Thermodilution cardiac output performed via a pulmonary artery catheter ( Royston and Ee.eley..,.199.3) is the standard technique employed by most ICUs. It is a variant of the dye dilution technique, utilizing cold to create a thermal deficit. The Stewart-Hamilton equation links cardiac output to changes in temperature as follows:
where V is the volume injected, TB and T are the blood and injectate temperatures respectively, K-, and K2 are computational constants, and TB(t)dt is the change in blood temperature as a function of time.
In practice, a 5- or 10-ml bolus of 5 per cent glucose is injected smoothly and rapidly (< 4 s) through the proximal (injectate) lumen of the catheter into the right atrium. A thermistor at the catheter tip detects the temperature change which is computed to produce a curve, the area under which is inversely proportional to the cardiac output. Both technical and physiological errors are commonplace and frequently unrecognized ( Gnaegiefal 1997). The thermistor must be freely within the pulmonary arterial lumen rather than abutting the wall. All circuit connections must be leak free to prevent loss of injectate, as volumes injected must be accurate. The injectate may occasionally reflux up the introducer sheath when the proximal port is contained within the sheath; alternatively, rapid infusions through the side port of the sheath may affect the readings. The computer settings must be correct. The injectate may be cooled in ice to increase the temperature difference and the signal-to-noise ratio; however, a 10-ml room temperature injectate is usually adequate. Large variations (up to 50 per cent) in cardiac output can be obtained by performing the measurements at end-inspiration and end-expiration. To improve reproducibility, it is currently recommended that the injections are spaced evenly across the respiratory cycle.
Anatomical or physiological conditions that preclude accurate measurement include intracardiac shunts, tricuspid regurgitation, and cardiac arrhythmias. Notwithstanding errors in the technique itself introduced by inappropriate mixing of the injectate or bidirectional flow of the indicator solution, left heart output may also vary markedly from right heart output in many of these situations.
Modified pulmonary artery catheters can provide continuous monitoring of mixed venous saturation via a fiber-optic probe sited at the catheter tip (e.g. Oximetrix, Abbott Laboratories), right ventricular ejection fraction measurement, or 'continuous' cardiac output monitoring (e.g. Intellicath, Baxter-Edwards). The last technique utilizes pulses of heat emitted every 30 to 60 s from a thermal filament located 14 to 25 cm from the catheter tip. The resulting rise in blood temperature is detected by a thermistor located 4 cm from the tip. A formula cross-correlates the thermistor temperature with the thermal energy input sequence. Cardiac output is computed from the area under the 'washout' curve and a modified Hamilton-Stewart equation. This technique is not strictly continuous, as updated values reflect an average of the previous 3 to 6 min. Correlation against the bolus thermodilution technique has not always been close.
Was this article helpful?