Georgia epidemiology report, Vol. 20, no. 10 (Oct. 2004)

October 2004

volume 20 number 10

Multi-Restaurant Outbreak of Hepatitis A Associated with Green Onions in Central and North Georgia, September October 2003

BACKGROUND Hepatitis A is characteristically a self-limited illness caused by infection of the liver with an enterically transmitted picornavirus, Hepatitis A virus (HAV). The incubation period ranges from 15-50 days, and in most cases is around 28-30 days. Symptoms include malaise, anorexia, nausea, fever and jaundice. The severity and duration of symptoms are variable and are associated with age. Infections in young children are often asymptomatic, whereas infections in older children and adults are typically symptomatic and can last for several weeks. HAV infection rarely produces a fulminant hepatitis except in some patients with underlying liver disease. Hepatitis A is not known to cause chronic liver disease. Persons at high risk for hepatitis A infection include: 1) men who have sex with men, 2) injecting and non-injecting drug users, 3) persons who travel to developing countries (or countries where hepatitis A is endemic), 4) and persons who work with nonhuman primates. Hepatitis A vaccine is recommended for persons at high-risk, and for children who live in communities with sustained high rates of disease. In most cases, hepatitis A infection is transmitted by personto-person contact, primarily through sexual and household contact. Foodborne outbreaks of hepatitis A are infrequent and have been attributed to eating certain foods (e.g., raw or undercooked shellfish, strawberries, green onions, chives, and other produce items) and by eating food contaminated by an infected food handler. This report describes a large foodborne outbreak of hepatitis A infections associated with eating green onions in restaurants throughout north and central
Georgia in mid-August, 2003.
INTRODUCTION During the week of September 22, 2003, the Georgia Division of Public Health (GDPH) received more than 30 reports of acute hepatitis A, which represented a sharp increase from the usual average of 10 cases per week. On September 26, GDPH began a formal investigation to identify the source of the outbreak. Initial inquiries identified three clusters of acute hepatitis A associated with eating at three different restaurants. Two of the restaurants were part of a national chain, and were located in neighboring counties. To maintain confidentiality, these restaurants will

be referred to as "Restaurant A" and "Chain B" restaurants. Simultaneous restaurant-associated outbreaks of hepatitis A identified in two other southeastern states raised suspicion that a common food item distributed to multiple states might be the source of this outbreak.
METHODS Cases of hepatitis A were identified primarily through laboratory reports received from commercial and hospital laboratories, in accordance with state notifiable disease reporting requirements. Additional case-finding activities included informing Georgia physicians about the increase in hepatitis A and asking them to report any confirmed or suspect cases of hepatitis A to GDPH immediately. As reports were received, cases were interviewed using an open-ended hypothesis-generating questionnaire to assess the date of onset of illness, presence of conventional risk factors (e.g., contact with a confirmed or suspect case, travel outside of the United States or Canada, an employee or child in day care), restaurant exposures, food-shopping habits, and exposures to food prepared outside the home during the 2-6 weeks before symptom onset. Although many ill persons reported eating at restaurants during their incubation period, no single restaurant or restaurant chain could account for more than a small proportion of illnesses. Three individual restaurants (Restaurant A and two Chain B restaurants) were named by four or more ill persons as places they ate during the two to six weeks before onset of illness. Information obtained from the hypothesis-generating questionnaire was used to narrow the focus and direct the next steps of the investigation.
Case Control Study: Restaurant A The hypothesis-generating questionnaire showed that many of the cases reported common restaurant exposures between August 13 and August 16, 2003. Therefore, we undertook a case control study among persons who ate at Restaurant A between August 13 and 16, 2003. Restaurant patrons were asked to recall menu items consumed during this period. A case was defined as a person with an acute onset of symptoms consistent with hepatitis A and/or serum IgM reactivity to HAV and who reported eating at Res-

