Purpose To retrospectively describe the association between thyroid hormones (TH) and platelet activation, as represented by mean platelet volume (MPV), in a cohort of patients hospitalized for COVID-19 with no known thyroid disease, and to correlate these data with the severity of COVID-19 and the occurrence of death/ARDS (Acute Respiratory Distress Syndrome). Methods 103 patients with real-time polymerase chain reaction (RT-PCR) testing-confirmed COVID-19 and hospitalized were enrolled. Serum samples were collected from patients upon admission before starting any treatment. Chi-squared test was used to determine the association between euthyroid sick syndrome (ESS) and COVID-19 severity. Multivariate logistic regression was performed to evaluate the best independent predictors of COVID-19 deaths/ARDS. Results 39/103 (37.9%) of patients were found to have ESS, and this condition was an independent predictor for the severity of COVID-19 (p = 0.003). Lower TSH and lower FT3/FT4 ratio correlated with higher MPV (p = 0,001 and p = 0.010), with an opposite trend with respect to what has been documented in non-COVID patients. Increasing MPV and lower FT3 significantly increased the risk, in COVID-19 patients, of an adverse outcome of death/ARDS. Conclusion Increased platelet activation, as represented by increased MPV, has already been reported to correlate with COVID-19 severity, possibly as a consequence of cytokine release. We demonstrated, in a cohort of 103 patients with COVID-19, that MPV is inversely correlated to TH levels, in particular in the case of ESS, where downregulation of TH axis may occur in case of systemic cytokine inflammation and more severe outcomes (death/ARDS). That ESS itself may directly cause platelet activation, as demonstrated by higher MPV in these patients, is an interesting hypothesis which deserves further investigation.