Dr Naveed Haider, Dr Simon Scammell
A young child with autism comes to your ED. He’s been limping for weeks. He’s already seen orthopaedics. His blood tests came back normal. His hip X-ray looked fine. He was sent home with reassurance – twice.
And then he stopped walking altogether.
We all know about scurvy from our medical school lectures – sailors, vitamin C, limes. But scurvy in a British child in 2026? Surely not. It sounds like something from a Charles Dickens novel.
Except it’s not. And if you don’t think about it, you’ll miss it – just like we nearly did.
Background — scurvy is back (was it ever really gone?)
Scurvy is caused by a lack of vitamin C (ascorbic acid). Your body can’t make it – it has to come from food. The daily requirement is tiny: just 15–45mg in children. That’s less than half an orange. Symptoms start after 4–12 weeks of very low intake.
So how does a child in the UK end up with scurvy?
The answer, more and more often, is autism and restrictive eating. Children with autism frequently have extreme food selectivity. It’s not “fussy eating” – it’s sensory, behavioural, and deeply rooted. Fruits and vegetables are usually the first things to go.
With autism and food restriction on the rise, children’s hospitals are seeing increasing cases of scurvy — and it’s no longer appropriate to view it as a historical disease. A 2024 case from Doncaster and Bassetlaw Teaching Hospitals NHS Trust describes a 14-year-old with autism who developed bilateral subperiosteal iliac haematomas causing lower limb deep vein thrombosis – a complication never previously described in scurvy.
What’s actually going wrong?
Think of vitamin C as the body’s builder and repairman. Its main job is making collagen – the scaffolding that holds blood vessels, bones, skin, and gums together. Without it, things fall apart fast.
Vitamin C is essential for an enzyme called prolyl hydroxylase. Don’t worry about the name – all it does is take loose, floppy collagen strands and cross-link them into strong, stable fibres. No vitamin C? The enzyme stops working. The collagen strands stay floppy. And floppy collagen means everything that depends on it starts to fail.
Blood vessels are basically tubes held together by collagen. When collagen weakens, the vessel walls become fragile and leaky – so you get petechiae, easy bruising, and bleeding gums. The vessels aren’t bursting dramatically. They’re just… slowly falling apart.
Bones are where it gets really dramatic in children. The metaphysis – the fast-growing zone at the end of long bones – is collagen-hungry. It’s building new bone constantly. When collagen fails here, the bone becomes structurally weak, and blood starts pooling under the periosteum (the tough lining wrapped around the bone). That’s your subperiosteal haematoma – and that’s exactly why these children scream when you touch their legs. It’s not the joint. It’s the bone itself.
Iron absorption also depends on vitamin C – it helps convert iron into the form the gut can actually absorb. So iron deficiency anaemia often comes along for the ride, even when the child is getting some iron in their diet.
The result is a child who looks like they have leukaemia, bone infection, or even non-accidental injury. But the cause is sitting in their diet history – if you remember to ask.
The case — how it might present
Picture a child with autism whose diet is severely restricted – commercially available caloric shakes, very little else. No fruits. No vegetables. For months. He first comes in with a limp. Transient synovitis seems like a reasonable call — he goes home. But he comes back, worse. Then worse again. Until eventually, he can’t walk at all.
When you examine him: thin, with reduced subcutaneous fat. Dental caries. Follicular hyperkeratosis on his scalp, and petechiae on his buttocks. You try to move his legs – he screams. No abdominal masses. No bruising overlying the skin.
You send bloods: iron deficiency anaemia, low vitamin D. But the CRP is normal. White cells normal. Coagulation normal. Calcium and bone profile normal. X-rays and hip ultrasound: normal. And you’re stuck.
The investigations – why you shouldn’t miss?
Scurvy can fool you in almost every direction.
Blood tests don’t help much. Vitamin C levels aren’t on a standard panel – you have to specifically request a serum ascorbic acid level. CRP is often normal. Coagulation, calcium, bone profile – all normal. The only hint in the blood is iron deficiency anaemia, which on its own points nowhere specific.
X-rays can look completely normal early on. The classic signs – Trümmerfeld zone, Frankel line, Wimberger ring sign, Pelkan spurs – are all late findings. In early or moderate scurvy, the X-ray is unremarkable.
