Abstract
Background: Infants often require life-saving mechanical ventilation, yet this can be injurious with adverse effects owing to ventilation-perfusion (VA/Q) mismatch and fluctuating carbon dioxide (CO2) levels. Furthermore, in those born extremely prematurely, abnormalities in gas exchange may result from an arrest in lung development, resulting in a smaller alveolar surface area (SA).Aims: (i) To describe current clinical practice in the UK with respect to carbon dioxide monitoring in newborn infants. (ii) To validate a microstream sidestream capnograph and assess the effects of non-invasive capnography monitoring on carbon dioxide control during newborn mechanical ventilation. (iii) To use volumetric capnography to calculate respiratory dead space volumes in various neonatal lung pathologies. (iv) To utilise relationships of VA/Q and apply a validated functional morphometric method of analysis to estimate the alveolar surface area in prematurely born infants, and further explore whether this composite index predicts long-term pulmonary morbidity such as the need for supplemental home oxygen.
Methods: A national survey of all level two and three neonatal intensive care units was undertaken to determine current practice in non-invasive monitoring of carbon dioxide. A microstream sidestream capnograph was validated against a ‘gold standard’ low dead space mainstream capnograph. Subsequently, the frequency of blood gas analyses and CO2 values were evaluated post introduction of sidestream capnography on the neonatal intensive care unit. Volumetric capnograms were constructed for ventilated infants with respiratory distress syndrome (RDS) and evolving bronchopulmonary dysplasia (BPD), with calculation of respiratory dead space volumes. In those born extremely prematurely, the fraction of inspired oxygen and corresponding transcutaneous oxygen saturation were measured, and VA/Q calculated by plotting those pairs on the oxyhaemoglobin dissociation curve. A validated method of functional morphometry, utilising Fick’s First Law of diffusion and the non-invasively calculated VA/Q ratio, was applied to determine the SA. Finally, to reflect the functioning gas exchanging SA, the fraction of alveolar dead space was deducted from the SA to give an adjusted SA index and thus provided an index of pulmonary morbidity.
Results: Eighty-two percent of UK neonatal units reported monitoring carbon dioxide during intubation. The sidestream capnograph performed similarly to the mainstream device [r = 0.85, p < 0.001], however there was no significant reduction in the frequency of invasive blood gas sampling with continuous use. End-tidal carbon dioxide (EtCO2) values better predicted blood CO2 (pCO2) levels in infants with less severe respiratory disease compared to those with more severe pulmonary disease [r2 = 0.66 versus r2 = 0.33]. Preterm infants with BPD requiring supplemental home oxygen had a larger pCO2 – EtCO2 gradient than those infants not requiring supplemental oxygen on discharge [11.1 versus 9.4 mmHg, p = 0.002]. Physiological dead space was higher in infants with RDS [median 5.7 ml/kg] and BPD [6.4 ml/kg] compared to infants with no respiratory disease [3.5 ml/kg; p < 0.001], however alveolar ventilation was maintained despite the higher dead space. The median adjusted SA was 647.9 cm2, and was lower in the extremely preterm infants that required supplemental home oxygen [637.7 cm2] compared to those that did not [799.1 cm2; p = 0.016]. An adjusted SA ≥ 688.6 cm2 had 86% sensitivity and 77% specificity in predicting the need for supplemental home oxygen [area under the receiver operating characteristic curve = 0.815; p = 0.017].
Conclusions: Carbon dioxide monitoring during intubation is performed in many neonatal units. Sidestream capnography was validated for use within the neonatal population. Volumetric capnography was used to describe larger volumes of physiological dead space in preterm infants. Preterm infants with bronchopulmonary dysplasia had increased volumes of alveolar dead space likely due to underlying developmental lung pathology. Utilising non-invasive physiological methods of volumetric capnography with functional morphometry enabled the estimation of an adjusted index of alveolar surface area. This index predicted the need for supplemental home oxygen and has potential to predict later pulmonary morbidity in those infants born extremely prematurely.
| Date of Award | 1 Aug 2023 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Anne Greenough (Supervisor) & Theodore Dassios (Supervisor) |