Validity of transcutaneous oxygen/carbon dioxide pressure measurement in the monitoring of mechanical ventilation in stable chronic respiratory failure.

The accuracy and precision of transcutaneous pressure measurements of oxygen (Ptc,O2) and carbon dioxide (Ptc,CO2) in the monitoring of nocturnal assisted ventilation in adult patients were evaluated. Transcutaneous measurements obtained with two analysers, Radiometer TINATCM3 (R) and Kontron MicroGas-7650 (K), were compared with arterial blood gases analysed in blood samples withdrawn simultaneously in 10 patients. Sensors were heated to 43 degrees C. Measurements of transcutaneous blood gases and arterial blood gases were collected six times at 1-h intervals. The data obtained with both instruments were similar and did not significantly change over the 5 h test period. Measurement of Ptc,O2 underestimated arterial oxygen tension (Pa,O2) and this underestimation increased with the level of Pa,O2 (p<0.01). Measurements of Ptc,CO2 overestimated arterial carbon dioxide tension (Pa,CO2) and this overestimation increased with the level of Pa,CO2 (p<0.05). These errors suggested an instrumental bias. Mathematical correction of this bias neutralized the error in accuracy and improved the precision (SD of the differences transcutaneous blood gases - arterial blood gases). An additional correction, suppressing the between-subject scattering, improved the actual precision: precision was reduced from 1.9 to 0.8 kPa (14.4 to 5.7 mmHg) (R) and from 1.7 to 0.5 kPa (13.1 to 3.7 mmHg) (K) for oxygen, and from 1.0 kPa (7.8 mmHg) (R) and 0.7 kPa (5.6 mmHg) (K) to 0.4 kPa (3.2 mmHg) for carbon dioxide (R and K). In conclusion, with these two successive corrections, transcutaneous oxygen and carbon dioxide provide a reliable estimation of blood gases to monitor nocturnal ventilation in adults with chronic respiratory failure.