Severely low testosterone in men with COVID-19: A case-control study

ABSTRACT

Introduction & Objectives: In patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection (COVID-19) reasons for sex disparity in disease severity are still unclear and circulating androgens could play a role. We investigated circulating sex steroids levels in a cohort of symptomatic patients with COVID-19 compared to a cohort of healthy men. Materials & Methods: Data of 286 patients with COVID-19 admitted to a single academic centre were compared to 305 voluntaryhealthy blood donors. Patients were further categorized according to disease severity as Group 1 mildly symptomatic and discharge home;Group 2 admitted in the internal medicine unit;Group 3 admitted to intensive care unit (ICU);and, Group 4 deceased because of COVID-19. Healthy controls were subdivided in SARS-CoV-2 negative and asymptomatic unaware SARS-CoV-2 positive. Health-related comorbidities were scored with the Charlson Comorbidity Index (CCI). Moreover, a validated composite risk score (Liang et al, 2020) was calculated to estimate the risk of developing critical illness in men with COVID-19. Hypogonadism was defined as a total testosterone (TT) level < 9.2 nmol/l. Logistic regression analysis tested the association between TT level and the risk of death due to COVID-19. Results: Overall, men with COVID-19 showed a higher burden of comorbidities than healthy controls and asymptomatic positive controls (CCI³2 in 66/286 (24%) vs. 0/281 (0%) vs. 0/24 (0%);p<0.0001). TT levels were significantly lower in patients with COVID-19 vs. asymptomatic vs. healthy controls (mean (IQR) 2.5 (1-4.7) nmol/L) vs. 11.8 (8.4-14.4) vs. 10.4 (8.1-13.4) nmol/L, respectively;p<0.0001). Of all, hypogonadism was observed in 257 (89.8%) patients, 9 (33%) asymptomatic and 42 (14.9%) healthy controls at hospital admission (p<0.0001). In as many as 243 (85%) patients, hypogonadism was secondary. Of patients, in Group 1 were 24 (4.5%), in Group 2 155 (29%), in Group 3 48 (8.9%), and in Group 4 31 (5.8%). Both Group 3 and 4 patients had significantly lower TT (1.0 (0.5,1.8) and 0.7 (0.3,2.3) nmol/L, respectively) compared to Group 2 (3.0 (1.8,5.7)) and Group 1 (3.9 (3.1,5.3) nmol/L) patients (p<0.0001). At logistic regression, a lower TT level was associated with a higher risk of death (OR 0.66;95%CI 0.45, 0.98) after accounting for the critical illness score. Of note, the lower the TT, the higher the risk of death for the same Critical-Ill COVID-19 score (Figure 1).(Figure Presented) Conclusions: We unveiled an independent association between SARS-CoV-2 infection status and hypogonadism already at hospital admission, with lower testosterone levels predicting the most severe clinical outcomes.