Hepatitis E: Epidemiology and Natural History

Abstract

Hepatitis E is a disease caused by infection with hepatitis E virus (HEV). The virus has four genotypes, named 1 to 4, with one shared serotype. Genotypes 1 and 2 infect only humans, whereas genotypes 3 and 4 primarily infect several mammalian animals, with occasional transmission to humans. Evidence of infection with HEV has been found in most parts of the world, with two distinct epidemiological patterns. In areas with high disease endemicity, primarily developing countries in Asia and Africa, the disease occurs as outbreaks and as sporadic cases of acute hepatitis, and is caused exclusively by infection with genotypes 1 or 2 HEV, which is acquired through fecal-oral route, usually through contamination of water supplies. The disease in these areas occurs most commonly in young adults, and is particularly severe in pregnant women and persons with pre-existing chronic liver disease; chronic infection has not been reported. In areas with lower endemicity, which are mainly developed areas with robust water supply and sanitation systems, occasional sporadic cases of locally-acquired genotype 3 or 4 HEV infection are observed. The affected persons are often elderly and have other coexisting illnesses. The reservoir of infection in these areas is believed to be in animals, such as pigs, wild boar and deer, with zoonotic transmission to humans, possibly through consumption of undercooked meat. Also, in these areas, persistent HEV infection has been well documented among immunosuppressed persons such as organ transplant recipients, and is believed to lead to chronic liver injury, including liver cirrhosis. Further work is needed to better understand the biological basis underlying these widely-differing epidemiological patterns.

‘Hepatitis E’ is the term used to describe liver disease resulting from infection with ‘hepatitis E virus’ (HEV). This virus consists of non-enveloped icosahedral virions measuring 27–34 nm in diameter, which contain a single-stranded RNA molecule with positive polarity surrounded by a viral capsid.^1,2 Its RNA genome has short untranslated regions at either end and three open reading frames that encode for (i) a polyprotein predicted to contain regions with enzymatic activities required for viral replication (ORF1), (ii) the viral capsid protein (ORF2), and (iii) a small phosphoprotein (ORF3) whose exact function remains unclear though several roles have been proposed.^1,2

Hepatitis E as a distinct entity was first suspected nearly 30 years ago, when a waterborne outbreak of acute hepatitis in the Kashmir valley in India was found to be related neither to hepatitis A virus nor to hepatitis B virus.³ Around the same time, sera from previous outbreaks of acute hepatitis in India, including a large outbreak in New Delhi that occurred during the year 1956,⁴ were also found to lack markers of acute hepatitis A or B.⁵ The confirmation of these observations came in 1983 through demonstration at immune electron microscopy of spherical virus-like particles in stool specimens collected from a human volunteer who had ingested a pooled fecal suspension from patients with acute hepatitis in Afghanistan.⁶

For nearly 15 years after its discovery, HEV was believed to infect only humans, and to circulate only in areas with weak water supply and sanitation systems that facilitate fecal-oral transmission. This seemed to fit well with the observations that hepatitis E was endemic in several developing countries, and that nearly all the cases with HEV infection in the developed countries were related to travel to disease-endemic countries.^7,8 Further, it was thought that the infection was short-lasting, manifesting mostly as self-limited viral hepatitis with occasional severe cases that presented with fulminant hepatic failure (FHF).

However, during the late 1990s, a few cases with HEV infection were reported among residents of developed countries who had not traveled to countries where hepatitis E was endemic.^9–11 Around the same time, HEV-like genomic sequences were identified in specimens from animals, in particular pigs.¹² Further work prompted by these observations has led to a major change in our understanding about HEV and its epidemiology, including the routes of transmission and clinical outcomes. Thus, we now know that HEV has a fair degree of genomic heterogeneity, has a wide host range with frequent cross-species transmission,¹³ circulates in most parts of the world, and causes persistent infection in some persons.¹⁴

Despite the fact that HEV was discovered only in late 20th century, its origin appears to be fairly ancient. Even in geographical areas where currently hepatitis E is infrequent, epidemics of acute hepatitis with features resembling those of HEV infection were common place during the 18th and 19th centuries.¹⁵ Further, a molecular-clock analysis of the currently-available HEV genomic sequences suggests origin from a common ancestor that existed between 500 and 1300 years ago.¹⁶

Virus and its Genotypes, etc

A detailed review of the structure, genomic organization and replication strategy of HEV are available in another piece in this issue of the Journal.¹⁷

HEV is classified in genus Hepevirus and family Hepeviridae.¹⁸ It consists of a group of closely-related viruses that, besides humans, infect other mammals such as pigs, rabbits, rats, deer, and mongoose.¹⁹ Based on analysis of viral genomic sequences, HEV isolates from various mammalian hosts cluster into at least four separate groups, known as genotypes 1 to 4, with several subgenotypes in each.^20,21 The genotypes differ in host specificity and geographical distribution (Figure 1). The genotype 1 and 2 strains of HEV infect only humans, and are restricted to areas where HEV disease is very common. In contrast, genotype 3 and 4 strains have a wider geographical spread, and a broader host range, causing frequent infection in several mammalian species with occasional transmission to humans, causing clinical disease.

