Article Text

Download PDFPDF

Transient bone marrow oedema in a child
  1. L Kröger1,
  2. P Arikoski1,
  3. J Komulainen1,
  4. R Seuri2,
  5. H Kröger3
  1. 1Department of Paediatrics, Kuopio University Hospital, FIN-70211 Kuopio, Finland
  2. 2Department of Radiology, Kuopio University Hospital, FIN-70211 Kuopio, Finland
  3. 3Department of Surgery, Kuopio University Hospital, FIN-70211 Kuopio, Finland
  1. Correspondence to:
    Dr L Kröger

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.

Transient bone marrow oedema (TBMO) is an uncommon condition associated with joint and bone pain on activity. The most common localisation is the proximal femur, and it mainly affects only one bone.1 Reports of bone marrow oedema in children are scarce.2–4


An 8 year old boy was referred to hospital owing to difficulty in walking. He had started ski jumping 2 weeks earlier, but his history showed no injuries in the extremities. He complained of pain in the knees, ankles, and wrists. Mild fever and signs of upper respiratory tract infection were detected and reactive arthritis was suspected. Treatment with naproxen was started and he was discharged. Ultrasound investigation showed only slight symmetrical oedema in the ankle joints. Haemoglobin, white blood cell count, and thrombocytes were all within normal limits.

During the next 2 weeks, the pain was localised in both feet and the patient was hardly able to walk. Palpation and compression of both feet was very painful. An x ray examination which was carried out 2 weeks after the first visit showed marked osteopenia in the tarsal area and in distal parts of the tibia and fibula (fig 1). A magnetic resonance imaging (MRI) scan was performed 4 weeks after the beginning of the symptoms showed marked bone marrow oedema (fig 1). Transient osteoporosis was suspected. Bone scintigraphy showed areas of hyperperfusion both in the feet and the hands. An MRI scan was subsequently obtained of the hands also, which also showed bone marrow oedema. Axial bone mineral density was measured by dual x ray absorptiometry and was normal. Serum calcium, phosphate, parathyroid hormone, alkaline phosphatase (AP), and vitamin D metabolites were normal. However, bone markers were raised (table 1).

Table 1

 Laboratory data at baseline and during follow up

Figure 1

 A radiograph showing marked osteoporosis in the tarsal and metatarsal bones. In the MRI scan, hyperintesity in the tarsal bones can be seen as a sign of oedema in T2 weighted images.

During follow up the pain became less intensive and after 3 months, the condition had resolved clinically, although an MRI scan was still abnormal. MRI normalised in 8 months. However, serum osteocalcin values remained raised, suggesting increased bone remodelling (table 1).


There are no data on the incidence of TBMO in children. However, it seems to be a relatively rare cause of foot and ankle pain also in adults.5 In another study of 1123 patients referred for MRI imaging of the foot, 72 patients with oedema-like bone marrow abnormalities were registered.6

The aetiology of TBMO is unknown but it may be associated with local vascular disturbances, microtrauma, bone contusion, or altered biomechanics.6,7 The x ray examination is usually either normal or shows localised osteoporosis, as in our case. However, osteoporosis is a rare finding in bone biopsies.8 Histological studies have suggested ischaemic origin.9

The definitive diagnosis of TBMO is made by MRI. Bone contusion and bone bruises normally manifest as focal areas of low signal intensity in T1 weighted images and increased signal intensity in T2 weighted images, whereas bone oedema shows diffuse changes of intensity.1 In our patient marked bone marrow oedema was seen in both feet and hands.

Because of the unknown aetiology of bone changes we measured biochemical markers of bone metabolism. Serum AP values were within the normal range but both osteocalcin and procollagen type I C-peptide were high and remained high, although the clinical condition was relieved. However, except for AP, interpretation of these results is difficult because there are no established normative data for bone markers in children.10

The treatment of TBMO is not well established. Core decompression and vasodilatator treatment with iloprost have been suggested.4,7 Conservative treatment aims at protecting bone from weight bearing, both to prevent collapse of the articular surface and provide relief from pain. In this patient the use of anti-inflammatory drugs and rest relatively rapidly relieved pain, although changes in the MRI scan were seen for several months.

The aetiology of TBMO in our patient remains unknown: his history disclosed no trauma, but it is possible that ski jumping and hence either bone contusion or abnormal stress might have been the predisposing factors.