Frozen-Permanent Section Discrepancy Rate in Oral Cavity and Oropharyngeal Squamous Cell Carcinoma

Abstract

Frozen section evaluation of head and neck squamous cell carcinoma (SCC) is critical for margin status and subsequent patient therapy. In this study, we retrospectively reviewed the rate of frozen-permanent section discrepancies in blocks with two frozen section levels compared to ≥ three levels in oral cavity and oropharyngeal SCCs. A search of the cases with both intraoperative frozen sections and corresponding permanent sections for SCCs in the oral cavity and oropharynx was performed. Frozen sections and permanent slides were compared. The nature of discrepancies was assigned to one of the following: change in diagnosis, margin status, or distance of the tumor from the margin. The cause of the discrepancy was designated as one of the following: block sampling, gross sampling, interpretation, or technical error. The pathologist experience, frozen section technical experience, and intraoperative impact of each discrepancy were also evaluated. A total of 654 frozen and corresponding permanent blocks were assessed. For 532 of the frozen section blocks, two levels were cut, while 122 frozen section blocks had ≥ three levels. Thirty-five frozen-permanent section discrepancies were observed (5.4% of all blocks). Among these, 2.5% had a possible or definitive intraoperative impact. The percentage of discrepancies in the ≥ three levels group (5.7%) was slightly higher than the two-level group (5.3%), and this difference was not statistically significant. For the two-level group, the overall block sampling error rate was 4.5%. This was not significantly different from the 4.1% block sampling error rate seen in the ≥ three levels group. The rate of block sampling discrepancy did not show significant differences based on attending or frozen section technical experience. A change in margin distance (closer margin detected on permanent) occurred in 4% of the blocks and involved 16% of the patients. This review of oral cavity and oropharynx SCCs frozen/permanent section discrepancies shows that the error rate is not significantly different depending on the number of levels cut. The results suggest that always performing more than two frozen section levels may not yield a decreased discrepancy rate. A change in margin distance occurred quite frequently, but only in rare cases it had a definitive impact on the intraoperative management. Given the importance of correct intraoperative diagnosis in patient management, additional levels may be warranted depending on the clinical scenario.