Comparison of three commercially available hollow fiber oxygenators: gas transfer performance and biocompatibility

The new generation of oxygenators have improved blood flow pathways that enable reduction in priming volume and, thus, hemodilution during cardiopulmonary bypass (CPB). We evaluated three oxygenators and two sizes of venous reservoirs in relation to priming volume, gas transfer, and blood activation. To compare priming volume, gas transfer, and biocompatibility of three hollow fiber oxygenators and two different size venous reservoirs, 60 patients were randomly allocated in groups to undergo cardiopulmonary bypass. In each group, an oxygenator with a different surface area and priming volume was used: 1.8 m2 and 220 ml (group 1, n = 23), 2.2 m2 and 290 ml (group 2, n = 20), and 2.5 m2 and 270 ml (group 3, n = 17). In groups 1 and 3, a large soft shell (1900 ml) venous reservoir was used, whereas in group 2, a smaller soft shell (600 ml) venous reservoir was used. Gas transfer was assessed by calculating the oxygen transfer rate for each group and per square meter for each oxygenator group. Partial arterial oxygen pressure (paO2) and partial arterial carbon dioxide pressure (paCO2) between the groups were assessed with forward stepwise regression analysis. Biocompatibility was evaluated through measurement of platelet numbers, complement activation products (C3b/c), coagulation (thrombin anti-thrombin III complex), and fibrinolysis (plasmin anti-plasmin complex). No differences were found in oxygen transfer rate per group. However, when correcting the oxygen transfer rate for surface area, group 1 demonstrated a higher oxygen transfer rate compared with group 2 (p < 0.05) at an FiO2 of 40 and 60% and compared with group 3 at an FiO2 of 60 and 70%. The regression analysis showed that the average arterial PO2 was the highest in group 3, i.e., 79.2 mm Hg higher than in group 1 (p < 0.001) and 73.5 mm Hg higher than in group 2 (p < 0.001). Group 3 also had the lowest average arterial pCO2, 0.57 mm Hg lower than in group 1 (p = 0.004) and 0.81 mm Hg lower than in group 2 (p < 0.001). During CPB, platelet numbers decreased significantly in all groups (p < 0.001), without differences between the groups. C3b/c levels increased in all groups during CPB. At cessation of CPB the C3b/c level in group 2 (398 nmol/L(-1)) was significantly higher compared to group 1(251 nmol/L(-1); p < 0.05) and group 3 (303 nmol/L(-1); p < 0.05). Thrombin anti-thrombin III complexes and plasmin anti-plasmin complex complexes increased during CPB to significantly high levels at cessation of CPB, but there were no differences between the groups. The oxygenator with the smallest surface area and lowest priming volume (group 1) had the highest oxygen transfer rate per square meter and showed the least blood damage, as depicted by complement activation. The oxygenator with the largest blood contact surface area and improved geometric configuration (group 3) showed the lowest oxygen transfer rate per square meter. However, this oxygenator elevated oxygen partial pressure the most and reduced carbon dioxide partial pressure the most. In group 2, where a smaller venous reservoir was used, the highest blood activation was observed.