Chronic Recurrent Multifocal Osteomyelitis: What is it and how Should it be Treated?

Hermann J Girschick; Christiane Zimmer; Guenter Klaus; Kassa Darge; Anke Dick; Henner Morbach

Disclosures

Nat Clin Pract Rheumatol. 2007;3(12):733-738. 

In This Article

The Case

The patient complained about pain in her spine at the age of 9 years. Her left knee, left hip and both feet were also painful. These symptoms were not constant, so it was 4 months before the patient was referred to a pediatrician. Examination revealed nothing remarkable, except dorsal back pain. Laboratory parameters, including C-reactive protein level, erythrocyte sedimentation rate, level of peripheral leukocytes, ferritin level, antinuclear antibody level, rheumatoid factor status, concentration of complement components C3 and C4, and serum levels of IgG, IgA, IgM, IgE and IgD, were all unremarkable. HLA-B27 was absent. Radiographs of the spine and chest were normal; however, MRI of the spine and pelvis revealed multiple inflammatory lesions of thoracic vertebrae numbers 6-11 and sacral vertebrae numbers 2 and 3. The patient was, therefore, referred to our institution.

In most cases, MRI of patients with chronic nonbacterial osteomyelitis (CNO) cannot completely rule out malignancy. The possibility of neuroblastoma was excluded using further diagnostic tests, including measurement of the urinary levels of homovanillic acid and vanillylmandelic acid and the serum level of neuron-specific enolase. A technetium bone scan showed the lesions already defined by MRI, in addition to a right metatarsal lesion and an epiphyseal lesion in the left distal femur. This multifocal pattern of lesions was indistinguishable from Langerhans' cell histiocytosis, another major differential diagnosis. A bone biopsy of the thoracic spine was, therefore, performed. The histological work-up revealed a lymphocytic and plasmacellular infiltrate, in addition to peritrabecular fibrosis. The immunohistology results showed that T lymphocytes, monocytes and plasma cells were present, which excluded malignancy from the diagnosis. Standard microbial and mycobacterial cultures and universal polymerase chain reaction, to amplify eubacterial and mycobacterial genes, showed no sign of bacterial infection.

Because the patient exhibited multiple bone lesions, a recurrent course, a biopsy consistent with chronic inflammation and no cultivable bacteria, chronic recurrent multifocal osteomyelitis (CRMO) was diagnosed.

The patient was administered naproxen (15 mg/kg/day); however, treatment had to be discontinued after 3 months because of the appearance of a putative allergic skin reaction. Ibuprofen was tried as an alternative therapy but was also not well tolerated and caused stomach pain. Sulfasalazine, a disease-modifying antirheumatic drug (DMARD), was then administered at a dose of 40 mg/kg/day. Sulfasalazine therapy was initially accompanied by a 3-week course of oral glucocorticoid treatment (prednisone 2 mg/kg/day for 1 week and then discontinued stepwise by 25% every 5 days). Significant clinical improvement was noted after 1 week; however, follow-up imaging studies, conducted 4 weeks after the initiation of sulfasalazine therapy, showed partial vertebral compressions of thoracic vertebrae numbers 8-10 (Figure 1A, B). MRI of the spine and extremities, performed after another 8 weeks of sulfasalazine treatment, showed a significant reduction of signal intensity in the T2-weighted images. The lesion in the epiphysis of the distal left femur, previously shown by the technetium bone scan, was also visualized, and thus confirmed, by this MRI. Therapy was discontinued at this time because the patient exhibited a minor elevation in the levels of liver enzymes.

MRI 6 months after diagnosis showed partial vertebral compressions, particularly of thoracic vertebrae numbers 8–10. (A) Despite anti-inflammatory treatment, T1-weighted imaging after intravenous gadolinium-DTPA revealed contrast enhancement of thoracic vertebrae numbers 8–11. (B) The strongly T2-weighted images, with fat-suppression (TIRM), no longer showed edematous changes of the bone marrow. MRI 1 year after the diagnosis of CNO revealed multiple compressions of thoracic vertebrae numbers 6–11 (C and D; arrows). (C) T1-weighted imaging after intravenous gadolinium-DTPA did not reveal contrast enhancement. (D) The TIRM images did not show bone edema. Arrows mark thoracic vertebrae numbers 6 and 11 in all images. Abbreviations: CNO, chronic nonbacterial osteomyelitis; DTPA, diethylene triamine penta-acetic acid; TIRM, turbo inversion recovery magnitude.

