Outbreak of listeria monocytogenes in an oncology unit associated with sandwiches consumed in hospital

Journal of Hospital Infection (2009) -, 1e5 Available online at www.sciencedirect.com Outbreak of Listeria monocytogenes in anoncology unit associated with sandwichesconsumed in hospital A. Shetty , J. McLauchlin K. Grant D. O'Brien T. Howard ,E.M. Davies a Cym Taf NHS Trust, Merthyr Tydfil, UKb Health Protection Agency Food Safety Microbiology Laboratory, Centre for Infections, London, UKc Velindre NHS Trust, Cardiff, UKd National Public Health Service for Wales, Cardiff, UK Received 11 October 2008; accepted 16 January 2009 In May 2003, two adult patients in an oncology unit were diag- nosed with listeriosis, and sandwiches consumed in the hospital were iden- tified as a common risk factor. Both patients were infected by the same strain of Listeria monocytogenes. Sandwiches collected from the hospital and external sandwich producer, as well as sites within the manufacturing Listeriosis; Sandwichesserved in hospitals environment, were contaminated by the same strain of L. monocytogenes.
Sandwiches consumed in other hospitals have been associated with smallclusters of listeriosis patients in the UK. This report describes the investi-gations following diagnosis of the two infections, and highlights a moregeneral problem with sandwiches sold in hospitals.
ª 2009 The Hospital Infection Society. Published by Elsevier Ltd. All rightsreserved.
laboratory animals in Cambridge in 1924.Laterit became apparent that the disease affects hu- The bacterium Listeria monocytogenes and the mans. The increased number of reported cases during the 1980s in several countries togetherwith evidence for foodborne transmission has re-newed interest in this disease. It most often af- * Corresponding author. Address: Cym Taf NHS Trust, Prince fects the unborn and newly delivered as well as Charles Hospital, Merthyr Tydfil, CF47 9DT, UK. Tel.: þ44 1685 the immunocompromised and elderly, but can oc- cur in healthy individuals. Listeriosis primarily 0195-6701/$ - see front matter ª 2009 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.jhin.2009.01.012 Please cite this article in press as: Shetty A et al., Outbreak of Listeria monocytogenes in an oncology unit associated with sand-wiches consumed in hospital, J Hosp Infect (2009), doi:10.1016/j.jhin.2009.01.012 presents as abortion, septicaemia or infections of amoxicillin and gentamicin for two weeks. She the central nervous system, with a high case fatal- made a good recovery and was sent home. Following ity rate.Although the numbers of reported cases a further course of chemotherapy, she was readmit- declined during the 1990s, there has been a recent ted a week later with hypothermia and hypotension increase in England and Wales and in other Euro- and died within 24 h. The second patient was a fe- pean countriThis increase has predominantly male aged 63 years with recurrent, progressive, met- occurred in patients aged >60 years with severe astatic breast carcinoma, admitted with pyrexia, lethargy, weight loss and loss of appetite. Treatment Although listeriosis is usually acquired from con- with ampicillin was commenced but she died within sumption of contaminated food, the epidemiology is a week. She was not treated with gentamicin as she complex. The ubiquitous nature of the bacterium had severe renal impairment. Since listeriosis is un- and the wide range in incubation periods (from 1 to usual (the total number of reported human listeriosis >90 days) between consumption of contaminated cases in Cardiff between 1995 and 2002 was three) food and onset of disease mean that investigations the cases were reported to the Consultant for Com- L. monocytogenes presents a potential hospital in-fection problem because of the possibility of cross- infection between neonates in delivery suites aswell as via the consumption of contaminated food An outbreak control meeting was called by the in hospitalSandwiches consumed in hospitals CCDC on 13 June 2003. Subtyping and national have been associated with three small clusters of lis- surveillance data for human listeriosis in England and Wales as well as subtyping results on L. mono- In May 2003, two patients in an oncology unit cytogenes was requested. Complete typing results were diagnosed with listeriosis, and sandwiches were available on 9 July 2003 and showed that the were identified as a common risk factor. This two blood culture isolates were indistinguishable report describes the investigations following the and were designated as serovar 1/2a, phage type diagnosis of these infections, and highlights a more general problem with sandwiches sold in hospitals.
