2 figures, 1 box omitted
A Salmonella serotype Enteritidis (SE) epidemic
emerged in the 1980s, when increasing numbers of infections were detected
in the Northeastern and Mid-Atlantic regions of the United States.1 In
the early 1990s, while SE rates in the Northeast began to decline, the SE
epidemic expanded to the Pacific region.2 Nationwide, the number
of SE isolates reported to CDC peaked at 3.8 per 100,000 population in 1995.
Although rates of culture-confirmed SE infection reported to CDC declined
to 1.9 by 1999, rates did not decline further through 2001, and outbreaks
continue to occur. Investigations of outbreaks and sporadic cases have indicated
repeatedly that, when a food vehicle is identified, the most common sources
of SE infection are undercooked and raw shell eggs.3,4 This report
describes two SE outbreaks associated with eating shell eggs and underscores
the need to strengthen SE-control measures.
During February-March 2001, outbreaks of gastroenteritis occurred among
inmates in four prison facilities of the South Carolina Department of Corrections
(SCDC). The first outbreak occurred in a men's facility (M1) on February 6.
The three other outbreaks, all occurring on March 2, affected a second men's
facility (M2) and two women's facilities (F1 and F2). Among 2,317 inmates
in the four prisons, 688 reported to prison infirmaries with gastrointestinal
symptoms (e.g., abdominal cramps, diarrhea, and nausea). Stool specimens from
ill inmates yielded SE phage types 2, 13a, and 23. No illness was reported
among SCDC staff members.
The South Carolina Department of Health and Environmental Control conducted
two case-control studies in M2 and F1, which shared a common kitchen. A case-patient
was defined as any SCDC inmate who reported to the prison infirmary with acute
gastrointestinal symptoms. Case-patients were selected at random from a list
of ill inmates. Controls were inmates without illness who were selected at
random from an inmate roster provided by the prisons and who were matched
by prison facility. A tuna salad served for lunch on March 2 was eaten by
88% of the male case-patients (odds ratio [OR] = 7.0; 95% confidence interval
[CI] = 1.8-30.5) and by 89% of the female case-patients (OR = 16.7; 95% CI
= 4.1-74.7). The tuna salad was prepared with eggs that were reportedly hard-boiled
by kitchen staff, who also were inmates. At the time of the outbreaks, all
eggs used by the four involved SCDC facilities were supplied from a single
vendor. Eggs supplied to M2 and F1 were traced back to the vendor's farm (Farm
A). In February 2001, eggs submitted from Farm A to the South Carolina Egg
Quality Assurance Program tested positive for SE phage types 2,13a, 22, 23,
and 28. Phage type 2 was the predominant SE strain isolated from both ill
patients and eggs from Farm A. To protect the inmates, SCDC switched to pasteurized
egg products in April 2001.
In June 2001, the Statistical Outbreak Detection Algorithm at CDC signaled
an increase in SE cases reported from North Carolina. The Division of Public
Health in North Carolina was alerted and began to review SE cases throughout
the state. The North Carolina State Laboratory of Public Health reported 51
cases in July and 31 in August, compared with 11 cases in each of those months
during 2000. Cases occurred throughout the state.
A case-control study was performed. A case was defined as culture-confirmed
SE in a resident of North Carolina reported during July 1–September
7, 2001. One to two neighbor controls were matched to each case. SE isolates
were subtyped by pulsed-field gel electrophoresis (PFGE) and phage typing.
Analysis of 53 patients and 78 controls indicated that illness was associated
with eating eggs (matched odds ratio [MOR] = 2.8; 95% CI = 1.1-9.5). Isolates
from 21 (40%) of 53 patients had PFGE pattern A. Analysis restricted to patients
with pattern A indicated a stronger association with egg consumption (MOR
= 10.7; 95% CI = 1.3-88.1). PFGE pattern A also was identified in isolates
from patients in the South Carolina SE outbreak. All isolates from SE patients
in both outbreaks that were PFGE pattern A also were phage type 13a. Among
14 random, nonoutbreak phage type 13a SE isolates tested subsequently at CDC,
seven distinct PFGE patterns were identified; none was PFGE pattern A. A traceback
of implicated eggs purchased from retail outlets in North Carolina was inconclusive
for implicating a farm.
D Drociuk, MSPH, S Carnesale, MD, G Elliot, LJ Bell, MD, JJ Gibson,
MD, South Carolina Dept of Health and Environmental Control. L Wolf, D Briggs,
B Jenkins, JM Maillard, MD, North Carolina Dept of Health and Human Svcs.
M Huddle, MPH, F Virgin, C Braden, MD, Div of Bacterial and Mycotic Diseases,
National Center for Infectious Diseases; P Srikantiah, MD, A Stoica, MD, T
Chiller, MD, EIS officers, CDC.
