Ureteral necrosis after kidney transplantation: Risk factors and impact on graft and patient survival

Background. Ischemia, the main cause of ureteral necrosis in renal transplantation, cannot alone explain the late occurrence of some fistulas beyond the first postoperative month. The aim of this study, performed on a cohort of 1,629 consecutive kidney transplantations, was to analyze the risk factors implicated in the occurrence of ureteral necrosis and its impact on graft and patient survival. Methods. Between January 1990 and December 2001, 1,629 renal transplantations were performed in the authors' center. All biologic and clinical data were computerized in a cross-audited and validated data bank (Donnees Informatisees et Validees en Transplantation). The parameters studied were donor age, gender, cause of death and serum creatinine before procurement; and recipient age, gender, initial disease, panel reactive antibody, retransplantation, cold ischemia time, delayed graft function, human leukocyte antigen incompatibilities, induction and maintenance immunosuppression, right or left kidney, number of arteries, site of transplantation and the presence or not of a double-J stent. The follow-up parameters were the number and timing of acute rejection episodes, cytomegalovirus (CMV) infection (viremia, polymerase chain reaction), and acute pyelonephritis. Ureteral histologic analysis was performed in 25 cases (necrosis, leukocyte infiltration, and CMV or BK virus inclusions). Uni- and multivariate statistical tests were used (alpha risk at 5%). All of the patients with ureteral necrosis had undergone neoureterocystostomy or ureteral anastomosis with the native ureter but with a systematic double-J stent. Results. Ureteral necrosis occurred in 52 of the 1,629 patients (3.2%) and was significantly and independently correlated with donor age (P = 0.041) and delayed graft function (P = 0.016). CMV infections were also higher in the necrosis group (P = 0.001), but donor CMV status was not statistically different between the two groups (36.2% vs. 36.7%). Ureteral histologic studies showed CMV and BK virus inclusions in 4 and 2 cases, respectively, and arterial and venous thrombosis in 4 and 16 cases, respectively. No pattern of ureteral rejection was observed. Ureteral necrosis did not affect the 10-year patient and graft survival, which were 87% and 66%, respectively, for the necrosis group and 86% and 58%, respectively, for the control group (P = not significant). Conclusions. The authors' data provide new information concerning a classic surgical complication after kidney transplantation. The link they have identified between the occurrence of ureteral necrosis, donor age, and delayed graft function reemphasizes the interdependence between surgical and medical complications in kidney transplantation.