Carbon dioxide clearance during apnoea. A reply.

Abstract

We thank Dr. Lyons for his observations on our paper 1, and offer the following thoughts in response. Firstly, the mouth was modelled as being open. As stated in the paper, we used the flow profile measured by Tusman et al. 2. Therefore, the inspiratory and expiratory times are from actual patients undergoing anaesthesia. We included data on a range of cardiogenic stroke volumes (5–40 ml) as reported clinically 2-4. We did not report results from models with different airway geometries in our publication, although preliminary data indicated detectable enhancement of carbon dioxide clearance in stenotic airways using transnasal humidified rapid insufflation ventilatory exchange (THRIVE), consistent with clinical data; we intend to publish these data shortly. Indeed, the clinical observation that THRIVE is efficacious in patients with stenoses 5, 6 suggests a driving pressure high enough to generate tracheal gas flow, which would be consistent with the driving pressure arising from cardiogenic oscillations. The effect of cardiogenic oscillations operating on the small airways is included in the model, not through replication of small airways per se but through the computer-controlled ‘diaphragm’ that acts on the model plenum. As Dr. Lyons notes, there is inter-patient variation in the rate of carbon dioxide clearance in patients treated with THRIVE 5, 7. However, this does not imply that oxygenation and clearance of carbon dioxide are not occurring simultaneously. Furthermore, different efficacies of oxygenation and carbon dioxide clearance do not speak to the validity of the model. Indeed, one would not expect accurate symmetry due to the inherent fluid characteristics and physiological non-linearities of the gases. Naturally there is also less variation in oxygenation success when measured using SpO2, and drawing conclusions about the symmetry of carbon dioxide clearance and oxygenation from clinical measures is complex. We would suggest caution in comparing clinical data from children 8 and adults 5, 7. There are obvious physiological and anatomical differences. Our model is of an adult airway and it was not our intention to imply that these mechanisms are applicable to the neonatal airway. Indeed, preliminary data from neonatal versions of our models, which we intend to publish in the future, show significant differences in the fluid flow compared with the adult model that was the subject of our publication. The preliminary laboratory results in the neonatal version of our model are consistent with the clinically observed lower rates of carbon dioxide clearance in children, which provides further validation of our model. We agree that there are differences in end-tidal and arterial blood gas measures and noted this in our paper. Finally, we note that the mechanisms we have proposed have been corroborated by further work 9. That said, there is undoubtedly more to understand and further clinical and modelling work will be beneficial in augmenting our current understanding of THRIVE in clinical practice.