IL-6 Inhibition in Critically Ill COVID-19 Patients Is Associated With Increased Secondary Infections

Abstract

BACKGROUND: Anti-inflammatory therapies such as IL-6 inhibition have been proposed for COVID-19 in a vacuum of evidence-based treatment. However, abrogating the inflammatory response in infectious diseases may impair a desired host response and predispose to secondary infections. METHODS: We retrospectively reviewed the medical record of critically ill COVID-19 patients during an 8-week span and compared the prevalence of secondary infection and outcomes in patients who did and did not receive tocilizumab. Additionally, we included representative histopathologic post-mortem findings from several COVID-19 cases that underwent autopsy at our institution. RESULTS: 111 patients were identified, of which 54 had received tocilizumab while 57 had not. Receiving tocilizumab was associated with a higher risk of secondary bacterial (48.1% vs. 28.1%, p=0.029 and fungal (5.6% vs. 0%, p=0.112) infections. Consistent with higher number of infections, patients who received tocilizumab had higher mortality (35.2% vs. 19.3%, p=0.020). Seven cases underwent autopsy. In 3 cases who received tocilizumab, there was evidence of pneumonia on pathology. Of the 4 cases that had not been given tocilizumab, 2 showed evidence of aspiration pneumonia and 2 exhibited diffuse alveolar damage. CONCLUSIONS: Experimental therapies are currently being applied to COVID-19 outside of clinical trials. Anti-inflammatory therapies such as anti-IL-6 therapy have the potential to impair viral clearance, predispose to secondary infection, and cause harm. We seek to raise physician awareness of these issues and highlight the need to better understand the immune response in COVID-19.

Cited by papers we hold 2

• Mucosal immune response in biology, disease prevention and treatment · 2025
  No notice
• The association of smoking status with SARS‐CoV‐2 infection, hospitalization and mortality from COVID‐19: a living rapid evidence review with Bayesian meta‐analyses (version 7) · 2020
  No notice

Reads

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.