Empiric Antibiotic Therapy in 1402 Children With Parapneumonic Effusion/Pleural Empyema in Germany: A Long-term Surveillance Study

Abstract

Objective: This study investigated empiric antibiotic treatment (EAT), guideline adherence, antibiotic streamlining and clinical outcomes in 1402 hospitalized children with pediatric parapneumonic effusion/pleural empyema (PPE/PE).

Methods: A nationwide surveillance study collected data on EAT, clinical course/outcome, pathogens, susceptibility testing and antibiotic streamlining of children with PPE/PE in Germany between 2010 and 2018. Subgroups were compared using χ2 test/Fisher exact test, Mann-Whitney U test and linear regression analysis adjusting for patient age where appropriate.

Results: Complete data on EAT were available for 1402 children. In children with monotherapy (n = 567) and in children with combination therapy of 2 antibiotics (n = 589), the most commonly used antibiotics were aminopenicillin/beta-lactamase inhibitor [138/567 (24.3%) and 102/589 (17.3%)] and cefuroxime [291/567 (51.3%) and 294/589 (49.9%)]. The most common combinations with these beta-lactams were macrolides, aminoglycosides and clindamycin. We observed no difference in clinical severity/outcome between EAT with aminopenicillin/beta-lactamase inhibitor and cefuroxime, neither when used in monotherapy nor when used in combination therapy of 2 antibiotics. Species diagnosis of Streptococcus pneumoniae (n = 192), Streptococcus pyogenes (n = 111) or Staphylococcus aureus (n = 38) in polymerase chain reaction or culture from pleural fluid or blood resulted in a switch to an appropriate narrow-spectrum beta-lactam therapy in 9.4%, 18.9 % and 5.2% of children. In a subset of children with reported bacterial susceptibility testing, penicillin resistance was reported in 3/63 (4.8%) of S. pneumoniae and methicillin resistance in S. aureus was reported in 10/32 (31.3%) of children.

Conclusion: This study points to antibiotic overtreatment in children with PPE/PE, particularly the frequent use of combinations of antibiotics. Children receiving combinations of antibiotics did not show differences in clinical outcomes. The low rate of children with streamlined antibiotic therapy even upon pathogen detection indicates a necessity for antibiotic stewardship measures in PPE/PE and the need of investigating other potential therapeutic strategies as anti-inflammatory therapy.