Efficacy of Higher Positive End-Expiratory Pressure Ventilation Strategy in Patients With Acute Respiratory Distress Syndrome: A Systematic Review and Meta-Analysis

Previous systematic reviews and meta-analyses assessing the pooled effects of higher positive end-expiratory pressure (PEEP) failed to show significantly reduced mortality in patients with acute respiratory distress syndrome (ARDS). Some new randomized controlled trials (RCTs) have been reported and an updated systematic review is needed to evaluate the use of higher PEEP in patients with ARDS. We searched MEDLINE, Cochrane Central Register of Controlled Trials (CENTRAL), EMBASE, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Igaku-Chuo-Zasshi, ICTRP, the National Institute of Health Clinical Trials Register, and the reference list of recent guidelines. We included RCTs to compare the higher PEEP ventilation strategy with the lower strategy in patients with ARDS. Two authors independently assessed the eligibility of the studies and extracted the data. The primary outcomes were 28-day mortality. The GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) methodology was used to evaluate the certainty of the evidence. Among the 6530 screened records, 16 randomized trials involving 4150 patients were included in our meta-analysis. When comparing higher PEEP versus lower PEEP, the pooled risk ratio (RR) for 28-day mortality was 0.85 (15 studies, n=4108, 95% CI 0.72 to 1.00, I2=58%, low certainty of evidence). Subgroup analysis by study participants with a low tidal volume (LTV) strategy showed an interaction (P for interaction, 0.001). Our study showed that in patients with ARDS, the use of higher PEEP did not significantly reduce 28-day mortality compared to the use of lower PEEP.


Introduction And Background
Acute respiratory distress syndrome (ARDS) is a common clinical syndrome with substantial morbidity and mortality in the intensive care unit (ICU) [1]. Most patients with ARDS require mechanical ventilation and the mortality was 28-47.8% [2]. In the management of patients with ARDS, the application of positive endexpiratory pressure (PEEP) is one method to improve oxygenation in patients. PEEP prevents atelectasis, increases functional residual capacity, recruits alveoli, redistributes extravascular lung water, and improves ventilation-perfusion matching. The higher PEEP usually improves the PaO 2 /FiO 2 ratio compared to a lower PEEP [3][4][5][6][7]. However, there is a potential risk of increasing dead space, carbon dioxide, and pressure injury.
The results of randomized controlled trials (RCTs) that evaluated the efficacy of higher PEEP were inconsistent [3][4][5][8][9]. The Cochrane Systematic Review of seven trials that involved 2,565 patients with ARDS showed that in-hospital mortality was not significantly reduced by higher PEEP [10]. Recent systematic reviews have demonstrated a trend toward a lower risk of mortality, but the difference was insignificant [6,11]. However, these studies did not include three RCTs published after 2018 [12][13][14].
Higher PEEP ventilation has potentially relevant benefits; however, the effects remain unclear. To update the Japanese ARDS guidelines 2016 [15], we conducted a systematic review and meta-analysis aimed to Measures of treatment effects were calculated using the Cochrane Review Manager software (version 5.3;Cochrane Collaboration,London,UK) for data synthesis and analysis. Dichotomous data (mortality, barotrauma) were expressed as RR with 95% CIs, and continuous data (QOL, VFD, LOS) were presented as mean difference (MD) with 95% CI.
The random-effects model was used for meta-analysis. The chi 2 test and the I 2 statistic were used to measure heterogeneity. We considered that a P-value of less than 0.1 and greater than 60% in the chi 2 and I 2 statistic tests, respectively, were significantly heterogeneous. To evaluate publication bias, funnel plots and Egger's tests were performed, where each comparison included more than ten studies. Egger's test was considered statistically significant at P< 0.05.