MRI 6 months after diagnosis showed partial vertebral compressions, particularly of thoracic vertebrae numbers 8-10. (A) Despite anti-inflammatory treatment, T1-weighted imaging after intravenous gadolinium-DTPA revealed contrast enhancement of thoracic vertebrae numbers 8-11.(B) The strongly T2-weighted images, with fat-suppression (TIRM), no longer showed edematous changes of the bone marrow. MRI 1 year after the diagnosis of CNO revealed multiple compressions of thoracic vertebrae numbers 6-11 (C and D; arrows). (C) T1-weighted imaging after intravenous gadolinium-DTPA did not reveal contrast enhancement. (D) The TIRM images did not show bone edema. Arrows mark thoracic vertebrae numbers 6 and 11 in all images. Abbreviations: CNO, chronic nonbacterial osteomyelitis; DTPA, diethylene triamine penta-acetic acid; TIRM, turbo inversion recovery magnitude.

Significant pain recurred 6 months later in the right knee and back. Arthritis was present in the right knee, which showed joint effusion. Since MRI revealed another fracture of the seventh thoracic vertebra, as shown by a local increase in signal intensity in the T2-weighted images, the symptoms were interpreted as a clinical flair. Another short-term glucocorticoid treatment period (prednisone 2 mg/kg/day for 1 week and then discontinued stepwise by 25% every 5 days) was successful in completely alleviating pain and arthritis. Low-dose prednisone treatment (0.1 mg/kg/day) was subsequently continued and a corset was applied to the trunk. Neither naproxen nor sulfasalazine treatment could be reintroduced because of the adverse effects previously experienced by the patient.

Concomitantly, the patient started to complain about chronic stomach pain. Gastroduodenoscopy showed mild mucosal inflammation. An infection by Helicobacter pylori was ruled out by microbial culture of the gastric mucosal biopsies and a C13 exhalation test; thus, omeprazole therapy was started. Another MRI 3 months later showed a stable situation, without further vertebral compression. Abdominal pain, however, persisted. A colonoscopy was performed (15 months after the diagnosis of CNO was made); considerable inflammation of the colon and terminal ileum was demonstrated. Crohn's disease was confirmed histologically by the presence of mucosal ulcers, fistulas and granulomas. The symptoms of Crohn's disease worsened significantly during the following weeks. A multimodal anti-inflammatory therapy, using local and systemic steroids (budesonide), local and systemic mesasalazine and azathioprine, was initiated after 1 week, when the histological work-up was received. The patient also complained of minor pain (2 on a scale of 0-10) in her feet. MRI was repeated another 1 week later and, as expected, significant compression of thoracic vertebrae (numbers 6-11) was still visible, without signs of active inflammation (Figure 1C, D). There was, however, inflammation of the metatarsal bones (os naviculare and cuneiforme mediale on the right side and os cuneiforme mediale on the left side) and a minor lesion in the left distal metaphysis of the femur (Figure 2). The patient was in clinical remission 4 months later.

MRI still revealed inflammatory lesions in the tarsal bones and femur 1 year after the diagnosis of CNO was made. T2-weighted images, with TIRM fat suppression, show bone marrow lesions or edema in the navicular and cuboid bones (A; arrows), including a small effusion in the ankle joint and minor lesion in the distal left femur metaphysis (B; arrow). T1-weighted images after intravenous gadolinium-DTPA administration revealed a signal enhancement of the lesions (not shown). Abbreviations: CNO, chronic nonbacterial osteomyelitis; DTPA, diethylene triamine penta-acetic acid; TIRM, turbo inversion recovery magnitude.

MRI still revealed inflammatory lesions in the tarsal bones and femur 1 year after the diagnosis of CNO was made. T2-weighted images, with TIRM fat suppression, show bone marrow lesions or edema in the navicular and cuboid bones (A; arrows), including a small effusion in the ankle joint and minor lesion in the distal left femur metaphysis (B; arrow). T1-weighted images after intravenous gadolinium-DTPA administration revealed a signal enhancement of the lesions (not shown). Abbreviations: CNO, chronic nonbacterial osteomyelitis; DTPA, diethylene triamine penta-acetic acid; TIRM, turbo inversion recovery magnitude.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.

processing....