[Accuracy of measurement and overestimation of CO2 of two capnometers intended for potential use in emergency medicine].

UNLABELLED

Capnometry, the noninvasive measurement of end-expiratory CO2 concentration (cCO2, vol%) or calculation of its respective partial pressure (pCO2; mmHg) is an established method. However, for prehospital settings, capnometry is still used very restrictively, mainly owing to the respective devices used. The prerequisite for their use is sufficient accuracy (+/-2 mmHg) and easy handling. Two special capnometers (STAT CAP. Nellcor: mainstream, semiquantitative estimation; Capnocheck 8200, BCI: sidestream, quantitative measurement, numeric display), developed recently for potential use in emergency medicine, are said to fit these criteria. Therefore, the objective of the present investigation was to assess the accuracy and precision of both devices, comparing methods under standardized in vitro (reference gases) and in vivo (intubated and ventilated patients) conditions.

METHODS

Both devices ("STAT CAP": pCO2 range, light bars; "Capnocheck 8200") were evaluated regarding the accuracy of pCO2 (Capnocheck) and the precision of the CO2 range (STAT CAP). Tests were performed with four dry gas mixtures (STPD) of defined composition and during ventilation of 20 intubated patients (BTPS). All measurements were compared with the alveolar gas monitor "AGM 1304" (Brüel & Kjaer, Denmark) as a reference method with a proven +/- 1 mmHg accuracy of pCO2 measurement.

RESULTS

The "Capnocheck" (BCI) presented an accuracy of the pregiven pCO2 of 0.7-1.4 mmHg (dry gas mixtures, STPD) and an overestimation of 0.2 +/- 4.1 mmHg (BTPS) during ventilation with pure oxygen; inaccuracy during ventilation with 70% N2O in O2 proved to be + 1.2 +/- 1.7 mmHg (BTPS). Nellcor's "STAT CAP" failed to reach the target value in 10% of analyses, as shown by the respective segment bar of the display.

CONCLUSION

Evaluation of the accuracy of capnometers must focus on the necessary pH2O correction and the possible effects exercised by O2 (and N2O) as well as the possible dependence on barometric pressure (if pCO2, mmHg, is the desired value). The "Capnocheck" showed an accuracy of more than 2 mmHg in dry gas mixtures as well as in humidified air. Concerning the practical use during constant artificial ventilation, the digital display and accuracy of the sidestream capnometer allow for reliable conclusions on patients' ventilation and circulation (CO2 elimination). The 90% accuracy of the segment bar display of Nellcor's "STAT CAP", per se covering only a rather broad range of 20 mmHg, obviously does not provide more than a rough overview. Therefore, the STAT CAP cannot be recommended for prehospital capnometry in the field. However, both the accuracy of the BCI capnometer (Capnocheck) and its numeric display and easy handling strongly recommend this device also for clinical use.