Background/Purpose: Pamidronate (PAM) may be effective in diminishing pain and permanent bone deformities in chronic non-bacterial osteomyelitis (CNO). Whole body magnetic resonance imaging (WB MRI) can be used to assess anatomical site, extent and inflammatory activity of bone lesions. The aim of this study was to evaluate the clinical and radiological outcome of children with CNO after one year of PAM treatment.

Methods: A retrospective evaluation of children diagnosed with CNO according to the Bristol criteria ADDIN RW.CITE{{216 Roderick,M. 2014}}(1). Children were treated with i.v. PAM 1mg/kg (max 60 mg/day) for 3 consecutive days every 3 months. We included 17 children (8 girls and 9 boys) median age 11 years (range 5-13). Medical history, clinical assessments and inflammatory biochemistry were obtained. WB MRI (1,5 Tesla) short tau inversion recovery (STIR) and T1-weighted images were performed at baseline, median -1 month (range -6-0) and after one year PAM treatment, median 12 months (range 9-19). Number of bone lesions are listed per patient (median, range).

Results: Comorbidities were JIA (n=3) and IBD (n=1); none of the children had psoriasis. All children had been treated with NSAID as first line treatment. Further medical history was antibiotics (n=1), methotrexate (n=7), steroids (n=7) and anti-TNFα treatment (n=3). Six children had elevated sedimentation rate at baseline, which normalized after one year.  At baseline there were in total 52 clinical symptomatic lesions, median 3 (range 1-8) and 113 radiological active lesions, median 6 (range 2-14). The most common sites of bone lesions were femur (19%), tibia (19%) and spine (12%). Six children had spinal lesions, median 1 (range 1-4), and 5 children had non-erosive lesions in relation to the sacroiliac (SI)-joints. After one year 11 children were still radiological active based on MRI with total number of 68 lesions, median 4 (range 1-21) (Table 1). Only one active spinal and one SI-joint lesion persisted. Ten children had new or worsening of lesions, median 2 (range 1-15). Three children had deformation of 11 vertebral bones (10 thoracic, 1 cervical) without sign of active inflammation. After one year the clinical symptoms in the radiological active group were; five with total pain relief, 5 with partial pain relief and 2 without any pain relief. In 6 children lesions resolved completely as assessed by MRI, but pain was still reported in 4 of them.

Conclusion: PAM given for one year is a potent second-line treatment for CNO. Axial lesions and SI-joint lesions respond well to PAM. Persistent pain or pain progression may exist despite total resolution as assessed by WB MRI. ADDIN RW.BIB (1) Roderick M. Efficacy of pamidronate therapy in children with chronic non-bacterial osteitis: disease activity assessment by whole body magnetic resonance imaging. Rheumatology (Oxford) 2014; 53(11):1973-1976  

Tab. 1. Distribution and number of active bone lesions before and after pamidronate treatment