Hepatitis E Virus in the Terai of Nepal

HEPATITIS E VIRUS IN THE TERAI OF NEPAL:

COMMUNITY-BASED COHORT STUDY OF PREGNANT WOMEN IN SARLAHI DISTRICT

AND INVESTIGATION OF THE 2014 OUTBREAK IN BIRATNAGAR, MORANG DISTRICT

by Lisa J. Krain, ScM

A dissertation submitted to Johns Hopkins University in conformity with the requirements for the degree of Doctor of Philosophy

Baltimore, Maryland July, 2016

ABSTRACT

In South Asia, hepatitis E virus (HEV) is an important cause of morbidity and mortality and poses a special threat to pregnant women. However, the burden of both infection and illness in the

Terai of Nepal have received scant attention.

We enrolled 2363 pregnant women in a community-based cohort study in Sarlahi District and followed threm through three months postpartum. Baseline anti-HEV IgG seroprevalence was

11.7% [95% confidence interval (CI): 10.2%-13.3%] and varied with age, socioeconomic factors, and meat consumption, but not with water source or latrine access. The incidence of HEV seroconversion was 32.3/1000 person-years (p*y) during pregnancy [95% CI: 14.8 to 61.3/1000 p*y] and 44.3/1000 p*y overall [95% CI: 27.1 to 68.4/1000 p*y]. One acute hepatitis E case

(genotype 1) was confirmed, but most hepatitis-like illnesses were not attributable to hepatitis A, B,

C, or E. Jaundice was linked to 1/2 maternal deaths. Earlier and extended serologic monitoring would permit improved surveillance of pregnancy outcomes and antibody kinetics. Environmental testing could help clarify exposure pathways and seasonal patterns.

A large outbreak of waterborne hepatitis E occurred in Biratnagar municipality in 2014. We tested sera collected from healthy adults in Biratnagar and Dharan before the outbreak and from patients seen at three hospitals in Biratnagar during the outbreak. The population seroprevalence of anti-HEV IgG was low in Biratnagar (~5.6%) before the outbreak. Among 450 oubtreak patients,

42.2% were IgG+, 14.7% IgM+, and 2.9% HEV Ag+. Acute HEV markers varied geographically and by hospital. The outbreak underscores the need for access to clean water, and for improved epidemiologic surveillance and response.

Hepatitis E virus threatens public health in the Terai. Rural and urban populations in Nepal remain highly vulnerable to infection, disease, and mortality due to HEV. Improvements in water, ii

sanitation, and hygiene are needed to prevent both sporadic and epidemic hepatitis E, especially as development of the Terai increases. Targeted interventions may help alleviate the increased burden of disease borne by pregnant women, individuals with chronic medical conditions, and members of socially- and economically-disadvantaged groups.

Dissertation Advisors & Readers:

Kenrad E. Nelson, MD (advisor) Professor, Department of Epidemiology Infectious Disease Epidemiology

Alain B. Labrique, PhD (committee chair) Associate Professor, Department of International Health Global Disease Epidemiology and Control

Lisa P. Jacobson, ScD Professor, Department of Epidemiology General Epidemiology and Methodology, STATEPI

Wikrom W. Karnsakul, MD Assistant Professor, Pediatrics Pediatric Gastroenterology

Stephan Ehrhardt, MD (alternate) Associate Professor, Department of Epidemiology Clinical Trials and Evidence Synthesis

William P. Ball, PhD (alternate) Professor, Department of Geography and Environmental Engineering Water Institute

iii

ACKNOWLEDGMENTS

This dissertation reflects the dedication, work, and support of a multitude of people, and I am grateful to and for all of you.

Ken Nelson has graciously shared his time, his vast experience, and his stories with me over many years. He has been a tireless advocate and a voice of reason. It has been a privilege to have him as an advisor and as a mentor.

Alain Labrique lives the mHealth he preaches, having supervised my work from Baltimore,

Brussels, Bangladesh, and even 30,000 feet. He has taught me much about the practice and politics of scientific research and public health, and I am fortunate to have had the chance to work with him.

I am enormously grateful to David Celentano, Lisa Jacobson, Fran Burman, and Peggy

Hayeslip for their encouragement, their wise advice, and their unwavering support throughout my doctoral training.

I thank my committee for providing useful suggestions for improving and extending the work. I appreciate the faculty members who generously provided guidance on my dissertation research and work in Nepal: Chris Heaney, Joanne Katz, Jim Tielsch, Luke Mullany, and Mark

Steinhoff.

The Nepal Nutrition Intervention Project – Sarlahi (NNIPS), under the expert leadership of

“Daak Saab” Subarna Khatry, Steve LeClerq, and Tirtha Raj Shakya, was instrumental in carrying out this research and was the key to its success. I appreciate the efforts of Senior Supervisors Uma

Shankar Sah and Shishir Var Shrestha, Supervisors Punya Dahal and Shiv Raj Bhattarai, and of the rest of the Flu study staff: TLIs, ANMs, Lab Team, Birth Team, Flu Data Collectors, and Data Center personnel. I thank the administrative and support staff in Kathmandu and Hariyon for their

assistance, and the drivers for their skill and dedication. The Dahals, , provided warm hospitality and tasty food. I am grateful to the entire NNIPS family, who sent out a search party on motorcycles through the bazaar and sugarcane fields of Hariyon to rescue me from a lovely family of “kidnappers” who’d invited me to their home for tea and dinner on a day I’d forgotten my phone!

I owe a special debt of gratitude to the thousands of women and infants who participated in the study, as well as to the other Sarlahi residents who introduced me to their language and customs with patience and enthusiasm. It is my hope that this work will return some of that goodwill to my former neighbors by contributing to improved health in the villages of Sarlahi and beyond.

My research collaborators on the Biratnagar outbreak study, Birendra Prasad Gupta and

John Ticehurst, braved numerous logistical and bureaucratic delays in order to see the project realized, and it has been a pleasure to work with them.

The Bill and Melinda Gates Foundation generously funded this research. The Walter Reed /

AFRIMS Research Unit – Nepal (WARUN) in Kathmandu, under the direction of Sanjaya Shrestha, and the Armed Forces Research Institute of Medical Sciences (AFRIMS) in Bangkok, provided laboratory and logistical support. I thank the administrative staff and lab techs in the Departments of Epidemiology, International Health, and Environmental Health at Johns Hopkins, in Global Health at Cincinnati Children’s Hospital, and at NNIPS and WARUN for their assistance in moving people, specimens, and resources around the world.

Johns Hopkins University librarians have been a vital, if unsung, force in my success as a researcher. I am also grateful to Allan Massie, the UCLA IDRE & Statistical Consulting Group, and many others who freely share their technical knowledge; the world is a better place for your generosity.

I have benefited tremendously from the collegiality of my fellow students, especially those in the former Silbergeld Lab Group and at NNIPS who showed me the ropes and shared with me the trials and tribulations of graduate school. I am especially grateful to my close colleagues and friends

Amy Peterson and Neha Naithani, whose humor, enthusiasm, insights, and friendship have served as sources of comfort and inspiration.

My parents and grandparents instilled in me a love of learning and a passion for doing good in the world, encouraged my curiosity, supported my education, and helped me navigate the challenges I encountered along the way. My husband Michael has done double duty on the professional and domestic fronts while I worked on my research in Baltimore, in Nepal, and in the family room of our house. His quiet thoughtfulness, his understanding, and the example he sets in his own work constantly remind me how fortunate I am to share my life with him. I am eternally grateful for my family’s love and support.

TABLE OF CONTENTS

ABSTRACT ...... II

ACKNOWLEDGMENTS ...... IV

LIST OF TABLES ...... XIII

LIST OF FIGURES...... XIV

INTRODUCTION ...... 1

EPIDEMIOLOGIC PATTERNS OF HEPATITIS E VIRUS INFECTION...... 2

Clinical Presentation of Hepatitis E ...... 2

Human-Associated Genotypes 1 and 2...... 3

Animal-Associated Genotypes 3 and 4 ...... 4

HYPOTHESES REGARDING REGIONAL DIFFERENCES IN THE NATURE AND IMPACT OF HEV ...... 6

Ecologic Influences on HEV Epidemiology Across Regions ...... 6

Genotypic and Subgenotypic Variation and Virulence ...... 7

FIGURES ...... 9

REFERENCES ...... 11

HEPATITIS E IN PREGNANCY: MATERNAL HEALTH ...... 23

EPIDEMIOLOGY OF HEPATITIS E IN PREGNANT WOMEN ...... 23

REGIONAL VARIATIONS IN OCCURRENCE AND SEVERITY OF HEPATITIS E IN PREGNANCY ...... 24

IMMUNE FUNCTION AND SUSCEPTIBILITY TO VIRAL INFECTION DURING PREGNANCY ...... 25

HYPOTHESES REGARDING MEDIATION OF HEV SEVERITY IN PREGNANT WOMEN ...... 26

CHALLENGES FOR SURVEILLANCE AND PREVENTION ...... 29

TABLE AND FIGURE...... 31

REFERENCES ...... 33

FETAL AND NEONATAL HEALTH CONSEQUENCES OF VERTICALLY-TRANSMITTED HEPATITIS E VIRUS INFECTION...... 38

ABSTRACT ...... 38

vii

INTRODUCTION ...... 39

HEPATITIS E AND MATERNAL-CHILD HEALTH ...... 39

MISCARRIAGE AND STILLBIRTH: THE ROLE OF ANTENATAL HEV INFECTION ...... 41

HEV INFECTION AND OUTCOMES AMONG LIVE-BORN INFANTS ...... 42

CHALLENGES IN UNDERSTANDING MOTHER-TO-CHILD TRANSMISSION OF HEV ...... 44

TABLE ...... 46

REFERENCES ...... 48

STUDY BACKGROUND AND EVOLUTION...... 57

MOTIVATIONS AND FRAMEWORK ...... 57

ONE DOOR CLOSES… ...... 58

FIGURE ...... 59

REFERENCES ...... 60

HEPATITIS E VIRUS IN PREGNANT WOMEN IN SARLAHI DISTRICT, NEPAL ...... 61

ABSTRACT ...... 61

BACKGROUND ...... 62

METHODS ...... 63

Study Setting - MaGIFT ...... 63

Study Enrollment ...... 64

Routine Blood Collection (Enrollment, Delivery, and Postpartum) ...... 65

Symptom Surveillance and Follow-Up of Acute Hepatitis Cases ...... 65

Serologic Surveillance and Testing ...... 67

Ascertainment of Pregnancy Outcomes and Maternal and Infant Mortality ...... 68

Feasibility and Sample Size Considerations ...... 68

Ethical and Human Subjects Considerations ...... 70

Data Management, Quality Assurance, and Quality Control ...... 71

Data Analysis...... 72

viii

RESULTS ...... 74

Study Enrollment ...... 74

Serologic Follow-Up ...... 74

Baseline Serostatus and Risk Factors for HEV Seropositivity ...... 75

Serologic Estimation of HEV Incidence ...... 77

Maternal Morbidity and Mortality ...... 78

Pregnancy and Infant Outcomes ...... 79

DISCUSSION ...... 80

Prevalence and Incidence of HEV Infection and Illness ...... 80

Maternal Mortality...... 85

Pregnancy Outcomes ...... 86

Risk Factors for Exposure to HEV ...... 87

Conclusions ...... 92

ACKNOWLEDGMENTS ...... 94

FIGURES AND TABLES ...... 95

REFERENCES ...... 105

CASE REPORT: HEPATITIS E VIRUS (HEV) IN AN ACUTELY-ILL PREGNANT WOMAN IN NEPAL, A PRELIMINARY PHYLOGENETIC ANALYSIS, AND A CAUTIONARY TALE ...... 114

ABSTRACT ...... 114

BACKGROUND ...... 115

METHODS ...... 116

Study Overview ...... 116

Identification of Acute Hepatitis Cases ...... 116

Laboratory Analyses ...... 117

Molecular and Phylogenetic Analysis ...... 117

CASE DESCRIPTION ...... 118

FURTHER INVESTIGATION ...... 119

Exposure History ...... 119

Serology ...... 119

Molecular Analysis and Phylogenetics ...... 120

DISCUSSION ...... 120

ACKNOWLEDGMENTS ...... 123

FIGURES AND TABLE...... 125

REFERENCES ...... 130

THE 2014 BIRATNAGAR, NEPAL HEPATITIS E OUTBREAK IN CONTEXT ...... 133

ABSTRACT ...... 133

BACKGROUND ...... 134

METHODS ...... 136

Study Populations ...... 136

Sample Storage and Shipment ...... 137

Serologic Testing ...... 137

Molecular and Phylogenetic Analyses ...... 138

Data Analysis...... 139

RESULTS ...... 139

Pre-Epidemic Healthy Adults ...... 139

Biratnagar Outbreak Patients ...... 140

Molecular and Phylogenetic Characterization of HEV Isolates ...... 142

DISCUSSION ...... 143

ACKNOWLEDGMENTS ...... 145

FIGURES AND TABLES ...... 147

REFERENCES ...... 158

DISCUSSION ...... 164

SUMMARY OF ORIGINAL RESEARCH ...... 164

Hepatitis E Virus in Pregnant Women in Sarlahi District, Nepal ...... 164

Case Report: Hepatitis E Virus (HEV) in an Acutely-Ill Pregnant Woman in Nepal ...... 165

The 2014 Biratnagar, Nepal Hepatitis E Outbreak ...... 166

PUBLIC HEALTH IMPLICATIONS ...... 168

DIRECTIONS FOR FUTURE RESEARCH ...... 172

Detection of HEV in the Environment and in Animals in the Terai ...... 172

Replication of HEV Findings from Bangladesh on cytokines, micronutrients, and arsenic ...... 173

Characterization of Antibody Responses to HEV During Pregnancy and Beyond ...... 175

REFERENCES ...... 177

APPENDIX 1: FLOWCHART OF DATA AND SAMPLE COLLECTION ...... 180

APPENDIX 2: SARLAHI COHORT STUDY FORMS AND DATA COLLECTION ...... 182

List of MaGIFT Forms and Scripts Used in the Hepatitis Substudy ...... 182

Form 14: Monthly Pregnancy Follow-Up Form (PFF) ...... 187

Form 51 v.3: Antibody and Hepatitis Substudy Form (AHSSF) ...... 191

Form 51 v.4: Antibody and Hepatitis Substudy Form (AHSSF) ...... 192

Form 66: Hepatitis E Substudy Pregnant Woman’s Consent Script ...... 193

Form 69: Laboratory Specimen Collection Log (Blood & Breastmilk) (LSBL) ...... 195

Form 71: Specimen Management Log (Blood) ...... 196

Form 78: Hepatitis E Symptomatic Woman’s Consent Script ...... 197

Form 79: HEV Exposure Assessment Questionnaire (HEV-EAQ) ...... 199

Form 80: Hepatitis Case Visit Form (HEV-CVF) ...... 208

Form 81: Hepatitis E Report Card (HERC) ...... 211

APPENDIX 3: FIELD LABORATORY SAFETY & INFORMATIONAL SIGNS ...... 212

Laboratory Biohazard & Emergency Contact Information ...... 212

Caution: Liquid Nitrogen ...... 213

Lab Safety ...... 214

Wear Closed-Toe Shoes in the Lab ...... 215

Specimen Chart for Ongoing NNIPS Studies ...... 216

CURRICULUM VITAE ...... 217

xii

LIST OF TABLES

Table 1. Seroprevalence of IgG antibodies to HEV among asymptomatic pregnant women recruited

from community-based cohorts or antenatal clinics...... 31

Table 2. Vertically-transmitted hepatitis E virus infection, morbidity, and mortality in live-born

infants of mothers with laboratory-confirmed antenatal HEV infection...... 46

Table 3. Case detection algorithm for acute hepatitis-like illness based on symptoms assessed at

monthly follow-up visits...... 96

Table 4. Baseline serologic results for markers of infection with hepatitis A, B, C, and E viruses

(HAV, HBV, HCV, and HEV)...... 99

Table 5. Baseline characteristics of full cohort, all women providing baseline blood samples, and

women seropositive for anti-HEV IgG at baseline...... 100

Table 6. Crude and adjusted odds ratios for the association of selected covariates with baseline

anti-HEV IgG seropositivity...... 103

Table 7. Serologic results of viral hepatitis assays for participant PG at three time-points...... 127

Table 8. Demographic and Serologic Characteristics of Pre-Outbreak (Healthy Adults) and

Outbreak Population Sample...... 149

Table 9. Clinical and biochemical comparison of healthy adults with outbreak patients by hepatitis

E virus (HEV) serostatus...... 154

Table 10. Demographic, serologic, and molecular characteristics of outbreak specimens positive for

HEV RNA by RT-PCR...... 155

xiii

LIST OF FIGURES

Figure 1. Global distribution of hepatitis E virus (HEV) genotypes 1-4 in humans...... 9

Figure 2. Phylogenetic tree of global hepatitis E virus (HEV) isolates, based on a portion of the

nucleotide sequence encoding the capsid protein...... 10

Figure 3. Changes in hormone levels and immune function are associated with an increased risk of

severe hepatitis E in late pregnancy...... 32

Figure 4. Conceptual framework for the study of hepatitis E viral infections among pregnant

women in Nepal...... 59

Figure 5. Location of study area in Sarlahi District, Nepal...... 95

Figure 6. Enrollment of study participants relative to the total female population of each village

development committee (VDC) and ward...... 97

Figure 7. Weekly specimen collection by specimen type over the duration of the study period...... 98

Figure 8. Baseline seroprevalence of anti-HEV IgG antibodies by selected exposures...... 101

Figure 9. Geographic variation in anti-HEV IgG seroprevalence and distribution of selected

covariates within the study area...... 102

Figure 10. Patterns of baseline anti-HEV IgG seroprevalence by meat consumption habits and

frequency in pregnant women in Sarlahi District, Nepal...... 104

Figure 11. Location of the study area in Sarlahi District, Nepal...... 125

Figure 12. Timeline of PG's hepatitis symptoms and serologic results relative to her pregnancy and

study participation...... 126

Figure 13. Preliminary phylogenetic tree of hepatitis E virus (HEV) isolated from pregnant woman

in Sarlahi District, Nepal, based on 189 nucleotides at the junction of open reading frames

1, 2, and 3...... 128

xiv

Figure 14. Alignment of 291-nucleotide sequences showing uncommon variants shared by Sarlahi

and Morocco hepatitis E virus (HEV) strains at positions 48 and 200...... 129

Figure 15. Map of Nepal with inset showing locations of Biratnagar (Morang District) and Dharan

(Sunsari District) along the Koshi Highway (H8) in southeastern Nepal...... 147

Figure 16. Epidemiologic curve showing course of 2014 Biratnagar hepatitis E outbreak...... 148

Figure 17. Boxplots showing age distribution of pre-epidemic healthy adults by city and anti-HEV

IgG serostatus...... 150

Figure 18. Venn diagram of Biratnagar epidemic patients who were seropositive for at least one

serologic marker of past or acute infection with hepatitis E virus (HEV)...... 151

Figure 19. Age distribution of Biratnagar outbreak patients and age-specific seroprevalence of

hepatitis E virus (HEV) infection biomarkers...... 152

Figure 20. Map of Biratnagar, Morang District, Nepal, showing geographic distribution and

hepatitis E serostatus of 405 outbreak patients from three hospitals...... 153

Figure 21. Phylogenetic tree of 540-nt partial hepatitis E virus (HEV) open reading frame 1 (ORF1)

sequences...... 156

Figure 22. Phylogenetic tree of 191-nt partial hepatitis E virus (HEV) open reading frame 2 (ORF2)

sequences...... 157

[Material in this chapter is excerpted/adapted, with permission, from an earlier publication. Copyright © 2014, American Society for Microbiology. Krain et al., Host Immune Status and Response to Hepatitis E Virus Infection, Clin Microbiol Rev 27(1): 139-165, doi:10.1128/CMR.00062-13.]

INTRODUCTION

LISA J. KRAIN, KENRAD E. NELSON, AND ALAIN B. LABRIQUE

Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland; Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland

Epidemics of jaundice with severe consequences in pregnant women have been recognized for many decades. 1-3 However, the virus that causes hepatitis E was not isolated until the early

1980s, when a Soviet virologist intentionally ingested virus excreted by ill patients in Central Asia and carried the incubating infection back to his laboratory in Moscow for identification by immune electron microscopy. 4 Researchers in the next decade described the genome sequences of hepatitis

E virus (HEV) strains from Burma, Pakistan, and Mexico, 5-11 now identified as representing HEV genotypes 1 and 2, the strains implicated in large outbreaks in developing countries. Genotypes 3 and 4, with a zoonotic reservoir, have more recently been found to cause human infection in countries throughout the world. 12 The discovery, sequencing, and phylogenetic analysis of

Hepeviridae from an ever-expanding range of places and host species 13-18 have provided important insights into the epidemiology and geographic patterns of HEV infection and disease, but have raised new questions as well.

EPIDEMIOLOGIC PATTERNS OF HEPATITIS E VIRUS INFECTION

Clinical Presentation of Hepatitis E

Hepatitis E is generally an acute, self-limiting illness, with full resolution of symptoms occurring within weeks (usually) to months (less commonly) of onset. Presenting symptoms are often non-specific, and resemble those seen in acute hepatitis A. 21 Clinically, patients suffering from acute hepatitis E typically present with combinations of symptoms such as fever, anorexia, nausea and/or vomiting, lassitude/weakness, dark urine, light (clay/ash-colored) stool, and jaundice

(yellowing of the skin and sclera). Pruritus and/or upper right quadrant pain may also be present.

22-24

Asymptomatic and subclinical HEV infections are common in both epidemic and sporadic transmission settings, and have been documented in diverse geographic regions. 25-35 HEV infections without overt symptoms have been detected in organ donors 36 and in contacts of case patients 33 in industrialized countries.

Several case reports have noted neurological symptoms during or shortly after acute infection, such as meningitis, 37 meningoencephalitis, 38 acute transverse myelitis, 39 Guillain-Barré syndrome, 40-46 and other peripheral neuropathies. 47, 48 However, these presentations appear to be relatively infrequent.

While most cases of infection with HEV are uncomplicated and self-limiting, some individuals with hepatitis E progress to acute liver failure (ALF). ALF, also called fulminant hepatic failure if onset is within 6-8 weeks of first symptoms, is often fatal. A disproportionate number of these severe cases occur in pregnant women, though men and women with pre-existing chronic liver disease or other medical problems appear to be at increased risk as well. 49-52 Rarely, HEV infections may be prolonged or chronic, though this phenomenon has been observed primarily 2

among patients with compromised immune systems – often transplant or cancer patients receiving immunosuppressive drugs.

Human-Associated Genotypes 1 and 2

Outbreaks and sporadic cases in developing countries Genotype 1 (HEV-1) is the primary cause of epidemic and sporadic cases of hepatitis E in developing countries in Africa and Asia, where it is transmitted primarily through fecally- contaminated water supplies. [Figure 1] Frequent HEV-1 outbreaks affecting tens of thousands of people in Central, South, and East Asia have been documented since the 1950s; the largest known

HEV epidemic to date, occurring from 1986-1988 in the Xinjiang region of China, sickened over

119,000 people and resulted in 707 documented fatalities, 414 of whom were pregnant women. 3

Although molecular data from Central and South America are limited, locally-acquired HEV-1 infections and small-scale outbreaks have been reported in Latin America as well. 53, 54

Genotype 2 (HEV-2) was first identified from cases in outbreaks in two rural towns in

Morelos, Mexico, in 1986-1987. 9, 55, 56 Other HEV-2 strains have appeared in Africa, where they have also been implicated in outbreaks. 57, 58

Age-specific incidence and patterns of transmission of HEV-1 and HEV-2 Although, like hepatitis A virus (HAV), HEV-1 and HEV-2 are enteric viruses commonly spread through fecally-contaminated water, the age-specific incidence of hepatitis E differs markedly from that of hepatitis A in most of the world. While HAV causes near-universal childhood infection in endemic areas, hepatitis E cases in most of the world peak in early- to mid-adulthood, with both antibody prevalence and attack rates of clinical illness highest among adults and little overt disease in children. 59, 60 Intrafamilial transmission of HEV appears relatively uncommon, 61 though some evidence of household transmission has been reported. 62-64 Slums and refugee camps

are particularly vulnerable to outbreaks, as access to clean water and sanitary waste disposal facilities is often limited. 30, 65-68 In addition, transient or migrant populations may repeatedly replenish the pool of susceptible individuals and/or import new infections.

Animal-Associated Genotypes 3 and 4

Host range Genotypes 3 and 4 (HEV-3 and HEV-4) have been found in other mammalian hosts, most notably wild and domestic swine, which serve as reservoirs for human infection. 69, 70 [Figure 2]

HEV-3 is prevalent among swine worldwide 71-86 as well as in other mammals, including deer and mongooses, 83, 87-89 though little is known about the sylvatic circulation of HEV.

The known animal reservoirs of HEV continue to multiply; novel members of the

Hepeviridae have been detected in wild and farmed rabbits, wild rats, birds, bats, and trout. 13, 15, 88-

106 Of these, only rabbit HEV strains have been shown to share a close phylogenetic relationship and possess antigenic similarity to human and swine-associated HEV. Newly-identified rabbit HEV strains appear to share a common ancestor with HEV-3 based on whole-genome and ORF2 sequence comparisons. 88, 93 However, rabbit strains contain a ~90-nucleotide insertion in ORF1 and cluster separately from most human and swine HEV-3 strains. 88, 95, 100

Genotype 4 is indigenous to South and East Asia, 18 where it has been detected in both wild and domesticated swine. 84, 98, 107-110 Short HEV sequence fragments obtained from sheep and cattle fecal samples, clustering with genotype 4, have been reported in the Chinese-language literature,

111, 112 but evidence for this expanded host range requires replication. 70 Serologic evidence of HEV infection obtained from a variety of mammals around the world, including dogs, goats, camels, cows, and buffalo, has yet to be followed by successful isolation and sequencing of HEV strains from these species. 14, 16, 17, 113, 114

Studies in eastern China have recently documented a shift in the HEV strains causing human illness; over the past decade or so, HEV-4 strains of zoonotic origin have overtaken HEV-1 strains as the major cause of hepatitis E in this region. 115-117 More recently, both imported and locally- acquired HEV-4 strains have been detected in Europe and the Americas. 118-124

In the relatively wealthy nations of Europe, North America, and East Asia, human infections with HEV-3 and HEV-4 are often asymptomatic, 33, 125 though more virulent strains appear to exist.

126, 127 In these settings, autochthonous HEV-3 appears to pose the greatest threat to people with chronic medical conditions that require immunosuppressive therapy or that directly compromise the immune system and/or liver. 128-130

Routes of exposure to animal-associated HEV Identified human exposures to HEV-3 and HEV-4 have occurred via consumption of undercooked pork or game meats, especially liver sausages or other organ meats, which are frequently contaminated with HEV. 70, 131-138 HEV-3 has also been detected in bivalves in sewage- contaminated waters, 139-141 and shellfish consumption has been identified as a possible risk factor for infection. 142, 143

However, cases without obvious food-linked exposures have been reported frequently, suggesting other environmental exposures may play a role. 144 Occupational exposures to farm animals, especially swine, have been associated with HEV infection in some studies in Africa,

Europe, and the United States 73, 133, 145, 146 but a recent study in Thailand found no difference in seroprevalence among swine workers compared with others in the local population. 147

Inadequately-treated manure from swine farms represents another potential source of environmental contamination. 148-150

A cross-sectional study of the U.S. population found that anti-HEV seroprevalence was elevated among those with pets at home. 35 In 2003, the pet cat of a Japanese man who developed hepatitis E was found to have anti-HEV antibodies; 151 subsequent to this case, a substantial proportion of sampled Japanese pet cats were found to be seropositive. 152 These findings suggest that some pet-related exposures – perhaps contaminated animal food, indirect contact with wild or domesticated animal reservoirs, or outdoor activity – could mediate some of the risk of human infection with zoonotic strains.

