To combat the continuing shortage of ideal donor hearts, we have used cardiac allografts from high-risk donors for critically ill recipients. We defined high-risk donor variables as age greater than 40 years, systemic (noncardiac) infection, cardiopulmonary resuscitation greater than 3 minutes, ischemic time longer than 5 hours, weight more than 20% less than that of the recipient, and requirements for high doses of inotropes. Of the 305 donors we have used, 73 (23.9%) have been high-risk, with 59/73 (80.8%) exhibiting one variable, 12/73 (16.4%) exhibiting two variables, and 2/73 (2.7%) exhibiting three variables. No correlation was found between the number of donor variables and a poor postoperative result. No infectious complications occurred in 17 patients receiving hearts from potentially infected donors. Hospital mortality rates (30 day) for recipients of high-risk donor versus non-high-risk donor hearts were 8.2% and 6.9%, respectively (not significant). The 1-, 6-, and 12-month actuarial survival rates were 91.7%, 81.2%, and 75.9% for the high-risk donor group and 93.5%, 80.3%, and 77.8% for the non-high-risk donor group (not significant). Among survivors with high-risk donor hearts, mean left ventricular ejection fractions were 0.54 +/- 0.08 at 3 months, 0.55 +/- 0.08 at 1 year, and 0.54 +/- 0.09 at 2 years after transplantation. These results suggest that accepting less than ideal donor hearts can be safe and might be considered when better options are not available.