Disclosure(s): No relevant financial relationship(s) to disclose.
First Author: Silvana Ellen Ribeiro Papp, n/a – University of Pittsburgh Co-Author: Rafael H. Melo, n/a Co-Author: Gabriela Bezerra de Freitas Diniz, MD – MD, Hospital Israelita Albert Einstein Co-Author: Amanda Pascoal Valle Felicio, N/A – MD, Hospital Israelita Albert Einstein Co-Author: Gokul Karthikeyan, BS – Medical Student, Drexel University
Introduction: The use of corticosteroids is recommended in the management of severe pneumonia in critically ill non-cancer patients; however, their role in oncologic populations remains uncertain. This study aimed to evaluate the association between corticosteroid use and short-term outcomes in critically ill cancer patients with pneumonia using a target trial emulation approach.
Methods: We conducted a retrospective target trial emulation using the TriNetX Global Research Network. Adults (18 - 80 years) with cancer and a diagnosis of pneumonia, requiring intensive care interventions (e.g., vasopressors, mechanical ventilation), were included. Patients with asthma, COPD, or COVID-19 were excluded. The intervention group received systemic corticosteroids (dexamethasone, hydrocortisone, methylprednisolone, prednisolone, or prednisone) within 1 month after pneumonia diagnosis; the control group did not. Propensity score matching (1:1) was performed. The primary outcome was 30-day all-cause mortality. Secondary outcomes included mechanical ventilation (MV) for 24–96h, intubation and hemodialysis.
Results: After propensity score matching, 25,362 patients were included in each cohort. The use of corticosteroids was associated with a higher 30-day mortality rate (17.8% vs. 13.6%; RR 1.31; 95% CI: 1.26–1.37; HR 1.33; 95% CI: 1.27–1.39; p< 0.001). The need for mechanical ventilation between 24 - 96 hours was also more frequent in the corticosteroid group (2.8% vs. 1.6%; RR 1.73; p< 0.001). Intubation occurred in 6.3% of patients receiving corticosteroids compared to 3.3% in the control group (RR 1.92; p< 0.001). Additionally, hemodialysis was slightly more common among corticosteroid users (0.2% vs. 0.1%; p=0.050).
Conclusions: In this large, real-world target trial emulation, corticosteroid use was associated with increased 30-day mortality and greater need for respiratory support among critically ill cancer patients with pneumonia.
