An outbreak of gastroenteritis linked to seafood consumption in a remote Northern Aegean island, February-March 2010.

Abstract

Raw seafood consumption has frequently been linked to gastroenteritis and Norovirus is a common cause. In February and March 2010, there was an increase cases of gastroenteritis on the remote island of Agios Efstratios in the Northern Aegean Sea, Greece. A massive increase in seafood consumption associated with a religious festival a few days prior to the outbreak suggested seafood as the vehicle of transmission.
An outbreak investigation team visited the island in order to document the outbreak and find epidemiological evidence for its source. The whole island was used as a cohort in the retrospective cohort study that was conducted.
Sixty-four of the 181 participants reported having had symptoms of gastroenteritis, of which 34 were considered primary cases. People who consumed any raw seafood imported to the island were 21.5 times (95% CI: 8.95-51.8) more likely to develop symptoms of gastroenteritis in the 72 hours following exposure. Consumption of local seafood was not found to be a risk factor.
Despite the lack of laboratory evidence, all four Kaplan's criteria were met and the outbreak was thought to have been caused by Norovirus. The outbreak investigation demonstrates epidemiological methods for use in a remote setting, where the means for laboratory and environmental investigation may be absent or limited.

Full text

Page 1

© I Karagiannis, M Detsis, K Gkolfinopoulou, D Pervanidou, T Panagiotopoulos, S Bonovas, 2010. A licence to publish this material has been
given to James Cook University, http://www.rrh.org.au

1




O R I G I N A L R E S E A R C H
An outbreak of gastroenteritis linked to seafood
consumption in a remote Northern Aegean
island, February-March 2010
I Karagiannis1, M Detsis1, K Gkolfinopoulou1, D Pervanidou1, T Panagiotopoulos2, S Bonovas1
1Hellenic Centre for Diseases Control and Prevention, Athens, Greece
2Department of Child Health, National School of Public Health, Athens, Greece

Submitted: 21 April 2010; Revised: 31 July 2010; Published: 4 October 2010
Karagiannis I, Detsis M, Gkolfinopoulou K, Pervanidou D, Panagiotopoulos T, Bonovas S
An outbreak of gastroenteritis linked to seafood consumption in a remote Northern Aegean island, February-March 2010
Rural and Remote Health 10: 1507. (Online), 2010
Available from: http://www.rrh.org.au

A B S T R A C T


Introduction: Raw seafood consumption has frequently been linked to gastroenteritis and Norovirus is a common cause. In
February and March 2010, there was an increase cases of gastroenteritis on the remote island of Agios Efstratios in the Northern
Aegean Sea, Greece. A massive increase in seafood consumption associated with a religious festival a few days prior to the
outbreak suggested seafood as the vehicle of transmission.
Methods: An outbreak investigation team visited the island in order to document the outbreak and find epidemiological evidence
for its source. The whole island was used as a cohort in the retrospective cohort study that was conducted.
Results: Sixty-four of the 181 participants reported having had symptoms of gastroenteritis, of which 34 were considered primary
cases. People who consumed any raw seafood imported to the island were 21.5 times (95% CI: 8.95–51.8) more likely to develop
symptoms of gastroenteritis in the 72 hours following exposure. Consumption of local seafood was not found to be a risk factor.
Conclusion: Despite the lack of laboratory evidence, all four Kaplan's criteria were met and the outbreak was thought to have
been caused by Norovirus. The outbreak investigation demonstrates epidemiological methods for use in a remote setting, where the
means for laboratory and environmental investigation may be absent or limited.

Key words: gastroenteritis, Greece, norovirus, public health.

Page 2

© I Karagiannis, M Detsis, K Gkolfinopoulou, D Pervanidou, T Panagiotopoulos, S Bonovas, 2010. A licence to publish this material has been
given to James Cook University, http://www.rrh.org.au

2

Introduction

Raw seafood consumption is known to be associated with
various public health threats. Outbreaks caused by
Norovirus, hepatitis A virus (HAV), Vibrio spp (more so
Vibrio parahaemolyticus and Vibrio cholerae) and a number
of other pathogens have frequently been documented1,2. Very
recently, a number of gastroenteritis cases clusters due to
Norovirus and linked to oyster consumption were reported
throughout Europe3.

