Editor,—Thomson’s recent review of negative pressure ventilation1 included results of our randomised controlled trial of 244 infants with neonatal respiratory failure.2 She commented that there was “a small overall benefit of CNEP [continuous negative extrathoracic pressure] but with non-significant increases in mortality, pneumothorax rate, and cranial ultrasound abnormalities in the CNEP group”. An important result not mentioned in her review was that the median duration of oxygen requirement was 33.6 days in the controls and 18.3 days in the infants treated with negative pressure (a difference of 15.3 days, 95% CI −0.2 to −30.4). We suggest, therefore, that negative pressure has an important role, particularly in the resolution phase of respiratory distress syndrome in preterm infants, and may reduce the risk of developing chronic lung disease. Newer methods for providing continuous nasal positive pressure support to preterm infants still fail to guarantee a transpulmonary pressure gradient, whereas negative pressure has provided this effectively, and quickly.3

We agree that, in school age children, the latest techniques for providing nasal mask ventilation may be preferable to negative pressure ventilation in the treatment of chronic respiratory failure. In addition, we have successfully used nasal mask positive pressure ventilation in three infants with congenital central hypoventilation syndrome.4

Our most frequent use of negative pressure ventilation is in the treatment of acute bronchiolitis. In infants with increasing oxygen requirements or recurrent apnoeic–hypoxaemic episodes, negative pressure ventilation may avoid the need for intubation and intensive care.5 The early use of negative pressure ventilation has resulted in a 0.4% intubation rate in our unit over the past four years. This compares favourably with published rates.

Finally, Thomson recommends monitoring during negative pressure ventilation with pulse oximetry and end tidal carbon dioxide. We find transcutaneous carbon dioxide monitoring more practical as this avoids the need for expired air sampling, and displaying and interpreting a capnogram.