protocols, case re- port forms, and complete, unedited data sets; standardized outcome definitions across trials (except for rescue thera- pies); and analyses based on the inten- tion-to-treat principle. To minimize the risk of overfitting and data-driven asso- ciations, we prespecified a limited num- ber of prognostic factors and potential effect modifiers for our statistical mod- els.²³We calculated RRs adjusted for im-

portant prognostic factors using log- binomial models¹⁶ and allowed for potential clustering effects by using ran- dom effects for recruiting hospitals.²⁴Our results proved robust in sensitivity analy- ses applying alternate statistical ap- proaches. We followed current recom- mendations for subgroup analyses in meta-analysis of individual-patient data,²⁵ thereby overcoming limitations of meta- analyses using aggregated data.^26-29All in- cluded trials met high methodological quality standards (concealed random- ization, explicit study protocols, and complete follow-up) and systematically collected data on important, potential ad- verse effects of high PEEP administra- tion by routinely documenting deaths, pneumothorax, use of vasopressors (he- modynamic instability) and rescue thera- pies (refractory hypoxemia), and dura-

tion of mechanical ventilation and intensive care. An independent data and safety monitoring committee was estab- lished to monitor and protect the safety of participants in each trial. The 3 ma- jor trials included 90 multidisciplinary intensive care units with international representation; these features enhance the generalizability of our findings.

The subgroup effect for ARDS at baseline meets all criteria for a cred- ible subgroup analysis.³⁰ We found a large and statistically significant (P=.02 for interaction) difference in RRs that was consistent across individual trials and efficacy outcomes. The hypoth- esis was generated a priori and was one of a small number tested. Exploring the effect of higher vs lower PEEP across quintiles suggests a threshold effect, rather than a progressive increase in

Table 4.Clinical Outcomes in All Patients and Stratified by Presence of ARDS at Baseline

Outcomes

All Patients With ARDS Without ARDS

No. (%)

Adjusted RR (95% CI)^a P

Value

No. (%)

Adjusted RR (95% CI)^a P

Value

No. (%)

