Reproducibility of the respiratory dead space measurements in mechanically ventilated children using the CO2SMO monitor

Objectives. To assess the reproducibility of respiratory dead space measurements in ventilated children. Design. Prospective study. Setting. University pediatric intensive care unit. Patients. Thirty-two mechanically ventilated children (0.13-15.4 years) who were clinically stable. Methods. The single-breath CO2 test (SBT-CO2) was recorded using the CO2SMO Plus from the mean of 30 ventilatory cycles during 1 h (at T0, T15, T30, T45, and T60). Airway dead space was determined automatically (Novametrix Medical Systems, USA), and manually by Bohr- Enghoff equations using data obtained by SBT-CO2. At the end of the study period, arterial blood gas was sampled in order to calculate alveolar and physiologic dead space. Intrasubject reproducibility of measurements was evaluated by the intraclass correlation coefficient. Two-way analysis of variance was used to evaluate the relationships between time and measurements. The two methods for calculating airway dead space were compared by using two-tailed Student's t-test and Bland-Altman analysis. Results. Airway dead space measurement had a good reproducibility during the 1-h period, whatever the method used (intraclass correlation coefficient: 0.84 to 0.87). No significant difference was observed with time. Airway dead space values from the SBT-CO2 method were smaller than those from Bohr-Enghoff equations. Physiologic dead space values from the SBT-CO2 method were similar to those from Bohr-Enghoff equations. Conclusion. The measurement of airway dead space by the CO2SMO Plus was reproducible over a 1-h period in children requiring mechanical ventilation, provided ventilatory parameters were constant throughout the study. SBT-CO2 analysis may provide a bedside non-invasive monitoring of volumetric capnography.