Zdenek Vrozina Profile picture
Jan 31 • 18 tweets • 3 min read • Read on X
Long COVID in children is often reduced to fatigue.
This study shows something far more concrete - measurable impairment in language skills - speaking, listening, and readingđź§µ
Study design.
N = 1,244 children (ages 3–18) in Hong Kong.
Three groups.
1. long COVID
2. COVID, fully recovered
3. never infected
Language was assessed using LEAP-Q, covering
speaking, listening, reading, writing - in two languages (Chinese + English).
Main finding.
Children with long COVID had significantly lower overall language proficiency than both recovered children and never-infected peers.
Crucial detail.
There was no significant difference between COVID without long COVID and no-COVID groups.
This directly challenges the claim that deficits are just due to lockdowns, masks, school disruption etc
Which domains were most affected?
The largest drops were in speaking and listening.
Reading was also reduced.
Writing showed less consistent differences.
Age matters.
No clear effect in kindergarten.
Clear deficits in primary school.
Strongest impact in secondary school students
Why this makes sense?
Older students are in the reading to learn phase. Language becomes the tool for complex thinking.
COVID often affects attention, working memory, and cognitive endurance - language performance suffers first.
Important nuance.
The negative effect appeared in both the first language and English.
This was not just missed school English, but a broader cognitive impact.
Limitations (fairly stated)
No pre-infection baseline testing.
Parent-reported measures (not clinical language tests).
Cross-sectional design.
Still, the pattern is consistent and developmentally plausible.
What this shows in real life?
Long COVID in children is not subjective or trivial.
It can mean measurable disruption of communication, learning, and daily functioning.
Estimated Long COVID prevalence in children (RECOVER / JAMA)?
Using the stricter PASC-probable definition
Ages 6–11 ~20%
Ages 12–17 ~14%
These are not marginal numbers!
And in the United States?
ACT scores were stable for ~30 years.
After 2019, the composite score drops sharply - to the lowest level in decades.
The decline is synchronous across math, reading, English, and science. And persists.
This is not normal fluctuation. Image
Image
At the same time, high school GPAs rise or remain stable while ACT scores fall.
That divergence matters - grades adapt, standardized tests don’t.
What’s declining is measured cognitive performance, not just grading standards. Image
Most countries lack national screening for pediatric long COVID.
Education and healthcare remain siloed.
Recognizing a biological driver would imply responsibility, cost, and abandoning the claim that children are fine.
Instead, the focus stays on learning loss and pandemic disruption -
not on population-level biological burden.
If public health continues to say children are fine,
it is effectively accepting loss of learning, function, and quality of life in a substantial fraction of a generation. @szupraha @msmtcr @ZdravkoOnline @adamkova_vera @adamvojtech86 @RobertPlaga
A public health call to action.
Stop treating pediatric long COVID as a taboo
monitor cognitive and language outcomes in schools
fund pediatric research and diagnostics
and
most importantly - reduce repeated infections as a preventable risk!
Xu at al., The Impact of Long COVID on Language Proficiency Across Different School Levels in Hong Kong. mdpi.com/2076-328X/15/4…

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Zdenek Vrozina

Zdenek Vrozina Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @ZdenekVrozina

Feb 6
If normal population plasma truly carries more low-grade inflammation, this study hints at a fork in the road.
Either we lower the bar and call it a new normal,
or this is a hidden population burden that will surface later as comorbiditiesđź§µ
A new study on the cytokine IL-32 after COVID-19 points directly at this uncomfortable question.
The authors analyzed nearly 1,000 healthy blood donors sampled before and during the pandemic, plus 212 hospitalized COVID-19 patients.
The result is consistent - plasma collected after 2020 shows systematically higher IL-32 levels compared to pre-pandemic plasma.
Read 9 tweets
Feb 5
This study does not tell us what exactly causes long COVID or ME/CFS, nor does it test clinical symptoms like PEM.
But it may tell us something just as important - what type of biological problem this likely is..đź§µ
The authors isolated immunoglobulins (IgG) from people with post-infectious ME/CFS, including post-COVID ME/CFS, and tested what these antibodies do to healthy cells.
In a subset of patients, these IgG alter the behavior of endothelial cells and their mitochondria.
Not by killing the cells or shutting down ATP production.
Read 17 tweets
Feb 2
This isn’t a new comparison.
For years, parallels between NeuroHIV and neuro-COVID/Long COVID have been discussed across fields.
What’s new is that they are now formally described as shared CNS mechanisms, not just analogy!🧵
Just a few years ago, parallels like
HIV - SARS-CoV-2
HAND - brain fog/neuro-LC
microglia - chronic inflammation
vasculature - cognition
were treated mainly as interesting analogies. With caution not to overstate them.
This new review formalizes the shift. These parallels are not Twitter pattern recognition, but convergent CNS phenotypes following viral insults.
A new review shows they are biologically grounded similarities.
Read 15 tweets
Feb 1
A new review links Alzheimer’s disease, Parkinson’s disease, and COVID-19 through a shared core - neuroinflammation + oxidative stress.
The same pathways, the same immune nodes, the same vulnerabilities of the brainđź§µ
Key players.
Microglia (the brain’s innate immune cells) and neutrophils (peripheral rapid responders).
When the blood–brain barrier (BBB) is disrupted, neutrophils enter the CNS and inflammation becomes self-amplifying.
Neutrophils can form NETs (neutrophil extracellular traps - DNA + histones + enzymes like MPO/elastase).
In the brain, NETs mean oxidative damage, mitochondrial stress, neuronal injury - and further microglial activation. A vicious cycle.
Read 19 tweets
Jan 30
Long-COVID monocyte signatures begin forming already in acute COVIDđź§µ
A new single-cell multiome study helps answer a key long-COVID question.
Where do the pathological monocyte programs seen later actually come from?
This paper maps how classical monocytes change already during the acute phase of COVID-19.
Using paired RNA + chromatin profiling, the authors analyze immune cells across disease severity from healthy controls to critical COVID.
A clear pattern emerges -
myeloid cells, especially classical monocytes, carry much of the severity-associated signal.
Read 19 tweets
Jan 30
The choroid plexus (ChP) has become one of the most consistent findings in post-COVID and long COVID neuroimaging.
It is enlarged.
But larger alone doesn’t yet tell us what is actually happening🧵
A new paper in Alzheimer’s & Dementia moves the ChP story one step further.
It’s no longer just about volume - the authors also measure choroid plexus blood flow (CBF) and link it to cognition and biomarkers.
Study design.
86 long COVID patients, 67 recovered COVID, 26 healthy controls.
MRI + pCASL perfusion imaging.
In a subset of long COVID patients, blood biomarkers (GFAP, p-tau217, etc)
Read 18 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(