Adjusted RR (95% CI)^a P

Value Higher

(n = 1136)PEEP

Lower (n = 1163)PEEP

Higher (n = 951)PEEP

Lower (n = 941)PEEP

Higher (n = 184)PEEP

Lower (n = 220)PEEP

Death in hospital 374 (32.9) 409 (35.2) 0.94

(0.86 to 1.04) .25 324 (34.1) 368 (39.1) 0.90

(0.81 to 1.00) .049 50 (27.2) 44 (19.4) 1.37

(0.98 to 1.92) .07 Death in ICU^b 324 (28.5) 381 (32.8) 0.87

(0.78 to 0.97) .01 288 (30.3) 344 (36.6) 0.85

(0.76 to 0.95) .001 36 (19.6) 37 (16.8) 1.07

(0.74 to 1.55) .71 Pneumothorax

between day 1 and day 28^c

87 (7.7) 75 (6.5) 1.19

(0.89 to 1.60) .24 80 (8.4) 64 (6.8) 1.25

(0.94 to 1.68) .13 7 (3.8) 11 (5.0) 0.72

(0.37 to 1.39) .33 Death after

pneumothorax^c 43 (3.8) 40 (3.5) 1.11

(0.73 to 1.69) .63 41 (4.3) 35 (3.7) 1.20

(0.79 to 1.81) .39 2 (1.1) 5 (2.3) 0.44

(0.08 to 2.35)^g .34 Days with

unassisted breathing between day 1 and day 28, median (IQR)^d

13 (0 to 22) 11 (0 to 21) 0.64

(−0.12 to 1.39)^e .10 12 (0-21) 7 (0-20) 1.22

(0.39 to 2.05)^e .004 17 (0-23) 19 (5.5-24) −1.74

(−3.60 to 0.11)^e .07

Total use of rescue

therapies^f 138 (12.2) 216 (18.6) 0.64

(0.54 to 0.75) !.001 130 (13.7) 200 (21.3) 0.63

(0.53 to 0.75) !.001 8 (4.4) 16 (7.3) 0.60

(0.25 to 1.43)^g .25 Death after rescue

therapy^f 85 (7.5) 132 (11.3) 0.65

(0.52 to 0.80) !.001 82 (8.6) 124 (13.2) 0.66

(0.52 to 0.82) !.001 3 (1.6) 8 (3.6) 0.37

(0.10 to 1.46)^g .15 Use of

vasopressors 722 (63.6) 759 (65.3) 0.93

(0.75 to 1.14)^g .49 627 (65.9) 647 (68.8) 0.90

(0.72 to 1.13)^g .37 95 (51.6) 111 (50.5) 0.92

(0.56 to 1.50)^g .72 Abbreviations: ARDS, acute respiratory distress syndrome; CI, confidence interval; ICU, intensive care unit; IQR, interquartile range; PEEP, positive end-expiratory pressure; RR,

relative risk.

aMultivariable regression with the outcome of interest as dependent variable; PEEP group, age, probability of dying in hospital derived from prognostic scores at baseline, severe sepsis at baseline, and trial as independent variables; and hospital as a random effect.

bPatients who died before being discharged from the intensive care unit for the first time up to day 60.

cDefined as the need for chest tube drainage.

dMedian number of days of unassisted breathing to day 28 after randomization, assuming a patient survives and remains free of assisted breathing for at least 2 consecutive calendar days after initiation of unassisted breathing.

eCoefficient from a corresponding linear regression model using the same independent variables and random effect as the above-described log-binomial model; for example, a coefficient of 1.22 means that patients in the group treated with higher PEEP have, on average, 1.22 days more of unassisted breathing during the first 28 days compared with patients in the group treated with lower PEEP.

fAs defined in each trial; rescue therapies included in the Assessment of Low Tidal Volume and Elevated End-Expiratory Pressure to Obviate Lung Injury and the Lung Open Ven- tilation to Decrease Mortality in the Acute Respiratory Distress Syndrome studies: inhaled nitric oxide, prone ventilation, high-frequency oscillation, high-frequency jet ventilation, extracorporeal membrane oxygenation, partial liquid ventilation, and surfactant therapy. Rescue therapies included in the Expiratory Pressure Study: prone ventilation, inhaled nitric oxide, and almitrine bismesylate.

gAdjusted odds ratio substitutes for relative risk, because the corresponding log-binomial model did not converge.

POSITIVE END-EXPIRATORY PRESSURE IN ACUTE LUNG INJURY AND ARDS

870 JAMA,March 3, 2010—Vol 303, No. 9(Reprinted) ©2010 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by TOKYO BAY MEDICAL CTR, jun kataoka on 04/23/2016

P/F > 200に限れば、Higher PEEP群で

⼈⼯呼吸器離脱期間が⻑い傾向である

CARING FOR THE

CRITICALLY ILL PATIENT

Higher vs Lower Positive End-Expiratory Pressure in Patients With Acute Lung Injury

and Acute Respiratory Distress Syndrome

Systematic Review and Meta-analysis

Matthias Briel, MD, MSc Maureen Meade, MD, MSc Alain Mercat, MD

Roy G. Brower, MD

Daniel Talmor, MD, MPH Stephen D. Walter, PhD Arthur S. Slutsky, MD

Eleanor Pullenayegum, PhD Qi Zhou, PhD

Deborah Cook, MD, MSc Laurent Brochard, MD

Jean-Christophe M. Richard, MD Francois Lamontagne, MD

Neera Bhatnagar, MLIS Thomas E. Stewart, MD Gordon Guyatt, MD, MSc

P ROTECTING LUNGS FROM VENTI - lation-induced injury is an im- portant principle in the man- agement of patients with acute lung injury or acute respiratory distress syndrome (ARDS). Although the criti- cal care community has generally en- dorsed lower tidal volumes and inspira- tory pressures, the optimal level of positive end-expiratory pressure (PEEP) remains unestablished. ^1,2 Experimental data suggest that PEEP levels exceed- ing traditional values of 5 to 12 cm H ₂ O can minimize cyclical alveolar collapse and corresponding shearing injury to the lungs in patients with considerable edema and alveolar collapse. ^3-5 For pa- tients with relatively mild acute lung in-

jury, however, potential adverse conse- quences of higher PEEP levels, including circulatory depression ⁶ or lung overdis- tension, ⁷ may outweigh the benefits. Sev- eral multicenter, randomized trials test- ing the incremental effect of higher levels

See also p 883 and Patient Page.

