How to interpret thyroid biopsy results: a three-year retrospective interventional radiology experience.

Results of thyroid biopsy determine whether thyroid nodule resection is appropriate and the extent of thyroid surgery. At our institution we use 20/22-gauge core biopsy (CBx) in conjunction with fine-needle aspiration (FNA) to decrease the number of passes and improve adequacy. Occasionally, both ultrasound (US)-guided FNA and CBx yield unsatisfactory specimens. To justify clinical recommendations for these unsatisfactory thyroid biopsies, we compare rates of malignancy at surgical resection for unsatisfactory biopsy results against definitive biopsy results. We retrospectively reviewed a database of 1979 patients who had a total of 2677 FNA and 663 CBx performed by experienced interventional radiologists under US guidance from 2003 to 2006 at a tertiary-care academic center. In 451 patients who had surgery following biopsy, Fisher's exact test was used to compare surgical malignancy rates between unsatisfactory and malignant biopsy cohorts as well as between unsatisfactory and benign biopsy cohorts. We defined statistical significance at P = 0.05. We reported an overall unsatisfactory thyroid biopsy rate of 3.7% (100/2677). A statistically significant higher rate of surgically proven malignancies was found in malignant biopsy patients compared to unsatisfactory biopsy patients (P = 0.0001). The incidence of surgically proven malignancy in unsatisfactory biopsy patients was not significantly different from that in benign biopsy patients (P = 0.8625). In conclusion, an extremely low incidence of malignancy was associated with both benign and unsatisfactory thyroid biopsy results. The difference in incidence between these two groups was not statistically significant. Therefore, patients with unsatisfactory biopsy specimens can be reassured and counseled accordingly.