Multidrug-Resistant Salmonella Heidelberg Associated with Mechanically Separated Chicken at a Correctional Facility

Introduction

S almonella infection is a significant cause of morbidity and mortality worldwide. In the United States, an estimated one million foodborne Salmonella infections occur each year, resulting in more than 19,000 hospitalizations and almost 400 deaths (Scallan et al., 2011). Most patients develop diarrhea, fever, and abdominal cramps 12 to 72 h after consumption of undercooked or contaminated food (CDC, 2011). Though more than 2,500 serotypes of Salmonella are known to cause illness in humans, severity varies by serotype (Jones et al., 2008) and may depend on host factors and strain characteristics. For example, multidrug-resistant (MDR) Salmonella strains have been associated with an increased risk of severe infection, hospitalization and death (Lynne et al., 2009).

On November 29, 2013, administrators at a local correctional facility notified the Tennessee Department of Health (TDH) of three ill inmates transported to an emergency room with high fever, diarrhea, and/or vomiting. TDH suspected a common source of infection and launched an investigation. Using a multidisciplinary team approach involving epidemiologists, environmental health specialists, and laboratory personnel, we sought to identify the source and prevent ongoing transmission.

Materials and Methods

Data on ill inmates were collected and a case–control study was conducted. Identifiable information was protected following TDH outbreak investigation guidelines. A case-patient was defined as an inmate reporting diarrhea (≥3 loose stools in a 24-h period) between November 28 and December 3, 2013. One healthy control subject was matched to a case-patient by group living quarter (i.e., pod). Case-patients and controls were interviewed with a standard questionnaire that focused on clinical symptoms, activities, and foods consumed during November 22–November 29. Descriptive and matched case–control analyses were performed using Epi Info Version 7.1.3 software (CDC, Atlanta, GA).

An environmental assessment, including a food flow diagram, was conducted at the jail kitchen. Two raw unopened chubs of mechanically separated chicken (MSC) were sampled. MSC lot numbers were recorded and provided to the U.S. Department of Agriculture–Food Safety Inspection Service (USDA-FSIS).

Human and food specimens were evaluated for enteric bacterial pathogens using standard laboratory methods (Isenberg, 2007; AOAC, 2007). Salmonella isolates were subtyped using pulsed-field gel electrophoresis (PFGE) and antimicrobial susceptibility testing by the TDH Division of Laboratory Services and the Centers for Disease Control and Prevention (CDC) Enteric Diseases Laboratory Branch. MDR isolates were defined as showing resistance to at least one antibiotic in at least three antimicrobial classes (NARMS, 2013). Pansusceptible isolates were identified as susceptible to all antimicrobials tested.

Education regarding infection-control measures and stool collection was provided to jail administrators, and kitchen staff were educated on proper food preparation and handling.

Results and Discussion

Our investigation of this correctional facility outbreak identified 23 inmates meeting the case definition. Mean age of cases was 35 years (range: 20–59); 83% were male and 91% were white. Mean duration of illness was 6.8 days (range: 3–12). Abdominal cramps, fatigue, or nausea were reported by 96% (n = 22), 90% (n = 21) reported chills or headache, and fever was reported by 83% (n = 19). Twenty cases and 20 controls from 9 of 10 affected pods were enrolled in the case–control study. We were not able to identify a matched control for the remaining three patients. Cases and controls did not differ with regard to gender or race. No single food item or activity was statistically associated with illness.

Salmonella Heidelberg with an indistinguishable PFGE pattern was isolated from nine different stool specimens (PFGE pattern JF6X01.0080). Stool samples were unavailable for remaining cases. Two of nine human isolates displayed resistance to a total of five drug classes including third-generation cephalosporins (Table 1). Seven isolates were pansusceptible. Two inmates with ceftriaxone-resistant Salmonella Heidelberg experienced bloody diarrhea, vomiting, and fever ≥103°F as well as longer duration of illness (11 and 12 days) compared to average illness duration of 5.7 days (range 3–8 days) for inmates with pansusceptible Salmonella Heidelberg infections. Although broad conclusions about increased disease severity cannot be made based on this small investigation, Salmonella Heidelberg infection has previously been demonstrated to cause invasive infections and mortality (Hoffman, 2013). Additionally, case-patients with antimicrobial-resistant nontyphoidal Salmonella infections have been shown to have more severe outcomes (Krueger et al., 2014) and increased hospitalization among patients infected with ceftriaxone-resistant strains. Similarly, we described greater severity with ceftriaxone-resistant Salmonella infection.

The environmental health assessment documented that food was prepared by jail inmates overseen by two kitchen supervisors with ServSafe^® Certification (NRAEF, 2012). Food flow included partitioning of MSC chubs into large portions for cooking. Approximately 70 pounds of MSC was cooked in a tilt skillet and frequently stirred with a large spoon. Cooked product was an ingredient in several entrees including breakfast gravy and lunch spaghetti. Likely contributing factors included cross-contamination from inadequate utensil sanitation and undercooking of MSC. Pansusceptible Salmonella Heidelberg indistinguishable from the outbreak strain was isolated from one MSC chub. Another MSC chub contained pansusceptible Salmonella Enteritidis. After traceback investigation identified the MSC manufacturer and outbreak details were shared, approximately 33,840 pounds of MSC was voluntarily recalled.

In this outbreak, we identified MDR Salmonella Heidelberg associated with MSC. Concurrently, a large multistate investigation traced Salmonella Heidelberg to a single chicken producer (Rothrock et al., 2015). Salmonellosis associated with poultry and poultry products, including Salmonella Heidelberg, are not adulterants but are of growing concern. Furthermore, some poultry products, such as MSC, are used in institutional settings that house vulnerable populations such as prisoners (HHS, 2009). Mitigating risks of these products include adherence to USDA guidelines during poultry processing and storage, and food handler and public education about proper handling (USDA HACCP, 1999; FDA, 2015). Control measures to further reduce human pathogens in raw foods can decrease the likelihood of foodborne outbreaks.