Cardiac output measurement by pulse dye densitometry: a comparison with the Fick's principle and thermodilution method.

OBJECTIVE: To evaluate the agreement between cardiac output (CO) measurements obtained by a new dye dilution technique using pulse dye densitometry (PDD) and thermodilution (TD) and the direct Fick method (F). DESIGN AND
SETTING: Prospective clinical study in a university hospital, cardiac surgery intensive care unit. PATIENTS: Fifty-eight cardiac surgery patients after admission to the intensive care unit (six were excluded due to a low pulse signal quality using the PDD method). MEASUREMENTS AND
RESULTS: Mean CO was 5.3+/-1.8 l/min for PDD, 5.7+/-1.68 l/min for TD, and 6.16+/-1.66 l/min for F. There was a good correlation between PDD and TD ( r(2)=0.93) and between PDD and F ( r(2)=0.77). Bias and precision between PDD and TD were -0.39+/-0.5 l/min and -0.69+/-0.85 l/min between PDD and F. In general, PDD determined lower CO values than TD and F. Especially in patients with CO below 5 l/min PDD underestimated CO in comparison to TD and F (bias and precision: -0.51+/-0.40 l/min and -0.83+/-1.0 l/min).
CONCLUSION: Comparison between PDD and TD showed good agreement for the normal to high CO range. However, agreement was poor in patients with low CO. In the latter patient group PDD showed relevant underestimation of CO compared to TD and F. Due to these limitations PDD cannot entirely replace the pulmonary artery catheter for CO determination.