[(18)F]FDG PET/CT imaging disproves renal allograft acute rejection in kidney transplant recipients with acute kidney dysfunction: a validation cohort.

PURPOSE: [(18)F]FDG PET/CT may predict the absence of acute allograft rejection (AR) in kidney transplant recipients (KTRs) with acute kidney injury (AKI). Still, the proposed threshold of 1.6 of the mean of mean standardized uptake values (mSUVmean) in the renal parenchyma needs validation. METHODS: We prospectively performed 86 [(18)F]FDG PET/CT in 79 adult KTRs who underwent per-cause transplant biopsy for suspected AR. Biopsy-proven polyoma BK nephropathies (n = 7) were excluded. PET/CT was performed 192 ± 18 min after administration of 254.4 ± 30.4 MBq of [(18)F]FDG. The SUV(mean) was measured in both upper and lower poles of the renal allograft. One-way analysis of variance (ANOVA) and Tukey's studentized range test were sequentially performed. The receiver operating characteristic (ROC) curve was drawn to discriminate "AR" from non-pathological ("normal" + "borderline") conditions. RESULTS: The median age of the cohort was 55 [43; 63] years, with M/F gender ratio of 47/39. The mean eGFR was 31.9 ± 14.6 ml/min/1.73m(2). Biopsies were categorized in 4 groups: "normal" (n = 54), "borderline" (n = 9), "AR" (n = 14), or "others" (n = 2). The median [min; max] mSUV(mean) reached 1.72 [1.02; 2.07], 1.97 [1.55; 2.11], 2.13 [1.65, 3.12], and 1.84 [1.57; 2.12] in "normal," "borderline," "AR," and "others" groups, respectively. ANOVA demonstrated a significant difference of mSUV(mean) among groups (F = 13.25, p < 0.0001). The ROC area under the curve was 0.86. Test sensitivity and specificity corresponding to the threshold value of 1.6 were 100% and 30%, respectively. CONCLUSION: [(18)F]FDG PET/CT may help noninvasively prevent inessential transplant biopsies in KTR with AKI.