Performance of transcutaneous PCO2 and pulse oximetry monitors in newborns and infants after cardiac surgery

We examined the effect of core and skin temperature on the accuracy of two pulse oximeters (Nellcor Symphony and Hewlett Packard saturation module, M1020A) and a transcutaneous PCO2 monitor (Fastrac Transcutaneous monitor) immediately after cardiac surgery in a group of newborns and infants. Seventy-nine sets of data were collected from 46 patients. Core temperatures ranged from 35.3 degrees C to 39.4 degrees C, skin temperatures ranged from 27.0 degrees C to 37.4 degrees C and core-skin temperature gradients ranged from 0.1 degree C to 10.1 degrees C. Data analysis consisted of comparing the difference between transcutaneous PCO2 and arterial PCO2 and the differences between oxygen haemoglobin saturation measured by both pulse oximeters and oxygen haemoglobin saturation measured by co-oximeter to core temperature, skin temperature and core-skin temperature gradients. The mean differences +/- standard deviations and limits of agreement for transcutaneous PCO2 and oxygen haemoglobin saturation measured by the Hewlett Packard and Nellcor pulse oximeters were 0.95 +/- 4.10 mmHg (-7.09 mmHg to 8.99 mmHg), -1.07 +/- 1.84% (-4.68% to 2.54%) and -1.23 +/- 2.23% (-5.60% to 3.14%) respectively. Analysis of correlation coefficients showed that the accuracy of the transcutaneous PCO2 monitor and the pulse oximeters were not affected by core temperature, skin temperature or core-skin temperature gradient in the ranges encountered. We therefore conclude that these devices are acceptably accurate and suitable for use in infants when core and skin temperatures and core-skin temperature gradient are in the range normally found after cardiac surgery.