taurant A between August 13 and 16. Controls were persons who ate at Restaurant A during this period and had no history of hepatitis symptoms (jaundice and/or dark urine) during September. Controls were identified through credit card records and by interviewing well dinner companions of cases.
Laboratory Commercial and hospital laboratories routinely bank individual patients' sera for a limited time. The GDPH collaborated with several laboratories to collect anti-HAV IgM positive residual serum during September and October for further analysis. Anti-HAV IgM serum specimens were submitted to the Centers for Disease Control and Prevention (CDC) for viral RNA sequencing. Viral sequences for Georgia cases were compared to each other as well as to sequences from other states that also reported restaurant-associated hepatitis A outbreaks. This methodology proved to be a useful adjunct to the epidemiologic investigation.
Environmental Environmental Health Specialists inspected Restaurant A and the two Chain B restaurants and found no significant restaurant violations. The three restaurants under investigation had one produce distributor (Distributor X) in common. Environmentalists inspected the produce distributor's facility and observed no food safety violations. When epidemiologic data showed that green onions were the source of the outbreak, the priority of the environmental investigation shifted to the traceback of the source of the green onions. We identified an out-of state produce supplier and reported our findings to the Food and Drug Administration (FDA), which can conduct national and international tracebacks and implement import bans.
RESULTS Between September 1 and October 15, 2003, 348 cases of acute hepatitis A were reported to GDPH, primarily in residents of central and northern Georgia. This was a sharp con-

The Georgia Epidemiology Report Via E-Mail
To better serve our readers, we would like to know if you would prefer to receive the GER by e-mail as a readable PDF file starting in 2004. If yes, please send your name and e-mail address to Gaepinfo@dhr.state.ga.us.

trast to the 37 cases of acute hepatitis A reported for the same period in 2002. Twelve cases were identified in out-of-state residents with a history of travel to Georgia during mid-August. The median age of cases was 38 years with a range of 1 to 89 years. Fifty-four percent of the cases were in males. Figure1 shows the dates of symptom onset for cases with onset between September 1 and October 15, 2003.
Case-Control Study Twelve persons with hepatitis A reported eating at Restaurant A during the incubation period. Of the 12 cases, 10 were interviewed in more detail and 9 reported dining there between August 13 and August 16, the period when the exposures were thought to have been most likely. We conducted a case-control study among persons eating at Restaurant A between August 13 and 16, 2003.
Eleven of the 12 cases and 63 control subjects agreed to an interview and were enrolled in the study. Although multiple foods were statistically associated with disease, the strongest association was with eating any of the three foods containing green onions. These foods included a Pico de Gallo salsa appetizer, an orzo & wild rice side dish, and a shrimp entre. Eighty-two per cent of cases but only 18% of controls ate foods that contained green onions, with a significant pvalue of less than 0.0001. In the initial analysis, cases were more likely than controls to have eaten cilantro, but stratification determined that green onions were independently associated with illness.
Laboratory results: Viral Sequencing CDC performed viral sequencing on 138 serum specimens from persons with laboratory-confirmed hepatitis A during September and October 2003. Eighty-eight percent were identical and were designated sequence "A". Cases with sequence A were scattered throughout northern and central Georgia, but were concentrated around the metropolitan Atlanta area and Macon (Bibb and Houston Counties) (Map 1.). We were able to locate serum specimens for five of the 12 cases that ate at Restaurant A; all five were identified as sequence A. Serum specimens for eight out-of-state cases who traveled to Georgia during the exposure period also yielded sequence A. The remaining specimens (n = 17) demonstrated several different viral sequences (B 1%, C 3%, D 2%, E 2%, F 1%, and H 1%) and probably represent cases unrelated to the incriminated green onions.
Environmental The environmental investigation revealed that green onions used by Restaurant A and the two "Chain B" restaurants on August 13 and for a few days thereafter came from same produce distributor, Distributor X, located at the Atlanta State Farmers' Market. Comparison of the delivery tickets from the produce distributor with the restaurant invoices during the exposure period

Number of Cases

10/13/2004

10/6/2004

9/29/2004

9/22/2004

9/15/2004

9/8/2004

9/1/2004

Figure 1. Acute Hepatitis A Cases by Date of Onset of Symptoms, September 1 - October 15, 2003, Georgia (n=271)
25 20 15 10
5 0
Onset of Symptoms
Hepatitis A Cases Associated With Green Onions In Georgia
9 -10/2003
Hep A Cases - Sequence A Counties Receiving Both Onion Lots Counties Receiving Onion Lot 34098 Counties Receiving Onion Lot 33958 County
N

