Access the full text.
Sign up today, get DeepDyve free for 14 days.
This study describes an outbreak investigation of 14 hepatitis A cases in the Netherlands. The hepatitis A virus (HAV) genotype IB sequences in cases were highly similar (459/460 nt). The origin of strains could be narrowed to Bulgaria based on information from EPIS-FWD. As an association with consumption of soft fruit was suspected, a case–control study was initiated using a questionnaire and a list of pictures of soft fruit available at the supermarket chain involved. Twelve out of 13 cases consumed a specific frozen raspberry/blueberry product shown on the list (OR 46.0, 95% CI 5.0–27). In multivari- able regression analysis this product was the only risk factor (aOR 26.6, 95% CI 2.0–263). Laboratory analyses could not demonstrate HAV-RNA in batches that had been on the market in the incubation period of patients. Trace back of frozen fruit showed that raspberries had been traded by a producer in Bulgaria. After withdrawal of the product from the supermarket no new cases were reported. Use of advertisement pictures of consumed food was helpful in this investigation. Suspicion of the source was strengthened by data from molecular typing and food trace back activities, underlining the importance of good (inter)national cooperation between public health and food safety organisations. Keywords HAV · Soft fruit · Raspberries · Virus · Outbreak Background endemic anymore, affecting mostly adults, with more severe course of infection (WHO/FAO 2008). Hepatitis A virus (HAV) is a faecal-orally transmitted patho- The notification rate in European Union (EU)/Euro- gen causing acute self-limiting hepatitis. Risk factors for pean Economic Area (EEA) member states for Hepatitis infection include exposure to infected persons, contaminated A shows a steadily decreasing trend over the past decades surfaces, food or water. Hepatitis A can re-emerge in regions (ECDC 2016). As a result introduction of HAV-contami- such as North America and Western Europe, where it is not nated food in EU/EEA countries may lead to diffuse out- breaks that are geographically and temporally dispersed. Between 2007 and 2012, EFSA and ECDC reported 14 Madelief Mollers and Ingeborg L.A. Boxman have shared first authorship. * Madelief Mollers Infectious Diseases, Research, Diagnostics and Screening, firstname.lastname@example.org Centre for Infectious Disease Control, National Institute for Public Health and the Environment, Antonie van National Coordination Centre for Communicable Disease Leeuwenhoeklaan 9, 3720 BA Bilthoven, The Netherlands Control, Centre for Infectious Disease Control, National Incidence and Crisis Centre, Netherlands Food Institute for Public Health and the Environment, Antonie van and Consumer Product Safety Authority (NVWA), Leeuwenhoeklaan 9, 3720 BA Bilthoven, The Netherlands Catharijnesingel 59, 3511 GG Utrecht, The Netherlands European Programme for Intervention Epidemiology Infectious Diseases, Epidemiology and Surveillance, Centre Training (EPIET), ECDC, Tomtebodavägen 11a, for Infectious Disease Control, National Institute for Public 171 65 Solna, Sweden Health and the Environment, Antonie van Leeuwenhoeklaan Laboratory for Feed and Food Safety, Netherlands Food 9, 3720 BA Bilthoven, The Netherlands and Consumer Product Safety Authority (NVWA), Catharijnesingel 59, 3511 GG Utrecht, The Netherlands Vol:.(1234567890) 1 3 Food and Environmental Virology (2018) 10:272–277 273 hepatitis A outbreaks in which there was strong evi- Outbreak Detection dence of food being the infection vehicle (EFSA 2014). More recently, three multinational outbreaks of hepatitis In April 2017, three cases of hepatitis A were notified by A affecting EU/EEA countries were reported with sus- the same Public Health Service (PHS), who had an onset of pected transmission through soft fruit (Gillesberg Lassen illness within 8 days. None of them had a history of travel et al. 2013; EFSA 2014; Sane et al. 2015). In the Nether- to endemic countries or MSM contact. However, cases were lands the incidence rate for hepatitis A is usually low with living in the same neighbourhood and bought food at the on average 0.8 cases per 100.000 population per year in same supermarket store. A local outbreak investigation was 2007–2016 (range 0.5–1.6) (RIVM 2016). The incidence inconclusive. rate increased