Human lymphocyte antigen (HLA)-identical sibling organs offer the best long-term outcomes for recipients of a renal transplant apart from an identical twin. Unlike cadaveric transplants, however, factors that affect long-term survival of these immunologically privileged grafts are not well described. We reviewed 108 HLA-identical transplants performed at our institution between January 1977 and February 1993. Variables chosen for graft survival analysis were: gender, age and ABO blood type of donors and recipients, panel reactivity antibodies (PRA), blood transfusions prior to transplant, pregnancies, and the underlying renal disease. Additionally, incidence of acute rejection (AR), timing of AR, serum creatinine levels at 1 wk and at 1 yr, and presence of hypertension were included in the analysis. Mean follow-up was 130.9 +/- 58.2 months (range 38-250 months). Actual 5-yr patient and graft survivals were 92 and 88%, respectively. Thirty-eight grafts were lost, and 22 recipients died during the observation period. Death was the main cause of graft failure. Cardiac events accounted for the majority of deaths. AR occurred in 46% and repeated rejections in 11% of recipients. Actuarial graft survival at 10 yr was poorer for patients with any AR (69%), and significantly worse with repeated AR (33%), compared to patients without AR (86%), p = 0.001). Sixty percent of all rejections and 88% of the first rejections occurred in the first 60 d post-transplantation. The first AR that occurred after 60 d was associated with poor graft survival (49 vs. 70%, p = 0.04). Recipients with renal diseases with potential to recur (membranous glomerulonephritis (MGN), membrano-proliferative glomerulonephritis (MPGN), focal and segmental glomerulonephritis (FSGN), polyarteritis nodosa (PAN), rapid progressive glomerulonephritis (RPGN), Henoch-Schoenlein purpura (HSP), diabetes mellitus (DM), interstitial nephritis, systemic lupus erythematosus (SLE) and chronic glomerulonephritis (CGN)) faired worse as a group than recipients with hypertensive nephrosclerosis (HTN), autosomal dominant polycystic kidney disease (ADPKD), Alport's, reflux or congenital dysplasia (68 vs. 96% at 10 yr, p = 0.0009). Poor patient survival was seen in diabetics (71 vs. 88% at 10 yr, p = 0.01). There was a trend to poorer graft survival in diabetic recipients when compared to non-diabetics (65 vs. 81% at 10 yr, p = 0.054). Elevated creatinine at 1 yr was associated with worse graft survival. Likewise, the magnitude of creatinine increase during the first year directly correlated with the risk of graft loss. Hypertensive patients were more likely to lose their grafts than normotensive recipients (72 vs. 86%, p = 0.04). Pre-transplant blood transfusion, pregnancy, and PRA level were not associated with increased graft failure or AR. Graft survival was not affected by gender, age, or ABO blood type of donors or recipients. In conclusion, better prevention and treatment of AR, hypertension, and cardiac disease should improve graft and patient survival. Close attention to recurrence of disease and subtle changes in the creatinine level during the first year might dictate early diagnostic and, hopefully, therapeutic interventions.