40

0

Georgia Department of Human Resources Division of Public Health Epidemiology

40
Created: Source: Projection:

80 Miles
February 2004 Hepatitis A Outbreak Investigation UTM 1983, Zone 16

-2 -

revealed that two lots from a single supplier were distributed to all three restaurants under investigation as well as over 100 other restaurants located in central and north Georgia. Further investigation by the Food and Drug Administration showed that the source of the green onions for the Georgia outbreak and the outbreaks in Tennessee and North Carolina was produce firms in Mexico. The FDA imposed an import ban on four Mexican produce firms linked to outbreaks in the United States until further investigation could be undertaken.
CONCLUSIONS AND RECOMMENDATIONS An outbreak of over 300 hepatitis A cases in Georgia in September and October of 2003 was traced to two lots of green onions grown

in Mexico and distributed to over 100 restaurants across the central and northern part of the state. Three other restaurant-associated hepatitis A outbreaks in Tennessee, North Carolina and Pennsylvania investigated in the fall of 2003 were also attributed to contaminated green onions. The Food and Drug Administration traced the source of the onions for the Georgia, Tennessee, and Pennsylvania outbreaks to Mexico.
Green onions have been implicated previously and are increasingly recognized as a relatively hazardous food for transmission of hepatitis A. They can be contaminated at any point between the farm and consumption, and the hepatitis A virus can persist in the environment for months and is relatively heat stable. Green onions are generally eaten raw or minimally cooked, and washing or rinsing onions is unlikely to prevent illness. Contamination of green onions at the

source is being addressed. The Mexican health and agricultural authorities are collaborating with the Food and Drug Administration to improve the safety of Mexican produce entering the United States. The FDA reports that the Mexican government has implemented a program of regular inspection of growers. However, the only way that individuals can completely eliminate the risk of acquiring hepatitis A from eating green onions is to cook them thoroughly or be immunized against hepatitis A.
This article was written by Julie A. Gabel, D.V.M., M.P.H., and Julie S. Wolthuis, M.P.H., M.S.W.

This outbreak investigation entailed a robust collaborative effort by many persons in GDPH and other state and federal agencies. A rapid multi-agency response was essential to quickly pinpoint the source of the outbreak and to control any potential for ongoing exposure. We would like to recognize the many individuals and agencies for their contribution to this outbreak investigation. If you have any questions regarding this outbreak or hepatitis A in general, please call 404-657-2588 and ask to speak with Dr. Julie Gabel or Julie Wolthuis.

Restaurant-Associated Hepatitis A Outbreak Investigation Team
Georgia Division of Public Health Epidemiology Branch Paul Blake, M.D., M.P.H. Susan Lance, D.V.M., M.P.H. Julie Gabel, D.V.M., M.P.H. Katie Arnold, M.D. Cherie Drenzek, D.V.M., M.S. Julie Wolthuis, M.P.H., M.S.W. Tammi Stigger, M.C.P.H. Cindy Burnett, M.P.H. Cassandra Neal, Data Manager
Georgia Public Health Laboratory Bill Shea, Administrative Operations Manager Mahin Park, Ph.D., Director, Microbiology Services Sandra Watkins, M.S.A., Clinical Laboratory Manager, Immunology

Kim Kilgour, B.S., Clinical Laboratory Associate, Immunology Jenuryl Fluellen, A.S., Manager, Central Accessioning and Data Processing
Health District 1-2 (Dalton) Jamie Cope, M.P.H., Epidemiologist Larry Staton, Communicable Disease Specialist
Health District 3-4 (Gwinnett) Heidi Davidson, M.P.H., Epidemiologist Crystal Witherspoon, M.P.H., Epidemiologist Lisa Jackson, Communicable Disease Specialist David Hornsby, Environmentalist
Health District 5-2 (Macon) German Gonzalez, M.D., Epidemiologist Phil McDevitt, Communicable Disease Specialist

Tony W. Hester and Bert Tilton, Environmentalists
Georgia Department of Agriculture
Food and Drug Administration Southeast Regional Office Chet Morris, Director of Cooperative Programs Branch Kim Livsey, Dan Redditt, and Alan Tart, Food Specialists Russell Zablan and John Lloyd, Consumer Safety Officers
Centers for Disease Control and Prevention, National Center for Infectious Diseases Anthony Fiore, M.D., M.P.H. Ian Williams, Ph.D. Omana Nainan, Ph.D. Gilberto Vaughan, Ph.D. Guo-liang Xia, Ph.D.

