2. Unpublished study in which flows in a cardiopulmonary bypass circuit were measured by lithium dilution and with an electromagnetic flow probe.


One of the problems in assessing the accuracy of LDCO by comparing it with thermodilution is the scatter of the thermodilution readings. In this study we measured flow in a cardiopulmonary bypass circuit using lithium dilution and compared these values with those obtained simultaneously by an electromagnetic flow probe.


The study was carried out in 16 patients during cardiopulmonary bypass. The venous blood from the patient was syphoned into a reservoir from which it was pumped through a membrane oxygenator by a Biomedicus centrifugal pump. The flow was measured with a TX40 Bioprobe flow transducer. This probe has an accuracy of ± 5%. The flow was measured up to 11 times in each patient by lithium dilution. The lithium chloride was injected just upstream of the membrane oxygenator and blood was sampled through the lithium sensor from the pipe returning blood from the oxygenator to the aorta.


The range of flows was 1.01 - 6.28 l/min. This was rather lower than the range in the initial clinical trial (4.75 - 8.3 l/min) but represented a larger percentage variation (a sixfold change compared with a doubling). The mean of the differences = 0.03 l/min with a standard deviation of the differences = 0.29 l/min.


This result compares favourably with the initial LDCO/thermodilution study in which the bias (thermodilution - LDCO ) was 0.3 l/min (SD = 0.5), and probably reflects the greater accuracy of the electromagnetic flow probe compared with thermodilution. This study was useful in terms of evaluating the LDCO system, but of course there is no need to use lithium for measurement of flows during cardiopulmonary bypass since these are always known either from the r.p.m. of the roller pump or the electromagnetic probe of a centrifugal pump.