UNLABELLED: Over the period 1980-1992 256 kidney transplantations were carried out in the Institute of Urology and Nephrology, Clinical Centre, Belgrade: 105 (41%) from cadaveric and 151 (59%) from alive related donors. The first kidney transplantation was performed in our Institution in 1974; however, in the first decade only 27 kidney transplantations were performed. Since 1987, thanks to an increasing number of living kidney donors, the number of transplantations continually increased, and after that period an average of 30 kidney transplantations are performed annually (Figure 1). The aim of the study was to establish the survival of patients and grafts, and factors influencing this survival, as well as to determine the causes of patients' death and graft loss. All the patients were followed-up in our outpatient department within at least 5 years to maximum 17 years. Drug combination therapies were changed in the observation period. From 1983 cyclosporin A (CyA) was added to azathioprine (Aza) and prednisolone (Pr). An increasing number of patients with high immunological risks necessitated the strongest initial immunosuppressive treatment with ALG in addition to Aza and Pr. CyA in a dose of 8 mg/kg b.w. was introduced when serum creatinine concentration fell below 300 mumol/L. The triple treatment including CyA, Aza and Pr was the most common maintenance immunosuppressive therapy in our patients. RESULTS: One and five years survived 95% and 75% of patients, and 84% and 52% of grafts. In assessing the impact of donor source, the year of transplantation, and age of donors we obtained the following results: Living related grafts survived better than cadaver grafts, especially during the first posttransplantation year (Figure 2). Furthermore, graft survival rates from 1987 to 1992 were significantly better than those from early period i.e. 1980 to 1986 (Figure 3). The significantly worse survival rate for grafts from donors older than 60 was noted than for grafts from younger donors. Searching for factors influencing the survival, non immunological and immunological differences between donors and recipients were analyzed. Our analysis showed that 50 living related donors were older than 60. In addition, the majority of them were 20 years older than their graft recipients. Two and more HLA mismatches were observed in 46% of our transplant patients, and 20 patients were highly sensitized. However, the immunological risks were higher in living related transplantations: different ABO blood groups, historical positive cross match reaction between donors and recipients (Table 1). A multivariate analysis using Cox proportional hazards model was performed to determine the important independent predictors of graft survival, and it revealed the following factors (Table 2): number of acute rejections, graft function at the end of the first month and until the end of the first posttransplant year, donors' age, and age and sex differences between donors and recipients. The occurrence of acute rejection at any time had a significant negative effect on graft survival. Since better HLA matching is likely to mean less early rejection it could be concluded that HLA matching influenced graft function and survival in our patients. Absence of acute rejection and delayed graft function or acute tubular necrosis were associated with an improvement of the graft function based on serum creatinine concentration, indicating that delayed graft function also influenced graft survival. The relative risk of graft loss was 2 times higher for patients receiving graft from donors older than 60. Until December 1997, when our analysis was done, of 256 kidney transplant patients 156 lost their grafts. The major causes of graft loss (Table 3) in the early period from 1980 to 1986 were non immunological such as acute tubular necrosis, vascular thrombosis and patients death with functioning graft. (ABSTRACT TRUNCATED)