Carbon dioxide tensions in manually ventilated, prone patients.

The distribution of arterial carbon dioxide tensions (PaCO2) in 288 anaesthetized, healthy patients in the prone position was investigated during non-monitored manual ventilation. Four equal groups of 72 patients were compared employing a conventional anaesthetic system with and without CO2-absorption and a modified Mapleson D rebreathing system using high and low fresh gas flows. No fundamental difference between the Mapleson D system and the circle system without CO2-absorption could be demonstrated. With the Mapleson D system a high fresh gas flow resulted in a loss of rebreathing characteristics, the scatter of PaCO2-values then approaching that of a conventional circle system with CO2-absorption. The results demonstrate that during manual ventilation a CO2-absorption results in unpredictable levels of PaCO2 with unacceptably wide ranges. On the other hand, prevention of hypocapnia could be obtained with the rebreathing systems using a low fresh gas flow. Hypercapnia was modest and similar in all groups, PaCO2 never exceeding 7.3 kPa. In terms of PaCO2-levels it seems, compared to recent literature, of no importance whether the patient is prone or supine or whether or not ventilation is monitored, performed by hand or by means of anaesthetic ventilators. Accurate levels of PaCO2 can probably only be obtained by frequent analyses of PaCO2, or monitoring of the end-tidal CO2 concentration.