E. coli O157:H7 Outbreak Associated with Consumption of Unpasteurized Milk, Kentucky, 2013

BACKGROUND:   In September 2014, the Kentucky Department for Public Health and local health departments investigated an outbreak of E. coli O157:H7 infections and pediatric Hemolytic Uremic Syndrome (HUS) cases all having a history of consumption of unpasteurized milk.

METHODS:   We established a case definition as: Any individual who has a positive laboratory result of E. coli O157:H7, Shiga toxin detected, E. coli Shiga-like, and/or has a clinically compatible case of Hemolytic Uremic Syndrome (HUS) with illness onset after 8/12/14. Case patients’ caregivers were interviewed with a standard foodborne illness questionnaire. Answers were analyzed to identify common exposures and establish a case definition.  A retrospective review of reported exposure information for all E. coli cases was conducted and compared to the “FoodNet Population Survey Atlas of Exposures, 2006-2007”. State-wide active surveillance for HUS cases was requested. The state public health laboratory prioritized all E. coliisolates for serotyping and pulsed-field gel electrophoresis (PFGE) analysis. An environmental assessment was conducted and environmental samples collected for laboratory analysis from a dairy which provided unpasteurized milk and milk products to a co-op common to the case patients (Dairy A and Co-op A).

RESULTS:   Five children from three families, residing in two counties in Kentucky were identified as part of this outbreak.  All cases reported consuming unpasteurized milk purchased from Co-op A, supplied by Dairy A.  Four of the 5 cases developed HUS.  Clinical specimens were obtained from four of the children, representing all affected families.  Laboratory testing of the clinical isolates identified E. coli O157:H7 PFGE Xbal pattern EXHX01.2401 and BlnI pattern EXHA26.0071 in 3 of 4 specimens.  This particular PFGE pattern combination was unique to Kentucky and to the PulseNet database, possibly indicating a unique source of infection. Milk samples from partially consumed and un-opened containers were collected from the case patients’ homes. Dairy A was visited on two separate occasions, where 25 environmental samples were collected, including manure, environmental swabs, and milk.  All 25 samples collected tested negative for E. coli O157:H7. 

CONCLUSIONS:   Although the environmental samples collected from the case patients’ homes and Dairy A were negative for E. coli O157:H7, epidemiologic evidence indicated that the consumption of unpasteurized milk from Dairy A, sold by Co-op A, was the probable source of this outbreak.  This was further supported by the unique PFGE pattern identified in the clinical specimens of unrelated case patients that were only linked by consumption of unpasteurized milk from Dairy A.