The gas transfer characteristics and gaseous microemboli (GME) production of five different bubble oxygenators (Polystan Venotherm, Harvey H-1700, Bentley BIO-10, Gambro 10 and Shiley S-100A HED) and one membrane oxygenator (Cobe CML) have been assessed during standardized clinical perfusion for open-heart surgery in 60 adult patients. The perfusionist attempted to maintain physiological levels of PaCO 2 (5 ± 1 kPa) and PaO2 (12 ± 2 kPa). Only 3% of blood gas values were within the normal range in the Bentley BIO-10 group compared with 17% for the Gambro 10, 20% for the Shiley S-100A HED, 31% for the Polystan Venotherm, 33% for the Cobe CML and 36% for the Harvey H-1700. The number of GME detected in the arterial line was significantly lower in the Cobe CML membrane oxygenator when compared with any of the five different bubble oxygenators (p < 0·001). The Polystan Venotherm released significantly less GME (p < 0·02) than the other oxygenators and the Shiley S-100A HED released significantly more GME (p < 0·02) than the other oxygenators except the Gambro 10. Low gas-blood flow ratios were not necessarily associated with low GME levels and inadequate oxygenation. This study provides meaningful data on which to base the choice of oxygenator, for clinical perfusions.

During 17 cardiopulmonary bypass (CPB) operations, arterial PO2 measurements were taken at 5 min intervals using four different types of PO2 sensor; three in-line (CardioMet-1000, Polytrode, and Gas-Stat) and one blood gas analyser (IL-1302). Comparisons between sensors were made relative to the CardioMet-1000. The three in-line sensors showed close agreement. The mean and standard deviation of the difference between the Gas-Stat and the CardioMet-1000 was 0.7 +/- 4.7 kPa, and between the Polytrode and the CardioMet-1000 was -1.4 +/- 4.0 kPa. Agreement between the IL-1302 blood gas analyser and CardioMet-1000 was not as good, the difference being -7.0 +/- 8.6 kPa.

The haemocompatibility of five different bubble oxygenators (Polystan venotherm, Harvey H-1700, Bentley BIO-10, Gambro 10 and Shiley S-100A HED) and one membrane oxygenator (Cobe CML) have been evaluated during standardized clinical perfusion for open-heart surgery in 48 adult patients. Control of arterial PO2 and PCO2 was an important feature of the evaluation protocol. Over the period of cardiopulmonary bypass (CPB) there was a marked difference in the mean percentage reduction in platelet count in the different oxygenator groups. Only 1% reduction in platelet count occurred with the Cobe CML membrane oxygenator group compared with, in the bubble oxygenator groups, 7% for the Gambro 10, 16% for the Harvey H-1700, 19% for the Shiley S-100A HED, 24% for the Bentley BIO-10 and 31% for the Polystan venotherm. The post bypass platelet count was significantly lower than the prebypass value in all oxygenator groups ( p < 0.05) except the Cobe CML and Gambro 10. The two oxygenator groups with the largest percentage reduction in platelet count (Polystan venotherm and Bentley BIO-1 0) demonstrated a significant reduction ( p < 0.05) in platelet aggregation over the period of bypass. Platelet depletion in the Harvey H-1700. Shiley S-100A HED, Bentley BIO-10 and Polystan venotherm oxygenators was associated with a significant fall ( p < 0.05) in mean platelet volume during the first 35 minutes of CPB due to the removal from the circulation of large, young, functionally more active platelets. Erythrocyte damage was minimal in all oxygenator groups and only a minor degree of leucopenia could be demonstrated during the first five minutes of CPB. Cardiotomy suction was not associated with significant changes in platelet numbers or platelet aggregation. When selecting the oxygenator for use in patients undergoing open-heart surgery, gas transfer characteristics and GME production together with the superior preservation of platelet numbers and function in the membrane oxygenator group and variable degree of platelet depletion and reduction in platelet aggregability demonstrated in the five bubble oxygenator groups, must be taken into account.