OBJECTIVE Video-assisted thoracoscopic debridement (VATD) is a well-established intervention to treat pediatric empyema. There is ongoing controversy at what stage in the treatment algorithm it should be utilized. To shed further light onto this debate, we reviewed our institutional experience looking for factors predicting treatment failure or complications of VATD. METHODS We retrospectively analyzed data on patients that had undergone VATD for empyema from 1995 to 2008. We used independent sample t tests and Chi-square tests (SPSS) for statistical analysis. RESULTS One hundred and fifty-two procedures in 151 patients [81 male (53.6%)] were identified. In 146 (96.7%) the etiology of the empyema was pulmonary, in 3 (1.98%) due to an infectious abdominal process and in 2 (1.3%) due to abdominal trauma. 118 patients (78.1%) were transferred from outside hospitals. 107 (70.1%) underwent VATD primarily, 44 (29.1%) following another procedure. The overall complication rate was 13.8%, most of which were minor. Treatment failures occurred in seven patients, resulting in three reoperations; two patients died. The average length of stay was 10.1 days, but was significantly longer if VATD followed another procedure or if a complication occurred. The risk for complications correlated with older age (6.2 vs. 8.8 years, p = 0.023) and lower hematocrit on admission (31.1 vs. 27.9%, p = 0.006). CONCLUSIONS VATD provided effective treatment for pediatric empyema. Complications were mostly minor, occurring more frequently in older patients and those with a lower admission hematocrit. Early VATD decreased the length of hospitalization.