The first possibility considered by the outbreak control team was of a coincidental finding since both patients were severely immunosuppressed asa result of their advanced malignancies and so atincreased risk of listeriosis. However, since listeri- Surveillance of listeriosis was derived from routine osis is rare, even in these types of patients, the data generated by national surveillance activities occurrence of two such cases linked closely in (National Public Health Service for Wales, and place and time would occur very rarely by chance.
HPA, unpublished data). Isolates of L. monocyto- The second possibility was of cross-infection be- genes were identified in the national reference tween patients. This was considered unlikely since laboratory (HPA Centre for Infections, London) by person-to-person spread has not been described phenotypic tests including the use of the API List- except during the neonatal period.Cross-infec- eria identification Strains were characterised tion was also unlikely since the patients were by serotyping, phage typing, DNA amplified frag- nursed on separate wards and had no direct con- ment length polymorphism, and DNA pulsed-field tact with each other. The final possibility was ex- gel electrophoresis.Food analysis was per- posure to a common source, and since listeriosis is predominantly a foodborne disease, the mostlikely common risk factor was contaminated food.
Environmental health officers (EHOs) investi- gated food consumption histories for the patients.
In the last week of May 2003, L. monocytogenes was Unfortunately one of the patients died before she isolated from blood cultures taken from two patients could be interviewed, but the next of kin was in an oncology unit, but nursed on different wards, eager to help, so the same EHOs interviewed the within two days of each other. Both patients had ad- second patient and her partner and the relative of vanced malignancy and were receiving palliative chemotherapy. One was a female aged 46 years One week prior to admission, both patients with disseminated adenocarcinoma of the right attended the same outpatients department and ovary, admitted with a high temperature, confusion had eaten sandwiches provided by the hospital (one and cerebellar symptoms. She was treated with was a ham salad and the other a tuna salad). This Please cite this article in press as: Shetty A et al., Outbreak of Listeria monocytogenes in an oncology unit associated with sand-wiches consumed in hospital, J Hosp Infect (2009), doi:10.1016/j.jhin.2009.01.012 Listeria outbreak from hospital sandwiches was the only risk factor identified. An investigation of sandwiches sold in the hospital was therefore manufacturing environment. Two other L. monocy- togenes strains were isolated from an additionalnine sandwiches and three environmental siteswithin the factory.
Supply of sandwiches from the company to the An outside supplier manufactured the sandwiches hospital was stopped on 25 June 2003. Following supplied to the hospital. EHOs investigated the this incident, an interim protocol for the purchasing supplier and the infection control nurse assessed of processed or prepared food for patients in the the reception, storage and delivery of sandwiches hospital was established. Each company was re- within the hospital. Sandwiches were also sent quired to provide a record of the following docu- from the hospital to the National Public Health ments: registration with the local authority; current Service Food and Water Laboratory in Cardiff for membership of a trade association such as the testing. The medical staff were informed of the British Sandwich Association; copies of their Hazard Analysis Critical Control Point (HACCP) documenta- Investigation of the factory revealed that rou- tion; evidence of appropriate training of staff; tine microbiological testing of their sandwiches six copies of quantitative microbiological testing; rec- months prior to this investigation had identified an ords on maintenance of temperature during trans- L. monocytogenes contamination 'problem', which port. Companies are also required to inform the coincided with the re-laying of the factory floor in hospital should the microbiological testing deviate December 2002. The hospital had not been made from the accepted levels and also if they had ever been served with an enforcement notice or been Investigations carried out in the hospital re- prosecuted. The Trust also demanded access for vealed that the sandwiches were stored in a cool inspection of the manufacturing premises.
