Abstract 16248: Comparison of a Non-Invasive Respiratory Volume Monitor to End-Tidal CO2 in Measuring Changes in Minute Ventilation

Abstract

Introduction: Monitoring respiratory performance is a challenge. Use of secondary indicators like oxygen saturation (SpO 2 ) and capnography (EtCO 2 ) have introduced delay in diagnosis and can produce false alarms. EtCO 2 in non-intubated patients has been problematic. We compared performance of an EtCO 2 monitor to a respiratory volume monitor (RVM), which provides accurate (error <10%) measurements of minute ventilation (MV), tidal volume & respiratory rate, to assess changes in respiratory status in non-intubated subjects. Methods: Continuous RVM (ExSpiron, Respiratory Motion, Waltham, MA) and capnography data (Capnostream 20, SmartCapnoLine Plus & Filterline Set, Covidien, Mansfield, MA) were collected from 39 subjects. The relationship between EtCO 2 & MV for each patient was quantified & EtCO 2 sensitivity to MV changes and mean EtCO 2 values were compared across the two EtCO 2 sampling techniques (paired t-test). Results: A strong negative correlation between MV changes and EtCO 2 changes was found (similar between nasal sampling (NS) and in-line sampling (IS) ( -0.6 ±0.1 v -0.7±0.1 mmHg/(L/min), p>0.1, Fig 1A)). IS readings were higher than NS for each patient. In the same patients (normal breathing) the average IS EtCO 2 was 36.7 ±0.7 vs 33.7±0.6 mmHg for NS (p<0.01) despite larger MV (8.74 ±0.5 vs 7.0±0.5 L/min). Similar disparity was present during hypo & hyperventilation (Fig 1B&C). Regardless of sampling methodology, EtCO 2 displayed poor resolution in response to large swings in MV (13 & 14 mmHg in response to 17 & 21 fold increase in MV; IS & NS, respectively). Conclusions: Under controlled experimental conditions & with an in-line sensor capable of capturing 100% of exhaled air, EtCO 2 did not report large swings in MV with fidelity. Commonly used nasal sampling introduced a systematic bias that could further delay informing clinicians of dangerously low MV. RVM provides a direct measure of MV, eliminating the lag between a change in MV and secondary measurements.