Accuracy of intraoperative frozen section diagnosis in head and neck surgery: experience at a university medical center.

We performed 2,210 intraoperative frozen sections on 258 patients from the Otolaryngology-Head and Neck Surgery Service in 1990 and 1991. Surgery involved a wide variety of benign and malignant lesions. Techniques included biopsies for diagnosis, simple excision, thyroid and salivary gland surgeries, lymph node biopsies, composite resections with radical neck dissections, laryngectomies, and skull base surgeries. During the 2-year period, 1,947 (88.1%) sections were requested for evaluation of surgical margins, 258 (11.7%) for diagnosis, and five (0.2%) cases for tissue identification. There was disagreement between the frozen section and permanent section in 46 (2.1%) cases, and the deferral diagnosis rate was 0.8%. Disagreements were the result of sampling errors in 33 and interpretive errors in 13 cases. There were six (0.3%) false-negative diagnosis of malignancy and four (0.2%) false-positive diagnoses of malignancy. Three of these had an impact on patient care as discussed. We recommend careful sampling and sectioning of small biopsies and the need for vigilant communication between surgeon and pathologist.