Analysis of risk factors for acute allograft rejection after heart transplantation

The diagnosis of cardiac allograft rejection is currently based on endomyocardial biopsy and histologic evaluation. Since the introduction of cyclosporine, cardiac biopsy has been recognized as the only reliable method of diagnosis. From July 1985 to June 1986, 22 transplant patients were studied during the early posttransplant period to evaluate other characteristics of allograft rejection. The immunosuppressive protocol included cyclosporine, azathioprine, prednisone, and prophylactic rabbit antithymocyte globulin. Acute rejection was established by standard histologic criteria. Twenty-four positive and 123 negative biopsy findings were obtained. Clinical variables (age, sex, immunosuppressive regimen, cyclosporine dosage, ECG voltage), biologic variables (level of cyclosporine, white blood cells, blood urea nitrogen, creatinine, bilirubin, prolactin, urinary polyamines, blood type, HLA typing), and immunologic variables (T-cell subsets) were analyzed in regard to the histologic diagnosis of acute rejection. Four variables (WBC, prolactin, total urinary polyamines, N-acetyl putrescine) were found to have a significant univariate association with acute rejection. These variables were studied by multivariate discriminant analysis. The only factors found to have independent predictive value for acute rejection were prolactin and N-acetyl putrescine peaks occurring before acute rejection. When we used these factors, a discriminant function correctly predicted acute rejection episodes, as well as negative biopsy results, in 74% of the cases in this group of patients.