Open Access Highly Accessed Open Badges Research article

Empyema associated with community-acquired pneumonia: A Pediatric Investigator's Collaborative Network on Infections in Canada (PICNIC) study

Joanne M Langley12*, James D Kellner34, Nataly Solomon56, Joan L Robinson56, Nicole Le Saux78, Jane McDonald9, Rolando Ulloa-Gutierrez1011, Ben Tan12, Upton Allen1314, Simon Dobson1011 and Heather Joudrey12

Author Affiliations

1 Department of Pediatrics, Dalhousie University, Halifax, Canada

2 IWK Health Centre, Halifax, Canada

3 Alberta Children's Hospital, Calgary, Canada

4 University of Calgary, Calgary, Canada

5 Stollery Children's Hospital, Edmonton, Canada

6 University of Alberta, Edmonton, Canada

7 Children's Hospital of Eastern Ontario, Ottawa, Canada

8 University of Ottawa, Ottawa, Canada

9 Montreal Children's Hospital, Montreal, Canada

10 University of British Columbia,Vancouver, Canada

11 Women's and Children's Hospital of British Columbia, Vancouver, Canada

12 University of Saskatchewan, Saskatoon, Canada

13 Hospital for Sick Children, Toronto, Canada

14 Hospital for Sick Children, University of Toronto, Toronto, Canada

For all author emails, please log on.

BMC Infectious Diseases 2008, 8:129  doi:10.1186/1471-2334-8-129

Published: 25 September 2008



Although the incidence of serious morbidity with childhood pneumonia has decreased over time, empyema as a complication of community-acquired pneumonia continues to be an important clinical problem. We reviewed the epidemiology and clinical management of empyema at 8 pediatric hospitals in a period before the widespread implementation of universal infant heptavalent pneumococcal vaccine programs in Canada.


Health records for children < 18 years admitted from 1/1/00–31/12/03 were searched for ICD-9 code 510 or ICD-10 code J869 (Empyema). Empyema was defined as at least one of: thoracentesis with microbial growth from pleural fluid, or no pleural fluid growth but compatible chemistry or cell count, or radiologist diagnosis, or diagnosis at surgery. Patients with empyemas secondary to chest trauma, thoracic surgery or esophageal rupture were excluded. Data was retrieved using a standard form with a data dictionary.


251 children met inclusion criteria; 51.4% were male. Most children were previously healthy and those ≤ 5 years of age comprised 57% of the cases. The median length of hospitalization was 9 days. Admissions occurred in all months but peaked in winter. Oxygen supplementation was required in 77% of children, 75% had chest tube placement and 33% were admitted to an intensive care unit. While similarity in use of pain medication, antipyretics and antimicrobial use was observed, a wide variation in number of chest radiographs and invasive procedures (thoracentesis, placement of chest tubes) was observed between centers. The most common organism found in normally sterile samples (blood, pleural fluid, lung biopsy) was Streptococcus pneumoniae.


Empyema occurs most commonly in children under five years and is associated with considerable morbidity. Variation in management by center was observed. Enhanced surveillance using molecular methods could improve diagnosis and public health planning, particularly with regard to the relationship between immunization programs and the epidemiology of empyema associated with community-acquired pneumonia in children.