Intraoperative frozen sections of specimens taken during thoracic surgery are widely seen as the gold standard. However, the accuracy of intraoperative cytology remains contentious. The study aims to estimate the value of intraoperative cytology by analyzing feasibility, accuracy, time requirements, and possible limitations when compared to standard frozen sections. To this end, we examined a total of 532 intraoperatively harvested specimens out of the 518 resected thoracic tumors from 360 patients between August 2016 and August 2017. The specimens were subject to intraoperative rapid cytology that was later counter compared to the final histology results. The mean time between the intraoperative harvesting and arrival at the laboratory was 2.23 min, and it took a further 3.5 min until the results were communicated to the surgeon. Cytologically, 218 cases (41%) were classified as malignant, 291 (55%) as benign, and 23 (4%) remained unclear. In 55 malignant cases, we observed additional benign formations. The final histological examination performed later yielded 267 malignant and 265 benign cases. Therefore, the sensitivity and specificity of rapid intraoperative cytology were 82% and 99%, respectively, with a negative/positive predictive value of 86%/99%. We conclude that the intraoperative rapid cytology is a fast, accurate, sensitive, and specific procedure for intraoperative decision making and is a distinctly helpful alternative or adjunct for the thoracic surgeon, providing that one is aware of the plausible limitations of this technique.
Keywords: Frozen section; Histological examination; Intraoperative cytology; Thoracic surgery; Tumors.
© 2021. The Author(s), under exclusive license to Springer Nature Switzerland AG.