The classical indicators of left ventricular (LV) performances have been derived from pressure-volume (PV) and stroke work-volume plots obtained during various loading or pharmacological interventions. More recently, the preload-adjusted maximal change in pressure over time (PAdP/dtmax), derived from single beat PV analysis, has been shown to reflect the LV systolic performance accurately in varying conditions of inotropy and afterload. The objective of this study was to address whether PAdP/dtmax is a valid indicator of LV function in the setting of large preload variations, compared with the classical dP/dtmax-end-diastolic volume (EDV) and stroke work-EDV (preload recruitable stroke work) relationships. Nine anaesthetized and mechanically ventilated rats were instrumented with a ventricular conductance catheter. Stepwise preload reduction was achieved by repeated blood withdrawals (up to a total of 5 ml). Steady-state and dynamic PV loops were recorded during brief occlusion of the inferior vena cava, and LV function parameters were derived from these recordings. Our results demonstrate that PAdP/dtmax behaved in a similar manner to preload recruitable stroke work, reflecting well-maintained LV contractility during controlled haemorrhage until mean arterial pressure decreased below 40 mmHg. In contrast, dP/dtmax-EDV increased significantly and exhibited a curvilinear response that was associated with a large inter- and intra-animal variability. In a model of acute preload reduction, PAdP/dtmax was found to be the best indicator of systolic LV function. Given its simplicity, this real-time index derived from single beat analysis should be tested further in clinical settings.