in 2017 due to an outbreak of hepatitis A A national outbreak investigation was initiated when a among men who have sex with men (MSM) which started fourth and fifth case were reported from two different geo- in 2016 (Freidl et al. 2017). Being a notifiable disease, graphical regions and the RIVM showed that all cases had cases are reported in a national electronic registration sys- been infected by an identical genotype IB strain (see Fig. 1, tem for infectious diseases (Osiris) including demographic timeline). and epidemiological data (Petrignani et al. 2014). Serum or faecal samples from ca. 70% of notified hepatitis A cases are sent to the National Institute for Public Health Methods and Environment (RIVM) for genotyping. Surveillance is intensified for cases with no travel history to endemic Case Finding countries and an unknown source of infection, by admin- istering a hypothesis generating questionnaire (Petrignani HAV in serum or faecal samples was genotyped as earlier et al. 2014). described (Grinde et al. 1997; Stene-Johansen et al. 2007). The present study describes the outbreak investigation Cases were identified as a cluster case if they had been in which molecular genotyping, a case–control study using infected by a genotype IB strain sharing at least 459/460 a list of food pictures, food analysis, and trace back led to nucleotides with the RIVM-HAV17-090 outbreak strain recall of the suspected food product. from 30th of March onwards. 4th Meeting outbreak team 1st Meeting 2nd Meeting 3th Meeting 5th Meeting 6th Meeting outbreak team outbreak team outbreak team outbreak team outbreak team 2 new cases 4 cases First 3 cases different 2 new in same 1 new case 1 new case 1 new case regions cases region April 1st May 1st June 1st July 1st Aug 1st EPIS-FWD: match soft fruit outbreak Canada Update 2 Update 2 EPIS RASFF 1st Message on EWRS and Preliminary RASFF analysis of the Source study shows investigation that 8 out of 8 (supermarket cases have products and eaten the suspected Develop Start case- suspected food employee) questionnaire control study product Withdrawal suspected product from supermarket Fig. 1 Timeline of the outbreak investigation on a possible food-borne hepatitis A outbreak with 14 cases in the Netherlands, 2017 (cases are shown by reporting date in contrast to Fig. 2 epicurve) 1 3 274 Food and Environmental Virology (2018) 10:272–277 Controls Results Controls were recruited from travel clinics of the PHS and Descriptive Epidemiology matched the demographic region of the cases. The aim was to include four controls for each case. If the travel clinic The outbreak cluster, including one secondary case, con- was visited by a couple, family or group, only one person sisted of seven males and seven females, with ages ranging was included in the study. No further matching between from 16 to 60 years. Nine cases had been infected by the cases and controls was performed. Only controls that had HAV IB outbreak strain RIVM-HAV17-090 (460 nt). Four never been HAV-vaccinated, were included. cases had been infected with nearly identical HAV variants with two different single nucleotide differences. Another case was epidemiologically linked to the cluster, being the Questionnaire and Statistical Analyses index case for a confirmed secondary case, but a diagnostic sample was not available for genotyping. The onset of illness An earlier described hypothesis generating questionnaire ranged between 30 March and 30 May 2017, but for most (EFSA 2014) was modified and focused on soft fruit con- cases the onset was within the first month of this period (see sumption (available upon request [in Dutch]). From the Fig. 2, Epicurve). No information on the severity of illness supermarket chain which was visited by all cases for their was known, but 5 out of 13 cases (38%) were admitted to groceries, a list of 24 advertisement pictures of products hospital. None of the patients died. available on the website of the specific supermarket chain was used. The list of advertisement pictures consisted of Genotyping and International Enquiry fresh raspberries and blackberries, cartons or plastic bags of frozen soft fruit, and bottles of juices, smoothies, dress- The outbreak genotype IB strain (RIVM-HAV17-090) was ings, and yoghurt drinks with fruit containing blackber- typical for samples from the northern part of the