It is seven in the morning. He gets out of bed and walks crooked. He does not cry, does not complain, sets his foot down in a way that is not his own, and you stand there watching him cross the corridor three times to work out whether you are imagining it.
You are not imagining it. And no, it is not a phase: limping, in a child, is always a symptom. But “it is not normal” does not mean “it is serious”. In the vast majority of cases there is something minor behind it, which passes on its own within a few days.
The hard part is telling the difference. That is why I will start from the end.
The rule that matters most
These are the reasons not to wait: call the paediatrician or go to the emergency department today, not on Monday.
- Fever, even a mild one, together with the limp.
- The child cannot put weight on that leg, or refuses to walk at all.
- The joint (hip, knee, ankle) is hot, swollen or red.
- He will not let you move the leg: as soon as you bend or rotate it, he stiffens and cries out.
- He is under 3 years old.
- The limp has lasted more than a week, or is getting worse instead of better.
- There is pain that wakes him at night, or pain in one precise spot on a bone.
- There are other signs: weight loss, paleness, constant tiredness, bruises you cannot explain.
Outside this list you have time to observe for a couple of days. Inside it, no.
And here something uncomfortable needs saying. An Italian study conducted in a large paediatric emergency department, on 485 children who came in with a non-traumatic limp, found that serious conditions were present in 6.2%, thirty children, but only 36.7% of those thirty showed at least one warning sign. Translated: warning signs are there to decide when to rush, not to decide that everything is fine. If the limp lasts, it should be seen in any case. In the same study the serious cases were concentrated mainly in the youngest children.
What usually happens
Now the part that helps you breathe, because that too is made of numbers.
One study followed 243 children under 14 who came to a paediatric emergency department with an acute limp and no trauma, checking their records again 18 to 21 months later. Median age 4.35 years, almost twice as many boys as girls, pain present in 80% of cases. In 40% there had been an illness in the preceding days: a flu, a bad cold, gastroenteritis.
The most frequent diagnosis was transient synovitis of the hip: 39.5%. Perthes' disease, 2%. And the figure that matters most to you: 77% of children were treated and discharged directly from the emergency department, without admission.
Most limps end with “rest and recheck in a few days”. That is the rule, not the exception.
Why the doctor's first question is about age
It is not a polite question. Age changes precisely the list of possible causes.
Under 3 years
This is the group where you need the most care, for a practical reason: the child cannot tell you where it hurts, and sometimes does not say anything at all. The most common causes remain trivial, such as a blister, a shoe that pinches, or a knock in the park that you did not see, but this group also includes the toddler's fracture: a fine, non-displaced fracture of the tibia, often after a minimal trauma, which may not show on the first X-ray.
Also in this group, a limp that has been “there forever” and did not appear suddenly is a different story: there the suspicion moves to hip development, which requires a different assessment, carried out calmly but carried out.
From 3 to 10 years
This is the territory of transient synovitis of the hip, which I explain further on. It is also the age group for Perthes' disease, also called Perthes' disease of the hip or avascular necrosis of the femoral head: they are the same thing. It is a temporary disruption of the blood supply to the head of the femur. A Danish study of almost two million children counted 2,374 cases diagnosed between 2 and 12 years, 81.6% of them boys, with an incidence of 12.1 per 100,000 live births, declining over recent decades. It is rare, but its first sign is often a limp that does not go away, with little or no pain.
From 11 years up
Here the diagnosis to fear is slipped capital femoral epiphysis, the slipping of the femoral head on the neck of the femur. It affects adolescents, more often boys and those who are overweight, and in the United States its incidence doubled over ten years: from 2.46 to 5.96 cases per 10,000 children between 2011 and 2020, in step with childhood obesity.
In this age group there are also much more common and much less worrying causes, which can cause a limp for weeks: Osgood-Schlatter disease of the knee and Sever's disease of the heel. These are inflammations of the growth cartilage, managed with graduated loading and they settle as the child grows.
What transient synovitis looks like
A French group analysed 995 children who came to the emergency department with a non-traumatic limp, with the diagnosis confirmed at the follow-up visit. Average age 5.3 years, 337 cases of transient synovitis. From this they extracted the four features that best distinguish it from everything else:
- age between 3 and 10 years;
- no fever;
- no local inflammatory signs (no redness, no visible swelling);
- sudden onset of the limp on waking.
