Article Text

Download PDFPDF

Positive effects of early continuous positive airway pressure on pulmonary function in extremely premature infants: results of a subgroup analysis of the COIN trial
  1. C C Roehr1,
  2. H Proquitté1,
  3. H Hammer1,
  4. R R Wauer1,
  5. C J Morley2,
  6. G Schmalisch1
  1. 1Clinic for Neonatology, CCM, Charité Universitätsmedizin Berlin, Berlin, Germany
  2. 2Neonatal Research, Royal Women's Hospital, Melbourne, Australia
  1. Correspondence to Dr C C Roehr, Clinic for Neonatology, Charité Universitätsmedizin Berlin, Chariteplatz 1, 10117 Berlin, Germany; christoph.roehr{at}


Objective Early continuous positive airway pressure (CPAP) may reduce lung injury in preterm infants.

Patients and methods Spontaneously breathing preterm infants were randomised immediately after birth to nasal CPAP or intubation, surfactant treatment and mechanical ventilation. Pulmonary function tests approximately 8 weeks post-term determined tidal breathing parameters, respiratory mechanics and functional residual capacity (FRC).

Results Seventeen infants received CPAP and 22 mechanical ventilation. Infants with early CPAP had less mechanical ventilation (4 vs 7.5 days; p=0.004) and less total respiratory support (30 vs 47 days; p=0.017). Post-term the CPAP group had lower respiratory rate (41 vs 48/min; p=0.007), lower minute ventilation (223 vs 265 ml/min/kg; p=0.009), better respiratory compliance (0.99 vs 0.82 ml/cm H2O/kg; p=0.008) and improved elastic work of breathing (p=0.004). No differences in FRC were found.

Conclusions Early CPAP is feasible, shortens the duration of respiratory support and results in improved lung mechanics and decreased work of breathing.

Statistics from


Mechanical ventilation is an independent risk factor for bronchopulmonary dysplasia; excessive baro- and volutrauma through mechanical ventilation can lead to over-distension of pulmonary structures and cause local as well as systemic inflammation and an acute phase inflammatory response.1

Comparative studies have suggested that avoidance of mechanical ventilation through the use of early continuous positive airway pressure (CPAP) with or without surfactant administration may reduce the risk of chronic lung disease. The recent CPAP or intubation at birth (COIN) trial showed that early use of CPAP in preterm infants was associated with less need for mechanical ventilation, fewer doses of surfactant, shorter duration of ventilation and significantly less oxygen requirement at day 28 of life.2

In light of the fact that local and systemic inflammatory responses are decreased when early mechanical ventilation is avoided, we assumed that the use of early CPAP would help reduce pulmonary injury in very low birth weight (VLBW) neonates.3 We hypothesised that early CPAP would help pulmonary structures adapt to breathing air and cause less local and systemic inflammation, leading to a reduced need for and lesser intensity of mechanical ventilation. This should result in improved pulmonary function post-term compared to infants initially managed with mechanical ventilation. The aim of this study was to compare pulmonary function at about 2 months of age post-term between two groups of infants treated according to the COIN protocol.

Patients and methods

Infants between 25+0 and 28+6 weeks gestation were studied between October 2003 and October 2006. Our patients (17 treated with early CPAP, and 22 treated with early mechanical ventilation) represented a subgroup of single centre recruits from the COIN study (which had a total population of 610 infants). The inclusion criteria, randomisation procedure and the treatment protocol were according to the COIN trial protocol, details of which are published elsewhere.2 All studied infants were inborn at a single level three perinatal centre and all intubated infants received surfactant according to our local protocol. Parental consent for randomisation was obtained before delivery and again before pulmonary function testing. The study was approved by the Ethics Committee of Charité Hospital (1943/Si 272). The analysis was stratified in hours 1 to 120 of life, over the course of hospitalisation and post-term lung function.

Protocol for pulmonary function analysis

Pulmonary function testing was performed in our pulmonary function laboratory at about 2 months post-term. Tidal breathing parameters (tidal volume (VT), respiratory rate (RR), minute ventilation (V'E)) were measured using a dead-space free flow through technique. Lung mechanics (respiratory compliance (Crs) and respiratory resistance (Rrs)) were measured by occlusion tests, and functional residual capacity (FRC) was determined by using the constant volume baby body plethysmograph (Jaeger, Würzburg, Germany). Crs was measured at the end of inspiration; 15 occlusion tests were usually performed.

