Thyroid Function During COVID-19 Illness and After Recovery

Endocrine Effects of the Coronavirus

Coronaviruses are known for their direct effects on endocrine glands, including the thyroid. In patients infected with the SARS-CoV virus, a relative of SARS-CoV-2, post-mortem signs of follicular and parafollicular cell damage in the thyroid have been observed. Additionally, the presence of coronaviruses has been post-mortem detected in the pituitary gland, and in patients with SARS-CoV infection, decreased staining of the pituitary gland for TSH (thyrotropin) was also observed. SARS-CoV-2 enters cells via the angiotensin-converting enzyme 2 (ACE2) receptor, which is abundantly expressed in the thyroid gland. Therefore, the hypothalamus-pituitary-thyroid axis may be very susceptible to disorders during COVID-19 illness. The study presented below aimed to clarify the extent to which thyroid dysfunction occurs in patients with COVID-19.

Methodology and Study Course

This was a cohort observational study conducted between March 9 and April 22, 2020, involving 456 patients with suspected COVID-19 illness. Blood samples were taken from the patients for standard testing: complete blood count, kidney function, albumin, C-reactive protein (CRP), cortisol, and thyroid function. The infection was confirmed by PCR testing based on a nasopharyngeal swab. Patients excluded from the study included those with a history of thyroid disease, those receiving replacement therapy, antithyroid treatment, or glucocorticoids.

Results

Among the 456 patients observed, COVID-19 was confirmed in 334 (73.2%). The average age in the infected group was 66.1 years, and in the non-infected group, it was 63.8 years. Both groups had a higher proportion of men (60.8% and 55.7%). By May 8, 40 (12%) patients from the COVID-19 group were hospitalized in intensive care units, and 95 (28.4%) had died. Patients were categorized according to fT4 (free T4) and TSH levels. The majority (86.5%) of COVID-19 patients were euthyroid, none were hyperthyroid (with low TSH and high fT4), and only a small number had overt hypothyroidism (0.6%). The characteristics did not differ significantly across all categories.

COVID-19 patients had lower TSH levels with a median of 1.03 mIU/l, while non-infected patients had a TSH median value of 1.48 mIU/l. COVID-19 patients also had a lower average fT4 level (12.6 vs. 13.11 pmol/l). In a small subset of patients (n = 55) who were re-tested after recovering from COVID-19, most results were classified as euthyroid, with only 2 patients showing lower TSH and normal fT4 (subclinical hyperthyroidism), and another 2 showing subclinical hypothyroidism (normal fT4 with slightly elevated TSH). None of the patients had overt thyrotoxicosis.

Discussion

The changes in thyroid function are most likely explained by the euthyroid sick syndrome, where initially there is a reduction in total and free T3 and an increase in reverse T3 without an increase in TSH. More severe or prolonged illness then causes a global reduction in TSH, fT4, and fT3. TSH suppression is likely associated with an elevation of pro-inflammatory cytokines, such as interleukin 6, which negatively correlate with TSH levels. Another factor may be cortisol, which suppresses TSH secretion. A third possible explanation is the direct cytopathic effect of SARS-CoV-2. Regardless of the cause of the functional changes, nearly all patients' TSH levels returned to normal following recovery from COVID-19. Thus, the functional changes are reversible post recovery.

Summary and Conclusion

This study monitored the acute effects of COVID-19 illness on thyroid function and evaluated whether the functional changes resolve after recovery. Most COVID-19 patients were euthyroid, and none suffered from thyrotoxicosis. However, COVID-19 patients had lower TSH and fT4 levels. Post-recovery and illness, thyroid function returned to normal.

(saz)

Source: Khoo B., Tan T., Clarke S. A. et al. Thyroid function before, during, and after COVID-19. J Clin Endocrinol Metab 2021 Jan 23; 106 (2): e803−e811, doi: 10.1210/clinem/dgaa830.