Background: Coronavirus disease 2019. is a highly contagious disease caused by the SARSCoV-2 virus, first confirmed in Wuhan, China, in 2019. Although it primarily affects the respiratory system, it also causes damage to several other organ systems. Its rapid spread and catastrophic global consequences on health and economy forced the entire health community to fight together for a solution. Brain injury biomarkers are objective indicators of both biological and pathological processes and thus can be used for diagnostic, preventive and therapeutic purposes. Ubiquitin carboxy-terminal hydrolase L1 is a protein found in the cytoplasm of neurons. Glial fibrillary acid protein is a structural protein of astrocytes. Both of these biomarkers have been extensively investigated in traumatic brain injuries, but also in nontraumatic pathological processes, such as Parkinson's and Alzheimer's disease. Testosterone is the primary male sex hormone responsible for regulating sex differentiation. Due to the more frequent occurrence of COVID-19 in men, testosterone has been extensively investigated in patients with COVID-19. The main goal of this study was to investigate the dynamics of GFAP and UCH-L1 during the hospitalization of patients with severe and critical COVID-19 in the Intensive Care Unit. Furthermore, secondary study objectives were to investigate whether GFAP and UCH-L1 levels were higher in patients who developed neurological sequelae during the hospital stay and to investigate whether testosterone levels were associated with GFAP and UCH-L1. Participants and methods: This retrospective study was conducted in the Intensive Care Unit of the Department for Anesthesiology, Reanimatology and Intensive Care, University Hospital of Split, Croatia in the period from January to May 2022. The study included 65 male patients between the ages of 18 and 65 who were treated in the ICU for severe and critical form of COVID-19. All included patients were intubated and mechanically ventilated upon admission to the ICU. Exclusion criteria were: female sex, active malignant disease in the past year, neuromuscular diseases, previous cerebrovascular incidents, autoimmune diseases, previously diagnosed hypogonadism, heart failure, liver failure, and kidney failure. Upon admission to the ICU blood samples were taken from the patients. Peripheral venous blood samples used in this study were taken on the first, seventh and fourteenth day after admission to the ICU, except in the case of transfer from the ICU or death of the patient. Blood samples were centrifuged at 3000x g for 10 minutes and then stored at -80°C. GFAP and UCH-L1 concentrations were determined from serum using a chemiluminescent immunochemical method (CMIA). Total testosterone levels were determined using an immunochemical method. All tests were performed according to the manufacturer's instructions. Results: There were no significant differences in the serum concentrations of brain injury biomarkers UCH-L1 and GFAP during a two-week stay in the ICU in male patients with severe and critical form of COVID-19. Patients with developed neurological sequelae had longer ICU stay and were mechanically ventilated for a longer period of time (p = 0.014 and p = 0.010). A negative correlation between UCH-L1 and serum testosterone levels was demonstrated for the first time (p < 0.001). A correlation between GFAP and testosterone has not been established. Among patients who survived the severe or critical form of COVID-19, patients with developed neurological sequelae have significantly higher levels of serum UCH-L1 (p = 0.022). A significant positive correlation was found between UCH-L1 and high-sensitivity troponin (p = 0.007). Conclusions: There was no significant difference in serum levels of GFAP and UCH-L1 in patients with severe and critical form of COVID-19 treated in the ICU for two weeks, while testosterone levels showed increase. UCH-L1 levels were higher at the admission in patients with developed neurological sequelae compared to patients without, however, GFAP levels at the admission did not differ between investigated groups. Also, UCH-L1 levels were negatively correlated with testosterone levels at all three measurement points, while GFAP levels did not show a correlation with testosterone. GFAP had a positive correlation with the age. The levels of UCH-L1, GFAP and testosterone did not differ between patients who survived and patients who died. A significant positive correlation was found between UCH-L1 and high-sensitivity troponin.