Clinical evaluation of nine hollow-fibre membrane oxygenators.

In a comparative study we investigated the performance characteristics of nine hollow-fibre oxygenators. In a clinical setting, 10 units of each type of oxygenator were tested for oxygen exchange, transoxygenator pressure drop, heat exchanger performance and blood trauma. The oxygenators included are Maxima PRF Plus, Affinity, Forte, Affinity NT, Quantum, Optima, Capiox 1.8, Hilite and Quadrox. Ninety patients scheduled for elective coronary bypass surgery were enrolled in the study. Cardiotomy suction blood was not allowed to mix with the circulating blood. Samples were taken preoperatively (t1), immediately after cooling and crossclamping of the aorta (t2), during rewarming (t3), and postoperatively (t4). The results showed that under clinical conditions all of the tested oxygenators are well capable of meeting the normal and above-normal oxygen needs. The Quantum and Affinity NT are the best performing oxygenators with respect to oxygen transfer. The Quantum oxygenator with a small membrane surface area of 1.9 m2 has also the highest oxygen transfer per square metre. A direct relationship between shunt fraction and blood flow could not be established in all oxygenators. The lowest shunt fraction at 4 litres blood flow was seen in the Maxima, Quantum and Affinity NT groups (15%). At near physiological PaO2-values, we noticed a negative influence on the shunt fraction. Large differences exist in pressure drop between the oxygenators under standardized conditions with circulating priming, as well as during bypass. The lowest pressure drop was noticed in the Quadrox and Affinity NT, the highest in the Quantum, Capiox and Optima. However, we did not measure any differences in pressure drop after cooling and during re-warming. Differences exist between the heat exchanger performances. The Quadrox oxygenator is the most efficient, while the Maxima has the smallest performance factor. It was not possible to calculate heat exchanger performance for the Hilite due to incorrect temperature measurement. The high pressure gradients in some of the devices did not result in higher haemolysis. Plasma free haemoglobin values were low in each oxygenator. There are no differences in platelet drop postoperatively. The influence on blood trauma of the higher pressure drop in some of the tested devices, in combination with the higher centrifugal pump revolutions needed to overcome this gradient, has to be studied with longer perfusion times.