display cabinet. Sandwiches for sale were placedat the top. Sandwiches for distribution to the wards were placed in a box on the base. The temperatureof the sandwiches varied during the day and at The company reacted as follows: supply of sand- times rose above the recommended 8 C (Food hy- wiches to vulnerable groups including hospitals, giene regulations England, Wales and Northern Ire- nurseries, schools was stopped; the suppliers of land 2006) to 11 C. The sandwiches were delivered raw materials were changed; the laboratory used to staff of a voluntary association who put these ei- for microbiological analysis was changed as it did not ther directly into a refrigerator or into a cooled dis- quantify any listeria detected; the main production play cabinet at 10 C. Sandwiches, sent to the line was dismantled, new rollers and belts were wards for patient consumption, were stored in installed; a deep clean of the factory was completed a fridge but removed and placed on hot food trol- and dry cleaning methods by dedicated cleaners leys for distribution to the patients. The tempera- instituted; the evaporator units were dismantled ture of one of these sandwiches was found to be and disinfected; the floor was repaired; and staff 19 C. This practice was stopped immediately.
retrained. These steps resulted in a reduction of the Staff were asked not to remove the sandwiches recovery of L. monocytogenes from within in the fac- from the refrigerator until they were required.
tory both in the environment and the finished prod- The voluntary group was advised that the temper- uct; however, the company could not sustain its ature in the cool cabinet should be lowered to al- activities and went out of business within six months.
low for opening and closing of the door.
No further cases of listeriosis due to the type Eight sandwiches (with a variety of fillings) were associated with these two cases were identified in randomly collected from the hospital and micro- England and Wales for the remainder of 2003 and biologically tested for the presence of listeria; L. monocytogenes was isolated from all sand-wiches tested at levels of <100 cfu/g.
The isolates from the two cases were indistin- guishable from those recovered from 49 sand-wiches and sandwich fillings (collected both from Identification of a common risk factor (consump- the hospital and at the factory) as well as 12 tion of the same brand of sandwiches in hospital) Please cite this article in press as: Shetty A et al., Outbreak of Listeria monocytogenes in an oncology unit associated with sand-wiches consumed in hospital, J Hosp Infect (2009), doi:10.1016/j.jhin.2009.01.012 and the recovery of infecting organism from spp. (5.6%) and L. monocytogenes (3.1%) were sandwiches in both the hospital and the food found in 950 hospital sandwiches sampled in factory together with sites within the factory Wales.In both studies, the presence of Listeria provide good evidence for this food as the vehicle spp. and L. monocytogenes was more often associ- ated with sandwiches that were supplied to hospi- The sandwich-manufacturing factory was inves- tals rather than those made on-site. The former tigated by sampling the various steps of sandwich study also highlighted the absence of a hazard production and the environment. The floor of the analysis system, the collection of sandwiches factory was thought to be the source of the from sites other than the hospital kitchen (e.g.
L. monocytogenes. The problem with contamina- wards, cafeterias or shops) and the storage/dis- tion had been identified in December 2002 when play of sandwiches above 8 C as factors associ- the floor was re-laid. It was incompletely sealed ated with the presence of Listeria spp. and and heavy machinery was dragged across, so wid- ening the gaps. At the time of the investigation it The Public Health Laboratory Service guidelines was found that the factory was cleaned in a non- for the microbiological quality of ready-to-eat systematic manner. The floors were first cleaned, foods at the point of sale state that <100 cfu and then the food contact surfaces were cleaned L. monocytogenes/g is acceptable, and >100 cfu with the same cloth. Contamination of food and is unacceptable/potentially hazardous.This out- break raised a number of questions regarding pro- manufacturing environments have been recognised vision of food for immunosuppressed patients, the as important in other outbreaks of listeri duty of care for foods provided in hospitals, and There have been four other small clusters of the provision of advice to vulnerable groups with patients associated with consumption of sand- wiches in different hospitals in the UK.
