Introduction

The application of positive end-expiratory pressure (PEEP) targeted towards improving oxygenation has become one of the cornerstones in the management of ARDS in children and adults since the first description by Ashbaugh et al in the 1960s . Pediatric ARDS (PARDS) is a manifestation of severe, life-threatening lung injury and occurs in up to 10 % of all children admitted to the pediatric intensive care unit (PICU) with mortality rates ranging up to 40 – 50 % . Ventilatory management of PARDS, guided by the principles of lung and diaphragm protective ventilation (LPV), entails limiting inspiratory volume, plateau (Pplat) and driving pressure (DP), individualized PEEP titration, and finding the balance between ventilatory under- and overassistance to prevent respiratory muscle atrophy . Remarkably, strong scientific evidence guiding the bedside team how to manage PEEP in (P)ARDS pediatric and adult ARDS is limited .

Rationale for using PEEP

Ventilator-induced lung injury (VILI) is caused by excessive lung stress and strain, with lung stress and strain being related to one another based on the specific lung elastance . During inspiration, high end-inspiratory transpulmonary pressure (lung stress) resulting from excessive pressures or volumes causes alveolar overdistention, whereas low end-expiratory airway pressures cause repetitive opening and closure of unstable alveoli, leading to surfactant dysfunction and parenchymal shear injury . Lung stress and strain are not homogenously distributed across the lung in ARDS. Amplified lung stress at the margins of aerated and atelectatic lung regions occurs, thereby further contributing to VILI . In addition, there is less lung volume available for tidal ventilation when there are atelectatic lung regions (i.e., the baby lung concept) . This will increase lung strain in aerated lung areas to which the tidal volume is preferentially distributed.

Experimental work showed that the use of PEEP in rats receiving high pressure ventilation attenuated lung injury . Years later, Burkhard proposed the “open lung concept”, i.e., the use of recruitment maneuvers (RM) followed by high PEEP to maintain alveolar patency and thereby reduce atelectrauma . Through the application of PEEP repetitive alveolar opening and closure is prevented, leading to a more homogeneous ventilation distribution and thus a reduction in (regional) overdistension as well as a reduction in lung stress at the margins of aerated and atelectatic lung regions. Furthermore, maintaining alveolar aeration will reduce intrapulmonary and improve oxygenation .

Side-effects of PEEP

In addition to the advantageous effects of PEEP, potential adverse effects on lung parenchyma and the cardiovascular system may also occur. PEEP itself may contribute to VILI by causing increased pressures and volume at end-inspiration, leading to (regional) alveolar overdistension and increased lung stress particularly if plateau or driving pressure limits are exceeded. Recruitment maneuvers with high inspiratory pressures to overcome the alveolar opening pressures are performed to recruit atelectatic lung regions. Recruited alveoli will only remain stable if the applied PEEP exceeds the alveolar closing pressure. If insufficient PEEP is set however, the alveolus will collapse again, redistributing the tidal volume towards aerated lung regions causing overdistension. On the other hand, when sufficient PEEP is applied there still may be regional overdistension, as the volume of aerated alveoli may increase more than newly aerated alveoli . Furthermore, dynamic and static end-inspiratory lung stress may increase when the increase in PEEP does not adequately recruit lung tissue . Higher PEEP may also increase alveolar dead space.

While a minimum level of PEEP is necessary because pulmonary vascular resistance (PVR) is increased in a collapsed lung, higher levels of PEEP increase the intrathoracic pressure that may result in decreased venous return, increased right atrial pressure and compression of the intra-alveolar vessels thereby also increasing PVR and causing right ventricle (RV) dysfunction because of the increased RV afterload. Assessment of the hemodynamic effects of PEEP can be done by measuring blood pressure or by point-of-care ultrasound (POCUS) although the latter requires some training.

