Continuous monitoring of ScvO2 by a new fibre-optic technology compared with blood gas oximetry in critically ill patients

a multicentre study

  • Molnar, Zsolt
  • Umgelter, Andreas
  • Toth, Ildiko
  • Livingstone, David
  • Weyland, Andreas
  • Sakka, Samir G.
  • Meier-Hellmann, Andreas

  • 1Department of Anaesthesia and Intensive Therapy, University of Pécs, 13 Ifjusag St., 7624 Pécs, Hungary

    2Department of Medicine II, Technical University of Munich, Munich, Germany

    3Department of Anesthesiology and Intensive Care, Sackler School of Medicine, Tel-Aviv University, Sheba Medical Center, Tel Hashomer, Israel

    4Department of Anesthesiology and Intensive Care Medicine, Klinikum Oldenburg GmbH, Oldenburg, Germany

    5Department of Anesthesiology and Intensive Care Medicine, Medical Center Cologne-Merheim, Cologne, Germany

    6Department of Anesthesiology, Intensive Care Medicine and Pain Management, HELIOS Klinikum Erfurt GmbH, Erfurt, Germany

e-mail: [email protected]

Received: 9 January 2007/Accepted: 17 May 2007/Published online: 19 June 2007

Electronic supplementary material The online version of this article (doi:10.1007/s00134–007–0743–7) contains supplementary material, which is available to authorized users.

Intensive Care Medicine 33(10):p 1767-1770, October 2007. | DOI: 10.1007/s00134-007-0743-7

Abstract

Objective:

The aim of this study was to compare the accuracy of the CeVOX monitor measuring continuous central venous saturation (ScvO2) with laboratory blood gas oximetry under clinical circumstances.

Design:

Prospective, multicentre, observational study.

Setting:

Five adult general intensive care units.

Patients and participants:

Fifty-three critically ill patients.

Interventions:

The fibre-optic probe was inserted into an ordinary central venous catheter's distal lumen. Blood samples were taken from this line via a Y-adapter every 8 h and ScvO2 was measured with a laboratory co-oximeter. Patients were observed for a maximum of 5 days. Results were compared using linear regression and the Bland and Altman plots.

Measurements and results:

The 526 matched pairs of ScvO2 showed a significant correlation between the two methods (r = 0.79, p< 0.001). Bland-Altman plots showed an overall mean bias of -0.3% and moderate agreement (lower and upper levels of agreement: -13.2% and 12.5%). Correlation for the first time point, and for differences between the first two time points for each method revealed good correlation: (n = 53): r = 0.79, p< 0.001; (n = 50): r = 0.58, p< 0.001, respectively.

Conclusion:

These results in a heterogeneous group of critically ill patients show that continuous ScvO2 monitoring by the CeVOX technology yielded results comparable with those obtained by laboratory co-oximetry and therefore can be relied on in everyday clinical practice.

Copyright © 2007 Springer. Part of Springer Science Business Media