Subgroup Analysis and Sensitivity Analysis
We pre-planned the following subgroup analyses for primary outcomes (28-day mortality): the Berlin definition or not, LTV strategy (LTV in the comparison group or not), and P/F ratio ≤200 mmHg at the time of inclusion. We performed sensitivity analyses that included only studies with a low risk of bias to evaluate the robustness of our inferences for the primary outcomes.

Assessment of the Certainty of the Evidence
To evaluate the quality of the evidence, the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) considerations were used classifying the quality as "high," "moderate," "low," or "very low." We used the methods and recommendations described in the Cochrane Handbook for Systematic Reviews of Intervention [20].
PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses; CENTRAL: Cochrane Central Register of Controlled Trials; CINAHL: Cumulative Index to Nursing and Allied Health Literature; ICTRP: International Clinical Trials Registry Platform; PEEP: positive end-expiratory pressure; ECMO: extracorporeal membrane oxygenation

Study Characteristics
The characteristics of this study are summarized in Table 1. ARDS was defined by the lung injury severity score (LISS) [8], AECC criteria [3,5,9,[21][22][23][24][25][26], Berlin definition [12][13][14][27][28][29], and the author's definition [4]. Most studies used the P/F ratio as an inclusion criterion. The intervention procedure varied across the studies. PEEP adjustments in the intervention groups were performed using the PEEP table in three of 16 studies [3][4]22], PEEP above the lower inflection point of the pressure-volume curve of the respiratory system (Pflex) in three of 16 studies [8][9]21], and compliance, including the decremental PEEP method, in four of 16 studies [13,[24][25][26]. The mean level of PEEP on the first day in the high PEEP group ranged from 10.0 to 18.7 cm H 2 O, and in the control group, it was 6.8-12.0 cm H 2 O. The levels of PEEP on Days 1,3,and 7 are described in Table 2.

Risk of Bias Assessment
The risk of bias for mortality was "low" when comparing higher PEEP and lower PEEP ( Figure 2). Blinding was not performed because of the nature of the intervention in all studies. Because unblinding does not influence objective outcomes, we evaluated mortality and P/F ratio as a low risk of bias. While subjective outcomes, such as QOL, VFD up to 28 days, hospital LOS, and barotrauma were assessed as "unclear" risk of bias related to blinding. For incomplete outcome data, one study was evaluated as "high" risk of bias due to the exclusion of seven patients after randomization [21]. For selective outcome reporting, studies for which protocols were not available or details could not be obtained at the trial registration site were rated as "unclear" risk of bias. In Brower, the protocol was changed during the trial, and the primary endpoint was switched from "28-day mortality" to "proportion of patients who died before discharge without respiratory

Certainty of Evidence
The certainty of evidence for mortality was downgraded by two levels for inconsistency (different direction of effect in studies and high heterogeneity) and imprecision (wide confidence interval) and was considered low. The details of the other outcomes are presented in Table 4.

Summary of Key Findings
The current systematic review and meta-analysis showed that a higher PEEP ventilation strategy did not significantly reduce the 28-day and longest follow-up mortality of adult patients with ARDS. While the P/F ratio was increased in the higher PEEP group, there was no significant effect on VFD at 28 days and hospital LOS. The higher PEEP ventilation strategy did not appear to significantly increase the risk of barotrauma.

