High frequency oscillation.

It is more than 15 years since high frequency oscillatory ventilation (HFOV) was first described as a technique of ventilation [ 1 ]. In the intervening period it has passed through a phase of being a physiological curiosity, where it could be demonstrated that normal gas exchange could be maintained using tidal volumes much smaller than deadspace at rates much greater than what was then considered appropriate for mechanical ventilation. As it was shown to work very effectively in normal lungs, the next logical step was to see if the system was a viable alternative ventilation strategy in the severely diseased lung. This led to HFOV being first used, principally as a rescue operation in neonates with respiratory failure in whom conventional ventilation techniques had failed. Several studies have shown that when changing from conventional mechanical ventilation (CMV) to HFOV in these infants, PaCOi tensions may be reduced rapidly from the previously increased values using lower airway pressures, although it has been difficult to show any dramatic improvement in 7aOi [2, 3]. Although HFOV has achieved some acceptance as an alternative ventilation technique in the management of newborn respiratory failure, there are still substantial gaps in our understanding of exactly how the system works, despite a considerable amount of research. In a recent review on high frequency ventilation, Froese and Bryan [4] pointed to the increasing number of publications on the subject, many repetitious, frequently with contradictory results; this has led to considerable confusion as to what role, if any, HFOV has as a mode of ventilation in clinical practice. All the original HFOV studies performed on animals with normal lungs were of short duration and the technology, in terms of ventilator design and circuitry, was relatively simple. If HFOV is to