Interinstitutional comparison of frozen section consultations. A college of American Pathologists Q-Probes study of 90,538 cases in 461 institutions.

OBJECTIVE To assess concordant, discordant, and deferred diagnosis rates from frozen sections; to determine reasons for discordance; to identify pathologic processes associated with discordant diagnoses; to determine false-positive or false-negative rates for neoplasms; and to identify anatomic sites associated with discordant frozen section diagnoses. DESIGN Q-Probes study of the College of American Pathologists. PARTICIPANTS Four hundred sixty-one institutions participating in the Q-Probes program from November 1, 1990, through March 31, 1991. MAIN OUTCOME MEASURES Concordant and discordant diagnosis rates. RESULTS The frozen section concordance rate for diagnoses from the aggregate group was 98.58% and the discordance rate was 1.42%, when uncorrected for deferred diagnoses. During the study period, participating institutions accessioned 1,693,331 surgical pathology cases; 90,538 of these cases were evaluated by frozen section consultation, resulting in the examination of 121,668 specimens and 148,506 frozen section blocks. The majority of the frozen section discordances occurred because of misinterpretation of the original frozen section (31.8%), presence of diagnostic tissue in permanent sections of the frozen block when the frozen section was negative (30.0%), and presence of diagnostic tissue in the portion of the specimen not sampled by the frozen section (31.4%). Of the discordant diagnoses, 67.8% had false-negative diagnoses for neoplasm. The pathology processes and anatomic sites represented in discordant diagnoses are also evaluated. CONCLUSIONS High diagnostic accuracy of frozen section consultations was demonstrated. Frozen sections are used to evaluate a variety of pathologic processes and anatomic sites.