Hepatitis E virus infection in Europe: surveillance and descriptive epidemiology of confirmed cases, 2005 to 2015

Hepatitis E virus (HEV) is an under-recognised cause of acute hepatitis in high-income countries. The purpose of this study was to provide an overview of testing, diagnosis, surveillance activities, and data on confirmed cases in the European Union/European Economic Area (EU/EEA). A semi-structured survey was developed and sent to 31 EU/EEA countries in February 2016, 30 responded. Twenty of these countries reported that they have specific surveillance systems for HEV infection. Applied specific case definition for HEV infection varied widely across countries. The number of reported cases has increased from 514 cases per year in 2005 to 5,617 in 2015, with most infections being locally acquired. This increase could not be explained by additional countries implementing surveillance for HEV infections over time. Hospitalisations increased from less than 100 in 2005 to more than 1,100 in 2015 and 28 fatal cases were reported over the study period. EU/EEA countries are at different stages in their surveillance, testing schemes and policy response to the emergence of HEV infection in humans. The available data demonstrated a Europe-wide increase in cases. Standardised case definitions and testing policies would allow a better understanding of the epidemiology of HEV as an emerging cause of liver-related morbidity.


Introduction
Hepatitis E virus (HEV) infection is one of the leading causes of acute viral hepatitis worldwide with four different genotypes (1-4) responsible for most human infections. HEV genotype 3 predominates in high-income countries, including those in Europe. Transmission of this genotype is usually zoonotic and has been linked to the consumption of pork products, and in some instances, shellfish [1][2][3]. Infection, which is most often asymptomatic, may cause an acute selflimiting hepatitis, with symptomatic infection more commonly reported among men older than 50 years of age [4,5]. Reports of chronic infection among immunocompromised people or those with pre-existing liver disease have been described [4].
There is evidence that HEV is an under-recognised pathogen in high-income countries, and that the incidence of confirmed cases has been steadily increasing over the last decade [6][7][8][9][10]. Although population studies have shown stable or decreasing seroprevalence rates [11,12], some countries have reported a consistently high seroprevalence and proportion of HEV-RNA-positive blood donors [13][14][15][16]. A systematic review conducted by the World Health Organization (WHO) reported HEV seroprevalence (as denoted by presence of IgG antibody) to range from 0.03% to 52% among the general population and blood donors in the WHO European Region, with the highest prevalence reported from studies in blood donors from France and the Netherlands [17]. Evidence of current HEV infection in up to 77% of patients presenting with symptomatic acute hepatitis suggests that the virus could be a considerable cause of liver morbidity in European countries [17]. However, the use of different serological test systems significantly influences the estimates of seroprevalence [18].
Europe-wide surveillance is not in place for hepatitis E, and the European Centre for Disease Prevention and Control (ECDC) has launched a number of activities that aim to better understand the current epidemiology, as well as the national monitoring systems for HEV infection within European Union/European Economic Area (EU/EEA) countries. A recent study involving HEV experts from 17 countries documented an increase in reported cases in nearly all EU/EEA countries that provided data, with France, Germany, England and Wales, and the Netherlands reporting more hepatitis E than hepatitis A notifications. The study also highlighted a lack of consistency in the reporting of cases, and wide variation in surveillance activity and case definitions [7]. The current study builds on this work to conduct a wider assessment of testing, diagnosis, surveillance activities and confirmed cases of HEV infection in the EU/EEA in order to inform the Europe-wide response to this emerging infection.

