Page 110 - Physiological based CPAP for preterm infants at birth Tessa Martherus

Chapter 4
from the large variability associated with application of rescue therapies (iPPV and additional caffeine) and differences in physical stimulation levels, which are difficult to standardize. Nevertheless, the close relation between CPAP level and FiO2 requirement was clearly indicated by changing CPAP levels. To maintain a similar SaO2, decreasing CPAP levels increased the FiO2 requirement, whereas increasing CPAP levels reduced the FiO2 requirement. This is consistent with the concept that higher CPAP levels increases the available area for gas exchange. Whilst supplemental oxygen stimulates spontaneous breathing and improves oxygenation, as too much oxygen can cause harm (through hyperoxia) it is important to limit the use of high O2 concentrations (10, 55, 56). Our results, and those of our previous study (30), indicate that this can be achieved through the use of high-CPAP, which increases lung aeration, improves gas exchange, and reduces the FiO2 requirement. Future studies are required to confirm the role of high-CPAP on improving lung aeration, and to investigate the interaction between CPAP level and FiO2, to find the optimal support strategy that can be tested in preterm infants at birth.
Contrary to our hypothesis, the dynamic high-CPAP strategy was not superior to high-CPAP. Indeed, our findings indicate that the dynamic high-CPAP group may have benefitted by remaining at the high CPAP level for longer and delaying the decrease. Whilst these lambs continued to have a high PBF and SaO2 following the reduction in CPAP, they needed a higher FiO2 to maintain their oxygenation levels and most probably their level of breathing activity. As preterm infants often only require a maximum of 8 cmH2O CPAP in the neonatal ward, it is necessary to decrease the CPAP level at some stage, but the questions of when and upon what indication, remains unknown. Clearly, the strategy of dynamic high-CPAP strategy needs further investigation, to identify the desirable moment for reducing the CPAP level.
Our model of spontaneously breathing preterm animals over time introduces some bias into the study because it incorporates strategies to stimulate lambs to breath. As a result, differential use of rescue interventions between groups effected the results by improving study outcomes in groups with higher rates of rescue interventions. As the interventions were required more frequently in the low-CPAP and dynamic high-CPAP group, mostly after reducing the CPAP level, the differences between these groups and the high-CPAP group were potentially reduced. The early drop-out of animals in the low-CPAP group, which occurred after they had reached the ethical endpoint where intubation was required, likely further reduced the difference between the groups. As such, high-CPAP levels likely benefitted these preterm sheep to a greater degree than we were able to demonstrate.
It is unclear why lambs in all groups experienced high pCO2 and low pH levels, although it is likely to be mostly metabolic in origin as indicated by the high lactate and large negative base excess levels. Indeed, at 15-30 min, lambs in the HCPAP group had a high SaO2 (>90%) while receiving a relatively low FiO2 (~30%), but still had high PCO2 levels. As the solubility and exchange potential for CO2 is almost 30 times greater than O2, a problem with CO2 exchange in the absence of a problem with O2 exchange is unlikely and so the high PCO2 is most likely
106