From the literature and our own comparative studies using different pulse contour cardiac output systems, we concluded that the accuracy (bias), precision (SD) as well as the tracking of changes in cardiac output by the PiCCO system is inferior to most of its competitors. During our use of the PiCCO system, several technical and patient-related limitations were uncovered by coincidence. The technical limitations were related to i) incorrect detection of heart beats, ii) incorrect detection of ejection phase, iii) no detection of damped arterial pressure tracings, all leading to incorrect computations of cardiac output. Patient-related problems were found during severe episodes of bleeding and cardio-pulmonary anatomical abnormalities. In most cardiothoracic patients, SVV or PPV to monitor preload dependency was only useful for a short time as most patients were weaned from the ventilator shortly after arrival in the ICU. In patients who are candidates for a heart assist device (intra-aortic balloon pump) a femoral arterial puncture for application of the PiCCO device is contra-indicated. We experienced, consistent with the literature, that measurement of GEDVI and ITBVI in cardiomyoplasty patients is irrelevant. Furthermore we have, based on theory and observation, the impression that the precision of these variables is dependent on SVV. From the foregoing we consider that the PiCCO system is of limited value in monitoring cardiothoracic patients. Copyright © 2008, Nederlandse Vereniging voor Intensive Care. All Rights Reserved.