Predictors of outcome of subdural empyema in children.

OBJECTIVE: The aim of this study was to report the etiology, clinical features, microbiology, surgical outcome, and predictors of outcome of spontaneous subdural empyema (SDE). METHODS: The authors conducted a retrospective study in a tertiary hospital. Children up to 18 years of age, with a diagnosis of SDE with infective etiology, were included in the present cohort. Patients with posttraumatic, postsurgery, and tubercular origin of SDE were excluded from the study. The Glasgow Outcome Scale was used for outcome assessment at the end of 3 months. For analysis purposes, the demographic data, clinical features, radiological data, microbiology, type of surgery, and complication data were categorized, and univariate and multivariable logistic regression analyses were performed to identify the factors associated with outcome. RESULTS: Ninety-eight children were included in the study and the mean age was 10.9 years. Otogenic origin (34.7%) was the most common source of infection, followed by meningitis (14.3%). The mean duration of symptoms was 12 days. Seventy-six children presented with Glasgow Coma Scale (GCS) score > 8 and the supratentorial location was the most common location. Almost 75% of the children underwent craniotomy or craniectomy and the rest had burr-hole evacuation. Beta-hemolytic Streptococcus (10%) was the most common organism isolated. Cerebral venous thrombosis (CVT; 10.2%) was the most frequent complication in this cohort. The other complications were infarction (6.1%), new-onset seizure (4.1%), and bone flap osteomyelitis (4.1%). Thirteen cases had a recurrence of pus collection, which was more common in the craniotomy group than in the burr-hole group. Age (p = 0.02), GCS score ≤ 8 (OR 8.15, p = 0.001), CVT (OR 15.17, p = 0.001), and presence of infarction (OR 7, p = 0.05) were strongly associated with unfavorable outcome. In multivariable logistic regression analysis, only GCS score ≤ 8 (p = 0.01), CVT (p = 0.02), and presence of infarction (p = 0.04) had a significant impact on unfavorable outcome. CONCLUSIONS: Prompt diagnosis and immediate intervention is the goal of management of SDE, especially in children as a delay in diagnosis can result in unconsciousness and secondary complications such as CVT and infarction, which adversely affect outcome.