Controversy exists regarding the risk factors for renal allograft loss in patients with systemic lupus erythematosus (SLE). This study is a retrospective evaluation of each of these independent risk factors in 80 renal transplants for ESRD secondary to SLE done at our institution between 1971 and 1994. Our entire non-diabetic cohort of 1,966 renal transplants is used as a comparison group. Our results showed equivalent graft survival rates between lupus patients and the cohort at 1, 5 and 10 years (P = 0.56). However, an analysis of cyclosporine-era cadaver grafts revealed that the lupus group had poorer 5-year graft survival than the cohort (41% vs. 71%, P = 0.02). Evaluation of cyclosporine-era lupus graft survival showed significantly improved outcome in living-related lupus recipients over cadaver grafts at five years (89% vs. 41%, P = 0.003). The majority of grafts lost in the lupus cadaver recipients were due to chronic rejection. Rejection was increased in lupus recipients: 69% of lupus patients experienced rejection in the first year compared to 58% of controls (P = 0.01). Stratified for age, sex, race and cyclosporine use, this difference remained significant (P = 0.003, relative risk 1.7). Nephrectomy, splenectomy and 3 to 6 months of pretransplant dialysis did not improve graft survival. A dialysis duration of greater than 25 months predicted worse graft survival (P = 0.01). Among lupus patients, PRA did not correlate with graft outcome (P = 0.5), and HLA-identical cadaver grafts had improved outcomes compared to cadaver grafts. We conclude that acute and chronic rejection are the major risk factors for graft loss in lupus patients. The superior outcome of living-related over cadaver grafts in lupus patients suggests an increased role for living-related grafts. Pretransplant dialysis, nephrectomy and splenectomy are not indicated.