Factors and outcome in BK virus nephropathy in a Hispanic kidney transplant population

Abstract

BK virus nephropathy (BKVN) is an increasingly recognized cause of kidney allograft loss and is thought to be related to the newer, more potent immunosuppressive agents. Conflicting information has been reported on risk factors for BK infection. To determine incidence, associated factors, and outcome of BKVN in our kidney transplant population in order to improve identification and management. Kidney transplants from January 2000 to December 2005 were retrospectively reviewed. Data were collected for patients with biopsy-proven BKVN including age, sex, body mass index (BMI), etiology of renal failure, other medical diseases, donor type, surgical complications, rejection and infection, time to diagnosis, induction, immunosuppressive and antiviral therapy, and clinical outcome. A control group of patients matched for sex, age, type of graft, etiology of kidney disease, and BMI, was established for comparison. During this period, 20 (4%) of 497 transplanted patients were diagnosed with BKVN. Thirteen (65%) were males, 8 (40%) were young adults (ages 21-40), and 18 (90%) received grafts from cadaveric donors (P=0.05). Twelve (60%) had hypertensive renal disease, 2 (10%) also had diabetes, and 16 (80%) had a BMI >25 (P=0.01). Lymphoceles occurred in 5 patients (25%). Mean creatinine level at diagnosis was 2.7 mg/dL and mean time to diagnosis was 23 months. Ten patients (50%) had leukopenia at or within a year before biopsy (P=0.001). Viruses other than BK occurred in 9 patients: varicella zoster virus in 3, cytomegalovirus in 2, herpes simplex virus in 1, molluscum contagiosum in 1, Epstein-Barr virus in 1, and human papillomavirus in 1. Eighteen patients (90%) had related rejection (P= 0.001) and 4 (20%) suffered allograft loss (P= 0.001). Basiliximab (living donors) and anti-thymocyte globulin (cadaver donors) were given for induction. All patients were on triple therapy; 15 on prednisone and sirolimus, with either tacrolimus in 8, cyclosporine in 4, mycophenolate in 1, or mycophenolate and tacrolimus in 2. The other 5 received prednisone with tacrolimus and mycophenolate. Graft loss occurred in 2 patients on tacrolimus and mycophenolate, 1 patient on tacrolimus and sirolimus, and 1 patient on cyclosporine and sirolimus. Immunosuppression was decreased in all patients. Two were given cidofovir for 6 months and had stable creatinine levels at the end of the study. Records were reviewed until April 2007. There were no deaths in this cohort. The number of rejections experienced by patients with BKV was much higher (P<0.0001), but the rate of graft loss was similar between the 2 groups (P=0.19). Viral co-infection was more frequent in patients with BKV (P=0.04). No episodes of leukopenia were reported for any of the patients in the control group (P=0.001). Immunosuppression with tacrolimus and sirolimus was more frequent in the BKV group, but this was not statistically significant (P=0.18, 0.28, respectively). The number of lymphoceles was larger in patients with BKV, but the difference was not statistically significant (P=0.35). BKVN is present in our transplant population and results in a high rate of allograft rejection with varying rates of graft loss. Associated factors were deceased donor and immunosuppression with potent agents, particularly tacrolimus and sirolimus. We also found a higher frequency of obesity, viral co-infection, and leukopenia. Routine screening and timely biopsy could prove cost-effective and significantly reduce morbidity.