Elizabethkingia Outbreak Source Remains a Mystery

During a session at the ASM Microbe 2016 conference on Monday, June 20, 2016, a Centers for Disease Control and Prevention (CDC) officer presented data on the investigation of the Elizabethkingia anophelis outbreak that started in Wisconsin and spread to two other states.

There are three Elizabethkingia species which are known to cause human illness: Elizabethkingia anophelis, Elizabethkingia meningoseptica, and Elizabethkingia miricola. These bacteria can be found in soil and water, and are known to be multidrug resistant. Historically, E. anophelis, which has been identified as the species causing this outbreak, was first known as Flavobacterium meningosepticum, then as Chryseobacterium. It was not until 2005 that it was placed under the genus Elizabethkingia, and in 2011 named as its own species, anophelis. Due to its misidentification, E. anophelis may account for more than 30% of E. meningoseptica cases identified before 2011, according to the CDC. E. anophelis has been reported in Hong Kong, Singapore, Central African Republic, and more recently, the United States. A recently published case-series showed that 17 out of 21 Elizabethkingia bacteremias in Hong Kong were E. anophelis. Most of these cases were hospital-acquired infections caused by genetically diverse isolates of the bacteria in a mostly adult population.

During her presentation, Maroya Walters, PhD, ScM, epidemiologist at the CDC, stated that between December 29, 2015 and January 5, 2016, a cluster of E. meningosepticum was isolated from six patients’ blood samples at different facilities in the state of Wisconsin. At this point, the Wisconsin Department of Health “initiated and performed state-wide surveillance in Wisconsin”, as well as “multiple nationwide calls for cases to figure out if this outbreak was truly localized in Wisconsin, or if it just happened to be detected in Wisconsin,” according to Walters. She further went on to explain that these cases were identified in Wisconsin because some of the isolates were in the same hospital for consecutive days at a time, and were identified by “an astute clinical microbiologist who recognized multiple isolates of the same unusual organism in a short period and alerted public health.”

Originally, there were ten patients, five men, five women, all white, with a median age of 73 years. Although E. anophelis is identified as a healthcare associated infection, only one of the original 10 patients had a (short) hospital stay, three had IV exposure, and seven had percutaneous exposures. Of the 10 patients, six died; however, due to these individuals’ serious comorbidities, cause of death is unknown. Members of Walters’ team began wondering if this was an actual outbreak, or if it was a change in clinical labs’ detection procedures, which could have led to “an artificial increase of Elizabethkingia.”

The CDC assessed the genetic relatedness of the isolates through PFGE and optical mapping. Healthcare practitioners were notified of the outbreak by the Wisconsin Department of Health and were urged to notify the Wisconsin Division of Public Health of any Elizabethkingia, Flavobacterium, or Chryseobacterium spp isolates. Walters explained that the Wisconsin Division of Public Health also “disseminated the antimicrobial susceptibilities for the outbreak strains, so that empiric therapies would actually cover Elizabthkingia, because initially they did not.” To report a case as part of the outbreak, the CDC investigation team set the limitations as follows: E. anophelis isolation from any body site must match the outbreak PFGE pattern, and specimens must have been collected on or after November 1, 2015.

After failing to identify common healthcare exposures among the cases, the team began looking at community exposures: contaminated products and contaminated tap-water (many patients had respiratory symptoms as well as cellulitis, which led the team to believe that water may be the culprit). As the investigation progressed, the number of patients increased. The investigation team then went on to look at person-to-person transmission as well as healthcare facility environments as potential outbreak sources. With the increase of cases, hospitalization rates went from 10% to 55%; however, as the number of those presenting with an E. anophelis infection increased, the number of patient deaths decreased to almost half.

The team began to simultaneously test several hypotheses as to the source of the outbreak. There was no evidence of person-to-person transmission within the healthcare settings. E. anophelis was identified in 1 out of 5 hospitals where the water and drains were tested; however, these samples did not match the outbreak strain. Food samples also tested negative, and were therefore discarded as a potential source of the outbreak. The team also cultured water and biofilm from patient homes. The only culture which was positive for the outbreak strain was found in a patient’s bathtub where the patient had soaked bedding in soap and water for at least a week. The rest of the patient’s home, including tap water and sink drains, all tested negative.

To date, this has been the largest Elizabethkingia outbreak; however, the source of the outbreak is yet to be identified. The CDC investigation team found that the tight geographic clustering of cases was inconsistent with nationally distributed products, such as food, medical products, and personal care products. According to Walters, “Elizabethkingia clinical isolates [were] more common than previously recognized. Passive surveillance from other states suggests less than 5 [cases were identified] per state per year.”

The Wisconsin Division of Public Health is currently performing a 5-year, statewide, retrospective, active laboratory surveillance for all bacteria previously misidentified as E. anophelis.