Objective: : To demonstrate the technical feasibility of CO2 removal with a commercial hemofilter and a replacement solution containing sodium hydroxide to replace bicarbonate.
Design: : Prospective animal experiment in sheep.
Subjects: : Seven mixed-breed female sheep.
Interventions: : Blood ultrafiltrate containing half of the metabolic production of CO2 was removed with a commercial hemofilter and a replacement solution containing sodium hydroxide was given as replacement. Minute ventilation was lowered to less than half of its baseline value. Ultrafiltration was stopped at 18 hrs, and Paco2 was allowed to increase for about 1 hr; at this time, the sheep were electively killed.
Measurements and main results: : Every 6 hrs, blood was sampled from the carotid artery, the pulmonary artery, and from the extracorporeal perfusion circuit (before the hemofilter, immediately after the hemofilter, and after mixing with the replacement solution). To maintain normocapnia, minute ventilation was reduced from 3.8 +/- 0.1 L/min to 1.9 +/- 0.7 L/min; Paco2 remained near constant during the study. The average blood pH, after mixing with the replacement solution, was 7.64 +/- 0.12. One hour after the ultrafiltration had stopped, Paco2 had increased from 36.7 +/- 4.2 torr (4.9 +/- 0.6 kPa) to 59.6 +/- 9 torr (7.9 +/- 1.2 kPa) (p < .01) and blood pH had decreased from 7.317 +/- 0.041 to 7.151 +/- 0.051 (p < .01).
Conclusion: : CO2 removal with bicarbonate ultrafiltration may be an effective treatment for patients with respiratory failure.