VIROLOGY

HEV is a small RNA virus, 32 to 34 nm in diameter, nonenveloped, and icosahedral. HEV RNA is approximately 7.2 kilobases in length, single- and positive-stranded, 5′-capped, and polyadenylated. The crystal structure of HEV-like particles that consist of capsid protein has been found to exhibit the protruding domain that is involved in the binding of HEV to susceptible cells and that contains some neutralization epitopes.⁷ Other investigators have found that each capsid protein contains 3 linear domains that form distinct structural elements: the continuous capsid (S), 3-fold protrusions (P1), and 2-fold spikes (P2) and have speculated that binding of the susceptible cell receptor to P1 may result in conformational changes that eventually lead to cell membrane penetration and genome release into the infected cell.⁸

The virus is currently classified in a separate genus named Hepevirus in the family Hepeviridae.⁹ The HEV genome contains three open reading frames (ORFs) (Fig. 80-1).⁶ ORF1 encodes nonstructural proteins, ORF2 encodes the viral capsid protein, and ORF3 encodes a protein of unknown function. Details of HEV replication in hepatocytes and its release from infected cells remain unknown; the application of the replicon system to HEV research may generate a better understanding of the intracellular events during virus replication.¹⁰ Phylogenetic analysis of HEV isolates shows the existence of four geographically distinct genotypes, termed genotypes 1 to 4 (Table 80-1).¹¹ Genotype 1 includes various isolates from Asia that have a nucleotide sequence homology of 92% to 99% (amino acid sequence homology 95% to 99%) with each other. These isolates have nucleotide homology of 75% (amino acid homology 86%) with genotype 2 isolates, which include a strain from Mexico and some isolates from western Africa. Isolates from the United States, classified as genotype 3, are 92% identical to each other but only 73.5% to 74.5% identical to genotypes 1 and 2.¹² Genotype 3 isolates have also been reported from human cases in several European countries (including the United Kingdom, France, the Netherlands, and Spain), Japan, and South America. Genotype 4 consists of isolates from some parts of China, Taiwan, Japan, and Vietnam. All HEV genotypes belong to a single serotype because they share at least one major serologically cross-reactive epitope.

Figure 80-1. The genome of hepatitis E virus. The three open reading frames—ORF1, ORF2, ORF3—are shown.

Table 80-1 Hepatitis E Virus Genotypes and Their Geographic Distribution

Adapted from Schlauder GG, Frider B, Sookoian S, et al. Identification of 2 novel isolates of hepatitis E virus in Argentina. J Infect Dis 2000; 182:294-7.

Swine HEV was identified in pigs in the midwestern United States¹³; the virus naturally infects pigs and induces transient viremia and antibodies that react with human HEV strains. A comparison of swine HEV with the Burmese and Mexican isolates of human HEV showed approximately 90% to 92% and 79% to 83% identity at the amino acid level in the ORF2 and ORF3 regions, respectively. HEV-like genomic sequences have been isolated from pigs in several parts of the world and occasionally from other animals, including deer and wild boar. In geographical regions where genotype 3 and 4 HEV has been reported among humans, swine isolates of HEV have also belonged to these genotypes. Also, swine HEV strains in Taiwan and Spain have greater genetic similarity to human HEV strains from these regions than to swine and human HEV isolates from other parts of the world, suggesting the possibility of zoonotic transmission. By contrast, swine HEV identified in India is genetically different from the HEV in patients from the same geographic region.^14,15 An HEV-like virus has been recovered from chickens and its genome partially sequenced; it appears to be noninfectious to mammals.¹⁶

EPIDEMIOLOGY

Several epidemics of hepatitis E affecting several hundred to several thousand persons have occurred on the Indian subcontinent and in southeast and central Asia, where this infection is endemic.^1,2,17–19 Outbreaks of hepatitis E also have been reported from the northern and western parts of Africa and the Middle East. Two small outbreaks were reported in Mexico in 1986 and 1987 (Fig. 80-2). Overall attack rates range from 1% to 15% and are higher for adults (3% to 30%) than for children (0.2% to 10%). The male-to-female ratio among cases has ranged from 1:1 to 4:1, but the outbreaks have been characterized by particularly high attack and mortality rates among pregnant women. The epidemics vary in nature, ranging from single-peaked, short-lived outbreaks to prolonged, multipeaked epidemics lasting more than a year (Table 80-2). In endemic areas, hepatitis E accounts for up to 50% to 70% of cases of sporadic acute hepatitis; these cases are demographically and clinically similar to those observed during disease outbreaks.

