Elizabethkingia Anophelis Sterile Site Infections Linked to a Large Primarily Community-Acquired Outbreak, Wisconsin, 2015-2016

Tuesday, June 6, 2017: 10:48 AM
420A, Boise Centre
Jordan L Dieckman , Wisconsin Department of Health Services, Madison, WI
Maroya S Walters , Centers for Disease Control and Prevention, Atlanta, GA
Lina I Elbadawi , Centers for Disease Control and Prevention, Atlanta, GA
Chulwoo Rhee , Centers for Disease Control and Prevention, Atlanta, GA
Krithika Srinivasan , Centers for Disease Control and Prevention, Atlanta, GA
Anna Kocharian , Wisconsin Department of Health Services, Madison, WI
Lindy Liu , Centers for Disease Control and Prevention, Atlanta, GA
Matthew Crist , Centers for Disease Control and Prevention, Atlanta, GA
Jeffrey P Davis , Wisconsin Department of Health Services, Madison, WI

BACKGROUND: Members of the genus Elizabethkingia are intrinsically multidrug-resistant Gram-negative bacilli and are rare, opportunistic human pathogens most frequently associated with healthcare associated-infections. However, a large outbreak of primarily community-acquired Elizabethkingia anophelis infections occurred among Wisconsin residents during December 2015 through May 2016. All patients were infected with the outbreak strain of E. anophelis. The source and mode of transmission of the outbreak strain remain unknown. To better understand the clinical course, we retrospectively reviewed medical records of patients with E. anophelis sterile site infection.

METHODS:  Medical records for the healthcare encounter during which E. anophelis was diagnosed and any subsequent hospitalization were retrospectively reviewed and abstracted using a standardized data collection instrument. We examined patient demographic features, underlying conditions, clinical presentation, clinical course and outcomes using EpiInfo 7.2.

RESULTS: Medical record abstraction was completed for 49 of the 58 patients with confirmed outbreak-related positive sterile site cultures. The initial positive sterile site specimens were collected from blood (94%), synovial fluid (2%), aortic wall (2%), and pleural fluid (2%) and were primarily collected in emergency room (67%) or inpatient (29%) settings. The median patient age was 77 (range: 19-101) years, 55% of patients were male. The median Charlson Comorbidity Index score was 4 (range: 0-8), and 73% of patients had ≥2 conditions considered in the CCI score. At the time of initial positive specimen collection, the most common symptoms reported were shortness of breath (43%) and weakness (43%); 2% had temperatures below 95oF and 29% had temperature above 100.4oF. Among the 47 (96%) patients who were hospitalized (duration range: 1-22 days), 22 were admitted into an intensive care unit (duration range: 1-19 days) and the most common primary admission diagnoses were pneumonia (15%), sepsis (15%), and cellulitis (11%). Among hospitalized patients, 21% died during their hospitalization. Patients who survived were discharged to home (59%), long-term care facility (16%), hospice (11%), assisted living (8%), independent living (3%), and inpatient rehabilitation (3%).

CONCLUSIONS: This outbreak was notable for community-acquired invasive infections that occurred among patients who, despite frequent underlying conditions, did not have obvious primary sites of infection to facilitate bloodstream invasion. Although initial clinical presentations were diverse with non-specific symptoms, recovery of E. anophelis from a sterile site resulted in serious, often fatal, illness, and should be regarded as a pathogen regardless of initial presentation. Public health officials should pursue investigation of clusters of E. anophelis infections when detected.