Methods
A semi-structured survey of EU/EEA countries was conducted during 2015/16. The survey was divided into four sections (surveillance, testing and diagnosis, data on diagnosed cases, and transfusion-associated HEV infections) to allow separate country-level respondents for each topic area with a total of 24 open and closed questions [19]. A pilot survey was conducted at a meeting of the ECDC HEV expert group in December 2015, and a revised survey was subsequently circulated by email to nominated representatives of ECDC's networks, the National Focal Points for Foodand Waterborne Diseases and Zoonoses as well as for Microbiology and Surveillance, in February 2016. Respondents were given three weeks to complete the survey (which could be completed electronically), and email and phone reminders were used to follow up with those who had not responded.
Returned survey data were manually extracted into Excel and double-checked. Any unclear responses were checked by email communication with the respective experts. Analyses were conducted in Excel and STATA version 13. The United Kingdom (UK) provided three separate responses (from England and Wales, Scotland, and Northern Ireland) which were considered as a single country response for all quantitative analyses and listed separately for qualitative analyses. A descriptive analysis was performed. The proportion of hospitalised cases was calculated using data exclusively from countries reporting hospital data, with the number of total cases as denominator. The Eurostat     population database from 2015 was used as source of population denominator data [20].

Results
All EU/EEA countries with the exception of Liechtenstein responded to the survey, giving a response rate of 30 out of 31. The data provided from countries that conduct surveillance was comprehensive, detailed and of high quality.

Hepatitis E surveillance systems
Of 30 responding countries, 20 had specific surveillance systems for HEV infection. The remaining 10 countries had more generic viral hepatitis syndromic surveillance with no plans to develop any specific surveillance for HEV infection. Of the 20 EU/EEA countries with specific systems for HEV infection, 15 had national surveillance systems, three had laboratory surveillance through national reference laboratories with unknown coverage, one had blood service surveillance with unknown coverage, and one had a sentinel surveillance system covering approximately half of the population.
Thirteen of 20 surveillance systems specific for HEV infection had a case definition for confirmed cases.
There was wide variation in the type of case definition used: seven used laboratory and clinical definitions, and six used a laboratory definition only ( Figure  1). Most surveillance systems collected demographic information as well as the date of notification and onset of disease, whereas morbidity and mortality data were collected less frequently (Table 1). Two countries' surveillance systems (Ireland and the UK (England and Wales)) differentiated between acute and chronic HEV infection: acute infection was defined as both HEV-IgM-positive and IgG-positive, or HEV-RNA-positive, and chronic infection was defined as HEV RNA persisting for at least 3 months.

Hepatitis E testing and diagnosis
Twenty-six countries conducted HEV testing in their own countries, with two countries sending samples abroad for testing and two reporting their testing availability as unknown. In 12 countries, there were less than five laboratories able to conduct diagnostic tests for HEV. Laboratories in 20 countries used more than one type of test (

Confirmed cases of hepatitis E virus infection
Twenty-two countries covering 469 million people (i.e. 91% of the EU/EEA population of 514 million people) provided data on confirmed cases of hepatitis E ( Figure  2). The total number of reported cases increased from 514 cases in 2005 to 5,617 cases in 2015, with 21,018 cases overall during the 11-year period studied ( Figure 2