HYPOTHESES REGARDING REGIONAL DIFFERENCES IN THE NATURE AND IMPACT OF HEV

Ecologic Influences on HEV Epidemiology Across Regions

Limited exposures, typically low doses of virus, and better overall health may protect most otherwise-healthy people in industrialized nations from clinically-apparent hepatitis caused by

HEV-3. 35 In developing countries in Asia and Africa, where HEV-1 predominates, human infections with HEV-3 and HEV-4 are rarely reported, and no large outbreaks have been attributed to these strains. However, HEV-3 and HEV-4 cases have been confirmed sporadically in local populations, in travelers to these regions, 119, 144 and in resident animal populations and locally-obtained meat. 107,

108, 114, 153, 154

Broad environmental, dietary, and overall health differences across settings and limited information on the distribution of viral subgenotypes in many parts of the world make it difficult to discern whether HEV-1 and HEV-2 are inherently more virulent in humans than HEV-3 and HEV-4, or whether the occurrence of epidemics and the incidence of illness and death reflect primarily exposure- and host-related risk factors. Similarly, it remains unclear to what extent subtle subgenotypic differences in HEV-1 strains may contribute to the variability in patterns of illness in endemic areas, and what can instead be explained by host vulnerabilities (e.g., genotype, nutritional

status, coinfections, immunocompetence) or by environmental determinants of timing and dose of exposures (e.g., seasonal rainfall patterns, sanitation systems, animal exposures).

Genotypic and Subgenotypic Variation and Virulence

Despite the challenge of confounding by ecologic factors, there is some evidence to suggest that differences in circulating virus strains may influence patterns and severity of illness. Studies of circulating subgenotypes and strains within geographic regions have identified several variants with increased or decreased virulence.

Renou et al. have suggested that HEV-3 (subtype 3c), which is common in Europe, may be less likely to produce symptomatic illness than genotype 1; 33 however, the better overall health status and typically non-waterborne exposures of individuals in industrialized countries may at least partly explain the relative lack of symptomatic cases in these settings. Cordoba and colleagues identified an attenuated swine HEV-3 strain from the United States with mutations in the sequences encoding the capsid protein that appeared to be associated with reduced viral replication. 155

On the other hand, successful in vitro cultivation of HEV-3 and HEV-4 strains with nucleotide substitutions associated with more severe illness was reported by Inoue and colleagues in Japan in 2009. 127 In a comparison of isolates from patients with milder and more severe hepatitis E, several substitutions in the RNA helicase and capsid protein-encoding domains, in particular, were found to be associated with severe hepatitis E leading to liver failure. 127 Mutations identified in HEV-3 isolates collected from several severe hepatitis E cases across Japan, including another substitution affecting the helicase domain, also suggest that these mutations may affect the pathogenetic potential of HEV strains. 126 A series of cases in France, who presented with more severe disease than typical HEV-3 cases from the same area, were attributed to HEV-4 strains

closely related to a Belgian swine isolate; however, specific mutations related to virulence in these strains were not identified. 123

A recent report by Shukla and colleagues described the integration of a nucleotide sequence from human host S17 ribosomal RNA into the HEV-3 genome in a patient with chronic HEV infection who had both neurologic and hepatic symptoms. 156 In contrast to other tested HEV viral strains, this recombinant virus (the Kernow-C1 strain) replicated in cell culture and also may have grown in extra-hepatic sites in the patient. 156, 157 Similarly, Nguyen et al. isolated an HEV-3 strain from a chronically-infected U.S. liver-transplant patient that contained an insertion derived from the human ribosomal S19 sequence, and this strain, too, predominated in cell culture compared with isolates lacking the insertion. 158 These observations suggest another source of genomic variation that may be important in the pathogenesis and epidemiology of HEV.

FIGURES

Figure 1. Global distribution of hepatitis E virus (HEV) genotypes 1-4 in humans. Genotypes 1 and 2 circulate in human populations and are transmitted primarily through fecally- contaminated water supplies. Genotype 2 was first identified in Mexico, but has subsequently been found across Africa. Genotypes 3 and 4, with reservoirs in swine and other species, are the dominant genotypes affecting human populations in industrialized countries. [Base map of Earth adapted and simplified from “Blue Marble” imagery (NASA Earth Observatory, http://visibleearth.nasa.gov/view_cat.php?categoryID=1484), projected using G.Projector v.1.5 (NASA Goddard Institute for Space Studies, http://www.giss.nasa.gov/tools/gprojector/).]

Figure 2. Phylogenetic tree of global hepatitis E virus (HEV) isolates, based on a portion of the nucleotide sequence encoding the capsid protein. Genotypes 1 and 2 (HEV-1 and HEV-2) circulate among humans, primarily in Africa and Asia, while Genotypes 3 and 4 (HEV-3 and HEV-4) have animal reservoirs, and zoonotic, often foodborne, transmission has often been implicated where a source of infection can be identified. Recently-discovered rabbit strains appear to form a closely-related clade. Sequences from open reading frame (ORF) 2, homologous to the 350bp segment of the Burma B1 strain used by Lu, 12 were identified from within the GenBank database (accession numbers in parentheses). Alignments were performed using the Basic Local Alignment Search Tool (BLAST), 19 and the tree display was modified with MEGA5.2. 20

REFERENCES

1 Virchow RLK. 1854. Archiv für pathologische Anatomie und Physiologie und für klinische Medizin. Springer.

2 Saint-Vel O, Delahaye VA. 1868. Traité des maladies des régions intertropicales. Adrien Delahaye, Libraire-Éditeur.

3 Teo CG. 2012. Fatal outbreaks of jaundice in pregnancy and the epidemic history of hepatitis E. Epidemiol Infect. 140:767-787.

4 Balayan MS, Andjaparidze AG, Savinskaya SS, Ketiladze ES, Braginsky DM, Savinov AP, Poleschuk VF. 1983. Evidence for a virus in non-A, non-B hepatitis transmitted via the fecal-oral route. Intervirology. 20:23-31.

5 Reyes GR, Purdy MA, Kim JP, Luk KC, Young LM, Fry KE, Bradley DW. 1990. Isolation of a cDNA from the virus responsible for enterically transmitted non-A, non-B hepatitis. Science. 247:1335- 1339.

6 Tam AW, Smith MM, Guerra ME, Huang CC, Bradley DW, Fry KE, Reyes GR. 1991. Hepatitis E virus (HEV): molecular cloning and sequencing of the full-length viral genome. Virology. 185:120- 131.

7 Bradley DW, Purdy MA, Reyes GR. 1991. Hepatitis E virus genome. Molecular features, expression of immunoreactive proteins and sequence divergence. J Hepatol. 13 (Suppl 4):S152-154.

8 Ray R, Aggarwal R, Salunke PN, Mehrotra NN, Talwar GP, Naik SR. 1991. Hepatitis E virus genome in stools of hepatitis patients during large epidemic in north India. Lancet. 338:783-784.

9 Huang CC, Nguyen D, Fernandez J, Yun KY, Fry KE, Bradley DW, Tam AW, Reyes GR. 1992. Molecular cloning and sequencing of the Mexico isolate of hepatitis E virus (HEV). Virology. 191:550- 558.

10 Uchida T, Suzuki K, Hayashi N, Iida F, Hara T, Oo SS, Wang CK, Shikata T, Ichikawa M, Rikihisa T, Mizuno K, Win KM. 1992. Hepatitis E virus: cDNA cloning and expression. Microbiol Immunol. 36:67-79.

11 Tsarev SA, Emerson SU, Reyes GR, Tsareva TS, Legters LJ, Malik IA, Iqbal M, Purcell RH. 1992. Characterization of a prototype strain of hepatitis E virus. Proc Natl Acad Sci U S A. 89:559-563.

12 Lu L, Li C, Hagedorn CH. 2006. Phylogenetic analysis of global hepatitis E virus sequences: genetic diversity, subtypes and zoonosis. Rev Med Virol. 16:5-36.

13 Kabrane-Lazizi Y, Fine JB, Elm J, Glass GE, Higa H, Diwan A, Gibbs CJ, Jr., Meng XJ, Emerson SU, Purcell RH. 1999. Evidence for widespread infection of wild rats with hepatitis E virus in the United States. Am J Trop Med Hyg. 61:331-335.

14 Arankalle VA, Joshi MV, Kulkarni AM, Gandhe SS, Chobe LP, Rautmare SS, Mishra AC, Padbidri VS. 2001. Prevalence of anti-hepatitis E virus antibodies in different Indian animal species. J Viral Hepat. 8:223-227.

15 Drexler JF, Seelen A, Corman VM, Fumie Tateno A, Cottontail V, Melim Zerbinati R, Gloza- Rausch F, Klose SM, Adu-Sarkodie Y, Oppong SK, Kalko EK, Osterman A, Rasche A, Adam A, Muller MA, Ulrich RG, Leroy EM, Lukashev AN, Drosten C. 2012. Bats worldwide carry hepatitis E- related viruses that form a putative novel genus within the family Hepeviridae. J Virol. 86:9134- 9147.

16 Vitral CL, Pinto MA, Lewis-Ximenez LL, Khudyakov YE, dos Santos DR, Gaspar AM. 2005. Serological evidence of hepatitis E virus infection in different animal species from the Southeast of Brazil. Mem Inst Oswaldo Cruz. 100:117-122.

17 El-Tras WF, Tayel AA, El-Kady NN. 2012. Seroprevalence of Hepatitis E Virus in Humans and Geographically Matched Food Animals in Egypt. Zoonoses Public Health. 60:244-251.

18 Okamoto H. 2007. Genetic variability and evolution of hepatitis E virus. Virus Res. 127:216-228.

19 Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment search tool. J Mol Biol. 215:403-410.

20 Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S. 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol. 28:2731-2739.

21 Khuroo MS. 1980. Study of an epidemic of non-A, non-B hepatitis. Possibility of another human hepatitis virus distinct from post-transfusion non-A, non-B type. Am J Med. 68:818-824.

22 Aggarwal R. 2011. Clinical presentation of hepatitis E. Virus Res. 161:15-22.

23 Labrique AB, Zaman K, Hossain Z, Saha P, Yunus M, Hossain A, Ticehurst JR, Nelson KE. 2010. Epidemiology and risk factors of incident hepatitis E virus infections in rural Bangladesh. Am J Epidemiol. 172:952-961.

24 Dalton HR, Stableforth W, Thurairajah P, Hazeldine S, Remnarace R, Usama W, Farrington L, Hamad N, Sieberhagen C, Ellis V, Mitchell J, Hussaini SH, Banks M, Ijaz S, Bendall RP. 2008. Autochthonous hepatitis E in Southwest England: natural history, complications and seasonal variation, and hepatitis E virus IgG seroprevalence in blood donors, the elderly and patients with chronic liver disease. Eur J Gastroenterol Hepatol. 20:784-790.

25 Aggarwal RA. 2013. Hepatitis e: clinical presentation in disease-endemic areas and diagnosis. Semin Liver Dis. 33:30-40.

26 Begum N, Devi SG, Husain SA, Kar P. 2009. Seroprevalence of subclinical HEV infection in pregnant women from north India: a hospital based study. Indian J Med Res. 130:709-713.

27 Christensen PB, Engle RE, Hjort C, Homburg KM, Vach W, Georgsen J, Purcell RH. 2008. Time trend of the prevalence of hepatitis E antibodies among farmers and blood donors: a potential zoonosis in Denmark. Clin Infect Dis. 47:1026-1031.

28 Dalton HR, Thurairajah PH, Fellows HJ, Hussaini HS, Mitchell J, Bendall R, Banks M, Ijaz S, Teo CG, Levine DF. 2007. Autochthonous hepatitis E in southwest England. J Viral Hepat. 14:304-309.

29 Gad YZ, Mousa N, Shams M, Elewa A. 2011. Seroprevalence of subclinical HEV infection in asymptomatic, apparently healthy, pregnant women in Dakahlya Governorate, Egypt. Asian J Transfus Sci. 5:136-139. 12

30 Guthmann JP, Klovstad H, Boccia D, Hamid N, Pinoges L, Nizou JY, Tatay M, Diaz F, Moren A, Grais RF, Ciglenecki I, Nicand E, Guerin PJ. 2006. A large outbreak of hepatitis E among a displaced population in Darfur, Sudan, 2004: the role of water treatment methods. Clin Infect Dis. 42:1685- 1691.

31 Khuroo MS, Kamili S, Yattoo GN. 2004. Hepatitis E virus infection may be transmitted through blood transfusions in an endemic area. J Gastroenterol Hepatol. 19:778-784.

32 Nicand E, Grandadam M, Teyssou R, Rey JL, Buisson Y. 2001. Viraemia and faecal shedding of HEV in symptom-free carriers. Lancet. 357:68-69.

33 Renou C, Pariente A, Cadranel JF, Nicand E, Pavio N. 2011. Clinically silent forms may partly explain the rarity of acute cases of autochthonous genotype 3c hepatitis E infection in France. J Clin Virol. 51:139-141.

34 Stoszek SK, Engle RE, Abdel-Hamid M, Mikhail N, Abdel-Aziz F, Medhat A, Fix AD, Emerson SU, Purcell RH, Strickland GT. 2006. Hepatitis E antibody seroconversion without disease in highly endemic rural Egyptian communities. Trans R Soc Trop Med Hyg. 100:89-94.

35 Kuniholm MH, Purcell RH, McQuillan GM, Engle RE, Wasley A, Nelson KE. 2009. Epidemiology of hepatitis E virus in the United States: results from the Third National Health and Nutrition Examination Survey, 1988-1994. J Infect Dis. 200:48-56.

36 Schlosser B, Stein A, Neuhaus R, Pahl S, Ramez B, Kruger DH, Berg T, Hofmann J. 2012. Liver transplant from a donor with occult HEV infection induced chronic hepatitis and cirrhosis in the recipient. J Hepatol. 56:500-502.

37 Despierres LA, Kaphan E, Attarian S, Cohen-Bacrie S, Pelletier J, Pouget J, Motte A, Charrel R, Gérolami R, Colson P. 2011. Neurologic disorders and hepatitis E, France, 2010. Emerg Infect Dis. 17:1510-1512.

38 Kejariwal D, Roy S, Sarkar N. 2001. Seizure associated with acute hepatitis E. Neurology. 57:1935.

39 Mandal K, Chopra N. 2006. Acute transverse myelitis following hepatitis E virus infection. Indian Pediatr. 43:365-366.

40 Kamani P, Baijal R, Amarapurkar D, Gupte P, Patel N, Kumar P, Agal S. 2005. Guillain-Barre syndrome associated with acute hepatitis E. Indian J Gastroenterol. 24:216.

41 Scharn N, Ganzenmueller T, Wenzel JJ, Dengler R, Heim A, Wegner F. 2013. Guillain-Barre syndrome associated with autochthonous infection by hepatitis E virus subgenotype 3c. Infection. [online March 2013, doi: 10.1007/s15010-013-0448-5].

42 Sharma B, Nagpal K, Bakki Sannegowda R, Prakash S. 2013. Hepatitis E with Gullain-Barre syndrome: still a rare association. J Neurovirol. 19:186-187.

43 Loly JP, Rikir E, Seivert M, Legros E, Defrance P, Belaiche J, Moonen G, Delwaide J. 2009. Guillain-Barre syndrome following hepatitis E. World J Gastroenterol. 15:1645-1647.

44 Sood A, Midha V, Sood N. 2000. Guillain-Barre syndrome with acute hepatitis E. Am J Gastroenterol. 95:3667-3668.

45 Cronin S, McNicholas R, Kavanagh E, Reid V, O'Rourke K. 2011. Anti-glycolipid GM2-positive Guillain-Barre syndrome due to hepatitis E infection. Ir J Med Sci. 180:255-257. 13

46 Maurissen I, Jeurissen A, Strauven T, Sprengers D, De Schepper B. 2012. First case of anti- ganglioside GM1-positive Guillain-Barre syndrome due to hepatitis E virus infection. Infection. 40:323-326.

47 Fong F, Illahi M. 2009. Neuralgic amyotrophy associated with hepatitis E virus. Clin Neurol Neurosurg. 111:193-195.

48 Kamar N, Bendall RP, Peron JM, Cintas P, Prudhomme L, Mansuy JM, Rostaing L, Keane F, Ijaz S, Izopet J, Dalton HR. 2011. Hepatitis E virus and neurologic disorders. Emerg Infect Dis. 17:173-179.

49 Coursaget P, Buisson Y, N'Gawara MN, van Cuyck-Gandre H, Roue R. 1998. Role of hepatitis E virus in sporadic cases of acute and fulminant hepatitis in an endemic area (Chad). Am J Trop Med Hyg. 58:330-334.

50 Mahtab MA, Rahman S, Khan M, Karim F. 2009. HEV infection as an aetiologic factor for acute hepatitis: experience from a tertiary hospital in Bangladesh. J Health Popul Nutr. 27:14-19.

51 Ghoshal UC, Somani S, Chetri K, Akhtar P, Aggarwal R, Naik SR. 2001. Plasmodium falciparum and hepatitis E virus co-infection in fulminant hepatic failure. Indian J Gastroenterol. 20:111.

52 Kumar A, Aggarwal R, Naik SR, Saraswat V, Ghoshal UC, Naik S. 2004. Hepatitis E virus is responsible for decompensation of chronic liver disease in an endemic region. Indian J Gastroenterol. 23:59-62.

53 Montalva Villalba MC, Rodriguez Lay LA, Chandra V, Corredor MB, Frometa SS, Moreno AG, Jameel S. 2008. Hepatitis E virus genotype 1, Cuba. Emerg Infect Dis. 14:1320-1322.

54 Garcia CG, Sanchez D, Villalba MC, Pujol FH, de Los Angeles Rodriguez Lay L, Pinto B, Chacon EP, Guzman MG. 2012. Molecular characterization of hepatitis E virus in patients with acute hepatitis in Venezuela. J Med Virol. 84:1025-1029.

55 Tavera C, Velazquez O, Avila C, Ornelas G, Alvarez C, Sepulveda J. 1987. Enterically transmitted non-A, non-B hepatitis--Mexico. MMWR Morb Mortal Wkly Rep. 36:597-602.

56 Velazquez O, Stetler HC, Avila C, Ornelas G, Hadler CC, Bradley DW, Sepúlveda J. 1990. Epidemic Transmission of Enterically Transmitted Non-A, Non-B Hepatitis in Mexico, 1986-1987. JAMA. 263:3281-3285.

57 Maila HT, Bowyer SM, Swanepoel R. 2004. Identification of a new strain of hepatitis E virus from an outbreak in Namibia in 1995. J Gen Virol. 85:89-95.

58 Nicand E, Armstrong GL, Enouf V, Guthmann JP, Guerin JP, Caron M, Nizou JY, Andraghetti R. 2005. Genetic heterogeneity of hepatitis E virus in Darfur, Sudan, and neighboring Chad. J Med Virol. 77:519-521.

59 Arankalle VA, Chadha MS, Chitambar SD, Walimbe AM, Chobe LP, Gandhe SS. 2001. Changing epidemiology of hepatitis A and hepatitis E in urban and rural India (1982-98). J Viral Hepat. 8:293- 303.

60 Arankalle VA, Tsarev SA, Chadha MS, Alling DW, Emerson SU, Banerjee K, Purcell RH. 1995. Age-specific prevalence of antibodies to hepatitis A and E viruses in Pune, India, 1982 and 1992. J Infect Dis. 171:447-450.

61 Arankalle VA, Chadha MS, Mehendale SM, Tungatkar SP. 2000. Epidemic hepatitis E: serological evidence for lack of intrafamilial spread. Indian J Gastroenterol. 19:24-28.

62 Teshale EH, Grytdal SP, Howard C, Barry V, Kamili S, Drobeniuc J, Hill VR, Okware S, Hu DJ, Holmberg SD. 2010. Evidence of person-to-person transmission of hepatitis E virus during a large outbreak in Northern Uganda. Clin Infect Dis. 50:1006-1010.

63 Khuroo MS, Dar MY. 1992. Hepatitis E: evidence for person-to-person transmission and inability of low dose immune serum globulin from an Indian source to prevent it. Indian J Gastroenterol. 11:113- 116.

64 Howard CM, Handzel T, Hill VR, Grytdal SP, Blanton C, Kamili S, Drobeniuc J, Hu D, Teshale E. 2010. Novel risk factors associated with hepatitis E virus infection in a large outbreak in northern Uganda: results from a case-control study and environmental analysis. Am J Trop Med Hyg. 83:1170- 1173.

65 Khatun F, Rasheed S, Moran AC, Alam AM, Shomik MS, Sultana M, Choudhury N, Iqbal M, Bhuiya A. 2012. Causes of neonatal and maternal deaths in Dhaka slums: implications for service delivery. BMC Public Health. 12:84.

66 Boccia D, Guthmann JP, Klovstad H, Hamid N, Tatay M, Ciglenecki I, Nizou JY, Nicand E, Guerin PJ. 2006. High mortality associated with an outbreak of hepatitis E among displaced persons in Darfur, Sudan. Clin Infect Dis. 42:1679-1684.

67 Gove S, Ali-Salad A, Farah MA, Delaney D, Roble MJ, Walter J, Aziz N. 1987. Enterically transmitted non-A, non-B hepatitis--East Africa. MMWR Morb Mortal Wkly Rep. 36:241-244.

68 Isaacson M, Frean J, He J, Seriwatana J, Innis BL. 2000. An outbreak of hepatitis E in Northern Namibia, 1983. Am J Trop Med Hyg. 62:619-625.

69 Meng XJ. 2010. Hepatitis E virus: animal reservoirs and zoonotic risk. Vet Microbiol. 140:256-265.

70 Meng XJ. 2011. From barnyard to food table: the omnipresence of hepatitis E virus and risk for zoonotic infection and food safety. Virus Res. 161:23-30.

71 Adlhoch C, Wolf A, Meisel H, Kaiser M, Ellerbrok H, Pauli G. 2009. High HEV presence in four different wild boar populations in East and West Germany. Vet Microbiol. 139:270-278.

72 de Deus N, Casas M, Peralta B, Nofrarias M, Pina S, Martin M, Segales J. 2008. Hepatitis E virus infection dynamics and organic distribution in naturally infected pigs in a farrow-to-finish farm. Vet Microbiol. 132:19-28.

73 de la Caridad Montalvo Villalba M, Owot JC, Corrreia B, Corredor MB, Flaquet PP, Frometa SS, Wong MS, Rodriguez Lay LD. 2013. Hepatitis E virus genotype 3 in humans and swine, Cuba. Infect Genet Evol. 14C:335-339.

74 de Souza AJ, Gomes-Gouvea MS, Soares MD, Pinho JR, Malheiros AP, Carneiro LA, Dos Santos DR, Pereira WL. 2012. HEV infection in swine from Eastern Brazilian Amazon: Evidence of coinfection by different subtypes. Comp Immunol Microbiol Infect Dis. 35:477-485.

75 Dell'Amico MC, Cavallo A, Gonzales JL, Bonelli SI, Valda Y, Pieri A, Segund H, Ibanez R, Mantella A, Bartalesi F, Tolari F, Bartoloni A. 2011. Hepatitis E virus genotype 3 in humans and Swine, Bolivia. Emerg Infect Dis. 17:1488-1490. 15

76 Dong C, Meng J, Dai X, Liang JH, Feagins AR, Meng XJ, Belfiore NM, Bradford C, Corn JL, Cray C, Glass GE, Gordon ML, Hesse RA, Montgomery DL, Nicholson WL, Pilny AA, Ramamoorthy S, Shaver DD, Drobeniuc J, Purdy MA, Fields HA, Kamili S, Teo CG. 2011. Restricted enzooticity of hepatitis E virus genotypes 1 to 4 in the United States. J Clin Microbiol. 49:4164-4172.

77 Hosmillo M, Jeong YJ, Kim HJ, Park JG, Nayak MK, Alfajaro MM, Collantes TM, Park SJ, Ikuta K, Yunoki M, Kang MI, Park SI, Cho KO. 2010. Molecular detection of genotype 3 porcine hepatitis E virus in aborted fetuses and their sows. Arch Virol. 155:1157-1161.

78 Ishida S, Yoshizumi S, Ikeda T, Miyoshi M, Goto A, Matsubayashi K, Ikeda H. 2012. Detection and molecular characterization of hepatitis E virus in clinical, environmental and putative animal sources. Arch Virol. 157:2363-2368.

79 Kanai Y, Tsujikawa M, Yunoki M, Nishiyama S, Ikuta K, Hagiwara K. 2010. Long-term shedding of hepatitis E virus in the feces of pigs infected naturally, born to sows with and without maternal antibodies. J Med Virol. 82:69-76.

80 Keawcharoen J, Thongmee T, Panyathong R, Joiphaeng P, Tuanthap S, Oraveerakul K, Theamboonlers A, Poovorawan Y. 2012. Hepatitis E virus genotype 3f sequences from pigs in Thailand, 2011-2012. Virus Genes. 46:369-370.

81 Lu YH, Qian HZ, Qin X, Jiang QW, Zheng YJ. 2013. Subtypes of genotype 3 hepatitis E virus in pigs. Vet J. 197(2):509-511.

82 Norder H, Sundqvist L, Magnusson L, Ostergaard Breum S, Lofdahl M, Larsen LE, Hjulsager CK, Magnius L, Bottiger BE, Widen F. 2009. Endemic hepatitis E in two Nordic countries. Euro Surveill. 14:pii=19211.

83 Reuter G, Fodor D, Forgach P, Katai A, Szucs G. 2009. Characterization and zoonotic potential of endemic hepatitis E virus (HEV) strains in humans and animals in Hungary. J Clin Virol. 44:277-281.

84 Sato Y, Sato H, Naka K, Furuya S, Tsukiji H, Kitagawa K, Sonoda Y, Usui T, Sakamoto H, Yoshino S, Shimizu Y, Takahashi M, Nagashima S, Jirintai, Nishizawa T, Okamoto H. 2011. A nationwide survey of hepatitis E virus (HEV) infection in wild boars in Japan: identification of boar HEV strains of genotypes 3 and 4 and unrecognized genotypes. Arch Virol. 156:1345-1358.

85 Temmam S, Besnard L, Andriamandimby SF, Foray C, Rasamoelina-Andriamanivo H, Heraud JM, Cardinale E, Dellagi K, Pavio N, Pascalis H, Porphyre V. 2013. High Prevalence of Hepatitis E in Humans and Pigs and Evidence of Genotype-3 Virus in Swine, Madagascar. Am J Trop Med Hyg. 88:329-338.

86 Zwettler D, Fink M, Revilla-Fernandez S, Steinrigl A, Winter P, Kofer J. 2012. First detection of hepatitis E virus in Austrian pigs by RT-qPCR. Berl Munch Tierarztl Wochenschr. 125:281-289.

87 Cooper K, Huang FF, Batista L, Rayo CD, Bezanilla JC, Toth TE, Meng XJ. 2005. Identification of genotype 3 hepatitis E virus (HEV) in serum and fecal samples from pigs in Thailand and Mexico, where genotype 1 and 2 HEV strains are prevalent in the respective human populations. J Clin Microbiol. 43:1684-1688.

88 Cossaboom CM, Cordoba L, Cao D, Ni YY, Meng XJ. 2012. Complete genome sequence of hepatitis E virus from rabbits in the United States. J Virol. 86:13124-13125.

