Response to Dr. Paul Drinka.

Abstract

We thank Dr. Drinka for his letter1 in response to our recent article2 and welcome the opportunity to provide additional details. There were no differences in interleukin (IL)-6 or tumor necrosis factor alpha (TNF-α) levels between euthyroid participants and individuals with subclinical hyperthyroidism (for IL-6: median 1.45 pg/mL, interquartile range (IQR) 0.85–2.24 pg/mL vs median 1.53 pg/mL, IQR 0.83–2.15 pg/mL; log-transformed age- and sex-adjusted P = .85; for TNF-α: median 4.64 pg/mL, IQR 3.34–6.23 pg/mL vs median 4.76 pg/mL, IQR 3.24–5.66 pg/mL; log-transformed age- and sex-adjusted P = .37). In Model 2 of the multivariate analysis,2 subclinical hyperthyroidism remained a significant independent risk factor for all-cause mortality when data were adjusted for log IL-6 (hazard ratio (HR) = 1.75, 95% confidence interval (CI) = 1.13–2.73, P = .01) or log TNF-α (HR = 1.6, 95% CI = 1.02–2.51, P = .04). We agree on the difficulty of correctly diagnosing subclinical hyperthyroidism based on a single result of low circulating levels of thyrotropin with normal thyroid hormone concentrations, because numerous conditions that characterize the so-called nonthyroidal illness, along with some pharmacological agents, may lead to low circulating thyrotropin.3 We also agree that repeating a thyroid function test would help in the diagnosis. As we stated in the comment of our article, only baseline thyroid function tests determined on a single blood withdrawal were available in our study. This could be a limitation of our results, which many epidemiological studies share, but the fact that IL-6 and TNF-α were comparable between euthyroid subjects and those with subclinical hyperthyroidism and the exclusion of subjects with low circulating concentrations of free triiodothyronine (T3) in our analysis further suggest that a nonthyroidal illness is unlikely to be the explanation for low thyrotropin levels in our subjects. Because inflammatory markers are not routinely measured in clinical practice, they cannot be considered as common tools for excluding the effect of a nonthyroidal illness in the diagnosis of subclinical hyperthyroidism. We rather believe that an accurate history of the patients, aimed at the identification of any underlying thyroid disease and potential clinical conditions (or therapeutic agents) that may result in low thyrotropin, be mandatory before a diagnosis of subclinical hyperthyroidism is made. In some cases, a further thyroid function test and measurements of free T3 circulating concentrations may help. On the other hand, a correct diagnosis of subclinical hyperthyroidism should not be missed, because this thyroid disorder may be associated with other negative clinical outcomes,4 and its treatment has been advocated, especially in elderly subjects.5 Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. Author Contributions: The authors contributed equally to the manuscript. Sponsor's Role: There was no sponsor.