ETCO₂: a predictive tool for excluding metabolic disturbances in nonintubated patients

The purpose of this study is to examine the relation between end-tidal carbon dioxide (ETCO₂) measurement and bicarbonate (HCO₃) level reflecting the patient's metabolic status.

This prospective cross-sectional study has been carried out during a 3-month period in a tertiary care university hospital's emergency department (ED). During the study period, every spontaneously ventilating ED patient requiring arterial blood gas analysis for any medical indication, regardless of presenting symptoms, had a simultaneous ETCO₂ measurement using a Medlab Cap 10 side stream capnograph. The demographics and clinical outcomes of the patients were recorded.

Of 399 eligible patients, 240 with possible metabolic disturbance were enrolled into the study. There was a statistically significant correlation between the value of ETCO₂ and HCO₃ levels (r = 0.506). The mean ET(CO)₂ level was statistically significantly lower in patients who died (26.5 ± 7.2, 95% confidence interval [CI], 24.2-28.6, vs 30 ± 7.5, 95% CI, 29-31; P = .007) and who had low bicarbonate levels (25.7 ± 6.7, 95% CI, 24.3-27.1, vs 31.6 ± 7.1, 95% CI, 30.4-32.8; P = .000). The value of ET(CO)₂ measurement to detect low bicarbonate level was found to be significant. The area under the receiver operating characteristic curve was 0.734, the (+) likelihood ratio for ETCO₂ less than or equal to 25 was 2.7, and the (-) likelihood ratio for ETCO₂ greater than or equal to 36 was 0.05.

ETCO₂ values correlate moderately with HCO₃ levels and thus might predict mortality and metabolic acidosis. Therefore, side stream capnograph can be used as a noninvasive diagnostic tool for ruling out suspected severe metabolic disturbance in the ED.