Optimising continuous positive airway pressure in preterm infants

Abstract

Background: Respiratory distress syndrome (RDS) and bronchopulmonary dysplasia (BPD) remain common complications of prematurity. Respiratory support may injure the lung and contribute to BPD. Non-invasive methods of respiratory support such as continuous positive airway pressure (CPAP) have the potential to reduce the risk of lung injury but randomised controlled trials comparing early CPAP with mechanical ventilation show no difference in the rates of BPD. In intubated infants, lung recruitment manoeuvres that allow for an ‘open lung’ improve oxygenation and ventilation but there is a paucity of data regarding CPAP strategies and how they impact upon lung behaviour. The series of studies described in this report aim to add to the knowledge regarding lung behaviour and potential complications in very preterm infants receiving CPAP. Method: Five studies of very preterm infants receiving CPAP were designed. These included three studies using respiratory inductive plethysmography and electrical impedance tomography; 1) a physiological study to determine if stepwise increases and decreases in CPAP level would demonstrate hysteresis and thus describe the changes in end-expiratory thoracic volume (∆EEV) and the distribution of ventilation (∆VT) in very preterm infants receiving CPAP from birth; 2) an observational study to describe ∆EEV and ∆VT during extubation to CPAP; and 3) a randomised pilot study comparing ∆EEV and ∆VT in infants extubated to currently used CPAP protocols versus a ‘CPAP recruitment’ manoeuvre. In addition, 4) a laboratory based study to determine whether the stable microbubble test could be used to determine within the first hour of life, which infants would successfully manage CPAP from birth and 5) a case-control study to determine risk factors for the development of pneumothoraces in very preterm infants were conducted. Results: These studies found that 1) it was possible to demonstrate hysteresis in many, but not all, very preterm infants exposed to stepwise increases and decreases in CPAP levels and that ∆EEV and ∆VT were often improved following this manoeuvre; 2) lung behaviour during extubation to CPAP was variable and unpredictable between and within infants; 3) after extubation, there were no differences in ∆EEV following a CPAP recruitment manoeuvre although ∆VT appeared to be more homogeneous when compared with a control group; 4) infants with a stable microbubble count ≥ 8 microbubbles/mm2 in gastric aspirate did not fail CPAP and 5) it may be possible to identify infants at highest risk of pneumothorax on the basis of their inspired fraction of oxygen in the first 12 hours of life. Conclusions: Lung behaviour at different stages of lung disease in very preterm infants receiving CPAP is complex, variable and unpredictable. It is possible to demonstrate hysteresis using CPAP, but optimising CPAP in very preterm infants at the bedside is difficult without tools that provide feedback regarding ∆EEV and ∆VT to clinicians. CPAP protocols that aid an ‘open lung’ require further investigation. Tests of endogenous surfactant status may be used to determine which infants may benefit from CPAP alone and which infants need escalation of respiratory support.