In vivo gas transfer performance of the intravascular oxygenator in acute respiratory failure

The intravascular oxygenator (IVOX) has undergone both animal and clinical trials. Data from the animal studies have demonstrated that the device is capable of transferring up to approximately 100 ml/min of oxygen and carbon dioxide. Initial data from the human trials suggest that gas transfer, although approaching these levels, varied widely in patients with respiratory failure. We studied the factors affecting gas exchange in 26 patients with severe acute respiratory failure who underwent intravenacaval support of gas exchange with IVOX. The patients underwent monitoring of IVOX gas transfer rates, hemodynamics, blood gases, and ventilation parameters at scheduled intervals following device insertion. All devices functioned following implantation. The mean value for O2 transfer was 64 +/- 21 SD ml/min (range 15-114 ml/min) and for CO2 transfer was 48 +/- 17 ml.min-1 (range 14-112 ml/min). CO2 transfer correlated positively with device surface area, cardiac output, and mixed venous Pco2 and negatively with duration of implantation. O2 transfer did not correlate with any patient factors probably due to error inherent in the measurement of this variable. Independent measurements of IVOX gas transfer by respiratory gas exchange in a subset of patients with normal values of mixed venous Pco2 were in good agreement with the routine measurements and indicated that the device provided up to 26% of gas exchange requirements in this subset. We conclude that IVOX transfers clinically useful amounts of oxygen and carbon dioxide in vivo. Factors that influence gas transfer include device surface area, PvCO2, cardiac output, and duration of implantation. Optimization of these factors (such as with permissive hypercapnea) could result in enhanced performance in vivo.