The hidden pandemic within the pandemic: How COVID-19 impacted sepsis – and what comes next

Sepsis is often called medicine's most under-recognised killer. COVID-19 didn't just add a new disease to the world's plate – it made an existing silent killer dramatically worse, and the data shows a repeatable pattern that will happen again in the next pandemic unless something changes.

US sepsis cases rose roughly 39% from 1.8 million in 2016 to 2.5 million in 2021 – about 20% of that increase came before the pandemic even started, and another 16% arrived during COVID itself. The pattern is not new, and it is not unique to coronavirus: when a pandemic arrives that humanity does not yet know how to treat, the endpoint for a large share of patients is the same – sepsis and multi-organ failure. The pathogen changes from outbreak to outbreak; the destination for the sickest patients does not.

Most people who die of COVID-19, seasonal flu, or a bacterial infection technically die of sepsis – their organs shut down under the weight of the body's own runaway inflammatory response. Yet, sepsis rarely makes headlines the way the triggering virus does. Every major respiratory pandemic in the last century has followed the same sepsis-amplification pattern, and the next one will too, unless health systems change how they prepare.

Cases

US sepsis hospitalisations rose roughly 39% between 2016 and 2021, split between a ~20% pre-pandemic rise and a further ~16% jump during COVID. In France, viral sepsis went from roughly 1% of all sepsis cases before the pandemic to roughly 10% during it.

The sepsis clinicians are trained to treat typically has a clear, identifiable source of infection – an abscess, an infected surgical site, a specific organ. Viral sepsis is a fundamentally different problem: there is no single focus to drain or excise, only a diffuse, ongoing inflammatory process that isn't easily shut off. It's the difference between treating a patient with a contained pocket of infection – where clinicians can perform blood purification once to stabilise the patient before surgery and again afterward to clear residual toxins – and treating a patient whose entire bloodstream is under continuous inflammatory assault with no single source to remove.

Mortality

In-hospital sepsis mortality rose 37.5% among patients 65 and older, and 44% among adults 18–64, between 2019 and 2021. The steepest relative jump in sepsis mortality was among younger adults – though, in absolute terms, older patients still die at higher rates overall (18.7 per 100 stays for 65+ versus 13.9 per 100 stays for 18–64 in 2021). An estimated 36,000 sepsis deaths occurred in 2020 alone, beyond what pre-pandemic trends would have predicted.

This is the recurring failure mode of every pandemic: we are never ready for the surge. Precious time is spent simply understanding a new pathogen, then more time developing vaccines and therapeutics – and while that work happens, hospitals are left without the staffing, protocols, or equipment to absorb the load.

Costs

US aggregate sepsis hospital costs jumped 36.5% in two years, from $38.2 billion to $52.1 billion. Strip out COVID-tagged stays and nearly all of that increase disappears – a clear signal that COVID itself, not a generic rise in sepsis burden, was the cost driver. Per-stay costs climbed too, up roughly 17% over the same two years – and severity, not volume, was the main driver.

Viruses don't just make people sick by themselves – they strip away the lungs' natural defenses, opening the door for ordinary bacteria that live harmlessly in the nose and throat to invade deeper tissue and cause secondary bacterial infection. So, virus first and bacteria second, is the throughline connecting pandemics a century apart – they all follow the same path.

On top of all this, pandemics strain the systems that sepsis treatment depends on: ICU beds, ventilators, antibiotics, dialysis capacity. Sepsis care is fundamentally a race against the clock, and pandemics slow the clock down at exactly the moment speed matters most.

1918 flu

The majority of deaths were ultimately caused by secondary bacterial pneumonia, not the influenza virus itself – the virus damaged the lungs first, and bacterial infection did the actual killing. No antibiotics existed yet, making secondary bacterial sepsis essentially untreatable.

2009 H1N1

Roughly 30% of ICU patients had a bacterial co-infection. Those coinfected patients needed vasopressors and mechanical ventilation far more often than patients with the virus alone, and saw substantially higher mortality.

Antibiotic and critical-drug shortages during COVID forced clinicians toward less-ideal alternatives, which likely worsened outcomes in some patients and fed the broader antimicrobial resistance crisis. Resistance is a slow-burn problem that pandemics accelerate: resistant infections are harder and more expensive to treat, and that burden compounds for years after the acute crisis has passed. Post-sepsis syndrome – lasting cognitive and physical effects seen in up to half of survivors – extends the true cost of pandemic-driven sepsis well beyond the initial hospitalisation, in much the same way long COVID has.

Underlying all of this is a coding problem: sepsis was formally coded in only about 9% of COVID deaths, while clinical and autopsy studies find sepsis present in 52–93% of those same deaths. The official numbers understate the true toll – by the data's own admission.

If a novel respiratory pathogen emerges with 1918-like severity, or the pandemic potential sometimes discussed around highly pathogenic avian influenza strains, the historical pattern suggests several things are likely: a larger initial spike in undercoded sepsis cases before clinicians and coding systems catch up – mirroring the 2020–2021 lag, likely worse if the pathogen is even less familiar than SARS-CoV-2 was. Additionally, disproportionate strain on ICU and antibiotic/consumable supply chains – worse than COVID if the next pathogen combines high case-fatality with high transmissibility, a combination COVID mostly avoided. Further, costs scaling faster than case counts, since severity drives cost per hospital stay – as it did during COVID, when per-stay costs rose even as clinical care matured and improved.

So, what can be done? Stockpiled ventilators, dialysis/CRRT capacity, and consumable cartridges, pandemic-specific sepsis surveillance codes that don't rely on clinicians retrofitting old bacterial-sepsis categories onto a viral disease, faster point-of-care diagnostics for rapid pathogen identification, pre-negotiated supply chain contracts for antibiotics and critical drugs, and public awareness campaigns. If governments want to save lives during a pandemic that will definitely come, the work starts today.

Global sepsis experts have described the pandemic's legacy as an opportunity – a chance to finally treat sepsis with the same urgency and infrastructure investment the world mobilised for COVID-19 itself. Sepsis already kills more people annually than COVID did at its peak. The next pandemic will simply make that number larger, unless sepsis readiness – equipment, supply chains, surveillance, and public awareness – is treated as core pandemic infrastructure, rather than an afterthought.

About the author

Dima Romashin is the founder and CEO of Efferon, a European biotech company developing blood purification technologies for critical care. He co-founded Efferon in 2016 and has led its business development, corporate strategy, and international expansion. A serial entrepreneur with experience across healthcare, finance, and manufacturing, Romashin focuses on bringing complex medical technologies from research into scalable clinical practice.