Severity associated with hepatitis E virus infection
Data from 14 countries reporting on the total number of hospitalisations related to HEV infection showed an increase from 85 hospitalised cases in 2005 to Year of reporting In most countries, it was not possible to ascertain the criteria for referral or for HEV test requests. While there may be an increasing awareness of locally acquired HEV infections, there is still concern that within parts of the clinical community, HEV is not considered as a possible cause of hepatitis unless there is a recent history of travel. It is therefore likely that the infection remains under-diagnosed. There also appears to be a gap in knowledge around the testing protocols used in outpatient and hospital settings within countries, which should be investigated further to understand when, who, and why patients are tested for HEV.
Twenty-two countries (comprising more than 90% of the total EU/EEA population) were able to provide data on confirmed cases for at least some of the period 2005-2015. These data demonstrate that the number of reported cases has been increasing year-on-year, with a more than threefold increase between 2011 and 2015. The majority of cases were reported by three countries (the UK, France and Germany) that have had surveillance systems in place since at least 2005. A small number of countries reported changes to surveillance systems during the study period; for example, Belgium created a national reference centre for HEV in 2011 and Portugal commenced web-based notification in 2014. However, there were no changes that could have contributed substantially to the increase in the number of confirmed cases. A greater awareness of and increased testing for HEV, particularly of patients presenting with acute hepatitis by general practitioners in outpatient settings, is likely to be one of the contributing factors e.g. through the inclusion of HEV in some standard diagnostic protocols. The rise in the number of symptomatic cases might also be related to the switch in virus subtypes that has been observed in some EU/EEA countries [7,21].
The concurrent increase in number of hospitalisations (a measure of increased HEV infection severity) and     gradual decrease in proportion of total cases hospitalised between 2005 and 2015 may suggest better outpatient diagnosis and an increase in the sensitivity of the diagnostic system, i.e. a change in testing practice from hospitals to community settings. For cases where clinical information was recorded, every second case was admitted to hospital and there were two to eight deaths associated with HEV infection annually during 2011-2015. These figures appear high for a disease that has previously been considered to be a mild selflimiting infection [22]. However, a testing bias and overestimation of the hospitalisation rate can be assumed because of a higher likelihood that cases of severe liver disease or other comorbidities are tested for HEV. Notably, only four countries were able to provide data on morbidity status since 2005 and only five were able to provide data on case fatality since 2005, although nearly half of the EU/EEA countries indicated that they are able to collect this information as of 2015. More data are needed to address the lack of knowledge in this area.

Country
As our results show, the number and proportion of autochthonous HEV infections has been increasing since 2005. Only a very small number of cases were known to be related to travel outside of the EU/EEA. Although genotype data were not requested from countries, it is known that autochthonous cases tend to be due to HEV genotype 3, the same genotype present in European pigs and consumed pork products; travelrelated cases would be expected to be infected with genotype 1 or 4 [2][3][4][5]. Lately, there appears to have been a change in circulating subtypes in Europe. For example, the number of subtype 3c recently exceeded the numbers of subtypes 3e, f and g in humans in England and Wales [21]. Such changes may be contributing to the emergence of HEV as an important infection in humans. Changes in the pattern of food preparation and consumption may also be implicated. A case-control study in England and Wales identified pork sausages and ham purchased in supermarkets as possible sources of infection, raising concerns about whether current practice in preparing these products is sufficient to prevent HEV infection [23]. More studies among humans, animal populations and on food are essential to informing on possible sources of HEV infection.
The main limitations of this study are the variation in the denominator data for reported cases of hepatitis E (most countries were only able to provide a subset of demographic information or a limited number of years of data) and the merging of data on reported cases of hepatitis E despite the different case definitions and testing approaches used by EU/EEA countries. These limitations have in part been addressed by showing data by year of surveillance introduction and by limiting the denominator to cases where the relevant information was known. The varying number of countries providing data inconsistently over the study period prompted us not to perform any trend analysis to avoid further introduction of bias. In the future these limitations could be addressed by standardisation of surveillance systems for HEV infection, harmonisation of the case definition as well as testing algorithms across EU/ EEA countries. This would enable better assessment of the burden of disease due to HEV in Europe and facilitate a better understanding of the actual number of performed tests or population coverage as a denominator, thus avoiding diagnostic and surveillance artefacts.
In summary, EU/EEA countries are at different stages in their surveillance and testing capacity for HEV, and there is no standardised European case definition. For the time period 2005-2015, at least 22 countries were able to report on cases infected with HEV, either through formal surveillance or existing systems of laboratory notifications. WHO's global health sector strategy on viral hepatitis asks countries to establish surveillance for viral hepatitis, particularly in blood donors [24]. Our study describes the availability of general hepatitis E surveillance in most EU/EEA countries. These data demonstrate a Europe-wide increase in reported cases, as well as a high number of reported HEV-related hospitalisations and a total of 28 deaths, between 2005 and 2015. They also provide a better understanding of the epidemiology and burden of HEV as an emerging cause of liver-related morbidity in EU/ EEA countries.