PubMed1989
In 203 consecutive gynecological operations where frozen sections were performed, 35.6% were from conditions of the ovary, 22.7% from the cervix, 18.2% from the endometrium, and 11.4% from the vulva. There were 0.5% false-positive, 1.0% false-negative and 2.0% deferred diagnosis. Incorrect interpretation was the cause of the single false-positive diagnosis, while the false-negative diagnoses were due to errors in block selection. The deferred diagnoses mainly occurred in gynecological conditions where diagnosis was difficult, required extensive sampling or a formal mitotic count. As in other surgical fields, gynecological frozen sections were used principally to guide the extent of surgery. The most valuable frozen sections were in those instances where the operation was affected most. These were on lymph nodes in cases of carcinoma of the vulva and cervix, myometrial lesions in young women where myomectomy was being considered, and ovarian tumours to distinguish primary from secondary tumours. Occasionally, frozen sections were also found useful to establish margins of vulval and cervical tumours, in hysterectomy specimens of endometrial carcinomas to determine prognostic factors, and in suspected recurrences and metastases of tumours to determine the adequacy of the biopsy material. Frozen sections in obviously benign conditions, e.g., ovarian cysts without papillary or solid areas, were found to be unnecessary. Frozen sections are contraindicated when only a small amount of crucial material is available, as the paraffin diagnosis may be compromised. Pathologists should have a clear idea of the role of frozen sections in gynecological surgery and work closely with the surgeon in the management of gynecological oncology patients.