Figure 1.

Distribution of HEV genotypes in viral isolates obtained from humans and animals (mainly pigs). The colors used for a country and the circle associated with it represent the predominant HEV genotypes of human and animal isolates, respectively, from that country. The figure is based on data from Okamoto, 2007.²⁰

Reproduced from: Aggarwal R, Jameel S. Hepatology 2011; 54: 2218–2226.

HEV-like genomic material has also been identified in birds, including chickens and turkeys.²² These isolates of HEV are genetically more divergent than the mammalian HEV isolates, appear incapable of transmission to mammals, and are placed in a separate genus named as ‘avian HEV’.¹⁸ More recently, genomic sequences resembling HEV have also been reported in a virus cultured from cutthroat trout, a species of freshwater fish.²³

Despite their fairly extensive genetic heterogeneity, the various HEV genotypes that infect mammals show serological cross-reactivity, with existence of a single serotype. This implies that one vaccine should offer protection against all the HEV genotypes that infect humans.

Geographical Distribution and Epidemiological Patterns

Serological data suggest that HEV infection occurs in most parts of the world.²⁴ However, two distinct epidemiological and clinical patterns are observed, with differences in the frequency of disease, virus genotypes causing human infection, routes of transmission, affected population groups and disease characteristics (Table 1).^25,26 Each of these patterns is described below.

Table 1.

Comparison of epidemiological features of hepatitis E in high-endemicity and low-endemicity areas.

Characteristics	Areas with high-endemicity rate	Areas with low-endemicity rate
Human disease	Large outbreaks, and a large proportion of cases with sporadic acute hepatitis	A small number of cases with sporadic acute hepatitis
Characteristics of diseased persons	Young, otherwise healthy, males > females	Mostly elderly, often with other comorbid conditions, males > females
Prevalent causative viral genotypes	Mostly genotypes 1 and 2, a few cases of genotype 4 in China	Genotype 3, occasionally genotype 4
Reservoir of infection	Human	Most likely animals (pigs, wild boars, deer)
Route of transmission	Fecal-oral, primarily contamination of water supplies	Consumption of undercooked meat (pig or deer), close contact with animals
Relationship with pregnancy	High disease attack rate and risk of fulminant disease among pregnant women	No reports of severe disease among pregnant women
Chronic infection	No reports of chronic hepatitis E	Chronic infection with hepatitis E virus reported among immunosuppressed persons, primarily organ transplant recipients receiving immunosuppressive drugs

Adapted from: Aggarwal R, Jameel S. Hepatitis E. Hepatology 2011; 54: 2218–2226.

Epidemiology in Areas with High-endemicity

Geographical Distribution

This epidemiologic pattern is observed primarily in developing countries in the Indian subcontinent, Southeast and Central Asia, the Middle East, and northern and western parts of Africa (Figure 2).^25,26 In these areas, hepatitis E occurs as distinct outbreaks of acute hepatitis, as well as frequent sporadic cases. Two small outbreaks were also reported in Mexico during 1986–87²⁷; however, no further outbreaks have since been reported from North America. Large outbreaks of hepatitis E occurred frequently in China during 1980s and early 1990s, but none have been reported in the recent years, suggesting a transition from the high-endemicity pattern to the low-endemicity pattern.

Figure 2.

Map showing areas with high-endemicity for hepatitis E.

Reproduced from: Aggarwal R, Naik S. J Gastroenterol Hepatol 2009; 24: 1484–93.

