Key points are not available for this paper at this time.
In order to predict kidney graft survival, the influence of independent prognostic factors can be examined multivariately and the factors combined into a prognostic index. Data on 7121 patients receiving an unrelated first and 1033 patients receiving an unrelated second transplant from nonliving donors, between 1 January 1984 and 31 December 1987, were analyzed to ascertain the most important prognostic variables up to 4.5 years posttransplantation. Factors found to be significant for graft survival were donor and recipient age and sex, recipient blood group, whether the recipient was diabetic, cold ischemic period, number of HLA-B and - DR mismatches, highest percent panel-reactive antibody, transplant center, and--for second transplants--duration of first graft. A risk score for graft failure, based on the prognostic factors, was developed using these factors and five risk groups (from excellent to very poor prognosis) were identified. This index was tested on an independent data set and showed a good fit when compared with the observed Kaplan-Meier graft survival: patients allocated by the risk score into the "excellent prognosis" group had an observed one-year graft survival of 90.4%, compared with a predicted value of 90.3% for first transplants. Corresponding results for second transplants were 86.2% (observed) and 86.0% (predicted). For the "very poor" prognosis group, the results were 73.4% (observed) and 74.4% (predicted), for first transplants, and 60.9% (observed) and 60.1% (predicted), for second transplants. A prognostic index can therefore identify patients likely to have a high or low graft survival, leading to improved decision-making and aiding the choice of patient management once a recipient has been transplanted.
Thorogood et al. (Fri,) studied this question.