A Potluck, 52 Cases and the First U.S. Outbreak of Enteroinvasive E. coli in 47 Years

On July 2, 2018, the North Carolina Division of Public Health learned that about three dozen members of an ethnic Nepali refugee community had been taken to hospitals with severe gastrointestinal illness. All had been at a potluck party on June 30. The state worked with the local health department and CDC to find the cause and stop further spread. CDC's molecular-guided laboratory testing identified the culprit as enteroinvasive Escherichia coli (EIEC), making this the first confirmed outbreak of EIEC in the United States in 47 years.

Who was ill

Investigators counted as a case anyone who ate food served at the party and then had diarrhea, vomiting, or a fever of at least 101°F (38.3°C). They found cases by reviewing medical records and by interviewing a convenience sample of attendees.

• About 100 people attended; 52 met the case definition.
• The median age was 31, ranging from 3 to 76.
• 28 (54%) were hospitalized, 13 (25%) had sepsis, and eight (15%) went to an intensive care unit.
• Everyone recovered, and no one caught the illness from a patient.

Forty-nine people, 35 of them ill, answered a questionnaire about symptoms, travel and what they ate, and were given advice on handwashing. Of the 35 ill people, 30 (86%) became sick on July 1, the day after the party. The median time from eating to first symptoms was 20.5 hours, with a range of 1 to 45.5 hours. Diarrhea affected 33 (94%): 27 (77%) described it as watery, 19 (54%) as mucoid, and two (6%) as bloody. Thirty-two (91%) had a fever.

The food

The meal included chicken curry, vegetable curry, rice, lentil soup, fried bread, hot and cold salads, and cake. No one reported imported food, and no one interviewed had traveled recently. No single dish was statistically linked to illness. The 37 people who ate chicken curry had a 47% higher risk of becoming ill than those who did not (risk ratio 1.47; 95% confidence interval 0.76–2.83), a difference that was not statistically significant. No leftover food was available to test.

Telling EIEC from Shigella

One hospital tested stool from 25 patients with a commercial multiplex PCR panel, and all were positive for "Shigella/EIEC." That panel cannot tell the two apart, and neither can routine clinical testing easily. At the North Carolina State Laboratory of Public Health, 24 of those specimens were negative by culture for Shigella and E. coli O157 and negative by PCR for the Shiga toxin genes stx1 and stx2. Specimens from 23 patients then went to CDC, which used a molecular-guided method to isolate colonies carrying ipaH, the gene that testing for Shigella/EIEC targets. In specimens from 12 patients these were identified as EIEC O8:H19, a serotype never reported with EIEC before.

Why it matters

EIEC causes dysentery and spreads through contaminated food or water and from person to person. Most infections occur in developing countries. The previous known U.S. outbreak, in 1971, was tied to imported cheese; more recent outbreaks in Italy (2012) and the United Kingdom (2014) were linked to contaminated vegetables. This investigation did not identify how the bacteria reached the party.

Shigella was the first suspect, because of the early PCR results and because EIEC is so rarely identified. But the picture did not fit. Shigella outbreaks in North Carolina usually spread from person to person in child care settings and cause milder illness. Because the two organisms are genetically and clinically so alike, only a molecular approach could separate them. The authors conclude that epidemiologists and laboratory scientists need to work closely together when findings do not match the first suspected cause.

Sources

• Herzig CTA, Fleischauer AT, Lackey B, et al. Notes from the Field: Enteroinvasive Escherichia coli Outbreak Associated with a Potluck Party — North Carolina, June–July 2018. MMWR Morb Mortal Wkly Rep 68(7). https://www.cdc.gov/mmwr/volumes/68/wr/mm6807a5.htm
• The report's epidemic-curve figure is not reproduced here.