Figure 1. Edema of the posterior paraspinal muscles is shown on axial T2-weighted image.
Figure 2. Enhancement of the posterior paraspinal muscles is seen on axial computed tomographic scan following intravenous contrast administration.
Figure 3. An epidural abscess with inflammatory change extending into the posterior paraspinal muscles is shown on sagittal (top) and axial (bottom) T1-weighted magnetic resonance images of the lumbar spine following intravenous gadolinium administration.
A repeated lumbar MRI (Figure 3) showed a large epidural abscess with erosion into the paraspinal muscles. The child was transferred to a tertiary care hospital for surgical drainage and has completely recovered. Staphylococcus aureus was grown from a culture specimen obtained directly from the abscess.
Spinal epidural abscess is rare in childhood. In this case, findings from the initial MRI and computed tomographic scan were confusing as the pus decompressed into the paraspinal muscles, resulting in a secondary pyomyositis. Rubin et al1 reviewed spinal epidural abscess in 1993, and they note that childhood spinal epidural abscess is usually of hematogenous origin, and S aureus is the usual etiologic agent (79% of all cases). It is seen more often in patients younger than 2 years or older than 12 years. The symptoms differ according to age. In a child younger than 2 years, neurological compromise at presentation is common. Older children often are seen for back pain and fever, although abdominal and hip pain are also presenting signs. The outcome correlates with the presence or absence of neurological signs at presentation. The definitive treatment is surgical drainage and antibiotic administration.
Obtaining an MRI is the diagnostic method of choice, although negative findings from MRI cannot exclude spinal epidural abscess.2 Most false-negative MRI findings seem to be secondary to motion artifact or adjacent similar-signal intensity from coexistent meningitis. To my knowledge, drainage into adjacent muscle groups, as in this case, has not been described in the literature.
Accepted for publication October 20, 1998.
Corresponding author: Don Seidman, MD, Elmhurst Pediatrics of the DuPage Medical Group, 103 Haven Rd, Elmhurst, IL 60126.
Radiological Case of the Month. Arch Pediatr Adolesc Med. 2000;154(4):415-416. doi: