ABSTRACT
BACKGROUND: Urine CXCL10 is a biomarker for renal allograft inflammation induced by rejection, urinary tract infection, or BK polyomavirus (BKPyV) replication. This study aimed to compare urine CXCL10 levels in different stages of BKPyV reactivation and to investigate urine CXCL10 as a biomarker for BKPyV replication. METHODS: We included 763 urine samples (235 patients) from an interventional, randomized trial obtained in the context of regular screening for urine CXCL10 levels. All urine samples had a complete urine sediment analysis, no rejection episode noted within 30 d before urine collection, and a urine decoy cell analysis was conducted within ±3 d. RESULTS: Urine CXCL10 levels were 2.31 ng/mmol in samples without BKPyV viruria, slightly rose to 4.35 ng/mmol with BKPyV viruria, and then markedly increased to 16.42 ng/mmol when decoy cells were detectable, but still in the absence of BKPyV DNAemia ( P < 0.001). The highest urine CXCL10 values were observed in samples with BKPyV DNAemia (median 42.59 ng/mmol). The area under the curve of urine CXCL10 levels to detect ≥3 decoy cells was 0.816. At a CXCL10 cutoff of 3 ng/mmol, the negative predictive value was 97%. The area under the curve of urine CXCL10 levels to detect BKPyV DNAemia was 0.882, with a negative predictive value of 99% at a CXCL10 cutoff of 3 ng/mmol. CONCLUSIONS: Urine CXCL10 levels are already significantly elevated in BKPyV viruria (especially with decoy cell shedding) and further increase with BKPyV DNAemia. Low urine CXCL10 values can rule out the presence of ≥3 decoy cells and BKPyV DNAemia with high certainty.
Subject(s)
BK Virus , Kidney Diseases , Kidney Transplantation , Polyomavirus Infections , Tumor Virus Infections , Humans , Biomarkers , Chemokine CXCL10/urine , Kidney Transplantation/adverse effects , Polyomavirus Infections/diagnosis , Tumor Virus Infections/diagnosis , UrineABSTRACT
BACKGROUND: Delayed graft function (DGF) and pretransplant donor-specific HLA-antibodies (DSA) are both regarded as risk factors for rejection and lower graft survival. However, the combined impact of DGF and DSA has not been studied in detail. METHODS: We investigated 375 deceased donor kidney transplantations, which had DSA assignment by single-antigen bead technology and which had surveillance biopsies at 3 of 6 months. Median follow-up time was 6.1 years. RESULTS: DGF occurred in 137 of 375 patients (37%), and DSA were present in 85 of 375 patients (23%). The incidence of DGF was similar in DSA-positive (DSApos)-patients and DSA-negative (DSAneg)-patients (40% versus 36%; P = 0.45). In DSAneg-patients, 5-year graft survival was not different with/without DGF (81% versus 83%; P = 0.48). By contrast, in DSApos-patients, 5-year graft survival was significantly lower with DGF (64% versus 79%; P = 0.01). Moreover, DSApos-patients with DGF had a higher 1-year incidence of subclinical rejection, which were mostly antibody-mediated or mixed rejection phenotypes. Graft loss due to rejection was significantly more frequent in DSApos-patients with DGF (5/34; 15%) compared to DSApos-patients without DGF (2/51; 4%), and DSAneg-patients with/without DGF (3/103; 3% and 4/187; 2%, respectively) (P = 0.005). In a multivariate Cox model, DSA with DGF was an independent predictor for graft (hazard ratio = 2.84 [95% confidence interval, 1.54-5.06]; P = 0.001) and death-censored graft loss (hazard ratio = 4.65 [95% confidence interval, 1.83-11.51]; P = 0.002). CONCLUSIONS: DGF has a much more detrimental impact in DSApos-patients than in DSAneg-patients, which is likely related to a higher incidence of antibody-mediated rejection. If possible, the combined risks of DGF and DSA should be avoided.
