BACKGROUND Diastolic dysfunction seen in early clinical transplant rejection has been difficult to demonstrate in experimental rodent models because of the inability to make sensitive in situ measurements of systolic and diastolic functions. We have developed a heterotopic heart transplant model with Fisher 344 and ACI rats (without immunosuppression), where in situ measurements of diastolic and systolic functions were made sequentially (daily) by use of an implanted left ventricular balloon. METHODS Syngeneic and allogeneic heterotopic heart transplants were performed. In situ function was determined by varying balloon volume to measure the developed pressure, the rates of pressure rise (+dp/dt) and pressure fall (-dp/dt), diastolic pressure-volume relationship, and the time constant of diastolic relaxation (tau). These results were compared with function measurements in transplanted hearts that were excised and perfused in a Langendorff mode (ex vivo) during the same posttransplantation period. RESULTS Histologic examination revealed that at day 3 after transplantation, allografts showed mild lymphocytic infiltration indicative of mild or early rejection, and by day 5, there was severe rejection with myocyte necrosis. By day 3, the slope of the diastolic pressure-volume relationship (ie, left ventricular stiffness) was significantly higher in allografts as compared with isografts (436 +/- 96 vs 177 +/- 26 mm Hg/mL, p < .05). Similarly, tau was significantly longer in allografts by day 3 after transplantation. Developed pressure and +dp/dt became significantly lower in allografts beginning on day 6. Function measurements made in the isolated perfused ex vivo hearts yielded the same results at day 3 after transplantation as the in situ group of hearts. CONCLUSION Using a chronically implanted left ventricular balloon, we have developed a heterotopic heart transplant model where sensitive measurements of systolic and diastolic functions can be made. With this preparation, the early changes in the diastolic dysfunction seen clinically can be reproducibly detected. Thus this model may be useful to study mechanisms and interventions during early transplant rejection.