Little information is available regarding donor-specific parameters that predict success or failure after heart transplantation. Furthermore, with increasing numbers of patients awaiting heart transplantation, there is tremendous pressure to expand the donor pool by stretching the margins of donor acceptability. To gain insight into donor-related and donor-recipient interrelated predictors of death after transplantation, 1719 consecutive primary transplantations performed at 27 institutions between Jan. 1, 1990, and June 30, 1992, were analyzed. Mean follow-up of survivors was 13.9 months, and actuarial survival was 85% at 1 year. By multivariable analysis, risk factors for death included younger recipient age (p = 0.006), older recipient age (p = 0.0005), ventilator support at time of transplantation (p = 0.0006), higher pulmonary vascular resistance (p = 0.02), older donor age (p < 0.0001), smaller donor body surface area (female donor heart placed into larger male patient) (p = 0.003), greater donor inotropic support (p = 0.01), donor diabetes mellitus (p = 0.01), longer ischemic time (p = 0.0003), diffuse donor heart wall motion abnormalities by echocardiography (p = 0.06), and, for pediatric donors, death from causes other than closed head trauma (p = 0.02). The overall 30-day mortality rate was 7% but increased to 11% when donor age exceeded 50 years and was 12% when inotropic support exceeded 20 μg/kg/min dopamine plus dobutamine and 22% with diffuse echocardiographic wall motion abnormalities. The interaction of donor risk factors was such that the heart of a smaller female donor given high-dose inotropes placed into a larger male recipient produced a predicted 30-day mortality rate of 26% and the heart of a 25-year-old male donor given high-dose inotropes with diffuse echocardiographic wall motion abnormalities transplanted into a 50- year-old male recipient led to a predicted 30-day mortality rate of 17%. This analysis supports cautious extension of criteria for donor acceptance but with an anticipated greater risk in the presence of diffuse echocardiographic wall motion abnormalities and long anticipated ischemic time, particularly in older donors given inotropic support.