Trancutaneous PCO 2 and End-Tidal PCO 2 in Ventilated Adults

Abstract

Continuous noninvasive ventilation monitoring can provide valuable information for the management of the ventilated adult patient. We evaluated the accuracy of transcutaneous PCO2 (PtcCO2) and end-tidal PCO2 (PetCO2) measurements as com- pared to arterial CO2 tension (PaCO2) measurements in adult patients on mechan- ical ventilators. METHODS: A convenience sample of 30 patients, ages 29 to 82 years, who were on mechanical ventilators in our Special Care Unit were included in the study. A PtcO2-PtcCO2 sensor was placed on the upper chest of each patient. A mainstream PetCO2 monitor-pulse oximeter was attached to the patient's arti- ficial airway according to manufacturer's recommendations. Arterial blood sam- ples were obtained and analyzed according to departmental protocol. RESULTS: 54 blood samples were obtained via indwelling catheters from the 30 patients (1-3 samples/patient) and PaCO2 was compared to PetCO2 and PtcCO2. The mean devia- tion (bias) for PetCO2 - PaCO2 was -5.13 and the precision was 8.25 when all samples were compared. PtcCO2 - PaCO2 produced an improved bias of -0.54 with a pre- cision of 5.23 in the total population studied. In patients with documented res- piratory disease, the 2-SD limits of agreement between PetCO2 and PaCO2 were -27.56 to 6.56 with a bias of -10.50 and a precision of 8.53. For PtcCO2 - PaCO2 in this group of patients, the 2-SD limits of agreement were -14.11 to 13.70 with a bias of -0.21 and a precision of 6.95. In patients without respiratory disease, the limits of agreement were closer, bias did not differ between the two monitoring tech- niques, and the precision was not dramatically different. CONCLUSIONS: In patients with respiratory pathology, PtcCO2 monitoring more accurately described PaCO2 and provided a better indication of ventilation status than did PetCO2. In ven- tilated patients without respiratory disease, both monitoring techniques were sim- ilarly accurate. These findings point to the importance of assessing patient pathol- ogy and instrumentation when instituting and utilizing noninvasive ventilation monitoring techniques.