Author Affiliations are listed at the end of this article.

Corresponding Author: Maureen Meade, MD, MSc, Department of Clinical Epidemiology and Biostatis- tics, Room 2C12, 1200 Main St W, Hamilton, ON L8N 3Z5, Canada ([email protected]).

Caring for the Critically Ill Patient Section Editor: Derek C. Angus, MD, MPH, Contributing Editor, JAMA ([email protected]).

Context Trials comparing higher vs lower levels of positive end-expiratory pressure (PEEP) in adults with acute lung injury or acute respiratory distress syndrome (ARDS) have been underpowered to detect small but potentially important effects on mortal- ity or to explore subgroup differences.

Objectives To evaluate the association of higher vs lower PEEP with patient- important outcomes in adults with acute lung injury or ARDS who are receiving ven- tilation with low tidal volumes and to investigate whether these associations differ across prespecified subgroups.

Data Sources Search of MEDLINE, EMBASE, and Cochrane Central Register of Con- trolled Trials (1996-January 2010) plus a hand search of conference proceedings (2004- January 2010).

Study Selection Two reviewers independently screened articles to identify studies randomly assigning adults with acute lung injury or ARDS to treatment with higher vs lower PEEP (with low tidal volume ventilation) and also reporting mortality.

Data Extraction Data from 2299 individual patients in 3 trials were analyzed using uniform outcome definitions. Prespecified effect modifiers were tested using multi- variable hierarchical regression, adjusting for important prognostic factors and clus- tering effects.

Results There were 374 hospital deaths in 1136 patients (32.9%) assigned to treatment with higher PEEP and 409 hospital deaths in 1163 patients (35.2%) assigned to lower PEEP (adjusted relative risk [RR], 0.94; 95% confidence interval [CI], 0.86-1.04; P =.25). Treatment effects varied with the presence or absence of ARDS, defined by a value of 200 mm Hg or less for the ratio of partial pressure of oxygen to fraction of inspired oxygen concentration (P =.02 for interaction). In patients with ARDS (n = 1892), there were 324 hospital deaths (34.1%) in the higher PEEP group and 368 (39.1%) in the lower PEEP group (adjusted RR, 0.90;

95% CI, 0.81-1.00; P =.049); in patients without ARDS (n=404), there were 50 hospital deaths (27.2%) in the higher PEEP group and 44 (19.4%) in the lower PEEP group (adjusted RR, 1.37; 95% CI, 0.98-1.92; P =.07). Rates of pneumotho- rax and vasopressor use were similar.

Conclusions Treatment with higher vs lower levels of PEEP was not associated with improved hospital survival. However, higher levels were associated with improved sur- vival among the subgroup of patients with ARDS.

JAMA. 2010;303(9):865-873 www.jama.com

©2010 American Medical Association. All rights reserved. (Reprinted) JAMA, March 3, 2010—Vol 303, No. 9 865

Downloaded From: http://jama.jamanetwork.com/ by TOKYO BAY MEDICAL CTR, jun kataoka on 04/23/2016

P/F≦200 P/F 201-300

⾮ARDS患者ではどうか︖

•

PEEPを上げることで、平均気道内圧を上昇させ、酸素化を 改善することができる可能性

•

⽐較的⾼めのPEEPをかけることで肺虚脱を予防させ、無気 肺を予防することができる可能性

→結果として、⼈⼯呼吸器期間を短縮させたり、抜管後の呼 吸不全を予防することができる︖

⾮ARDS患者における

⽣理学的に考えうるPEEPのメリット

⾮ARDS患者における

PEEPの“有無”を検討したRCT

Crit Care Med. 2008;36 (8):2225-2231. 