Eastern ries and/or raspberries. This list was based on results from Mediterranean region (RIVM 2017). The sequence was the national registration questionnaire (Osiris) indicating shared on EPIS-FWD (epidemic intelligence information frequent consumption of blackberries and/or raspberries system for food- and waterborne diseases) (5 May 2017) by cases (10/14). Cases and controls filled in the question- to be able to identify more cases and clues. This revealed naire and the image list either by mail or in the presence of a match with an identical sequence detected in cases of a an officer of the regional PHS. Data were analysed using HAV outbreak associated with consumption of mixed fro- a univariable and multivariable logistic regression model zen berries in Canada, May–June 2016 (PHAC 2016). The (SAS version 9.4 SAS Institute Inc., USA). Canadian outbreak was shown to be associated with HAV- contaminated blackberries from Bulgaria by viral analyses. In addition, the origin of the HAV strain was confirmed Food Analyses independently with information from Italy through EPIS- FWD. It was shown that the strain had been detected earlier Twenty-seven cartons containing a mix of frozen blueber- in clinical samples of patients from Bulgaria. The interna- ries and raspberries were collected by inspectors of the tional community within the EU was notified through EWRS Netherlands Food and Consumer Product Safety Authority (early warning and response system) (10 May 2017) and (NVWA) and transported and stored frozen until analysis RASFF (food and feed safety alerts) (11 May 2017). at the Laboratory for Feed, Food and Product Safety of the NVWA in Wageningen, the Netherlands. The 27 cartons Risk Factors Analyses represented 13 different batches known to be present on the market in the incubation period of the patients, one to Thirteen of the 14 cases, including the non-confirmed four cartons per batch. Of each carton three subsamples case, and 29 controls completed the questionnaire and of 25 grammes were analysed. Analysis for the presence image list. The median age of the cases was 29 years (range of HAV was performed according to ISO15216-2 under 16–60 years) compared to 27 years (range 20–54 years) of accreditation of the Dutch Council for Accreditation. Prior the controls. Fifty-four percent were men compared to 45% to reverse-transcription real-time polymerase chain reac- of controls. tion (RT qPCR) all RNA samples were treated to reduce Based on the questionnaires, frozen mixed fruit was inhibitory substances and increase the detectability of the consumed by all cases, but by only 24% of the controls target RNA using the OneStep PCR Inhibitor Removal Kit (OR 38.5, 95% CI 5.0–853). In addition, there was a clear (Zymo Research) (Boxman et al. 2016a). association with consumption of frozen raspberries, mixed 1 3 Food and Environmental Virology (2018) 10:272–277 275 Hepas A outbreak in frozen so fruit, 30 March-30 May 2017, the Netherlands 2 Possible case Conﬁrmed case Date of disease onset Fig. 2 Epicurve of a hepatitis A outbreak related to frozen soft fruit with date of disease onset of cases between 30 March–30 May 2017, the Netherlands frozen fruit or consumption of smoothies made of mixed a blend of frozen fruit as risk factor, which was a blend of fruit (Table 1), but for none of the 94 other food items. raspberries, blackberries, strawberries, cherries, blueber- Twelve out of 13 cases recalled to have consumed ries and red currants (Table 1). In multivariable regression one particular raspberry/blueberry product shown on the analysis, the specific carton with mixed frozen blueberries advertisement picture list (OR 46.0, 95% CI 5.0–427). and raspberries was the only remaining risk factor (aOR Univariable logistic regression analyses further identified 26.6, 95% CI 2.0–263). Table 1 Food items consumed a Cases (n = 13) Controls (n = 29) OR 95% CI aOR 95% CI within 6 weeks before onset of illness by hepatitis A cluster Questionnaire cases (n = 13) in comparison b c d Frozen mixed fr uit 13 (100%) 7 (24%) 38.5 (5.0–853) to controls (n = 29), the Frozen raspberries 9 (69%) 8 (26%) 5.6 (1.3–23.6) NS Netherlands, April–May 2017 Mixed fruit smoothie 4 (31%) 1 (3%) 12.4 (1.2–126) NS Image list Carton of frozen blue- 12 (92%) 6 (21%) 46.0 (5.0–427) 26.6 (2.0–263) berries/raspberries Carton of frozen fruits 9 (69%) 3 (10%) 