Diagnostic performance and concordance of rapid on-site evaluation using touch preparations of core needle biopsies...

Rapid on-site evaluation (ROSE) using touch preparations (TP) of core needle biopsies (CNB) are increasingly utilized in the diagnostic workup of renal masses. This study examines the concordance between TP and corresponding CNB in both adequacy and diagnostic interpretation.

A retrospective review was conducted of 415 renal mass CNB with corresponding ROSE using TP, performed between July 2011 and October 2020. ROSE adequacy and impressions were compared with final CNB diagnoses. Impressions were categorized as inadequate, negative, atypical/suspicious, or positive for malignancy or neoplasm.

Of the 336 procedures deemed adequate by ROSE, 320 (95%) corresponded to a satisfactory CNB, with a sensitivity and positive predictive value (PPV) of 86% and 95%. Specificity and negative predictive value (NPV) were 62% and 35%. ROSE impressions were recorded in 386 cases, with an overall concordance rate of 82% between ROSE impression and final CNB diagnosis. A "positive" ROSE impression with a malignant or neoplastic CNB diagnosis had a sensitivity and PPV of 67% and 97%, while a "nonpositive" impression with a nonmalignant or non-neoplastic CNB diagnosis had a specificity and NPV of 91% and 36%. Discordance was most often due to absent or scant lesional material on TP despite sufficient lesional cells in CNB, or misinterpretation of normal renal parenchyma as lesional tissue, and vice versa.

ROSE using TP is a valuable adjunct in renal mass evaluation, offering high predictive value for specimen adequacy and impression, and guiding biopsy workflows.