Using imprint and frozen sections in determining the surgical strategies for thyroid pathologies

To compare the diagnosis of thyroid pathological findings by using frozen section and preoperative imprint.

From January 1998 to December 1999, one hundred and three consecutive patients admitted to Department of Surgery, Gazi University Medical Faculty for thyroid operation, were prospectively included into the study. In all patients, imprint and frozen section were carried out from their suspected nodules and the results were compared with the permanent biopsy. Those findings used to find out sensitivity, specificity, negative predictive value (NPV) and positive predictive value (PPV) of frozen section and imprint. Because of the definitive differentiation of malignant or benign lesions could not be done, three patient were excluded from the study.

In our study; sensitivity, specificity, PPV, NPV and accuracy for imprints was found as 83.3%, 97.7%, 83.3%, 97.7% and 96% (respectively), while respective corresponding data for frozen section were 81.8%, 96.6%, 75%, 97.7% and 95%. In one patient, the pathological finding of follicular carcinoma as found by permanent section was accurately diagnosed by imprint, but missed by frozen section. Both the imprint and frozen section failed to diagnose two papillary carcinoma. Overall cancer rate among 103 patients was 12 % (nine papillary carcinoma, one Hürthle cell carcinoma and two follicular carcinoma).

Both the frozen sections and imprints showed high sensitivity rate and similar accuracy rates. So, if there is no opportunity for frozen section, the imprints could be used. Furthermore, the use of imprints is less time consuming and less expensive than that of frozen sections.