Accuracy of definitive rapid onsite evaluation cytopathology diagnoses: Assessment of potentially critical diagnoses as a quality assurance measure

Intraprocedural rapid onsite evaluation (ROSE) of cytology specimens enhances cytopathology practice. More recently, ROSE diagnoses, like frozen section (FS) diagnoses, have guided immediate clinical decisions. In this study, we evaluated the diagnostic accuracy of definitive ROSE diagnoses in our quality assurance system over a 52-month period. Cytopathology cases with ROSE from January 2017 to April 2021were retrieved from our laboratory information system. After excluding cases that were deferred or nondiagnostic/unsatisfactory, each definitive ROSE diagnosis (ie, negative for malignant cells or positive for malignant cells) was categorized as having agreement or disagreement with the final diagnosis. For comparison, concordance of FS diagnoses from the same time period were tabulated and compared to those of ROSE diagnoses by using χ2 testing with P < 0.05 considered statistically significant. Of the 1649 ROSE diagnoses, there were 15 disagreements (0.9%) with 1 final moderate interpretive disagreement (0.06%). By comparison, of the 17,469 FS diagnoses, there were 141 disagreements (0.8%) with 49 final moderate or major interpretive disagreements (0.3%). The remaining disagreements were minor. There were no statistically significant differences in the rates of final moderate and major interpretive disagreements. The final interpretive disagreement rates for definitive ROSE and FS diagnoses were similar in this study. Given the expanding role of ROSE and its use for immediate clinical decisions in some cases, monitoring the accuracy of definitive diagnoses may serve as an initial quality assurance measure. •Intraprocedural rapid onsite evaluation (ROSE) of cytology specimens enhances cytopathology practice and some definitive ROSE diagnoses have guided immediate clinical decisions.•However, quality assurance (QA) of ROSE diagnoses is not required currently.•This QA study compared the definitive ROSE diagnoses to the final cytopathologic diagnoses. The disagreed ROSE diagnoses were tabulated and the data were compared to those of frozen section (FS) diagnoses.•There were no statistically significant differences in the rates of moderate and major interpretive disagreements for ROSE and FS diagnoses.•Monitoring definitive ROSE diagnoses, which can influence critical clinical decisions, may represent the initial step in ROSE QA and have potential patient-safety benefits.