Etiology and Treatment of Pediatric Pleural Empyema.

Abstract

Parapneumonic effusions (PE) complicate pediatric community-acquired pneumonia (CAP) in 40% of cases. It is estimated that empyema forms in more than half of such cases but evidence-based data regarding appropriate treatment is limited. This study evaluates treatment modalities as they relate to duration of hospitalization and cost of care in such cases. A quality improvement based retrospective chart review was undertaken to evaluate outcomes for empyema that were diagnosed between 12/00–3/04 at Children's Mercy Hospital in Kansas City, MO. Identification of empyema was based on strict criteria that was independently confirmed by 2 evaluating physicians. Cases were included if ultrasound and or CT showed pleural fluid loculation and septation OR pleural fluid was grossly purulent OR bacteria was identified on pleural fluid culture. Data abstracted included demographic data, radiographic imaging studies, pleural fluid analysis, treatment modality, length of hospital stay (LOS), and cost of care. 96 cases of CAP with PE were reviewed; 57 cases were classified as pneumonia, uncomplicated PE, 5 cases as necrotizing pneumonia/abscess but w/o empyema and 34 cases of empyema. Children with empyema ranged in age from 17 months–16 years (mean 5 years) and 20 were boys (59%). In 27 empyema cases, pleural fluid evaluation was performed. Empyema patients more often had neutrophil >90% and glucose <20 mg/dl. Those with LDH >10,000 were more likely to be bacteriologically confirmed (7/11). Bacteriologic diagnosis was confirmed in 38% of cases; gram-positive pathogens predominated and S. pneumoniae was most commonly identified. Ten cases occurred annually and S. pneumoniae remained consistent despite widespread implementation of PCV in our community. Treatment included 4 modalities: Group 1 (n=2): antibiotics (A) only; Group 2 (n=14): A + thoracostomy tube; Group 3 (n=10): A + tube + alteplase fibrinolysis; and Group 4a/b (n=3/5): VATS (early ≤7 d sxs; late >7 days sxs). Patients in Group 2 had the greatest failure rate and longest stay (11.5 days) as well as cost of care. Patients in Group 3 and 4a had the shortest LOS and cost of care (7 days and median cost $21,062). Empyema was identified in 35% of cases of CAP with PE and the annual incidence remained stable over the last 3 years. Cases caused by S. pneumoniae persist despite PCV and may represent non vaccine strains (type 1 and 3). Among invasive interventions, tube thoracostomy alone had longer LOS and more failures. Early VATS and intrapleural fibrinolysis have shorter stays and cost. More evidence based investigation is necessary to confirm these results; utilization of a strict definition of empyema would facilitate such work.