89 Kanai Y, Miyasaka S, Uyama S, Kawami S, Kato-Mori Y, Tsujikawa M, Yunoki M, Nishiyama S, Ikuta K, Hagiwara K. 2012. Hepatitis E virus in Norway rats (Rattus norvegicus) captured around a pig farm. BMC Res Notes. 5:4.

90 Cossaboom CM, Cordoba L, Dryman BA, Meng XJ. 2011. Hepatitis E virus in rabbits, Virginia, USA. Emerg Infect Dis. 17:2047-2049.

91 Geng J, Fu H, Wang L, Bu Q, Liu P, Wang M, Sui Y, Wang X, Zhu Y, Zhuang H. 2011. Phylogenetic analysis of the full genome of rabbit hepatitis E virus (rbHEV) and molecular biologic study on the possibility of cross species transmission of rbHEV. Infect Genet Evol. 11:2020-2025.

92 Geng J, Wang L, Wang X, Fu H, Bu Q, Zhu Y, Zhuang H. 2011. Study on prevalence and genotype of hepatitis E virus isolated from Rex Rabbits in Beijing, China. J Viral Hepat. 18:661-667.

93 Geng Y, Zhang H, Li J, Huang W, Harrison TJ, Zhao C, Zhou Y, Lian H, Wang Y. 2013. Comparison of hepatitis E virus genotypes from rabbits and pigs in the same geographic area: no evidence of natural cross-species transmission between the two animals. Infect Genet Evol. 13:304-309.

94 Geng Y, Zhao C, Song A, Wang J, Zhang X, Harrison TJ, Zhou Y, Wang W, Wang Y. 2011. The serological prevalence and genetic diversity of hepatitis E virus in farmed rabbits in China. Infect Genet Evol. 11:476-482.

95 Izopet J, Dubois M, Bertagnoli S, Lhomme S, Marchandeau S, Boucher S, Kamar N, Abravanel F, Guerin JL. 2012. Hepatitis E virus strains in rabbits and evidence of a closely related strain in humans, France. Emerg Infect Dis. 18:1274-1281.

96 Jirintai S, Jinshan, Tanggis, Manglai D, Mulyanto, Takahashi M, Nagashima S, Kobayashi T, Nishizawa T, Okamoto H. 2012. Molecular analysis of hepatitis E virus from farm rabbits in Inner Mongolia, China and its successful propagation in A549 and PLC/PRF/5 cells. Virus Res. 170:126- 137.

97 Lhomme S, Dubois M, Abravanel F, Top S, Bertagnoli S, Guerin JL, Izopet J. 2013. Risk of zoonotic transmission of HEV from rabbits. J Clin Virol. 58:357-362.

98 Liu P, Li L, Wang L, Bu Q, Fu H, Han J, Zhu Y, Lu F, Zhuang H. 2012. Phylogenetic analysis of 626 hepatitis E virus (HEV) isolates from humans and animals in China (1986-2011) showing genotype diversity and zoonotic transmission. Infect Genet Evol. 12:428-434.

99 Wang S, Cheng X, Dai X, Dong C, Xu M, Liang J, Dong M, Purdy MA, Meng J. 2013. Rabbit and human hepatitis E virus strains belong to a single serotype. Virus Res. 176:101-106.

100 Wang S, Dong C, Dai X, Cheng X, Liang J, Dong M, Purdy MA, Meng J. 2013. Hepatitis E virus isolated from rabbits is genetically heterogeneous but with very similar antigenicity to human HEV. J Med Virol. 85:627-635.

101 Zhao C, Ma Z, Harrison TJ, Feng R, Zhang C, Qiao Z, Fan J, Ma H, Li M, Song A, Wang Y. 2009. A novel genotype of hepatitis E virus prevalent among farmed rabbits in China. J Med Virol. 81:1371- 1379.

102 Hirano M, Ding X, Li TC, Takeda N, Kawabata H, Koizumi N, Kadosaka T, Goto I, Masuzawa T, Nakamura M, Taira K, Kuroki T, Tanikawa T, Watanabe H, Abe K. 2003. Evidence for widespread infection of hepatitis E virus among wild rats in Japan. Hepatol Res. 27:1-5.

103 Johne R, Heckel G, Plenge-Bonig A, Kindler E, Maresch C, Reetz J, Schielke A, Ulrich RG. 2010. Novel hepatitis E virus genotype in Norway rats, Germany. Emerg Infect Dis. 16:1452-1455.

104 Koma T, Yoshimatsu K, Yasuda SP, Li T, Amada T, Shimizu K, Isozumi R, Mai LT, Hoa NT, Nguyen V, Yamashiro T, Hasebe F, Arikawa J. 2012. A survey of rodent-borne pathogens carried by wild Rattus spp. in Northern Vietnam. Epidemiol Infect. 141:1876-1884.

105 Purcell RH, Engle RE, Rood MP, Kabrane-Lazizi Y, Nguyen HT, Govindarajan S, St Claire M, Emerson SU. 2011. Hepatitis E virus in rats, Los Angeles, California, USA. Emerg Infect Dis. 17:2216- 2222.

106 Batts W, Yun S, Hedrick R, Winton J. 2011. A novel member of the family Hepeviridae from cutthroat trout (Oncorhynchus clarkii). Virus Res. 158:116-123.

107 Kulkarni MA, Arankalle VA. 2008. The detection and characterization of hepatitis E virus in pig livers from retail markets of India. J Med Virol. 80:1387-1390.

108 Vivek R, Kang G. 2011. Hepatitis E virus infections in swine and swine handlers in Vellore, Southern India. Am J Trop Med Hyg. 84:647-649.

109 Wang H, He Y, Shen Q, Wang X, Yang S, Cui L, Ren L, Sun G, Hua X, Shao S, Zhang W. 2012. Complete genome sequence of the genotype 4 hepatitis E virus strain prevalent in swine in Jiangsu Province, China, reveals a close relationship with that from the human population in this area. J Virol. 86:8334-8335.

110 Zhang W, Yang S, Ren L, Shen Q, Cui L, Fan K, Huang F, Kang Y, Shan T, Wei J, Xiu H, Lou Y, Liu J, Yang Z, Zhu J, Hua X. 2009. Hepatitis E virus infection in central China reveals no evidence of cross- species transmission between human and swine in this area. PLoS One. 4:e8156.

111 Hu GD, Ma X. 2010. [Detection and sequences analysis of bovine hepatitis E virus RNA in Xinjiang Autonomous Region]. Bing Du Xue Bao. 26:27-32.

112 Wang Y, Ma X. 2010. [Detection and sequences analysis of sheep hepatitis E virus RNA in Xinjiang autonomous region]. Wei Sheng Wu Xue Bao. 50:937-941.

113 Sanford BJ, Emerson SU, Purcell RH, Engle RE, Dryman BA, Cecere TE, Buechner-Maxwell V, Sponenberg DP, Meng XJ. 2012. Serological Evidence for a Hepatitis E Virus-Related Agent in Goats in the United States. Transbound Emerg Dis. [Online August 2012; doi: 10.1111/tbed.12001]

114 Shukla P, Chauhan UK, Naik S, Anderson D, Aggarwal R. 2007. Hepatitis E virus infection among animals in northern India: an unlikely source of human disease. J Viral Hepat. 14:310-317.

115 Zhang W, He Y, Wang H, Shen Q, Cui L, Wang X, Shao S, Hua X. 2010. Hepatitis E virus genotype diversity in eastern China. Emerg Infect Dis. 16:1630-1632.

116 Zheng Y, Ge S, Zhang J, Guo Q, Ng MH, Wang F, Xia N, Jiang Q. 2006. Swine as a principal reservoir of hepatitis E virus that infects humans in eastern China. J Infect Dis. 193:1643-1649.

117 Geng JB, Wang MR, Wang L, Wang J, Yang ZG, Cheng Y, Qiao F, Wang M. 2012. Genetic characteristics and pathogenicity of human hepatitis E virus in Nanjing, China. World J Gastroenterol. 18:965-970.

118 Garbuglia AR, Scognamiglio P, Petrosillo N, Mastroianni CM, Sordillo P, Gentile D, La Scala P, Girardi E, Capobianchi MR. 2013. Hepatitis E virus genotype 4 outbreak, Italy, 2011. Emerg Infect Dis. 19:110-114.

119 Fogeda M, Avellon A, Cilla CG, Echevarria JM. 2009. Imported and autochthonous hepatitis E virus strains in Spain. J Med Virol. 81:1743-1749.

120 Colson P, Romanet P, Moal V, Borentain P, Purgus R, Benezech A, Motte A, Gérolami R. 2012. Autochthonous infections with hepatitis E virus genotype 4, France. Emerg Infect Dis. 18:1361-1364.

121 Colson P, Swiader L, Motte A, Ferretti A, Borentain P, Gérolami R. 2012. Circulation of almost genetically identical hepatitis E virus of genotype 4 in France. J Clin Virol. 55:181-183.

122 Hakze-van der Honing RW, van Coillie E, Antonis AF, van der Poel WH. 2011. First isolation of hepatitis E virus genotype 4 in Europe through swine surveillance in the Netherlands and Belgium. PLoS One. 6:e22673.

123 Jeblaoui A, Haim-Boukobza S, Marchadier E, Mokhtari C, Roque-Afonso AM. 2013. Genotype 4 Hepatitis E Virus in France: An Autochthonous Infection With a More Severe Presentation. Clin Infect Dis. 57:e122-126.

124 Tessé S, Lioure B, Fornecker L, Wendling MJ, Stoll-Keller F, Bigaillon C, Nicand E. 2012. Circulation of genotype 4 hepatitis E virus in Europe: first autochthonous hepatitis E infection in France. J Clin Virol. 54:197-200.

125 Takahashi M, Tamura K, Hoshino Y, Nagashima S, Yazaki Y, Mizuo H, Iwamoto S, Okayama M, Nakamura Y, Kajii E, Okamoto H. 2010. A nationwide survey of hepatitis E virus infection in the general population of Japan. J Med Virol. 82:271-281.

126 Takahashi K, Okamoto H, Abe N, Kawakami M, Matsuda H, Mochida S, Sakugawa H, Suginoshita Y, Watanabe S, Yamamoto K, Miyakawa Y, Mishiro S. 2009. Virulent strain of hepatitis E virus genotype 3, Japan. Emerg Infect Dis. 15:704-709.

127 Inoue J, Takahashi M, Mizuo H, Suzuki K, Aikawa T, Shimosegawa T, Okamoto H. 2009. Nucleotide substitutions of hepatitis E virus genomes associated with fulminant hepatitis and disease severity. Tohoku J Exp Med. 218:279-284.

128 Kamar N, Abravanel F, Selves J, Garrouste C, Esposito L, Lavayssière L, Cointault O, Ribes D, Cardeau I, Nogier MB, Mansuy JM, Muscari F, Peron JM, Izopet J, Rostaing L. 2010. Influence of immunosuppressive therapy on the natural history of genotype 3 hepatitis-E virus infection after organ transplantation. Transplantation. 89:353-360.

129 Kamar N, Selves J, Mansuy JM, Ouezzani L, Peron JM, Guitard J, Cointault O, Esposito L, Abravanel F, Danjoux M, Durand D, Vinel JP, Izopet J, Rostaing L. 2008. Hepatitis E virus and chronic hepatitis in organ-transplant recipients. N Engl J Med. 358:811-817.

130 Pischke S, Suneetha PV, Baechlein C, Barg-Hock H, Heim A, Kamar N, Schlue J, Strassburg CP, Lehner F, Raupach R, Bremer B, Magerstedt P, Cornberg M, Seehusen F, Baumgaertner W, Klempnauer J, Izopet J, Manns MP, Grummer B, Wedemeyer H. 2010. Hepatitis E virus infection as a cause of graft hepatitis in liver transplant recipients. Liver Transpl. 16:74-82.

131 Colson P, Borentain P, Queyriaux B, Kaba M, Moal V, Gallian P, Heyries L, Raoult D, Gérolami R. 2010. Pig liver sausage as a source of hepatitis E virus transmission to humans. J Infect Dis. 202:825- 834.

132 Meng XJ, Wiseman B, Elvinger F, Guenette DK, Toth TE, Engle RE, Emerson SU, Purcell RH. 2002. Prevalence of antibodies to hepatitis E virus in veterinarians working with swine and in normal blood donors in the United States and other countries. J Clin Microbiol. 40:117-122.

133 Perez-Gracia MT, Mateos ML, Galiana C, Fernandez-Barredo S, Garcia A, Gomez MT, Moreira V. 2007. Autochthonous hepatitis E infection in a slaughterhouse worker. Am J Trop Med Hyg. 77:893- 896.

134 Teo CG. 2010. Much meat, much malady: changing perceptions of the epidemiology of hepatitis E. Clin Microbiol Infect. 16:24-32.

135 Matsubayashi K, Kang JH, Sakata H, Takahashi K, Shindo M, Kato M, Sato S, Kato T, Nishimori H, Tsuji K, Maguchi H, Yoshida J, Maekubo H, Mishiro S, Ikeda H. 2008. A case of transfusion- transmitted hepatitis E caused by blood from a donor infected with hepatitis E virus via zoonotic food-borne route. Transfusion. 48:1368-1375.

136 Feagins AR, Opriessnig T, Guenette DK, Halbur PG, Meng XJ. 2007. Detection and characterization of infectious Hepatitis E virus from commercial pig livers sold in local grocery stores in the USA. J Gen Virol. 88:912-917.

137 Berto A, Grierson S, Hakze-van der Honing R, Martelli F, Johne R, Reetz J, Ulrich RG, Pavio N, Van der Poel WH, Banks M. 2013. Hepatitis e virus in pork liver sausage, france. Emerg Infect Dis. 19:264-266.

138 Chaussade H, Rigaud E, Allix A, Carpentier A, Touzé A, Delzescaux D, Choutet P, Garcia-Bonnet N, Coursaget P. 2013. Hepatitis E virus seroprevalence and risk factors for individuals in working contact with animals. J Clin Virol. [Online Spetember 2013; doi: 10.1016/j.jcv.2013.08.030]

139 Crossan C, Baker PJ, Craft J, Takeuchi Y, Dalton HR, Scobie L. 2012. Hepatitis E virus genotype 3 in shellfish, United Kingdom. Emerg Infect Dis. 18:2085-2087.

140 Diez-Valcarce M, Kokkinos P, Soderberg K, Bouwknegt M, Willems K, de Roda-Husman AM, von Bonsdorff CH, Bellou M, Hernandez M, Maunula L, Vantarakis A, Rodriguez-Lazaro D. 2012. Occurrence of human enteric viruses in commercial mussels at retail level in three European countries. Food Environ Virol. 4:73-80.

141 Donia D, Dell'Amico MC, Petrinca AR, Martinucci I, Mazzei M, Tolari F, Divizia M. 2012. Presence of hepatitis E RNA in mussels used as bio-monitors of viral marine pollution. J Virol Methods. 186:198-202.

142 Renou C, Moreau X, Pariente A, Cadranel JF, Maringe E, Morin T, Causse X, Payen JL, Izopet J, Nicand E, Bourliere M, Penaranda G, Hardwigsen J, Gérolami R, Peron JM, Pavio N, Angh F. 2008. A national survey of acute hepatitis E in France. Aliment Pharmacol Ther. 27:1086-1093.

143 La Rosa G, Muscillo M, Vennarucci VS, Garbuglia AR, La Scala P, Capobianchi MR. 2011. Hepatitis E virus in Italy: molecular analysis of travel-related and autochthonous cases. J Gen Virol. 92:1617- 1626.

144 Rolfe KJ, Curran MD, Mangrolia N, Gelson W, Alexander GJ, L'Estrange M, Vivek R, Tedder R, Ijaz S. 2010. First case of genotype 4 human hepatitis E virus infection acquired in India. J Clin Virol. 48:58-61.

145 Adjei AA, Tettey Y, Aviyase JT, Adu-Gyamfi C, Mingle JA, Nartey ET. 2010. Unexpected elevated alanine aminotransferase, aspartate aminotransferase levels and hepatitis E virus infection among persons who work with pigs in Accra, Ghana. Virol J. 7:336.

146 Galiana C, Fernandez-Barredo S, Garcia A, Gomez MT, Perez-Gracia MT. 2008. Occupational exposure to hepatitis E virus (HEV) in swine workers. Am J Trop Med Hyg. 78:1012-1015.

147 Hinjoy S, Nelson KE, Gibbons RV, Jarman RG, Mongkolsirichaikul D, Smithsuwan P, Fernandez S, Labrique AB, Patchanee P. 2012. A Cross-Sectional Study of Hepatitis E Virus Infection in Healthy People Directly Exposed and Unexposed to Pigs in a Rural Community in Northern Thailand. Zoonoses Public Health. [Online December 2012; doi: 10.1111/zph.12030]

148 Kasorndorkbua C, Opriessnig T, Huang FF, Guenette DK, Thomas PJ, Meng XJ, Halbur PG. 2005. Infectious swine hepatitis E virus is present in pig manure storage facilities on United States farms, but evidence of water contamination is lacking. Appl Environ Microbiol. 71:7831-7837.

149 McCreary C, Martelli F, Grierson S, Ostanello F, Nevel A, Banks M. 2008. Excretion of hepatitis E virus by pigs of different ages and its presence in slurry stores in the United Kingdom. Vet Rec. 163:261-265.

150 Ziemer CJ, Bonner JM, Cole D, Vinje J, Constantini V, Goyal S, Gramer M, Mackie R, Meng XJ, Myers G, Saif LJ. 2010. Fate and transport of zoonotic, bacterial, viral, and parasitic pathogens during swine manure treatment, storage, and land application. J Anim Sci. 88:E84-94.

151 Kuno A, Ido K, Isoda N, Satoh Y, Ono K, Satoh S, Inamori H, Sugano K, Kanai N, Nishizawa T, Okamoto H. 2003. Sporadic acute hepatitis E of a 47-year-old man whose pet cat was positive for antibody to hepatitis E virus. Hepatol Res. 26:237-242.

152 Okamoto H, Takahashi M, Nishizawa T, Usui R, Kobayashi E. 2004. Presence of antibodies to hepatitis E virus in Japanese pet cats. Infection. 32:57-58.

153 Begum N, Polipalli SK, Husain SA, Kar P. 2010. Molecular analysis of swine hepatitis E virus from north India. Indian J Med Res. 132:504-508.

154 Arankalle VA, Chobe LP, Walimbe AM, Yergolkar PN, Jacob GP. 2003. Swine HEV infection in south India and phylogenetic analysis (1985-1999). J Med Virol. 69:391-396.

155 Cordoba L, Huang YW, Opriessnig T, Harral KK, Beach NM, Finkielstein CV, Emerson SU, Meng XJ. 2011. Three amino acid mutations (F51L, T59A, and S390L) in the capsid protein of the hepatitis E virus collectively contribute to virus attenuation. J Virol. 85:5338-5349.

156 Shukla P, Nguyen HT, Torian U, Engle RE, Faulk K, Dalton HR, Bendall RP, Keane FE, Purcell RH, Emerson SU. 2011. Cross-species infections of cultured cells by hepatitis E virus and discovery of an infectious virus-host recombinant. Proc Natl Acad Sci U S A. 108:2438-2443.

157 Shukla P, Nguyen HT, Faulk K, Mather K, Torian U, Engle RE, Emerson SU. 2012. Adaptation of a genotype 3 hepatitis E virus to efficient growth in cell culture depends on an inserted human gene segment acquired by recombination. J Virol. 86:5697-5707.

158 Nguyen HT, Torian U, Faulk K, Mather K, Engle RE, Thompson E, Bonkovsky HL, Emerson S. 2012. A naturally-occurring human/hepatitis E recombinant virus predominates in serum but not in faeces of a chronic hepatitis E patient and has a growth advantage in cell culture. J Gen Virol. 93:526- 530.

HEPATITIS E IN PREGNANCY: MATERNAL HEALTH

Lisa J. Krain, Kenrad E. Nelson, and Alain B. Labrique

EPIDEMIOLOGY OF HEPATITIS E IN PREGNANT WOMEN

Severe hepatitis E disproportionately affects pregnant women, with a high incidence of acute liver failure and reported case-fatality rates of ~15 to 30%. 1-3 This perplexing and grave feature, which is distinctive of hepatitis E epidemics, has been noted for decades (perhaps centuries) but remains poorly explained. 4 Recent work in Bangladesh suggests that acute hepatitis, with HEV the probable etiologic agent in a majority of cases, may underlie an astonishing ~10 to

25% of pregnancy-related deaths. 5-7 In addition to the effects on maternal health, fetal and newborn health may also be compromised, with elevated risk of miscarriage, stillbirth, premature delivery, and neonatal jaundice. 7-15 Vertical transmission appears to be common among mothers with symptomatic hepatitis E, with deleterious effects on fetal and neonatal health. 14, 16, 17 Because few studies have addressed the issue, it is unclear to what extent the fetuses and neonates of women with subclinical or asymptomatic infection are at risk. 18 There are not yet any reliable data on whether HEV is transmissible via breast milk. Geographic discrepancies in the frequency and severity of clinical hepatitis E in pregnant and postpartum women persist.

REGIONAL VARIATIONS IN OCCURRENCE AND SEVERITY OF HEPATITIS E IN PREGNANCY

Although severe liver disease among pregnant women, with high mortality, is the hallmark of epidemics of HEV-1 in Asia and Africa, there have been reports of severe hepatitis among pregnant women infected with HEV-3 in countries where locally acquired genotype 1 infections do not occur. Clinical hepatitis E attributed to HEV-3 was recently reported in a pregnant woman in

France. 19 HEV-3-linked acute liver failure also occurred in a nonpregnant Spanish woman whose medical history was devoid of known risk factors for severe hepatitis E. 20 The woman reported long-term use of hormonal contraceptives (norgestrel/ethinyl estradiol) and was found to have hepatic adenomas upon examination, which the authors speculated may reflect elevated estrogen levels, mimicking pregnancy, prior to the onset of hepatitis E. 20

Whether these reports suggest that pregnancy or pregnancy-like states may represent a risk factor for more severe clinical disease even in high-income countries where HEV-3 and HEV-4 circulate, or whether the paucity of such reports in the face of 10 to 20% population seroprevalence supports the increased virulence of genotype 1 viruses for pregnant women, is not certain. Severe infections during pregnancy in developing countries may be primarily a result of greater virulence of the human-adapted HEV-1 and HEV-2 genotypes, increased susceptibility of pregnant women in developing countries where these genotypes are endemic, or the epidemiologic conditions of exposure to these viruses. 21, 22 Also, in many developed- country settings, awareness of hepatitis E as a plausible diagnosis may be low, leading to missed cases and underdiagnosis of this infection.

Also unexplained is the contrast between the apparently infrequent and unremarkable occurrence of hepatitis E in women in Egypt 23 and the disproportionately severe hepatitis E observed in pregnant women sub-Saharan Africa and in South and Southeast Asia, despite the fact that HEV-1 predominates in all of these regions. 24 The different patterns of illness seen among pregnant women, paralleling regional differences in age-specific attack rates and seroprevalence, 25- 24

27 remain perplexing. [Table 1] The study of this range of possible outcomes to HEV infection is challenging, given that most HEV research occurs either in acute outbreak settings or in clinical contexts where only severely ill patients are typically referred. The prospective identification of

HEV seroconverters and the study of subsequent disease outcomes require the longitudinal surveillance and follow-up of large pregnancy cohorts – an expensive prospect for a pathogen that has yet to achieve standing even as a “neglected” disease.

IMMUNE FUNCTION AND SUSCEPTIBILITY TO VIRAL INFECTION DURING PREGNANCY

Pregnancy is associated with changes in sex hormone levels and immune system function

[Figure 3] that serve primarily to protect the fetus from attack by the maternal immune system but also exert strong effects on the immune response to pathogens. In addition to maternally mediated changes in the biochemical milieu, fetal hormones may also affect susceptibility to infection.

Pregnancy-related changes in body chemistry and fetal and placental influences have been variously linked to increases in the severity of certain infections, including Plasmodium falciparum malaria, leprosy, influenza, varicella, viral hemorrhagic fevers, and measles 28 and decreases in the severity of some autoimmune diseases, e.g., rheumatoid arthritis 29 during pregnancy.

During pregnancy, a shift from T helper cell type 1 (Th1)-dominated to T helper cell type 2

(Th2)-dominated immune responses, or “Th2 bias,” has been hypothesized to help protect the fetus by suppressing macrophage activation. 30 Th1-type responses are marked by increases in IFN-γ synthesis, and this type of response appears to help protect against parasitic infections. 31 Th2 responses, on the other hand, are tipped toward anti-inflammatory cytokines, such as IL-4, IL-6, and

IL-10, and increases in antibody production. 32, 33 These changes in the immunologic environment are not instantaneous, but vary with increasing estradiol and progesterone levels, 32 becoming most apparent as the pregnancy approaches term.

Extrapolation from serial measurements of IFN- γ and IL-6 (as biomarkers of Th1- and Th2- type responses, respectively) in 35 healthy pregnant women in Quebec suggested that Th1-type responses prevail until the mid-second trimester, with IFN- γ decreasing and IL-6 increasing from the 10th to 40th weeks of gestation. 34 Serial serum samples from a cohort of 50 healthy pregnant and postpartum women in New York City showed elevated levels of TNF-α throughout pregnancy compared with 6 months postpartum. 35 These women also experienced increases in levels of β- defensins and IP10 and decreases in IFN- γ during gestation. 35, 36

A study of 111 pregnant, 126 postpartum, and 86 nonpregnant healthy women in Japan indicated that natural killer (NK) cell activity was markedly reduced by the third trimester of pregnancy but relatively high during the first trimester and the first month postpartum. 37 Similarly, the Viral Immunity in Pregnancy study in New York City suggested a nadir in NK cell activity in the third trimester. 36 Caution is warranted, however, in generalizing biomarker findings from healthy pregnant women in industrialized countries to women in resource-limited settings.

HYPOTHESES REGARDING MEDIATION OF HEV SEVERITY IN PREGNANT WOMEN

The interplay of these factors in hepatitis E during pregnancy is complex and not yet well understood. Pal et al. 38 measured cytokine responses in cells from pregnant and nonpregnant women with HEV in India, as well as from healthy pregnant and nonpregnant controls. They found a Th2 bias among pregnant women with HEV, as well as an increased lymphoproliferative response to stimulation relative to that in healthy pregnant women. However, as their study was cross- sectional, infection-induced changes in cytokine and lymphoproliferative responses, as opposed to preexisting differences that predispose to disease, cannot be excluded.

In addition to mediating shifts in the immune system, pregnancy-associated hormones may also directly influence viral replication. 39 Poorer outcomes in HEV-infected pregnant patients have

been associated with higher viral load, 8, 40 but not consistently. 41 Elevated sex steroid hormones in women with HEV-associated acute liver failure, relative to those with milder illness, may suggest that higher estradiol and progesterone levels are a risk factor predisposing women to poorer outcomes; however, the elevated hormone levels may also be triggered by the infection state itself. 39

Host factors such as nutritional status, which may affect and be affected by pregnancy, may also contribute to the immune response to HEV infection in pregnant women. In contrast to the findings of Pal et al., 38 preliminary work in Bangladesh has suggested that the immune profiles of pregnant women who develop clinical hepatitis E may not reflect the Th1-Th2 shift observed in other cohorts of pregnant women, and it is possible that this is related to micronutrient deficiencies. 42 Pregnant women elsewhere in South Asia are often deficient in several micronutrients, and seasonal food availability may exert a major influence on maternal micronutrient status. 43 A proteome analysis conducted in a cohort of pregnant women in southern

Nepal identified changes from the first to third trimesters in the expression of proteins, some of which had not previously been reported in studies of healthy pregnant women. 44 Several proteins whose expression differed significantly between the first and third trimesters were associated with immune function and inflammation, including gelsolin, pregnancy zone, plasma protease C1 inhibitor, and various complement-related subcomponents, as well as other proteins whose functions remain uncertain. 44 Environmentally and genetically mediated differences in immune system modulation over the course of pregnancy could help explain regional differences in the observed hepatitis E incidence and severity during pregnancy.

