Abstract

Objectives: Less-invasive and continuous cardiac output monitors recently have been developed to monitor patient hemodynamics. The aim of this study was to compare the accuracy, precision, and trending ability of noninvasive bioreactance-based Starling SV and miniinvasive pulse-power device LiDCOrapid to bolus thermodilution technique with a pulmonary artery catheter (TDCO) when measuring cardiac index in the setting of cardiac surgery with cardiopulmonary bypass (CPB).

Design: A prospective method-comparison study.

Setting: Oulu University Hospital, Finland.

Participants: Twenty patients undergoing cardiac surgery with CPB.

Interventions: Cardiac index measurements were obtained simultaneously with TDCO intraoperatively and postoperatively, resulting in 498 measurements with Starling SV and 444 with LiDCOrapid.

Measurements and Main Results: The authors used the Bland-Altman method to investigate the agreement between the devices and four-quadrant plots with error grids to assess the trending ability. The agreement between TDCO and Starling SV was qualified with a bias of 0.43 L/min/m² (95% confidence interval [CI], 0.37–0.50), wide limits of agreement (LOA, −1.07 to 1.94 L/min/m²), and a percentage error (PE) of 66.3%. The agreement between TDCO and LiDCOrapid was qualified, with a bias of 0.22 L/min/m² (95% CI 0.16–0.27), wide LOA (−0.93 to 1.43), and a PE of 53.2%. With both devices, trending ability was insufficient.

Conclusions: The reliability of bioreactance-based Starling SV and pulse-power analyzer LiDCOrapid was not interchangeable with TDCO, thus limiting their usefulness in cardiac surgery with CPB.