Portable gas analyser Cosmed K4b2 compared to a laboratory based mass spectrometer system

The purpose of this study was to systematically test the accuracy of an automated, portable, gas analysis system, the Cosmed K4b2 with a laboratory based mass spectrometer system, the Morgan EX670 across a number of gas and ventilation parameters.

Eight subjects (mean+/-SE) age, 23.7+/-1.1 y, height, 1.78+/-0.01 m, mass, 74.4+/-2.1 kg performed a V.O2max test and a submaximal exercise test at 150, 200, 250 and 300 Watts (W), on an SRM cycle ergometer. The Morgan EX670 system and the K4b2 were randomly connected in series, using the same breath for the calculation of gas and ventilatory parameters.

The K4b2 system reads significantly higher than the Morgan EX 670 for both VO2 and V.CO2 at 250 (VO2/V.CO2: p<0.05, p<0.002), and 300 W (VO2/V.CO2: p<0.002, p<0.005). Unsystematic bias between the 2 analysers varies between 1% and 16% and systematic bias between 3% and 8%.

There are some significant unsystematic and systematic differences between these 2 systems and laboratories should endeavour to utilise either one or the other piece of equipment to test their subjects.