19.5 (3.6–104) NS Bag of frozen fruits 5 (38%) 2 (7%) 8.4 (1.4–52.1) NS NS not significant aOR = adjusted OR obtained from the multivariable model Any fr uit The OR was calculated by hand by adding 1 extra person to each cell in a 2 × 2 table, because all cases had consumed this product Not possible to calculate Results for one case are missing Raspberries, blueberries, blackberries, strawberries, cherries and red currants 1 3 Number of Hepas A cases 276 Food and Environmental Virology (2018) 10:272–277 for typing of diagnostic samples (70%) (Petrignani et al. Trace Back and Viral Analyses of Soft Fruit Samples 2014). Through this surveillance system, several, mainly small, food-borne hepatitis A outbreaks have been identi- The suspected raspberry/blueberry product was exclusively sold at shops of one supermarket chain in the Netherlands. fied in recent years, e.g. outbreaks associated with semi- dried tomatoes (Petrignani et al. 2010; Fournet et al. 2012), Requested trace back data for this particular product in the incubation period of the cases (half February until half soft fruit (EFSA 2014), as well as mussels (Boxman et al. 2016b). In the present outbreak, cases were identified as a April) revealed that 24 batches had been on the market, and that in this particular period raspberries were traded by a cluster despite the large (inter)national outbreak in MSM that occurred at the same time (Freidl et al. 2017), as out- producer in Bulgaria whereas the blueberries were traded by producers in other countries. Herewith the raspberries break strains were different. The fact that all patients went to the same supermarket became the suspected food type, as the country trading the raspberries corresponded to the most likely country origin chain for their groceries, led to the idea of adding a list of the supermarket’s advertisement pictures of raspberry and of the outbreak HAV strain. No common suppliers could be identified between this and the Canadian outbreak (PHAC blackberry containing products. The use of pictures has earlier shown to be helpful in recall of consumption of sus- 2016), personal communication). Analyses of samples were restricted to those batches that pected products (Guzman-Herrador et al. 2014). Also in the present study, the picture list had a clear added value, as it had been on the market in the exposure period of the cases. As raspberries were the suspected fruit type in the mix, three identified one specific product, i.e. the brand and the com- position, whereas the questionnaire only identified mixed subsamples of raspberries only were weighed from each of the 27 cartons and analysed. None of the 81 subsamples frozen fruit as a risk factor. Especially the fact that almost all cases ate the same product was strong. Trace back data tested positive for HAV-RNA. Quality assurance parameters showed an extraction efficiency of the added murine noro- indicated the raspberries in the box of frozen raspberries and the box of mixed frozen fruit originated from the same virus process control of 8.2 ± 4.7% (range 4–16%) and no inhibition in the RNA samples was seen. producer in Bulgaria. Interestingly, most cases recalled consumption of the Based on the epidemiological findings, specified batches to which cases most likely had been exposed were classified frozen soft fruit products five or more times a week. It sug- gests that such consumption patterns increased the risk of as potentially unsafe for human consumption by the NVWA. The retailer subsequently initiated a total withdrawal of infection in these cases, particularly when the contamination of these raspberries is low, possibly too low to be detected the product (12 June 2017) for replacement with batches coded with later production dates. No new cluster cases by viral analyses and also maybe too low to infect a person after a single exposure only. Due to low numbers we were were identified thereafter. International alerts and sharing of sequences in the international databases did not result in not able to explore this possible dose–response association any further. identification of additional cases. Detection of the viral strain in the implicated food is dif- ficult and often hampered by absence of leftovers, absence Discussion of an adequate method for extraction or heterogeneous dis- tribution of low amounts of virus in different batches. In