スペイン2施設での⾮盲検化RCT

P

胸部Xp正常でP/F⽐＞250の挿管⼈⼯呼吸器管理患者

I

PEEP 5-8 cmH2O N=66

C

PEEP 0 cmH2O N=65

O

院内死亡率

⾮ARDS患者における

PEEPの“有無”を検討したRCT

TBIが3割で、

意識障害がメイン 肺炎少ない

Primary outcomeの院内死亡率は変わりなし。

VAPや低酸素発症率（P/F<175）を有意に低下させた。無気肺は有意差なし。

⾮ARDS患者における

PEEPの“有無”を検討したRCT

Crit Care Med. 2008;36 (8):2225-2231. 

⾮ARDS患者における

Lower PEEP VS Higher PEEP

systematic review & meta-analysis

今回の論⽂と同グループの論⽂

PaO2/FiO2（5研究）

低酸素⾎症の発症（2研究）

⾎圧（2研究）

低⾎圧（2研究）

High PEEP群で酸素化は良好、

低酸素⾎症の発症は有意に低い 酸素化については異質性⾼い

⾎⾏動態の変化については

⼗分に評価できず

PEEP設定は研究によりバラバラである

院内死亡率（7研究）

28⽇死亡率（2研究）

⼈⼯呼吸期間（3研究）

ARDSの発症（4研究）

肺炎の発症（3研究）

無気肺の発症（3研究）

圧外傷の発症（7研究）

ARDSの発症は

High PEEP群で有意に低い

死亡含めその他のアウトカムに差なし

いずれも異質性⾼い

国際的な観察研究にて、非ARDS患者における設定されるPEEP中央値 は、

1998年 5cmH2Oから2016年 7cmH2Oとなっており、

高めに設定される傾向になっている

根拠は乏しい中で

近年PEEPは⾼めに設定される傾向にある

Intensive Care Med (2020) 46:444–453

•

動物実験では⾼いPEEPが既存の肺障害が進⾏させたり、新規肺 障害を発⽣させる可能性が指摘されている

Eur Respir J. 2004;23 (1):122-128 Crit Care. 2006;10(4):R122- R128 Anesthesiology. 2009;110(6):1341-1347

•

⾼いPEEPが⾎⾏動態に影響を与え、⼿術中の輸液や昇圧剤投与 が増加したとの報告がある

Lancet. 2014;384(9942):495-503.

JAMA. 2019;322(18):1829.

•

無駄に⾼いPEEPを設定することで、ウィーニングが遅れ、⼈⼯呼吸器 期間が⻑くなる可能性がある

BMJ. 2011;342:c7237- c7237.

⾮ARDS患者におけるPEEPのデメリット

⾮ARDS患者においては

あえて⾼めのPEEPをかける

必要はないのでは︖

本⽇の論⽂

RELAx trial

JAMA. 2020;324(24):2509-2520.

PReVENT trialと同じグループによる研究

JAMA. 2018 Nov 13;320(18):1872-1880.

論⽂の PICO

P

^{気管挿管されている⾮}

^ARDS

^患者

I ^{Lower PEEP}

^（

^PEEP

^︓

⁰

^〜

^{5 cmH 2 O}

^）

C Higher PEEP

（

PEEP

︓

8 cmH ₂ O

）

O Primary Outcome: 28

⽇までの⼈⼯呼吸器⾮装着期間

Methods

Study design

•

オランダ8病院のICU

•

研究期間︓2017年10⽉〜2019年12⽉

•

アカデミックな施設、⾮アカデミックな施設共に含む

•

多施設Open-label RCT

•

⾮劣性試験

•

中間解析なし

•

独⽴した委員会により、事前指定された3時点で試験及び有害 事象のチェックが⾏われた

Inclusion criteria

•

ICU⼊室

•

侵襲的⼈⼯呼吸器管理が必要、もしくは開始されている

•

ランダム化から24時間以上の⼈⼯呼吸管理が必要と予測される

Exclusion criteria①

•

18歳未満

•

臨床的にARDSと診断、またはARDSの可能性がありP/F⽐＜200mmHg

•

⾎⾏再建に失敗した⼼筋梗塞による進⾏性の⼼筋虚⾎

•

増悪傾向でコントロール不良の頭蓋内圧上昇(>18mmHg)

