Metaphyseal osteomyelitis in children: how often does MRI-documented joint effusion or epiphyseal extension of edema indicate coexisting septic arthritis?

Joint effusions identified by MRI may accompany osteomyelitis and determining whether the joint effusion is septic or reactive has important implications on patient care.

Determine the incidence of epiphyseal marrow edema, joint effusions, perisynovial edema and epiphyseal non-enhancement in the setting of pediatric metaphyseal osteomyelitis and whether this may be used to predict coexisting septic arthritis.

Following IRB approval, we retrospectively evaluated children who underwent MRI and orthopedic surgical consultation for suspected musculoskeletal infection between January 2011 and September 2013. Criteria for inclusion in the study were microbiologically/pathologically proven infection, MRI prior to surgical intervention, long bone involvement and age 0–18 years. MRI exams were independently reviewed by two faculty pediatric radiologists to confirm the presence of appendicular metaphyseal osteomyelitis, to evaluate extent of edema, to determine subjective presence of a joint effusion and to assess perisynovial edema and epiphyseal non-enhancement. Any discrepant readings were reviewed in consensus. Charts and operative notes were reviewed to confirm the diagnosis of osteomyelitis and septic arthritis.

One hundred and three joints with metaphyseal osteomyelitis were identified (mean age: 7.1 years; M:F 1.3:1), of whom 53% (55/103) had joint effusions, and of those, 75% (41/55) had surgically confirmed septic arthritis. The incidence of coexisting septic arthritis was 40% in the setting of epiphyseal edema, 74% in epiphyseal edema and effusion, 75% with perisynovial edema, 76% with epiphyseal non-enhancement and 77% when all four variables were present. Of these, the only statistically significant variable, however, was the presence of a joint effusion with a P-value of <0.0001 via Fisher exact test. Statistical significance for coexisting septic arthritis was also encountered when cases were subdivided into intra-articular vs. extra-articular metaphyses (P-value  =  0.0499). No statistically significant difference was found between patients younger than 24 months and those older than 24 months.

Patients with joint effusions identified by MRI, in the setting of metaphyseal osteomyelitis, should be presumed to have septic arthritis until proven otherwise. Epiphyseal extension of edema, perisynovial edema and epiphyseal non-enhancement in the setting of metaphyseal osteomyelitis are not helpful predictors in differentiating reactive and pyogenic joint effusions. Osteomyelitis at a site with an intra-articular metaphyses, however, is more likely to have concurrent septic arthritis.