Clinical usefulness of end-tidal CO2 measured using a portable capnometer in patients with respiratory disease.

Abstract

This study aimed to evaluate the correlation and agreement between end-tidal CO2 (EtCO2 ) measured with the novel portable capnometer (CapnoEye®) and partial pressure of arterial carbon dioxide (PaCO2 ) levels in patients with respiratory diseases and to compare the efficacy of EtCO2 and PvCO2 in predicting PaCO2 levels. We analyzed the correlation and the agreement between EtCO2 and PaCO2 and between PvCO2 and PaCO2 using Pearson's moment correlation coefficient in patients with type 1 and type 2 respiratory failure and both groups overall. A total of 100 samples were included that comprised 67 men (67%). The mean age of the subjects was 77 ± 13 years. Chronic obstructive pulmonary disease (COPD) (43%) was the most common disease. There was a high correlation between EtCO2 and PaCO2 (r = 0.88; p < 0.0001). Sixty-six PvCO2 samples were obtained, and there was a high correlation between PvCO2 and PaCO2 (r = 0.81; p < 0.0001). Regarding type 2 respiratory failure, there was a high correlation between EtCO2 and PaCO2 (r = 0.81). The Bland-Altman analysis between PaCO2 and EtCO2 revealed a bias of 5.7 mmHg, with limits of agreement ranging from -5.1 mmHg to 16.5 mmHg. In contrast, the analysis between PaCO2 and PvCO2 revealed a bias of -6.8 mmHg, and the limits of agreement ranged from -22.13 mmHg to 8.53 mmHg. EtCO2 measured by CapnoEye® was significantly correlated to PaCO2 levels in patients with respiratory diseases. Moreover, CapnoEye® may be more useful for predicting hypercapnia conditions in which respiratory diseases are compared with measure PvCO2 .