Titrating PEEP

Traditionally, PEEP is set at an arbitrary level based on local practices and medical culture . This is most likely influenced by that fact that no single method of PEEP titration that has been shown to be associated with better patient outcomes. Initially, the primary goal of setting PEEP was to improve oxygenation by reducing intrapulmonary shunt. However, improved oxygenation does not always translate to improved oxygen delivery. In fact, the advantages of a better oxygenation can be attenuated by a reduced cardiac output because of higher intrathoracic pressures with higher PEEP although this reduction appears to be clinically irrelevant in children . Lower cardiac output leads to a reduced perfusion of poorly or non-ventilated alveoli, thereby decreasing intrapulmonary shunt making it difficult to interpret improvements in oxygenation .

A much used approach for titrating PEEP is based on the ARDSNetwork PEEP/FiO ₂ tables . These tables were developed through expert consensus with the aim to the balance the negative hemodynamic effects of higher PEEP and oxygen toxicity with higher FiO ₂ . Two versions of the tables exist, one with a lower PEEP/FiO ₂ combination and one a higher PEEP/FiO ₂ combination. Both tables have been tested in several trials in adult ARDS, but none of those trials showed benefit of one table over the other in reducing mortality . Not surprisingly, through a meta-analysis of published studies, it was concluded that in an unselected group of patients higher PEEP was not associated with reduced mortality because of heterogeneity of treatment effect (i.e., some patients having benefit while others having harm from higher PEEP because of patient heterogeneity with respect to the amount of recruitable lung), underscoring the need for individualized PEEP titration .

Pediatric critical care practitioners tend to use low levels of PEEP and inherently accept higher FiO ₂ . However, these practices may negatively affect patient outcome. Observational studies reported an association of higher mortality and longer duration of mechanical ventilation when applied PEEP was lower than the Acute Respiratory Distress Syndrome Net (ARDSNet) Low PEEP/High FiO2 table as compared to on-protocol or PEEP higher than protocol . Wong et al observed especially in moderate-severe PARDS a significant reduction after implementation of a lung protective ventilation protocol which resulted in higher PEEP and lower Vt .

Personalized PEEP titration

Over the years, the philosophy has progressed more towards the goal of lung protection for setting PEEP. Instead of aiming for “best PEEP” – which might be seen as an impossible-to-achieve goal since it is a very complicated topic, it might be better to focus on “better PEEP” balancing the potential benefits and harms. The potential for benefit with higher PEEP is directly related to the potential for alveolar recruitment, hence the net benefit will ultimately be determined by the balance between the advantageous effects, the size of the associated tidal volume and the hemodynamic effects . This means that the one size fits all approach to simply set higher PEEP in all patients cannot be indiscriminately applied. For example, the Alveolar Recruitment Trial (ART) that was performed in adults with moderate-to-severe ARDS reported higher mortality among subjects who were randomized to a strategy composed of a stepwise recruitment maneuver with inspiratory pressures up to 60 cmH ₂ O and a decremental PEEP titration targeting PEEP > 2 cmH ₂ O above the point of maximal respiratory system compliance compared with the control group who was managed with a PEEP/FiO ₂ grid . In one small pediatric retrospective study it was found that only half of the patients had an improvement in the Vd/Vt ratio that was used as indicator to identify a favorable to a PEEP increase . Secondary analysis of adult data showed that mortality was only reduced in patients showed improvement in oxygenation following the application of higher PEEP levels, indicating that these patients had potential for lung recruitablity . Similarly, from an individual patient data meta-analysis it was concluded that higher PEEP was only associated with reduced mortality among patients with baseline P/F < 200 mmHg . In a small pediatric study, Smallwood and colleagues found a in retrospective review of 76 patients with acute hypoxemic respiratory failure (AHRF) who had an oxygen saturation index (OSI) > 5 that only about half of the patients had an improvement in oxygenation following routine increases in PEEP .

Thus, it is important to establish if a patient is a so-called PEEP-responder, i.e., is the application of higher levels of PEEP beneficial in this individual patient. Such patients may be identified through a variety of measurements including respiratory system mechanics such as static or dynamic respiratory system compliance, stress index, identification of the lower inflection point of the pressure – volume curve, esophageal-pressure, metrics for oxygenation and dead space, and by imaging techniques such as computed tomography (CT), electrical impedance tomography) or lung ultrasound, but so far there are no clear pediatric data to suggest the best approach to individualize PEEP .

