Estimation of arterial PCO2 from a lung model during ramp exercise in healthy young subjects

The aim of this study is to propose a new approach to estimate non-invasively arterial carbon dioxide partial pressure (P(a)CO2) approach was based on the reconstruction of alveolar gas composition over each breath from a tidally ventilated lung model (P(M)(CO2)). Eight healthy young subjects were studied during a ramp exercise test on a cycle ergometer. Arterial samples were drawn at rest and every minute during the exercise test for determination of P(a)CO2 . P(a)CO2 was compared with indirect estimates of P(CO2) : P(M)(CO2), end-tidal P(CO2) (P(ET)(CO2)) and an empirical equation involving P(ET)(CO2) and tidal volume (P(J)(CO2)). The difference between estimated and measured P(a)CO2 on the whole ramp exercise was -0.3+/-1.9mmHg for P(M)(CO2), 1.0+/-2.2mmHg for P(ET)(CO2) and -1.7+/-1.7mmHg for P(J)(CO2) . P(ET)(CO2) and P(J)(CO2) were significantly different from actual P(a)CO2 (P<0.001). It is concluded that, on the basis of the bias, the breathing lung model gave better estimates of P(a)CO2 than the two other indirect methods during ramp exercise.