The content you’re about to read or listen to is at least two years old, which means evidence and guidelines may have changed since it was originally published. This content item won’t be edited but there will be a newer version published if warranted. Check the new publications and curriculum map for updates

Author: Charlotte Davies / Editor: Liz Herrieven / Codes: MuC9, SLO5 / Published: 11/02/2020

Elsie is a four year old who has come into the Emergency Department because her parents are concerned that she is not walking. Three days ago she was playing with her aunt and hurt her leg. Since then she has not been weight-bearing and her parents have been carrying her everywhere.

She has been given regular paracetamol and ibuprofen at home for the pain. She has been off her food but is drinking well. She has no diarrhoea or vomiting. Her observations are: RR 18, SpO2 98% on room air, heart rate 89, capillary refill <2 seconds, temperature 37.5oC, GCS 15.

We’ve all seen limping children. Maybe some of us have even had limping children. Some departments have excellent management strategies and pathways. Some don’t. Here’s some of our thoughts, musings and suggestions.

What are your differential diagnoses at this point?

There are lots of things that can cause a child not to weight-bear. There is a history of trauma here so careful examination is needed to determine which bit of the leg might need imaging. The delay in presentation should make you question whether there are any concerns about non-accidental injury – if in doubt, you will need to speak to the Paediatricians and your safeguarding team.

Clinical Examination Findings

The child looks very well. She is holding her hip in flexion and abduction. She is very tender on palpation of the greater trochanter and the iliac crest, but has no femoral pain. She will not move her leg, even when encouraged to do so by tickling her foot. When you try and move her leg she is sore on all hip movements but has a good range of movement.

Because of her history of trauma and pain, you send her for a pelvis and hip x-ray. These x-rays are normal.

What next?

The history of trauma given doesn’t account for the leg positioning in the child and the amount of pain that she is in. A fracture has been excluded so it would be sensible to think of atraumatic causes of a limping child. Perhaps the trauma history was a red herring? There are lots of causes for limp in children and many hospitals have detailed algorithms to follow.
Perthes’ disease or avascular necrosis of the hip is unlikely as the radiographs are normal. The signs of Perthes’ may be subtle ranging from almost normal, with a slightly wider epiphysis, to fragmentation and widening of the femoral neck (coxa magna). In Perthes’, the joint capsule is under maximum pressure in the position of flexion and internal rotation – the child may hold their hip in external rotation.

A SUFE is unlikely – she’s too young! Slipped upper femoral epiphyses normally affect 10-15 year olds, boys more often than girls. Half of those affected have bilateral disease. Make sure you do the “frog view” which can reveal an early posterior slip. You may need to look carefully – the S line may be useful. Klein’s line will normally intersect with the lateral portion of the femoral epiphysis (Trethowan’s sign).

Transient synovitis or irritable hip is a possibility. Irritable hip is a common cause of limp in children and again is more common in boys than girls. There is no reliable way of differentiating irritable hip from septic arthritis. Clinical suspicion, supplemented with blood test results, is important and this formulates the kocher criteria.

The other common causes of limp are unlikely in Elsie because of the normal investigations so far and the short history.

Figure 1

The protocol on Paediatric Pearls website has some very clear guidelines .

Toddler’s Fracture normally affect younger children (1-3 years old). This is usually after a fall and normally affect the distal 1/3 of the tibia. Sometimes a long leg cast is needed. The x-ray findings can be subtle, and many places treat clinical symptoms without definite abnormal radiology.

“Other” causes of limp tend to be higher on paediatrician’s minds, but that’s no reason why we in the ED shouldn’t consider them too, especially in the child with slightly more chronic, or non-isolated symptoms. Could there be a limb-length discrepancy. Primary bone malignancy is rare – Ewing’s sarcoma and osteogenic sarcoma are often found in older children, but limping may be a presenting symptom of leukaemia. Stroke and other neurologic symptoms are a possibility – there’s normally a risk factor like sickle cell disease that makes you suspicious.

Rickets is starting to become more common again in the UK. A wrist x-ray is often chosen to see whether there are the classic features of splaying and concavity of the metaphysis, and irregularity or fraying of the physis. A chest x-ray would demonstrate the rachitic rosary appearance.

Elsie had some blood tests which showed white cells of 8.9, neutrophils of 5.5 and a CRP of 95. She had two red flags or risk factors (non-weight-bearing and a raised ESR equivalent) and a presumed diagnosis of septic arthritis was made. She was taken to theatre and 5ml of pus was aseptically aspirated from her hip, confirming the diagnosis.

Septic Arthritis

Septic arthritis is a differential diagnosis for the limping child in all age groups. Because the hip is very vascular, haematogenous spread of infection can happen easily, making it a common joint to be affected.

Figure 2: BMJ | 28 AUGUST 2010 | VOLUME 341

Pain is the most common presenting symptom of septic arthritis followed by joint swelling, fevers, sweats and rigors. Clinical suspicion is the most important diagnostic tool in diagnosis in these cases as even apyrexial patients with a normal white cell count can have septic arthritis.

Figure 3: Kocher’s Criteria for diagnosing Septic Arthritis is well validated, and “NEWT” is a useful mnemonic for remembering it.

N – Non weight bearing
E – ESR >50
W – WCC > 12
T – Temperature > 38.5oC

Likelyhood of septic arthritis
4/4 – 99%
3/4 – 93%
2/4 – 40%
1/4 – 3%

Many places now use CRP instead of ESR. A high CRP has a strong association with septic arthritis, and a high CRP in a non-weight-bearing child gives a 74% probability of the diagnosis.

X-rays may show an effusion, if the effusion is big enough. Ultrasound normally identifies the presence of an effusion and some radiographers can tell you whether it is thick fluid (pus) or not. Fluid may be present in both transient synovitis and septic arthritis.


• If the history of trauma doesn’t match the severity of the clinical findings, follow the atraumatic limp in children guidelines.
• Red flags for limping children are non-weight-bearing, raised ESR or CRP, raised WCC and history of fever.
• A limping child has septic arthritis until proven otherwise.
• If in doubt, always refer the child to orthopaedics for review.

Further Reading