There have been very few studies of how to wean babies from CPAP, even though it is something we do all the time.
A recent tendency, which seems to have arisen without any supportive data, and which never made much sense to me, is to take the baby off CPAP for a specified period, such as for one hour every 8 hours,and then gradually prolong the duration of these weaning episodes. The alternative is to wean the pressure to a minimal level (I would call 4 cmH2O minimal, below this there is little or no measurable positive pressure in the posterior pharynx) and then stop the CPAP if the baby is comfortable, only restarting if there are signs of poor tolerance.
David Todd and his colleagues in Australia have just published (on-line first) a trial comparing these 2 approaches. Todd DA, Wright A, Broom M, Chauhan M, Meskell S, Cameron C, et al. Methods of weaning preterm babies <30 weeks gestation off CPAP: a multicentre randomised controlled trial. Archives of Disease in Childhood – Fetal and Neonatal Edition. 2012. http://fn.bmj.com/content/early/2012/05/17/adc.2011-300133.abstract
They randomized 177 stable babies, less than 30 weeks gestation who were on CPAP to stopping the pressure and restarting only if there is poor tolerance, compared to 2 groups who had weaning periods as described above, one of the weaning period groups had low flow nasal cannulae during the weaning period (another strange practice that has developed with no physiologic rationale and no supporting clinical data), the other did not.
Although the study was of modest size, with between 50 and 60 in each group, the results were dramatic. In fact it was stopped early by the data safety committee because all of the respiratory outcomes were significantly better in group who had immediate stopping of CPAP without the weaning periods; and the differences were large.
There were fewer days on CPAP (which is probably to be expected by trial design) and this translated into fewer days of oxygen, less BPD, fewer days of hospitalization and in tandem with that earlier postmenstrual age at discharge.
This result is consistent with 2 other trials which have never been fully published, but which both show improved respiratory outcomes with slightly different comparisons. The 2 trials are back-to-back abstracts in the European Journal Of Pediatrics November 2006 supplement, page 48. The first by Singh et al randomized 112 infants less than 1500 gr birthweight to a progressive reduction in pressure or to increasing periods of CPAP. Infants on the weaning by pressure schedule had many fewer days of CPAP, and more successful weaning. The 2nd by Soe et al randomized 98 infants less than 32 weeks to weaning the pressure or weaning by increasing periods off CPAP. Babies in the pressure weaning group were more likely to wean successfully, and had less BPD.
The practice of progressively prolonging weaning periods has no scientific basis, has been shown to be associated with worse outcomes, and should stop.