See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/7843709

Botulism and Preserved Green Olives

Article  in  Emerging infectious diseases · June 2005

DOI: 10.3201/eid1105.041088 · Source: PubMed

CITATIONS READS
32 330

9 authors, including:

Lucia Pastore Celentano Fortunato D'Ancona

European Centre for Disease Prevention and Control Istituto Superiore di Sanità
40 PUBLICATIONS   903 CITATIONS    120 PUBLICATIONS   2,042 CITATIONS   

SEE PROFILE SEE PROFILE

Marco Massari Fabrizio Anniballi

Istituto Superiore di Sanità Istituto Superiore di Sanità
71 PUBLICATIONS   2,511 CITATIONS    68 PUBLICATIONS   913 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

SpIDnet project View project

Broad Spectrum Project View project

All content following this page was uploaded by Fabrizio Anniballi on 01 June 2014.

The user has requested enhancement of the downloaded file.

 LETTERS

and 120 participants (89%) reported a about H. pylori infection and gastroin- 5. United Nations Development Programme.
variety of gastrointestinal symptoms, testinal illness and to introduce diag- Human Development Report Georgia,
2000. [cited 2004 Feb 22]. Available from
including epigastric pain. Persons nosis of this infection and appropriate http://www.undp.org.ge/nhdr2000/NHDR-
with active infection were no more treatment for it into standard medical GEO2000.pdf
likely to report epigastric pain (75 practice. In addition, rigorous popula- 6. Gatta L, Ricci C, Tampieri A, Vaira D. Non-
[78%] of 97) than persons without tion surveys that include children are invasive techniques for the diagnosis of
Helicobacter pylori infection. Clin
infections (30 [79%] of 38). One par- needed to identify high-risk groups of Microbiol Infect. 2003;9:489–96.
ticipant reported gastric cancer and persons for targeted public health 7. Gold BD. Helicobacter pylori infection in
was H. pylori–positive. Seven partici- interventions (7). children. Curr Probl Pediatr Adolesc Health
pants reported a family history of gas- Care. 2001;31:247–66.
tric cancer; all were H. pylori–posi- Acknowledgments
Address for correspondence: Jeremy Sobel,
tive. Five participants, of whom 3 We thank John Heinrich for facilitat-
Centers for Disease Control and Prevention,
were H. pylori–positive, reported a ing the use of an accurate point-of-care test
1600 Clifton Rd, Mailstop A38, Atlanta, GA
history of gastric surgery for peptic of active H. pylori infection in the field. We
30333, USA; fax: 404-639-2205; email: jso-
ulcer disease. More rural participants also thank Maiko Chokheli, Ekaterina
bel@ cdc.gov
(23 of 27) reported a history of rup- Jhorjholiani, and Tamuna Zardiashvili for
tured ulcers or a family member with assistance with data collecting.
gastric cancer than did urban partici-
This study was supported by a grant
pants (74 of 108). The frequency of
from the Biotechnology Engagement
reported gastrointestinal symptoms
was similar between urban and rural
Program (BTEP) #13, Department of
Health and Human Services. Dr. Gold is
Botulism and
participants.
supported in part by a grant from the Preserved Green
