Review Article

Hepatitis E—A Concise Review of Virology, Epidemiology, Clinical Presentation and Therapy

M. C. Donnelly; L. Scobie; C. L. Crossan; H. Dalton; P. C. Hayes; K. J. Simpson


Aliment Pharmacol Ther. 2017;46(2):126-141. 

In This Article

Genotype 3 and 4 Infection


HEV genotypes 3 and 4 are recognised to infect both humans and animals, in contrast with genotype 1 and 2; pigs, deer and wild boar have all been identified as reservoirs of infection. Genotype 4 infection mainly occurs in South-East Asia.[39,41] Genotype 3 HEV is the most prevalent genotype causing autochthonous (locally acquired) infection in developed countries.[5] Many seroprevalence studies have been undertaken in Europe, and the results have shown a high variability in seroprevalence rates. A recent meta-analysis identified 73 studies of HEV seroprevalence in Europe; estimates of seroprevalence ranged from 0.6% to 52.5%, with rates increasing with age but unrelated to gender.[60] In the USA, seroprevalence for anti-HEV is around 6%,[61] in the UK 3–16%[62] and in some regions of France up to 52%.[63] In England, the number of confirmed (symptomatic) cases of nontravel associated HEV infection has steadily increased over the past 14 years, from 124 in 2003 to 958 in 2015.[64] However, a study from South-East England suggested that there are 80 000–100 000 infections per year in England, the majority of which are asymptomatic.[65] Autochthonous (locally acquired) HEV infection is not a benign condition, with mortality rates up to 27% reported in patients with underlying chronic liver disease (Figure 1).[66]

Interesting geographical variations in genotype 3 HEV infection have been observed in France where there is considerable variation in seroprevalence by region from 8–86%;[67] very high seroprevalence occurs in the southwest, southeast and northeast of the country.[66–68] The reason for this interesting observation is unclear, but contaminated foodstuffs in the food chain are likely to explain these geographical differences in part. There appears to be no correlation with potential transmission routes (eg, location of pig farms). Although Scotland is a relatively low seroprevalence region, seroprevalence rates are also variable.[69] This geographical variation in HEV infection is worthy of further consideration and investigation. As in France, the main pig-rearing/farming region is located in the North East of Scotland, in contrast to the area of peak HEV seroprevalence.[69,70] In a study of patients with decompensated chronic liver disease from the UK and France, HEV was significantly more common in the French cohort compared with the UK cohort (7.9% vs 1.2% respectively; P=.003).[66] Potential explanations for this include the quantum of circulating HEV in these respective regions, and the exposure to contaminated foodstuffs.

Many countries have undertaken epidemiological studies of HEV seroprevalence in their respective blood donor populations ( Table 2 ).[71–80] Most countries report increasing incidence of HEV infection due to increased awareness of HEV, increased testing for the virus and a true increase in the numbers of new infections. Data from the Netherlands and Scotland suggest that this increase in incidence appears to be in younger patients, in contrast with previous observations that autochthonous HEV infection predominantly affects older (>60 years) males.[69,81] The results of a large survey of hepatitis E infection in French blood donors have recently been reported. Overall IgG seroprevalence was 22.4%, with an IgM seroprevalence of 1%.[67] IgM seroprevalence was highest in those patients living in the south of France and in those patients who self-reported consumption of pork liver sausage, pate and wild boar meat. The presence of HEV RNA was not reported upon in this large study, and there was no information given to suggest that any recipients of products from donors who were IgM positive developed active HEV infection.

HEV genotype 4 is endemic in China, Japan and Indonesia, and recently cases have been reported in Western countries including Belgium, Germany and France.[82–84] There was an outbreak of genotype 4 infection in Italy in 2011.[85] This outbreak was not directly linked to travel or infection by imported foods, raising the possibility of newly imported strains. All patients affected in this outbreak were male (mean age 59 years), and fatigue was the most frequently reported symptom.

Importantly, the presence of HEV antibodies does not protect from further infection. A French group performed a longitudinal study of multi-transfused immunocompetent patients in France. In this study, one seropositive patient demonstrated an increase in IgG level and HEV RNA reappearance, suggesting that reinfection does occur. The rates of reinfection and association with HEV antibodies warrants further investigation.[86]

Transmission: Foodborne

Genotype 3 and 4 infections are most commonly transmitted via contaminated foodstuffs. These foodstuffs include porcine liver and sausage products, shellfish, green vegetables and strawberries. In the UK, a questionnaire-based study identified that risks for autochthonous infection include consumption of processed pork products, including pork pies (OR 6.33), sausages (OR 7.59) and ham (OR 10.98).[87] In a study of blood donors exposed to HEV infection in Southern France, foodstuffs associated with positive antibodies against HEV on multivariate analysis were uncooked pork liver sausages, offal and mussels.[73] To support this, HEV has been found to be highly prevalent amongst global pig populations. For example, in Germany, 33% of wild boar and 50% of domestic pigs are seropositive for anti-HEV IgG.[88,89] HEV infection has been identified in more than 80% of some pig herds in the USA, Canada and in the UK (England),[90,91] although Scottish herds have a lower seroprevalence of around 62%.[92] In a more recent report on English and Northern Irish pigs, 93% of slaughter age animals were seropositive.[90]

