Monitoring of carbon dioxide in ventilated neonates: a prospective observational study

Objective To assess the reliability, accuracy and precision of distal end-tidal capnography (detCO 2 ) in neonates compared with transcutaneous (tcCO 2 ) carbon dioxide measurements. Design Observational, prospective clinical study. Setting Neonatal intensive care unit at Medical University of Vienna. Participants Conventionally ventilated neonates with a body weight between 1000 g and 3000 g. Intervention End-tidal partial pressure of CO 2 was measured in distal position using the separate lumen of a double-lumen endotracheal tube connected to an external side-stream capnometer. Three consecutive detCO 2 and tcCO 2 values were recorded simultaneously and compared with simultaneous arterialised partial pressure of CO 2 (paCO 2 ) measurements in each patient. Main outcome measures Reliability, accuracy and precision of detCO 2 and tcCO 2 measurements compared with paCO 2 in neonates. Results Twenty-five neonates were included with a median (range) weight at enrolment of 1410 (1010–2980) g, from which 81 simultaneous measurements of detCO 2 , tcCO 2 and paCO 2 were obtained. The mean (SD) of paCO 2 , detCO 2 and tcCO 2 was 45.0 (8.6) mmHg, 42.4 (8.4) mmHg and 50.4 (20.4) mmHg, respectively. The intraclass correlation between paCO 2 and detCO 2 and between paCO 2 and tcCO 2 reached 0.80 (95% CI 0.71 to 0.87, p<0.001) and 0.59 (95% CI 0.43 to 0.72, p<0.001), respectively. In the Bland-Altman analysis, bias and precision of detCO 2 with respect to paCO 2 amounted to −2.68 mmHg and 10.62 mmHg (95% CI 8.49 to 14.51), respectively. Bias and precision of tcCO 2 with respect to paCO 2 amounted to 5.39 mmHg and 17.22 mmHg (95% CI 13.21 to 23.34), respectively. Conclusion DetCO 2 had better reliability, accuracy and precision with paCO 2 than tcCO 2 in ventilated neonates without severe lung diseas. Trial registration number NCT03758313 .