Performance of a minimally invasive uncalibrated cardiac output monitoring system (Flotrac/Vigileo) in haemodynamically unstable patients.

Background: Early haemodynamic assessment is of particular importance in the evaluation of haemodynamically compromised patients, but is often precluded by the invasiveness and complexity of the established cardiac output (CO) monitoring techniques. The FloTrac/Vigileo system allows minimally invasive CO determination based on the arterial pressure waveform derived from any standard arterial catheter, and the algorithm underlying CO calculation was recently modified to allow a more precise estimate of aortic compliance.
Methods: Using the new software, we studied 25 haemodynamically unstable patients who had a radial artery catheter and underwent invasive haemodynamic monitoring with the PiCCO system. PiCCO-derived transpulmonary thermodilution and pulse contour CO (reference-CO) were compared with the CO values obtained with the FloTrac/Vigileo system (AP-CO). Reported CO values are indexed to body surface area. Agreement between reference-CO and AP-CO recorded during routine clinical care was assessed using Bland-Altman statistics.
Results: Overall bias between the reference-CO and the AP-CO (n=324) was 0.68 litre min(-1) m(-2) with a high percentage error of +/- 58.8% (95% limits of agreement +/- 1.94 l min(-1) m(-2)). There was a significant difference (P<0.001) between the radial and the femoral mean arterial pressures, and bias was significantly larger for a mean pressure difference of >5 mm Hg (0.93 vs 0.57 litre min(-1) m(-2), P=0.032). No connection was found between the norepinephrine dose and the CO agreement.
Conclusions: Despite the updated algorithm, AP-CO still showed a limited agreement with the reference-CO and systematically underestimated the CO so that the method is not suitable to replace invasive CO monitoring at present.