Premium
Improving ROSE: Discrepant touch preparation and histology findings in cytology of renal masses: A 10‐year retrospective review
Author(s) -
Shachner Tracy R.,
Nodit Laurentia,
Hubbard Elizabeth,
Van Meter Stuart
Publication year - 2019
Publication title -
diagnostic cytopathology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.417
H-Index - 65
eISSN - 1097-0339
pISSN - 8755-1039
DOI - 10.1002/dc.24253
Subject(s) - medicine , biopsy , malignancy , radiology , predictive value , gold standard (test) , renal cell carcinoma , diagnostic accuracy , cytology , predictive value of tests , pathology
Background The number of “renal incidentalomas” is on the rise due to increasing use of radiologic studies. Image‐guided core needle biopsies (CNB) with touch preparations are performed to guide specimen collection and triage of sample for additional studies. Results allow the clinical team to make appropriate treatment decisions. Design Our electronic database was searched for a 10‐year period to identify 180 image‐guided biopsies of renal masses with rapid on‐site evaluations (ROSE) and corresponding biopsy/resection specimens. Touch preparations were classified as non‐diagnostic, negative/benign, adequate/positive for malignancy/oncocytic predominance, or atypical. These results were compared to the final diagnosis on the biopsy or resection specimen (if available). Diagnostic accuracy, sensitivity, specificity, and positive and negative predictive values were determined. Non‐diagnostic cases and cases in which ROSE and final diagnosis were discordant were reviewed by cytopathologists blinded to the original interpretation to reconcile discrepancies and highlight interpretation pitfalls. Results A ROSE diagnosis was rendered in 133 of 180 cases; 47 cases were non‐diagnostic. Of the 133 diagnostic cases, the ROSE diagnosis was concordant with the core biopsy final diagnosis in 125 cases, yielding a diagnostic accuracy of 94%. The overall sensitivity was calculated to be 80.1%; specificity 72.4%; positive predictive value 94%; and negative predictive value 41.2%. Conclusions Touch preparation slides are vital but imperfect tools in evaluating renal masses. In our study, distinction between malignant and benign samples was accomplished in most cases (94% accuracy), but there are limitations. Awareness of interpretation pitfalls allows informed decisions to be made regarding specimen collection and patient management.