During 1990-2001, state and territorial health departments reported
677 SE outbreaks, which accounted for 23,366 illnesses, 1,988 hospitalizations,
and 33 deaths (CDC, unpublished data, 2002). Among the 309 outbreaks reported
with a confirmed vehicle of transmission, 241 (78.0%) were associated with
shell eggs, accounting for 14,319 illnesses. Of these, 10,406 illnesses occurred
during 1990-1995, and 3,913 occurred during 1996-2001. The overall decrease
in SE incidence and the decrease in the number of illnesses related to egg-associated
SE outbreaks during the last decade might be attributed in part to the implementation
of prevention measures, including on-farm control programs, egg refrigeration,
and consumer and food worker education. However, reported cases did not decline
during 1999-2001, and outbreaks associated with shell eggs continue to occur.
In the South Carolina outbreak, eggs from a farm that tested positive
for SE in February 2001 were distributed to the prisons in March, despite
the farm's participation in a voluntary, state-sponsored SE-control program.
This farm withdrew from the state program in April 2001. Phage type 2 was
the most common SE strain isolated in the South Carolina outbreak. This uncommon
phage type, which has accounted for 3% of SE outbreaks with reported phage
type since 1985, also was found on Farm A. Cases in the outbreak in North
Carolina shared the same SE PFGE pattern and phage type (13a) as some of the
South Carolina outbreak cases, suggesting a possible link to the same farm.
Eggs that reportedly were hard-boiled and used in a tuna salad were
the implicated vehicle in the South Carolina outbreak. A recent study demonstrated
that unless SE-containing eggs are exposed to boiling water until the yolk
is completely solidified, SE can survive the cooking process.5 Cross
contamination of the tuna salad by inmate food handlers also was possible.
To achieve sustained decreases in egg-associated SE illnesses, a concerted
prevention effort is needed from farmers to consumers.6 A key factor
in this effort is the implementation of farm-based measures to reduce SE contamination
of eggs during production. The implementation of such control programs in
Northeastern states in the early 1990s might have contributed to subsequent
decreases in human SE isolation rates in New England and Mid-Atlantic regions.7 One important control measure is microbiologic testing of hen houses
for the presence of SE. If SE is found on a farm during routine environmental
testing, eggs may be diverted to pasteurization. Evidence suggests that proper
implementation and oversight of farm-based control programs can result in
a reduction of SE infections among flocks in poultry houses.8 Farm
participation in current SE-control programs is voluntary, and the components
of programs vary. Future shell-egg safety measures should include greater
participation in farm-based control programs with microbiologic testing.
Both outbreaks described in this report occurred in the Southeastern
region of the United States. Compared with declining rates of SE infections
in other regions of the United States, the incidence of SE in Southeastern
states increased by 50% from 2000 to 2001. Ongoing surveillance of SE outbreaks
will be necessary to detect changes in trends of SE infection in this region.
Expansion of SE-prevention measures will be an important part of efforts to
prevent SE infections in the Southeast. This includes actively encouraging
farms to participate in SE-control programs, promotion of proper refrigeration
of eggs during storage and transportation, and education of food handlers
and consumers about food preparation. Retail and consumer buyers can specify
that suppliers provide only eggs produced from farms managed under an SE-control
program that is recognized by a state regulatory agency or a state poultry
The outbreak in South Carolina prisons was the largest SE outbreak in
2001. Because persons residing in institutions depend entirely on their institutions
for meals, the supply of contaminated foods to these settings can place large
populations at risk for developing foodborne diseases. Persons residing in
institutions, especially elderly persons in nursing homes or assisted living
facilities, are at higher risk for dying from outbreak-associated SE infections.9 During 1990-2001, a total of 83 SE outbreaks occurred in institutional
settings,* representing 12% of reported SE outbreaks. Of the 33 outbreak-associated
deaths, 22 (67%) occurred in institutional facilities, underscoring the importance
of using pasteurized egg products or in-shell pasteurized eggs for all recipes
requiring pooled, raw, or undercooked shell eggs for institutionalized persons.
Additional information about preventing SE infections associated with
eating raw or undercooked shell eggs is available at
http://www.cfsan.fda.gov/~dms/fs-eggs4.html. Information for retail and food service establishments
and institutional facilities about handling and cooking shell eggs is available
in the Food Code at
References: 9 available
*Institutional settings include nursing homes, independent living facilities,
assisted living facilities, childcare settings, campus cafeterias, prisons
and correctional facilities, and shelters.
Outbreaks of. JAMA. 2003;289(5):540-541. doi:10.1001/jama.289.5.540