An evaluation of thermodilution cardiac output measurement using the Swan-Ganz catheter.

Errors in thermodilution cardiac output measurement were quantitated to determine the order of accuracy of routine measurements performed by unskilled personnel. In vitro and in vivo studies were undertaken to examine factors affecting the volume and temperature of the injectate, catheter thermistor and computer performance, effect of respiration, use of cold (0-4 degrees C) versus ambient temperature (20-25 degrees C) injectate, and the interpretation of measurements. Ambient temperature injectate incurred unacceptably large errors; cold injectate (injections were timed with respiration) produced variations in performance by equipment and personnel which accounted for only 2% of the variation between successive measurements. Real changes in cardiac output and inherent variability of the downslope of the thermal curve, necessitating an empirically based calculation, account for up to 10% variation between successive measurements. When cold injectate was used, and the average of three corrected measurements taken, thermodilution cardiac output measurements were within 10% of a simultaneous dye dilution measurement.