•

くも膜下出⾎後の遅発性脳虚⾎

•

壊死性筋膜炎

•

治療不良な重症貧⾎（ エホバの証⼈など）

•

以前にこのRCTに参加したことがある

•

同じclinical endpointを持つ他のRCTに参加している、またはprimary endpointを損なう可能性のある介⼊がなされている

Exclusion criteria②

•

今回のICU⼊室直前に12時間以上の挿管⼈⼯呼吸器管理

•

ランダム化1時間以上前からの挿管⼈⼯呼吸器管理

•

妊娠が疑われる/確定している

•

BMI＞40の肥満

•

GOLD Ⅲ-ⅣのCOPD

•

元々拘束性肺疾患(胸部画像で慢性的な間質浸潤影)あり

•

パルスオキシメーターが不正確に出る（CO中毒など）

•

⼈⼯呼吸器期間を延⻑させそうな神経疾患（ギラン・バレー症候群や⾼位 脊髄病変, ALS, MS, MG）

•

ECMO使⽤中

•

インフォームドコンセントなし

Randomization/Blinding

•

ICU⼊室1時間以降にランダム化

•

1:1でランダム割り付け

•

各施設の研究者が、中央でパスワードで保護されて暗号化され たウェブベースの専⽤⾃動ランダム化システムを使⽤

•

最⼤8⼈の患者のブロックでランダム化した

•

割り付けの盲検化は治療の性質上不可能

Intervention

Higher PEEP Lower PEEP

PEEP 8cmH2O, FiO2 0.21-0.6で開始 PEEPは基本的に8cmH2Oで固定

SpO2 92〜96%となる様、FiO2を調整

PEEP 5cmH20, FiO2 0.21-0.6で開始

FiO2 0.21-0.6の間で, SpO2＞92％または

PaO2>60mmHgを保てる最⼩のPEEP値(最⼩ 0 cmH2O)で継続

PEEPを下げる際は、15分毎に1ずつ下げる

▪SpO2 88〜91%が5分間続く場合 FiO2を0.6まで適宜上げる

それでもSpO2＜92%の場合、PEEPを1ずつ上げる

▪⾎⾏動態不安定時（昇圧剤増量が必要な時）

1-2時間PEEPを5cmH2Oに設定する。

⾎⾏動態が安定したら、元の設定に戻す。（higher PEEPは8に、lower PEEPは再度段階的にPEEPを下 げる）

▪重症低酸素時（SpO2≦88%, PaO2≦55mmHg）

適宜pulmonary toiletなど施⾏ 診療医によってFiO2を1.0まで増量可、PEEP増量可

Intervention

⽬標酸素化

通常 ︓SpO2 92〜96%, PaO2 60〜85mmHg 以下の場合︓SpO2 94〜96%, PaO2 68〜85mmHg

※組み⼊れ後に以下の病態に発展した患者

・⾎⾏再建に失敗した⼼筋梗塞による進⾏性の⼼筋虚⾎

・増悪傾向でコントロール不良の頭蓋内圧上昇(＞18mmHg)