Relation With Previous Systematic Reviews
Previous systematic reviews of the effect of higher PEEP on patients with ARDS have not reported a significant mortality benefit with high PEEP [11,26]. Our systematic review included several recent RCTs [12][13][14][25][26][27], but the results remained unchanged. A sensitivity analysis that excluded studies with a high risk of bias showed similar results. However, these results should be carefully interpreted. This was because of the high heterogeneity in the included studies. In some RCTs, the control group had non-LTV ventilation, and the intervention group had open-lung strategy ventilation that combined higher PEEP with LTV. This heterogeneity may be important for interpreting the results. In our review, we conducted a subgroup analysis to examine whether there was an interaction between LTV and higher PEEP. This subgroup analysis suggested that there was an interaction when combined with LTV. Walkey et al. reported similar results to our analysis, which involved a recent RCT [6]. These results suggest that not only high PEEP but also a combination of LTV may be more effective. However, since the main effect was not significant in this study, the results of the subgroup analysis must be interpreted with caution.
We performed a subgroup analysis of the definition of ARDS and differences in severity at inclusion (P/F≤200) and examined the interaction on the effect of higher PEEP, but found no interaction on mortality.
A meta-analysis of individual patient data from three trials [3][4][5] by Briel et al. showed that patients with ARDS (P/F≤200 mmHg) may benefit from higher PEEP than patients with mild ARDS [7]. There are several possible reasons for this difference in results; one reason is that we performed a subgroup analysis by P/F ratio at the inclusion criteria. To maintain the randomness of covariates between subgroups, it is necessary to perform subgroup analysis with or without the inclusion criteria of severity or stratified randomization by severity. Briel et al. adjusted for confounders to equalize the groups, but this may have been insufficient [7]. The second reason is the difference in the number of included studies. In our study, we included 12 RCTs in the subgroup analysis. Under the assumption that there was no bias, the pooled effect size would have been closer to the true value if the number of included studies increased. However, the sample size was insufficient for detecting small effects in the subgroups in our study. Until further evidence is accumulated, it is not clear whether higher PEEP should be applied depending on the severity of the target patients.

Implications for Practice and Further Study
Higher PEEP did not show a significant reduction in mortality, but it did show a consistent effect on improving oxygenation [10][11]26]. A higher PEEP usually improves the PaO 2 /FiO 2 compared to a lower PEEP.
PEEP prevents atelectasis, increases functional residual capacity, and improves ventilation-perfusion matching. Our study also showed that oxygenation could be improved without increasing pressure injury. In the intervention groups of most of the included studies, the combination of higher PEEP with LTV did not increase the occurrence of clinically objectified barotrauma. Therefore, a high PEEP may be safer for patients with ARDS when combined with LTV.
It is not clear from this study which method of setting PEEP was the best; there were many different methods of setting PEEP in each study, including oxygenation, stress index, transpulmonary pressure, decremental PEEP adjustment after a recruitment maneuver, or pressure-volume curves. These subgroup analyses could not be performed because of the small number of studies.

Limitations
The most important limitation was heterogeneity. There was variation in PEEP levels in the intervention group (PEEP at Day 1; 10-16.3 cmH 2 O). In the control group, there was a similar variation in PEEP levels (PEEP at Day 1; 6.5-12.0 cmH 2 O). The setting of the target PEEP level and the PEEP level of the control population may affect the effect of higher PEEP. Second, the P/F ratios of patients at inclusion also varied. Therefore, we performed a subgroup analysis using P/F ratios and found no interaction. Third, the definition of ARDS also varied among the studies, and subgroup analysis by ARDS definition did not show any interaction. Fourth, there may have been variabilities in important common interventions such as recruitment maneuvers and prone positions. The key common interventions were not specified by the studies; we could not verify whether there was an interaction between these subgroups. A meta-analysis by Walkey et al. reported that they did not detect any interaction by performing a protocolized recruitment maneuver [6].

Conclusions
Low-certainty evidence showed that the use of higher PEEP compared to the use of lower PEEP did not reduce mortality. However, the use of higher PEEP in patients with ARDS might be justified because it improves oxygenation without increasing harm and is associated with reduced mortality when combined with LTV. Further studies are needed to determine which subpopulations with higher PEEP are effective. #13 TI "lung protective ventilatory strateg*" OR AB "lung protective ventilatory strateg*" OR TI LPVS OR AB LPVS #14 TI "alveolar recruit*" OR AB "alveolar recruit*" OR TI "recruitment maneuve*" OR AB "recruitment maneuve*" #15 S8 OR S9 OR S10 OR S11 OR S12 OR 13 OR S14

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.