During partial left heart bypass (LHBP), the flow delivered by an assist device is easy to measure while residual left ventricular function (LVI) seems difficult to assess. In this study, we have attempted to define the separate right and left ventricular function during LHBP. In 6 anesthetized dogs, following thoracotomy and systemic heparinization, aorta and left atrium were cannulated and connected to the servo-controlled roller pump (modified Stockert-System). Following saline infusion (30 ml/kg), LHBP was started and maintained at 50 ml/min/kg throughout 6 h. Standard hemodynamic parameters were continuously monitored. Cardiac output, blood gas analyses, hemoglobin and activated coagulation time were measured at regular intervals. LVI was calculated as the difference between cardiac output and assist-flow rate. Other derived variables were obtained using standard formulas. The Wilcoxon rank-test was used for the statistical analysis. The results, as median and 25th-75th percentile, are summarized in the graphics 1-6. Under the experimental conditions of this study, the flow performance and stroke work of the right ventricle remained unchanged, while the work-unloaded left ventricle maintained only a part of systemic perfusion. Neglecting the physiological shunt and its changes, which influences the difference between the left and right ventricular output, the simple formula to assess LVI during LHBP seems plausible.