Stragegies to reduce surface area requirements for carbon dioxide removal for an intravenacaval gas exchange device

Experimental and clinical use of the intravascular oxygenator (IVOX), an intravenacaval gas exchange device, in acute respiratory failure yielded a CO2 transfer of 40-70 ml/min (approximately 30% of adult CO2 production) at normocapnia. Although significant, this rate of CO2 removal is not clinically useful. To maximize CO2 transfer, given the same gas exchange properties and structure design of the IVOX, the authors analyzed the effects of permissive hypercapnia (stepwise increase in arterial blood pCO2 up to 100 mmHg) and active blood mixing (with an intraaortic balloon pump) on different sizes of IVOX (sizes 7, 8, and 9 mm, surface area 0.21, 0.32, and 0.41 m2, respectively) using a previously established ex vivo circuit to model the human vena cava. The CO2 net transfer coefficient (KCO2) was averaged for all sizes and applied to extrapolate the surface area requirements under different pCO2 and with active blood mixing. Results showed that KCO2 increased in a linear relationship with blood flow. Increases in blood flow and blood pCO2 further increase CO2 removal and decrease surface area requirements. For blood flow at 4.0 L/min, the membrane surface area required for 150 ml/min CO2 removal at blood pCO2 of 40 mmHg is 1.76 m2, but can be decreased to 0.47 m2 at blood pCO2 of 80 mmHg, and further to 0.42 m2 with additional active blood mixing. A 0.42 m2 surface area is associated with an O2 transfer of 80 ml/min without and 107 ml/min with active blood mixing. It is concluded that CO2 removal by IVOX alone is limited by insufficient surface area and the resistance in the blood-surface boundary layer. The combination of permissive hypercapnia, adequate blood flow, and active blood mixing can substantially improve CO2 removal and can therefore achieve clinically significant CO2 removal by intravenacaval gas exchange devices during severe respiratory failure.