This is where MRI comes in. MRI can detect early changes in scurvy – the most common finding is diffuse multifocal decreased T1-weighted signal and increased T2-weighted signal within the bone marrow, with the metaphysis most affected. Subperiosteal haematomas appear heterogeneous on MRI with associated oedematous changes – and the appearance of blood products in subperiosteal fluid should raise suspicion of scurvy.
In this presentation, MRI shows subperiosteal fluid collections, heterogeneous bone marrow signal, and periosteal thickening in the femora – findings that, in the right clinical context, point directly to scurvy.
The differential – what else should you consider?
MRI findings in scurvy are nonspecific – the differential includes haematological malignancy, osteomyelitis, and metastatic disease, and can lead to additional invasive investigations.
Septic arthritis / osteomyelitis – always first. Ruled out by normal CRP, white cells, no fever, no effusion. Leukaemia / bone tumour – genuinely considered; MRI in scurvy and leukaemia can look almost identical. JIA – inflammatory markers flat throughout. Rickets – low vitamin D may be present, but rickets doesn’t explain petechiae or the MRI picture. Transient synovitis – reasonable initially, but it should improve in 1–2 weeks. Progressive worsening is a red flag.
Scurvy is only considered once the dietary history is carefully reviewed – and yet, it’s the answer all along.
Child protection – you need to think about this
When a child presents like this – bedridden, screaming in pain, with petechiae on his buttocks – safeguarding should be your first thought, before any imaging is done. Your child protection antenna goes up. And it should.
Good practice in this scenario looks like this: full child protection assessment, detailed social history, and early discussion with your named nurse for safeguarding and social services – before the MRI result is even back.
The MRI, when it comes, helps clarify the picture. Subperiosteal haematomas are not the typical finding in NAI – you would more commonly expect spiral/metaphysial fractures, corner fractures, and subdural haemorrhages. In scurvy, none of those are present. The dietary history, reviewed carefully, explains everything.
But here’s where it gets nuanced – and worth pausing on. In a presentation like this, the concern isn’t always about deliberate harm. The more important question is often: is this family neglecting their child, or are they struggling to manage the genuinely challenging feeding behaviour that comes with autism – without enough support? That distinction matters. Restrictive eating in autism is increasingly common. Scurvy, however, is not. When a child reaches the point of being unable to walk, somewhere the support has failed – whether that is the family, the professionals around them, or both. The answer isn’t always a safeguarding referral. Sometimes it’s asking: who is supporting this family, and is it enough?
The lesson: think about scurvy before you escalate to a full NAI pathway – mislabelling a family as neglectful when they are overwhelmed is just as damaging as missing genuine harm. When in doubt, discuss with your named doctor for child protection. Don’t do this alone.
Treatment – the satisfying part
Oral vitamin C (ascorbic acid) 300–500mg/day in divided doses.
Vitamin C supplementation results in rapid resolution of symptoms. [2] Most children start to improve within days. Within two weeks, many are running around. This is one of those rare moments in paediatrics where the treatment works faster than the diagnosis took.
Top tips 🌟
- Always take a dietary history – especially in autism, ARFID, or any neurodevelopmental condition. Ask: “Does he eat any fruit or vegetables at all?”
- Normal bloods don’t rule out scurvy – you need a serum vitamin C level. Request it specifically.
- Normal X-ray doesn’t rule out scurvy – early disease has no X-ray changes. MRI is far more sensitive.
- Petechiae + bone pain + restricted diet = think scurvy before leukaemia or NAI. But consider all three.
- Low vitamin D and scurvy often travel together – both reflect the same restricted diet.
Summary
Scurvy is making a comeback in 21st-century Britain. Children with autism and restricted diets are most at risk. It mimics transient synovitis, leukaemia, osteomyelitis, and NAI. Blood tests and X-rays can be entirely normal. MRI is the key investigation. Treatment is simple, cheap, and works fast. The secret to getting there? A thorough dietary history and the courage to think the unexpected.
Key learning points
- Scurvy is a 21st-century diagnosis – consider it in any child with a restricted diet, especially autism or ARFID
- Think scurvy: bone pain + petechiae + anaemia + normal inflammatory markers + dietary restriction
- MRI, not X-ray, is the investigation of choice – look for subperiosteal haematomas and heterogeneous marrow signal
- Always consider safeguarding – scurvy can mimic NAI, but nutritional neglect itself may require a safeguarding response
Dr Naveed Haider – ST4 Paediatrics (South Yorkshire); Reviewed by: Dr Simon Scammell – Consultant Paediatrician (Barnsley Hospital)
X handle: @drnaveed83