Back

Pneumococcal empyema in Toronto, Canada, 1995–2006

Abstract number: 1733_700

Lee T., Green K., McGeer A., Low for the Toronto Invasive Bacterial Diseases Network D.

Background: Several recent studies have documented increases in the rate of pneumococcal empyema (PEMP). We examined trends in the occurrence of PEMP in Toronto, Canada from 1995 to 2005.

Methods: Population based surveillance for invasive pneumococcal disease (IPD) in residents of Toronto and Peel region, Ontario, Canada (population 3.9M) has been on-going since 1995. PEMP includes cases with a positive pleural fluid culture, and bacteraemic cases with a clinical diagnosis of empyema. Data are collected from patients and their physicians. Antibiotic histories are available for patients from 2000 on.

Results: From 1995 to 2005, 4,496 episodes of IPD have been identified: 114 (2.5%) are PEMP. The incidence of empyema increased from 0.22/100,000/y in 1995–7 to 0.34/100,000/y in 2003–5, while the incidence of all IPD decreased from 13.5 to 7.9/100,000/y. Patients with PEMP were not significantly different from others with IPD in age (median 56y vs 50y, P=.14), gender (61% vs 55% male, P=.22), or proportion that were healthcare acquired (6.0% vs 4.2%, P=.34), but were somewhat more likely to have a chronic underlying illness (68% vs 59%, P = 0.05). Isolates from PEMP were more likely to be of serotype (ST) 1 (5/32 PEMP vs 96/3,902 other, P = 0.001) and ST12F (6/80 vs. 95/3,854, P = 0.02); they were also more likely to be resistant to levofloxacin (4/101 vs 29/3,933, P = 0.004) and erythromycin (17/101 vs 439/3,956, P = 0.08) but not penicillin (5/101 vs. 154/3,956, P = 0.60). Patients with PEMP were more likely to failing out-patient antibiotic therapy when admitted with IPD (11EMP of 127 cases failing antibiotics, vs 52EMP of 1,343 other cases, P = 0.01). Patients who were failing fluoroquinolone (FQ) outpatient therapy were more likely to have a diagnosis of empyema than other patients (4/26, 15% vs 65/2,275, 3%, P = 0.008). All 4 EMP patients with FQ resistant isolates were failing FQ: none of these patients died, compared to 14/58 (24%) other patients (P=.26). Similarly, 2/3 EMP patients with erythromycin resistant isolates were failing macrolides (ML); neither of these two patients died, compared to 14/62 (23%) other patients.

Conclusion: The incidence of PEMP is increasing. Some serotypes are more likely to cause PEMP than others. Patients with pneumococcal infection who are failing ML or FQ therapy because their isolate is resistant are more likely to present with empyema, but less likely to die than other patients, perhaps because these antibiotics control systemic but not local disease.

Session Details

Date: 31/03/2007
Time: 00:00-00:00
Session name: European Society of Clinical Microbiology and Infectious Diseases
Subject:
Location: ICC, Munich, Germany
Presentation type:
Back to top