The primary purpose of this study was to evaluate parameters used for the measurement of regional myocardial contractile function in the setting of left ventricular (LV) asynchrony. Secondarily, we tested whether the peak negative value of left ventricular dP/dt (-dP/dt) can be used to estimate global LV end-systole during asynchrony. In seven anesthetized (Isoflurane) swine the left anterior descending coronary artery was cannulated and perfused at constant blood flow rates. To produce LV asynchrony, dobutamine (D) was infused into the perfusion system. This was repeated later during coronary hypoperfusion (HYPO) sufficient to produce regional contractile dysfunction. The amount of LV wall thickening during systole (% WT, sonomicrometry) was calculated using either -dP/dt or the closure of the aortic valve (AO, electromagnetic flow probe) for estimating the timing of global LV end-systole. % WT was compared to other parameters which are not dependent upon the timing of global LV end-systole, including the amplitude of the first harmonic of the Fourier transform (AMP) and regional myocardial work (WI) estimated form the left ventricular pressure-wall thickness relationship. A close correlation between global LV end-systole defined by the AO or -dP/dt existed during control, D or HYPO. During HYPO + D no such relationship was found (r = .22, NS), and % WT calculated using -dP/dt as an estimate of end-systole was underestimated when compared to % WT calculated by use of the AO to estimate end-systole (2.9 +/- 6.8% vs 6.3 +/- 6.6%, p less than .05). % WT, AMP, and WI showed similar results during control, D and HYPO. However, During HYPO increased the AMP from .59 +/- .23 mm to .76 +/- .32 mm and WI from 67 +/- 20 mm Hg*mm to 95 +/- 24 mm Hg*mm (p less than .05), respectively. This increase in regional myocardial function, however, was not detected by % WT (10.5 +/- 6.4% vs 6.3 +/- 6.6%). Thus, during left ventricular asynchrony, the measurement of LV -dP/dt to estimate the timing of global LV end-systole is inappropriate and can lead to inaccuracies in the measurement of regional contractile function. Parameters such as AMP or WI are advantageous since global LV end-systole does not need to be accurately defined.