Background
Hashimoto thyroiditis is the most common cause of hypothyroidism in the United States [
1]. With a 5-10 time preference over men, the reported prevalence in white women is in the 1-2% range [
2].
Etiology and pathogenesis of Hashimoto thyroiditis are still elusive. Moreover, little is known about progression of euthyroid to hypothyroid Hashimoto's. At least in children, disease progression from euthyroid to hypothyroid Hashimoto thyroiditis has been suggested [
3]. Also, available evidence relating the progression of sub-clinical to overt hypothyroidism in adults has been rated as good [
4]. Hence, it is conceivable that a euthyroid stage of Hashimoto thyroiditis exists and that progression to a full-blown disease stage is a matter of time.
Since there is growing evidence that unrecognized hypothyroidism is deleterious, early diagnosis of Hashimoto thyroiditis would be advantageous in predicting thyroid failure. Specifically, it is well known that maternal thyroid status assessment and treatment improves fetal outcomes and neuropsychological developmental of the newborn [
5].
The University of Wisconsin Thyroid Multidisciplinary Clinic is a large referral site for thyroid diseases in the Midwest. A continuously increasing number of thyroid biopsies are being performed every year for cancer screening. Yet, Hashimoto thyroiditis is being too frequently diagnosed. The prevalence of Hashimoto thyroiditis is reported to be approximately twice that of type 1 diabetes. However, the prevalence of Hashimoto thyroiditis confirmed by cytology has never been documented in a large cohort of patients with ultrasound detectable nodules. To evaluate different aspects of thyroid physiopathology, a database of clinical features, ultrasound images and cytology results of patients referred for fine needle aspiration of thyroid nodules was prospectively developed. In this paper we probed our database for the frequency and characteristic of patients diagnosed with Hashimoto thyroiditis while being referred for thyroid cancer screening.
Discussion
This is the first study to show such a high prevalence of Hashimoto thyroiditis diagnosed by ultrasound-guided FNA cytology on a somewhat large cohort of patients. Based on our study, the prevalence of cytology-proven Hashimoto thyroiditis appears to be >10% in patients with thyroid nodules. Given that the prevalence of thyroid nodules on ultrasonography or autopsy data is as high as 50% [
7], such a high prevalence of Hashimoto thyroiditis diagnosed by cytology is noteworthy. More strikingly, the prevalence of euthyroid, non-previously diagnosed, cytology-proven Hashimoto thyroiditis (euthyroid autoimmunity), appears to be >5% in our study (Figure
1 and
2). This condition has not been previously defined. For comparison, its prevalence is similar to that of type 2 diabetes which is considered to be a health care crisis.
Weetman [
2] reported clinical Hashimoto thyroiditis prevalence rate at 1 in 182 or 0.55% in the US. In the UK, Tunbridge et al [
8] reported an overall Hashimoto thyroiditis prevalence of 0.8%. However, based on our study, the cytology of Hashimoto thyroiditis seems to be much more prevalent, at 13.4%. This difference may be partially explained by the fact that for diagnosing clinical Hashimoto thyroiditis, abnormally elevated TSH, low thyroid hormones [
2,
8] and the confirmatory presence of thyroid autoantibodies are usually accounted for. We hypothesized then, that cytological diagnosis of Hashimoto's may precede clinical diagnosis. Interestingly however, in most organ specific autoimmune diseases, humoral immunity heralds tissue infiltrative damage. Hence, we expected the diagnosis of Hashimoto's by cytology to be less but not more common than the antibody diagnosis especially on early stages of the disease (Figure
2).