End-expiratory alveolar pressure is a marker of the end-expiratory pressure that the chest wall exerts. In a single-center study of adult patients with acute lung injury or ARDS it was found that a ventilator strategy using esophageal pressures to estimate end-expiratory transpulmonary pressure and titrate PEEP significantly improved oxygenation and respiratory system compliance . Thus, esophageal pressure guided PEEP titration may enhance individualized PEEP setting, but has not been systematically investigated in PARDS, although it is likely that much higher levels of PEEP would be needed than are used in routine pediatric practice to achieve a trans-pulmonary pressure at end-exhalation near 0, the therapeutic target when using esophageal pressure guided PEEP . Furthermore, whether esophageal pressure guided PEEP titration actually improves patient outcomes remains unclear as the outcomes of a multi-center trial performed in adult patients with moderate to severe ARDS observed no differences in mortality or days free from mechanical ventilation .

Amato and colleagues showed through multilevel mediation analyses of data from 3562 patients with ARDS enrolled in nine randomized controlled trials (RCT) that higher driving pressure (i.e., scaling Vt to respiratory system compliance) was associated with increased mortality, and that reducing driving pressure by limiting Pplat and increasing positive end-expiratory pressure (PEEP) was associated with increased survival . This suggests that improvement in respiratory system compliance or a reduction in driving pressure with same Vt may be used as measure of lung recruitment to evaluate changes in PEEP. Reanalysis of the Assessment of Low Tidal Volume and Elevated End-Expiratory Volume to Obviate Lung Injury (ALVEOLI) and Expiratory Pressure (ExPress) trials showed that changes in driving pressure were more strongly associated with mortality than changes in PaO ₂ /FIO ₂ after changes in PEEP . Interestingly, and somewhat in contrast with the concept of individualization, Chiumello et al found that the high PEEP/FiO ₂ table was the only strategy that consistently provided higher PEEP to patients with severe ARDS and greater recruitablity by CT and lower PEEP to patients with mild ARDS and less recruitablity compared to PEEP setting based on limiting plateau pressure and stress index, or end-expiratory transpulmonary pressure .

Conclusions

Lower PEEP than recommended is being used in the PICU, especially in children with severe acute respiratory failure. This practice is associated with harm and thus underscores the need for higher PEEP. Following its most recent update, the Pediatric Acute Lung Injury Consensus Conference (PALICC) suggests that PEEP should be titrated to oxygenation/oxygen delivery, hemodynamics, and compliance measured under static conditions as compared to other clinical parameters or any of these parameters in isolation in patients with PARDS . They further suggested that PEEP levels should typically be maintained at or above the lower PEEP/higher FiO ₂ table from the ARDS Network protocol as compared to PEEP levels lower than the lower PEEP/ higher FiO2 table, using this table to maintain sufficient oxygenation as a starting point, with subsequent individually titrating the level of PEEP as even the lower PEEP/FiO2 table may still lead to alveolar hyperinflation . During PEEP titration, exceeding plateau pressure and/or driving pressure limits should be avoided. Further research is needed on how to improve PEEP selection in mechanically ventilated children .

Directions for future research

• •
  

  Identify patients who will most likely benefit from higher positive end-expiratory pressure.

  
  
• •
  

  Identify the most optimal approach to setting positive end-expiratory pressure.

  
  
• •
  

  The role of positive end-expiratory pressure in cardiopulmonary interactions.

Funding support statement

MK received lecture fees from Getinge and Vyaire and consultant fees from Metran. MK’s research is financially supported by NIH/NHBLI, ZonMW, Stichting Beatrix Kinderziekenhuis, Fonds NutsOhra, UMC Groningen, TerMeulen Fonds/Royal Dutch Academy of Sciences and VU university medical center. MK received unrestricted technical research support from Vyaire and Applied Biosignals.