Community-acquired staphylococcal musculoskeletal infection in infants and young children: necessity of contrast-enhanced MRI for the diagnosis of growth cartilage involvement.

OBJECTIVE

Previous studies have reported that contrast-enhanced sequences do not increase the sensitivity of MRI for the diagnosis of pediatric osteomyelitis and are not needed in the absence of edema on unenhanced MRI sequences. Invasive skeletal infections due to community-acquired Staphylococcus aureus are increasingly encountered in infants and young children and have a proclivity for involvement of both the unossified growth cartilage and the metadiaphyseal bone marrow of the extremities. The study objective is to assess the diagnostic efficacy of contrast-enhanced and unenhanced MRI sequences for the diagnosis of community-acquired S. aureus extremity skeletal infection in infants and young children.

MATERIALS AND METHODS

A retrospective review was conducted of the clinical charts and imaging studies of patients younger than 18 months diagnosed with invasive community-acquired S. aureus skeletal infections from 2001 to 2009 at a large children's hospital. Sensitivity was calculated for the detection of skeletal infection on contrast-enhanced and unenhanced MRI sequences. The p values were calculated using the Fisher exact score method. The kappa value for interobserver reliability was determined.

RESULTS

Community-acquired S. aureus skeletal infections were noted in 34 extremity sites in 25 patients, five of whom had more than one site of disease. The affected skeletal sites were metaphyseal or metadiaphyseal bone marrow only in 16 cases (47%), unossified growth cartilage only in nine cases (26%), and both the unossified growth cartilage and metaphyseal or metadiaphyseal bone marrow in nine cases (26%). In seven of the nine cases of isolated involvement of the unossified growth cartilage, the cartilage appeared normal on unenhanced sequences and the diagnosis was made only by the demonstration of hypoenhancing or nonenhancing foci in the cartilage after gadolinium-based contrast agent administration. In five of the nine cases of infection of both the unossified growth cartilage and metaphyseal or metadiaphyseal bone marrow, neither the cartilage nor bone marrow appeared abnormal on unenhanced sequences. Therefore, 12 cases of skeletal infection would have been missed without the inclusion of contrast-enhanced sequences. Follow-up extremity radiographs were available for 10 patients, eight (80%) of whom exhibited growth disturbances.

CONCLUSION

Skeletal infection caused by community-acquired S. aureus in infants and young children manifests differently than in older children, including a propensity for involvement of the unossified growth cartilage and potentially occult nature of both cartilage and bone marrow involvement on unenhanced MRI sequences.