this Combined analyses of data has led to the conclusion that outbreak the contaminated batch(es) might even not have been among the tested batches. Only samples of 13 of the 24 frozen raspberries from Bulgaria were the most likely source for this food-borne hepatitis A virus outbreak. Molecular batches that were present on the market during the incuba- tion period of the cases were available for testing, also indi- typing showed that cases clustered by being infected by (nearly) the same (459/460 nt) HAV strain. This strain was cating the high through-put of these products on the market. It is expected that outbreaks related to soft fruit keep most likely associated with Eastern Mediterranean countries, but the geographical association could be further narrowed occurring and being reported (Tavoschi et al. 2015). The berry production has experienced an increase in the last to Bulgaria by typing information of cases from Bulgaria (EPIS-FWD) as well as cases in a Canadian outbreak asso- decade and this food commodity often receives no or only minimal (industrial) processing. The FAO/WHO (FAO/ ciated with Bulgarian fruit (PHAC 2016). Our case–control study identified a raspberry/blueberry product, containing WHO 2012) has produced a Codex guideline for better hygienic practices to reduce the illness by viral contami- Bulgarian raspberries, consumed by 92% of the cases (aOR 26.6, 2–263). nated food, including specific recommendation for berries production in a separate Annex. Food business operators Identification of cluster cases that are geographically and temporarily dispersed strongly depend on national sur- and trade associations have an important role in providing specific instructions and training of personnel for control veillance, which in the Netherlands has a high coverage 1 3 Food and Environmental Virology (2018) 10:272–277 277 of general principles of food hygiene to the control of viruses in of viruses. In countries where HAV is endemic, risk com- food (CAC/GL 79–2012). munication should focus on preventive measures of which FAO/WHO. (2008). Food and Agriculture Organization of the United hand washing after using the toilet at the farm level and Nations/World Health Organization. Viruses in food: Scientific providing good equipped sanitary facilities to the amount of advice to support risk management activities: Meeting report. Microbiological Risk Assessment Series, No. 13. Retrieved seasonal influx of workers seems the most effective (FAO/ Feburary 17, 2012. http://www.who.int/foods afety /publi catio ns/ WHO 2008). Other options, already implemented by some micro /Virus es_in_food_MRA.pdf. countries is to harvest berries mechanically, or to boil frozen Fournet, N., Baas, D., van Pelt, W., Swaan, C., Ober, H., Isken, L., et al. berries before use (Gillesberg Lassen et al. 2013), but boil- (2012). Another possible food-borne outbreak of hepatitis A in the Netherlands indicated by two closely related molecular sequences, ing does alter the taste and texture. July to October 2011. Euro Surveill, 17(6), 20079. In conclusion, the retailer was willing to recall the impli- Freidl, G. S., Sonder, G. J., Bovee, L. P., Friesema, I. H., van Rijckevor- cated product by the strong epidemiologically link of the sel, G. G., Ruijs, W. L., et al. (2017). Hepatitis A outbreak among cases with frequent consumption of one specific product. men who have sex with men (MSM) predominantly linked with the EuroPride, the Netherlands, July 2016 to February 2017. Euro Since the recall, no new cases with the same genotype were Surveill. https://doi.or g/10.2807/1560-7917.ES.2017.22.8.30468. reported in the Netherlands. The present outbreak investiga- Gillesberg Lassen, S., Soborg, B., Midgley, S. E., Steens, A., Vold, L., tion showed the strength of advertisement pictures in addi- Stene-Johansen, K., et al. (2013). Ongoing multi-strain food-borne tion to questionnaires to assist recall of implicated food in hepatitis A outbreak with frozen berries as suspected vehicle: four Nordic countries affected, October 2012 to April 2013. Euro Sur - outbreak studies. Data from molecular typing and food trace veill, 18(17), 20467. back activities strengthened the link to the suspected source. Grinde, B., Stene-Johansen, K., Sharma, B., Hoel, T., Jensenius, M., This underlines the importance of good (inter)national coop- & Skaug, K. (1997). Characterisation