Staphylococcus aureus dissemination to a preexisting subdural hematoma.

Abstract

To The Editors: Subdural empyema in infants and children is relatively uncommon and is usually caused by trauma, meningitis or sinusitis. Dissemination to a preexisting subdural effusion as a cause of subdural empyema is rare in adults and children. Recently we treated a patient who had Staphylococcus aureus dissemination to a preexisting subdural hematoma caused by child abuse. This 11-month-old male infant was admitted to our intensive care unit with fever and generalized seizures for 1 hr before admission. He had a history of fever for 7 days and a varicella rash. On admission the child's temperature was 102°F, blood pressure 121/58 mm Hg and respiratory rate 66/min. The skin showed multiple crusted lesions consistent with varicella. Otolaryngologic and systemic examinations were normal. Fundoscopic examination was normal. On neurologic examination the patient had bilateral hyperreflexia and positive Babinski sign. Laboratory tests were: erythrocyte sedimentation rate 120 mm/h; white blood cell count 9400/mm3 with a differential count of 56% neutrophils, 8% band forms, 21% lymphocytes and 12% monocytes. Hemoglobin was 6.3 g/dl; hematocrit 20% and platelet count 338 000/mm3. A computerized tomography scan of the head showed bilateral subdural effusions, left greater than right, and a mass effect with midline shift from left to right. A craniotomy was performed. Thick membrane overlaid the bilateral hematomas which consisted of altered blood on the left side and a lesser amount of thick sanguinous fluid on the right side. The cerebrospinal fluid glucose was 15 mg/dl and protein was 303 mg/dl; clotting of the fluid made a cell count impossible. Intravenous cefotaxime and nafcillin were initiated. Cerebrospinal fluid cultures grew S. aureus susceptible to methicillin. Seven days later the patient had repeat surgery for recurrence of bilateral subdural collections. Frank pus was irrigated and the epidural space was drained bilaterally. Subdural fluid cultures revealed S. aureus and the treatment was changed to vancomycin and rifampin for 2 weeks. The postoperative course was complicated by failure of the wound to heal, with persistent purulent drainage. A skeletal radiographic survey was normal and no retinal hemorrhages were noted. The clinical suspicion of child abuse was supported by results of an investigation by the local social and child protective services. After 3 weeks the child was discharged in a stable condition to foster parents. Subdural empyema in infants younger than 1 year of age is rare. In a series of 32 cases with subdural empyema in adults and children admitted during a 12-year period, there was only one infant younger than 1 year.1 In a 10-year review of patients with subdural empyema by McIntyre et al.2 the youngest patient was 10 years old. Subdural empyema in children may be associated with an infected subdural effusion in cases of meningitis, spread from a contiguous site (sinusitis or otitis), head trauma and complication of neurosurgery.1 A less frequent cause is bacteremic spread of a remote infection to the subdural space. In a MEDLINE search of the literature we could find only four previous case reports and all were adults. In these cases dissemination occurred after urinary tract infection or after gastrointestinal procedures.3, 4 In two infants infected bilateral subdural effusions without any apparent cause were described as a part of a larger series of patients with subdural effusions. Child abuse could have caused these cases.1 Common organisms isolated from subdural empyema are anaerobic and aerobic streptococci, S. aureus, anaerobes and less commonly Gram-negative organisms.1 Cerebrospinal fluid from patients with subdural empyema shows pleocytosis between 20 and 5200 white blood cells, high protein (40 to 2000 mg/dl) and low glucose. In our case the presence of a membrane overlying the hematoma and the black appearing blood on one side as opposed to the pink blood on the other side suggests that the hematomas were not recent and they were of different ages. The bilaterality and the different ages of the hematomas raised the suspicion of child abuse, which was later confirmed by social services. Since there was no other adjacent infection we assumed that the hematomas were infected during a transient S aureus bacteremia, possibly arising from a varicella lesion. Itzhak Levy, M.D.; Sunil Sood, M.D. Division of Infectious Diseases Scheider Children's Hospital Long Island Jewish Medical Center New Hyde Park, NY