・くも膜下出⾎後の遅発性脳虚⾎

・壊死性筋膜炎

・治療不良な重症貧⾎（エホバの証⼈など）

•

呼吸器モード︓⾃動で

FiO2

と

PEEP

が調整されなければ、モード 設定に制限はなし

•

⼀回換気量︓理想体重×

6

〜

8 ml

で設定

•

換気回数︓動脈⾎ガスで

pH 7.35

〜

7.45

になる様設定

※

代謝性アシドーシス

/

アルカローシス例では試験者の判断で 低

/

⾼

CO2

を許容して管理する

その他の管理 ー⼈⼯呼吸器ー

＜Weaningプロトコール＞

FiO2≦0.4、もしくは前⽇より改善している場合に、PSVに変更可能か毎⽇評価 そして指導ナースや医師が、患者が覚醒していてPSVに変更可能と判断したら変更

▪SBT開始基準

・反応があり協⼒的 ・適切な咳嗽反射あり ・FiO2≦0.4でP/F⽐≧200

・呼吸数 8-30回/分で呼吸窮迫兆候なし ・⾎⾏動態安定 ・コントロール不良の不整脈なし ・直腸温＞36℃

▪SBTクリア基準

・呼吸数＜35回/分 ・SpO2>90% ・⼼拍数や⾎圧増加＜20％

・不安徴候や発汗がない

30分以上SBTクリア後に、治療者の判断で抜管を⾏う。

※Lower PEEP群では、そのままの最低PEEPでSBT

※Higher PEEP群ではPEEP 8cmH2OのままSBT、または5cmH2Oで1〜2時間SBT

その他の管理 ー⼈⼯呼吸器ー

＜気管切開基準＞

挿管10⽇⽬以降に以下の場合、考慮する。

・14⽇以上の⼈⼯呼吸器管理 ・著明なICU-AW

・GCS＜7かつ/または嚥下/咳嗽反射不良、痰喀出不良

・肺予備能低下 ・抜管後の再発性呼吸不全 ・挿管失敗

・⻑期のWeaning不成功

その他の管理 ー⼈⼯呼吸器ー

＜鎮静、鎮痛＞

analgo-sedationを行い、看護師により1日3回評価された

・鎮静：RASS -2〜0を目標に行う

・鎮痛：NRS、VAS、CCPOT、BPSで評価

その他の管理 ー鎮静・鎮痛ー

＜SOD /SDD＞

⼈⼯呼吸器管理が48時間以上必要 and/or ICUに72時間以上滞 在することが予想される全ての患者に、院内感染予防として

SODまたはSDDを⾏われた。

＜DVT予防＞

抗凝固⾮使⽤患者ではその地域のガイドラインに従ってDVT予 防が⾏われた。

その他の管理

SOD：selective oropharyngeal decontamination SDD：selective decontamination of digestive tract

＜補液＞normovolemiaをターゲットに、尿量≧0.5ml/kg/h⽬標に適宜 補液を施⾏。晶質液が膠質液より好ましいとする。

＜栄養＞ICU⼊室後、可能な限り早期に低カロリー・⾼タンパク(1.2-

1.7g/kg/⽇)を開始する。経腸栄養を優先し、薬剤投与でも胃管 不耐性であれば⼗⼆指腸チューブを使⽤する。4⽇⽬までタンパ ク摂取が⽬標に届かなければ、追加で静脈栄養を開始する。

