K. N. Edwards1, N. Bhutiani1, A. R. Quillo1 1University Of Louisville,Department Of Survery, Division Of Surgical Oncology,Louisville, KY, USA
Introduction: Graves’ Disease is the most common autoimmune disorder in the United States and is the underlying cause of 50-80% of cases of hyperthyroidism. To date, there has been no true consensus as to the risk thyroid nodules pose for thyroid cancer in patients with Graves’ disease. This study aimed to assess the prevalence of thyroid carcinoma in patients with Graves’ disease who underwent thyroidectomy at our institution and identify possible predictive factors of thyroid carcinoma in those with Graves’.
Methods: Patients undergoing thyroidectomy at our institution between 2011 and 2016 who enrolled in our prospectively collected research database were identified and classified according to whether they carried a diagnosis of Graves’ disease and then whether malignancy was present on final pathology. They were then compared along demographic parameters, results of pre-operative ultrasound and fine needle aspirate (FNA) findings, and final pathology. Two-tailed student t-tests were performed to determine the significance of any noted differences between comparison groups.
Results: A total of 363 patients met inclusion criteria. Of these, 113 patients underwent thyroidectomy for Graves’ disease, and 250 for other reasons. Overall, the incidence of thyroid cancer in these two groups did not differ significantly (p=0.16). Among patients with Graves disease, 16 (14%) had evidence of malignancy on final pathology while 97 had benign findings. However, in patients with both Graves’ disease and the presence of thyroid nodules on preoperative ultrasound, 31% of patients were determined to have thyroid cancer upon final pathology evaluation. In addition, patients with Graves’ disease were less likely to have undergone pre-operative FNA than patients without Graves’ (p<0.0001). Finally, among patients who underwent ultrasound, those with Graves’ disease and cancer who were more likely to have nodules than those who had Graves’ disease alone (75% vs. 42%, p=0.016).
Conclusion: Patients with Graves’ disease who had evidence of malignancy on final pathology are more likely to have nodules visible on ultrasound than those without any evidence of malignancy. A more frequent use of FNA in patients with Graves’ disease undergoing thyroidectomy who have nodules visible on ultrasound should be considered given the correlation between presence of nodules and malignancy to potentially optimize decisions regarding extent of surgical intervention. In addition, the overall rate of malignancy in patients with both Graves’ disease and nodules in our database was 31%, higher than the overall risk of malignancy in multinodular goiter in the general public. For this reason, surgical treatment over other treatment options for Graves’ should be strongly considered in all patients with both Graves’ disease and nodules.