Review question: how do different positive end-expiratory pressure (PEEP) levels, or different approaches for selecting PEEP levels, compare on their effects in preterm infants with respiratory distress syndrome (RDS) or bronchopulmonary dysplasia (BPD)?

Background: infants born preterm (before 37 weeks' gestational age) often have respiratory failure and require medical support to achieve adequate pulmonary gas exchange (removing carbon dioxide from the blood and replenishing it with oxygen). RDS and BPD are complications of premature babies that describe early (RDS) and late (BPD) respiratory failure as a result of preterm birth. Conventional mechanical ventilation (the use of a breathing machine) following endotracheal intubation (placement of a breathing tube in the windpipe) is a commonly used therapy for both complications. While this therapy can improve pulmonary gas exchange and lead to benefits, it may simultaneously injure the lungs and cause harms. Gas exchange is achieved by several ventilator settings that help cycle heated, humidified and (sometimes) oxygen-enriched air in and out of the lungs. One of these settings is called PEEP, which is a continuous distending pressure applied throughout the respiratory cycle. It plays an important role in keeping the lungs open, enabling all areas of the lungs to undergo gas exchange. While the use of PEEP is generally accepted, it is not clear which PEEP levels maximize benefits while minimizing possible harms. An appropriate PEEP level may also be best achieved by an individualized approach in which a good PEEP level is determined on a patient-to-patient basis.

Study characteristics: we organized our study to compare the effect of low versus high PEEP levels, which we defined as: less than 5 cmH₂O (low) versus 5 cmH₂O or greater (high). Our primary outcomes were death by hospital discharge (for RDS) or two years of age (for BPD) and neurodevelopmental impairment (problems with growth and development of the brain or central nervous system) at two years of age. We identified four trials. Two trials with 28 infants compared different PEEP levels in infants with RDS. Two trials with 44 infants compared approaches for individually selecting PEEP levels. The compared approaches were: routine, typical practice (control) versus a "lung-recruitment maneuver." With this maneuver, investigators slowly increased and then decreased PEEP levels while monitoring for improvements in oxygenation to guide PEEP level selection. The evidence is up to date to 14 February 2018.

Key results: two trials comparing different PEEP levels in RDS did not report on any patient important outcomes. They compared the short-term effect of different PEEP levels on pulmonary gas exchange, and found no differences between low and high PEEP levels. The two trials comparing the lung-recruitment maneuver to control reported survival to hospital discharge or the development of BPD and found no clear difference between groups. However, they did suggest that an infant receiving the maneuver may have required less time on conventional mechanical ventilation and had better oxygenation for a period after the maneuver. The quality of the evidence for these results was low because they were small studies and vulnerable to bias from limitations in the study methods.

Conclusions: evidence-based guidance for PEEP level selection in preterm infants continues to be limited. We are unable to make recommendations for clinical practice based on our findings. This review should raise awareness of the lack of high-quality research evidence guiding the selection of PEEP levels in preterm infants and encourage further research. Additional research evaluating lung-recruitment maneuvers as an approach for individually selecting PEEP levels would be particularly valuable, given our findings of possible but uncertain benefit with this approach.