In an attempt at differentiating better the apparent stages of Hashimoto thyroiditis, we divided our cytology-proven Hashimoto's cohort into three subgroups: clinically hypothyroid, sub-clinical and euthyroid. For these distinctions, we used the clinical definition of hypothyroid Hashimoto thyroiditis as that occurring in patients usually with TSH greater than 10 ng/dl at diagnosis, with low free thyroid hormones and on thyroid hormone replacement. Based on NHANES III study [
9] normative data for TSH distribution (reference population of 13,344, 95% of TSH between 0.45-4.12 ng/dl), abnormally high TSH (but usually less than 10 ng/dl) with normal thyroid hormones was used to define sub-clinical hypothyroidism. Patients with those characteristics were assigned to the sub-clinical subgroup in our study. A third subgroup classified as euthyroid included patients with normal thyroid hormones and normal TSH.
Aside from the unexpected observation of the high prevalence of Hashimoto thyroiditis by cytology, especially in euthyroid patients, the lack of cytological correlation with TPO autoantibody positivity is noteworthy. As mentioned before, the hallmark in the diagnosis of Hashimoto thyroiditis is the presence of TPO autoantibodies [
10]. Yet, only about half of the patients tested for anti-TPO in the
euthyroid subgroup were positive. On the other hand and although non-statistically significant because of the small number reported, the clinically hypothyroid subgroup had most (13 out of 16 tested) patients positive for anti-TPO. Similarly, Poropatich et al., [
11] found that anti-TPO and/or antithyroglobulin antibody titers were present in only 50% of the patients with euthyroid, cytology-proven Hashimoto thyroiditis, a finding never reproduced by these or other authors in the literature.
Given the wide range of normal values for TSH (1 fold) and the variability on the presence of TPO autoantibodies, it is conceivable that early Hashimoto's autoimmune process might be clinically missed. These issues, together with the awareness that sub-clinical and clinical hypothyroidism associates with cardiovascular and neuropsychiatric morbidities, make finding high prevalence of Hashimoto thyroiditis on cytology, especially in
euthyroid patients clinically significant [
12‐
14].
An important aspect of our findings is the fact that most of the patients in our cohort are pre-menopausal females. Past and more recent studies found that the risk of poor obstetrical outcome is increased with relative thyroxine deficiency [
15‐
18]. Benhadi et al. recently showed that the risk of miscarriage, fetal and neonatal death is increased with higher TSH levels. Moreover, the risk of fetal loss occurred even when maternal free thyroxine levels were normal [
16]. Based on our results of high prevalence of cytology-proven Hashimoto thyroiditis and especially high prevalence of
euthyroid disease in pre-menopausal women, there might be need of at least follow up of thyroid function on cytologically-proven Hashimoto's in pre-conception through delivery stages.
Positive Hashimoto's cytology clearly represents a state in which inflammation of thyroid epithelium may lead to tissue remodeling. Hence the possibility of acquiring mutations while cells are dividing might be greater. Some studies have suggested that inflammation of thyrocytes could be linked to thyroid cancer [
19‐
21]. For example, Larson et al. showed that phosphatidylinositol 3-kinase phosphorilates Akt, which in turn suppresses pro-apoptotic signals and promotes tumorigenesis, was increased in both Hashimoto thyroiditis and well differentiated thyroid cancer [
19]. Furthermore, Borrello et al., showed that RET/PTC oncogene, when exogenously expressed in normal thyrocytes, induces expression of a large set of genes involved in inflammation and tumor invasion (cytokines, matrix-degrading enzymes and adhesion molecules) [
20]. There seem to be however, controversial opinions regarding the association of Hashimoto thyroiditis and thyroid cancer. Some studies find Hashimoto thyroiditis as associated with thyroid cancer [
22,
23] as examples, while others do not report an association [
24,
25] as examples. Since higher TSH level in patients with thyroid nodules has been found to be associated with risk of differentiated thyroid cancer [
26,
27], we hypothesized that active remodeling of thyroid epithelium in cytological Hashimoto's may explain in part the increased risk for differentiated thyroid cancer observed in patients with
elevated but within normal TSH [
26,
27].
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
All authors participated in the design of the study. AS and JCJ performed the statistical analysis. JCJ conceived of the study. All authors read and approved the final manuscript.