Division of Public Health http://health.state.ga.us
Kathleen E. Toomey, M.D., M.P.H. Director
State Health Officer

Epidemiology Branch http://health.state.ga.us/epi
Paul A. Blake, M.D., M.P.H. Director
State Epidemiologist
Mel Ralston Public Health Advisor

Georgia Epidemiology Report Editorial Board
Carol A. Hoban, M.S., M.P.H. Editor
Kathryn E. Arnold, M.D. Paul A. Blake, M.D., M.P.H. Susan Lance-Parker, D.V.M., Ph.D. Kathleen E. Toomey, M.D., M.P.H. Angela Alexander - Mailing List Jimmy Clanton, Jr. - Graphic Designer

Division of Public Health Two Peachtree St., N.W. Atlanta, GA 30303-3186 Phone: (404) 657-2588 Fax: (404) 657-7517

Georgia Department of Human Resources

Please send comments to: Gaepinfo@dhr.state.ga.us

-3 -

The Georgia Epidemiology Report Epidemiology Branch Two Peachtree St., NW Atlanta, GA 30303-3186

PRESORTED STANDARD U.S. POSTAGE
PAID ATLANTA, GA PERMIT NO. 4528

October 2004

Volume 20 Number 10

Reported Cases of Selected Notifiable Diseases in Georgia Profile* for July 2004

Selected Notifiable Diseases
Campylobacteriosis Chlamydia trachomatis Cryptosporidiosis E. coli O157:H7 Giardiasis Gonorrhea Haemophilus influenzae (invasive) Hepatitis A (acute) Hepatitis B (acute) Legionellosis Lyme Disease Meningococcal Disease (invasive) Mumps Pertussis Rubella Salmonellosis Shigellosis Syphilis - Primary Syphilis - Secondary Syphilis - Early Latent Syphilis - Other** Syphilis - Congenital Tuberculosis

Total Reported for July 2004
2004 63 2856 15 0 73 1329 5 16 71 4 2 0 0 1 0 318 63 2 6 4 16 0 29

Previous 3 Months Total

Ending July

2002

2003 2004

197

237

176

8581

9063

8445

28

35

39

23

10

5

237

209

218

4858

4589

3761

22

18

31

117

121

71

146

202

169

2

11

21

2

6

3

9

5

4

0

0

0

10

6

6

0

0

0

583

636

639

370

395

189

28

30

10

73

131

35

164

209

24

177

201

72

2

5

0

170

137

93

Previous 12 Months Total

Ending in July

2002

2003

2004

685

773

653

36805

39060

37319

166

138

161

64

39

31

996

987

933

20614

20222

17416

112

82

119

792

553

723

528

645

782

11

33

45

4

12

10

46

34

28

4

1

2

27

36

32

0

0

1

2044

2258

2388

1356

2062

809

119

121

123

320

472

385

750

850

473

872

896

651

21

14

3

645

595

569

* The cumulative numbers in the above table reflect the date the disease was first diagnosed rather than the date the report was received at the state office, and therefore are subject to change over time due to late reporting. The 3 month delay in the disease profile for a given month is designed to minimize any changes that may occur. This method of summarizing data is expected to provide a better overall measure of disease trends and patterns in Georgia.

** Other syphilis includes latent (unknown duration), late latent, late with symptomatic manifestations, and neurosyphilis.

AIDS Profile Update

Report Period
Latest 12 Months: 10/03-09/04 Five Years Ago: 10/99-09/00 Cumulative: 07/81-09/04

Total Cases Reported* <13yrs >=13yrs Total

9

1,872 1,881

6

1,365 1,371

230

29,666 29,896

Percent Female
27.4
27.4
18.7

Risk Group Distribution (%) MSM IDU MSM&IDU HS Blood Unknown

34.0

6.7

2.1

14.6

1.5

41.1

31.7

11.9

3.5

19.9

2.1

30.8

46.3

16.3

5.2

14.3

1.9

16.0

Race Distribution (%) White Black Other

21.3 75.7

3.0

20.9 75.9

3.2

32.7 64.7

2.6

MSM - Men having sex with men

IDU - Injection drug users

HS - Heterosexual

* Case totals are accumulated by date of report to the Epidemiology Section

- 4 -