Leak during manual neonatal ventilation and its effect on the delivered pressures and volumes: an in vitro study.

BACKGROUND

Mask leak is a frequent problem during manual ventilation. Our aim was to investigate the effect of predefined leaks on delivered peak inflation pressure (PIP), positive end-expiratory pressure (PEEP) and tidal volume (V(t)) when using different neonatal manual ventilation devices.

METHODS

A neonatal-lung model was ventilated at different respiratory rates (RRs, 40, 60, 80/min) using a mechanically operated self-inflating bag (SIB) and a manually operated T-piece resuscitator (PIP = 20 cm H(2)O, PEEP = 5 cm H(2)O). Four open tubes of different lengths, which produced up to 90% leak, were consecutively attached between the ventilation device and the lung model. A pneumotachograph was used to measure pressures, flow and volume.

RESULTS

With increasing leak (0-90%) PIP and PEEP decreased significantly (p < 0.001) for both devices. Using the SIB, the mean ± SD PIP fell from 20.1 ± 0.3 to 15.9 ± 7 cm H(2)O and PEEP fell from 5.0 ± 0 to 0.3 ± 0.5 cm H(2)O, leading to an increased pressure difference (Δp); V(t) increased from 8.8 ± 0.7 to 11.1 ± 0.8 ml (p < 0.001). With increasing RRs, the leak-dependent changes were significantly lower (p < 0.001). Using the T-piece resuscitator, PIP dropped independent of RRs from 20.3 ± 0.5 to 18.5 ± 0.6 cm H(2)O and PEEP from 5.1 ± 0.4 to 4.0 ± 0 cm H(2)O, while Δp and V(t) did not differ significantly.

CONCLUSION

The decrease in PIP and PEEP with increasing leak is RR dependent and distinctly higher when using an SIB compared to a T-piece device. In contrast to V(t) delivered with the SIB, V(t) delivered by the T-piece resuscitator was nearly constant even for leaks up to 90%.