Archives of Disease in Childhood | 2019
GP10\u2005Chronic nonbacterial osteomyelitis; the irish experience
Abstract
Background Chronic nonbacterial osteomyelitis (CNO) is a rare autoinflammatory disease affecting bone with an estimated prevalence of 1 in 105. Untreated CNO can result in complications such as vertebral compression fractures and leg length discrepancy. Limited data on different clinical phenotypes and efficacy of treatments (such as NSAIDs, steroids, methotrexate, bisphosphonates and biologic agents) makes prognosis challenging. Objectives To describe the clinical phenotype of an Irish cohort of patients with CNO including treatment response. Methods Retrospective chart review of current patients attending the National Centre for Paediatric Rheumatology and the Rheumatology Department in the Children’s University Hospital. Clinical charts, radiology and histology reports were reviewed. All data was coded and statistical analysis was performed in R. Results Clinical charts of 37 patients with CNO were reviewed. The median age at onset was 8.5 years, median follow-up was 3 years and the median number of sites 3 (1–21). F:M was 2.7:1. 94% had multifocal disease. 35% had a personal history of an associated disease such as psoriasis, inflammatory arthritis or inflammatory bowel disease. A first- or second-degree family history of an associated disease was present in 54%. All patients underwent whole-body MRI prior to diagnosis. All patients with unifocal disease underwent biopsy to outrule infection or malignancy. Treatment escalation beyond NSAIDs was required in 51% with biologic agents being used in 84% of those requiring second-line treatment. The indications for second-line treatment were in keeping with recent CARRA guidelines; persistent active disease on NSAIDS (n=12), the presence of spinal lesions (n=5), the presence of a physeal lesion (n=1) and pre-existing JIA (n=1). Biologic agents led to symptomatic improvement in all patients while none of the patients in this cohort responded to bisphosphonates; response to methotrexate was variable. Conclusions There was a low incidence of unifocal disease and a high incidence of either a personal or family history of associated diseases in this cohort. Those with unifocal disease were less likely to have an associated inflammatory disease or to require second-line treatment. The rate of progression to biologic treatment was higher than that reported most other cohorts. Multinational collaborative consistent data collection with clearly defined outcomes is required to ascertain the association between different CNO phenotypes and outcomes.