Rate-controlled blood withdrawal was used to reduce cardiac preload and consequently stroke volume in patients with normal cardiac function. Twelve patients with asymptomatic hereditary hemochromatosis, undergoing regular phlebotomy therapy, volunteered for the study. An average volume of 375 ml was withdrawn in an average period of 6.4 min. Finger pressure was continuously measured by a Finometer device which includes the Beatscope software for deriving the stroke volume from the blood pressure waveform. Blood withdrawal resulted in reduction of the stroke volume estimates (from 94.0 +/- 5.2 to 80.7 +/- 5.3, P < 0.05) together with a reduced pulse pressure (from 53.0 +/- 3.5 to 47.1 +/- 3.2, P < 0.05). No significant changes in heart rate (75.2 +/- 3.7 versus 78.3 +/- 4.5 beats/min) were observed. Calculated cardiac output was reduced while calculated total peripheral resistance was elevated after blood withdrawal. Beat-to-beat analysis demonstrated a significant linear regression between most of the hemodynamic indices and the volume withdrawn. The highest correlation coefficients were found for the stroke volume (0.88 +/- 0.01, P < 0.001) and the pulse pressure (0.80 +/- 0.04, P < 0.001) corresponding also to the highest slopes for the lines relating these measures to the relative blood volume withdrawn. The non-invasive estimation of finger blood pressure can be used to derive simple on-line indices (pulse pressure, stroke volume using the Modelflow) of cardiac preload, which are of major interest in the monitoring of cardiovascular status.