Reliability of Deceased Donor Procurement Biopsy Images at the Time of Kidney Allocation
1Yale University School of Medicine, New Haven
2University of Utah School of Medicine, Salt Lake City
3Perelman School of Medicine at the University of Pennsylvania, Philadelphia
4Saint Barnabas Medical Center, Livingston
5Hartford Hospital, Hartford
6Wayne State University, Detroit
7University Hospital, Ulm, Germany.
Meeting: 2018 American Transplant Congress
Abstract number: B116
Keywords: Allocation, Biopsy, Donors, Kidney transplantation, non-heart-beating
Session Information
Session Name: Poster Session B: Kidney Deceased Donor Allocation
Session Type: Poster Session
Date: Sunday, June 3, 2018
Session Time: 6:00pm-7:00pm
Presentation Time: 6:00pm-7:00pm
Location: Hall 4EF
Histological findings of procurement deceased-donor kidney biopsies contribute to organ allocation decisions. Representative images of pathology on biopsy slides are often uploaded into United Network of Organ Sharing (UNOS) for optional clinician review but it is unclear how reliable these images are as a clinical tool and if they should be made available to clinicians. Using standardized forms, we compared reports between three experienced and expert renal pathologists for kidney biopsy images uploaded to UNOS system.
A random sample of biopsy reports stratified by UNOS-reported acute tubular injury (ATI) status was used to obtain 60 distinct image sets (32 without ATI and 28 with ATI). Fleiss kappa was used to determine inter-rater agreement.
Mean (±SD) donor age was 49 ±14, 57% were male, and 33% had at least one kidney discarded. There was no inter-rater agreement for the finding of ATI [Kappa 0.03 (95%CI -0.12, 0.18)]; minimal agreement for interstitial fibrosis [0.36 (0.22, 0.50)] and arteriolar hyalinosis [0.29 (0.15,0.44)]; weak agreement for interstitial inflammation [0.48 (0.33,0.62)], tubular atrophy [0.43 (0.29,0.57)], and arterial intimal fibrosis [0.41(0.35,0.48)]; and moderate agreement for glomeruli thrombi [0.63 (0.38, 0.88)] and glomerulosclerosis [0.72 (0.29, 1.00)].
There was minimal to moderate inter-rater agreement between pathologists, raising concern about the utility and reliability of uploaded biopsy images. Although further research is needed to understand how UNOS-uploaded images are being used clinically, these images have been shown to be an unreliable tool when read by expert pathologists. The field may consider precluding the availability of these images to clinicians in the future as they may influence clinical decisions inappropriately.
CITATION INFORMATION: Mansour S., Hall E., Reese P., Jia Y., Thiessen-Philbrook H., Moeckel G., Weng F., Revelo P., Khalighi M., Trivedi A., Doshi M., Schroppel B., Parikh C. Reliability of Deceased Donor Procurement Biopsy Images at the Time of Kidney Allocation Am J Transplant. 2017;17 (suppl 3).
To cite this abstract in AMA style:
Mansour S, Hall E, Reese P, Jia Y, Thiessen-Philbrook H, Moeckel G, Weng F, Revelo P, Khalighi M, Trivedi A, Doshi M, Schroppel B, Parikh C. Reliability of Deceased Donor Procurement Biopsy Images at the Time of Kidney Allocation [abstract]. https://atcmeetingabstracts.com/abstract/reliability-of-deceased-donor-procurement-biopsy-images-at-the-time-of-kidney-allocation/. Accessed May 9, 2025.« Back to 2018 American Transplant Congress