CDC counted 109 children with influenza-associated encephalopathy last season

In January 2025, CDC received several reports of deaths among children with a severe form of influenza-associated encephalopathy called acute necrotising encephalopathy [s1]. Because no national surveillance for influenza-associated encephalopathy exists, the agency could not look the answer up. It had to ask.

CDC requested notification of paediatric cases from clinicians and health departments for the rest of the 2024–25 season [s1]. The resulting count, published on 25 September, is the first systematic US tally of a condition that has been described for decades but never routinely counted.

What the count found

Of 192 reports of suspected influenza-associated encephalopathy submitted to CDC, 109 (57%) were categorised as meeting the case definition [s1]. Of those 109, 37 were subcategorised as acute necrotising encephalopathy and 72 as other influenza-associated encephalopathy [s1]. The remaining 82 reports did not meet the criteria and were categorised as other influenza-associated neurologic disease [s1].

The median age of children with influenza-associated encephalopathy was 5 years [s1]. Fifty-five percent were previously healthy [s1]. Seventy-four percent were admitted to intensive care, and 19% died [s1]. Among the subset with acute necrotising encephalopathy, 41% died [s1].

Among children with encephalopathy who were eligible for vaccination, only 16% had received the 2024–25 influenza vaccine [s1].

The denominator problem

These figures need to be read for what they are: a voluntary, solicited case series assembled mid-season, not incidence. There is no national surveillance for the condition [s1], which means there is no denominator and no way to know what fraction of actual cases the 109 represent. The count is almost certainly an undercount, and the direction and size of the undercount are unknown.

A solicited series also carries ascertainment bias toward severe cases. Clinicians report the children who die or end up in intensive care more reliably than those who recover, which would inflate the apparent case fatality. The 19% and 41% mortality figures are properties of this series, not established case fatality ratios for the condition.

Two other numbers should be held loosely. That 55% were previously healthy [s1] is striking, but "previously healthy" is a clinical judgement recorded under acute conditions. And the 16% vaccination figure [s1] cannot on its own establish that vaccination prevents encephalopathy — a case series has no comparison group. It is consistent with a protective effect; it does not measure one.

The season it sat in

The 2024–25 season was severe. A companion MMWR report published the same day found that the highest number of influenza-associated paediatric deaths reported in a single season since national notifiability began in 2004 — excluding the 2009–10 H1N1 pandemic — occurred in 2024–25 [s2].

Through 13 September 2025, 280 influenza-associated paediatric deaths had been reported, a national rate of 3.8 deaths per million children [s2]. Median age at death was 7 years, and 56% of the children who died had at least one underlying medical condition [s2]. Influenza A viruses were associated with 240 of the deaths (86%) [s2].

Two figures from the deaths report bear on clinical practice. Forty percent of children who died were treated with influenza antiviral medications [s2] — meaning the majority were not. And among the 208 paediatric decedents with available data who were eligible for influenza vaccination, 89% were not fully vaccinated [s2].

What CDC asks clinicians to do

The report's operational recommendation is a diagnostic one: health care providers should consider influenza-associated encephalopathy in children with encephalopathy or altered level of consciousness and a recent or current febrile illness when influenza viruses are circulating [s1].

That framing matters because acute necrotising encephalopathy progresses fast, and the differential for an acutely encephalopathic febrile child is long. A condition that no surveillance system counts is also a condition that clinicians are not systematically primed to look for.

CDC's standing recommendation is that all children aged 6 months and older receive annual influenza vaccination to prevent influenza and associated complications, potentially including severe neurologic disease such as influenza-associated encephalopathy and acute necrotising encephalopathy [s1]. The agency recommends that everyone aged 6 months and older without contraindications receive the influenza vaccine each year, ideally by the end of October [s2].

The word "potentially" in the first of those sentences is CDC's own, and it is the right one. The evidence that vaccination reduces influenza cases is strong; the evidence that it reduces encephalopathy specifically rests, at present, on the inference that preventing the infection prevents its complications.

What to watch: whether the ad hoc reporting mechanism used in 2024–25 becomes a standing surveillance system, and whether the 2025–26 season produces a comparable signal.

This article is informational and is not medical advice.

Sources

• [s1] "Pediatric Influenza-Associated Encephalopathy and Acute Necrotizing Encephalopathy — United States, 2024-25 Influenza Season," MMWR Morbidity and Mortality Weekly Report, 25 September 2025. https://doi.org/10.15585/mmwr.mm7436a1
• [s2] "Influenza-Associated Pediatric Deaths — United States, 2024-25 Influenza Season," MMWR Morbidity and Mortality Weekly Report, 25 September 2025. https://doi.org/10.15585/mmwr.mm7436a2