Hepatitis E as a catalyst for coagulopathy may also contribute to maternal morbidity and mortality. Hemorrhage, often postpartum hemorrhage (PPH), is the leading proximal cause of maternal death in developing countries. 45, 46 If HEV infection disrupts clotting, this may further 27

increase the likelihood of uncontrolled bleeding during the peripartum period. As PPH remains one of the major causes of maternal mortality globally, especially in areas of endemicity, it may be that some proportion of peripartum hepatitis E is misclassified as delivery-associated PPH; this and other manifestations of HEV infections in pregnancy may explain the underrecognition of HEV etiologies in emergency obstetric complications.

Women admitted to a tertiary care hospital in New Delhi, India, with acute HEV (n = 132) had significantly elevated prothrombin times relative to those of women admitted for other acute viral hepatitides (n = 88). 10 The HEV-infected women had a greater incidence of antepartum (but not postpartum) hemorrhage, including gastrointestinal bleeding, and elevated maternal mortality.

In marked contrast to other studies, the women with acute HEV in this study were admitted with a lower average gestational age than women with other acute viral hepatitis. 10 Puri et al., 47 working in the same hospital in New Delhi in subsequent years, conducted a case-control study of third- trimester pregnant women with acute HEV and coagulopathy, comparing those who did (n = 13) and did not (n = 25) suffer postpartum bleeding. Women with postpartum hemorrhage had approximately five times the odds of having experienced hepatic encephalopathy as women without PPH and had 20 times the odds of a gastrointestinal bleed, though only gastrointestinal bleeding remained a significant clinical predictor of PPH after controlling for other factors. Fully a third of women admitted to the hospital with acute HEV infections experienced postpartum hemorrhage. 47

Work by Geng et al. 48 suggests possible mechanisms by which HEV infection may contribute to dysregulation of coagulation, which may pose a particular hazard to maternal health.

Their study of HEV-1, HEV-4, and rabbit HEV ORF3 protein interactions with human liver cell proteins found that HEV ORF3 may interact with several clotting-related pathways. Specifically, they propose that down-regulation of fibrinogen production may precipitate hemostasis, eventually 28

depleting clotting factors, and that several other protein pathways may also be disturbed. 48

Poliakova et al. 49 also observed dysregulated coagulation among HEV patients with acute liver failure, although their study did not focus specifically on pregnant women.

CHALLENGES FOR SURVEILLANCE AND PREVENTION

Despite emerging evidence explaining the role(s) of hepatitis E in maternal and neonatal morbidity and mortality, HEV-related deaths in pregnant women, especially if unaccompanied by frank jaundice, may be not be identified or recorded as such. Careful examination of the proximal and distal causes of maternal death, in clinical studies and, importantly, in community-based studies, may help elucidate the contribution of HEV infection and associated risk factors to maternal death. Maternal verbal autopsy methods used in both rural and urban Bangladesh have revealed a higher-than-anticipated prevalence (9.8 to 25%) of jaundice and/or acute viral hepatitis- like illnesses among women who died from all pregnancy-related causes during the maternal period. 5-7 Similar investigations, especially if supported by serologic and/or molecular confirmation of etiology, could help clarify the role of HEV in maternal death in other settings and suggest new avenues for prevention of maternal mortality.

The long-awaited release of a commercial vaccine against hepatitis E in China 50 is a hopeful sign for prevention of HEV-associated morbidity and mortality among pregnant and postpartum women. However, the vaccine has not yet been tested for safety and efficacy in large populations of pregnant women or in places where HEV genotype 1 is the primary cause of human infection. 6

Although a post hoc study of 34 pregnant women inadvertently included in the Chinese vaccine trial revealed no obvious differences in immunogenicity or safety, 51 the “hormonal and immunological milieu of pregnant women is strikingly different from that of women who are not pregnant”, 52 and

thus dedicated studies are needed to establish more clearly the risks, benefits, and optimal timing of vaccination in the context of pregnancy, for both the mother and her fetus. 52

TABLE AND FIGURE

Table 1. Seroprevalence of IgG antibodies to HEV among asymptomatic pregnant women recruited from community-based cohorts or antenatal clinics. Abbreviations: see table footnote b.

Figure 3. Changes in hormone levels and immune function are associated with an increased risk of severe hepatitis E in late pregnancy. (a) Changes in hormone levels and immune function over the course of pregnancy. Pregnancy is characterized by marked changes in hormone levels and corresponding shifts in immune status, which vary over the course of gestation. Levels of estradiol and progesterone are at their lowest early in pregnancy, while an increase in human chorionic gonadotropin (hCG) levels occurs and then begins to wane during the first trimester. As estradiol and progesterone levels continue to rise over the course of pregnancy, the dominant T helper 1 (Th1) immune regimen, associated with proinflammatory activity, is gradually superseded by T helper 2 (Th2)-biased responses associated with anti-inflammatory activity. Regulatory T-cell (Treg) activity also increases over the course of pregnancy. (Adapted fromFig.2 of reference 52 with permission from Elsevier.) (b) Distribution by trimester of hospital admissions of pregnant women with acute hepatitis during the historic Delhi hepatitis E virus (HEV) epidemic of 1954 to 1955. An association of greater hepatitis E severity with more advanced pregnancy is a frequent observation during outbreaks and among sporadic cases. Alterations in both hormone levels and immune function during pregnancy appear to contribute to these outcomes. Data are from reference 15. [Note that some early-pregnancy cases may not be classified as cases in pregnant women if the pregnancy is not detected, potentially biasing this apparent association to some extent.]

REFERENCES

1 Basnyat B. 2010. Neglected hepatitis E and typhoid vaccines. Lancet 376:869. doi:10.1016/S0140- 6736(10)61403-1.

2 Kumar A, Beniwal M, Kar P, Sharma JB, Murthy NS. 2004. Hepatitis E in pregnancy. Int. J. Gynaecol. Obstet. 85:240–244. doi:10.1016/j.ijgo.2003.11.018.

3 Rein DB, Stevens G, Theaker J, Wittenborn JS, Wiersma ST. 2012. The global burden of hepatitis E virus genotypes 1 and 2 in 2005. Hepatology 55:988–997. doi:10.1002/hep.25505.

4 Teo CG. 2012. Fatal outbreaks of jaundice in pregnancy and the epidemic history of hepatitis E. Epidemiol. Infect. 140:767–787. doi:10.1017/S0950268811002925.

5 Khatun F, Rasheed S, Moran AC, Alam AM, Shomik MS, Sultana M, Choudhury N, Iqbal M, Bhuiya A. 2012. Causes of neonatal and maternal deaths in Dhaka slums: implications for service delivery. BMC Public Health 12:84. doi:10.1186/1471-2458-12-84.

6 Labrique AB, Sikder SS, Krain LJ, West KP, Jr, Christian P, Rashid M, Nelson KE. 2012. Hepatitis E, a vaccine-preventable cause of maternal death. Emerg. Infect. Dis. 18:1401–1404. doi:10.3201/eid1809.120241.

7 Gurley ES, Halder AK, Streatfield PK, Sazzad HM, Nurul Huda TM, Hossain MJ, Luby SP. 2012. Estimating the burden of maternal and neonatal deaths associated with jaundice in Bangladesh: possible role of hepatitis E infection. Am. J. Public Health 102:2248–2254. doi:10.2105/AJPH.2012.300749.

8 Bose PD, Das BC, Kumar A, Gondal R, Kumar D, Kar P. 2011. High viral load and deregulation of the progesterone receptor signaling pathway: association with hepatitis E-related poor pregnancy outcome. J. Hepatol. 54:1107–1113. doi:10.1016/j.jhep.2010.08.037.

9 Ahmed RE, Karsany MS, Adam I. 2008. Acute viral hepatitis and poor maternal and perinatal outcomes in pregnant Sudanese women. J. Med.Virol. 80:1747–1748. doi:10.1002/jmv.21284.

10 Patra S, Kumar A, Trivedi SS, Puri M, Sarin SK. 2007. Maternal and fetal outcomes in pregnant women with acute hepatitis E virus infection. Ann. Intern. Med. 147:28–33. doi:10.7326/0003-4819- 147-1-200707030-00005.

11 Rasheeda CA, Navaneethan U, Jayanthi V. 2008. Liver disease in pregnancy and its influence on maternal and fetal mortality: a prospective study from Chennai, Southern India. Eur. J. Gastroenterol. Hepatol. 20:362–364. doi:10.1097/MEG.0b013e3282f246d6.

12 Rathi U, Bapat M, Rathi P, Abraham P. 2007. Effect of liver disease on maternal and fetal outcome—a prospective study. Indian J. Gastroenterol.26:59–63.

13 Tsega E, Hansson BG, Krawczynski K, Nordenfelt E. 1992. Acute sporadic viral hepatitis in Ethiopia: causes, risk factors, and effects on pregnancy. Clin. Infect. Dis. 14:961–965. doi:10.1093/clinids/14.4.961.

14 Khuroo MS, Kamili S. 2009. Clinical course and duration of viremia in vertically transmitted hepatitis E virus (HEV) infection in babies born to HEV-infected mothers. J. Viral Hepat. 16:519–523. doi:10.1111/j.1365-2893.2009.01101.x.

15 Naidu SS, Viswanathan R. 1957. Infectious hepatitis in pregnancy during Delhi epidemic. Indian J. Med. Res. 45:71–76.

16 Khuroo MS, Kamili S, Jameel S. 1995. Vertical transmission of hepatitis E virus. Lancet 345:1025– 1026. doi:10.1016/S0140-6736(95)90761-0.

17 Dahiya M, Kumar A, Kar P, Gupta RK. 2005. Acute viral hepatitis in third trimester of pregnancy. Indian J. Gastroenterol. 24:128–129.

18 Krain LJ, Atwell JA, Nelson KE, Labrique AB. 2014. Fetal and neonatal health consequences of vertically-transmitted hepatitis E virus infection. Am. J. Trop. Med. Hyg., 90(2): 365–370. doi:10.4269/ajtmh.13-0265

19 Anty R, Ollier L, Péron JM, Nicand E, Cannavo I, Bongain A, Giordanengo V, Tran A. 2012. First case report of an acute genotype 3 hepatitis E infected pregnant woman living in South-Eastern France. J Clin Virol. 54:76-78.

20 Mateos Lindemann ML, Morales JG, Fernández-Barredo S, Rodríguez-Domínguez M, García de la Hoz F, Halfon P, Pérez Gracia MT. 2010. Fulminant hepatitis E in a woman taking oral contraceptive medication. Am J Trop Med Hyg. 82:12-15.

21 Dalton HR, Bendall R, Ijaz S, Banks M. 2008. Hepatitis E: an emerging infection in developed countries. Lancet Infect Dis. 8:698-709.

22 Nelson KE, Kmush B, Labrique AB. 2011. The epidemiology of hepatitis E virus infections in developed countries and among immunocompromised patients. Expert Rev Anti Infect Ther. 9:1133- 1148.

23 Stoszek SK, Abdel-Hamid M, Saleh DA, El Kafrawy S, Narooz S, Hawash Y, Shebl FM, El Daly M, Said A, Kassem E, Mikhail N, Engle RE, Sayed M, Sharaf S, Fix AD, Emerson SU, Purcell RH, Strickland GT. 2006. High prevalence of hepatitis E antibodies in pregnant Egyptian women. Trans R Soc Trop Med Hyg. 100:95-101.

24 Navaneethan U, Al Mohajer M, Shata MT. 2008. Hepatitis E and pregnancy: understanding the pathogenesis. Liver Int. 28:1190-1199.

25 Stoszek SK, Engle RE, Abdel-Hamid M, Mikhail N, Abdel-Aziz F, Medhat A, Fix AD, Emerson SU, Purcell RH, Strickland GT. 2006. Hepatitis E antibody seroconversion without disease in highly endemic rural Egyptian communities. Trans R Soc Trop Med Hyg. 100:89-94.

26 Fix AD, Abdel-Hamid M, Purcell RH, Shehata MH, Abdel-Aziz F, Mikhail N, el Sebai H, Nafeh M, Habib M, Arthur RR, Emerson SU, Strickland GT. 2000. Prevalence of antibodies to hepatitis E in two rural Egyptian communities. Am J Trop Med Hyg. 62:519-523.

27 Meky FA, Stoszek SK, Abdel-Hamid M, Selim S, Abdel-Wahab A, Mikhail N, El-Kafrawy S, El-Daly M, Abdel-Aziz F, Sharaf S, Mohamed MK, Engle RE, Emerson SU, Purcell RH, Fix AD, Strickland GT. 2006. Active surveillance for acute viral hepatitis in rural villages in the Nile Delta. Clin Infect Dis. 42:628-633.

28 Jamieson DJ, Theiler RN, Rasmussen SA. 2006. Emerging infections and pregnancy. Emerg. Infect. Dis. 12:1638–1643. doi:10.3201/eid1211.060152.

29 Doria A, Iaccarino L, Arienti S, Ghirardello A, Zampieri S, Rampudda ME, Cutolo M, Tincani A, Todesco S. 2006. Th2 immune deviation induced by pregnancy: the two faces of autoimmune rheumatic diseases. Reprod. Toxicol. 22:234–241. doi:10.1016/j.reprotox.2006.04.001.

30 Romagnani S. 1997. The Th1/Th2 paradigm. Immunol. Today 18:263–266. doi:10.1016/S0167- 5699(97)01070-0.

31 Klein SL. 2004. Hormonal and immunological mechanisms mediating sex differences in parasite infection. Parasite Immunol. 26:247–264. doi:10.1111/j.0141-9838.2004.00710.x.

32 Robinson DP, Klein SL. 2012. Pregnancy and pregnancy-associated hormones alter immune responses and disease pathogenesis. Horm. Behav. 62:262–271. doi:10.1016/j.yhbeh.2012.02.023.

33 Vargas-Villavicencio JA, De León-Nava MA, Morales-Montor J. 2009. Immunoendocrine mechanisms associated with resistance or susceptibility to parasitic diseases during pregnancy. Neuroimmunomodulation 16:114–121. doi:10.1159/000180266.

34 Aris A, Lambert F, Bessette P, Moutquin JM. 2008. Maternal circulating interferon-gamma and interleukin-6 as biomarkers of Th1/Th2 immune status throughout pregnancy. J. Obstet. Gynaecol. Res. 34:7–11. doi:10.1111/j.1447-0756.2007.00676.x.

35 Kraus TA, Sperling RS, Engel SM, Lo Y, Kellerman L, Singh T, Loubeau M, Ge Y, Garrido JL, Rodríguez-García M, Moran TM. 2010. Peripheral blood cytokine profiling during pregnancy and post-partum periods. Am. J. Reprod. Immunol. 64:411–426. doi:10.1111/j.1600-0897.2010.00889.x.

36 Kraus TA, Engel SM, Sperling RS, Kellerman L, Lo Y, Wallenstein S, Escribese MM, Garrido JL, Singh T, Loubeau M, Moran TM. 2012. Characterizing the pregnancy immune phenotype: results of the Viral Immunity and Pregnancy (VIP) study. J. Clin. Immunol. 32:300–311. doi:10.1007/s10875- 011-9627-2.

37 Hidaka Y, Amino N, Iwatani Y, Kaneda T, Mitsuda N, Morimoto Y, Tanizawa O, Miyai K. 1991. Changes in natural killer cell activity in normal pregnant and postpartum women: increases in the first trimester and postpartum period and decrease in late pregnancy. J. Reprod. Immunol. 20:73–83. doi:10.1016/0165-0378(91)90024-K.

38 Pal R, Aggarwal R, Naik SR, Das V, Das S, Naik S. 2005. Immunological alterations in pregnant women with acute hepatitis E. J. Gastroenterol. Hepatol. 20:1094–1101. doi:10.1111/j.1440- 1746.2005.03875.x.

39 Jilani N, Das BC, Husain SA, Baweja UK, Chattopadhya D, Gupta RK, Sardana S, Kar P. 2007. Hepatitis E virus infection and fulminant hepatic failure during pregnancy. J. Gastroenterol. Hepatol. 22:676–682. doi:10.1111/j.1440-1746.2007.04913.x.

40 Kar P, Jilani N, Husain SA, Pasha ST, Anand R, Rai A, Das BC. 2008. Does hepatitis E viral load and genotypes influence the final outcome of acute liver failure during pregnancy? Am. J. Gastroenterol. 103:2495–2501. doi:10.1111/j.1572-0241.2008.02032.x.

41 Saravanabalaji S, Tripathy AS, Dhoot RR, Chadha MS, Kakrani AL, Arankalle VA. 2009. Viral load, antibody titers and recombinant open reading frame 2 protein-induced TH1/TH2 cytokines and cellular immune responses in self-limiting and fulminant hepatitis E. Intervirology 52:78–85. doi:10.1159/000214862.

42 Labrique AB, Klein S, Kmush B, Ali H, Engle R, Schulze K, West KP, Jr, Purcell RH, Nelson KE. 2012. Immunologic dysregulation and micronutrient deficiencies associated with risk of intrapartum hepatitis E infections in pregnant Bangladeshi women. FASEB J. 26:127.4.

43 Jiang T, Christian P, Khatry SK, Wu L, West KP, Jr. 2005. Micronutrient deficiencies in early pregnancy are common, concurrent, and vary by season among rural Nepali pregnant women. J. Nutr. 135:1106–1112.

44 Scholl PF, Cole RN, Ruczinski I, Gucek M, Diez R, Rennie A, Nathasingh C, Schulze K, Christian P, Yager JD, Groopman JD, West KP, Jr. 2012. Maternal serum proteome changes between the first and third trimester of pregnancy in rural southern Nepal. Placenta 33:424–432. doi:10.1016/j.placenta.2012.02.009.

45 Lozano R, Naghavi M, Foreman K, Lim S, Shibuya K, Aboyans V, et al. 2012. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 380:2095–2128. doi:10.1016/S0140-6736(12)61728-0.

46 Khan KS, Wojdyla D, Say L, Gulmezoglu AM, Van Look PF. 2006. WHO analysis of causes of maternal death: a systematic review. Lancet 367:1066–1074. doi:10.1016/S0140-6736(06)68397-9.

47 Puri M, Patra S, Singh P, Malhotra N, Trivedi SS, Sharma S, Kumar A, Sarin SK. 2011. Factors influencing occurrence of postpartum haemorrhage in pregnant women with hepatitis E infection and deranged coagulation profile. Obstet. Med. 4:108–112. doi:10.1258/om.2011.110031.

48 Geng Y, Yang J, Huang W, Harrison TJ, Zhou Y, Wen Z, Wang Y. 2013. Virus host protein interaction network analysis reveals that the HEV ORF3 protein may interrupt the blood coagulation process. PloS One 8:e56320. doi:10.1371/journal.pone.0056320.

49 Poliakova AM, Novozhenina EB, Danil’iants IL. 1994. The hemostatic system of patients with a severe course of viral hepatitis E. Ter Arkh 66:19–21.

50 Proffitt A. 2012. First HEV vaccine approved. Nat. Biotechnol. 30:300. doi:10.1038/nbt0412-300a.

51 Wu T, Zhu FC, Huang SJ, Zhang XF, Wang ZZ, Zhang J, Xia NS. 2012. Safety of the hepatitis E vaccine for pregnant women: a preliminary analysis. Hepatology 55:2038. doi:10.1002/hep.25522.

52 Klein SL, Jedlicka A, Pekosz A. 2010. The Xs and Y of immune responses to viral vaccines. Lancet Infect. Dis. 10:338–349. doi:10.1016/S1473-3099(10)70049-9.

53 Amer AF, Zaki SA, Nagati AM, Darwish MA. 1996. Hepatitis E antibodies in Egyptian adolescent females: their prevalence and possible relevance. J Egypt Public Health Assoc. 71:273-284.

54 Begum N, Devi SG, Husain SA, Kar P. 2009. Seroprevalence of subclinical HEV infection in pregnant women from north India: a hospital based study. Indian J Med Res. 130:709-713.

55 Surya IG, Kornia K, Suwardewa TG, Mulyanto, Tsuda F, Mishiro S. 2005. Serological markers of hepatitis B, C, and E viruses and human immunodeficiency virus type-1 infections in pregnant women in Bali, Indonesia. J Med Virol. 75:499-503.

56 Mizuo H, Suzuki K, Takikawa Y, Sugai Y, Tokita H, Akahane Y, Itoh K, Gotanda Y, Takahashi M, Nishizawa T, Okamoto H. 2002. Polyphyletic strains of hepatitis E virus are responsible for sporadic cases of acute hepatitis in Japan. J Clin Microbiol. 40:3209-3218.

57 Caron M, Kazanji M. 2008. Hepatitis E virus is highly prevalent among pregnant women in Gabon, central Africa, with different patterns between rural and urban areas. Virol J. 5:158.

58 Cevrioglu AS, Altindis M, Tanir HM, Aksoy F. 2004. Investigation of the incidence of hepatitis E virus among pregnant women in Turkey. J Obstet Gynaecol Res. 30:48-52.

59 Hannachi N, Hidar S, Harrabi I, Mhalla S, Marzouk M, Ghzel H, Ghannem H, Khairi H, Boukadida J. 2011. [Seroprevalence and risk factors of hepatitis E among pregnant women in central Tunisia]. Pathol Biol (Paris). 59:e115-118.

60 Traoré KA, Rouamba H, Nébié Y, Sanou M, Traoré AS, Barro N, Roques P. 2012. Seroprevalence of fecal-oral transmitted hepatitis A and E virus antibodies in Burkina Faso. PLoS One. 7:e48125.

61 Huang F, Ma T, Li L, Zeng W, Jing S. 2013. Low seroprevalence of hepatitis E virus infection in pregnant women in Yunnan, China. Braz J Infect Dis. [Online September 2013; doi:10.1016/j.bjid.2013.02.006]

62 Oncu S, Oncu S, Okyay P, Ertug S, Sakarya S. 2006. Prevalence and risk factors for HEV infection in pregnant women. Med Sci Monit. 12:CR36-39.

[This review is republished with the permission of the American Society of Tropical Medicine and Hygiene, from Krain et al., 2014, Fetal and Neonatal Health Consequences of Vertically-Transmitted Hepatitis E Virus Infection, Am. J. Trop. Med. Hyg., 90(2): 365–370, doi:10.4269/ajtmh.13-0265. Permission conveyed through Copyright Clearance Center, Inc. License #3916661390018.]

FETAL AND NEONATAL HEALTH CONSEQUENCES OF VERTICALLY-TRANSMITTED HEPATITIS E VIRUS INFECTION

Lisa J. Krain, Jessica E. Atwell, Kenrad E. Nelson, and Alain B. Labrique

ABSTRACT. Hepatitis E virus (HEV) infections lead to tens of thousands of deaths annually, mostly in developing countries. Hepatitis E poses a significant threat to the health of expectant mothers, a well-noted epidemiologic feature of the disease, but the contribution of vertically- transmitted HEV infection to fetal and neonatal morbidity and mortality has received limited attention. Evidence assembled to date suggests that mother-to-child HEV transmission may be frequent and deleterious to the fetus and newborn in pregnancies affected by hepatitis E. Further work is required to resolve key questions: 1. What risks do subclinical maternal HEV infections and infections early in pregnancy pose to fetal health and development? 2. Does vertical transmission occur during labor and/or breastfeeding and contribute appreciably to neonatal morbidity and mortality? 3. How do treatment decisions for severely-ill mothers affect fetal and neonatal outcomes? 4. Can maternal vaccination effectively prevent vertical transmission of HEV?

INTRODUCTION

Although hepatitis E virus (HEV) is sometimes referred to as an ‘emerging’ infectious agent,

HEV is well-established as a major cause of acute viral hepatitis (AVH) worldwide.1 Of the more than 20 million infections estimated to occur globally each year ~70,000 result in death.2 The vast majority of these deaths occur in resource-poor countries in Asia, Africa and Latin America, where exposure to fecally-contaminated water results in outbreaks and sporadic cases of hepatitis E.2,3

Hepatitis E in these locations is nearly always attributable to the human-associated HEV genotypes

1 (in Asia, Africa, and Latin America) and 2 (in sub-Saharan Africa and Mexico). However, the globally-ubiquitous zoonotic HEV strains (genotypes 3 and 4) have more recently been identified as a cause of sporadic hepatitis in medically-vulnerable patients and in the general population in high- income countries.4-14

High case-fatality ratios among pregnant women, particularly during the third trimester of pregnancy, remain an almost pathognomonic feature of hepatitis E epidemics caused by HEV genotype 1.15-21 There is mounting evidence that hepatitis E is an important contributor to maternal morbidity and mortality in South Asia, even outside of periodic large outbreaks.22-26 In the early literature on hepatitis E, there was much conjecture about the extent to which maternal hepatitis E threatened fetal and neonatal health, beyond catastrophic maternal illness or death. Emerging evidence from epidemiologic and clinical studies now suggests that vertical transmission of HEV may occur frequently among mothers ill with hepatitis E and contribute to serious perinatal health outcomes along with the effects of maternal morbidity and mortality.

HEPATITIS E AND MATERNAL-CHILD HEALTH

Epidemic jaundice marked by an excess of illness and death among pregnant women and their infants has been documented at least as far back as 18th century Europe.15,27 The first

(retrospectively) serologically-confirmed hepatitis E outbreak occurred in Delhi, India, in the mid-

1950s,17,28,29 although molecular evidence suggests that HEV may already have been circulating in humans for several hundred years.30,31 A hospital-based study during the Delhi epidemic documented a ~10% maternal case-fatality rate, along with miscarriage, stillbirth, or neonatal death in 56% of infants of women with HEV infection. In addition, jaundice was reported among both expiring and surviving infants.16

Numerous studies conducted over the three decades since HEV was identified as a cause of infectious hepatitis32 have continued to support high rates of maternal, fetal, and neonatal illness and death in affected pregnancies. According to one recent model, HEV may be responsible for

~2,400-3,000 stillbirths each year in developing countries,2,33 with many additional fetal deaths linked to antenatal maternal mortality.34 Pre-term delivery in mothers with hepatitis E is common and is associated with poorer neonatal survival.33,35 During a 2002 outbreak in the Central African

Republic, all of the pregnant women with serologically-confirmed hepatitis E (n=7) delivered prematurely; three of these babies were stillborn (one macerated), and another died within minutes of delivery.36 Newborns whose mothers had acute hepatitis at the time of delivery comprised half (4/8) of the fatalities in a 1993-1994 outbreak in Islamabad, Pakistan.21 Pregnant women with jaundice during a 2008-2009 hepatitis E outbreak in Tongi, Bangladesh, were more than twice as likely as non-jaundiced pregnant women to miscarry or deliver a stillborn baby.37 In two separate, hospital-based prospective studies in New Delhi and Chennai, India, ~15% to >50% of the live-born infants of mothers with hepatitis E died within the first week postpartum.33,38

During a 2010-2011 outbreak in Sudan, 14 intrauterine deaths and 9 premature deliveries were reported among 39 pregnant hepatitis E cases.39

To date, only hepatitis E caused by HEV genotype 1 has consistently been observed to yield these effects in pregnancy. However, HEV genotype 2 was implicated in acute liver failure in a 40

pregnant woman during an outbreak in Namibia,40 and the potential of genotypes 2-4 to cause adverse outcomes in pregnant women, given exposure, remains uncertain.

