Background: Intravenous sufentanil-midazolam and inhalational sevoflurane are widely used for anesthetic induction in children undergoing cardiac surgery. However, knowledge about their effects on hemodynamics and cardiac efficiency remains limited due largely to the lack of direct monitoring method. We used a minimally invasive technique, the pressure recording analytical method (PRAM), to directly monitor hemodynamics and cardiac efficiency, and compared the effects of the two anesthetic regimens in children undergoing ventricular septal defect repair. Methods: Forty-four children (2.3 +/- 0.9 years) were randomized into two groups to receive either intravenous sufentanil (1 mu g/kg) and midazolam (0.2 mg/kg) (Group SM) or 2.0 MAC sevoflurane (Group S) to complete induction after sedation was obtained with 2.0 MAC sevoflurane. Systemic hemodynamic data recorded by PRAM included heart rate (HR), systolic (SBP) and mean (MBP) blood pressure, stroke volume index (SVI), cardiac index (CI), systemic vascular resistance index (SVRI), the maximal slope of systolic upstroke (dp/dt(max)) and cardiac cycle efficiency (CCE) after sedation obtained; 1, 2, and 5 minutes after induction achieved; 1, 2, 5, and 10 minutes after intubation. Results: HR and SVRI showed a decrease in Group SM but an increase in Group S (P-time*group < 0.0001) in the study period. SVI and CCE showed an increase in Group SM but a decrease in Group S (P-time*group < 0.0001). SBP, MBP, and CI were related to time after polynomial transformation, showing an increase after intubation in Group SM but a decrease in Group S (P-time2*group < 0.0001). Conclusion: PRAM provides meaningful and direct monitoring of hemodynamic parameters as well as cardiac efficiency during the dynamic period of anesthetic induction in children undergoing cardiac surgery. As compared to inhalational sevoflurane, intravenous sufentanil-midazolam exerts more favorable effects on systemic hemodynamics and cardiac efficiency during anesthetic induction in this group of patients.