その他の管理

Primary outcome

・28⽇⽬までの⼈⼯呼吸器⾮装着期間

（28⽇⽬までに死亡した場合は0⽇とする）

Secondary outcome

•

ICU、院内の滞在期間

•

ICU、院内での28⽇^*1、90⽇死亡率

•

⽣存者での⼈⼯呼吸器期間^*2

•

肺合併症（ARDSへの進⾏、VAP、重度無気肺、重度低酸素、気胸）

•

重度低酸素や無気肺に対する治療必要性（リクルート⼿技や腹臥位、気 管⽀鏡での気道開通）

•

昇圧剤使⽤期間

•

鎮静剤使⽤期間

*1 *2は最初のプロトコールには含まれなかったが、更新版にて追加された

Statistical analysis

•

同意を撤回した患者と追跡不能となった患者を除いたランダム割 り付け群にて⼀次解析を施⾏（ITT解析）

•

PEEP titration strategyを⾏えた患者でper-protocol解析を施⾏

【脱落基準】

lower PEEP群︓最初の2⽇間で1⽇4回の測定のうち2回以上で PEEP＞5cmH2Oかつ、FiO2≦0.6またはSpO2＞92%

higher PEEP群︓⾎⾏動態が不安定な時を除き、最初の2⽇間で1

⽇4回の測定のうち2回以上でPEEP＜ 8cmH2O

Sample size

・先⾏研究¹⁾²⁾から、higher PEEP群で28⽇⽬までの⼈⼯呼吸器⾮装着期間は 平均16⽇（SD 10）と推定

・⾮劣性マージンを-10%と設定^＊

・α＝0.05, power＝80%, 1:1のランダム割り付け

・脱落率を10%と設定

N＝980⼈が必要

＊⾮劣性マージン-10%︓⼈⼯呼吸器期間の差が12時間以下または⼈⼯呼吸器離脱期間 の差が1.6⽇以下となる

1) JAMA. 2018;320(18):1872-1880. , 2) JAMA. 2018;319(10):993-1001.

Statistical analysis

Primary outcome

⽚側95%CI＞0.90（⾮劣性マージン10%）で⾮劣性を評価

・GAMLSSを使⽤し、平均⽐を評価

・95%CIはzero-inflated β分布, デルタ解析により計算

・⾮劣性の場合は、lower PEEP群の優越性についても調査（階層 的閉検定で95%CIを計算）

▪サブグループ解析

治療効果の同質性は、zero-inflated β 分布を参考にしたGAMLSSでサ ブグループと治療の交互作⽤のテストをすることにより調査された。

Secondary outcome

優越性を評価 両側検定

Risk ratioや95%CI(Wald尤度⽐近似検定、x²検定)で評価

・⼈⼯呼吸器期間、ICU/院内滞在期間、28⽇/90⽇死亡率︓

Kaplan-Meier 曲線、95%CI(Cox⽐例ハザードモデル、Schoen -feld残差により評価)

・⾎管作動薬や鎮静薬の使⽤⽇数︓独⽴したt検定での平均差で⽐較 p値は多重検定としてHolm-Bonferroni法により調整された

Statistical analysis

primary outcomeとsecondary outcomeは感度分析がなさ れ、以下の調整を受けて再評価された。

・混合効果モデル︓年齢、性別、APACHE Ⅳスコア

・変量効果モデル︓施設

⽣存者の⼈⼯呼吸期期間、ICU⼊室期間、在院⽇数

・競合リスク︓抜管前死亡、ICU退室、退院

・Fine-Grayモデルで再評価

・95%CIを⽤いた部分分布ハザード⽐で報告

解析ソフト︓R software ver

.3.6.3(R Core T

eam)

Statistical analysis

Results

2017年10⽉ – 2019年12⽉

980⼈がランダム化割り付け 同意撤回、追跡不能のため

⼀次解析は

lower PEEP 476⼈

higher PEEP 493⼈

プロトコール通りに試験終了して per-protocol解析されたのは

lower PEEP 411⼈

higher PEEP 411⼈

患者背景①

APACHE Ⅳはhigher PEEP

群でやや⾼い（90.0 vs. 83.5）

他の背景因⼦に差なし

平均年齢は66歳

平均BMIは約26

ARDSリスクのある患者は40%

Septic shockは約10%

患者背景②

内科的ICU⼊室が80%

挿管の理由

・呼吸不全︓30%

・⼼肺停⽌︓25%

・意識障害︓15%

・予定術後︓15%

・気道保護︓10%

ランダム化までの

⼈⼯呼吸器期間は0.6時間

（開始後すぐに組み込まれ

ている）

TV平均︓両群 約7.0 ml/kg

プラトー圧︓両群 約20 cmH₂O

Driving pressure︓lower PEEP群で⾼い

（14 cmH₂O vs 13 cmH₂O）

FiO2︓両群 0.50 P/F⽐︓210程度

lower PEEP︓平均PEEP 5.0 cmH₂O higher PEEP︓平均PEEP 7.0 cmH₂O

患者背景③

day1〜day5のPEEP値推移

higher PEEP群︓PEEP値はランダム化以降、平均 8cmH2O

lower PEEP群︓PEEP値は平均でday1 1.8cmH20, day2 1.5cmH2O, day3 2.5cmH2O 4⽇⽬以降平均PEEP 5cmH2O

