Background: Thyroid nodules are a common finding in clinical practice. Among classic risk factors for thyroid cancer, thyroid scintigraphy has traditionally been attributed prognostic value, with cold nodules implying greater risk. However, research supporting this assumption is of considerable age and possibly influenced by selection bias. In this study, we aimed to calculate the risk of malignancy in cold and hot nodules.
Material and Methods: All thyroid nodules that underwent both thyroid scintigraphy and pathologic characterisation (cytology and/or histology) in a 5-year period at a tertiary centre were retrospectively analysed. Cancer rates were calculated in cold and hot nodules. Furthermore, rates of malignancy were calculated taking into account several established and more controversial risk factors for thyroid carcinoma, in order to identify subgroups with greater risk for cancer.
Results: 343 thyroid nodules were included for study. Cancer rates were 7.7% in cold nodules (N=248) and 5.3% in hot nodules (N=95). Thyrotropin levels were lower in hot nodules (P=0.000), and levels were higher in cancerous cold nodules compared with benign cold nodules (P=0.014). A cancer rate of 26.7% was noted in cold nodules with elevated anti-thyroglobulin levels. In all other subgroup analyses, the rate of malignancy in cold nodules was never higher than cancer rates suggested by literature for nodules in the general population. Although similar observations were made for hot nodules, no definite conclusions were drawn as there were too few hot nodules to perform statistical tests.
Conclusion: Our findings suggest that cold nodules are not high-risk nodules by definition, as their cancer rates were not notably higher than the risk of malignancy proposed in literature for nodules in the general population. Therefore, we discourage the use of thyroid scintigraphy for the selection of cold nodules for further pathologic characterisation.