In the West of Scotland, 92% of tested wild caught mussels were PCR positive for HEV and consumption of undercooked/raw shellfish is another viable route of transmission.[46] Likewise, numerous studies have indicated HEV contamination of soft fruits, likely via exposure to contaminated water. In Quebec, Canada swine HEV was detected in strawberries (1.6% of samples tested).[47] A separate study of the European berry fruit supply chain identified HEV in 2.6% of berries (raspberries) at point of sale.[93] A more recent study identified HEV in 5% of irrigation water samples from leafy green vegetable production chains.[94]

In view of the seroprevalence in animal populations described above, it seems logical that HEV can be transmitted to humans via the consumption of contaminated foodstuffs. Several studies have identified actual transmission of HEV to humans via the consumption of contaminated foodstuffs. One study investigated the role of figatellu (a traditional pig liver sausage eaten in France and commonly consumed raw): anti-HEV IgM or HEV RNA was positive in seven of 13 individuals who consumed raw figatellu, compared with zero of five individuals who did not eat figatellu.[95] Genetic links were identified between HEV RNA sequences recovered from supermarket figatellu and sequences recovered from patients eating the same product, providing firmer evidence that human HEV infection is associated with figatellu consumption. More recently, Lee reported upon a 55 year old man who was found to be HEV RNA positive at 22 months after liver transplantation.[48] The patient was Muslim, and therefore the potential route of food-borne transmission unclear. Phylogenetic analysis confirmed the patients HEV sequence to belong to camelid HEV; the patient owned a camel farm and subsequently confirmed regular consumption of camel meat and milk, making transmission via camel products the most likely source. In China, where mixed farming is common practice, a high prevalence of active HEV infection in cows was identified, and Huang demonstrated that HEV is excreted into milk that is produced by infected cows.[96] Therefore, HEV-contaminated cows' milk is another potential zoonotic source; gavage of infected milk to rhesus macaques resulted in active HEV in infection as confirmed by HEV RNA in blood and faeces. However, in milk samples collected from dairy farms (ie, not mixed farms) in Germany, no HEV RNA was detected.[97] The exact contribution of zoonotic HEV infection via dairy milk, and the potential contamination via mixed farming remains to be established.

Transmission: Via Blood Products

Transmission of genotype 3 and 4 HEV by transfusion of blood products (including red cells, platelets and even pathogen-inactivated [Intercept treated] fresh frozen plasma) that are HEV-infected has also been reported in many Western (and some Asian) countries.[65,98,99] The incubation period for genotype 3 HEV infection in immunosuppressed patients with blood-borne HEV infection has been demonstrated to be 50–60 days, compared with less than 30 days for immunocompetent patients with genotype 1 infection.[100] Many countries have undertaken seroprevalence studies in their blood donor populations ( Table 2 ).[71–80]

In Scotland, an increase in seroprevalence of genotype 3 infections has been observed; studies of the Scottish blood donor population in 2012 revealed 1 in 14 500 donors to be viraemic. More recent data, have shown donor viraemia in Scotland to have increased significantly, including in younger donors (Petrik J. Personal communication).

In South-East England, retrospective screening of 225 000 individual blood donations identified HEV RNA in 79 samples, equating with a prevalence of viraemia of 1 in 2848 donations.[65] Of all RNA-positive samples undergoing genotyping, genotype 3 virus was identified in all cases. A total of 79 donations from viraemic donors had been used to prepare 129 blood components, 62 of which had been transfused. Of the recipients of these components, 42% had evidence of infection and 10 patients developed persistent infection. This study also suggested increased levels of circulating virus were associated with increased risk of infection.

A study of Catalonia (Spain) blood donors reported a prevalence of anti-HEV IgG of 19.96% (Wantai assay), with a HEV RNA positivity rate of 0.03%, or one per 3333 donations.[71] Study of the blood donor population in southwestern France found that anti-HEV IgG was detectable in 52.5% of blood donors, with seroprevalence increasing with age and associated with rural residence.[63] Another group studied the presence of HEV RNA in manufacturing plasma pools from North America, Europe, the Middle East and Asia.[101] Asian pools were most frequently positive for HEV RNA and had higher viral loads, and there was no evidence of HEV in pools tested from the Middle East, presumably relating to the low rates of pork consumption in this region.

