Background and objective: The best place for the electrode of transcutaneous measurement of oxygen tension (tcPO2) and carbon dioxide tension (tcPCO2) during general anaesthesia was investigated in three different locations. Methods: Fifteen patients for major abdominal surgery in the supine position were studied. The electrode of the TCM4™ (Radiometer, Copenhagen, Denmark) was put on the chest, upper arm or forearm. TcPO2, tcPCO2, end-tidal carbon dioxide tension (etCO2), percutaneous oxygen saturation (SaO2), arterial oxygen tension (PaO2 ) and arterial carbon dioxide tension (PaCO2) were simultaneously measured at four different etCO2 concentrations and inhaled oxygen percentages and the location of the electrode was changed to other places to repeat the same measurement. In total, 12 measurements for each patient and 60 measurements for each place were performed. Results: TcPO2 correlated better than SaO2 (R2 = 0.58) with PaO2 (R2 = 0.76), and tcPCO2 correlated well with PaCO2 (R2 = 0.76) and etCO2 (R2 = 0.74) when the electrode was put on the chest, while not on the upper arm or forearm (R2 < 0.5). However, limits of agreement were too big to use tcPO2 (bias, −67.9; limits of agreement, 16.5, −152.3) and tcPCO2 (bias, −0.47; limits of agreement, 8.7, −9.64) as surrogate measures of PaO2 and PaCO2, respectively even when the electrode was put on the chest. Conclusions: When the electrode was put on the chest, tcPO2 and tcPCO2 might be available as non-invasive monitors of oxygenation and CO2 status during general anaesthesia, while the absolute values were not interchangeable with PaO2 and PaCO2, respectively.