INTRODUCTION: During the coronavirus disease 2019 (COVID-19) pandemic, low- and middle- income countries had less access to monoclonal antibodies, such as tocilizumab (TCZ), compared to high-income countries. This retrospective cohort study aimed at evaluating the impact of a delayed TCZ administration on patient outcomes, and at determining the optimum timing of TCZ initiation for COVID-19 pneumonia in Serbia. METHODOLOGY: The study included 150 patients who received TCZ at a tertiary referral center. The outcomes analyzed in this study were the need for an intensive care unit (ICU) treatment and mortality. RESULTS: The multiple Cox proportional hazard model suggested that the delay in TCZ administration was an independent predictor of needing ICU treatment and mortality. The receiver operating characteristic (ROC) curve showed that patients who received TCZ after 7.5 days since the onset of symptoms had 74.4% higher chances of needing ICU treatment. Receiving TCZ after 9.5 days since the onset of symptoms, increased the chances of mortality by 78.9%. The multiple Cox proportional hazard model suggested that TCZ administration after 7.5 days since the onset of symptoms increased the hazard for ICU admission by 24.5%
and the hazard of mortality increased by 46.1% after 9.5 days since the onset of symptoms. CONCLUSIONS: This study emphasizes the importance of timely administration of TCZ in COVID-19 pneumonia. Better outcomes were observed when TCZ was administered up to 7.5 days since the onset of symptoms.