Article Text

Download PDFPDF
Fifteen-minute consultation: The limping child
  1. Jonathan Adamson1,
  2. Thomas Waterfield2
  1. 1 Emergency Department, Birmingham Women’s and Children’s NHS Foundation Trust, Birmingham, UK
  2. 2 Royal Belfast Hospital for Sick Children, Belfast, UK
  1. Correspondence to Dr Jonathan Adamson, Emergency Department, Birmingham Women’s and Children’s NHS Foundation Trust, Birmingham, UK; jonadamson{at}


The limping child is a common presentation to paediatric services. In most instances the cause is benign with few, if any, investigations required. There is, however, always that concern that the limping child may have an underlying limb-threatening or life-threatening disease. This poses a challenge to clinicians, who must find that balance between correctly identifying disease early and avoiding the risks and harms of overinvestigation. In this article we discuss the diagnostic approach to the limping child and present a structure for assessment, investigation and risk management.

  • limp
  • orthopaedics
  • general paediatrics

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Limp is a term that describes an abnormal gait caused by pain, weakness or deformity, most commonly characterised by a shortened ‘stance’ phase on one leg, in order to offload the affected side and prevent or reduce pain.1 Traumatic causes of limp include soft tissue injuries and fractures. In these instances, the history of significant trauma is clear with a temporal relationship between injury and symptoms.

Atraumatic limp is common in children, with a reported annual incidence of 1.5–3.6 per 1000.2 3 The causes of an atraumatic limp are varied and are summarised in table 1.1–8 In assessing the child with an atraumatic limp, the aims are to (1) identify children at high risk of limb-threatening or life-threatening disease, (2) attempt to make a positive diagnosis, and (3) manage risk in those where the diagnosis is uncertain initially.

Table 1

Differential diagnoses: atraumatic limp


The assessment of the limping child involves a detailed history and physical examination, as summarised below. The crucial ‘not-to-miss’ diagnoses are infection, malignancy and non-accidental injury. The presence of fever or systemic illness should alert you to the possibility of a significant disease, whereas a short history in a well child is more reassuring.

The examination is especially important as children often poorly localise leg pain, especially hip and knee pain. One common pitfall is to fail to correctly identify the source of the limp. This invariably leads to either overinvestigation or an incorrect diagnosis. Time should be taken to examine the child, making use of analgesia and distraction. As part of the examination, it is important to look for signs of rare but serious causes of ‘leg pain’; these include signs of underlying malignancy, abdominal pathology and testicular torsion in boys.

These pitfalls are outlined in table 2.

Table 2



  • Age.

  • Duration.

  • Complete inability or refusal to weight-bear: crawling but not walking is more typically due to a lower leg issue (eg, toddler’s fracture).

  • Trauma.

  • Fever or systemic illness.

  • Bruising, bleeding and pallor.

  • Preceding illness.

  • Morning stiffness.

  • Pain: site, severity, and exacerbating or relieving factors.

  • Additional sites of pain: other joints and bone pain.

  • Previous limping episodes and their cause (if known).


  • General appearance: well or unwell?

  • Temperature.

  • Bruising, bleeding and pallor.

  • Gait assessment: note that running may exaggerate a subtle limp.

  • Joints: remember that knee pain may be referred from the hip and thigh pain may be referred from the lumbar spine, so examine the sacroiliac joints and spine. Pain on flexion or midline tenderness may indicate discitis. Abduction and internal rotation of the hip are most commonly restricted in hip pathology.

  • Posture: excessive lordosis may indicate discitis.

  • Lymphadenopathy.

  • Neurology: ataxia and weakness.

  • Bony tenderness.

  • Abdomen, scrotum and inguinal area: masses, appendicitis and testicular pain.

Identifying serious disease

Following the initial assessment, it is important to consider if the child has a life-threatening or limb-threatening cause for their limp (table 1). Of the possible differential diagnoses, there are a number of specific localised limb diseases to be especially mindful of.

Septic arthritis/osteomyelitis

Septic arthritis and/or osteomyelitis should be suspected in any child with a limp and fever (or history of fever). Septic arthritis is an infection of a joint; definitive diagnosis is made by positive cultures from the joint. In children, the route of infection is most commonly haematogenous spread, and as such disproportionately affects the joints in which the highly vascular metaphysis is intra-articular.1 The incidence is 4–5 per 100 000 in developed countries, and the most common causative organism is Staphylococcus aureus.8 The introduction of the Haemophilus influenzae type B (Hib) vaccine has all but eliminated Hib as a cause.8 In neonates, group B streptococcus and Escherichia coli should be considered.9

Prompt investigation and treatment with surgical washout and antibiotics is required to avoid joint destruction.1 Parenteral antibiotics are started after bacteriological samples have been obtained; choice of empirical antibiotic will depend on local protocols but would typically be directed against S. aureus and may include cephalosporin and/or clindamycin. Vancomycin may be used empirically in areas with high prevalence of methicillin-resistant Staphylococcus aureus (MRSA).