Olives
To our knowledge, this is the first
National Institutes of Health, NIDDK
survey of H. pylori infection in
(DK-53708).
Georgia. We found a high rate of To the Editor: In March 2004, a
infection with H. pylori. Participants total of 16 suspected cases of botulism
also reported a very high rate of dys- Katrina Kretsinger,* Jeremy Sobel,*
were reported to the Italian National
peptic symptoms, although these were Nato Tarkhashvili,†
Institute of Health by hospitals in 3
not correlated with infection. This Neli Chakvetadze,†
adjoining regions in central and
small convenience sample survey has Marina Moistrafishvili,†
southern Italy (Molise, Campania,
several limitations, however. First, Merab Sikharulidze,†
and Puglia). Initial investigation
our participants did not constitute a Ben D. Gold,*‡
showed that all patients had eaten at
systematically selected population Marina Chubinidze,†
the same restaurant in Molise on
sample. Second, rural populations and Paata Imnadze†
February 22 or 24, 2004. The restau-
were underrepresented. Finally, we *Centers for Disease Control and
Prevention, Atlanta, Georgia, USA;
rant provided reservation lists for
used neighborhood or village of resi- those dates (the restaurant was closed
†National Center for Disease Control,
dence as a marker for socioeconomic Tbilisi, Republic of Georgia; and ‡Emory on February 23). It also provided a list
status without specific income infor- University School of Medicine, Atlanta, of foods that had been served each
mation from the participants. Georgia, USA evening. Persons on the reservation
Therefore, socioeconomic status mis- lists were contacted and asked to pro-
classification possibly occurred and References vide the names of others who had
the association between infection and been at their tables to ensure that as
1. Suerbaum S, Michetti P. Helicobacter
socioeconomic status may not be pylori infections. N Engl J Med. 2002; many diners as possible were traced.
accurate. 347:1175–86. Of 73 persons who had been identi-
Nevertheless, it is unlikely that we 2. Frenck R, Clemens J. Helicobacter in the
developing world. Microbes Infect. fied as having eaten at the restaurant
substantially over- or underrepresent- on either evening, 66 were successful-
2003;5:705–13.
ed infection prevalence in the general 3. Brown LM. Helicobacter pylori: epidemi- ly contacted and interviewed in per-
population. Despite the limitations of ology and routes of transmission. son or by telephone about symptoms
this study, our results clearly indicate Epidemiol Rev. 2000;22:283–97.
4. Skarbinski J, Walker HK, Baker LC, and food consumed at the restaurant.
that H. pylori is a serious public For purposes of the investigation, a
Kobaladze A, Kirtava Z, Raffin TA. The
health problem in Georgia. There is a burden of out-of-pocket payments for probable case-patient was defined as a
pressing need to educate medical pro- health care in Tbilisi, Republic of Georgia. person who had dined at the restaurant
fessionals and the general public JAMA. 2002;287:1043–9.
on February 22 or 24 and had experi-

Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 11, No. 5, May 2005 781
LETTERS

enced diplopia or blurred vision and at and effect modification (2). Foods providing training and periodic moni-
least 1 of the following symptoms: associated with illness with p values toring of those involved in small-scale
dysphagia, dry mouth, dysarthria, <0.20 were considered in the model. preparation to ensure that disease risks
upper/lower extremity weakness, dys- In a univariate analysis in which from improperly prepared or stored
pnea, and severe constipation. Those all 28 patients were considered, the foods are minimized.
who met the probable case definition RR of illness was higher among din-
and had laboratory-confirmed botu- ers who ate home-preserved green Acknowledgments
lism were considered definite case- olives in salt water (RR 5.2, 95% CI We thank the staff of the local health
patients. 1.4–19.8), ate cream pastries (RR 2.5, unit and Carabinieri force for their logistic
We tested for botulinum neurotox- 95% CI 1.8–3.4), and drank home- support of the field investigation, Primula
in in serum and spores in stool sam- made lemon liqueur (RR 2.1, 95% CI Semprini for providing laboratory results
ples as described (1). Serum or stool 1.3–3.4). After multivariate analysis, for food samples, Alain Moren for scien-
specimens from 24 patients with ≥2 only the risk associated with eating tific support and advice, and Nancy
symptoms were sent to the Italian green olives remained significant (RR Binkin for her assistance in reviewing the
National Institute of Health for testing. 5.2, 95% CI 1.4–19.8). manuscript.
Twenty-eight persons reported ≥2 None of the food items served on
Amy Cawthorne,*
symptoms (42% attack rate); 25 February 22 or 24 was available for
Lucia Pastore Celentano,*†
(89%) were considered probable sampling, and none of the other 13
Fortunato D'Ancona,†
cases and 3 (11%) were considered food samples obtained from the
Antonino Bella,† Marco Massari,†
confirmed cases. Two members of the restaurant was positive for C. botu-
Fabrizio Anniballi,† Lucia Fenicia,†
restaurant owner’s family and 1 linum. However, the pH of a jar of
Paolo Aureli,†
employee were among the probable olives that had been prepared at the
and Stefania Salmaso†
case-patients. Onset of symptoms same time as those eaten on February
*European Programme for Intervention
occurred 4 hours to 6.5 days after eat- 22 and 24 was 6.2, far above the level Epidemiology Training, Rome, Italy; and
ing at the restaurant (median 36 of 4.6 required to prevent growth of †Istituto Superiore di Sanità, Rome, Italy
hours). Twenty persons (71%) had C. botulinum. No salinity testing was
been seen in emergency rooms, 15 performed by the local laboratory, and References
(53%) were admitted to a hospital, inadequate storage during transit 1. Food and Drug Administration. Bacterio-
and 18% were admitted to intensive made it impossible to conduct salinity logical analytical manual online. January
care. None required ventilatory sup- and water activity tests at the national 2001 [cited 2005 Feb 10]. Available from
http://www.cfsan.fda.gov/~ebam/bam-
port, and no deaths occurred. reference laboratory.
mm.html
The main symptoms reported by Interviews with the restaurant pro- 2. Zou G. A modified poisson regression
28 probable and confirmed patients prietors indicated that the olives were approach to prospective studies with binary
included dry mouth in 25 (89%), dys- prepared on site during the fall of 2003 data. Am J Epidemiol. 2004;159:702–6.
3. Fenicia L, Ferrini AM, Aureli P. Epidemia
phagia in 25 (89%), severe constipa- from local olives. After soaking in salt
di botulismo da olive nere. Industrie
tion in 22 (79%), and blurred vision water for 35 days, the olives had been Alimentari. 1992;31:307–8.
in 27 (96%). Three weeks after onset decanted into jars, and salt water had 4. Padua L, Aprile I, Lo Monaco M, Fenicia L,
of symptoms, 15 (68%) reporting been replaced with fresh water. Neither Anniballi F, Pauri F et al. Neuro-
physiological assessment in the diagnosis
severe constipation, 11 (41%) report- the amount of salt used in the salt
of botulism: usefulness of the single-fiber
ing blurred vision, 10 (40%) report- water mixture nor the pH at any stage EMG. Muscle Nerve. 1999;22:1388–92.
ing dry mouth, and 11 (44%) report- was standardized during preparation. 5. Endoh M, Okuno R, Shimojima Y, Murata
ing dysphagia still had these symp- Both epidemiologic evidence and I, Sekine H, Kokubo Y. Botulism, Japan.
Infectious agents surveillance report. [cited
toms. Of the 24 patients for whom information obtained regarding prepa-
2005 Feb 10]. 2000; 21:54.Available from
rectal swabs were available, 3 were ration of the olives strongly suggest http://idsc.nih.go.jp/iasr/21/241/tpc241.
culture-positive for Clostridium botu- that they were the likely source of the html
linum type B. None of 5 serum sam- outbreak. This outbreak highlights the
ples tested positive. previously documented risk associated Address for correspondence: Amy Cawthorne,
Food-specific attack rates, relative with improperly prepared olives (3–5). European Programme for Intervention
Epidemiology Training, Centro Nazionale di
risks (RRs), and 95% confidence In Italy and elsewhere in Europe, an
Epidemiologia Sorveglianza e Promozione
intervals (CIs) were calculated. A increasing trend favors traditional
della Salute, Reparto di Epidemiologia delle
Poisson model with robust error vari- foods and preparation methods over Malattie Infettive, Istituto Superiore di Sanità,
ance was used to estimate RR with large-scale industrial products. This Viale Regina Elena 299, 00161 Rome, Italy;
adjustments for possible confounding outbreak underlines the importance of fax: 39-06-4423-2444; email: cawthorn@iss.it

782 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 11, No. 5, May 2005

View publication stats