The current (contracted) annual value of sand- All five incidents were associated with sandwiches wich sales into the National Health Service (NHS) is with mixed fillings produced outside the hospital and £13 m, which equates to w16 million sandwiches contaminated at the food production sites. The at an average price of £0.80 (source: NHS Supply same strains of L. monocytogenes were isolated Chain). This does not account for the sandwiches from the food and food production areas, although that are bought from non-contracted suppliers or these differed in each of the five instances. In all made in-house. Clearly, sandwiches are commonly five, large numbers of people (including vulnerable consumed by patients in UK hospitals.
populations in hospitals) were exposed but only The vast majority of sandwiches are safe and a few serious systemic illnesses resulted. Low attack the small number of individuals affected by the rates in outbreaks of listeriosis have been previously outbreaks/incidents described above highlights recognised.Foods associated with outbreaks usu- the low risk associated with them. Nevertheless, ally have the common features of being highly pro- sandwiches have the potential for contamination cessed, with refrigerated shelf-lives, capable of with L. monocytogenes, either because they in- supporting the growth of L. monocytogenes and con- clude several components and are subject to ex- sumed without further cooking, all of which apply to tensive handling during preparation of the filling pre-prepared sandwiches.The UK Food Standards and sandwich assembly, or as a result of cross-con- Agency currently advises the immunocompromised tamination from the environment. As sandwiches (as well as pregnant women) not to eat soft ripened are ready-to-eat foods, this underlines the impor- cheeses of the brie, camembert or blue-veined tance of high quality ingredients, hygienic manu- bought cooked-chilled meals or ready-to-eat poul- storage in the maintenance of product safety.
try.No specific advice is available concerning The current and recent outbreaks highlight the po- tential for sandwiches contaminated with L. mono- As a result of this outbreak and similar in- cytogenes to cause severe infection in vulnerable cidents, two recent microbiological studies have people. Furthermore, microbiological food studies focused on the quality of sandwiches served to demonstrate that sandwiches served in hospitals selected vulnerable groups in the UK. A UK-wide can be contaminated with low levels of this patho- study of >3000 sandwiches served in hospitals and gen. L. monocytogenes has been found in a range residential/care homes found Listeria spp. in 7.6% of chilled ready-to-eat foods, including pre-packed of sandwiches, with 2.7% of samples containing L. monocytogenes at <10 cfu/g and one at cheeses, cooked sliced meats and smoked salmon.
20 cfu/g.Similar contamination rates for Listeria All foods consumed by hospital patients should be Please cite this article in press as: Shetty A et al., Outbreak of Listeria monocytogenes in an oncology unit associated with sand-wiches consumed in hospital, J Hosp Infect (2009), doi:10.1016/j.jhin.2009.01.012 Listeria outbreak from hospital sandwiches free from potential pathogens, including L. mono- 6. McLauchlin J, Audurier A, Taylor AG. Aspects of the epid- cytogenes, and those responsible for procuring emiology of human Listeria monocytogenes infections inBritain 1967 sandwiches for hospitals should ensure the safety e1984; the use of serotyping and phage typing.
of vulnerable patients in their care. Businesses 7. Farber JM, Peterkin PI, Carter AO, Varughese PV, Ashton FE, manufacturing sandwiches should always aim to Ewan EP. Neonatal listeriosis due to cross-infection con- ensure that L. monocytogenes is absent from their firmed by isoenzyme typing and DNA fingerprinting. J Infect product. The British Sandwich Association recom- 8. Graham JC, Lanser S, Bignardi G, Pedler S, Hollyoak V. Hos- mend a target level of <10 cfu/g for L. monocyto- pital-acquired listeriosis. J Hosp Infect 2002;51:136e139.