MISCARRIAGE AND STILLBIRTH: THE ROLE OF ANTENATAL HEV INFECTION

Despite the ample epidemiologic evidence that maternal hepatitis E may result in adverse consequences to the fetus, few studies have been able to address whether these outcomes are solely the result of maternal health complications, or whether fetal HEV infection also plays a role.

Etiologic studies of pregnancy loss and stillbirth are complicated by numerous logistical, technical, cultural, and ethical obstacles.41-44 The prospective design, sample size and frequency of perinatal observations and follow-up would require considerable resources and advanced clinical and research infrastructure. As a result, the task of demonstrating a contribution of fetal HEV infection to intrauterine or intrapartum death has rarely been attempted.

Early research in China failed to detect HEV antigen (HEV Ag) in any of 17 stillborn fetuses whose mothers were infected during a 1986-1988 outbreak.45 In a case-control study in Egypt,

HEV Ag was detected in 5% of aborted fetal tissue samples, but absent from the cord blood of live- born, full-term infants.46 Anti-HEV IgM was detected in 3%, and HEV RNA in 16% of mothers who aborted, but in none of the mothers who delivered live infants.46 While these Egyptian results support an association of HEV with abortion, they fall short of implicating antenatal HEV transmission to the fetus as a cause of fetal death.

Results in other animals have been similarly inconclusive. Experimental inoculation studies of HEV genotype 1 in pregnant rhesus macaques47,48 and of genotype 3 in swine49 have failed to provide clear evidence of vertical transmission or of adverse pregnancy outcomes attributable to

HEV. Of four pregnant macaques inoculated with an epidemic HEV strain isolated from an Indian patient in 1990, three delivered healthy infants (two of them following elevation of liver enzymes),

while the remaining macaque delivered a macerated fetus five days prior to the onset of hepatitis.48

Consistent with other experimental studies of HEV involving animal models,50 the disease provoked in these pregnant animals was less severe than that seen in most pregnant women who come to clinical attention.

Investigation of adverse gestational outcomes among naturally-infected animals has also yielded results that are challenging to interpret. HEV genotype 3 RNA was amplified from the livers of ~16% of aborted porcine fetuses on two South Korean farms.51 Concomitant detection of porcine circovirus-2 (PCV2) in all of these fetal pig samples casts doubt on the role of HEV in these deaths, but also raises the important question of whether coinfection with immunosuppressive viruses may facilitate fetal infection with HEV.51

In South and Central Asia and sub-Saharan Africa, where genotypes 1 and 2 predominate, no human studies have yet systematically evaluated the association of fetal HEV infection with miscarriage or stillbirth. Therefore, it remains unclear whether some of the increased risk of miscarriage and stillbirth reported in these diverse settings may be attributable to vertically- transmitted infection or whether it is the result of maternal complications of hepatitis E alone.

HEV INFECTION AND OUTCOMES AMONG LIVE-BORN INFANTS

Morbidity and mortality among neonates born to mothers with hepatitis E may be explained, to a large extent, by preterm delivery and other pre- and perinatal stresses caused by the maternal response to infection. However, a small body of evidence suggests that vertically- transmitted infections also contribute directly to infant morbidity and mortality in the early postnatal period.

A landmark study published in 1995 by Khuroo and colleagues was the first to document mother-to-child transmission of HEV using serologic and molecular methods.52 HEV RNA was 42

found in the cord blood of 5/8 Kashmiri infants whose mothers had serologic evidence of infection preceding delivery.52 All five of these infants had elevated alanine aminotransferase (ALT) at birth.

Two infants died within a day of delivery.52 Serum viremia, IgM antibodies, and hepatitis persisting for several weeks in two of the surviving infants suggested that these results could not be explained solely by contamination of cord blood with maternal blood, and that vertically-transmitted HEV could, in fact, cause illness in neonates.52

Subsequent studies have found a similarly high prevalence of vertically-transmitted HEV infection [Table 2],53-57 which frequently, but not always, results in disease. Hepatitis, either icteric or anicteric, may be present from birth in a substantial proportion of neonates born to mothers with hepatitis E.58,59 Recently, a second-trimester fetal HEV infection associated with ascites was reported to have resolved in utero, resulting in a healthy infant who was born at 38 weeks of gestation.60 However, evidence of severe necrosis in liver tissue samples from neonatal autopsies suggests that some babies, like their mothers, experience fulminant hepatic failure (FHF) as a result of HEV infection.53

In contrast to perinatally-transmitted hepatitis B virus infections, which may be lifelong if preventive measures are not initiated at birth,61 there have been no reports of persistent HEV infection in infants born to mothers with hepatitis E.62 This is consistent with the natural history of hepatitis E in adults, which is typically self-limiting.63 Chronic HEV infections have been reported primarily in immunosuppressed and immunocompromised populations.9,63-82 They have not been documented in pregnant women or infants, though some authors have speculated that among immunocompromised pregnant women or neonates, such infections could, plausibly, be of concern.83 In addition, pregnant women may be coinfected with other hepatotropic pathogens,54,56,84-89 but how such coinfections influence vertical HEV transmission and outcomes has not been studied. 43

CHALLENGES IN UNDERSTANDING MOTHER-TO-CHILD TRANSMISSION OF HEV

Limited surveillance and reporting have presented obstacles to understanding the consequences of HEV infections on maternal, fetal, and neonatal outcomes. Misclassification of HEV infection due to barriers to medical care,90,91 failure to consider hepatitis E in differential diagnosis,92-95 or the use of insensitive assays,96 may obscure the impact of HEV on pregnancy outcomes.

In addition, study populations in the reported literature are largely hospital-based and skewed towards women with more severe illness and thus a predisposition to having worse fetal and/or neonatal outcomes. Nonetheless, results from the largest prospective studies of women with hepatitis E paint a picture of relatively poor pregnancy outcomes even in women with milder illness, and not only among those with acute liver failure.33,97

The extent to which asymptomatic infections influence pregnancy outcomes has not yet been studied systematically. In addition, there are no reliable data on whether HEV can be transmitted through breast milk. Population-based serologic surveillance of pregnant women and follow-up of pregnancy and neonatal outcomes may help to address these issues. Furthermore, studies are needed across a wider variety of settings. Differences in viral characteristics,98 exposure patterns,99 underlying population health and nutrition status,100 host genetics,101 and other host factors may modify the effects of maternal HEV infection on fetal and neonatal outcomes.

Further investigation of the mechanisms of HEV pathogenesis in pregnant women would help in understanding the role of transplacental transmission in fetal loss and stillbirth. The timing of HEV infection relative to pregnancy may also be a critical variable. Although the preponderance of severe hepatitis E in pregnant women occurs during the third trimester, how this observed maternal response relates to vertical transmission and fetal viability has not yet been elucidated.

It has been suggested that in utero fetal infection may itself contribute to adverse maternal outcomes.53 Such “upside-down” vertical effects have been posited to occur with other viral infections as well. Studies of murine gammaherpesvirus 68 in mice, for example, have found that fetal inflammatory responses to viral replication in the placenta may predispose the mother to morbidity and reduce the mother’s capacity for sustaining the pregnancy.102,103 Hormone receptor- moderated inflammatory responses at the feto-maternal interface have been associated with pregnancy outcomes in human hepatitis E.97

There are currently no adequate treatments for hepatitis E in pregnancy. Experimental use of ribavirin has recently shown promise in treating severe, acute hepatitis E in non-pregnant patients,104-106 but this drug is typically contraindicated in pregnancy due to “significant embryocidal and/or teratogenic effects” and fetal harm.107 Some researchers have suggested expedited delivery or pregnancy termination could be considered to preserve the life of the mother.53,108 Whether this approach would prevent death in women who present with severe disease has not been studied systematically. Given the high rates of miscarriage, stillbirth, and premature delivery in pregnancies affected by severe hepatitis E, the net impact of such a strategy on neonatal morbidity and mortality is also uncertain.

New HEV vaccines show promise in preventing hepatitis E,109,110 and they may help obviate the need for treatment of severe illness. However, follow-up data indicate that HEV infections can still occur among vaccinated adults,111 and only incidental data on safety and efficacy in pregnant women are available.112 Evaluating the effectiveness of these vaccines in preventing maternal disease and death, and in reducing the burden of fetal loss, premature delivery, and neonatal morbidity and mortality, should stand among global maternal-child health priorities.

TABLE

Study Live HEV Hepatitis E Icteric Neonatal Characteristics births infections cases hepatitis E deaths tested in neonates in neonates cases in # neonates # (% of live # (% of live # (% of live # (% of live births) births) births) births) Srinagar, India51: 33 22 (67%) 20 mothers with 22 (67%) 20 (61%) 11 (33%) AVH, 16 with FHF ≥17 RNA+; ≥21 IgM+ New Delhi, 18 6 (33%) no data no data no data India52(see also 55): 19 mothers with 4 RNA+; 3 AVH, 9 with FHF IgM+ New Delhi, 6 3 (50%) 1 (17%) 1 (17%) 0 (0%) India56: 8 mothers with 3 RNA+ AVH, 14 with FHF Accra, Ghana57: 1 1 (100%) 1 (100%) 1 (100%) 0 (0%) 1 mother with AVH, 2 with FHF

Table 2. Vertically-transmitted hepatitis E virus infection, morbidity, and mortality in live-born infants of mothers with laboratory-confirmed antenatal HEV infection.

AVH = acute viral hepatitis due to HEV, FHF = fulminant hepatic failure (acute liver failure) due to HEV. *“Laboratory-confirmed antenatal HEV infection” means detection in maternal serum of anti-HEV IgM, HEV RNA, or both, at any time during the pregnancy or at delivery; most cases occurred in the 3rd trimester.

Note: Among mothers, serological testing was performed for HAV, HBV, and HCV infections in the Srinigar and New Delhi-based studies. No coinfections were reported by Khuroo and colleagues in Srinigar; 2 HBV + HEV coinfections and 1 HCV + HEV coinfection were reported by Kumar and colleagues and by Singh and colleagues in each of the two different New Delhi-based studies. The mother of the tested neonate in the Ghana study by Bonney and colleagues was initially suspected to have a malaria + viral hepatitis coinfection, but confirmation of a malaria diagnosis was not reported. Aside from the Ghana study, none of the papers provided sufficient detail to assign specific outcomes to the infants of mothers with coinfections.

Search strategy and selection criteria

We searched the PubMed database over the date range 1966 - September 2012 using the

Medical Subject Headings (MeSH) terms [("Infant, Newborn" OR "Fetus" OR "Infectious Disease

Transmission, Vertical" OR "pregnancy outcome") AND ("Hepatitis E" OR "Hepatitis E virus")].

We performed free-text searches for variants of these terms as well as “jaundice” and “non-A,

non-B hepatitis” in both PubMed and Google Scholar. After screening abstracts for broad

relevance, we downloaded full-text papers and manually searched the references to identify

additional sources. Several historic papers known to the authors were included for context.

New papers received via topical HEV publication alerts from PubMed, and several suggested by

reviewers, were also added. Although we did not explicitly restrict our search to English-

language papers, all of the papers included were published in English. In this paper, we survey

the broader literature, but emphasize those mammalian studies in which both maternal and

fetal/neonatal diagnoses or outcome investigations were supported by immunologic and/or

virologic assays.

Box 4.1. Search strategy and selection criteria.

REFERENCES

1 Labrique, A. B., Kuniholm, M. H. & Nelson, K. E., 2010. The Global Impact of Hepatitis E: New Horizons for an Emerging Virus. Ch. 4 in Emerging Infections, Vol. 9 (eds W. M. Scheld, M. L. Grayson, & J. M. Hughes), pp.53-88. ASM Press.

2 Rein, D. B., Stevens, G., Theaker, J., Wittenborn, J. S. & Wiersma, S. T., 2012. The global burden of hepatitis E virus genotypes 1 and 2 in 2005. Hepatology 55(4): 988-997, doi:10.1002/hep.25505.

3 Teshale, E. H., Hu, D. J. & Holmberg, S. D., 2010. The two faces of hepatitis E virus. Clin Infect Dis 51(3): 328-334, doi:10.1086/653943.

4 Meng, X. J., 2011. From barnyard to food table: the omnipresence of hepatitis E virus and risk for zoonotic infection and food safety. Virus Res 161(1): 23-30, doi:10.1016/j.virusres.2011.01.016.

5 Teo, C. G., 2010. Much meat, much malady: changing perceptions of the epidemiology of hepatitis E. Clin Microbiol Infect 16(1): 24-32, doi:10.1111/j.1469-0691.2009.03111.x.

6 Kuniholm, M. H., Purcell, R. H., McQuillan, G. M., Engle, R. E., Wasley, A. & Nelson, K. E., 2009. Epidemiology of hepatitis E virus in the United States: results from the Third National Health and Nutrition Examination Survey, 1988-1994. J Infect Dis 200(1): 48-56, doi:10.1086/599319.

7 Colson, P., Borentain, P., Queyriaux, B., Kaba, M., Moal, V., Gallian, P., Heyries, L., Raoult, D. & Gérolami, R., 2010. Pig liver sausage as a source of hepatitis E virus transmission to humans. J Infect Dis 202(6): 825-834, doi:10.1086/655898.

8 Dalton, H. R., Bendall, R., Ijaz, S. & Banks, M., 2008. Hepatitis E: an emerging infection in developed countries. Lancet Infect Dis 8(11): 698-709, doi:10.1016/S1473-3099(08)70255-X.

9 Nelson, K. E., Kmush, B. & Labrique, A. B., 2011. The epidemiology of hepatitis E virus infections in developed countries and among immunocompromised patients. Expert Rev Anti Infect Ther 9(12): 1133-1148, doi:10.1586/eri.11.138.

10 Matsubayashi, K., Kang, J. H., Sakata, H., Takahashi, K., Shindo, M., Kato, M., Sato, S., Kato, T., Nishimori, H., Tsuji, K., Maguchi, H., Yoshida, J., Maekubo, H., Mishiro, S. & Ikeda, H., 2008. A case of transfusion-transmitted hepatitis E caused by blood from a donor infected with hepatitis E virus via zoonotic food-borne route. Transfusion 48(7): 1368-1375, doi:10.1111/j.1537- 2995.2008.01722.x.

11 Pischke, S., Suneetha, P. V., Baechlein, C., Barg-Hock, H., Heim, A., Kamar, N., Schlue, J., Strassburg, C. P., Lehner, F., Raupach, R., Bremer, B., Magerstedt, P., Cornberg, M., Seehusen, F., Baumgaertner, W., Klempnauer, J., Izopet, J., Manns, M. P., Grummer, B. & Wedemeyer, H., 2010. Hepatitis E virus infection as a cause of graft hepatitis in liver transplant recipients. Liver Transpl 16(1): 74-82, doi:10.1002/lt.21958.

12 Sakata, H., Matsubayashi, K., Takeda, H., Sato, S., Kato, T., Hino, S., Tadokoro, K. & Ikeda, H., 2008. A nationwide survey for hepatitis E virus prevalence in Japanese blood donors with elevated alanine aminotransferase. Transfusion 48(12): 2568-2576, doi:10.1111/j.1537-2995.2008.01910.x.

13 Takahashi, M., Nishizawa, T., Yoshikawa, A., Sato, S., Isoda, N., Ido, K., Sugano, K. & Okamoto, H., 2002. Identification of two distinct genotypes of hepatitis E virus in a Japanese patient with acute hepatitis who had not travelled abroad. J Gen Virol 83(Pt 8): 1931-1940. 48

14 Xia, H., Wahlberg, N., Belak, S., Meng, X. J. & Liu, L., 2011. The emergence of genotypes 3 and 4 hepatitis E virus in swine and humans: a phylogenetic perspective. Arch Virol 156(1): 121-124, doi:10.1007/s00705-010-0818-6.

15 Teo, C. G., 2012. Fatal outbreaks of jaundice in pregnancy and the epidemic history of hepatitis E. Epidemiol Infect 1-21, doi:10.1017/s0950268811002925.

16 Naidu, S. S. & Viswanathan, R., 1957. Infectious hepatitis in pregnancy during Delhi epidemic. Indian J Med Res 45(Suppl.): 71-76.

17 Viswanathan, R., 1957. Certain epidemiological features of infectious hepatitis during the Delhi epidemic, 1955-1956, in Hepatitis Frontiers (Henry Ford Hospital International Symposium), eds. Hartman, F. W., G. A. LoGrippo, J. G. Mateer, & J. Barron. pp. 207-210 (Churchill, Detroit, Michigan).

18 Khuroo, M. S., Teli, M. R., Skidmore, S., Sofi, M. A. & Khuroo, M. I., 1981. Incidence and severity of viral hepatitis in pregnancy. Am J Med 70(2): 252-255.

19 Khatun, F., Rasheed, S., Moran, A. C., Alam, A. M., Shomik, M. S., Sultana, M., Choudhury, N., Iqbal, M. & Bhuiya, A., 2012. Causes of neonatal and maternal deaths in Dhaka slums: implications for service delivery. BMC Public Health 12 84, doi:10.1186/1471-2458-12-84.

20 Boccia, D., Guthmann, J. P., Klovstad, H., Hamid, N., Tatay, M., Ciglenecki, I., Nizou, J. Y., Nicand, E. & Guerin, P. J., 2006. High mortality associated with an outbreak of hepatitis E among displaced persons in Darfur, Sudan. Clin Infect Dis 42(12): 1679-1684, doi:10.1086/504322.

21 Rab, M. A., Bile, M. K., Mubarik, M. M., Asghar, H., Sami, Z., Siddiqi, S., Dil, A. S., Barzgar, M. A., Chaudhry, M. A. & Burney, M. I., 1997. Water-borne hepatitis E virus epidemic in Islamabad, Pakistan: a common source outbreak traced to the malfunction of a modern water treatment plant. Am J Trop Med Hyg 57(2): 151-157.

22 Labrique, A. B., Sikder, S. S., Krain, L. J., West, K. P., Jr., Christian, P., Rashid, M. & Nelson, K. E., 2012. Hepatitis E, a Vaccine-Preventable Cause of Maternal Death. Emerg Infect Dis 18(9): 1401- 1404, doi:10.3201/eid1809.120241.

23 Khuroo, M. S., Duermeyer, W., Zargar, S. A., Ahanger, M. A. & Shah, M. A., 1983. Acute sporadic non-A, non-B hepatitis in India. Am J Epidemiol 118(3): 360-364.

24 Khuroo, M. S. & Kamili, S., 2003. Aetiology, clinical course and outcome of sporadic acute viral hepatitis in pregnancy. J Viral Hepat 10(1): 61-69, doi:398 [pii].

25 Kumar, S., Ratho, R. K., Chawla, Y. K. & Chakraborti, A., 2007. The incidence of sporadic viral hepatitis in North India: a preliminary study. Hepatobiliary Pancreat Dis Int 6(6): 596-599.

26 Gurley, E. S., Halder, A. K., Streatfield, P. K., Sazzad, H. M., Nurul Huda, T. M., Hossain, M. J. & Luby, S. P., 2012. Estimating the Burden of Maternal and Neonatal Deaths Associated With Jaundice in Bangladesh: Possible Role of Hepatitis E Infection. Am J Public Health, doi:10.2105/ajph.2012.300749.

27 Purcell, R. H. & Ticehurst, J., 1988. Enterically transmitted non-A, non-B hepatitis: epidemiology and clinical characteristics. In Viral Hepatitis and Liver Disease (ed A. J. Zuckerman) pp. 131-137. New York: Alan R. Liss.

28 Wong, D. C., Purcell, R. H., Sreenivasan, M. A., Prasad, S. R. & Pavri, K. M., 1980. Epidemic and endemic hepatitis in India: evidence for a non-A, non-B hepatitis virus aetiology. Lancet 2(8200): 876-879.

29 Arankalle, V. A., Chadha, M. S., Tsarev, S. A., Emerson, S. U., Risbud, A. R., Banerjee, K. & Purcell, R. H., 1994. Seroepidemiology of water-borne hepatitis in India and evidence for a third enterically- transmitted hepatitis agent. Proc Natl Acad Sci U S A 91(8): 3428-3432.

30 Purdy, M. A. & Khudyakov, Y. E., 2010. Evolutionary history and population dynamics of hepatitis E virus. PLoS One 5(12): e14376, doi:10.1371/journal.pone.0014376.

31 Purdy, M. A. & Khudyakov, Y. E., 2011. The molecular epidemiology of hepatitis E virus infection. Virus Res 161(1): 31-39, doi:10.1016/j.virusres.2011.04.030.

32 Balayan, M. S., Andjaparidze, A. G., Savinskaya, S. S., Ketiladze, E. S., Braginsky, D. M., Savinov, A. P. & Poleschuk, V. F., 1983. Evidence for a virus in non-A, non-B hepatitis transmitted via the fecal-oral route. Intervirology 20(1): 23-31.

33 Patra, S., Kumar, A., Trivedi, S. S., Puri, M. & Sarin, S. K., 2007. Maternal and fetal outcomes in pregnant women with acute hepatitis E virus infection. Ann Intern Med 147(1): 28-33, doi: 10.7326/0003-4819-147-1-200707030-00005.

34 Beniwal, M., Kumar, A., Kar, P., Jilani, N. & Sharma, J. B., 2003. Prevalence and severity of acute viral hepatitis and fulminant hepatitis during pregnancy: a prospective study from north India. Indian J Med Microbiol 21(3): 184-185.

35 Tsega, E., Hansson, B. G., Krawczynski, K. & Nordenfelt, E., 1992. Acute sporadic viral hepatitis in Ethiopia: causes, risk factors, and effects on pregnancy. Clin Infect Dis 14(4): 961-965.

36 Goumba, C. M., Yandoko-Nakouné, E. R. & Komas, N. P., 2010. A fatal case of acute hepatitis E among pregnant women, Central African Republic. BMC Res Notes 3 103, doi:10.1186/1756-0500-3- 103.

37 International Centre for Diarrhoeal Disease Research, Bangladesh, 2009. Outbreak of hepatitis E in a low income urban community in Bangladesh. ICDDR,B Health Sci Bull 7(3): 14-20.

38 Rasheeda, C. A., Navaneethan, U. & Jayanthi, V., 2008. Liver disease in pregnancy and its influence on maternal and fetal mortality: a prospective study from Chennai, Southern India. Eur J Gastroenterol Hepatol 20(4): 362-364, doi:10.1097/MEG.0b013e3282f246d6.

39 Rayis, D. A., Jumaa, A. M., Gasim, G. I., Karsany, M. S. & Adam, I., 2013. An outbreak of hepatitis E and high maternal mortality at Port Sudan, Eastern Sudan. Pathog Glob Health 107(2): 66-68, doi:10.1179/2047773213y.0000000076.

40 Maila, H. T., Bowyer, S. M., and Swanepoel, R., 2004. Identification of a new strain of hepatitis E virus from an outbreak in Namibia in 1995. J Gen Virol 85: 89-95.

41 Goldenberg, R. L. & Thompson, C., 2003. The infectious origins of stillbirth. Am J Obstet Gynecol 189(3): 861-873, doi:10.1067/s0002-9378(03)00470-8.

42 Goldenberg, R. L., McClure, E. M., Saleem, S. & Reddy, U. M., 2010. Infection-related stillbirths. Lancet 375(9724): 1482-1490, doi:10.1016/s0140-6736(09)61712-8.

43 Kelley, M., Rubens, C. E. & the GAPPS Review Group, 2010. Global report on preterm birth and stillbirth (6 of 7): ethical considerations. BMC Pregnancy Childbirth 10 Suppl 1 S6, doi:10.1186/1471- 2393-10-s1-s6.

44 Lawn, J. E., Blencowe, H., Pattinson, R., Cousens, S., Kumar, R., Ibiebele, I., Gardosi, J., Day, L. T., Stanton, C. for the Lancet's Stillbirths Series Steering Committee, 2011. Stillbirths: Where? When? Why? How to make the data count? Lancet 377(9775): 1448-1463, doi:10.1016/s0140- 6736(10)62187-3.

45 Zhai, Q., Zhang, L. F., Liang, R. M., Zhang, Y. H., Sun, S. Y. & Cao, X. Y., 1994. Immunohistochemical study of hepatitis E virus antigen in liver tissues from patients with hepatitis E. Chin Med J (Engl) 107(3): 222-224.

46 El-Esnawy, N. A., Ali, M. A., Bayoumi, F. S., Abo-El-Khir, A. & Abdel-Wahab, K. S., 2001. Waterborne viruses associated with repeated abortion. J Egypt Public Health Assoc 76(5-6): 487-503.

47 Tsarev, S. A., Tsareva, T. S., Emerson, S. U., Rippy, M. K., Zack, P., Shapiro, M. & Purcell, R. H., 1995. Experimental hepatitis E in pregnant rhesus monkeys: failure to transmit hepatitis E virus (HEV) to offspring and evidence of naturally acquired antibodies to HEV. J Infect Dis 172(1): 31-37.

48 Arankalle, V. A., Chadha, M. S., Banerjee, K., Srinivasan, M. A. & Chobe, L. P., 1993. Hepatitis E virus infection in pregnant rhesus monkeys. Indian J Med Res 97 4-8.

49 Kasorndorkbua, C., Thacker, B. J., Halbur, P. G., Guenette, D. K., Buitenwerf, R. M., Royer, R. L. & Meng, X. J., 2003. Experimental infection of pregnant gilts with swine hepatitis E virus. Can J Vet Res 67(4): 303-306.

50 Krawczynski, K., Meng, X. J. & Rybczynska, J., 2011. Pathogenetic elements of hepatitis E and animal models of HEV infection. Virus Res 161(1): 78-83, doi:10.1016/j.virusres.2011.03.007.

51 Hosmillo, M., Jeong, Y. J., Kim, H. J., Park, J. G., Nayak, M. K., Alfajaro, M. M., Collantes, T. M., Park, S. J., Ikuta, K., Yunoki, M., Kang, M. I., Park, S. I. & Cho, K. O., 2010. Molecular detection of genotype 3 porcine hepatitis E virus in aborted fetuses and their sows. Arch Virol 155(7): 1157-1161, doi:10.1007/s00705-010-0679-z.

52 Khuroo, M. S., Kamili, S. & Jameel, S., 1995. Vertical transmission of hepatitis E virus. Lancet 345(8956): 1025-1026.

53 Khuroo, M. S., 2006. Association of severity of hepatitis E virus infection in the mother and vertically transmitted infection in the fetus. JK-Practitioner 13(2): 70-74.

54 Kumar, A., Beniwal, M., Kar, P., Sharma, J. B. & Murthy, N. S., 2004. Hepatitis E in pregnancy. Int J Gynaecol Obstet 85(3): 240-244, doi:10.1016/j.ijgo.2003.11.018.

55 Dahiya, M., Kumar, A., Kar, P. & Gupta, R. K., 2005. Acute viral hepatitis in third trimester of pregnancy. Indian J Gastroenterol 24(3): 128-129.

56 Singh, S., Mohanty, A., Joshi, Y. K., Deka, D., Mohanty, S. & Panda, S. K., 2003. Mother-to-child transmission of hepatitis E virus infection. Indian J Pediatr 70(1): 37-39.

57 Bonney, J. H. M., Kwame-Aryee, R. A. D., Obed, S. D., Tamatey, A. A. D., Barnor, J. S. D., Armah, N. B. D., Oppong, S. A. D. & Osei-Kwasi, M. D., 2012. Fatal hepatitis E viral infection in pregnant women in Ghana: a case series. BMC Res Notes 5(1): 478, doi:10.1186/1756-0500-5-478. 51

58 Khuroo, M. S. & Kamili, S., 2009. Clinical course and duration of viremia in vertically transmitted hepatitis E virus (HEV) infection in babies born to HEV-infected mothers. J Viral Hepat 16(7): 519- 523, doi:10.1111/j.1365-2893.2009.01101.x.

