AMSTERDAM — At the European Respiratory Society (ERS) Congress 2025, respiratory experts debated whether to consider viral sepsis a distinct clinical threat that demands specific recognition or a dangerously ambiguous term that risks misattributing symptoms and delaying life-saving antibiotics.

While the COVID-19 pandemic proved that a virus can indeed cause a dysregulated host response leading to life-threatening organ dysfunction, experts disagree on the utility of viral sepsis as a broad clinical diagnosis.

Joseph P. Mizgerd, ScD, professor of medicine and virology at Boston University Chobanian & Avedisian School of Medicine, Boston, argued that viruses are among the deadliest pathogens and are sufficient to cause sepsis, while Karl Hagman, MD, senior consultant in infectious diseases at the University of Gothenburg, Gothenburg, Sweden, countered that the current clinical tools for diagnosing sepsis are not validated for viruses, creating a high risk for misdiagnosis and clinical harm.

Nicolas Roche, MD, PhD, professor of respiratory medicine at Paris Cité University and head of the Respiratory Medicine Department at Cochin Hospital, both in Paris, France, said getting to the bottom of the debate is important because whether viral sepsis is real or not, and what the mechanisms of it can inform what clinicians can do about it in terms of diagnosis and therapeutic approaches.

Everyone Agrees Viral Sepsis Exists

No one disputed that a virus can cause sepsis, a life-threatening organ dysfunction that arises when the body has a dysregulated response to infection. Mizgerd opened his argument by noting that 80% of the session’s audience, surveyed before the session began, already believed that viral sepsis is a real threat.

Mizgerd presented data showing that while viruses were documented in a smaller percentage of ICU sepsis cases in one large international study, the associated mortality was the same regardless of the microbial cause. He stressed that respiratory infections are the most common trigger for sepsis and that focused microbiological studies reveal a significant presence of viruses. “My conclusion from these really micro-focused studies is that half or more of patients with severe pneumonia have a viral infection,” he said.

But whether the virus is a cause or a consequence of illness remains unclear. Mizgerd pointed to the success of antiviral treatments as proof of causality. Citing a systematic review of more than 1.2 million patients with COVID-19, he noted that the antiviral drug remdesivir significantly reduced mortality, even in patients on mechanical ventilation. “This really suggests that the antiviral therapies are helpful and that the viruses are causing [the sepsis], not just [acting as] consequences,” he said.

But Some Have Concerns

Hagman agreed that viruses can cause sepsis but challenged the clinical utility of the term “viral sepsis,” arguing that it could lead to a harmful misdiagnosis, and raised four critical concerns with the broader diagnostic label.

First, the clinical scoring systems used to identify sepsis, such as the Sequential Organ Failure Assessment (SOFA) and quickSOFA (qSOFA), were validated in patient cohorts with suspected bacterial infections. “The SOFA score and the qSOFA that we use to assess the clinical criteria for sepsis are not validated for viral infections,” he explained.

Second, the definition of a dysregulated host response is ambiguous in the context of a viral infection. Much of the tissue damage in viral disease is caused by the host’s own immune response, which is necessary for viral clearance. “It’s not clear when the host response to a virus should be considered dysregulated and when it’s just what we would expect,” Hagman said.

Third, he questioned whether viral sepsis is a clinically relevant term, as it groups together infections like influenza, Ebola, and COVID-19, all of which have vastly different characteristics and require highly specific treatments. “I’m not sure that it’s helpful for the clinician,” he said.

His final and most pressing concern was the risk for misattribution. He described a common clinical scenario: an elderly man with fever, cough, and low oxygen saturation who tests positive for influenza. “How sure are we that he really has a viral sepsis?” Hagman asked. “Could it be a coinfection with both a virus and a bacterium? Or does he have a secondary bacterial infection?”

Overdiagnosing viral sepsis, he warned, could lead to delaying antibiotics and administering ineffective antivirals.

What Should Physicians Do?

For clinicians on the front lines, the debate boils down to what to do when a patient presents with sepsis. “The most important thing in the ER [emergency room] is that you give them antibiotics, because time is very important with bacteria,” Hagman said.

This approach also makes sense from a biological standpoint. “Just because somebody has a virus doesn’t mean they don’t have a secondary bacterial infection,” Mizgerd said.

Both speakers also agreed that initiating broad antiviral therapy is not currently feasible. Unlike broad-spectrum antibiotics, antivirals are highly specific in their action. “You can’t really just give a patient an antiviral and hope that it might work, because an anti-influenza and an anti-coronavirus are very different,” Mizgerd explained.

Ultimately, the consensus highlighted the urgent need for improved and faster diagnostics, specifically tools that can rapidly distinguish between bacterial and viral pathogens and characterize the specific nature of the host’s immune response. Such advances would enable clinicians to administer rapid, precision-guided, and individualized therapies.

Mizgerd, Hagman, and Roche reported having no relevant financial relationships.

Manuela Callari is a freelance science journalist specializing in human and planetary health. Her work has been published in The Medical Republic, Rare Disease Advisor, New Scientist, The Guardian, MIT Technology Review, and others.