Prediction of BK viremia by urine viral load in renal transplant patients: An analysis of BK viral load results in paired urine and plasma samples.

BK virus (BKPyV)-associated nephropathy (BKPyVAN) may affect up to 10% of renal transplant recipients, causing graft failure in the absence of intervention. The dilemma in monitoring BKPyVAN in renal transplant patients has been that only testing urine BK viral load represents higher sensitivity (earlier detection) but lower specificity, while testing plasma BK viral load represents lower sensitivity (later detection) but higher specificity. However, blindly testing both urine and plasma inevitably contributes to unnecessary medical cost. We analyzed 1030 paired urine and plasma BKPyV viral load results and identified a reliable urine BKPyV viral load cutoff (4.0 log IU/mL) that can predict BKPyV viremia with 99.7% negative predictive value (NPV). We propose a cost-effective screening algorithm to first only monitor the urine BKPyV levels until the viral load reaches 4.0 log IU/mL, and then only monitor plasma with higher frequency. This approach ensures 98.7% sensitivity of catching the earliest BKPyV viremia onset, and 100% sensitivity of detecting the critical BKPyV viremia. In addition, we identified a urine BKPyV viral load cutoff of 6.7 log IU/mL as predictive of critical BKPyV viremia (defined as plasma viral load >4.0 log IU/mL) with 100% sensitivity and 100% NPV.