Infants monitored by pulse oximeter (Novametrix, Bad Ems, Germany) were studied during natural sleep or after having received chloral hydrate (50 mg/kg) orally before testing. Sleeping infants were measured in a supine position with the neck in a neutral position.


No statistically significant differences in baseline characteristics between the CPAP and the mechanical ventilation (MV) group were found (table 1), except that significantly fewer doses of surfactant were applied in the early CPAP group (p<0.001), and multiple doses were more often required in the MV group (p=0.008). Five of the 17 (30%) infants from the early CPAP treatment group received no ventilation during the first 120 h of life. Patients from the early CPAP group who were mechanically ventilated required significantly lower inflation pressures, less ventilation during hours 1–120 and fewer total days of mechanical ventilation while in hospital, as well as shorter total respiratory support.

Table 1

Patient characteristics

No statistically significant differences were seen in pulmonary (FiO2 requirement, pneumothorax) or extra-pulmonary outcomes (intraventricular haemorrhage, cystic periventricular leucomalacia, necrotising enterocolitis, retinopathy of prematurity) between the two groups. However, there was a trend towards an increased need for supplementary oxygen at 28 days of life, which was higher in the MV group compared to the CPAP group (68% vs 41%; p=0.092).

Table 2 shows results from pulmonary function testing. On the day of measurement there were no statistically significant differences in post-conceptional age or body weight between the groups. Patients from the CPAP group had a significantly lower RR (p=0.007) and V’E (p=0.009) and significantly greater Crs (p=0.008). Thus, the elastic work of breathing per minute given by was significantly lower in the CPAP group compared to the MV group (p=0.004).

Table 2

Pulmonary function parameters measured at about 2 months


Pulmonary function differs significantly between VLBW infants initially supported by early CPAP compared to those managed with early intubation and mechanical ventilation. Infants treated with early CPAP needed less ventilatory support and consequently had significantly better pulmonary mechanics. There was no significant difference in lung growth, indicated by an almost identical FRC.

The identical FRC indicates that at the time of measurement, infants from both groups were able to inflate their lungs to the same extent. However, the improved Crs in the early CPAP group may be explained by better alveolarisation or less alveolar arrest. This is supported by recent experimental work by Polglase and coworkers, but this cannot be validated by our non-invasive measurements alone.4 No difference in airway conductivity was seen, as measured by Rrs, which mainly assess the resistance of the upper airways. Differences between groups might have emerged if the forced partial expiratory flow technique had been used on all our patients, as it allows assessment of small airway conductivity, but this was not possible. Regarding breathing patterns and lung mechanics, we found a lower RR and V’E in the CPAP group. As regards energy expenditure, the significantly higher Crs and higher respiratory time constant in the CPAP group resulted in favourable changes in breathing pattern, with a reduced RR and reduced V’E. These findings are in agreement with the concept of improved respiratory mechanics.5

Our study has several limitations. First, the data are from a subgroup of the COIN trial with only small patient numbers in each treatment group. This limits statistical power when comparing differences in outcome parameters and pulmonary function and increases the risk of type II error. Second, despite strict randomisation it cannot be excluded that the infants in the MV group were more severely ill than the infants in CPAP group, which would contribute to the poorer pulmonary function test results in the MV group. However, apparent differences between groups (Apgar, birth weight, gestational age) were statistically not significant. Therefore, we believe that the clearly significant differences in lung function testing post-term can be attributed to differences in the initial treatment. We believe that by avoiding early alveolar injury, the fragile lungs of VLBW infants were inflated more gently, allowing FRC to develop without injury. Hence, if these infants become ventilator dependent, they can be treated with lower pressures and for a shorter period. This non-invasive approach during the first ‘golden hours’ of life had a lasting effect, as evidenced by the significant improvements in lung function in those infants treated with early CPAP.

To conclude, early CPAP is feasible, shortens the total duration of respiratory support and results in improved lung mechanics and decreased work of breathing at about 8 weeks post-term.


The authors wish to thank Mrs S Wilitzki for carrying out the pulmonary function tests, and Dr M Pandu and Mrs J Blank for helping with the data bank management.



  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the University Ethics Committee, Charité Universitätsmedizin Berlin, Berlin, Germany.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.