Two Multistate Outbreaks of Shiga Toxin--Producing Escherichia coli Infections Linked to Beef from a Single Slaughter Facility --- United States, 2008

Two Multistate Outbreaks of Shiga Toxin--Producing Escherichia coli Infections Linked to Beef from a Single Slaughter Facility --- United States, 2008
Weekly
May 14, 2010 / 59(18);557-560


During May--August 2008, state and local health and agriculture departments, the U.S. Department of Agriculture's Food Safety and Inspection Service (FSIS), and CDC investigated two multistate outbreaks of Shiga toxin--producing Escherichia coli O157 (STEC O157) with distinct pulsed-field gel electrophoresis (PFGE) patterns. Investigations into each outbreak included epidemiologic analysis of food exposures, microbiologic testing, and food distribution tracebacks. This report summarizes the results of those investigations. During May 27--August 25, 2008, a total of 99 persons (64 from the first outbreak and 35 from the second outbreak) from 18 states had confirmed illness with an STEC O157 isolate indistinguishable from the outbreak PFGE patterns. A case-control study conducted as part of the first investigation found a statistically significant association with purchase of ground beef from one large grocery chain (matched odds ratio [mOR] = 9.3). Traceback investigations for both outbreaks led to the same slaughter facility, resulting in multiple nationwide recalls of ground beef, intact beef, and beef products used to produce ground beef. This is the first report of two distinct STEC O157 outbreaks traced to a single slaughter facility and the first documented report of outbreaks linked to STEC O157 contamination of intact beef cuts ground by a retail chain. To help reduce the risk for outbreaks of STEC O157, the public health community should continue to educate consumers regarding the proper measures to take when handling and consuming ground beef.

Outbreak 1

On June 14, 2008, the Ohio Department of Health posted a cluster of seven STEC O157 isolates with an indistinguishable PFGE pattern (pattern 1)* on PulseNet, the national molecular subtyping network for foodborne disease surveillance. Most of the ill patients resided in an urban area of central Ohio. At approximately the same time, the Michigan Department of Community Health was interviewing patients in an STEC O157 cluster with the same PFGE pattern as the patients in Ohio. Several Michigan patients reported consuming ground beef before their illness, and some reported purchasing it at one of several stores belonging to a large national retail grocery chain (chain A). On June 18, state and local health and agriculture departments, FSIS, and CDC's OutbreakNet Team initiated an investigation to determine the extent and source of the outbreak.

A confirmed case was defined as illness in a person with an STEC O157 isolate having both a PFGE pattern indistinguishable from PFGE pattern 1 and multiple-locus variable-number tandem repeat analysis (MLVA) pattern 1,† and illness onset (or specimen collection date, if onset date was unavailable) after May 26. Sixty-four confirmed cases were identified in 12 states: Alabama (one case), Florida (one), Georgia (eight), Indiana (three), Kentucky (one), Michigan (23), New Jersey (one), New York (one), Ohio (21), Texas (one), Utah (two), and West Virginia (one). Illness onset dates ranged from May 27 to August 7 (Figure). Median age of patients was 21 years (range: 1--71 years); 40 (63%) were female. Thirty-two (59%) of 54 patients with available information were hospitalized, and two developed hemolytic uremic syndrome. No deaths were reported.

During June 20--25, the Ohio and Michigan state health departments and CDC conducted a case-control study examining ground beef and several other exposures linked to STEC O157 infections in previous outbreaks. All patients in Ohio and Michigan identified through June 25 were eligible for interview. Controls were identified through reverse-digit directory and matched on neighborhood, as identified by reverse directory, and age group (<18, 18--60, and >60 years); controls were asked about their food exposures for the week preceding the onset date of their matched case-patient. Data were analyzed using a conditional logistic regression model with exact methodologies because of the small sample size. Twenty-five case-patients and 65 controls from Michigan and Ohio were interviewed with a goal of three controls per case. Twenty-one case-patients were matched: 18 with three controls, two with two controls, and one with one control. Illness was significantly associated with purchasing ground beef from a store owned by chain A (mOR = 9.3) (Table). Seven samples of ground beef purchased at chain A stores, collected from patient homes, yielded STEC O157 with PFGE pattern 1 and MLVA pattern 1.

Traceback investigation of the ground beef from chain A identified a single large beef slaughter facility in Nebraska that supplied beef to chain A stores in outbreak-associated areas of Michigan and Ohio. On June 25, chain A issued a recall of all ground beef sold by its Michigan and Ohio stores in the regions where patients resided.§ On June 30, the slaughter facility initiated a nationwide recall, which was expanded July 3 to encompass beef products used to produce ground beef totaling 5.3 million pounds.¶

Outbreak 2

On July 28, 2008, PulseNet alerted CDC regarding a cluster of STEC O157 isolates from eight states with an indistinguishable PFGE pattern (pattern 2).** Initial telephone calls by local and state health departments revealed that several patients reported consuming undercooked ground beef. On July 29, local and state health and agriculture departments, FSIS, and CDC initiated an investigation to determine the extent and source of the outbreak.

A confirmed case was defined as illness in a person with an STEC O157 isolate indistinguishable from both PFGE pattern 2 and MLVA pattern 2 with an isolation date on or after July 1. Thirty-five cases were identified in eight states: Colorado (one case), Connecticut (one), Massachusetts (eight), New Jersey (two), New York (two), Ohio (seven), Pennsylvania (eight), and Virginia (six), with illness onset dates ranging from June 29 to August 25 (Figure). Median age was 18.5 years (range: 1--70 years). Median age was 18.5 years (range: 1--70 years). Of the 35 patients, 19 (54%) were female. Nineteen (63%) of 30 patients with available information were hospitalized, and one developed hemolytic uremic syndrome. No deaths were reported.

Thirty patients were interviewed using the questionnaire used in outbreak 1. Twenty-four (80%) patients reported eating ground beef in the home, and 13 (54%) of those reporting ground beef consumption indicated that it was raw, bloody, or pink. Seventeen (57%) of 30 patients had exposure to ground beef purchased from one of two grocery chains (chain B and chain C). One sample of raw ground beef purchased at chain B (a regional upscale chain) and three samples of ground beef purchased at chain C (a national upscale chain), all collected from patient homes, and one sample of intact beef collected from chain B before grinding yielded STEC O157 with PFGE pattern 2 and MLVA pattern 2.

Traceback investigations indicated that the same slaughter facility linked to outbreak 1 was the sole source of the contaminated beef from chain B and was one of multiple suppliers to chain C. On July 28, chain B announced a retail-level recall of ground beef products.†† Chain C announced a recall on August 8,§§ and the slaughter facility initiated a nationwide recall of beef products linked to outbreak 2. This recall was expanded on August 14 to include approximately 1.36 million pounds of additional intact beef cuts.¶¶

Slaughter Facility Control Measures

open here to see the full-text:
Two Multistate Outbreaks of Shiga Toxin--Producing Escherichia coli Infections Linked to Beef from a Single Slaughter Facility --- United States, 2008