Outbreaks of Hepatitis E

The outbreaks of hepatitis E vary in size, and can be large, affecting several hundred to several thousand cases each.^3,4,28,29 Most of the outbreaks in endemic areas can be traced to contamination of drinking water supplies with human fecal matter. These outbreaks often follow heavy rainfall and floods, which facilitate mixing of human excreta with sources of drinking water,^3,4 or occur during dry summer months when water flow in rivers and streams is reduced, with consequent increase in concentration of fecal contaminants.^28,30 In Southeast Asia, disposal of human excreta into rivers is common, and water from the same rivers is used for drinking, cooking and personal hygiene at downstream locations; this may account for the frequent recurrence of outbreaks.³⁰

Another mode of contamination of water supplies in developing countries is at the level of peripheral water distribution system.³¹ This happens when old, leaky water pipes pass through soil that is contaminated with sewage, and water supply is intermittent. The negative pressure produced in pipes during periods of no flow allows inward suction of surrounding contaminants.

In recent years, several outbreaks of hepatitis E have been reported from areas with conflict, violence and major human displacement.^32,33 In these outbreaks too, limited availability of facilities for safe drinking water and proper disposal of human feces, for instance in refugee camps, have been the main factors responsible for spread of HEV infection.

In addition to water, contaminated food may be expected to serve as a vehicle for transmission of HEV in areas where the disease is highly-endemic. However, few food-borne outbreaks have been reported. This may be related to the inherent difficulty of relating an outbreak to a particular meal or food item, because of a relatively long and variable incubation period (2–10 weeks) of hepatitis E.

The outbreaks of hepatitis E vary widely in time-course. Several outbreaks, particularly those that follow a short and well-defined period of water contamination, have been unimodal lasting only a few weeks, indicating absence of secondary (person-to-person) spread.^3,4 However, some other outbreaks have been multi-peaked and have lasted for over a year; in these, the secondary peaks have generally been related to continued water contamination, rather than to person-to-person transmission.²⁸ Some workers have recently suggested existence of secondary spread³⁴; however, these data have been questioned.³⁵ During the outbreaks, overall attack rates have varied widely from 1% to above 15%; this is most likely related to the intensity of water contamination.

Relationship of Hepatitis E with Host Characteristics, Including Pregnancy

During waterborne outbreaks of hepatitis E, where exposure to the virus may be ubiquitous, disease attack rates are the highest among adolescents and young adults, and are lower among children.³⁶ This observation is probably explained at least partly by a higher proportion of asymptomatic infections, than to rarity of infection, in children. Men are generally affected more often than women, possibly due to their greater risk of exposure to contaminated water; however, a greater propensity for clinical disease among those infected cannot be ruled out.

A typical feature of hepatitis E outbreaks is the occurrence of a higher disease attack rate and a higher mortality rate among pregnant women. During a large epidemic in Kashmir, India, the disease attack rates were 8.8%, 19.4%, and 18.6% among pregnant women in the first, second, and third trimesters, respectively; these rates were significantly higher than those among non-pregnant women (2.1%) and men (2.8%).³⁷ In addition, the risk of development of FHF among those with symptomatic hepatitis E was higher among pregnant women (22%) than among men (2.8%) and non-pregnant women (0%).

Once FHF appears, the mortality rate appears to be similar among pregnant women with hepatitis E and pregnant women with other causes of severe liver injury.³⁸ However, since HEV infection during pregnancy is associated with high rates of symptomatic disease and FHF, it accounts for a large proportion of cases with liver failure among pregnant women in the endemic areas.³⁶ The reason underlying the association of hepatitis E and pregnancy is unknown, though immunological or hormonal factors have been proposed.^39–42

Hepatitis E during pregnancy is also associated with prematurity and low birth weight.⁴³ The children born to such mothers frequently suffer from hypoglycemia and jaundice, and have an increased perinatal mortality.

Sporadic Hepatitis E in High-endemicity Areas

In areas where hepatitis E outbreaks occur, a large proportion of cases with acute sporadic hepatitis are related to HEV infection, irrespective of the age group. For instance, in some studies from the Indian subcontinent, HEV infection accounted for up to 70% of adult cases with sporadic hepatitis.^44–46 The patients with sporadic hepatitis E in these regions share several epidemiological and clinical characteristics with those observed in patients with epidemic hepatitis E. These include predominant affliction of adolescents and young adults, the association between pregnancy and severe disease, and clinical presentation as acute hepatitis, with occasional cases of FHF.⁴⁷

Reservoir of Infection and Routes of Transmission

In areas where hepatitis E is highly-endemic, only genotype 1 or 2 HEV have been identified from human cases. Since such strains circulate only in humans and have not been identified in animals, the reservoir of infection in these areas appears to reside in humans with HEV infection, either symptomatic or asymptomatic. Genotype 3 and 4 HEV, which circulate in animals, have not been identified in human cases from these areas, providing evidence against zoonotic transmission from an animal reservoir.

