The present study was designed to analyze the usability of a commercially available, transcutaneous PCO2 (TcPCO2) sensor for monitoring noninvasive positive pressure ventilation (NPPV). Twenty-six hemodynamically stable patients with intra-arterial radial catheters were assessed. After stabilization of TcPCO2, arterial blood was analyzed and results were compared with TcPCO2 at time of sampling. To evaluate the drift of the signal, samples were taken hourly in five patients for 4 h while continuously recording TcPCO2. Finally, to assess for the response of the sensor to changes in PaCO2, six patients underwent continuous TcPCO2 recording while initiating or interrupting NPPV; arterial samples were analyzed before the event, and 1, 3, 5, 7, 9, and 20 min afterwards.
Results: TcPCO2 and PaCO2 were tested over a range of 26 to 71 mm Hg, and were found to be closely correlated (r=0.968, p<0.0001); mean bias was 0.75 mm Hg. There was no significant drift of TcPCO2 as compared with PaCO2 over 4 h. The time of response of TcPCO2 to initiation or interruption of NPPV was <60 s. An estimation of the lag time averaged 5+/-3 min (range, 1 to 9 min).
Conclusion: TcPCO2 in hemodynamically stable adults was in excellent agreement with arterial measurements. The time of response to a change in ventilation was compatible with the aim of clinical monitoring of patients under NPPV.