Title: The Changing Epidemiology of Musculoskeletal Infection in Children
Author(s): *Omar Gafur, Lawson Copley, MD, Tyler Hollmig, Lori Thornton, Shellye Crawford, Richard Browne; Children's Medical Center, Dallas
Purpose: To evaluate the current epidemiology of musculoskeletal infection at a tertiary pediatric medical center and to propose an alternative paradigm with which to categorize pediatric musculoskeletal infection based on severity of illness factors.
Methods: Children with musculoskeletal infection who were treated between January 1, 2002 and December 31, 2004 were retrospectively studied. A comparison was made to the epidemiology which was reported at the same institution in 1982 with respect to incidence, causative organism, and location of infection. Diagnoses of osteomyelitis, septic arthritis, pyomyositis, and deep abscesses requiring surgical drainage were established for each child based on magnetic resonance imaging (MRI) and intra-operative findings. A hierarchical disease classification system was proposed to account for the differences in clinical manifestations in musculoskeletal infection which were noted between the historical comparison group and the recent retrospective group. Children were subcategorized based on the identification in primary, secondary, tertiary or quaternary musculoskeletal tissues. An evaluation of disease severity within each diagnostic category and subcategory was based on mean values of C-reactive protein and erythrocyte sedimentation rate (ESR) at admission, causative organism (percent MRSA), number of surgical procedures, intensive care unit admissions, identification of deep venous thrombosis, and length of hospitalization. ANOVA and Tukey multiple comparison tests were performed to evaluate statistical significance.
Results: 554 children with musculoskeletal infection were studied retrospectively (osteomyelitis - 212, septic arthritis - 118, pyomyositis - 20, and abscess - 204). In comparison to the historical comparison group, the incidence of osteomyelitis and septic arthritis increased 6 and 2.2 fold, respectively (Table 1). We also noted: 1) the emergence of MRSA as the causative organism in children with osteomyelitis ¡V 20%, septic arthritis ¡V 4%, pyomyositis ¡V 25%, and deep abscesses ¡V 55% and 2) the disappearance of Hemophilus influenza type b(Table 2). Additionally, pyomyositis and deep abscesses accounted for 40% of our musculoskeletal infection cases. With our proposed disease classification model, we identified a linear increase in severity of illness parameters according to a hierarchy of tissue involvement (osteomyelitis>septic arthritis>pyomyositis>abscess). We further identified linear increases in severity of illness parameters with increasing number of tissue types involved within each diagnostic sub-category (Figure 1). The greatest severity of illness was noted in children who had osteomyelitis, septic arthritis, pyomyositis, and an abscess.
Discussion: The epidemiology of musculoskeletal infection has evolved at our institution. The change in incidence of osteomyelitis, the emergence of CA-MRSA, and the recognition of the clinical significance of pyomyositis and deep abscesses have been noteworthy changes from the previous experience. We have proposed a new paradigm of pediatric musculoskeletal infection concurrent with the changes in epidemiology and clinical manifestations of disease severity identified at our institution. This study has led us to modify our guidelines of evaluation and treatment to obtain an MRI early in the diagnostic process, to make a concerted effort to obtain tissue samples and identify the causative organism, and to utilize empiric antibiotic coverage to treat MRSA in children with osteomyelitis, pyomyositis and deep abscesses.