How does variation of the bag volume in inert gas rebreathing cardiac output measurements influence the reproducibility in patients with pulmonary diseases

Background: Cardiac output (CO) is an important hemodynamic parameter, however its determination is difficult in daily clinical routine. Non-invasive inert gas rebreathing (IGR) showed promising results in recent investigations with the volume of the rebreathing bag (V b ) being an important factor. The aim of our study therefore was to evaluate the influence of different V b on the reproducibility of IGR. Methods: The collective consisted of 45 patients (age 26 to 88 years). The CO was determined in patients with obstruction (group A), restriction (group B) and pulmonary healthy controls (group C). For V b of 2200 ml, 1700 ml and 1200 ml two repeated measurements were taken each. The determination of lung function was performed using bodyplethysmography. Results: Pulmonary obstruction was diagnosed in 12 patients (FEV 1 52±21%) and restriction in 11 patients (VC 61±16%). The mean CO did neither differ between the groups (p=0.1) nor for the different V b (p=0.2). The mean bias between the repeated measurements was 0.2±0.9l/min for V b =2200ml, 0±0.7l/min for V b =1700ml and 0.3±0.7l/min for V b =1200ml. There was no statistically significant difference between the groups for the different V b =2200ml (p=0.7), 1700ml (p=0.4) und 1200ml (p=0.2). Conclusion: The reproducibility of IGR is not negatively affected by V b , so that it can be varied between 1200 und 2200 ml. This is especially important when V b has to be reduced due to incomplete inspiration. The trend to a worse reproducibility at extreme volumes at rest should be further investigated. For now, measurements should only be compared directly when identical V b were used.