Figure 80-2. Geographic distribution of areas endemic for hepatitis E virus (red areas).

Table 80-2 Epidemiologic Features of Hepatitis E

In nonendemic regions, hepatitis E is related mostly to travel to HEV-endemic regions and accounts for fewer than 1% of cases of acute viral hepatitis. Solitary cases or small series of cases with indigenously-acquired cases of acute hepatitis E have been reported from several countries in Europe, North America, and Australia. A report from the United Kingdom described 40 patients with hepatitis E and no history of travel to an area endemic for this disease. HEV infection was thought to be acquired locally, and the patients were described as a group of “autochthonous” hepatitis E cases.²⁰ Although most patients found to have HEV infection in this series had jaundice, a few had anicteric illness with nonspecific symptoms or asymptomatic aminotransferase elevations. The number of cases peaked in spring and summer, and disease appeared to be more common in residents of coastal and estuarine areas. The disease showed seasonal variations with peaks in spring and summer and no cases occurring during November and December. Case series with similar characteristics have been described from southwest France and the Netherlands.^21,22 Table 80-3 highlights differences in the epidemiologic and clinical features of HEV genotypes 1 and 3.

Table 80-3 Comparison of Epidemiologic and Clinical Features Associated with HEV Genotypes 1 and 3

PATHOGENESIS

Understanding of virologic, serologic, and pathologic events in the pathogenesis of HEV infection (Fig. 80-3) is based on data from human patients and experimentally infected primates (cynomolgus macaques, rhesus monkeys, and chimpanzees). The virus enters the host primarily through the oral route, although the mechanism(s) by which the virus reaches the liver is unknown. In human volunteers, the incubation period after oral exposure is four to five weeks, and HEV is detected in the stool approximately one week before the onset of illness and for up to two weeks thereafter. In clinical cases, the fecal shedding of the virus is observed until about four weeks after the onset of illness. HEV RNA can be detected in serum for two weeks after the onset of illness in virtually all patients. In experimentally infected primates, HEV RNA has been found in serum, bile, and feces a few days before elevation of the serum alanine aminotransferase (ALT) level. HEV antigen (HEVAg) is expressed in hepatocytes as early as seven days after infection and may be identified in more than 50% of the cells, but the number of HEVAg-positive cells decreases sharply when serum ALT levels increase markedly.³⁶ The onset of ALT elevation in the serum and of histopathologic changes in the liver generally corresponds with the appearance of anti-HEV in serum.

Figure 80-3. The time course of events during hepatitis E virus (HEV) infection (based on studies in human subjects and in experimentally infected primates). ALT, serum alanine aminotransferase level; anti-HEV, antibody to HEV; HEVAg, HEV antigen; IgG, immunoglobulin G; IgM, immunoglobulin M.

Patients with acute hepatitis E show cellular immune responses to HEV proteins^37–39 but the number of cytokine-producing HEV-reactive CD8⁺ cells is not increased.⁴⁰ Helper T cell type 1/type 2 (Th1/Th2) cytokine balance is biased toward Th2 in pregnant women,⁴¹ in whom hepatitis E is particularly severe. Although the temporal concordance of the HEV-specific immune response with the onset of liver pathology suggests that liver injury is immune mediated, further studies are needed to define the role of the immune response in the induction of liver injury; no evidence for cytopathic properties of HEV itself exists.

