Serum testosterone mirrors inflammation parameters in females hospitalized with COVID-19

While low testosterone (T) was described as a predictor of unfavorable coronavirus-disease 19 (COVID-19) outcome in men, data concerning the role of T in women with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection are scant and limited to small cohorts. This study investigated the relationship between serum T values and outcomes of COVID-19 in a large female hospitalized cohort.

One-hundred-sixty-eight adult women (median age 77, range 18-100 years; 154 in post-menopause) hospitalized for COVID-19 were assessed for PaO2/Fio2 ratio, serum T and inflammatory parameters.

Median duration for hospital stay was 14.2 days (range 1-115) with overall mortality of 26% (n = 44). Subjects who died were significantly older (p < 0.001), had significantly more comorbidities (p = 0.015) and higher serum T (p = 0.040), white blood cells (p = 0.007), c-reactive protein (CRP; p < 0.001), interleukin-6 (IL-6; p < 0.001), procalcitonin (PCT; p < 0.001), lactate dehydrogenase (LDH; p = 0.001), D-dimer (p = 0.035), fibrinogen (p = 0.038) and lower serum free-triiodothyronine (FT3; p < 0.001) and luteinizing hormone (LH; p = 0.024) values. In post-menopausal women, significant associations were observed between T levels and serum CRP (rho: 0.23; p = 0.002), IL-6 (rho: 0.41; p < 0.001), LDH (rho: 0.34; p < 0.001), D-Dimer (rho: 0.21; p = 0.008), PCT (rho: 0.26; p = 0.001) and HDL cholesterol (rho:  - 0,22, p = 0.008). In multivariate regression analyses, serum T maintained the significant association with mortality after correction for age, coexistent comorbidities and serum LH and FT3, whereas it was lost after correction for inflammatory parameters.

In females, high serum T levels might be a mirror of inflammatory phenotype and worse COVID-19 course.