BACKGROUND: It is essential to control hemodynamics in cardiac surgery. Patients are often monitored extensively in order to optimize hemodynamic performance. However, pre-operative values are normally unknown. Furthermore, hemodynamic goals may seem arbitrary and the lack of an evidence-based consensus may lead to both under- and over-treatment. The aim of this study was to evaluate the variables most commonly used for hemodynamic guidance in the post-operative period. METHODS: Ten patients scheduled for elective cardiac surgery were followed with invasive hemodynamic monitoring the night before surgery. All data were recorded automatically and electronically. RESULTS: We found considerable inter-patient differences and intra-patient variation. The greatest intra-patient variation was found in the cardiac index (CI), ranging from 1.9 to 5.3 l/min/m(2). Four patients had periodic CI <2.4 l/min/m(2). Eight patients showed SpO2 values < or =92, four of them in more than 15% of the observations. Six patients had an SvO2 <70% in more than 40% of the observations and two an SvO2 < 64% in more than 20% of the observations. CONCLUSIONS: This study is unique because hemodynamic reference data in cardiac surgery patients have not been published previously. The intra-patient variations were unexpectedly high in most hemodynamic variables and demonstrate the difficulties of using hemodynamic parameters as a guidance for treatment and indicate that goal-oriented therapy using currently accepted values may result in over-treatment in some patients.