Cardiac output derived from arterial pressure waveform analysis without calibration vs. thermodilution in septic shock: evolving accuracy of software versions

Abstract

Background and objective: We studied the evolution of software in the accuracy of the FloTrac/Vigileo system to measure cardiac output less invasively from arterial pressure waveform analysis without calibration, in comparison with pulmonary artery catheter-derived thermodilution measurements, in patients with septic shock and presumed alterations in vascular tone.

Methods: Nine patients who received a pulmonary artery catheter and were on mechanical ventilation and in sinus rhythm were monitored by the FloTrac/Vigileo. Paired cardiac output measurements by both techniques were analysed for 86 measurements in four patients using the 1.07 software version and 73 measurements in five subsequent patients using the later 1.10 version.

Results: For the 1.07 version, bias was -1.6 L min, precision 1.6 L min, limits of agreement -4.8-1.5 L min and error 48%. Measurements correlated at partial r equal to 0.32 (P = 0.003). For the 1.10 version, bias was -1.2 L min, precision 1.1 L min, limits of agreement -3.5-1.0 L min and error 32%. Measurements correlated at partial r equal to 0.90 (P < 0.001 vs. version 1.07). Differences were inversely related to mean cardiac output (P < 0.001, generalized estimating equations), particularly for software version 1.07 vs. 1.10 (P = 0.017, generalized estimating equation). Changes in thermodilution cardiac output over the course of time were also better tracked by the FloTrac/Vigileo when applying the latest software (P < 0.001, generalized estimating equation).

Conclusions: Evolving software versions are thus better able to account for the effect of vascular tone on cardiac output measurements by less invasive waveform analyses without calibration (FloTrac/Vigileo), so that the latter may become useful in the haemodynamic monitoring of septic shock.