On 19 February 2010, the Hellenic Centre for Disease
Control and Prevention (HCDCP) was informed about an
increased number of patients with gastroenteritis symptoms
on the remote island of Agios Efstratios, in the Northern
Aegean sea, Greece. Because seafood is massively
consumed on Ash Monday throughout Greece (on 15
February 2010), seafood items were the suspected vehicles
of the outbreak. The island’s two physicians reported that the
symptoms, which were mostly mild, were compatible with a
viral disease. In the immediate following days, a structured
trawling questionnaire was used to interview some patients
about water and food items consumption in the 3 days
preceding symptom onset. This revealed that 4 of 5 patients
(80%) had consumed raw seafood items allegedly imported
to the island from nearby Limnos and shared with locals as a
treat on Ash Monday. These items could not be backtraced.

According to the preliminary information available, all four
Kaplan’s criteria4,5 were met and, although there were no
laboratory results for Norovirus, it is highly likely to have
been the causative pathogen. Kaplan’s criteria are: (i) a mean
(or median) illness duration of 12 to 60 hours; (ii) a mean (or
median) incubation period of 24 to 48 hours; (iii) more than
50% of affected people having vomiting; and (iv) no
bacterial agent found.

An outbreak investigation team was formed by the HCDCP.
It was decided to conduct an analytical epidemiological
study in the island, given that the population of the island is
small and, therefore, it could be used as a cohort; in this way,
the biases pertaining to the selection of a control group of a
possible case-control study would be avoided, and it would
be possible to calculate an attack rate. Due to the lack of
laboratory infrastructure on the island, environmental
samples had not been collected until the point at which the
HCDCP was informed about the outbreak. The investigation
team arrived on the island of Agios Efstratios on
2 March 2010 in order to document the outbreak and
discover epidemiological evidence of its source.

According to the authority of Greek Public Health Law
3370/2005, in public health emergencies or outbreaks the
HCDCP intervenes immediately, using any means
available to protect public health. With the delegation of this
legal framework, such urgent outbreak investigations are in
full compliance with the Elsinki Declaration, and can be
carried out without further ethical approval.

Methods

A retrospective cohort study was carried out in order to
identify individual risk factors for the acquisition of
gastroenteritis symptoms. The cohort was defined as the
island’s entire population. People who were absent on
Monday 15 February 2010 were excluded from the study.
Demographic information was collected and participants
asked about their food and water consumption, animal
ownership, contact with animals and the presence of any
symptoms of gastroenteritis since 15 February 2010. A
primary case of gastroenteritis was defined as any person
who had at least two episodes of diarrhoea or vomiting in the
72 hours following 12.00 h Monday 15 February. Any
person who had these symptoms later than that time met the
case definition of a secondary case of gastroenteritis.
Participants were interviewed face-to-face using a structured
questionnaire. Rate ratios and corresponding 95%
confidence intervals (CI) were calculated. Data were entered
into EpiData Entry v3.1 (http://www.epidata.dk/) and

Page 3

© I Karagiannis, M Detsis, K Gkolfinopoulou, D Pervanidou, T Panagiotopoulos, S Bonovas, 2010. A licence to publish this material has been
given to James Cook University, http://www.rrh.org.au

3
statistical analysis was performed with Stata v11 software
(http://www.stata.com/stata11/).

Laboratory tests

Island physicians were advised that should more cases occur,
clinical samples were to be tested for Norovirus and bacteria.

Environmental tests

Testing of left over food for viruses and bacteria was
recommended.

Results

Descriptive epidemiology

According to the 2001 census, the island had
314 inhabitants; however, the president of the Municipal
District advised that the current population was
approximately 210. During the visit of the the HCDCP,
officials found there were 197 current residents, with an
additional 8 people who were temporarily absent. Of the
available population, 181 (91.9%) advised they were on the
island on 15 February 2010. The remaining 16 were
excluded from the study.

Eighty-seven of the participants were female (48.1%).
Participants’ median age was 42 years (range: 3 months–
97 years).

Sixty-four participants (35.4%) reported symptoms of
gastroenteritis occuring since 15 February 2010. Thirty-four
(53.1%) of these were primary cases, according to the case
definition used, and the remaining 30 were secondary cases.
The distribution of cases by symptoms of onset date is
shown (Fig1).

The most common symptoms reported were diarrhoea
(81.2%), abdominal pain (79.7%), vomiting (68.8%), fatigue
(52.4%), myalgia (44.4%), nausea (35.9%), fever (29.7%),
shivering (26.6%) and headache (6.7%). Only one patient
(1.6%) reported blood in the stool.

