Transcutaneous PCO2 monitoring in critically ill adults: clinical evaluation of a new sensor.

OBJECTIVE: In critically ill patients, arterial blood gas analysis is the gold standard for evaluating systemic oxygenation and carbon dioxide partial pressure. A new miniaturized carbon dioxide tension Pco2-Spo2 single sensor (TOSCA, Linde Medical Sensors AG, Basel, Switzerland) continuously and noninvasively (transcutaneously) monitors both Paco2 and oxygen saturation by pulse oximetry (Spo2). The present study was designed to investigate the usability and the accuracy of this device in critically ill patients.

DESIGN: Prospective clinical investigation.

SETTING: A 20-bed, university-affiliated, surgical intensive care unit.

PATIENTS: Patients admitted after major surgery, multiple trauma, or septic shock equipped with an arterial catheter.

INTERVENTIONS: The heated (42 degrees C) sensor was fixed at the earlobe using an attachment clip. Transcutaneous Pco2 (TcPco2) measurements were correlated with Paco2 values (measured using a blood gas analyzer). In addition, the differences between Paco2 and TcPco2 values were evaluated using the method of Bland-Altman.

MEASUREMENTS AND MAIN RESULTS: We studied 55 patients, aged 18-80 (mean 57 +/- 15) yrs. A total of 417 paired measurements were compared. Correlation between TcPco2 and Paco2 was r = .86 (p < .01) in the Paco2 range of 24-101 mm Hg. Mean bias (+/-sd) between the two methods of measurement (Bland-Altman analysis) was 1.2 +/- 6.0 mm Hg with TcPco2 slightly overestimating arterial carbon dioxide tension. Nineteen percent of the measured values were outside of the acceptable clinical range of agreement of +/-7.5 mm Hg.

CONCLUSIONS: The present study suggests that Paco2 can be acceptably assessed by measuring TcPco2 using the TOSCA Pco2-Spo2 sensor.