Thyroid nodules are common clinical findings, with a prevalence reaching 70%. Most thyroid nodules are benign, while the incidence of malignancy is around 5% (11). Multiple clinical, pathological, laboratory and radiological factors were proposed to differentiate benign from malignant nodules; however, the reported predictors are inconclusive.
TSH is a major regulator of thyroid function. Multiple studies have evaluated it as a predictor of malignancy in thyroid nodules with conflicting results (6–10, 12). Most studies showed that a higher serum concentration of TSH is associated with an increased risk of thyroid cancer. However, studies investigating this relationship in euthyroid nodules are limited (13). In the present study, the TSH level was higher in malignant euthyroid nodules compared to benign euthyroid nodules. TSH levels within the higher end of the reference range increased the likelihood of thyroid malignancy by 1.54 times. Baser et al. evaluated 1433 patients who underwent thyroidectomy and reported that malignant thyroid nodules had elevated TSH levels compared to the benign group (p < 0.001) (13). The TSH level threshold above which the risk of thyroid malignancy increases was 1 mIU/L (14). Conflicting these studies, Castro et al. reported that serum TSH level was not associated with increased malignancy risk among 327 thyroid lesions suspicious for follicular neoplasm or Hürthle cell neoplasms (12). Singh et al. found no association between Hashimoto's thyroiditis, which has high TSH levels, and the incidence of malignancy in a meta-analysis conducted in 1999 (15). Moreover, Holm et al. found that patients with uncontrolled hypothyroidism had no increased risk of thyroid malignancy after two decades of follow-up (16).
It is unclear whether higher TSH levels increase the malignant potential of the thyroid nodules or whether the malignant nodules produce higher levels of TSH. It is suggested that TSH modulates the thyroid cell function, growth, and expression of specific proteins after binding to its receptors of the follicular cells. This leads to the continuous stimulation and proliferation of the thyroid nodules by high TSH levels, which results in cellular alternations and modulation of thyroidal gene expression (17, 18). Animal experiments on mice and golden hamsters showed that TSH stimulation is associated with thyroid malignancy, as overstimulation of TSH leads to hyperplasia and, eventually, thyroid cancer (6). Moreover, this hypothesis is supported by the fact that patients with well-differentiated thyroid cancer benefit from TSH suppressive treatment with levothyroxine as it decreases disease progression, recurrence rates, and cancer-related mortality (19). On the other hand, recent genetic studies aimed to search for sequence variants that link TSH with thyroid malignancy. Two variants, located on 9q22.33 and 14q13.3, were associated with thyroid cancer; both have also been associated with low serum TSH (20). Moreover, an inverse relationship between TSH receptor mRNA and cancer progression was documented by Shi et al (21).
TSH level was measured for each Bethesda category separately in the present study. TSH level was significantly correlated with the Bethesda system category. The higher the Bethesda category, the higher the TSH level was, and the highest levels were found in the malignant categories (V-VI). Similarly, Baser et al. reported that TSH levels increased gradually as the Bethesda category did, rising from Bethesda category II to VI (13).
Malignant thyroid nodules are smaller in size than benign in the present study. Moreover, our study showed the large nodule size (> 4cm) decreased the possibility of thyroid cancer by 24%. Our findings correlate with Castro et al. and Baser et al., who reported that smaller nodules were more frequently seen in patients with malignancy (12, 13). On the other hand, Sahin et al. and Chung et al. found no association between the size of the thyroid nodule and its malignant potential (22, 23).
In addition to TSH, low FT4 levels were reported as an independent predictor of thyroid malignancy in the literature (24). Our study had no significant association between FT4 levels and thyroid malignancy. Extremes of age and the male gender were also reported to increase the risk of thyroid malignancy (3). However, we did not find a significant correlation between age and gender with thyroid malignancy. A meta-analysis conducted by Zhao et al. found that obesity is significantly associated with an increased risk of thyroid cancer (25). In our study, the univariate analysis showed a significant association between high BMI and the risk of thyroid malignancy; however, the multivariate analysis did not support this finding.
Our study has a few limitations, such as being a retrospective single-center study which may limit its generalizability. Moreover, it only included patients who underwent thyroidectomies. Patients with Bethesda categories I-III who did not have other indications for surgical intervention were not included in the study. However, this is the first study to evaluate the association between TSH levels and thyroid malignancy in SA.