Analytical epidemiology

The comsumption of raw seafood imported to the island was
found to be statistically significantly associated with the
development of symptoms among primary cases. People who
consumed raw smooth Venus clams, oysters, scallops, clams
and sea squirts were more likely to develop symptoms of
gastroenteritis in the following 3 days than those who did not
consume these raw products. Far fewer people reported
having eaten these products cooked. The consumption of
local seafood items, such as sea urchins, limpets, squid,
octopus, crab and sea anemone was not found to be a risk
factor for acquiring symptoms of gastroenteritis (Table 1).

All imported seafood consumed raw were grouped together;
people who consumed any of these food items were
21.5 times (95% CI: 8.95-51.8) more likely to develop
symptoms of gastroenteritis within 3 days, compared with
those who did not consume raw imported seafood items
(Table 1). Twenty-nine of 34 primary cases (85.3%) and 2 of
26 secondary cases (7.7%) had consumed raw seafood on
15 February 2010.

People who consumed locally produced pasta were also
found to be more likely to become ill than those who did not
(Table 1). However, this relationship was no longer present
after stratifying by consumption of raw imported seafood
items (data not shown). Possession of, or contact with,
animals was not found to be associated with disease
acquisition.

Laboratory tests

No samples were obtained from patients visiting the
peripheral medical practice of Agis Efstratios.

Environmental tests

No seafood remained in Agios Efstratios for for testing.

Page 4

© I Karagiannis, M Detsis, K Gkolfinopoulou, D Pervanidou, T Panagiotopoulos, S Bonovas, 2010. A licence to publish this material has been
given to James Cook University, http://www.rrh.org.au

4


Figure 1: Distribution of primary (n=33) and secondary (n=29) cases of gastroenteritis by symptoms onset in Agios
Efstratios, Greece, February–March 2010 (n=62).


Discussion

An outbreak of gastroenteritis occurred in a very remote
island in the Northern Aegean Sea in February 2010. In most
cases the onset of symptoms was between 15 and
18 February 2010, although cases continued to appear within
the island population until well into March 2010. Even
though Norovirus is transmitted to humans through
consumption of infected food items, it can also spread from
person to person, usually via the fecal–oral route6; however,
airborne transmission has also been suspected7.

The seasonality of Norovirus infections among humans has
been described and further analysed elsewhere8. In the 2010
northern hemisphere winter, a number of clusters of
gastroenteritis cases due to Norovirus that could be linked to
oyster consumption were reported throughout Europe3. There
is no evidence that the western European outbreaks were
linked to the present Greek report but information on
backtracing of seafood was limited.

In Greece, during Lent or at least on its first day, Ash
Monday, it is customary to abstain from eating meat and
instead to consume seafood. In the year of this report, Ash
Monday was on 15 February 2010, and the consumption of
raw seafood items was found to be associated with the
acquisition of an infection.

The fact that all Kaplan’s criteria were met in the outbreak
described, coupled with the descriptive epidemiology
evidence suggesting person-to-person transmission of the
disease (Fig1), makes it likely that Norovirus was the
causative agent of the outbreak. Indeed, 29 of the island’s
34 primary cases (85.3%) reported having consumed raw,
imported seafood on Ash Monday. This was the case for
only 2 of the 26 secondary cases (7.7%).

Page 5

© I Karagiannis, M Detsis, K Gkolfinopoulou, D Pervanidou, T Panagiotopoulos, S Bonovas, 2010. A licence to publish this material has been
given to James Cook University, http://www.rrh.org.au

5

Table 1: Risk factors for the acqusition of gastroenteritis symptoms following Ash Monday 15 February 2010 in Agios
Efstratios, Greece

Exposed
Total
86
Unexposed
Total
Risk factors
Cases
17
AR%
19.8
Cases
17
AR%
18.1
RR 95% CI
Sex (ref: male)
Age (ref: ≥65 years)
0-4 years
5-14 years
15-64 years
Water consumption at home
Tap
Bottled
Food consumption
Locally produced pasta
Imported pasta†
Cured meat products (eg ham)†
Bottled pasteurized milk†
Pork†
Locally produced cheese
Imported cheese†
Veal†
Egg
Chicken
Rice†
Lamb
Lettuce†
Cucumber†
Tomato†
Cabbage†
Fruits†
Sweets†
Fish
Cornstarch halva
Tahini halva†
Fish roe†
Filled vineyard leaves
Legumes† (lentils, chickpeas,
dry beans)
Seafood
Raw smooth Venus clams†
Cooked smooth Venus clams
Raw oysters†
Cooked oysters†
Raw scallops†
Cooked scallops†
Raw clams†
Cooked clams†
Cooked mussels†
Sea squirts†
Sea urchins
Limpets
Squid
Octopus
Cuttlefish†
94 1.09 0.6-2.0

