Comparison of End-Tidal, Arterial, Venous, and Transcutaneous P_CO2

Discussion

Our findings showed that simple and noninvasive measurements of P_ETCO2 with a capnometer estimated P_aCO2. However, the limits of agreement were wide. Therefore, care providers must pay attention to the characteristics and errors of these devices.

P_ETCO2 is often measured in closed circuits in patients undergoing mechanical ventilation. In the absence of significant changes in cardiac output or the ventilation/perfusion ratio, P_aCO2 can then be estimated with clinically acceptable sensitivity and specificity in comparison with P_ETCO2.^5,6 However, accurate measurement of P_ETCO2 has been problematic in patients who are not intubated because it is difficult to collect pure respiratory gas. Several recently developed transportable capnometers can measure P_ETCO2 easily and noninvasively in patients who are not intubated, and a recent study showed that they could provide reliable P_ETCO2 values compared with P_aCO2.⁷ We therefore examined the usefulness of this approach in terms of the principle, experience of use, and measured values. A previous study showed that P_ETCO2 was 2–5 mm Hg lower than P_aCO2 during general anesthesia in subjects without lung disease,⁸ and we observed that P_ETCO2 was 6 ± 6 mm Hg lower than P_aCO2. This study revealed wide limits of agreement in each group, which suggests that the measurement of P_ETCO2 with a capnometer requires attention to the characteristics and errors of these devices.

P_tcCO2 is more accurate than other measurements in any patient group. However, P_tcCO2 is complicated to measure and the required equipment is expensive, and therefore this technique is less widely used than pulse oximetry. There is thus a need to develop devices that can easily estimate P_aCO2.

A previous study reported that P_aCO2 measurements at the time of COPD exacerbation could be replaced with P_v̄CO2 measurements.⁹ However, although P_v̄CO2 may be a useful measure, venous blood sampling is mildly invasive. Our results indicate that P_ETCO2 agreed with P_aCO2 as well as or better than P_v̄CO2 in any subject group, suggesting that P_ETCO2 may be a more useful measure than P_v̄CO2.

The relative advantages and disadvantages of the capnometer and the transcutaneous carbon dioxide monitor are shown in Table 2. The capnometer is small and easy to carry, allowing it to be used regardless of location. In addition, measurements can be performed within a few minutes, allowing its use even during out-patient clinical practice. It can also be used in settings outside of the hospital or out-patient clinic where blood gas analysis is not an option, such as during house calls. We observed no measurement-related complications in this study, and the patient burden was very small. Furthermore, the only equipment required is disposable mouthpieces, allowing the capnometer to measure P_ETCO2 at low cost. The capnometer has a disadvantage, however, in that the measured value changes depending on the patient's breathing pattern, and measured values of inhaled oxygen may be affected by oxygen flow. The developer of the device states that supplemental O₂ does not significantly affect the examination value at a flow ≤ 2 L of nasal oxygen. However, administration must be interrupted in cases of higher oxygen flows. In our study, we interrupted oxygen delivery at the time of measurement, and we therefore did not consider the relationship between oxygen flow and the measured value. Further studies are needed to clarify the effect of oxygen administration.

Another disadvantage of the capnometer is its inability to continuously monitor and record measurement values inside the equipment. Transcutaneous carbon dioxide monitors provide such measurements, and our results indicate that P_tcCO2 is more accurate than P_ETCO2. Transcutaneous carbon dioxide monitors also measure P_tcCO2 continuously, and they can be used in patients with altered mental status, making them useful in ICUs and operating rooms. However, the measurement method is complicated and lengthy, and errors are caused by mixing of air. Furthermore, low-temperature burns have also been reported,¹¹ and the device must therefore be moved periodically to avoid this complication. This technique is limited further by the relative expense of the equipment and accessories. In addition, measurements obtained with transcutaneous carbon dioxide monitors tend to be higher than the P_aCO2 when the measured value is ≥ 50 mm Hg.⁴ We confirmed that measurement of P_ETCO2 with a capnometer could provide a simpler approach than measurement of P_tcCO2.

This study had several limitations. First, the capnometer measurement depends on the subject's breathing pattern, thus the examiner must be familiar with the measurement method. Furthermore, caution is needed because expired air may leak from the corners of the patient's nose and mouth. Although P_ETCO2 was measured only once in this study, it may be possible to reduce error by measuring P_ETCO2 several times and using the average value. Second, numerous factors reportedly weaken the correlation between P_ETCO2 and P_aCO2 in mechanically ventilated patients,^11,12 including severe respiratory failure, obesity, pregnancy, breathing frequency at the time of measurement, and posture,^13–15 and these factors should also be taken into consideration when measuring P_ETCO2 with a capnometer. Third, some patients with hypercarbic respiratory failure underwent repeated measurements, which may be a source of bias. Finally, although we drew from either the right or left antecubital vein and took the tourniquet time as the minimum necessary time, we attempted to reduce the bias as much as possible by having the same researcher draw blood in the same way for all samples. In contrast, P_v̄CO2 is affected by local conditions, such as muscle conditions, and does not represent the partial pressure of central venous carbon dioxide, which may partly explain why P_v̄CO2 had a poor agreement with P_aCO2. P_ETCO2 and P_tcCO2 have shown agreements with P_aCO2 commensurate with the increasing accuracy of the measuring instruments, and these measurements have become clinically applied as a substitute for P_aCO2. However, caution is required for its application, and use without knowing the advantages and disadvantages of this method may result in erroneous results and improper clinical interpretation. Further studies are needed to assess their suitability in different diseases and clinical situations.