Objective
: The electrochemical approach with CO
2
electrodes is the gold standard for the measurement of transcutaneous CO
2
(PtCO
2
). However, this system requires the continuous calibration of the device, the remembranization of the electrodes, an infrastructure with specialized personnel, and a still position during the measurement. This paper presents a wearable device for PtCO
2
measurement with an optical approach (non-dispersive infrared sensor).
Methods
: The device first measures the environmental CO
2
, then brings the skin temperature to 42 °C, and finally measures the actual PtCO
2
. 30 volunteers (20 men, 10 women, mean age 27.2 ± 7.0) were enrolled to validate the new device against the gold standard Sentec Digital monitor. The protocol consisted in 10 minutes of spontaneous breathing, 2 minutes of rebreathing in a 2 L bag to induce hypercapnia, and 12 minutes of recovery. This was repeated for each volunteer because the gold standard probe was placed once on the forearm and once at the earlobe. Also, a conversion method from ppm readings of the new device to mmHg was implemented and evaluated.
Results
: The new device detects a response to the stimulus of hypercapnia with a time delay comparable to that of the gold standard. Before reaching a baseline, both devices require an assessment time approximatively in the order of minutes. The Sentec device goes back to the baseline after the end of the stimulus, while the new device saturates. After conversion of values into mmHg, the Intraclass Correlation Coefficient analysis on median values results in a coefficient of 0.75. The new device has a median root means square error of 2.38 mmHg (IQR=1.59 mmHg). The median of the mean absolute error is 2.05 mmHg (IQR=1.64 mmHg) and the uncertainty is ±4.63 mmHg (IQR=2.38 mmHg).
Conclusion
: This work demonstrates that a wearable device to measure PtCO
2
based on an optical sensor is technically feasible.
