Article Text


Refusal to walk in an afebrile well toddler

Statistics from

Q1: What is the likely diagnosis and what do the magnetic resonance imaging studies (figs 2 and 3; see p 570) show?

The likely diagnosis is L5–S1 discitis. Magnetic resonance imaging of the spine (fig 2) shows bulging both anteriorly and posteriorly of the L5–S1 disc with erosion of the anterior corners of both the L5 and S1 vertebra. There is evidence also of some oedema within the anterior aspects of the vertebral bodies. The radiological appearances are more likely to be consistent with a mild discitis. Repeat magnetic resonance imaging of the spine (fig 3) a year later shows resolution of the bony changes and a mild reduction in the disc signal at the L5–S1 level with also some reduction in the degree of bulging.

Q2: What are the alternative differential diagnoses of refusal to walk in a well afebrile toddler?

The diagnosis depends on the onset of walking difficulties. An afebrile well toddler who is slow to achieve the walking milestone might be a bottom shuffler, have benign familial hypermobile joint syndrome, have congenital hip dislocation, cerebral palsy, myopathy, neuropathy, or have inherited mild forms of defects in collagen synthesis such as osteogenesis imperfecta type IA or Ehlers-Danlos syndrome type III. An afebrile well toddler presenting with sudden onset of walking difficulties, as in our case, has to be investigated. Conditions that should be considered are shown in box 1. Acute lymphoblastic leukaemia (ALL) may present with musculoskeletal complaints. Clinical features suggestive of ALL are hepatosplenomegaly, anaemia, and bruising. There may be characteristic bony changes on radiography. In tuberculosis of the spine, the disc space is usually reduced in association with vertebral collapse. Areflexia or hyporeflexia and raised cerebrospinal fluid protein are suggestive of peripheral neuropathy. Investigations that should be performed in a toddler with refusal to walk, as in our patient, are full blood count, blood film, C reactive protein, erythrocyte sedimentation rate, blood culture, plain radiograph of the spine, bone scan, and spinal magnetic resonance imaging. In our case the x ray findings and magnetic resonance imaging appearances are consistent with the diagnosis of discitis.

Box 1: Main causes of sudden refusal to walk in an afebrile well toddler

  • Joint: septic arthritis, transient synovitis.

  • Bone: osteomyelitis, tuberculosis of the spine, bone tumour.

  • Nervous system: Guillain-Barré syndrome, spinal cord tumour.

  • Others: acute lymphoblastic leukaemia, Hodgkin’s lymphoma, neuroblastoma.


Discitis, an uncommon, usually benign and self limiting disease in children, tends to present with refusal to walk, stand, or sit.1 Other symptoms are fever, back pain, limping, abdominal pain, nausea, and irritability. Physical signs are difficult to elicit in the very young. It is an important observation by parents that our patient disliked having his nappy changed when flexion of the spine was obviously involved. Crawling was not a problem as he was comfortable with lumbar spinal extension. The pathophysiology of discitis is not fully understood. Most authors believe in an infectious etiology where there is haematogenous spread of the infecting organisms to the disc space via vascular channels in the cartilaginous vertebral endplates and in the immature discs. Others feel that discitis is caused by trauma, namely a partial dislocation of the epiphysis secondary to a flexion injury.2,3 Investigations show raised erythrocyte sedimentation rates and C reactive proteins. White blood cell count may be normal. Interestingly, the inflammatory markers of our case started to fall before the start of antibiotics. This may support the theory of a non-infectious aetiology. Studies have shown that only about 28% of blood cultures4 and only 25% of disc space biopsy specimens are culture positive (positive cultures usually yield Staphylococcus aureus).1 Plain anteroposterior and lateral spinal radiographs should be taken initially in any case of suspected discitis, although radiographic changes may be delayed for 2–8 weeks after the onset of symptoms. Bone scans may be abnormal one week after symptoms appear, but a normal bone scan does not exclude discitis5 as illustrated in our case. The sensitivity in diagnosing discitis is greater with computed tomography and magnetic resonance imaging compared with bone scans6 with magnetic resonance imaging becoming the diagnostic tool of choice. Treatment is controversial. Most authors recommend bed rest with or without external immobilisation and analgesia.4 While most routinely use antibiotics, others reserve antibiotics for patients with septicaemia. Antibiotics should be broad spectrum and penicillinase resistant (preferably a second generation cephalosporin), beginning with parenteral antibiotics when the child is toxic, and once non-toxic, conversion to oral therapy until the three week course is complete.6 Discitis does not seem to result in any adverse long term sequelae. The prognosis in our case is good as x rays do not show any significant disruption of the disc spaces.

Final diagnosis



View Abstract

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.

Linked Articles