59 Mansoor, M., Raza, H. & Tariq, R., 2011. Feto-Maternal Outcome in HEV Infection. Annals of King Edward Medical University 17(1).

60 Pradhan, M., Anand, B. & Singh, A., 2012. Hepatitis E Virus Infection Causing Isolated Fetal Ascites: A Case Report. Fetal Diagn Ther, doi:10.1159/000337612.

61 Beasley, R. P., Hwang, L. Y., Lee, G. C., Lan, C. C., Roan, C. H., Huang, F. Y. & Chen, C. L., 1983. Prevention of perinatally transmitted hepatitis B virus infections with hepatitis B virus infections with hepatitis B immune globulin and hepatitis B vaccine. Lancet 2(8359): 1099-1102.

62 Begum, N., Polipalli, S. K., Husain, S. A., Kumar, A. & Kar, P., 2010. Duration of hepatitis E viremia in pregnancy. Int J Gynaecol Obstet 108(3): 207-210, doi:10.1016/j.ijgo.2009.09.023.

63 Aggarwal, R., 2011. Clinical presentation of hepatitis E. Virus Res 161(1): 15-22, doi:10.1016/j.virusres.2011.03.017.

64 Gérolami, R., Moal, V. & Colson, P., 2008. Chronic hepatitis E with cirrhosis in a kidney-transplant recipient. N Engl J Med 358(8): 859-860, doi:10.1056/NEJMc0708687.

65 Kamar, N., Selves, J., Mansuy, J. M., Ouezzani, L., Péron, J. M., Guitard, J., Cointault, O., Esposito, L., Abravanel, F., Danjoux, M., Durand, D., Vinel, J. P., Izopet, J. & Rostaing, L., 2008. Hepatitis E virus and chronic hepatitis in organ-transplant recipients. N Engl J Med 358(8): 811-817, doi:10.1056/NEJMoa0706992.

66 Lockwood, G. L., Fernandez-Barredo, S., Bendall, R., Banks, M., Ijaz, S. & Dalton, H. R., 2008. Hepatitis E autochthonous infection in chronic liver disease. Eur J Gastroenterol Hepatol 20(8): 800- 803, doi:10.1097/MEG.0b013e3282f1cbff.

67 Haagsma, E. B., van den Berg, A. P., Porte, R. J., Benne, C. A., Vennema, H., Reimerink, J. H. & Koopmans, M. P., 2008. Chronic hepatitis E virus infection in liver transplant recipients. Liver Transpl 14(4): 547-553, doi:10.1002/lt.21480.

68 Dalton, H. R., Bendall, R. P., Keane, F. E., Tedder, R. S. & Ijaz, S., 2009. Persistent carriage of hepatitis E virus in patients with HIV infection. N Engl J Med 361(10): 1025-1027, doi:10.1056/NEJMc0903778.

69 Kamar, N., Abravanel, F., Selves, J., Garrouste, C., Esposito, L., Lavayssière, L., Cointault, O., Ribes, D., Cardeau, I., Nogier, M. B., Mansuy, J. M., Muscari, F., Peron, J. M., Izopet, J. & Rostaing, L., 2010. Influence of immunosuppressive therapy on the natural history of genotype 3 hepatitis-E virus infection after organ transplantation. Transplantation 89(3): 353-360, doi:10.1097/TP.0b013e3181c4096c.

70 Legrand-Abravanel, F., Kamar, N., Sandres-Saune, K., Garrouste, C., Dubois, M., Mansuy, J. M., Muscari, F., Sallusto, F., Rostaing, L. & Izopet, J., 2010. Characteristics of autochthonous hepatitis E virus infection in solid-organ transplant recipients in France. J Infect Dis 202(6): 835-844, doi:10.1086/655899.

71 Renou, C., Lafeuillade, A., Cadranel, J. F., Pavio, N., Pariente, A., Allègre, T., Poggi, C., Pénaranda, G., Cordier, F., & Nicand, E., for the ANGH, 2010. Hepatitis E virus in HIV-infected patients. AIDS 24(10): 1493-1499, doi:10.1097/QAD.0b013e32833a29ab.

72 Halac, U., Béland, K., Lapierre, P., Patey, N., Ward, P., Brassard, J., Houde, A. & Alvarez, F., 2011. Chronic hepatitis E infection in children with liver transplantation. Gut 61(4):597-603, doi:10.1136/gutjnl-2011-300708.

73 Dalton, H. R., Keane, F. E., Bendall, R., Mathew, J. & Ijaz, S., 2011. Treatment of chronic hepatitis E in a patient with HIV infection. Ann Intern Med 155(7): 479-480, doi:10.1059/0003-4819-155-7- 201110040-00017.

74 Kamar, N., Garrouste, C., Haagsma, E. B., Garrigue, V., Pischke, S., Chauvet, C., Dumortier, J., Cannesson, A., Cassuto-Viguier, E., Thervet, E., Conti, F., Lebray, P., Dalton, H. R., Santella, R., Kanaan, N., Essig, M., Mousson, C., Radenne, S., Roque-Afonso, A. M., Izopet, J. & Rostaing, L., 2011. Factors associated with chronic hepatitis in patients with hepatitis E virus infection who have received solid organ transplants. Gastroenterology 140(5): 1481-1489, doi:10.1053/j.gastro.2011.02.050.

75 Jagjit Singh, G. K., Ijaz, S., Rockwood, N., Farnworth, S. P., Devitt, E., Atkins, M., Tedder, R. & Nelson, M., 2011. Chronic hepatitis E as a cause for cryptogenic cirrhosis in HIV. J Infect 66(1):103- 106, doi:10.1016/j.jinf.2011.11.027.

76 Colson, P., Dhiver, C., Poizot-Martin, I., Tamalet, C. & Gérolami, R., 2011. Acute and chronic hepatitis E in patients infected with human immunodeficiency virus. J Viral Hepat 18(3): 227-228, doi:10.1111/j.1365-2893.2010.01311.x.

77 Motte, A., Roquelaure, B., Galambrun, C., Bernard, F., Zandotti, C. & Colson, P., 2012. Hepatitis E in three immunocompromized children in southeastern France. J Clin Virol 53(2): 162-166, doi:10.1016/j.jcv.2011.11.012.

78 Halac, U., Béland, K., Lapierre, P., Patey, N., Ward, P., Brassard, J., Houde, A. & Alvarez, F., 2012. Cirrhosis due to Chronic Hepatitis E Infection in a Child Post-Bone Marrow Transplant. J Pediatr 160(5): 871-874 e871, doi:10.1016/j.jpeds.2012.01.028.

79 Pischke, S., Stiefel, P., Franz, B., Bremer, B., Suneetha, P. V., Heim, A., Ganzenmueller, T., Schlue, J., Horn-Wichmann, R., Raupach, R., Darnedde, M., Scheibner, Y., Taubert, R., Haverich, A., Manns, M. P., Wedemeyer, H. & Bara, C. L., 2012. Chronic Hepatitis E in Heart Transplant Recipients. Am J Transplant 12(11): 3128-3133, doi:10.1111/j.1600-6143.2012.04200.x.

80 Lhomme, S., Abravanel, F., Dubois, M., Sandres Saune, K., Rostaing, L., Kamar, N. & Izopet, J., 2012. Hepatitis E Virus Quasispecies and the Outcome of Acute Hepatitis E in Solid-Organ Transplant Patients. J Virol 86(18):10006-10014, doi:10.1128/jvi.01003-12.

81 Scobie, L. & Dalton, H. R., 2013. Hepatitis E: source and route of infection, clinical manifestations and new developments. J Viral Hepat 20(1): 1-11, doi:10.1111/jvh.12024.

82 Kamar, N., Rostaing, L. & Izopet, J., 2013. Hepatitis e virus infection in immunosuppressed patients: natural history and therapy. Semin Liver Dis 33(1): 62-70, doi:10.1055/s-0033-1338115.

83 Caron, M., Bouscaillou, J. & Kazanji, M., 2012. Acute risk for hepatitis E virus infection among HIV- 1-positive pregnant women in central Africa. Virol J 9: 254, doi:10.1186/1743-422x-9-254. 53

84 Jaiswal, S. B., Chitnis, D. S., Asolkar, M. V., Naik, G. & Artwani, K. K., 1996. Aetiology and prognostic factors in hepatic failure in central India. Trop Gastroenterol 17(4): 217-220.

85 Jaiswal, S. P., Jain, A. K., Naik, G., Soni, N. & Chitnis, D. S., 2001. Viral hepatitis during pregnancy. Int J Gynaecol Obstet 72(2): 103-108.

86 Jilani, N., Das, B. C., Husain, S. A., Baweja, U. K., Chattopadhya, D., Gupta, R. K., Sardana, S. & Kar, P., 2007. Hepatitis E virus infection and fulminant hepatic failure during pregnancy. J Gastroenterol Hepatol 22(5): 676-682, doi:10.1111/j.1440-1746.2007.04913.x.

87 Kar, P., Jilani, N., Husain, S. A., Pasha, S. T., Anand, R., Rai, A. & Das, B. C., 2008. Does hepatitis E viral load and genotypes influence the final outcome of acute liver failure during pregnancy? Am J Gastroenterol 103(10): 2495-2501, doi:10.1111/j.1572-0241.2008.02032.x.

88 Strand, R. T., Franque-Ranque, M., Bergstrom, S. & Weiland, O., 2003. Infectious aetiology of jaundice among pregnant women in Angola. Scand J Infect Dis 35(6-7): 401-403.

89 Tsega, E., Krawczynski, K., Hansson, B. G. & Nordenfelt, E., 1993. Hepatitis E virus infection in pregnancy in Ethiopia. Ethiop Med J 31(3): 173-181.

90 Sikder, S. S., Labrique, A. B., Ullah, B., Ali, H., Rashid, M., Mehra, S., Jahan, N., Shamim, A. A., West, K. P. & Christian, P., 2011. Accounts of severe acute obstetric complications in rural Bangladesh. BMC Pregnancy Childbirth 11: 76, doi:10.1186/1471-2393-11-76.

91 Filippi, V., Richard, F., Lange, I. & Ouattara, F., 2009. Identifying barriers from home to the appropriate hospital through near-miss audits in developing countries. Best Pract Res Clin Obstet Gynaecol 23(3): 389-400, doi:10.1016/j.bpobgyn.2008.12.006.

92 Hamid, S. S., Jafri, S. M., Khan, H., Shah, H., Abbas, Z. & Fields, H., 1996. Fulminant hepatic failure in pregnant women: acute fatty liver or acute viral hepatitis? J Hepatol 25(1): 20-27.

93 Aggarwal, R., 2003. Hepatic encephalopathy in pregnancy. Indian J Gastroenterol 22 (Suppl 2): S78- 80.

94 Jayanthi, V. & Udayakumar, N., 2008. Acute liver failure in pregnancy: an overview. Minerva Gastroenterol Dietol 54(1): 75-84.

95 Devarbhavi, H., Kremers, W. K., Dierkhising, R. & Padmanabhan, L., 2008. Pregnancy-associated acute liver disease and acute viral hepatitis: differentiation, course and outcome. J Hepatol 49(6): 930-935, doi:10.1016/j.jhep.2008.07.030.

96 Mast, E. E., Alter, M. J., Holland, P. V. & Purcell, R. H., 1998. Evaluation of assays for antibody to hepatitis E virus by a serum panel. Hepatitis E Virus Antibody Serum Panel Evaluation Group. Hepatology 27(3): 857-861, doi:10.1002/hep.510270331.

97 Bose, P. D., Das, B. C., Kumar, A., Gondal, R., Kumar, D. & Kar, P., 2011. High viral load and deregulation of the progesterone receptor signaling pathway: association with hepatitis E-related poor pregnancy outcome. J Hepatol 54(6): 1107-1113, doi:10.1016/j.jhep.2010.08.037.

98 Mishra, N., Walimbe, A. & Arankalle, V., 2012. Hepatitis E virus from India exhibits significant amino acid mutations in fulminant hepatic failure patients. Virus Genes 1-7, doi:10.1007/s11262- 012-0833-7.

99 Stoszek, S. K., Abdel-Hamid, M., Saleh, D. A., El Kafrawy, S., Narooz, S., Hawash, Y., Shebl, F. M., El Daly, M., Said, A., Kassem, E., Mikhail, N., Engle, R. E., Sayed, M., Sharaf, S., Fix, A. D., Emerson, S. U., Purcell, R. H. & Strickland, G. T., 2006. High prevalence of hepatitis E antibodies in pregnant Egyptian women. Trans R Soc Trop Med Hyg 100(2): 95-101, doi:10.1016/j.trstmh.2004.12.005.

100 Labrique, A. B., Zaman, K., Hossain, Z., Saha, P., Yunus, M., Hossain, A., Ticehurst, J. R. & Nelson, K. E., 2010. Epidemiology and risk factors of incident hepatitis E virus infections in rural Bangladesh. Am J Epidemiol 172(8): 952-961, doi:10.1093/aje/kwq225.

101 Mishra, N. & Arankalle, V. A., 2011. Association of polymorphisms in the promoter regions of TNF- alpha (-308) with susceptibility to hepatitis E virus and TNF-alpha (-1031) and IFN-gamma (+874) genes with clinical outcome of hepatitis E infection in India. J Hepatol 55(6): 1227-1234, doi:10.1016/j.jhep.2011.03.023.

102 Cardenas, I., Means, R. E., Aldo, P., Koga, K., Lang, S. M., Booth, C. J., Manzur, A., Oyarzun, E., Romero, R. & Mor, G., 2010. Viral infection of the placenta leads to fetal inflammation and sensitization to bacterial products predisposing to preterm labor. J Immunol 185(2): 1248-1257, doi:10.4049/jimmunol.1000289.

103 Cardenas, I., Mor, G., Aldo, P., Lang, S. M., Stabach, P., Sharp, A., Romero, R., Mazaki-Tovi, S., Gervasi, M. & Means, R. E., 2011. Placental viral infection sensitizes to endotoxin-induced pre-term labor: a double hit hypothesis. Am J Reprod Immunol 65(2): 110-117, doi:10.1111/j.1600- 0897.2010.00908.x.

104 Gerolami, R., Borentain, P., Raissouni, F., Motte, A., Solas, C. & Colson, P., 2011. Treatment of severe acute hepatitis E by ribavirin. J Clin Virol 52(1): 60-62, doi:10.1016/j.jcv.2011.06.004.

105 Karim, F., Rahman, S., Al-Mahtab, M. & Ahmed, F., 2012. Successful treatment of severe hepatitis E with ribavirin: a case report from Bangladesh. Euroasian J Hepato-Gastroenterol 2(1): 56-57, doi:10.5005/jp-journals-10018-1034.

106 Pischke, S., Hardtke, S., Bode, U., Birkner, S., Chatzikyrkou, C., Kauffmann, W., Bara, C. L., Gottlieb, J., Wenzel, J., Manns, M. P. & Wedemeyer, H., 2013. Ribavirin treatment of acute and chronic hepatitis E: a single-centre experience. Liver Int 33(5): 722-726, doi:10.1111/liv.12114.

107 State of Florida Department of Health Central Pharmacy, 2010. In National Library of Medicine (NLM) DailyMed Database, 2010. [http://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=3b8b0c65-e466-4662-9e85- 5a253d31ebe8]

108 Navaneethan, U., Al Mohajer, M. & Shata, M. T., 2008. Hepatitis E and pregnancy: understanding the pathogenesis. Liver Int 28(9): 1190-1199, doi:10.1111/j.1478-3231.2008.01840.x.

109 Shrestha, M. P., Scott, R. M., Joshi, D. M., Mammen, M. P., Jr., Thapa, G. B., Thapa, N., Myint, K. S., Fourneau, M., Kuschner, R. A., Shrestha, S. K., David, M. P., Seriwatana, J., Vaughn, D. W., Safary, A., Endy, T. P. & Innis, B. L., 2007. Safety and efficacy of a recombinant hepatitis E vaccine. N Engl J Med 356(9): 895-903, doi:10.1056/NEJMoa061847.

110 Zhu, F. C., Zhang, J., Zhang, X. F., Zhou, C., Wang, Z. Z., Huang, S. J., Wang, H., Yang, C. L., Jiang, H. M., Cai, J. P., Wang, Y. J., Ai, X., Hu, Y. M., Tang, Q., Yao, X., Yan, Q., Xian, Y. L., Wu, T., Li, Y. M., Miao, J., Ng, M. H., Shih, J. W. & Xia, N. S., 2010. Efficacy and safety of a recombinant hepatitis E

vaccine in healthy adults: a large-scale, randomised, double-blind placebo-controlled, phase 3 trial. Lancet 376(9744): 895-902, doi:10.1016/S0140-6736(10)61030-6.

111 Zhang, J., Zhang, X. F., Zhou, C., Wang, Z. Z., Huang, S. J., Yao, X., Liang, Z. L., Wu, T., Li, J. X., Yan, Q., Yang, C. L., Jiang, H. M., Huang, H. J., Xian, Y. L., Shih, J. W. K., Ng, M. H., Li, Y. M., Wang, J. Z., Zhu, F. C. & Xia, N. S., 2013. Protection against hepatitis E virus infection by naturally aquired and vaccine incude immunity. Clin Microbiol Infect, doi:10.1111/1469-0691.12419.

112 Wu, T., Zhu, F. C., Huang, S. J., Zhang, X. F., Wang, Z. Z., Zhang, J. & Xia, N. S., 2011. Safety of the hepatitis E vaccine for pregnant women: A preliminary analysis. Hepatology 55(6): 2038, doi:10.1002/hep.25522.

STUDY BACKGROUND AND EVOLUTION

MOTIVATIONS AND FRAMEWORK

The preceding chapters have provided most of the context for the work described in the remainder of this dissertation, i.e.: hepatitis E is a major public health problem in developing countries and its effects in pregnancy may be especially serious, but many unanswered questions remain about global-scale, regional, and interpersonal variations in hepatitis E occurrence and outcomes.

That the era of hepatitis E vaccines had recently dawned – at least in theory – was another major motivating factor for initiation of the study. Two candidate vaccines had recently been tested and found efficacious in controlled trials,1,2 but neither had been made available to high-risk populations in South Asia. The fact that the first vaccine was tested in Nepali army conscripts1 and subsequently shelved added what many observers viewed as an ethical insult to the unmitigated injury of high HEV-associated morbidity and mortality in the region.3 We expected that the results of our work would help strengthen the evidence base (and perhaps galvanize the international resolve) needed to evaluate hepatitis E vaccine as a means for preventing maternal, fetal, and neonatal illness and death across South Asia.

An ongoing study in southern Nepal provided the ability to collect data from a large cohort of Nepali women prospectively, taking advantage of a well-established field research infrastructure.

Johns Hopkins University investigators have a long history of work in Sarlahi District. The Nepal

Nutrition Intervention Project Sarlahi (NNIPS) was established by Johns Hopkins in 1989 to investigate the potential reductions in mortality that obtained by Vitamin A supplementation in children.4 The research and organizational infrastructure have since been expanded to undertake work on a variety of maternal and child health concerns, and the organization is well-respected in

Sarlahi. Working within the organization and with the ongoing Mother’s Gift Maternal Influenza

Immunization Field Trial (MaGIFT) would allow us to obtain detailed demographic, behavioral, and biological information over time.

In this way, not only would we be able to establish baseline prevalence and incidence of hepatitis E in pregnant women in this part of Nepal, but we could also begin to examine risk factors for infection and explore potential mediators and moderators of illness severity and the association of HEV infections with pregnancy outcomes. [Figure 4] Based on estimates from Bangladesh and preliminary data from an earlier phase of the MaGIFT study in Nepal, we expected to observe a large number of cases and of infections, giving us at least modest power to conduct nested case- control analyses of immunologic, hormonal, and nutritional biomarkers that might be associated with predisposition to better or poorer outcomes.

ONE DOOR CLOSES…

Once the cohort study finally got off the ground, it proceeded apace, with a few minor hiccups and changes of protocol. I spent a cumulative six months in the field, variously accompanying field staff to interviews and blood draws, helping to implement a biosafety / good clinical practice initiative [see Appendix 3], checking study forms, and practicing my limited Nepali with many patient residents of the town of Hariyon.

The study began winding down in the spring of 2014, bringing with it the very welcome revelation that not very many women had become seriously ill with hepatitis, and the corresponding (somewhat more challenging) realization that there would be far fewer cases for our nested biomarker studies than anticipated.

No sooner had I returned home from my final stint in Sarlahi than reports of a jaundice outbreak began appearing in Nepali news media,5,6 this time in Biratnagar, a large Terai city on the 58

Indian border, a few districts to the east. A return trip to investigate the outbreak first-hand was contemplated and contacts were made, but the trip was shelved at the last minute due to logistical and political concerns, and an inability to secure the necessary IRB approvals within a short timeframe. However, one of my contacts in Biratnagar directed a Nepali virologist who had been involved in the outbreak response to Johns Hopkins, and a collaboration was born. Undeterred by bureaucratic and administrative delays and logistical challenges, not to mention the magnitude 7.87 earthquake that devastated Nepal on April 25, 2015, we finally met in Baltimore in the summer of

2015 to analyze specimens obtained during the 2014 outbreak.

The remaining chapters detail the conduct and results of the cohort study in Sarlahi and offer additional description of, and perspective on, the Biratnagar outbreak.

FIGURE

Figure 4. Conceptual framework for the study of hepatitis E viral infections among pregnant women in Nepal.

REFERENCES

1 Shrestha MP, Scott RM, Joshi DM, Mammen MP, Jr., Thapa GB, Thapa N, et al. (2007) Safety and efficacy of a recombinant hepatitis E vaccine. N Engl J Med 356(9): 895-903. [doi:10.1056/NEJMoa061847]

2 Zhu FC, Zhang J, Zhang XF, Zhou C, Wang ZZ, Huang SJ, et al. (2010) Efficacy and safety of a recombinant hepatitis E vaccine in healthy adults: a large-scale, randomised, double-blind placebo- controlled, phase 3 trial. Lancet 376(9744): 895-902. [doi:10.1016/S0140-6736(10)61030-6]

3 Basu S, Andrews J, & Smith-Rohrberg D. (2006) Populations who test drugs should benefit from them. Nature 440(7084): 605. [doi:10.1038/440605d]

4 Pradhan EK, Katz J, LeClerq SC, & West KP, Jr. (1994) Data management for large community trials in Nepal. Control Clin Trials 15(3): 220-234.

5 Subedi B. (2014) Viral fevers spreading in Biratnagar. Republica 27 April 2014. [http://www.myrepublica.com/portal/index.php?action=news_details&news_id=73649#sthash.vvH wkAg2.dpuf Accessed 17 February 2016].

6 Himalayan Times. (2014) Government plans to combat jaundice. Himalayan Times 4 May 2014 Accessed 25 March 2016].

7 United States Geological Survey. (2015) M7.8 - 36km E of Khudi, Nepal. [http://earthquake.usgs.gov/earthquakes/eventpage/us20002926#general Accessed 8 June 2016].

HEPATITIS E VIRUS IN PREGNANT WOMEN IN SARLAHI DISTRICT, NEPAL

Lisa J. Krain

ABSTRACT

In South Asia, hepatitis E virus (HEV) has recently been recognized as an important contributor to maternal morbidity and to maternal, fetal, and infant mortality. A community-based, observational cohort study was conducted in Sarlahi District, Nepal, from 2012-2014, using existing clinical trial infrastructure to enroll 2363 pregnant women and follow them through delivery or three months postpartum. Active symptom surveillance and serologic testing were used to detect sporadic acute “hepatitis-like” illnesses and subclinical or asymptomatic viral hepatitis infections.

Pregnant women in Sarlahi have low immunity to HEV and are vulnerable to hepatitis E during pregnancy. Baseline anti-HEV IgG seroprevalence was 11.7% [95% confidence interval (CI): 10.2%-

13.3%] and varied with age, village, caste/ethnicity, land ownership, and diet (meat consumption).

Water sources and latrine access were not associated with seropositivity. Incidence of HEV seroconversion was 32.3/1000 person-years (p*y) during pregnancy [95% CI: 14.8 to 61.3/1000 p*y] and 44.3/1000 p*y overall [95% CI: 27.1 to 68.4/1000 p*y]. Postpartum seroreversion was common following HEV seroconversion in pregnancy. One acute case of hepatitis E due to genotype

1 HEV was confirmed by serology, but most “hepatitis-like” illnesses could not be attributed to acute hepatitis A, B, C, or E. Jaundice of uncertain etiology was linked to one of two maternal deaths during the study period. Mid-to-late pregnancy HEV infection was not clearly associated with adverse pregnancy outcomes, but study design and sample size limited detection. Serologic monitoring beginning prior to (or earlier in) pregnancy and longer postpartum follow-up would permit improved surveillance of both pregnancy outcomes and antibody kinetics. Environmental testing could help clarify exposure pathways and seasonal patterns of infection in the Terai.

BACKGROUND

Community-based studies have recently established that sporadic hepatitis caused by the hepatitis E virus (HEV) are an important contributor to maternal morbidity and mortality in

Bangladesh.1,2 Nearby, Nepal is no stranger to enterically-transmitted hepatitis outbreaks, nor their devastating consequences for some expectant mothers.3-10(cf. Biratnagar chapter) The untimely death of a pregnant woman – one of many – from acute liver failure during a 1987 outbreak ultimately helped to demonstrate the etiologic role of hepatitis E virus in the recurrent Kathmandu Valley epidemics of enterically-transmitted non-A-non-B hepatitis.11

Still, hepatitis E-related illness and death in Nepali pregnant women, especially in the lull between epidemics, has rarely been more than a blip on a very busy maternal health radar. A handful of hospital-based studies in Kathmandu have called attention to the poor outcomes of severely ill pregnant women with acute hepatitis E12 and attempted to quantify the role of hepatitis

E in severe maternal illness and death. 13,14 These hospital-based studies and case series capture some of the most concerning outcomes associated with hepatitis E, but provide only a limited glimpse of the broader epidemiology of HEV infection.

Previous population-based studies were conducted primarily within the densely-populated

Kathmandu Valley. Results generally suggest high population exposure to HEV among adults, with recent seroprevalence estimates near or above 50%.15,16 Several older studies were conducted in the late 1990s in advance of the contentious17-19 Phase II testing in Nepal of a vaccine20 that was promptly abandoned by its commercial sponsor,21 despite its demonstrated efficacy. As Nepali

Army and Police conscripts were offered up as study subjects, those data also have a decidedly male skew.22,23

A new recombinant vaccine,24 tested and produced in China, has been approved there since

2012.25,26 Though this vaccine currently lacks World Health Organization pre-qualification for use in other countries, there have been calls to expedite the process in order to address public health crises.27 The vaccine has been subject to limited safety and efficacy testing in pregnant women,28,29 but pregnant women are among the potential targets if the vaccine is cleared for global use.

With this context in mind, and with the benefit of an established field study site in southern

Nepal, we undertook this study to begin to address outstanding issues:

 What is the burden of hepatitis E virus infections during pregnancy in a community setting

in Nepal, considering a range of maternal, fetal, and neonatal outcomes?

 Do documented rates of HEV infection and illness in the Kathmandu Valley reflect the

epidemiologic patterns elsewhere in Nepal?

 What factors place women at risk of – or protect against – infection, disease, and death?