※両群ともday3までで半数が抜管または死亡している

平均値 最低値 最⾼値

実際のPEEP設定

ランダム化後 Day 1

Day 2 Day 3

<5 5 <5

<5 <5

8 8

8 8

>8 >8

>8 >8

>8 >8

>8 >8

Higher PEEP

Lower PEEP

Lower PEEP Higher PEEP

Higher PEEP群で、FiO2は若⼲低くなっているように⾒える

day1〜day5のFiO2値推移

平均値 最低値 最⾼値

Primary outcome

lower PEEP群はhigher PEEP群と⽐較し、

28⽇⽬までの⼈⼯呼吸器⾮装着期間は⾮劣性

（17.7⽇

vs 16.7⽇ AR 0.41; 95%CI -1.16 –

1.98）

Per-protocol解析でもlower PEEP群はhigher PEEP群に⾮劣性 (14.3⽇ vs 13.5⽇)

優越性試験では有意差を認めなかった（p=0.22）

Primary outcome 〜サブグループ解析〜

Primary outcome 〜サブグループ解析〜

いずれのサブグループでも

28 ⽇⽬の⼈⼯呼吸器離脱期間は有意差なし

術後、⼼肺停⽌後、呼吸不全、

ただし BMI

＞

30 、

LIPS ≧ 4 、 APACH Ⅳ score ≧ 86

では⾮劣性は⽰されなかった

Secondary outcome

Secondary outcome

VAPやARDS、無気肺変わりなし

lower PEEP群でレスキュー治療の必要性が有意に増える

（RR 1.35; 95%CI 1.02-1.79, p=0.54 ）

特にリクルートメント⼿技が多く⾏われた

Secondary outcome

Secondary outcome

昇圧剤・鎮静剤使⽤、ICU/院内滞在期間、死亡率に 有意差なし

Kaplan-Meier Curve

⼈⼯呼吸器離脱 ICUからの退室

Kaplan-Meier Curve

死亡 ⽣存退院

Day5でのトータルバランスは 有意差なし（p=0.728）

Day1までのバランスはhigher PEEPで多い傾向

平均 3010ml vs 2597ml (p=0.05)

IN -OUTバランス

Discussion

•

lower PEEP群はhigher PEEP群と⽐較し、28⽇⽬までの⼈

⼯呼吸器⾮装着期間は⾮劣性であり、ICU/病院滞在期間、

死亡率、肺合併症、昇圧剤や鎮静剤使⽤期間に有意差はな かった

•

本研究は、⾮ARDS患者においてlower PEEP群がhigher PEEP群に対しての⾮劣性かを調査した最⼤のRCT

・バイアスが⼩さい

・追跡不能数が少ない

・アカデミックな施設、⾮アカデミックな施設が共に参加

・ランダム化が⼈⼯呼吸開始1時間以内に⾏われている

本研究の強み

•

研究の性質上盲検化ができない

•

挿管とランダム化の間隔が1時間と⾮常に短い

→⼀部の患者が脱落した

Limitation

•

ARDSを除いた、背景が多様な患者群が含まれていた

→サブグループ解析では何らかの関連は明らかでなかった

•

higher PEEP群の⽅がAPACHE Ⅳスコアが⾼かった

→APACHEⅣスコアを調整した解析でもアウトカムの変化は なかった

•

PEEP 8cmH₂0とより低いPEEPを⽐較したが、PEEP8cmH₂0 は通常のケアでない可能性がある

→ただし近年の臨床研究にて⾮ARDS患者でも⾼めのPEEPを 使⽤する傾向がある

Limitation