1Humanitas Clinical and Research Center; [email protected]; 2IRCCS, Humanitas Clinical and Research Center, Rozzano, Milan, Italy;
Objective: To evaluate the association between testosterone (T) values and clinical outcome of COVID-19.
Methods: This is a retrospective single-center study performed on 221 male patients (median age 70.5 years, range 26-93) with COVID-19 and hospitalized in our Institution between November 1st 2020 and January 31st 2021. Patients height, weight, BMI, medical history and comorbidities were recorded; at admission, patients underwent sampling for complete blood count, inflammatory markers (C-reactive protein, Ferritin, IL-6, LDH) and gonadal hormones status (LH, T). Depending on T levels, subjects were stratified in 3 groups: normal (T > 12 nmol/l; 17 cases), borderline (T 8-12 nmol/l; 28 cases) and low (T<8nmol/l; 176 cases); central hypogonadism was defined in subjects with low T and LH below 9.4 mIU/ml (112 cases). Acute respiratory insufficiency (ARI) was defined by P/F ratio below 300 (arterial partial O2 pressure /fraction of inspired oxygen ratio) at hospital admission (Berlin definition).
Results: Compared to subjects with higher T values, patients in the low T group were significantly older (P=0.001) and had higher IL-6 (P=0.001), C-reactive protein (P<0.001), LDH (P<0.001), ferritin (P=0.012), lower P/F ratio (P=0.001) with increased prevalence of ARI (P<0.001) and mortality rate (P=0.009). When central and primary hypogonadism were compared, the latter group was found to be significantly older (P<0.001) and pluricomorbid (P=0.015), with higher mortality rates (P=0.025). In the multivariate regression analyses, ARI and in-hospital mortality significantly were associated with hypogonadism (OR 4.56, CI 1.22-17.06; P=0.02) and testosterone values (OR 0.74, CI 0.63-0.88; p 0.001), respectively, independently of age, comorbidities and inflammation.
Conclusions: This study provides convincing evidence that testosterone levels predict a negative outcome of SARS-CoV2-related pneumonia.