METHODS

Study Setting - MaGIFT

The Mother’s Gift Maternal Influenza Immunization Field Trial (MaGIFT; NCT01034254) was conducted in 9 village development committees (VDCs) of Sarlahi District, in the Terai

(lowlands) of Nepal. MaGIFT and the Hepatitis Substudy were carried out in the field through the

Johns Hopkins University-affiliated Nepal Nutrition Intervention Project – Sarlahi (NNIPS) and its field headquarters in Hariyon, Sarlahi District. [Figure 5]

The nine VDCs where the study took place had a combined population of just under 100,000 people in 2011.30 The study area has few improved roads; however, it is bisected by the East-West

(Mahendra) Highway, the major road traversing the Nepali Terai. In May 2014, several VDCs in the study area were merged and officially reclassified as municipalities as part of an urbanization initiative.31

From April 2012 through April 2013, MaGIFT participants were individually randomized to receive influenza vaccine or saline placebo between 17-34 weeks of gestation. At the end of this controlled trial phase, all participants received the active vaccine. Women with new pregnancies identified from late April 2013 onward were also offered immunization (without randomization, at scheduled clinics) and followed through delivery or the end of the follow-up period in May 2014.

The hepatitis substudy spanned the period from October 2012 until the end of follow-up in May

2014 and was open to all women eligible for participation in MaGIFT.

Study Enrollment

Married women ages 15-40 living in the 9 participating VDCs of Sarlahi District were identified through a baseline community census conducted in 2011 and ongoing reports by field workers assigned to sectors in each ward of the participating VDCs. With consent, field data collectors visited these married women on a 5-week schedule and offered pregnancy tests to women who reported a missed menstrual period.

Field data collectors reported pregnancies to the field headquarters in Hariyon. Each pregnant woman was then visited by a trained team leader-interviewer (TLI). Eligible women (≤34 weeks gestation, planned to deliver within the study area, had not received current flu vaccine, no allergies to vaccine components, not previously a MaGIFT participant) were invited to join MaGIFT and the hepatitis substudy. [Appendix 1]

Upon obtaining informed consent, [Appendix 2, Form 66] the TLI administered questionnaires concerning the woman’s health and reproductive histories, routine behaviors, and

living situation. Additional questions about possible hepatitis-related risk factors focused on water, sanitation, and hygiene practices; occupation; food preparation and consumption; travel history; and animal exposures. [Appendix 2, Form 79] All women were offered basic antenatal care packages, including iron supplements, deworming tablets, and a safe birthing kit.

Routine Blood Collection (Enrollment, Delivery, and Postpartum)

Auxiliary nurse midwives (ANMs) trained in phlebotomy collected ~5mL of venous blood from participants by at each of three event-contingent time-points: upon enrollment, after delivery, and at 3 months postpartum. Specimens were labeled and placed on ice packs in insulated coolers and transported from the field by motorcycle to the field laboratory in Hariyon within 4-6 hours of collection. At the field laboratory, specimens were centrifuged, and ~2.5mL of serum derived from each blood sample was aliquoted into four cryovials. These cryovials were labeled and placed in liquid nitrogen shippers for storage and shipment.

The aliquots were batched and transported in liquid nitrogen dewars to the Walter

Reed/AFRIMS Unit Nepal (WARUN) in Kathmandu. The first set of aliquots was used at WARUN to conduct serologic testing for the hepatitis substudy. The other aliquots were shipped in liquid nitrogen shippers to Cincinnati Children’s Hospital for distribution to MaGIFT investigators in the

United States. [Appendix 1]

Symptom Surveillance and Follow-Up of Acute Hepatitis Cases

As part of MaGIFT, field workers visited women weekly to ask about a limited set of symptoms. After enrollment, each woman was visited by a TLI at 30-day intervals to ask about broader symptoms of illness and any care sought, using the Monthly Pregnancy Follow-Up Form

(PFF) [Appendix 2, Form 14]. The PFF recorded the number of days in the past 30 that a symptom was experienced, along with a code for type of provider (e.g., traditional healer, health post, medical

clinic) if care was sought. To adapt the PFF for the Hepatitis Substudy in this phase of MaGIFT, additional symptoms relevant to hepatitis were included.

Hepatitis-related symptoms (nausea or vomiting, fever, yellow eyes or skin, dark urine or light stools, upper abdominal pain, or strange/unappealing smell to food) reported at monthly visits yielded a weighted symptom score, calculated on the PFF. Scores above the threshold of 4, indicating a presumptive case of acute hepatitis, were reported by the interviewer to the field office by telephone at the end of the visit. [Table 3]

Upon receipt of a telephone report of a possible hepatitis case, auxiliary nurse midwives

(ANMs) scheduled a case follow-up visit within 24 hours. At this visit, after obtaining informed consent for the follow-up procedure, [Appendix 2, Form 78] the ANMs complete a rapid anti-HEV

IgM assay, a 5mL venous blood draw, and a brief questionnaire about treatments sought, and they provided a referral for local medical care if warranted. The Case Visit Form [Appendix 2, Form 80] included questions about previous care or advice sought for the reported symptoms, treatments recommended or administered (pills, herbs, teas, spiritual practices), and perceived effectiveness of treatments.

Two drops (~30µL) of whole blood from the venous blood draw were tested in the field with the Wantai Rapid HEV IgM test (Wantai BP Co., Beijing, China), carried out according to the manufacturer’s instructions.

All symptomatic women were informed of the rapid HEV test result, and told about local medical care options. Women testing positive for acute HEV infection were also provided with a

“Hepatitis E Report Card” (HERC) [Appendix 2, Form 81] that they could present to a local medical facility if they sought treatment. The ANMs explained the typical course of illness in simple terms,

and recommended seeking care at a medical facility if symptoms worsened or if the woman had concerns about her pregnancy.

Blood samples from symptomatic women were transported to the field laboratory and processed following the same procedures as the routine blood samples, but the entire volume was designated for the hepatitis substudy.

Serologic Surveillance and Testing

Routine serologic testing of baseline, post-delivery, and 3 month postpartum samples, and diagnostic serology for cases detected by syndromic surveillance, were performed by the WARUN under the terms of a collaborative research agreement.

All participants’ baseline (pregnancy enrollment/PE) samples were tested for antibodies to hepatitis A, B, C, and E, as well as for acute or recent infection with HEV using Wantai ELISA kits

(Beijing Wantai BP Enterprise Co., Beijing, China). Specimens testing positive for anti-HBc were also tested for HBsAg to determine if the infection was likely recent or chronic. All women, regardless of baseline serostatus, were tested for antibodies to HEV at all three timepoints to capture seroconversions and possible seroreversions.

Hepatitis follow-up samples drawn from women identified through syndromic surveillance as presumptive cases underwent diagnostic screening. These samples were tested for anti-HAV

IgM, HBsAg, anti-HCV, and anti-HEV IgG and IgM (Wantai HAV IgM ELISA, AiD HBsAg ELISA, AiD

HCV ELISA, HEV IgG ELISA, and HEV IgM ELISA, Beijing Wantai BP Enterprise Co., Beijing, China) to identify hepatitis virus(es) causing the symptoms. Results of the Wantai Rapid HEV IgM assays performed by study staff in the field in Sarlahi were compared with the results of the Wantai HEV

IgM assays conducted in the laboratory at WARUN in Kathmandu. Acute hepatitis E was diagnosed in a symptomatic woman when either the rapid field test or the laboratory test for anti-HEV IgM

returned a positive result. Personnel at WARUN also attempted to detect viral RNA in serum from women with acute hepatitis symptoms and/or positive anti-HEV IgM results using reverse transcriptase polymerase chain reaction (RT-PCR) methods described elsewhere,32,33 and sent viremic samples to the Armed Forces Medical Institute of Research (AFRIMS) in Bangkok, Thailand, for sequencing.

Ascertainment of Pregnancy Outcomes and Maternal and Infant Mortality

In addition to limited weekly and extensive monthly symptom surveys and hepatitis case follow-up, maternal health, pregnancy outcomes, and infant health were monitored as part of the

MaGIFT trial framework.

Upon notification of labor or delivery by a family member or field worker, a trained Birth

Assessment Team was dispatched by motorcycle to the woman’s residence. Maternal and infant birth assessment forms were used to collect information about the health of the mother-child pair prior to, during, and after delivery, and to record any complications or apparent birth defects. In the cases of multiple births, separate forms were completed for each infant.

Hospitalizations occurring during the study period were reported by field data collectors and by members of the birth assessment team to the study PIs. Maternal and infant deaths identified by field staff during follow-up were recorded, and verbal autopsy forms were completed with the aid of relatives or neighbors, when possible, to ascertain causes of death.

Feasibility and Sample Size Considerations

Conducting the proposed work through an ongoing field trial was intended to make efficient use of existing study infrastructure and minimize added burden to study participants. The project also stood to benefit from good community rapport and interest in research resulting from both

long-term engagement of the MaGIFT team in a number of other health interventions in the region, and also the employment of local men and women as field workers.

Target enrollment during the April 2012-April 2013 phase of MaGIFT was 1850 pregnant women. Based on prior NNIPS work in Sarlahi, fewer than 10% of eligible participants were expected to decline participation, and under 5% were expected to be lost to follow-up. There was some concern that higher-than-anticipated enrollment during the previous phase of the trial, due primarily to a large number of prevalent pregnancies enrolled into the cohort at study inception, could reduce yield in this second phase, as women were ineligible to contribute more than one pregnancy to the trial. (J. Tielsch, pers. comm.) The enrollment targets were increased to 2300 when the purely observational phase was added following completion of the randomization phase.

Based on population-based incidence rates of roughly 60 seroconversions / 1000 person- years (p*y) observed in a rural Bangladeshi cohort (overall and in women)34 and ~64 infections /

1000 p*y observed in Kathmandu (in a primarily male population),23 and an assumption that each of 1000 participants would contribute ~9 months of person-time over the study period, approximately 45-48 HEV seroconversions would have been expected if all participants were susceptible at enrollment. Assuming a 20% prevalence of protective immunity at baseline,35 the number of seroconversions would instead have been be about 36-39. Of these, we estimated that

~1/3 to 1/2 would be acutely symptomatic and detected by syndromic surveillance, yielding approximately 12-20 cases of acute hepatitis E. We expected to detect a similar number of non-

HEV-attributable hepatitis-like illnesses through the symptom surveillance system. A sample size of 1000 women would be sufficient to estimate the incidence of seroconversion and symptomatic illness with a precision of ±6.2% (at the 95% confidence level; with 1400 women, to within 5.3%).36

We expected to detect viral RNA in one or more samples from confirmed cases, allowing us to examine HEV strain(s) circulating and causing illness in the Terai. 69

Were the study period to have coincided with a sizable outbreak of hepatitis E, our a priori predictions of seroconversion and clinical illness incidence would have underestimated the number of observed cases, but our incidence estimates would be higher than expected. Conversely, if only sporadic cases were detected and no outbreaks were to occur during the study period, our estimates of HEV infection and illness would likely be low relative to what one might expect over a longer period of observation that included both sporadic cases and occasional outbreaks.

Ethical and Human Subjects Considerations

Study protocols and forms for the hepatitis substudy were reviewed and approved by the

Johns Hopkins Bloomberg School of Health IRB (#2458), the Nepal Health Research Council (NHRC,

#92/2011), and Walter Reed Army Institute of Research (#1973). The parent study, the Mother’s

Gift Field Trial of Maternal Influenza Immunization in Asia (MaGIFT), is a registered Phase III clinical trial (clinicaltrials.gov ID #NCT01034254) with a data safety monitoring board and an independent monitor. Its principal investigators and study staff have had a long history of commitment to health projects in the region. Most field workers were women hired locally, reducing potential cultural or linguistic barriers.

The study was designed with local cultural practices and reported preferences in mind, and attempts were made to minimize the burden of participation. Venous blood collection was substituted for the fingerstick blood collection system used successfully in Bangladesh because venipuncture was reported to be more acceptable to participants in Sarlahi (J. Tielsch, pers. comm.).

Using aliquots from venous blood draws instead of fingerstick samples also reduced the total number of needlesticks participants received. Confidentiality was maintained to the extent possible in this community setting. Data collection forms and biological samples were indexed by coded study ID number, and identifiers were kept for follow-up purposes in a secure location in the field office in Hariyon. 70

Participation in the hepatitis substudy was not designed to yield any direct benefits to participants, as no standard treatments are currently available for acute hepatitis E during pregnancy. Furthermore, due to off-site laboratory testing arrangements, detection of antibodies to other hepatitis viruses was delayed and results could not be provided in real time. Although a pentavalent vaccine including hepatitis B is part of the routine childhood immunization schedule in

Nepal,37 its coverage in Sarlahi is lax (J. Katz, pers. comm.) and no monovalent hepatitis B vaccine is currently available in Nepal to treat infants born to mothers with chronic hepatitis B. Thus, unfortunately, we would have been unable to intervene to prevent vertical transmission of hepatitis

B even if a chronic infection could have been detected in a pregnant woman in a timely manner.

Women whose rapid test result was positive for HEV received information from study staff and a card they could take with them if they sought medical care. Study staff referred all ill women, regardless of test results, to the nearest medical clinic for supportive care, as indicated. Indirect benefits to participants included contribution to the empirical evidence base describing illness and death from HEV in their community, and helping to inform possible interventions, including the potential introduction of a vaccine that has been approved in China.

Data Management, Quality Assurance, and Quality Control

General forms and protocols for the study were field-tested and refined during the first phase of MaGIFT in Sarlahi. Hepatitis Substudy-specific forms and protocols, including the exposure assessment questionnaire and symptom scoring procedures, were adapted from those used successfully in the Matlab study areas in rural Bangladesh.38 From July-October 2012, all Hepatitis

Substudy data collection instruments were translated into Nepali, pre-tested, and refined by study personnel in an iterative process to ensure clarity, usability, and sensitivity to local custom.

All field workers involved in the study were trained for their roles. Venipuncture was performed by auxiliary nurse-midwives (ANMs) trained in phlebotomy. All study personnel handling biological samples were trained in bloodborne pathogen precautions.

Training was provided in the use of Hepatitis Substudy-specific forms and protocols.

Written instructions and examples were documented in the MaGIFT Manual of Operations. Weekly meetings of field workers, team leaders, and other personnel based in Hariyon provided a means for reviewing protocols, troubleshooting any problems, and providing feedback.

All forms contained participants’ study identification number, date, and worker initials for tracking purposes. Paper forms completed in participants’ villages were returned to the field office in Hariyon for preliminary review of completeness, legibility, and consistency by team leaders before being transported to the NNIPS Data Center in Kathmandu.39,40 Entry of form data into the

MaGIFT database was done by staff at the NNIPS Data Center in Kathmandu. The staff at the Data

Center in Kathmandu have historically had a low data error rate over time,40 and corrective action was taken when a database updating problem was discovered.41

Laboratory results from WARUN and AFRIMS were sent to hepatitis substudy investigators for integration with the study database using participants’ study identification number. WARUN used appropriate positive and negative controls for all assays performed.

Data Analysis

Stata v.11 (StataCorp, College Station, Texas) was for statistical analyses. Sessions were logged and archived.

Prevalence, Incidence, and Related Estimates Seroprevalence for each of HAV, HBV, HCV, and HEV at a given time-point (enrollment, delivery, 3 months postpartum) was estimated as the number of women with seropositive assay results, divided by the total number of consenting women tested at that time-point.

To estimate the incidence of seroconversion during pregnancy, the number of women seronegative for anti-HEV IgG at baseline who subsequently tested positive at the post-delivery blood draw were divided by the amount of person-time contributed by susceptible women between those two points less one-half the person-time contributed by seroconverters. The incidence of seroconversion during the postpartum period was estimated as the number of women seronegative at the post-delivery timepoint (excluding women seropositive at baseline) who subsequently tested positive at the three month postpartum timepoint, divided by the amount of person-time contributed by susceptible women during this time period. We calculated incidence over additional intervals (e.g., baseline to 3 months postpartum) for comparison with published results from other studies. Person-time for susceptible women was calculated by subtracting the date of one blood draw from the date of the previous blood draw and converting the result to years.

We estimated the incidence of hepatitis-like illness using the number of cases defined by above-threshold Hepatitis Score as the numerator and the number of person-months ascertained by completed follow-up forms as the denominator. The proportion of acute viral hepatitis cases attributable to HEV was estimated by dividing the number of symptomatic women with lab- confirmed acute Hepatitis E by the total number of women with acute hepatitis-like illness. The case-to-infection ratio was estimated by dividing the number of confirmed hepatitis E cases by the total number of women with baseline anti-HEV IgM plus anti-HEV IgG HEV seroconverters detected through routine follow-up serology from baseline to the post-delivery timepoint.

RESULTS

Study Enrollment

A total of 2,363 women consented to participate in the hepatitis substudy within MaGIFT, and 5 women declined consent. Of these, 956 women joined the hepatitis study and 5 declined during the randomized, controlled influenza vaccine trial period of MaGIFT (8 October 2012- 24

April 2013) and 1407 women enrolled during the post-randomization observational phase (25

April 2013 – 2 May 2014). Forms for 29 women exist but have not yet been entered into the study database, and forms for another 8 women are missing; these 37 women and their data are not included in the enrollment figures or in the data analyses. Enrollment of study participants relative to the total female population of each VDC and ward is shown in Figure 6.

Serologic Follow-Up

Blood collection over the course of the study is summarized in Figure 7. Baseline maternal blood specimens were collected from 1669 participants at a median gestational age (estimated from self-reported last menstrual period) of 19.4 weeks (range: 6.7-34.4 weeks). Of the 117 other women for whom baseline specimen collection attempts were recorded: 108 declined to provide baseline specimens, and 9 did not provide specimens for other reasons. The rest of the enrolled women (n=577) were not approached to provide a sample due to the transition between the RCT phase and the observational phase of the study or the end of the study period, or they had missing records.

Delivery specimens were collected from 811 women a median of 13 days following the end of the pregnancy, with >90% of these specimens collected within 4 weeks of the pregnancy outcome date. Of the 470 other women for whom delivery specimen collection attempts were recorded: one woman was deceased, 327 were not met, 50 declined to provide specimens, and 92

women did not provide specimens for other reasons. The remainder of the women (n=1082) did not deliver in time to give a blood sample prior to the end of the study period, or they had missing records.

Three-month postpartum specimens were collected only from women who joined the study as part of the randomized, controlled trial, and not from women who joined during the later observational phase. These specimens were obtained from 659 women. One additional woman was deceased, 149 were not met, 38 declined to provide specimens, and 46 did not provide specimens for unknown reasons. The remainder of the women either did not reach 3 months postpartum prior to the end of the study, or they had missing records.

A total of 792 women provided both baseline and delivery blood samples, 643 women provided both baseline and 3 month postpartum samples, and 480 women provided samples at all three timepoints (baseline, delivery, and 3 months postpartum).

Baseline Serostatus and Risk Factors for HEV Seropositivity

Among 1669 women with baseline specimens, 195 or 11.7% [95% confidence interval (CI):

10.2%-13.3%] were positive for anti-HEV IgG (one of whom was also anti-HEV IgM positive).

Nearly all were positive for IgG antibodies to HAV, roughly one-quarter had anti-HBc antibodies and about 2% were positive for HBsAg, and none was anti-HCV IgG positive. [Table 4]

Characteristics of the full cohort, women who provided baseline specimens, and women seropositive for anti-HEV IgG at baseline are shown in Table 5, and Figure 6 depicts study enrollment relative to local population. Seropositivity for anti-HEV IgG was associated in univariate analyses with age, parity, education, caste, household land ownership, and meat-eating, but not with household latrine presence or type or ward-level latrine prevalence, nor with primary

drinking water source. [Figure 8] Village- and ward-level variation in baseline seroprevalence of anti-HEV IgG, in population by caste, and in distribution of latrines are shown in maps in Figure 9.

In multivariate logistic regression analysis, [Table 6] increasing age was associated with increased odds of baseline seropositivity. Both socioeconomic variables included in the final model were independently associated with serologic status, with traditionally lower caste/religious affiliation (especially Shudra, Muslim, and those with another or no reported caste) and lack of agricultural land ownership both linked to increased odds of seropositivity. Although residents of

Hariyon VDC had the highest seroprevalence of anti-HEV IgG overall, only residence in

Dhungrekhola (and no other VDCs) was statistically-significantly protective [OR 0.47; 95% CI: 0.23-

0.95] after adjustment for other covariates.

Meat-eating [Figure 10] was consistently and fairly strongly associated with seropositivity in both univariate and multivariate analyses. Inclusion of individual meat consumption patterns in the multivariate models (i.e., specific meats eaten or frequency of consumption of specific meats) did not yield effects as strong as the general binary vegetarian/meat-eater variable. Eating fish appeared to underlie much of the association between meat-eating and seropositivity, as its inclusion in models produced the largest decrease in the general meat-eating effect size. However, addition of a “dose” variable based on frequency of fish consumption in the prior 30 days did not improve the models, whereas both pig consumption and buffalo consumption appeared to show some differences in effect with increasing consumption. While the direct association of frequency of pig consumption with seropositivity represented a statistically-significant trend on its own, it did not improve the models in multivariate analyses. The association of buffalo meat consumption frequency with seropositivity, while statistically significant at some levels, was inconsistent in direction, driven by a large number of seropositive women who had eaten buffalo exactly 2 times in the prior month. Very few women ate only a single type of meat, and no women ate only fish or 76

only pig; the collinearity of several of the specific binary meat-consumption variables and the unclear dose-response relationships for specific meat consumption frequencies were additional reasons for inclusion of the general vegetarian/meat-eater variable in the model rather than individual meat types.

Serologic Estimation of HEV Incidence

Excluding those women who had positive anti-HEV IgG serology at baseline, we observed 9 seroconversions during pregnancy (from baseline to delivery) and 8 seroconversions in the postpartum period (from delivery to 3 months postpartum). An additional 3 seroconversions occurred between enrollment and 3 months postpartum among women who did not provide post- delivery samples. The 9 seroconversions, occurring among 695 susceptible women who gave both baseline and delivery blood samples an average of ~20 weeks apart, represents an approximate incidence proportion of 1.3% (or 1.4% if the woman who was seropositive for anti-HEV IgM and

HEV RNA at baseline were also counted as a seroconverter).

Accounting for actual person-time contributed by susceptible women during each of these periods, and subtracting ½ the person-time contributed by seroconverters, these observations yield seroincidence estimates of 32.3/1000 person-years (p*y) [95% CI: 14.8 to 61.3/1000 p*y] during pregnancy and 93.3/1000 p*y [95% CI: 40.3-183.8/1000 p*y] during the first three months postpartum. The incidence among women for whom only baseline and 3 month postpartum specimens were obtained (n=141) was 34.5/1000 p*y [95% CI: 7.1-100.9/1000 p*y].

Considering person-time contributed to serologic surveillance by the full cohort, regardless of distribution relative to pregnancy or absence of a specimen at a single time-point, we obtain an overall incidence estimate of 44.3/1000 p*y [95% CI: 27.1-68.4/1000 p*y] based on 20 seroconverters over 451.4 p*y. Looking at baseline and 3 month postpartum serostatus only (and

excluding women who failed to provide a specimen at either of these timepoints), we obtain an estimate of 26.3/1000 p*y [95% CI: 12.0-49.9/1000 p*y] based on 9 seroconverters over 342.2 p*y.

Seroreversion from anti-HEV IgG+ to anti-HEV IgG- was also observed in our cohort. Among

90 women seropositive at baseline who provided post-delivery samples, 6 were seronegative following delivery. Among 62 women seropositive at delivery who provided 3 month postpartum samples, 5 were seronegative at three months postpartum. Among the 9 women who seroconverted between enrollment and delivery, 5 provided samples at 3 months postpartum, and all 5 of these were seronegative; however, the woman who was viremic and seropositive for anti-

HEV IgG and IgM at baseline maintained her IgG seropositivity at 3 months postpartum.

Maternal Morbidity and Mortality

Pregnant participants contributed a total of 8716 completed follow-up visits at roughly monthly intervals, each covering symptoms experienced during the prior 30-day period. Using the symptom-based case definition shown in Table 3, 15 pregnant women were identified as potential cases of hepatitis-like illness, for an incidence of approximately 20.7 / 1000 person-years.

Only one of these 15 symptomatic cases was confirmed to be hepatitis E, both by rapid field test (IgM ELISA) and by later serologic and molecular testing. [See case report for further details.]

Two other symptomatic cases had anti-HEV IgG, but were anti-HEV IgM negative, and neither had a usable baseline sample to determine whether or not the positive IgG result was linked to the recent illness or represented infection in the more distant past. None of the acutely symptomatic women tested positive for anti-HAV IgM, anti-HBsAg, or anti-HCV antibodies.

If we count only the single lab-confirmed case to be caused by HEV, this yields a hepatitis E incidence of roughly 1.37 / 1000 person-years. If we relax the case definition and assume that all 3

HEV IgG+ cases represented hepatitis E, we obtain an estimate of 4.13 / 1000 person-years.

There were 2 maternal deaths recorded among women enrolled in the cohort. One of these deaths occurred on the day of childbirth, and no specific cause was recorded; the woman had reported dark urine but no other hepatitis-like symptoms at a follow-up visit three weeks earlier.

The other death occurred in a woman who had delivered a child 7 weeks earlier, and this death was associated with “cough, body pain, and jaundice”. In the 3-4 months preceding her death, while still pregnant, she reported persistent nausea, weakness, upper abdominal pain, and olfactory disturbances, but never jaundice, and these combinations of symptoms never exceeded the threshold for hepatitis case detection used in the study.

Pregnancy and Infant Outcomes

Primary Birth Outcomes Among the 2363 cohort participants, 1801 pregnancy outcomes were recorded through 2

May 2014. 1729 (96%) of these were live births (singleton or twin), 31 were stillbirths, 36 were miscarriages, and 5 were abortions. All of the nine mothers who seroconverted for HEV during pregnancy, as well as the one woman confirmed to have acute hepatitis E, gave birth to live infants.

Gestational Age and Birthweight The median duration of pregnancy, based on self-reported last menstrual period, was 39.6 weeks in the 1801 women for whom pregnancy outcome was known. Among live births (n=1729),

196 (11.3%) were preterm (<37 weeks gestation). Two of the nine HEV seroconverters during pregnancy (22.2%) had preterm deliveries, one at 30.7 weeks and the other at 36.3 weeks; the association of preterm births with HEV seroconversion was not statistically significant (odds ratio

2.5, 95% confidence interval 0.5-12.3). In 1244 women with live births for whom baseline anti-

HEV IgG testing was completed, maternal baseline HEV seropositivity was associated with increased odds of preterm birth compared with baseline seronegativity (odds ratio 1.9, 95% confidence interval 1.2-3.0). The size and statistical significance of this apparent association were

diminished slightly when socioeconomic factors (caste and household land ownership), maternal age, and parity were taken into account (odds ratio 1.6, 95% confidence interval 0.9-2.6).

Among all singleton live births visited within 48 hours of delivery (n=1164), the prevalence of low birth weight (<2500g) was 21.6%. This was similar to the 25.0% prevalence of low birth weight among babies born to mothers who seroconverted for HEV during pregnancy, and the difference was not statistically significant (χ2=0.02, p=0.88). None of the babies born to seroconverting mothers had very low birth weight (<2000g). No association was found between maternal baseline HEV serostatus and low birth weight.

DISCUSSION

Prevalence and Incidence of HEV Infection and Illness

This community-based, longitudinal study evaluated sporadic hepatitis E virus infection and illness among pregnant and postpartum women in the Terai of Nepal, where hepatitis E has not previously been well-characterized. The baseline anti-HEV IgG seroprevalence of 11.7% in pregnant women in Sarlahi is considerably lower than among adults living in the Kathmandu

Valley15,16,23 and in some areas of India and Bangladesh,35,42,43 where estimates ranging from 25% to

>75% have been reported with comparably sensitive assays. However, it is similar to reports of

13.7% and 10% in two recent cohorts of pregnant women drawn from the JiVitA study area in northern Bangladesh.44,79 It is also comparable to seroprevalence estimates in healthy adults in the eastern Terai cities of Dharan and Biratnagar (cf. Biratnagar outbreak chapter).