Histopathologic features of hepatitis E are similar to those of other forms of acute hepatitis and include ballooned hepatocytes, acidophilic bodies, focal parenchymal necrosis, and inflammatory infiltrates in the lobules and enlarged portal tracts. Nearly one half of patients with hepatitis E have cholestatic hepatitis, characterized by canalicular bile stasis and gland-like transformation of parenchymal cells, with less marked hepatocytic degeneration and necrosis.⁴² Submassive or massive necrosis and collapse of liver parenchyma are seen in patients with severe liver injury. Chronic HEV viremia with genotype 3 virus has been reported in some renal and liver transplant recipients in Europe, with elevated aminotransferase levels and no evidence of infection with other hepatotropic viruses.⁴³ Viremia and liver enzyme elevation appeared nearly at the same time, and liver biopsies showed evidence of liver injury. A case report has suggested that such infection can lead to chronic hepatitis and cirrhosis.⁴⁴

CLINICAL FEATURES

The most common recognizable form of HEV infection is acute icteric hepatitis. The clinical manifestations (Table 80-4) are similar to those observed in patients with acute hepatitis A or B.⁴⁵ The illness usually is insidious in onset. An initial prodromal phase lasts one to four days and is characterized by a varying combination of flu-like symptoms, fever, mild chills, abdominal pain, anorexia, nausea, aversion to smoking, vomiting, clay-colored stools, dark urine, diarrhea, arthralgias, asthenia, and a transient macular skin rash. Most prodromal symptoms tend to diminish with the onset of jaundice (icteric phase), but dark urine, light stool color, and itching persist for a varying amount of time. Physical examination reveals jaundice and a mildly enlarged, soft, and slightly tender liver. Some patients have splenomegaly. Laboratory test abnormalities include bilirubinuria, conjugated hyperbilirubinemia, and marked elevations in serum aminotransferase and gamma glutamyl transpeptidase levels. Elevation of the serum ALT level may precede symptoms, and the magnitude of elevation does not correlate with the severity of liver injury. Mild leukopenia and relative lymphocytosis may occur. As the illness subsides, serum aminotransferase levels return to normal, followed by a decline in the serum bilirubin, which usually returns to normal levels by six weeks. Ultrasonography may demonstrate normal findings or show a mildly enlarged liver, increase in hepatic parenchymal echogenicity, gallbladder wall edema, prominence of portal venules, and a slightly enlarged spleen. The main use of ultrasonography is to exclude biliary obstruction as the cause of jaundice.

Table 80-4 Clinical Features of Hepatitis E

Acute hepatitis E usually is self-limited. A few patients have a prolonged course with marked cholestasis (cholestatic hepatitis), including persistent jaundice lasting 2 to 6 months, prominent itching, and marked elevation of the serum alkaline phosphatase level, ultimately with spontaneous resolution. Reported case-fatality rates have ranged from 0.5% to 4%; however, these rates seem to be overestimated because they are based on data from hospitals. Population surveys during outbreaks have reported lower mortality rates of 0.07% to 0.6%.¹⁷

Some HEV-infected persons may exhibit only nonspecific symptoms resembling those of an acute viral febrile illness with serum aminotransferase elevations but without jaundice (anicteric hepatitis), and some may remain entirely asymptomatic. Anicteric and asymptomatic infections may occur more frequently than icteric disease, because a large proportion of HEV-seropositive persons in endemic areas do not recall ever having had jaundice. The lower attack rates among children than adults during HEV outbreaks may represent a higher frequency of anicteric or subclinical HEV infections in the younger age group. In a small proportion of patients, the disease is severe and associated with subacute or fulminant hepatic failure.

Pregnant women, particularly those in the second or third trimester, are affected more frequently during hepatitis E outbreaks than are others in the population and have a worse outcome, with mortality rates of 5% to 25%. In an epidemic in Kashmir, India, clinical hepatitis E developed in 17.3% of pregnant women, compared with 2.1% of nonpregnant women and 2.8% of men, all ages 15 to 45 years.⁴⁶ Among the pregnant women, attack rates during the first, second, and third trimesters were 8.8%, 19.4%, and 18.6%, respectively. Fulminant hepatic failure developed in approximately 22% of the affected pregnant women, with an increased frequency of abortions, stillbirths, and neonatal deaths. The pathogenic elements that lead to severe liver damage during pregnancy, especially during the last trimester, remain unknown.

