Abstract: PUB343
Diagnostic Performance of Polymerase Chain Reaction in Diagnosing BK Virus-Associated Nephropathy: A Decade-Long, Single-Center Experience
Session Information
Category: Transplantation
- 2102 Transplantation: Clinical
Authors
- Kew, Clifton E., The University of Alabama at Birmingham, Birmingham, Alabama, United States
- Wei, Tina Ruoting, The University of Alabama at Birmingham, Birmingham, Alabama, United States
- Abujar, Sheikh, The University of Alabama at Birmingham, Birmingham, Alabama, United States
- Tamhane, Ashutosh, The University of Alabama at Birmingham, Birmingham, Alabama, United States
- Chen, Jin, The University of Alabama at Birmingham, Birmingham, Alabama, United States
- Taylor, Christy, The University of Alabama at Birmingham, Birmingham, Alabama, United States
- Agarwal, Gaurav, The University of Alabama at Birmingham, Birmingham, Alabama, United States
Group or Team Name
- BK Group.
Background
BK viremia has been linked with BK virus-associated nephropathy (BKVAN), a significant risk for kidney transplant inury and loss. Biopsies are costly, inconvenient, and associated with complications. Monitoring kidney transplant patients for BKVAN, the less invasive plasma BK Polymerase Chain Reaction (BKPCR) test can be used to assess for viral activity. We examined the overall accuracy and performance measures of BKPCR in diagnosing BKVAN.
Methods
We included all kidney transplants at our institution between 01/2013 - 12/2023 and had a for cause or surveillance kidney biopsy (the gold standard) performed. The BKPCR diagnostic criteria for BKVAN (PCRDX) was either BKPCR >1000 IU/mL on 2 determinations 3 weeks apart or BKPCR >10000 IU/mL once. Any immunohistochemical staining for BK virus on pathology was considered evidence of BKVAN. Overall accuracy and various performance measures of PCRDX were calculated.
Results
Overall, 316 (19.5%) of the 1624 kidney transplants met the BK PCR criteria of BKVAN. Mean age was 50 years at transplantion, 59.5% were men, 59.9% were African-American, and median time on dialysis was 122 months. Of the 115 patients who were biopsy-positive 110 (true positives) were PCRDX(+) while of the 1509 patients who were biopsy-negative 1303 (true-negatives) were PCRDX(-); 5 patients were false-negative and 206 were false-positive. The overall accuracy of PCRDX was 87.0% (95% CI: 86.2 – 87.4), sensitivity was 95.7% (95% CI: 89.8 – 98.4) and specificity was 86.3% (95%CI: 85.9 – 86.6). The positive predictive value was 34.8% (95% CI: 32.7 – 35.8) whereas the negative predictive was 99.6% (95% CI: 99.1 – 99.9). The C- statistic (area under the curve) for PCRDX to predict BKVAN was 91.0% (95% CI: 88.9 – 93.1). The positive likelihood-ratio (LR) was 7.0 (95% CI: 6.4 – 7.3) and negative LR was 0.05 (95% CI: 0.02 – 0.12).
Conclusion
We conclude that in this large cohort of kidney transplants, utilizing set criteria for BK viremia had good accuracy in diagnosing BKVAN especially in ruling-out BKVAN.
Agreement between biopsy and BKPCR in diagnosing BKVAN
| Biopsy (+) n(%) | Biopsy(-) n(%) | Total | |
| PCRDX(+) | 110 (95.7) | 206 (13.7) | 316 (19.5) |
| PCRDX(-) | 5 (4.3) | 1303 (86.3) | 1308 (80.5) |
| Total | 115 | 1509 | 1624 |
Funding
- Commercial Support – MEMO Therapeutics AG