2
2
2
45
45
45
4.44
4.44
4.44
0 n/a
3
29
18
108
16.7
26.9
3.75
6.04
0.68-20.6
1.50-24.5

27
10
145
51
18.6
19.6
7
24
36
130
19.4
18.5
0.96
1.06
0.45-2.02
0.55-2.06

17
5
2
10
12
11
16
5
14
10
1
8
0
1
6
12
14
2
5
0
18
24
1
2
52
59
28
72
48
50
93
27
73
53
8
43
31
10
29
66
66
30
46
11
74
106
6
27
32.7
8.47
7.14
13.9
25.0
22.0
17.2
18.5
19.2
18.9
12.5
18.6
0
10.0
20.7
18.2
21.2
6.67
10.9
0
24.3
22.6
16.7
7.41
16
28
31
22
21
22
17
28
19
23
32
25
33
32
27
21
19
31
28
33
15
9
32
31
124
117
148
103
124
128
85
145
103
120
168
132
145
166
147
110
110
146
130
167
104
72
172
151
12.9
23.9
21.0
21.4
16.9
17.2
20.0
19.3
18.5
19.2
19.1
18.9
22.8
19.3
18.4
19.1
17.3
21.2
21.5
19.8
14.4
12.5
18.6
20.5
2.53
0.35
0.34
0.65
1.48
1.28
0.86
0.96
1.04
0.98
0.66
0.98
0
0.52
1.13
0.95
1.23
0.31
0.50
0
1.69
1.81
0.90
0.36
1.39-4.62
0.14-0.87
0.09-1.34
0.33-1.29
0.79-2.76
0.67-2.44
0.46-1.59
0.41-2.26
0.56-1.94
0.50-1.92
0.10-4.21
0.48-2.01
n/a
0.08-3.42
0.51-2.48
0.50-1.81
0.66-2.28
0.08-1.24
0.21-1.23
n/a
0.91-3.13
0.89-3.67
0.15-5.51
0.09-1.42

23
0
17
0
17
0
14
0
6
1
24
9
4
3
0
31
0
23
1
21
0
19
1
7
1
108
34
58
36
8
74.2
n/a
73.9
0
81.0
n/a
73.7
0
85.7
100
22.2
26.5
6.90
8.33
0
11
33
15
32
16
33
19
32
27
32
9
24
29
30
33
148
178
154
176
157
178
159
176
171
177
70
144
120
142
170
7.43
18.5
9.74
18.2
10.2
18.5
12.0
18.2
15.8
18.1
12.9
16.7
24.2
21.1
19.4
9.98
n/a
7.59
0
7.94
n/a
6.17
0
5.43
5.53
1.73
1.59
0.29
0.39
0
5.45-18.3
n/a
4.43-13.0
n/a
4.48-13.2
n/a
3.74-10.2
n/a
3.43-8.60
4.04-7.57
0.85-3.50
0.81-3.10
0.11-0.77
0.13-1.22
n/a

Page 6

Table 1: cont’d

Risk factors
© I Karagiannis, M Detsis, K Gkolfinopoulou, D Pervanidou, T Panagiotopoulos, S Bonovas, 2010. A licence to publish this material has been
given to James Cook University, http://www.rrh.org.au

6
Exposed
Total
Unexposed
Total Cases
0
2
0
0
1
29
AR%
0
15.4
n/a
n/a
8.3
76.3
Cases
33
31
33
33
32
5
AR%
18.6
18.8
18.5
18.5
19.3
3.55
RR 95% CI
Lobster†
Shrimps†
Crawfish†
Crab
Sea anemone (fried)
Any raw imported seafood†
AR%, Attack rate ; n/a, not applicable; RR, relative risk.
†Food items not produced on the island.
1
13
0
0
12
38
177
156
178
178
166
141
0 n/a
0.82
n/a
n/a
0.43
21.5
0.22-3.05
n/a
n/a
0.06-2.90
8.94-51.8

