In vivo carbon dioxide clearance of a low-flow extracorporeal carbon dioxide removal circuit in patients with acute exacerbations of chronic obstructive pulmonary disease

BACKGROUND

Veno-venous extracorporeal carbon dioxide removal allows clearance of CO₂ from the blood and is becoming popular to enhance protective mechanical ventilation and assist in the management of acute exacerbations of chronic obstructive pulmonary disease, including the prevention of intubation. The main factor determining CO₂ transfer across a membrane lung for any given blood flow rate and venous CO₂ content is the sweep gas flow rate. The in vivo characteristics of CO₂ clearance using ultra-low blood flow devices in patients with acute exacerbations of chronic obstructive pulmonary disease has not been previously described.

METHODS

Patients commenced on extracorporeal carbon dioxide removal for acute exacerbations of chronic obstructive pulmonary disease recruited to a randomized controlled trial of non-invasive ventilation versus extracorporeal carbon dioxide removal had pre- and post-membrane circuit gases measured after each increment of sweep gas flow to allow calculation of the transmembrane CO₂ clearance. This was compared with the clearance reported by the device and also corrected to inlet PCO₂ to allow characterization of the CO₂ clearance of the device at different sweep gas flow rates.

RESULTS

CO₂ clearance was calculated using both the transmembrane CO₂ whole-blood content difference and CO₂ clearance reported by the device. The two methods demonstrated a linear relationship and agreement with a bias of 14 mL/minute (SD = ±10) and an R² of 0.92. The membrane CO₂ clearance was non-linear with nearly two thirds of total clearance achieved with sweep gas flow below 2 L/minute (VCO₂ of 40 ± 16.7 mL/minute) and a plateau above 5 L/minute sweep gas flow (VCO₂ 64 ± 1 2.4 mL/minute).

CONCLUSION

The extracorporeal carbon dioxide removal device used in the study provides efficient clearance of CO₂ at low sweep flow rates which then plateaus. This has implications for how the device may be used in clinical practice, particularly during the weaning phase where the final discontinuation of the device may take longer than anticipated. (ClinicalTrials.gov: NCT02086084, registered 13 March 2014, https://clinicaltrials.gov/ct2/show/NCT02086084 ).

In-vitro performance of a low flow extracorporeal carbon dioxide removal circuit

INTRODUCTION

Extracorporeal gas exchange requires the passage of oxygen and carbon dioxide (CO₂) across an artificial membrane. Current European Union regulations do not require the transfer to be assessed in models using clinically relevant haemoglobin, making it difficult for clinicians to understand the CO₂ clearance of a membrane, and how it changes in relation to sweep gas flow through the membrane. The characteristics of membrane CO₂ clearance are described using a single membrane at different sweep gas flows in an in vitro model with clinically relevant haemoglobin concentrations using three separate methods of calculating CO₂ clearance.

METHODS

To define the CO₂ removal characteristics of the extra-corporeal CO₂ removal (ECCO₂R) device, we devised an in-vitro gas exchange circuit formed by a dedicated ECCO₂R circuit (ALung, Pittsburgh, USA) in series with two membrane oxygenators. The system was primed with donated expired human red cells provided by the local blood bank. The experimental set-up allowed constant CO₂ input (via one membrane oxygenator) with variable removal from a portion of the blood in a manner which was analogous to that seen in vivo. Blood gases were measured from different ports in the circuit in order to measure the experimental membrane CO₂ clearance (VCO₂).

RESULTS

Results demonstrate that the relationship between VCO₂ and gas flow at a constant blood flow of 0.4 L/minute with a haemoglobin of 7 g/dL increases sharply from a gas flow of 0 to 2 L/min but plateaus at gas flows >4 L/minute. VCO₂, calculated using three different methods, showed a strong linear correlation with minimal bias.

CONCLUSIONS

The CO₂ clearance of the membrane used in this bench test is non-linear. This has implications for clinical practice, especially during the weaning phase of the device.