As indicated above, most outbreaks of hepatitis E have been traced to contamination of drinking water supplies. In contrast, the route of transmission of HEV infection in sporadic cases in highly-endemic areas is less clear, though fecal contamination of water or food appear to be responsible in most cases. Some of the sporadic cases may in fact represent small common point-source outbreak among persons who had come together for a short period of time.

Person-to-person transmission of HEV appears to be relatively infrequent, in both epidemic and sporadic settings.^48,49 During outbreaks of hepatitis E, secondary attack rates among household contacts of patients with hepatitis E cases have been much lower than those observed among susceptible household contacts of patients with hepatitis A, another enterically-transmitted form of hepatitis. Even when multiple cases occur in a family, the time interval between these is shorter than the minimum incubation period, indicating exposure to a common waterborne primary source rather than person-to-person spread⁴⁸ though a recent report has questioned this finding.³⁴ In the sporadic setting too, secondary spread of HEV appears to be infrequent.⁴⁹

Transmission of hepatitis E through materno-fetal transfer and transfusion of blood and blood products is known. A proportion of healthy blood donors in endemic regions have detectable HEV viremia and transfusion of their blood has been shown to lead to HEV infection among recipients.⁵⁰ Similarly, a proportion of babies born to mothers with acute hepatitis E during the third trimester of pregnancy have showed evidence of HEV infection in the form of detectable HEV RNA or IgM anti-HEV antibodies in cord blood.⁵¹ However, the contribution of these routes to the overall burden of hepatitis E appears to be small.

Seroprevalence Data

IgG anti-HEV antibodies have been shown to appear following infection with HEV infection and appear to persist for several years.⁵² Thus, their detection in a person's blood is believed to indicate prior exposure to HEV, and their prevalence may provide an epidemiological marker of frequency of exposure to HEV in a population. In the high-endemicity countries, anti-HEV antibody prevalence rates are higher than those in the low-endemicity countries but are much lower than that expected from the frequent occurrence of outbreaks which indicates frequent opportunities for water contamination. Also, the differences in antibody prevalence rates between countries at times do not fit in well with other available data. For instance, the seroprevalence rates in India⁵³ are lower than those in Egypt,⁵⁴ even though outbreaks of hepatitis E are common in the former and are virtually unknown in the latter. The reason for this is unclear.

Data on duration of persistence of anti-HEV antibodies are limited. In a study from India, nearly half of those who had been affected during a hepatitis E outbreak had detectable anti-HEV antibodies 14 years later, and the other half tested negative.⁵⁵ However, in another study, IgG anti-HEV titers showed a rapid decline.⁵⁶

Data on efficacy of subunit vaccines to induce protection against hepatitis E indicates that anti-HEV antibodies do provide protection against HEV infection. However, information on protective titers is not available.

Epidemiology in Low-endemicity Regions

This epidemiological pattern of HEV-related disease is observed in western Europe, United States, Canada, Australia and New Zealand and several developed countries in Asia, such as Japan, Taiwan, Hong Kong, etc.^25,26 In these areas, large outbreaks of hepatitis E are unheard of and only a few limited food-borne outbreaks have been reported.^57,58 In these areas, sporadic disease due to HEV infection is also rare, forming only a minority of all cases with acute viral hepatitis.

Characteristics of Affected Subjects

Most of the data from these areas are in the form of small case series.^59–62 In contrast to disease seen in highly-endemic areas, the patients with hepatitis E in low-endemicity areas are generally elderly, and often have other coexisting diseases, including diabetes, cardiovascular disease and prior liver disease.⁶³ In a series from the UK, the disease showed seasonal peaks in spring and summer and no cases during winter.⁶⁴ In another series, the disease appeared to be more common among residents of coastal and estuarine areas⁶⁵; the exact reason underlying this relationship is unclear.

Viral Genotype and Role of Zoonotic Transmission

Genomic sequencing of viral isolates from human cases in these areas has revealed exclusive circulation of genotype 3 and 4 HEV. Genotype 1 and 2 HEV have not been identified, except in persons who had recently traveled to high-endemicity areas.