A number of seafood items were consumed in Agios
Efstratios for the island has a very long fishing tradition.
Locally fished items include a number of fish species, sea
anemones, sea urchins, limpets, squids, octopuses, crabs and
lobsters. All but the latter were reported to have been fished
locally in February 2010. Mussels, oysters, clams, smooth
Venus clams and scallops are foreign to the island and
always imported. All but mussels were eaten either raw or
cooked; only cooked mussels were eaten. Although thermal
processing can kill pathogens, those who consumed cooked
mussels were 5.4 times more likely to develop symptoms
than those who did not. Although the thermal processing
could not be observed at the time of the outbreak
investigation, this finding may partly reflect the fact that
people who ate cooked mussels on Ash Monday were more
likely to also consume raw seafood at the same time.

Interviewees reported that the shellfish were alive at the time
of consumption, an observation often confirmed by pouring
fresh lemon juice on the shellfish. If the shellfish is still
alive, it reacts to the acid liquid and many consider this to be
a sign that the shellfish is safe to eat. However, the presence
of transmittable pathogens in seafood does not necessarily
kill the shellfish, and the public should be made aware of
this.

Among the limitations of the study, is the fact that laboratory
and environmental sampling and backtracing were not
optimal. The island of Agios Efstratios has only one
peripheral medical office with two physicians.


Conclusion

Despite the limitation, there was strong epidemiological
evidence that seafood items were the vehicle of the outbreak.
In this cohort study set in a very closed population, people
who consumed imported seafood items were 21.5 times
more likely to develop symptoms of gastroenteritis within
3 days. This report emphasises the role and importance of
the use of epidemiological methods in outbreak investigation
in any such remote settings, where laboratories support may
be absent or limited.

Acknowledgements

The authors acknowledge the contribution of colleagues
Costas Danis, Theologia Sideroglou and Theodoros Lytras
for their fruitful comments before the study onset. Also
thanked are Nikolaos Manoloudakis and Paraskevas
Vrochides, physicians at the Agios Efstratios local medical
practice, for providing the first information on the outbreak,
as well as the president of the Municipal District of Agios
Efstratios, Charalampos Makris, for his assistance during the
authors’ time on the island.

References

1. Potasman I, Paz A, Odeh M. Infectious outbreaks associated with
bivalve shellfish consumption: a worldwide perspective. Clinical
Infectious Diseases 2002; 35: 921-928.

Page 7

© I Karagiannis, M Detsis, K Gkolfinopoulou, D Pervanidou, T Panagiotopoulos, S Bonovas, 2010. A licence to publish this material has been
given to James Cook University, http://www.rrh.org.au

7
2. Butt AAA, Aldridge KE, Sanders CV. Infections related to the
ingestion of seafood Part I: viral and bacterial infections. Lancet
Infectious Diseases 2004; 4(4): 201-212.

3. Westrell T, Dusch V, Ethelberg S, Harris J, Hjertqvist M,
Jourdan-da Silva N et al. Norovirus outbreaks linked to oyster
consumption in the United Kingdom, Norway, France, Sweden and
Denmark, 2010. Eurosurveillance 2010; 15(12): pii=19524.
Available: http://www.eurosurveillance.org/ViewArticle.aspx?
ArticleId=19524

4. Kaplan JE, Feldman R, Campbell DS, Lookabaugh C, Gary GW.
The frequency of a Norwalk-like pattern of illness in outbreaks of
acute gastroenteritis. American Journal of Public Health 1982;
72(12): 1329-1332.
5. Turcios RM, Widdowson MA, Sulka AC, Mead PS, Glass RI.
Reevaluation of epidemiological criteria for identifying outbreaks
of acute gastroenteritis due to norovirus: United States, 1998–2000.
Clinical Infectious Diseases 2006; 42(7): 964-969.

6. Patel MM, Hall AJ, Vincé J, Parashar UD. Noroviruses: a
comprehensive review. Journal of Clinical Virology 2009; 44: 1-8.

7. Marks PJ, Vipond IB, Carlisle D, Deakin D, Fey RE, Caul EO.
Evidence of airborne transmission of Norwalk-like virus (NLV) in a
hotel restaurant. Epidemiology and Infection 2000; 124: 481-487.

8. Lopman B, Armstrong B, Atchison C, Gray J. Host, weather and
virological factors drive norovirus epidemiology: time-series
analysis of laboratory surveillance data in England and Wales.
PLoS ONE 2009; 4(8): e6671