Taken together, in children over 12 months with a limp lasting ten days or less, these four features have a specificity of 98.2% and a positive likelihood ratio of 19.6, and in validation no serious diagnosis was missed.
If you recognise that picture, it is the most likely explanation. That does not entitle you to make the diagnosis at home: those four features are for a doctor who has already examined the child and ruled out the rest.
Synovitis or septic arthritis: the question that keeps you awake
This is the real fork in the road, because the two look alike at the start and their consequences are opposite. Septic arthritis is an infection inside the joint and needs urgent treatment.
The best-known tool for telling them apart is the Kocher criteria, born from a study of children with an acutely painful hip. The four independent predictors are: history of fever, inability to bear weight, ESR of at least 40 mm/h and white blood cell count above 12,000/mm³. The estimated probability of septic arthritis rises as follows:
| Criteria present | Estimated probability of septic arthritis |
|---|---|
| 0 | less than 0.2% |
| 1 | 3.0% |
| 2 | 40.0% |
| 3 | 93.1% |
| 4 | 99.6% |
Look at the numbers: two criteria out of four already mean 40%. That is why nobody plays at waiting.
But honesty is needed here, because this table circulates online as if it were a test to do at home, and it is not. First: two of the four criteria are blood tests, so you cannot apply it. Second, and more important: it works less well than it seems. The authors' own validation on a new population saw accuracy fall from 0.96 to 0.86. An Australian prospective study of 535 children with acute limp, where septic arthritis was 2.6% of cases, measured an accuracy of 0.72 for the Kocher criteria and 0.78 for their version with CRP. And in an Asian series of 101 children, with all four criteria present the real probability of septic arthritis was 59%, not the 99% in the original table.
The same Australian study, however, identifies two signs that you can observe yourself, and these are the most informative of all:
- a marked restriction of movement compared with the healthy side (the strongest sign of all);
- the inability to bear weight.
With a warning that the authors state in black and white: their absence is not enough to rule anything out. None of the signs tested was reliable enough to say “then it is nothing”.
“He fell, but the X-ray is negative”
This happens very often in the youngest children, and it does not mean the child is pretending. Fine, non-displaced fractures of the tibia in this age group can remain invisible on the first X-ray: a dedicated review notes that X-rays are often negative at the first assessment and that the diagnosis rests mainly on history and clinical examination. In a series of 39 children treated on clinical suspicion alone despite a negative X-ray, the fracture was confirmed at follow-up in 41% of cases.
The most up-to-date review of this type of fracture brings reassuring news: they do not displace and heal anyway, whatever the treatment. In fact, the traditional circular plaster cast was associated with more skin problems than a removable brace or simple observation, and for that reason management now varies a lot from centre to centre.
So: a negative X-ray with a child who still will not bear weight is not a contradiction. It is a reason to recheck, not to file the case away.
“He complains about his knee”: look at the hip
It is perhaps the most costly trap in paediatric orthopaedics. Hip pain in children and adolescents very often shows up at the knee or the thigh, and if you stop at the knee you lose precious time.
The rule in paediatric orthopaedics is clear: every child who complains of hip, thigh or knee pain should have a hip examination. And a “pulled groin muscle” in a child is so rare that it can only be accepted after everything else has been ruled out. In slipped capital femoral epiphysis, diagnostic delay is common and it has a price: the deformity progresses, and a stable form can become unstable, with more serious complications.
If you also notice that the foot tends to turn outwards when he walks, or that the hip opens outwards when you bend it, say so explicitly to the doctor.
When the suspicion moves elsewhere
There are presentations where the limp is the tip of something that is not orthopaedic. A study of 29 children who were later found to have a tumour, and who initially reached paediatric rheumatology with the wrong diagnoses, isolated the features that should have shifted the suspicion:
- non-articular bone pain (68% of cases);
- back pain as the main symptom (32%);
- bone tenderness on pressure (29%);
- significant general symptoms: fever, tiredness, weight loss (32%).