Reservoir of Infection and Mode of Transmission

The mode of transmission of infection cannot be definitively identified in most of the cases with hepatitis E in these regions, though epidemiological data suggest that zoonotic spread may be responsible.^64,65 This mode of transmission was initially suspected because of the observation that genomic sequences of HEV isolates from two autochthonous cases in the United States were more closely-related to swine HEV than to human HEV isolates.^9,11 Subsequent experimental studies showed that human genotype 3 and 4 HEV could infect experimental pigs, and that swine HEV isolates could be transmitted to primates, which served as a surrogate for humans.^11,66

More direct evidence for zoonotic transmission came from a cluster of cases in two Japanese families that had shared a meal consisting of inadequately-cooked deer meat a few weeks prior to the onset of illness.⁶⁷ The genomic sequences of HEV isolated from these cases and from the left-over frozen meat were identical implicating the latter as the source of infection. Further, the genomic sequences of HEV from the deer meat had a high (99.7%) sequence identity with HEV isolates from a wild boar and another wild deer from the same forest, indicating transmission of HEV between various animal species and from deer to humans.⁶⁸

More information in favor of zoonotic transmission of HEV to humans has come from studies on commercial meat products available for human consumption in several countries in low-endemicity regions. In one study, nearly 2% of commercial packets of pig liver sold in Japanese grocery stores were found to contain genotype 3 or 4 HEV,⁶⁹ which shared a high degree of sequence homology with HEV isolated from a previous human case. In France, pig liver sausages have been shown to contain HEV genomic material and their consumption has been epidemiologically linked to human disease.⁷⁰ In the United States too, genotype 3 HEV RNA has been identified in more than 10% of pig liver packages sold in grocery stores.⁷¹ Further, a large proportion of Japanese patients with sporadic hepatitis E admitted to have eaten uncooked or undercooked pig livers.⁷²

These data suggest that genotype 3 and 4 circulate freely among pigs and other animal species in these areas and may be transmitted to the humans through exposure to contaminated meat products or other forms of close contact.

An additional source of HEV could be the consumption of inadequately-cooked shellfish. Mussels harvested in coasts off Scotland and in the Mediterranean area have been shown to contain HEV RNA and may thus be capable of transmitting this infection.^73,74

Seroepidemiological Data

Several seroepidemiological studies from areas from developed countries have shown seroprevalence rates of anti-HEV antibodies that far exceed those expected from the infrequent occurrence of clinical disease. For instance, in sera collected during the Third National Health and Nutrition Examination Survey (NHANES III) (1988–1994) in the United States, the seroprevalence rate was found to be 21%⁷⁵; further analysis showed that male gender, non-Hispanic white ethnicity, residence in the Midwest and/or in metropolitan areas, having a pet in the home, and consuming liver or other organ meats more than once per month were associated with HEV seropositivity. Occupational groups with more frequent contact with animals, such as veterinarians, have higher seroprevalence rates than in general population.⁷⁶

Natural History and Clinical Outcomes of Hepatitis E Virus Infection

Clinical outcomes of HEV infection vary widely (Table 2). In both high-endemicity and low-endemicity areas, a large proportion of persons who have detectable anti-HEV antibodies do not recall having had jaundice or other symptoms of liver disease ever in life. This indicates that a majority of those who acquire HEV infection either have no clinical consequences or have a minor illness without discernible manifestations of liver injury. This outcome appears to be more common among children with HEV infection than in adults.

Table 2.

Clinical outcomes associated with hepatitis E virus infection.

Inapparent or asymptomatic infection

Anicteric hepatitis (biochemical abnormalities but no jaundice)

Icteric hepatitis (similar to other forms of viral hepatitis)

Severe hepatitis leading to fulminant hepatic failure

Acute on chronic liver disease

The incubation period in HEV infection varies from 2 to 10 weeks.⁴ The spectrum of clinical consequences of HEV infection is quite wide, and may depend on genotype of the infecting virus.

Clinical Outcomes in High-endemicity Areas

The most common clinical manifestation associated with HEV infection is that resembling acute hepatitis due to other hepatitis viruses, including hepatitis A or B.^36,77 The illness begins with a prodrome which is followed within a few days by an icteric phase. It is generally self-limited, with most patients improving spontaneously in a few weeks. Occasional patients develop prolonged cholestasis that manifests with persistence of jaundice and marked itching. The outcome is generally good with symptoms eventually disappearing after a few months.

Some affected persons develop severe liver injury that may present as FHF (also known as acute liver failure), which is characterized by development of hepatic encephalopathy, or subacute hepatic failure. As indicated above, this outcome is more common during pregnancy. These conditions have a poor prognosis and may be fatal, if liver transplantation is not available.

