OBJECTIVE To analyse the clinical characteristics of patients who died on the Stanford heart transplant waiting list and to develop a method for risk stratifying status 2 patients (outpatients). METHODS Data were reviewed from all patients over 18 years, excluding retransplants, who were accepted for heart transplantation over an eight year period from 1986 to 1994. RESULTS 548 patients were accepted for heart transplantation; 53 died on the waiting list, and 52 survived on the waiting list for over one year. On multivariate analysis only peak oxygen consumption (peak VO2: 11.7 (SD 2.7) v 15.1 (5.2) ml/kg/min, P = 0.02) and cardiac output (3.97 (1.03) v 4.79 (1.06) litres/min, P = 0.04) were found to be independent prognostic risk factors. Peak VO2 and cardiac index (CI) were then analysed in the last 141 consecutive patients accepted for cardiac transplantation. All deaths and 88% of the deteriorations to status 1 on the waiting list occurred in patients with either a CI < 2.0 or a VO2 < 12. In those with a CI < 2.0 and a VO2 < 12, 38% died or deteriorated to status 1 in the first year on the waiting list. Patients with CI > or = 2.0 and a VO2 > or = 12 all survived throughout follow up. Using a Cox's proportional hazards model with CI and peak VO2 as covariates, tables were constructed predicting the chance of surviving for (a) 60 days and (b) 1 year on the waiting list. CONCLUSIONS These data provide a basis for risk stratification of status 2 patients on the heart transplant waiting list.