Linton R, Turtle M, Band D, O'Brien T, Jonas M. In vitro evaluation of a new lithium dilution method of measuring cardiac output and shunt fraction in patients undergoing venovenous extracorporeal membrane oxygenation. Critical Care Medicine 1998; 26: 174-177.

The model consisted of an ECMO circuit set up in parallel with a 'patient' circuit, which consisted of a 50 litre reservoir of saline circulated at 3 l/min through tubing by a pump which represented the heart; the 'cardiac output' was therefore constant throughout these experiments. A second pump in the ECMO circuit drew saline from the 'patient' circuit and passed it through a membrane oxygenator. The flow from the membrane oxygenator either returned directly to the 'patient' circuit or was diverted via a third pump back into the ECMO circuit, so producing a shunt.

By adjusting the flows of the second (ECMO) and third (shunt) pumps shunt fractions of 12, 25 and 50% were produced at each of three ECMO flow rates (0.5, 1 and 2 l/min). Lithium chloride (0.15 mmol) was injected just downstream of the membrane oxygenator and its concentration-time curves recorded simultaneously in the flow returning to the saline reservoir and in the flow just upstream of the membrane oxygenator using lithium selective electrodes. Analysis of these curves allowed shunt flow and 'cardiac output' to be calculated and compared with the flows delivered by the pumps.

Nine pairs of curves were recorded, one pair for each combination of ECMO and shunt flows. Mean 'cardiac output' derived from the lithium dilution curves was 2.98 ± 0.18 (± SD) l/min, compared with a delivered pump flow of 3 l/min. Measured shunt flow = 0.008 + 1.09 x actual shunt flow (R = 0.997). The figure shows examples of two sets of curves.

This method has now been used successfully in a patient [Linton RAF, Turtle M, Band DM, O'Brien TK, Jonas MM, Linton NWF. A new technique for measuring cardiac output and shunt fraction during venovenous extracorporeal membrane oxygenation. Perfusion 1999; 14: 43-48].