HEV infection in persons with pre-existing liver injury may present with a more severe illness, often with ‘acute-on-chronic liver failure’ syndrome.^78–80 The mortality rate among such patients is higher than that observed in patients with usual hepatitis E.⁸⁰

Clinical Outcomes in Low-endemicity Regions

The illness associated with hepatitis E in low-endemicity areas is more variable, possibly in part because (i) these patients are older, (ii) they more often have associated liver, cardiac, respiratory or malignant disease, and (ii) the diagnosis of hepatitis E in them is often based on serological tests for HEV infection undertaken because of detection of liver function test abnormalities rather than symptoms of liver disease. Besides jaundice, these patients often have symptoms such as asthenia, fever, joint and muscle pains, and abdominal pain.⁸¹ In addition, some patients have headache, nausea and vomiting, loss of appetite, loss of weight, bowel disturbances and purpuric skin rash.

In a study from the United Kingdom, complications were more common in patients with hepatitis E than in those with hepatitis A.⁸² Also, patients with hepatitis E in these areas appear to have a higher mortality rate than that observed with this disease in high-endemicity areas.⁶³ However, these findings were possibly related to the older age and more frequent presence of coexisting diseases in patients with hepatitis E.

FHF due to HEV infection in pregnant women has not been reported from these areas. Whether this is related to the relative infrequency of HEV infection among women of child-bearing age in these areas or a true difference due to biological differences between various HEV genotypes is unclear.

Some recent series have shown evidence of hepatitis E among persons living in high-income countries who were initially diagnosed as drug-induced livery injury and were later retrospectively tested for serological evidence of recent HEV infection.^83–85

A specific feature of HEV infection in low-endemicity areas is the occurrence of chronic HEV infection, defined as persistence of HEV RNA in a person's blood beyond 3 or 6 months, in immunosuppressed persons.⁸⁶ Such infection is associated with evidence of chronic liver injury, leading to fibrosis and even cirrhosis. Chronic HEV infection has not yet been reported from high-endemicity areas; thus, whether such infection occurs with genotype 1 or 2 HEV remains unclear. The subject of chronic hepatitis E has been reviewed in detail in a separate piece in this issue of the Journal (Kamar N, Izopet J, Dalton HR. Chronic hepatitis E virus Infection and treatment. J Clin Exp Hepatol 2013;3:134-140) and hence is not discussed further here.⁸⁷

Another unusual feature of HEV infection in these areas is the occurrence of a wide variety of neurological syndromes in association with genotype 3 HEV infection.^88,89 However, these syndromes are not seen in patients with acute infection due to genotype 1 or 2 HEV infection in high-endemicity areas.

Disease Burden Due to Hepatitis E

Due to wide variations in the epidemiology of hepatitis E based on geographical region and viral genotypes, it is difficult to get an accurate estimate of disease burden due to hepatitis E. An attempt has been made to model the annual disease burden of HEV genotypes 1 and 2 for nine of the 21 regions defined for the Global Burden of Diseases, Injuries, and Risk Factors Study (the GBD 2010 Study), which represent 71% of the world's population.^90,91 This revealed an estimated 20.1 million (95% credible interval: 2.8 million–37.0 million) incident HEV infections across the selected regions, with 3.4 (0.5–6.5) million symptomatic cases, 70,000 (12,400–132,732) deaths and 3000 (1892–4424) stillbirths in the selected regions. However, the model was very sensitive to estimates of age-specific incidence of HEV disease. The use of these data for policy making is also limited by the wide credible intervals of the estimates.

Prevention of Hepatitis E

Two subunit vaccines have been developed against HEV infection and have been shown to be highly protective against clinical hepatitis E in clinical trials.^92,93 These will be reviewed in a future issue of the Journal. Currently only one of these vaccines has been licensed and that too only in China. Though several additional pieces of information are needed before these vaccines can be used for prevention of hepatitis E in general population, viz. the burden of disease in various populations, efficacy and safety of these vaccines in high risk groups (e.g. pregnant women, patients with prior liver disease), efficacy of vaccines in post-exposure setting, duration of protection offered by the vaccines, need for boosters and cost-benefit ratio of vaccines, it appears that these would play a major role in the control of hepatitis E in future.

In the meanwhile, our knowledge about epidemiology of hepatitis E allows us to devise and use non-vaccine approaches for prevention of this disease. In high-endemicity areas, these primarily include provision of safe drinking water, proper disposal of human feces, and education about personal hygiene. In addition, during outbreaks, boiling and chlorination of water should be useful. In contrast, in low-endemicity areas with zoonotic transmission, sanitary handling and proper cooking of pig and deer meat may be important.

Conflicts of interest

The author has none to declare.