Transmission: Via Solid Organ Transplantation

Transmission of HEV infection can occur via liver transplantation and transplantation of non-hepatic grafts. Schlosser described a case of a 73 year old man in whom HEV transmission occurred after transplantation of a HEV-infected liver from a donor with occult HEV infection.[102] At the point of donor death, alanine aminotransferase was 4x upper limit of normal. The patient developed rapid graft cirrhosis and died from decompensated liver disease and septic shock. HEV was diagnosed at the time of hepatic decompensation; retrospective testing of a stored serum sample from 150 days post-transplant was also positive for HEV RNA. Pre-mortal blood and liver tissue from the donor confirmed that the patients serum was HEV PCR negative, but HEV RNA was detected in high concentrations in the liver tissue of the donor. Sequence data from recipient serum and donor liver tissue were concordant, suggesting transmission of the virus via the transplanted liver. More recently, there has been a report of HEV transmission via renal grafts.[103] The donor kidneys were transplanted into two separate recipients, and of note donor LFTs had been abnormal prior to transplantation. The first infected recipient presented at 9 months with deranged LFTS. Retrospective analysis detected negative HEV RNA until the day of transplantation, with positive RNA from the first month post-transplant. The second recipient was identified after biomonitoring. Donor serum was positive for HEV RNA; genotype 3f was identified and this genotype was also identified in both recipients, providing evidence for transmission via nonhepatic solid organs.

Clinical Features

Immunocompetent individuals clear HEV promptly, usually within a few weeks. In these patients, HEV infection usually runs a mild course and is often asymptomatic. However, in the immunocompromised, HEV infection is often more difficult to clear, and about 60% of these patients go on to develop chronic HEV infection.[6,7] Chronic HEV infection may lead to complications such as liver cirrhosis.

Clinical Features of Genotype 3 HEV Infection. In 2008, it was established that genotype 3 HEV infection may progress to chronic infection in patients who are immunocompromised (eg, patients with HIV, leukaemia, high dose steroid therapy) and solid organ transplant recipients.[6,104,105] To date, chronic HEV infection has not been documented with genotype 1 or genotype 2 HEV infection. Chronic infection with genotype 4 HEV was reported in 2015, and there has only been one case reported in the literature to date.[106]

Chronic HEV infection (genotype 3) is generally defined as persisting serum HEV RNA and elevated liver enzymes 6 months after the acute phase,[6] although some units use a 3 month cut-off to define chronicity.[107] In Toulouse, France between January 2004 and December 2009, 50 cases of HEV infection in solid organ transplant patients were identified: 32 kidney transplant recipients, three kidney-pancreas recipients and 15 liver transplant recipients. A total of 45.7% of the kidney recipients in this cohort developed chronic HEV infection.[108] In a retrospective study analysing stored plasma from 2919 HIV-infected patients, three female patients were identified to have had HEV infection: two patients had acute HEV infection and one patient had chronic HEV infection for >4 years (all infections genotype 3a).[109] In addition to chronic HEV infection, reactivation of resolved infection has also been reported. One patient who had undergone stem cell transplant for leukaemia and was therefore immunosuppressed. Versluis described a patient who had initially cleared HEV infection, evidenced by 53 days with an undetectable HEV RNA.[110] At the time of allogeneic stem cell transplantation, HEV RNA was detectable, although viral load was low. Viral reactivation post-stem cell transplant was based upon rising HEV RNA levels and identical HEV-ORF1b sequences. This patient later cleared infection with a reduction in immunosuppression and ribavirin therapy.

In organ transplant recipients, chronic HEV genotype 3 infection may lead to cirrhosis and liver failure within 1–2 years.[7] This may in turn lead to a requirement for re-transplantation. In transplant patients, anti-HEV IgM and IgG may be negative and therefore RNA testing by polymerase chain reaction (PCR) must be employed. Pas demonstrated that anti-HEV IgM could only be detected in 7/16 immunocompromised patients compared with 18/18 immunocompetent patients in the acute phase of infection, suggesting a delayed immune response and abnormal IgM antibody kinetics in the immunocompromised group.[111]

A UK/French study also looked at the role of HEV infection in patients with decompensated chronic liver disease.[66] Acute HEV infection (genotype 3) was identified in a minority (3.2%) of patients with decompensated chronic liver disease, and there were no differences in mortality between patients with and without HEV infection. It is likely that HEV genotype 3 infection in patients with chronic liver disease confers an adverse prognosis (similar to that of other insults causing decompensation), but this effect is less than that seen with HEV genotype 1 infection as seen in a cohort of Indian patients.[59] The role of the genotype of HEV in acute-on-chronic liver failure/decompensated chronic liver disease and its effect on patient outcome is therefore less certain, and requires further investigation.