Using continuous overnight pulse oximetry to guide home oxygen therapy in chronic neonatal lung disease

Aim The aims of this study are: (i) to survey the knowledge of paediatric clinicians using overnight continuous pulse oximetry data to guide management of infants with chronic neonatal lung disease (CNLD); (ii) to assess the ability of paediatric clinicians to interpret overnight continuous pulse oximetry data; and (iii) to describe the overnight oximetry interpretation practices of paediatric respiratory specialists. Methods Paediatric clinicians from three tertiary teaching hospitals completed an anonymous survey regarding overnight continuous pulse oximetry in chronic neonatal lung disease. Using a modified Delphi technique, paediatric respiratory specialists participated in a concordance exercise and discussions to establish consensus interpretations for 25 oximetry studies. Paediatric clinicians were invited to complete the same exercise as a comparison. Results Self-rated knowledge from 74 surveyed clinicians was proportional to clinical experience. Twenty paediatric clinicians and nine paediatric respiratory specialists completed the oximetry exercise with scores of 64% (kappa = 0.25) and 80% (kappa = 0.45), respectively. Individual parameters like a mean peripheral arterial haemoglobin saturation (SpO(2)) below 93% and percentage time spent below SpO(2) 93% correlated poorly with the consensus interpretations. Paediatric respiratory specialists instead relied on visual analysis of SpO(2) waveforms, utilising the frequency and depth of desaturations to guide management. Conclusion Interpretation of overnight oximetry data is variable amongst both paediatric clinicians and respiratory specialists. This likely reflects inadequate evidence defining clinically significant intermittent hypoxaemia, whether in terms of desaturation duration, frequency or nadir.