Time to Antibiotics and Mortality in Immunocompromised versus Non-Immunocompromised Patients with Suspected Sepsis.

Abstract

Timely antibiotics are key determinants of sepsis survival and are presumed to be especially critical in immunocompromised patients. However, evidence supporting this assumption is limited. We identified all adults treated for suspected sepsis in the emergency departments of nine U.S. hospitals, 2015-2024. We identified immunocompromised patients using diagnosis codes, supplemented with clinical data to define a severely immunocompromised subgroup. We used multivariable logistic regression to assess associations between time-to-antibiotics (primary analysis: 1-3 vs 0-1h; secondary analysis: 3-6 vs 0-3h) and in-hospital mortality, stratified by immune status and sepsis severity. Among 39,842 hospitalizations with suspected sepsis, 20,721 occurred in non-immunocompromised patients and 19,121 in immunocompromised (2,283 severe). Overall, antibiotic administration at 1-3 vs 0-1h was associated with increased mortality in non-immunocompromised (OR 1.33, 95% CI 1.12-1.59) but not immunocompromised patients (OR 1.08, 95% CI 0.94-1.25). Increased risk was limited to septic shock, where delayed antibiotics were associated with higher mortality in both non-immunocompromised (OR 1.41, 95% CI 1.12-1.76) and immunocompromised patients (OR 1.21, 95% CI 1.002-1.47). In contrast, no association was observed in sepsis without shock regardless of immune status, including for antibiotic administrations at 3-6 vs 0-3h. Effect estimates were similar for mild-moderate and severe immunocompromise. Short delays in antibiotic administration were associated with increased mortality in septic shock but not in sepsis without shock, with no evidence of greater vulnerability among immunocompromised patients. These findings suggest antibiotic urgency should be guided primarily by clinical severity rather than immune competence.