Distinguishing septic arthritis from other causes

A common conundrum in managing children with atraumatic limp is whether the limp is caused by a benign condition, such as transient synovitis, or septic arthritis. Kocher et al 10 11 produced an evidence-based risk stratification based on the presence or absence of four features: fever >38.5°C, inability to weight-bear, erythrocyte sedimentation rate (ESR) >40 and white cell count >12 x 10^9/L (table 3).

Table 3

Risk of septic arthritis

It has since been shown that C-reactive protein (CRP) >20 can be used instead of ESR.12 Based on these findings, any child with fever and an inability to weight-bear confers enough of a risk to justify further investigation, whereas in the child with no fever and in whom pain is mild the risk is minimal with little justification for blood tests.

Slipped upper femoral epiphysis

This is a condition that affects children over 10 years old and is slightly more common in boys and those with an elevated body mass index.6 13 In slipped upper femoral epiphysis (SUFE; also known as slipped capital femoral epiphysis), the femoral head displaces, usually posteriorly, in relation to the metaphysis.1 Prompt diagnosis of SUFE is crucial to avoiding long-term deformity.14

SUFE can present as knee pain, being the primary feature in as many as 15%. Atraumatic knee pain should prompt consideration of hip pathology, and those with discomfort on internal rotation of the hip should be X-rayed. Not considering the hip in a child with knee pain is a common pitfall1 (table 2). Furthermore, frog lateral projections will provide the best view for identifying SUFE; it is possible to miss a SUFE if only an anteroposterior film is taken. Figures 1 and 2 demonstrate the radiographical appearance of SUFE.

Figure 1

Slipped upper femoral epiphysis. Klein line is drawn on the anteroposterior radiograph along the superior aspect of the femoral neck and should intersect the epiphysis. (Image courtesy of John M Flynn, MD, via

Figure 2

Anteroposterior (AP) and frog-leg lateral radiographs showing slipped upper left femoral epiphysis. Note that the AP film looks essentially normal (and Klein line would intersect the epiphysis). Frog-leg lateral view is often more sensitive, as the typical slip is posterior and medial. Here you can see misalignment of the femoral epiphysis and femoral neck on the frog-leg lateral view. (Case courtesy of Dr M Osama Yonso,, rID: 17079.)

Perthes disease

Perthes disease is a condition that typically affects boys aged 4–815 and is characterised by avascular necrosis of the capital femoral epiphysis. It is self-limiting, with revascularisation and new bone formation occurring over around 2–4 years.

In the early stages of vascular occlusion, soft tissue changes including synovitis occur in the hip and there may be no radiological changes. A suspected transient synovitis which does not resolve within 2 weeks should prompt further imaging looking for Perthes, either by technetium bone scan or MRI.1 The subsequent phases of Perthes are fragmentation, during which time the femoral head gradually extrudes leaving it vulnerable to deformity, and reconstitution. Typical X-ray changes progress from sclerosis to fragmentation, and eventually to flattening of the femoral epiphysis.16

The purpose of treatment is to ensure that femoral head deformity, a result of excessive extrusion of the femoral head and likely to lead to degenerative arthritis in mid-adult life,17 does not occur during revascularisation. It is therefore aimed at ‘containment’ of the femoral head within the acetabulum and can be surgical or non-surgical. Figure 3 demonstrates the radiographical appearance of Perthes disease.

Figure 3

Anteroposterior radiograph showing Perthes disease of the left hip. Note the fragmentation and flattening of the left capital femoral epiphysis. (By J Lengerke - Praxis Dr Lengerke, Public Domain,

Managing risk

Once the assessment is complete, including consideration of the potential causes and common pitfalls (tables 1 and 2), a decision has to be made with regard to further investigation and management. Those with signs and symptoms of significant disease will require investigations (table 4) and onward referral. Those with a short history (under 72 hours) and no worrying features can initially be managed without further investigation (figure 4). It is essential to provide safety netting advice and to arrange follow-up.

Table 4

First-line investigations

Figure 4

Flow diagram: suggested approach to a limping child. AP, anteroposterior; SUFE, slipped upper femoral epiphysis; XR, X-ray; FBC, full blood count; CRP, C-reactive protein

The most common diagnosis in the group of children who are well, mobilising and under 10 is transient synovitis of the hip (‘Irritable Hip’). This is the most common cause of atraumatic limp in children and typically occurs between the ages of 2 and 8. It is more common in boys.3 5 There is often a history of recent or concurrent viral illness.18 19 A definitive diagnosis would be demonstrated by proven joint effusion with exclusion of other causes (eg, negative microbiology on joint aspiration); in practice, transient synovitis is often made as a presumptive clinical diagnosis without invasive testing. Another significant group are those given no formal diagnosis but managed symptomatically and have a self-limiting course.


The limping child can represent a challenge for clinicians. While most have a self-limiting and benign cause, a minority have significant underlying disease. Uncertainty around diagnosis and difficulties in assessing the child can lead to overinvestigation or failure to identify serious underlying pathology. A structured approach to the assessment, including a consideration of risk, can help identify those requiring further investigation and treatment from those who can be safely discharged without investigation.



  • Contributors JA and TW planned the manuscript. JA wrote the body of the text, tables and added the figures. TW edited the document and the tables. Both authors have agreed to the full content.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Commissioned; externally peer reviewed.

  • Patient consent for publication Not required.