Multivariate analysis of risk factors for sensitization and early rejection episodes in a donor-specific transfusion plus azathioprine protocol.

A total of 205 patients who underwent donor-specific transfusions with azathioprine (DST+AZA) treatment from 1982 to 1987 included 25 patients (12%) who developed antibodies cytotoxic to donor T cells (+T warm crossmatch). When we examined a series of 14 variables in relation to the risk of sensitization using a univariate analysis, panel-reactive antibody (PRA) pre-DST of greater than 40%, pregnancy, female sex, male-to-female DST, and HLA antigen match all were significant (P less than .05) risk factors. Of these variables, only PRA pre-DST greater than 40% proved to be statistically significant (P less than .002) in a multivariate logistic regression model (overall P less than .001). Of those patients who went on to transplant with the donor-specific kidney, 41/170 (24%) developed early on or before day 5 posttransplant) rejection episodes. Univariate analysis of 17 variables in relation to early (less than 5 days) rejection episodes indicated no significant variables. Nonetheless using a multivariate logistic regression model (overall P = .09), HLA match (P less than .004), perioperative transfusions (P less than .05), and azathioprine withdrawal for greater than 1 week during DST (P less than .05) all correlated with the incidence of early rejection episodes. In a smaller multivariate model involving 47 patients for whom MLC/blocking factor data were available, the presence of plasma blocking factor at the time of transplant was associated with a decreased incidence of early rejection (P = .07). We conclude that the factors that influence the rate of T+ sensitization are distinct from those that influence the rate of early rejection episodes in a DST+AZA protocol.