Diagnostic Discrepancies between Frozen and Permanent Section Diagnoses in Pretransplant Donor Kidney Biopsies: Quality Assurance Study of 1887 Cases
K Sasaki, P Randhawa, AJ Demetris, E Sasatomi. University of Pittsburgh Medical Center, Pittsburgh, PA
Background: Pretransplant donor kidney biopsies are mandatory when a donor kidney is the marginal category. In such cases, frozen section evaluation quickly provides useful information.
Design: Evaluation of donor kidney biopsy includes three histologic parameters (interstitial fibrosis; Ci, arteriosclerosis; Cv, arteriolosclerosis; Ah) scored as 0-3 according to Banff 97 criteria, counting the percentage of glomerulosclerosis (Gs), and describing additional pathologic findings such as glomerular abnormality, capillary thrombosis, infarct, acute tubular necrosis (ATN), and tumor. For this study, the types of, and reasons for, discrepancies between frozen and permanent section diagnoses were analyzed in 1887 pretransplant donor kidney biopsies obtained between March 1994 and April 2008. Discrepancies were then classified as major (difference of score ≥ 1 in Ci, Cv, or Ah and difference of percentage of Gs (≥ 20% GS vs < 20 % GS)), minor (difference of score < 1 in Ci, Cv, or Ah) and miscellaneous (capillary thrombosis, glomerular abnormality, ATN, infarct, and tumor).
Results: 157 kidneys among 1887 pretransplant donor kidney biopsies (8.3%) showed discrepancies between frozen and permanent section diagnoses. 78/165 discrepancies were considered major (47.3 %), 68/165 minor (41.2%), and 19/165 miscellaneous (11.5 %). The most frequent major discrepancies were Ah (35.9%) and Gs (35.9%), followed by Cv (21.8 %) and Ci (5.1%). Minor discrepancies included Cv (47.1 %), Ah (27.9 %) and Ci (25.0 %). Miscellaneous discrepancies included glomerulopathy (78.9 %), capillary thrombosis (10.5 %), and ATN (10.5 %). Most Ah lesions were underestimated (87.2 %) in frozen section. All the Gs-related discrepancies were attributable to sampling variability (100 %). Cv was falsely estimated (overestimate; 49.0 %, underestimate; 51.0 %) most often attributed to sampling variability. Ci was often overestimated (71.4 %) mainly due to misinterpretation of interstitial edema. Most glomerular abnormality-related discrepancies were due to underestimation of Kimmelstiel-Wilson lesions (93.3 %).
Conclusions: Major discrepancies between frozen and permanent section evaluation of donor biopsies is uncommon, but not rare and most often includes underestimation of Ah, overestimation of Ci, and sampling variability in Gs and Cv scoring. Surgical pathologists should be aware of these potential pitfalls and limitations for an accurate evaluation of pretransplant donor kidneys.
Category: Kidney (does not include tumors)
Monday, March 22, 2010 9:30 AM
Poster Session I Stowell-Orbison/Surgical Pathology/Autopsy Awards Poster Session # 206, Monday Morning