Infectious, inflammatory, and clinical outcomes among patients prescribed trifluoperazine or amoxapine: a retrospective cohort study using real-world data.

INTRODUCTION: The rise of antimicrobial resistance and limited development of new antibiotics have increased interest in drug repurposing. Preclinical evidence suggests that certain psychotropic drugs, including trifluoperazine (TFP) and amoxapine (AXPN), possess antimicrobial or immunomodulatory properties. However, their effects on infection risk in clinical populations are largely unknown. This study evaluated whether TFP or AXPN use is associated with altered risks of infectious illnesses, systemic inflammation, and mortality compared to patients not taking antidepressants. METHODS: This retrospective cohort study utilized TriNetX, a federated network of de-identified electronic health records from over 75 million patients. Patients prescribed TFP, AXPN, fluoxetine, or no antidepressants/antipsychotics (general control) were assigned to mutually exclusive cohorts. Age- and sex-matched comparisons were conducted for four groups. Assessed outcomes included bacterial and viral infections (e.g., Clostridioides difficile), severe acute respiratory syndrome coronavirus 2, an etiological agent of COVID-19, as well as for pneumonia and sepsis, inflammatory markers (C-Reactive Protein, Erythrocyte Sedimentation Rate, ferritin, and procalcitonin), Intensive Care Unit admission, and mortality. Risk ratios (RR), odds ratios (OR), 95% confidence intervals, and Kaplan-Meier survival analyses were performed for significant findings. RESULTS: Overall, 2,177 TFP and 834 AXPN patients met the inclusion criteria. TFP was associated with significantly lower rates of ENT/dental infections across all three age groups compared with general controls (RR 0.316-0.405, all p ≤ 0.002). It was also consistently associated with increased leukocytosis across all age strata. Reduced COVID-19 incidence was observed primarily in older adults (70-90 years) versus general controls. AXPN showed fewer associations overall, with a significant reduction in C. difficile infection (RR 0.458, p = 0.025) and lower leukocytosis (RR 0.685, p = 0.012) in the 70-90 age group compared with fluoxetine. Most other outcomes, including pneumonia, sepsis, and other inflammatory markers, did not differ significantly when compared to general control or fluoxetine groups. DISCUSSION: These exploratory findings identify potential signals linking TFP and AXPN to infection susceptibility. Given the study design, these observations are hypothesis-generating and require confirmation in future prospective studies.