Abstract
Oxygenation impairment can be assessed non-invasively by determining the degree of right-to-left shunt and ventilation/perfusion (V-A/Q) inequality. These indices have been used in sick newborn infants, but normative values have not been reported which are essential to determine the magnitude of the abnormality. We, therefore, aimed to measure the shunt and V-A/Q in infants with no history of respiratory conditions and determine if there was any effect of supine or prone position and the reproducibility of the data. Data were analysed from infants who had undergone a hypoxic challenge and in a subset who had been assessed in the supine or prone position. Transcutaneous oxygen saturations (SpO(2)) were recorded at fractions of inspired oxygen (FIO2) of 0.21 and 0.15. Two independent raters used a computer software algorithm which analysed and fitted paired data for FIO2 and SpO(2) and derived a curve which represented the best fit for each infant's data and calculated the shunt and V-A/Q. The raters ability to interpret the SpO(2) value which corresponded to a given FIO2 was compared. The downwards displacement of the FIO2 versus SpO(2) curve was used to estimate the degree of right-to-left shunt and the rightwards shift of the curve was used to calculate the V-A/Q ratio. The mean (SD) gestational age of the 145 infants was 39 (1.6) weeks, their birth weight was 2990 (578) gms and median (range) postnatal age at measurement 3 (1-8) days. The mean (SD) V-A/Q ratio was 0.95 (0.21). None of the infants had a right-to-left shunt. No significant differences were found in V-A/Q in the supine compared to the prone position. The intraclass correlation coefficient of V-A/Q between two independent raters was 0.968 (95% CI 0.947-0.980), p < 0.001. Right-to-left shunt and V-A/Q ratio in healthy newborn infants were similar in the prone compared to the supine position.