The present study aimed to assess whether impedance cardiography (IC) can correctly identify the optimal interventricular (VV) pacing interval in cardiac resynchronization therapy (CRT). Twenty four patients received a biventricular pacemaker and underwent IC for cardiac output (CO) measurements to identify the optimal VV interval. Invasive measurements of left ventricular (LV) dP/dt(max) were used as a reference. During optimization the VV interval was changed with 20 ms steps from +80 (LV pre-excitation) to-80 ms (RV pre-excitation). The optimal VV interval was defined as the one that resulted in the highest LV dP/dt(max) value and the highest CO obtained by IC, respectively. During simultaneous biventricular pacing both LV dP/dt(max) and CO increased (mean 16.6% and 16.2%, respectively) as compared to baseline. Biventricular pacing with optimized VV intervals resulted in a further absolute increase of LV dP/dt (max) and CO (5.6% and 41.3%, respectively). The average decrease in LV dP/dt(max) was 79.6 +/- 51.6 mmHg/s when the optimal VV interval was programmed according to the IC measurements. Cross spectral analysis showed no correlation between the optimal VV intervals identified by the two methods (p > 0.05) and identical optimal VV intervals were identified in only six of the 24 patients. When broader VV time intervals were compared the correlation between the two methods was statistically significant (p = 0,0166). In conclusion, the use of IC for VV interval optimization is questionable since these optimized time intervals do not seem to correlate well with those obtained by measuring LV dP/dt.