DIAGNOSIS

The diagnosis of human HEV infection is based on detection of either HEV RNA in stool and serum specimens with use of a reverse transcription–polymerase chain reaction assay or demonstration of a virus-specific host immune response (see Fig. 80-3). Tests to detect HEV RNA are available only in research laboratories. Enzyme immunoassays (EIAs) for the detection of immunoglobulin M (IgM) and IgG antibodies to HEV have been developed using recombinant HEV antigens expressed in Escherichia coli or insect cells, synthetic peptides corresponding to immunogenic epitopes of HEV, and protein expressed from a synthetic gene encoding multiple linear antigenic epitopes from the ORF2 and ORF3 regions.^47–49 The presence in serum of IgM anti-HEV indicates acute infection, whereas detection of IgG anti-HEV may indicate the convalescent phase or past infection. IgM anti-HEV appears in the early phase of clinical illness, lasts 4 to 5 months, and can be detected in 80% to 100% of cases during outbreaks of acute hepatitis E. IgG anti-HEV appears in serum a few days after IgM anti-HEV, and the titer of IgG anti-HEV increases during the convalescent phase and remains high for at least one to four and one half years; the exact duration of persistence of IgG anti-HEV in serum is not known. Although several commercial kits for the detection of IgM and IgG anti-HEV are available in various countries, none is currently licensed for clinical use in the United States. Use of different target antigens from different HEV strains and of different expression systems to produce recombinant proteins makes comparison of different tests problematic; the sensitivity of various assays has ranged from 17% to 100% when such assays are applied to a panel of coded specimens from nonendemic areas.⁵⁰

TREATMENT AND PREVENTION

Acute hepatitis E usually is self-limited and requires only supportive care and no specific intervention. Isolation of infected persons is not indicated, because person-to-person transmission is uncommon. Patients with fulminant hepatitis need measures to control cerebral edema and consideration of liver transplantation (see Chapter 93). In pregnant women, no proven benefit to terminating the pregnancy has been documented; postpartum hemorrhage resulting from deranged coagulation requires treatment with fresh-frozen plasma.

Prevention of hepatitis E in endemic areas depends primarily on the supply of clean drinking water and strict attention to sewage disposal. In an epidemic setting, measures to improve the quality of water—as simple as boiling water—have been shown to lead to rapid abatement in the number of new cases. The occurrence of large epidemics of hepatitis E among adults in endemic areas suggests that anti-HEV is not fully protective or that antibody levels in serum decline with time, gradually reaching a nonprotective level. In the few studies that have evaluated the role of immune globulin manufactured in endemic areas for pre- or postexposure prophylaxis, no significant reduction in disease rates with use of this agent was found.

Cloning of the HEV genome and subsequent availability of recombinant proteins have led to the development of candidate vaccines. Immunization with recombinant HEV capsid protein in HEV-susceptible primates has shown protection against hepatitis and viremia, although viral excretion is not prevented.⁵¹ An experimental HEV DNA vaccine has also been shown to induce serum anti-HEV that is protective against challenge with a heterologous HEV strain in cynomolgus macaques.⁵²

A vaccine using a recombinant truncated form of HEV capsid protein, produced in Spodoptera frugiperda cells infected with a recombinant baculovirus and adsorbed on aluminium hydroxide, has undergone a phase 2, double-blind, randomized, placebo-controlled safety and efficacy trial among human volunteers in Nepal.⁵³ In this trial, nearly 2000 young adults, almost all (>99%) male, were randomly assigned to receive three doses of either the recombinant HEV protein or a matched placebo (at 0, 1, and 6 months). The study subjects underwent active surveillance for clinical acute hepatitis E for a period extending over two years. The vaccine showed 95.5% protective efficacy against hepatitis E disease among those who received three doses and 88.5% efficacy in an intention-to-treat analysis among those who had received only the first dose. A high IgG anti-HEV level was observed in all the vaccinated volunteers at one month after the third vaccine dose, but only 56% of them had this high an antibody titer by the end of the study. More studies are needed to determine the efficacy of the vaccine among pregnant women, the duration of its protective efficacy, its effect on subclinical HEV infection and fecal shedding of the virus, and thus its impact on transmission of the virus. The vaccine may be useful for travelers to HEV-endemic regions and for pregnant women and persons with chronic liver disease who reside in such regions. Finally, cost considerations will determine whether this new and promising vaccine can be used in populations living in developing countries where hepatitis E is endemic and vaccination is likely to be the most beneficial.