of an epidemic of hepatitis eration between public health and food safety organisations A virus involving intravenous drug abusers-infection by needle sharing? Journal of Medical Virology, 53(1), 69–75. to fully profit from each other’s expertise. Guzman-Herrador, B., Jensvoll, L., Einoder-Moreno, M., Lange, H., Myking, S., Nygard, K., et al. (2014). Ongoing hepatitis A out- Acknowledgements Marieke Timmer, Titia Kortbeek, Stefanie Schoss, break in Europe 2013 to 2014: imported berry mix cake suspected Judith LeBlanc, Mariska Petrignani, Wendy Tas, Mirjam Knol; NVWA to be the source of infection in Norway. Euro Surveill, 19(15), Laboratory for technical assistance (Claudia Jansen, Kyara Edelenbos- Klunder, Geke Hägele, Nico Nendos, Sanne Ockers, Patricia Ramzan, Petrignani, M., Harms, M., Verhoef, L., van Hunen, R., Swaan, C., van Sandra Verheij-Laan, Kariene Wennekes, Hüseyin Yilmaz, Ans Zwart- Steenbergen, J., et al. (2010). Update: a food-borne outbreak of kruis-Nahuis); travel clinics of Public Health Services (PHS). hepatitis A in the Netherlands related to semi-dried tomatoes in oil, January–February 2010. Euro Surveill, 15(20), 19572. Open Access This article is distributed under the terms of the Crea- Petrignani, M., Verhoef, L., Vennema, H., van Hunen, R., Baas, D., van tive Commons Attribution 4.0 International License (http://creat iveco Steenbergen, J. E., et al. (2014). Underdiagnosis of foodborne hep- mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- atitis A, the Netherlands, 2008–2010. Emerging Infectious Dis- tion, and reproduction in any medium, provided you give appropriate eases, 20(4), 596–602. https ://doi.org/10.3201/eid20 04.13075 3. credit to the original author(s) and the source, provide a link to the PHAC. (2016). Public Health Agency of Canada. Public Health Creative Commons license, and indicate if changes were made. Notice: Outbreak of Hepatitis A infections; consumers advised not to eat Nature’s Touch Organic Berry Cherry Blend frozen fruit. Retrieved August 3, 2017 from http://www.phac-aspc.gc.ca/ phn-asp/2016/hepat itisa -eng.php. References RIVM. (2016). Meldingen wet publieke gezondheid. Infectieziekten Bulletin, 28(10), 411. Boxman, I., Hägele, G., & Jansen, C. (2016a). Improvement of virus RIVM. (2017). HAVNET global network for hepatitis A reference labo- extraction from soft fruit by implementing a PCR inhibitor raties. Retrieved May 5, 2017 from http://www.rivm.nl/en/Topic removal kit. Abstract T4-12, IAFP, presented 1 August 2016 IAFP, s/H/HAVNE T. . St Louis, MO. Sane, J., MacDonald, E., Vold, L., Gossner, C., Severi, E., & Investi- Boxman, I., Verhoef, L., Vennema, H., Ngui, S., Friesema, I., White- gation, O., T (2015). Multistate foodborne hepatitis A outbreak side, C., et al. (2016b). International linkage of two food-borne among European tourists returning from Egypt-need for rein- hepatitis A clusters through traceback of mussels, the Netherlands, forced vaccination recommendations, November 2012 to April 2012. Euro Surveill, 21(3), 30113. https ://doi.org/10.2807/1560- 2013. Euro Surveill, 20(4), 21018. 7917.ES.2016.21.3.30113 . Stene-Johansen, K., Tjon, G., Schreier, E., Bremer, V., Bruisten, S., ECDC. (2016). Hepatitis A virus in the EU/EEA, 1975–2014. A system- Ngui, S. L., et al. (2007). Molecular epidemiological studies show atic review of seroprevalence and incidence comprising European that hepatitis A virus is endemic among active homosexual men surveillance data and national vaccination recommendations. in Europe. Journal of Medical Virology, 79(4), 356–365. https :// Technical report. doi.org/10.1002/jmv.20781 . EFSA. (2014). European food safety authority. Tracing of food items Tavoschi, L., Severi, E., Niskanen, T., Boelaert, F., Rizzi, V., Liebana, in connection to the multinational hepatitis A virus outbreak in E., et al. (2015). Food-borne diseases associated with frozen ber- Europe. EFSA Journal, 12(9), 3821. https ://doi.org/10.2903/j. ries consumption: A historical perspective, European Union, 1983 efsa.2014.3821. to 2013. Euro Surveill, 20(29), 21193. FAO/WHO. (2012). Food and Agriculture Organization of the United Nations/World Health Organization, guidelines on the application 1 3
Food and Environmental Virology – Springer Journals
Published: May 4, 2018
Access the full text.
Sign up today, get DeepDyve free for 14 days.