The purpose of this study was to assess the agreement and consistency between gas exchange variables measured by two online metabolic systems during an incremental exercise test. After obtaining local ethics approval and informed consent, 15 healthy subjects performed an incremental exercise test to volitional fatigue using the Bruce protocol. The Innocor (Innovision, Denmark) and CardiO₂ (Medical Graphics, USA) systems were placed in series, with the Innocor mouthpiece attached to the pneumotach of the CardiO₂. Metabolic data were analysed during the last 30 seconds of each stage and at peak exercise. There were non- significant differences (p > 0.05) between the two systems in estimation of oxygen consumption (VO₂) and in minute ventilation (V_E). Mean Cronbach’s alpha for VO₂ and V_E were 0.88 and 0.92. The Bland-Altman analysis revealed that limits of agreement were -0.52 to 0.55 l.min^-1 for VO₂, and -8.74 to 10.66 l.min^-1 for V_E. Carbon dioxide production (VCO₂) and consequently respiratory exchange ratio (RER) measured by the Innocor were significantly lower (p < 0.05) through all stages. The CardiO₂ measured fraction of expired carbon dioxide (FeCO₂) significantly higher (p < 0.05). The limits of agreement for VO₂ and V_E are wide and unacceptable in cardio-pulmonary exercise testing. The Innocor reported VCO₂ systematically lower. Therefore the Innocor and CardiO₂ metabolic systems cannot be used interchangeably without affecting the diagnosis of an individual patient. Results from the present study support previous suggestion that considerable care is needed when comparing metabolic data obtained from different automated metabolic systems.