Several recent reports have demonstrated an increased incidence of allograft renal vascular thrombosis in patients receiving cyclosporine alone or as part of multiple drug regimens when compared with patients receiving azathioprine (AZA) and prednisone (P). To determine whether CsA therapy is indeed a risk factor for renal artery or vein thrombosis, we examined the incidence of these complications in 224 adult renal allograft recipients who were prospectively randomized and stratified by risk to treatment with either CsA-P (n = 117) or AZA-P-antilymphocyte globulin (n = 107) between September 1980 and October 1983, and in 452 adult and 87 pediatric patients on triple (AZA-P-CsA) or quadruple (AZA-P-CsA-ALG) therapy protocols between July 1984 and November 1987. In the randomized trial, one of 107 AZA-P-ALG patients (0.9%) and two of 117 CsA-P patients (1.7%) developed renal vein thrombosis (P = 0.94), and there were no cases of arterial thrombosis. Though CsA levels were elevated in one of the two CsA-treated patients at the time of their events, and both these patients demonstrated other predisposing factors for thrombosis. In the triple/quadruple therapy era, there were no cases of renal vein thrombosis, and the only case of renal artery thrombosis occurred in a pediatric recipient who was not receiving CsA at the time. These data, when taken together with a critical review of the conflicting literature, strongly suggest that factors other than immunosuppression with CsA, including surgical technique, allograft rejection, use of multiple artery and/or pediatric donor kidneys, and postoperative hypotension, are important in the pathogenesis of allograft renal vascular thrombosis. It seems possible, however, that high initial dosing of CsA might trigger this complication in the early posttransplant period when other predisposing factors are present.