genes in sandwiches at the point of productio 9. Dawson S, Evans M, Willby D, Bardwell J, Chamberlain N, Written food safety procedures for hospitals Lewis D. Listeria outbreak associated with sandwich con- should ensure that fridges are monitored regularly sumption from a hospital retail shop, United Kingdom. Euro to keep temperatures below 5 C, in line with re- 10. McLauchlin J. The identification of Listeria species. Int cent recommendations from the British Sandwich 11. Seeliger HPR, Hohne K. Serotyping of Listeria monocyto- Temperature control should be maintained from genes and related species. In: Bergan T, Norris JR, editors.
production until food is served, and sandwiches Methods in Microbiology, vol. 13. London: Academic Press; should be consumed as close as possible to their 12. Doumith M, Buchrieser C, Glaser P, Jacquet C, Martin P. Dif- production date. Foods which carry a greater risk ferentiation of the major Listeria monocytogenes serovars by multiplex PCR. J Clin Microbiol 2004;42:3819e3822.
blue-veined cheeses e should not be given to 13. McLauchlin J, Audurier A, Gerner-Smidt P, et al. WHO study on subtyping Listeria monocytogenes: results of phage-typ-ing. Int J Food Microbiol 1996;32:289e299.
14. Guerra MM, Bernardo FA, McLauchlin J. Amplified fragment length polymorphism (AFLP) analysis of Listeria monocyto-genes. Syst Appl Microbiol 2002;25:456e461.
15. Graves LM, Swaminathan B. PulseNet standardized protocol We thank the Environmental Health Officers and for subtyping Listeria monocytogenes by macrorestriction and pulsed-field gel electrophoresis. Int J Food Microbiol2001;65:55e62.
16. Roberts D, Greenwood M. Practical food microbiology. 3rd 17. Anonymous. Human listeriosis linked to hospital sand- wiches: implication for procurement and storage. Health Protection Report 2008;2:29. August.
18. McLauchlin J. The pathogenicity of Listeria monocytogenes: a public health perspective. Rev Med Microbiol 1997;8:1e14.
19. Anonymous. Food sense no. 1. Food safety. PB0551. London: Ministry of Agriculture, Fisheries and Foods; 1994.
20. Little CL, Barrett NJ, Grant K, McLauchlin J. Microbiological 1. Murray EGD, Webb RA, Swann MBR. A disease of rabbits safety of sandwiches from hospitals and other health care es- characterised by a large mononuclear leucocytosis, caused tablishments in the United Kingdom with a focus on Listeria by a hitherto undescribed bacillus Bacterium monocyto- monocytogenes and other Listeria species. J Food Prot 2008; genes (n.sp.). J Pathol Bacteriol 1926;29:407e439.
2. Farber JM, Peterkin PI. Listeria monocytogenes, a food- 21. Meldrum RJ, Smith RM. Occurrence of Listeria monocyto- borne pathogen. Microbiol Rev 1991;55:752e811.
genes in sandwiches available to hospital patients in Wales, 3. Gillespie I, McLauchlin J, Grant KA, et al. Changing pattern United Kingdom. J Food Protect 2007;70:1958e1960.
of human listeriosis in England and Wales, 2001e2005.
22. Gilbert RJ, De Louvois J, Donovan T, et al. A working group of the PHLS Advisory Committee for Food and Dairy Prod- 4. Denny J, McLauchlin J. Human Listeria monocytogenes in- ucts. Guidelines for the microbiological quality of some fections in Europe: an opportunity for improved pan-Euro- ready-to-eat foods sampled at the point of sale. Commun pean surveillance. Euro Surveill 2008;13:pii8082.
5. McLauchlin J. The relationship between Listeria and listeri- 23. British Sandwich Association Manufacturer Code of Practice, Please cite this article in press as: Shetty A et al., Outbreak of Listeria monocytogenes in an oncology unit associated with sand-wiches consumed in hospital, J Hosp Infect (2009), doi:10.1016/j.jhin.2009.01.012

Source: http://fichero.veterinariaudec.cl/concepcion/wp-content/uploads/2009/05/shetty-2009.pdf