In 40% of cases the tests already done contained abnormalities that had not been recognised. This is not an invitation to anxiety: it is the reason why pain that wakes him at night, a child who is losing weight, or pain that sits in the bone rather than the joint deserves to be described clearly to the paediatrician, without playing it down.
At the opposite extreme there are growing pains, which have a typical pattern: evening or night-time, in both legs, with the child running around the next morning as if nothing had happened. If that is your situation, we have covered it in full in growing pains: what they really are.
What a physiotherapist can do, and what not
Let us say it plainly, even though it is counterintuitive for us: for an acute limp, the physiotherapist is not the first door. A doctor comes first, to rule out causes that require antibiotics, X-rays, an orthopaedic surgeon or admission. Anyone who suggests exercises for a child who has been limping for three days without anyone having examined him is doing him harm.
Physiotherapy comes afterwards, and in these cases it really helps: when the diagnosis is in place and it is one of those managed with graduated loading, exercise and timing management, such as Osgood-Schlatter, Sever's disease, recovery after a minor injury, or a gait that remains asymmetric even after the pain has gone.
With Ri-Hub Junior you can start with a free twenty-minute introductory consultation with a physiotherapist on the Italian professional register who works with children and teenagers: you show us how he walks, we talk about it, and we tell you frankly whether the right place is us or whether you first need to go through the paediatrician. For children up to 10 years old, sessions last 50 minutes and cost 59,90 €; from 11 to 18 years old you move to the standard session at 29,90 €.
Two things we do not do, and we prefer to tell you before: we do not make diagnoses in place of the doctor for an acute limp, and we do not treat complex neurological conditions online, which require in-person assessment and specialised centres.
In practice, tonight
If he has a fever, is not putting his foot down, has a hot or swollen joint, or is under 3 years old: call now. If he has none of these, keep him calm for a couple of days, watch how he gets up tomorrow morning, and note on your phone when it started. That date is the first thing they will ask you.
And if after a week he is still limping, do not wait for it to pass: get him seen.
The most common cause of all has a name and a precise course: Transient synovitis of the hip.
If instead the limp has always been there, it is worth measuring the legs: Child with one leg shorter.
If he does not limp but just tires, the threshold is a different one: Child who gets tired walking.
Sources
- The limping child: epidemiology, assessment and outcome, Journal of Bone and Joint Surgery (Br), 1999
- Atraumatic Limping Child, a Challenge for Pediatricians: An Observational Age-Related Study in a Pediatric Emergency Department, Children, 2024
- Transient synovitis of the hip: Development and validation of a new diagnostic algorithm, Acta Paediatrica, 2024
- Differentiating between septic arthritis and transient synovitis of the hip in children: an evidence-based clinical prediction algorithm, JBJS Am, 1999
- Validation of a clinical prediction rule for the differentiation between septic arthritis and transient synovitis of the hip in children, JBJS Am, 2004
- Test characteristics of clinical findings and clinical decision rules for the diagnosis of septic arthritis in children with an acute limp presenting to the emergency department: a prospective observational study, Emergency Medicine Journal, 2025
- Factors distinguishing septic arthritis from transient synovitis of the hip in children. A prospective study, JBJS Am, 2006
- C-reactive protein of ≥ 20 mg/L and ultrasound finding of an effusion ≥ 7 mm has a high specificity and sensitivity in diagnosing paediatric hip septic arthritis, Archives of Orthopaedic and Trauma Surgery, 2023
- Incidence trends and risk factors for Perthes' disease in children born between 1985 and 2016: a Danish nationwide register-based study, Bone & Joint Journal, 2025
- Decade-long Trends in Incidence of Slipped Capital Femoral Epiphysis in the United States: A Nationwide Database Analysis of Over 33 Million Patients, JAAOS Global Research & Reviews, 2024
- Slipped capital femoral epiphysis: the importance of early diagnosis, Pediatric Annals, 2006
- Toddler's fracture: an updated guide on evaluation and management, Journal of Pediatric Orthopaedics B, 2025
- The Toddler's Fracture, Pediatric Emergency Care, 2022
- Toddler's fracture: presumptive diagnosis and treatment, Journal of Pediatric Orthopaedics, 2001
- Malignancies in children who initially present with rheumatic complaints, Journal of Pediatrics, 1999
Last source check: September 2026.