Our finding of relatively low baseline seroprevalence and incidence in pregnant women in a country where very high rates of infection with HEV have previously been documented is somewhat surprising. Typically higher seroprevalence estimates among men, which tend to elevate general population estimates if not adjusted for sex distribution, likely explain some of the 80

discrepancy. A 1994 study conducted in Kathmandu showed divergence of seroprevalence by sex with increasing age, with an increase in men’s seroprevalence but a decrease in women’s seroprevalence from ages 20-30 and overall seropositivity in 16% of men and 9.5% of women over age 19.22 That study used an older, considerably less-sensitive assay,80 so the absolute seroprevalence estimates they obtained are not directly comparable to ours (and may underestimate seroprevalence by as much as 200%); however, the observation of greater seroprevalence in men and the different age-specific seroprevalence patterns are still of interest.

The men involved in the Kathmandu study were mostly members of the military or police forces,22 who may have had increased risk of HEV exposure relative to men in the general population due to their outdoor occupations and housing in barracks, biasing the population estimates. However, the decline in seropositivity with increasing age among women in Kathmandu,22 which also contributed to the differences by sex and age, is not a common finding. By contrast, in our cohort of pregnant women, seropositivity was greater in older age groups through the second and third decades of life.

A more recent study comparing age-specific population seroprevalence estimates in Nepal

(Kathmandu), Bangladesh, and France did not report results by sex, precluding consideration of sex-by-age interactions.15

The decentralized nature of the water supply in rural areas of the Terai, with most women obtaining water for drinking and other household uses from local tubewells or ring wells, may also help explain the relatively low exposure in Sarlahi as compared with urban areas in both the

Kathmandu Valley and in northern India, where water is more commonly supplied by pipe and has ample opportunities to become contaminated as it is distributed. Rural-urban differences in seroprevalence among pregnant women have been studied elsewhere, with seroprevalence in rural women lower in Gabon,81 higher in Turkey, 82 and similar to that in urban women in Qingdao and

Weihei, China.83 Major differences in study populations and geographic factors make it difficult to generalize findings from these studies to the Nepali setting.

Jaundice (representing both icterus and a variety of non-icteric complaints) is reported frequently in Bangladesh,45 but it was uncommon in our southern Nepali cohort . Even among the

15 women who met the criteria for our case definition of hepatitis-like illness, only 4 reported jaundice as a symptom. As has been noted in other settings, the case-to-infection ratio in our cohort was low, with only one symptomatic case of HEV occurring for 20 apparently-asymptomatic seroconverters. Although we kept our case definition intentionally broad so that it would be more sensitive to milder hepatitis E cases without frank jaundice, we did not capture any additional seroconverters this way.

The incidence of serologically-detected, asymptomatic or subclinical HEV infection in the cohort was 44.6/1000 person-years (p*y), substantially less than the 99/1000 p*y incidence reported in young adult populations in Kathmandu 23 and the 64/1000p*y in a general population in southern Bangladesh.34 Although our estimate at first appears comparable to the 46/1000 p*y incidence recently estimated in northern Bangladeshi pregnant women, 44 Kmush et al. defined seroconverters as women who were seropositive at 3 months postpartum but seronegative at earlier points (excluding from their pool of seroconverters those women who may have seroconverted and subsequently seroreverted within that timeframe). Applying a similar definition of seroconversion to the Sarlahi cohort, we obtained a much lower estimate of 26.3/1000p*y.

As in northern Bangladesh, where only one of 40 detected seroconversions occurred between early and late pregnancy and the rest occurred postpartum, 44 we observed a much greater incidence of seroconversion from delivery to 3 months postpartum than from baseline enrollment to delivery. The reasons for this difference in rate are unclear.

In the Sarlahi cohort, 5/5 women who had seroconverted between baseline and delivery were again seronegative when re-tested at 3 months postpartum. Seroreversion could perhaps explain a small portion of the discrepancy observed by Kmush et al., as their testing for antibodies at earlier points was conditioned on anti-HEV IgG seropositivity at 3 months postpartum. Thus, some women who seroconverted during early pregnancy could have seroreverted during the late pregnancy/early postpartum period, and those infections would have gone undetected. Transience of seropositivity would not explain the differences in rates observed in the Sarlahi cohort, however, as specimens collected from all participants at all three study timepoints were tested prospectively for IgG.

Completeness of serologic follow-up with the Nepali women was limited considerably by several factors: the MaGIFT study dates and changes to protocol during the study period; Nepali women’s greater participation in activities outside the home relative to women in Bangladesh, such that a large number of women were not met on multiple occasions when the ANMs visited to collect samples; and participants who declined to provide samples for various reasons. Women who reported staying in their homes all day were slightly more likely to be seronegative at baseline than women who regularly went out. If engagement in activities outside the home were a risk factor for infection, more non-seroconverters than seroconverters might have been at home to provide follow-up specimens, artificially depressing our incidence estimates. Whether or how this would have a differential effect on women providing samples following delivery vs. at 3 months postpartum is not clear, but missing data (if differentially missing by serostatus and/or season) could affect the validity of our incidence estimates.

Behavioral changes associated with pregnancy could potentially change HEV exposure patterns. Some of these factors (such as reduced outdoor exertion) might be expected to decrease potential exposure to HEV, while others (such as increased water intake and/or more frequent 83

need to urinate) might be expected to increase it. Our behavioral exposure data were collected only at enrollment and may not adequately capture changes over the course of the pregnancy and postpartum periods. Changes in food intake patterns during pregnancy resulting from morning sickness, increased nutritional needs, or cultural traditions are likewise poorly reflected in our data.

Seasonal patterns of infection could also potentially account for some of the incidence difference in the Sarlahi cohort, as follow-up periods were not evenly distributed over the calendar year due to MaGIFT study dates. During the 1981-1982 hepatitis E epidemic in Kathmandu, hospital-based case counts were highest during the monsoon, and in 1980, prior to the epidemic, a peak in cases from July-September was still evident.4 In southern Bangladesh, a higher incidence of seroconversion was detected during the monsoon than during the pre-monsoon season.34 In

Vellore and Lucknow, India, HEV genotype 1 RNA was detected substantially less frequently in sewage samples collected during the monsoon season than in other seasons.84,85 While these environmental observations could concievably reflect dilution of the sewage stream by rainwater during the monsoon rather than a true reduction in viral shedding in the human population, fewer symptomatic hepatitis E cases presented to the hospital in Vellore during the monsoon as well.85

In the Sarlahi cohort, postpartum follow-up person-time was more likely than pregnant person-time to occur during the monsoon season in the Sarlahi cohort due to the study’s October

2012 start and the gap in new enrollments from May-July 2013. A slight majority of intervals during which participants seroconverted fell entirely outside the monsoon season (even with a 2-4 week lag to allow for delayed development of antibodies). The single confirmed acute hepatitis E case occurred during the winter season. Determining seasonality of infection in our study with a relatively small number of seroconverters was not possible to do with high confidence, and infrequent and dynamically-scheduled serologic testing made it more difficult to pinpoint when infections were most likely to occur. If a seasonal pattern could be detected, both exposure-related 84

and host-related factors might play a role. Micronutrient levels vary by season among pregnant women in Sarlahi.86,87 Zinc deficiencies and anemia (which peak in summer in Sarlahi 86), as well as vitamin D deficiencies (which peak in winter in Sarlahi 86), were associated with HEV seroconversion in Bangladeshi pregnant women.44

Maternal Mortality

Neither of two maternal deaths reported during the hepatitis substudy, one peripartum and one postpartum, was conclusively attributable to hepatitis E. Verbal autopsy for the participant who died 7 weeks postpartum mentioned jaundice, and that woman had a history of symptoms consistent with hepatic illness (including dark urine and upper abdominal pain) in the months prior to death. Although a small number of delayed cases have been documented in non-pregnant adults,46 liver failure resulting from acute hepatitis E typically occurs within days to a few weeks of symptom onset. Although we cannot rule in or out HEV as a potential contributor to this woman’s death, other causes of liver dysfunction in pregnancy and adulthood, many of which present with similar symptoms,47 may be more likely.

To determine whether reports of jaundice or hepatitis as causes of death may have been unusually low during the study period, and to consider a larger sample, we reviewed all maternal deaths that had been documented by verbal autopsy in an extended MaGIFT cohort from January

2011 onward (n=22). Among the 22 recorded maternal deaths, 6 were attributed to suicide or accidental injuries, 4 to postpartum bleeding or abdominal abscess, 2 to jaundice, 2 to preexisting conditions (heart valve defect, asthma), 2 to tuberculosis, 1 to liver cancer, 1 to fever and kidney failure, 1 to fever and edema of unknown etiology, 1 to neurologic dysfunction associated with a fall and bleeding, and 2 to unknown causes.

The overall prevalence of jaundice as either a cause or as a precursor of death (as mentioned in free narrative or in response to questions about prior jaundice diagnosis and yellow eyes) was 13.6% (3/22). This was somewhat higher than the 9.8% noted in a population in rural

Bangladesh;1 however, the proportion of jaundice-related deaths attributable to hepatitis E is unclear, and may be lower than in that Bangladeshi population. Of the 3 jaundice-associated maternal deaths, one was part of our hepatitis substudy cohort, as described above, and two were not. One of the deceased women had a 5-month history of jaundice that did not respond to treatment, possibly a result of hepatitis B virus infection or another condition. The other woman’s death was attributed to a diagnosed-but-unrepaired heart-valve defect, and the jaundice may have been secondary to heart failure or even to treatments prescribed by the traditional healers or hospitals she visited.

The recorded deaths were distributed unevenly over time, with more reported in 2011 than in 2012, and more in 2012 than in 2013, suggesting possibly-incomplete ascertainment of the more recent maternal deaths, but jaundice-associated deaths were not obviously underrepresented in the hepatitis substudy cohort. While we cannot definitively determine all of the causes contributing to these deaths, and we cannot rule out some contribution of HEV infection, jaundice-associated mortality in this population of pregnant women is attributable to multiple causes.

Pregnancy Outcomes

We were unable to robustly evaluate the potential impact of hepatitis E on pregnancy outcomes because of the systematic exclusion from the MaGIFT clinical trial of women who miscarried prior to randomization, poorer postpartum sample collection among women who did not have live births, and relatively few seroconversions and even fewer clinical cases among enrolled women. A greater proportion of infants born to seroconverting mothers than to seronegative mothers were delivered preterm, but this association was not statistically significant. 86

Otherwise, in our small sample of women infected with HEV during pregnancy, most of whom were not ill, and none of whom developed life-threatening disease, we observed no greater occurrence of adverse fetal or neonatal outcomes.

Risk Factors for Exposure to HEV

Socioeconomic factors, including traditionally lower caste / non-caste affiliation and lack of family land ownership, were strongly associated with increased seroprevalence of anti-HEV IgG as has been observed elsewhere.42 The relationship of lower status and lack of land ownership to elevated seroprevalence in Sarlahi did not appear to be explained, as might be expected, by differences in household water sources, nor by individual access to latrines, nor by ward-level access to latrines. Few women purified their drinking water in any way prior to consumption, and those who did mostly used a cloth filter to remove sediment. Soap was widely available in homes and nearly all women reported good hand hygiene.

We did not ascertain all potentially-relevant details of water sources, such as tubewell depth, that may vary with economic means and be associated with likelihood of water contamination. It is also unclear whether either individual- or ward-level lack of latrine access is an appropriate proxy for potential exposure to infectious human waste resulting from open defecation, as these variables were neither strongly nor statistically-signficantly associated with

HEV seropositivity. Local sanitation customs, geographic controls on water and waste runoff, or ward boundaries that do not correspond well to natural “neighborhood” boundaries may partly explain the apparent lack of association between water, sanitation, and hygiene-related variables and anti-HEV IgG serostatus in our analyses.

Many of the young women in the study were recently-married, and some may have only recently moved into the study area; thus, it is possible that local environment at the time of study

enrollment was a poor proxy for lifetime exposure. However, many of the women did reside within the study area or within nearby villages prior to marriage, and given strong economic, caste, and ethnicity-based influences on marriage customs, it is not clear that their exposures would change markedly.

Among participants who self-identified as belonging to one of the four broad Hindu varna- based (caste) groups, seroprevalence increased steadily from Brahmin (5.9%) to Chhetri (8.0%) to

Vaishya (10.4%) women, before rising sharply to 17.6% among Shudra women. The bounds of the

“Shudra” designation have been defined variously, with the label traditionally applying to the lower-status groups still considered “pure” under the varna system 74 and others extending it to groups identified by tradition as “impure”, “water-unacceptable” (meaning ), or “untouchable” and now more often called “Dalit” (“oppressed”). 75 Only four women identified themselves as belonging to groups other than those defined by the four-varna system (or religion, in the case of Islam), and we lacked caste-identification data for another 69 women, who were grouped with them in an

“other/unknown category” which had the highest seroprevalence of anti-HEV IgG (24.7%). Some of the “other” groups may include Dalit communities as well as “Adivasi Janajati” (“indigenous peoples”). Janajati are those belonging to ethnic groups tied to various regions of Nepal, some of which are not predominantly Hindu, and others of which practice Hinduism but which, for historical reasons, were not as tightly bound to the formal varna system.74 Both Dalit and Janajati groups have historic legacies of “social exclusion and discrimination” in Nepal, 74 and these groups remain underrepresented in government and in other positions of influence. In tandem with their disenfranchisement, Dalit and Janajati communities continue to lag on a variety of indicators of economic development and health, including latrine access, antenatal care, and infant mortality. 74

In our cohort, however, the prevalence of open defecation was higher among Shudra, Muslim, and

Vaishya women than among women of “Other/Unknown” caste, and access to latrines appeared to show little association with HEV seropositivity, in any case.

Muslims in Sarlahi (who, as non-Hindus, also do not fit within the traditional varna system), like the Shudra and “other” groups, had a greatly-elevated seroprevalence of anti-HEV IgG (20.9%) relative to higher caste-based groups, even after adjustment for several other covariates. This stands in contrast to an observation in Balinese pregnant women, where Muslims had a lower seroprevalence of anti-HEV IgG than Hindus.48 Muslims in Bali, as in Sarlahi, are a religious minority in an area with a large, caste-stratified Hindu majority. Further investigation is needed to determine what specific socioeconomic, cultural, phylogeographic, or other factors might account for these observed differences in anti-HEV IgG seroprevalence in Nepali vs. Balinese Muslim women.

In both Indonesia and Nepal, as in India, waterborne HEV-1 has been implicated as the predominant cause of outbreaks and clinical illness.4,49-55 An earlier study in western India found that even where genotype 4 HEV was prevalent among swine, human cases from the same region were all attributable to genotype 1.56 Thus, exposure in these regions South and Southeast Asia might be expected to be less heavily dependent on dietary practices or animal exposures than in places like North America, Europe, and urban East Asia, where HEV-3 and HEV-4 are the dominant circulating genotypes and there is no sustained, local HEV-1 transmission.

In the context of direct animal-to-human foodborne transmission, Muslims and others

(including some Hindu Brahmins) who eschew pork might be less frequently exposed to HEV than the general population. In Bali, HEV genotype 4 has been detected in swine, 57 and foodborne transmission via pork is suggested by the authors of the Bali study as an explanation for the lower seroprevalence of HEV antibodies among Muslim women on that island. 48 In Nepal, HEV antibodies

(of unknown genotype) have been detected in swine near Kathmandu,58 though a more recent survey detected no HEV in manure runoff at 41 small pig farms along Dhobi Khola in the

Kathmandu Valley. [Kerry Morrison et al., unpublished data] No human HEV-3 or HEV-4 cases have yet been reported in Nepal, however, and it is unclear what role, if any, zoonotic reservoirs may play in HEV transmission in South Asia.

Despite the high seroprevalence of anti-HEV IgG in Muslim women in Sarlahi, we still observed a modest (but not statistically-significant) association of HEV seropositivity with pig meat consumption, as well as a strong association of strict vegetarianism with seronegativity that persisted even after controlling for caste and other socioeconomic indicators (land ownership and literacy). It is plausible that there could be concurrent circulation of both waterborne (genotype 1) and zoonotic HEV in southern Nepal, with exposure to the latter having gone previously undetected due to limited molecular surveillance and/or lower virulence of those strains. Scant (and murky) regional data exist on potential HEV reservoirs in domesticated animals other than swine, with one report of ubiquitous but not specifically reactive antibodies in sera obtained from goats, buffalo, and even chickens in North India.59 HEV has, however, been documented in buffalo, cattle, sheep, and goats in other regions,60-62 as well as in milk from infected cows.63

The segregation of genotypes by host species observed in India by Arankalle and colleagues56 would seem to suggest that separate patterns of circulation and transmission may exist among livestock and among humans there. They initially speculated that amino acid substitutions present in swine HEV-4 isolates might reduce the potential for cross-species transmission,64 but moderated their stance when they subsequently demonstrated that these same strains produced intermittent serum viremia and provoked an immune response in experimentally-inoculated rhesus macaques.65 Human HEV-4 infections have been documented elsewhere, including recently in India,66 so it cannot be generally assumed that South Asian swine HEV strains pose no threat of 90

infection to humans. The much higher seroprevalence of anti-HEV IgG in swine handlers than in other, comparable adults in Vellore, India,67 is also suggestive of zoonotic transmission within a

South Asian context. The possibility of HEV transmission directly from pigs or from buffalo deserves further attention.

Another possible mediator of the association of meat-eating with seropositivity in this setting is the consumption of fish, shellfish, or snails, likely harvested from contaminated local surface waters. HEV has been detected in shellfish in coastal waters off Korea,68 China,69 Spain,70 and Italy,71 most likely as a result of untreated effluent from swine farms reaching coastal estuaries.

Some bivalves may contain high concentrations of pathogens, including HEV, because they filter microbial particles from the surrounding water.71

While Nepal is a landlocked country with no coastal estuaries, the Terai is crossed by numerous streams and rivers, many seasonally-fed by meltwater from the Himalayas and by monsoonal rains, with heavy sediment burdens. A large irrigation channel diverted from the

Bagmati River –also used for livestock and recreation -- traverses the study area. Although an earlier dietary survey conducted by NNIPS found that fish and snails were infrequently consumed by women in Sarlahi District on a year-round basis,72 44% of the women in our cohort reported eating fish or snails at least once in the prior month and 7% reported eating these meats roughly once a week or more (≥4 times in the past 30 days), consistent with an independent survey in 2010 that found that the vast majority of households had consumed fish in the prior year.73 In Sarlahi, where industrial-scale pig farms are rare but open defecation is common, human waste containing

HEV-1 virions may contaminate the surface waters from which fish and snails are harvested. It is plausible that foodborne transmission could occur if fish and snails from these surface waters are not handled carefully and cooked thoroughly. If exposure to HEV via fish, shellfish, and snails were

common, however, one might expect to see an even higher seroprevalence of anti-HEV IgG in this population.

It is important to note, in considering the observed association of meat-eating with seropositivity, that there is a strong possibility of residual confounding by unmeasured or inadequately measured factors such as ethnicity and religious affiliation* or extended family relationships that may be associated both with dietary customs and with possible human-waste exposures related to neighborhood of residence.

Conclusions

Hepatitis E virus circulaties and causes illness among pregnant women in Sarlahi District,

Nepal. While pregnant women in Sarlahi appear to be at reduced risk of HEV relative to populations in the Kathmandu Valley, the incidence of HEV infection remains substantial and represents a threat to maternal health in southern Nepal. As rates of HEV infection and hepatitis E elsewhere in Asia, including in the Kathmandu Valley, vary seasonally and from year to year,34,55,76 we cannot be certain whether the 1.5-year snapshot captured by this study is wholly representative. Our estimates may be also affected by the exclusion of women with early pregnancy loss and by incomplete serologic and symptomatic follow-up data; the true prevalence, incidence, and impact of HEV on pregnancy in Sarlahi may be somewhat greater than our estimates suggest.

Furthermore, the risks of maternal HEV infection and illness in Sarlahi appear to be borne unequally by populations of different social and economic standing. Given the limited access to improved sanitation and water sources in this area, the links between these factors and individual

* Some “1,250 ethnic/caste and 207 religious groups” were documented during the 2011 Census in Nepal. 75

and community-level HEV exposure must be examined further. The dual (and seemingly contradictory) findings of increased anti-HEV IgG seroprevalence among both Muslim women and among non-vegetarians, especially pork and buffalo consumers, may hint at parallel HEV-1 and zoonotic HEV transmission pathways in the Nepali Terai. If this were true – and particularly if the production and persistence of antibodies varied with nature or frequency of exposure – it would have important implications for modeling population-level HEV dynamics and the occurrence of outbreaks. However, our study did not capture the full range of exposures that may lead to the higher seroprevalence of HEV antibodies among both Muslim and non-Muslim women from disadvantaged groups relative to women from traditionally higher Hindu castes. Environmental sampling of surface waters and groundwater sources and testing of commonly-encountered animals in Sarlahi (including fish and snails) could help clarify where and how exposures to HEV occur in this population and what measures might be most effective in reducing the risk of infection.

The increasing urbanization of this part of Sarlahi brings with it the unwelcome potential for wider spread of hepatitis E (e.g., through development of contamination-prone community water supply systems, greater waste runoff into multi-use irrigation channels, or increased reliance on surface water when local wells run dry due to high demand and inadequate groundwater recharge). In this population, with its apparently low immunity to hepatitis E, many women would be susceptible to infection and illness in the face of greater population exposure to HEV. Thus, although hepatitis E does not currently appear to be as grave a threat to pregnant women and their infants in northern Sarlahi as it is in Kathmandu, exposure patterns may change as Sarlahi itself changes. Ensuring that communities have access to clean water and sanitary facilities, and rectifying the disparities that currently exist, will help prevent hepatitis E (and other waterborne diseases) from becoming a greater threat to the residents of Sarlahi and other Terai districts.

ACKNOWLEDGMENTS

We are grateful to Kenrad Nelson, Alain Labrique, the MaGIFT PIs, the staff of NNIPS in Sarlahi and in Kathmandu, WARUN, and the study participants for their contributions to the study. The research was funded by Grant #50274 from the Bill and Melinda Gates Foundation.

FIGURES AND TABLES

Figure 5. Location of study area in Sarlahi District, Nepal. Top left: Sarlahi District (shaded light yellow) in the Terai region of southern Nepal, bordering India; Center: MaGIFT study area (shaded light yellow) in north-central Sarlahi District; Top right: Nine participating Village Development Committees (VDCs) comprising MaGIFT study area along East-West Highway (H1), with Nepal Nutrition Intervention Project Sarlahi (NNIPS) field headquarters in Hariyon marked by red dot; Bottom right: NNIPS field headquarters and eye clinic in Hariyon VDC, prior to 2012-2013 addition of student and staff lodging above.

Presence in Past 30 Days of... Score

Nausea and/or vomiting 1

Fever 1

Jaundice (yellow eyes/skin) 4

Dark urine and/or light-colored stool 2

Pain in upper right abdomen 1

Smell of food strange or unappealing 1

TOTAL HEPATITIS SCORE (sum of score column)

If Total Hepatitis Score ≥4, case is reported by phone to auxiliary nurse-midwife at field headquarters, who schedules follow-up visit as soon as possible (within 24-48 hours). Table 3. Case detection algorithm for acute hepatitis-like illness based on symptoms assessed at monthly follow-up visits.

Figure 6. Enrollment of study participants relative to the total female population of each village development committee (VDC) and ward. Top: Total female population of each VDC (left) and ward (right) from 2011 Nepal Population Census. 77. 78 Center: Number of study participants enrolled by VDC (left) and ward (right). Bottom: Study participation as a proportion of the total female population of each VDC (left) and ward (right). Note: total female population includes females of all ages, regardless of pregnancy status, while study enrollment was limited to pregnant women; bottom maps thus reflect variations in both study enrollment yield and local pregnancy prevalence. 97

Figure 7. Weekly specimen collection by specimen type over the duration of the study period. Specimen collection began 8 October 2012 and ended 2 May 2014. The gap in baseline specimen collection in summer 2013 was due to the transition from randomized, controlled vaccine trial (RCT) enrollment to open vaccine clinics / observational enrollment and follow-up in the main MaGIFT study. This change in protocol necessitated institutional review and approval of new consent forms and retraining of staff. The low number of specimens collected one week each October reflects a break for the holiday of Dashain. Strikes that occurred around the time of Nepal’s Constituent Assembly Election, held on 19 November 2013, also caused brief disruptions in sample collection.

Viral Infection | Biomarker Test Result Total Tested Assay + — ‽ Hepatitis A Virus | Anti-HAV IgG 1619 46 3 1669 Wantai HAV IgG ELISA (97.1%) (2.8%) (0.2%)

Hepatitis B Virus | Anti-HBC 436 1177 56 1669

Wantai anti-HBC ELISA (26.1%) (70.5%) (3.4%)

Hepatitis B Virus | HBSAg 26 359 9 394 Wantai AiD HBsAg ELISA (6.6%) (91.1%) (2.3%) Hepatitis C Virus | Anti-Core/NS 0 1669 0 1669 Wantai AiD anti-HCV ELISA (100%) Hepatitis E Virus | Anti-HEV IgG 195 1471 3 1669 Wantai HEV IgG ELISA (11.7%) (88.1%) (0.2%) Hepatitis E Virus | Anti-HEV IgM 1 354 0 355 Wantai HEV IgM ELISA (0.3%) (99.7%) Table 4. Baseline serologic results for markers of infection with hepatitis A, B, C, and E viruses (HAV, HBV, HCV, and HEV). All assays manufactured by Beijing Wantai Biological Pharmacy Enterprise Co., Ltd., Beijing, China. Abbreviations and symbols: +: positive test result (signal-to-cutoff ratio > 1), —: negative test result (signal- to-cutoff ratio < 1), ‽: indeterminate test result, ELISA: enzyme-linked immunosorbent assay, IgG: immunoglobulin G, IgM: immunoglobulin M, HAV: hepatitis A virus, HBV: hepatitis B virus, HCV: hepatitis C virus, HEV: hepatitis E virus, HBc: hepatitis B virus core antigen, HBsAg: hepatitis B virus surface antigen, NS: non-structural protein.

Characteristic All Enrolled Women All Women Tested Anti-HEV IgG+ at Baseline Women at Baseline n = 2363 1669 195 Age (yrs) Median 22.2 Median 22.2 Median 23.9 Mean 23.0 (SD 4.9) Mean 23.0, (SD 4.8) Mean 24.6 (SD 5.6) <20 707 (29.9%) 497 (29.8%) 45 (23.1%) 20-24 931 (39.4%) 670 (40.1%) 67 (34.4%) 25-29 514 (21.8%) 358 (21.5%) 47 (24.1%) 30-34 149 (6.3%) 100 (6.0%) 26 (13.3%) 35+ 62 (2.6%) 44 (2.6%) 10 (5.1%) Parity Primipara 1170 (49.5%) 835 (50.0%) 74 (37.9%) Multipara 1192 (50.5%) 834 (50.0%) 121 (62.1%) VDC Sasapur 172 (7.3%) 118 (7.1%) 14 (7.2%) Hariyon 527 (22.3%) 365 (21.9%) 59 (30.3%) Dhungrekhola 323 (13.7%) 230 (13.8%) 18 (9.2%) Karmaiya 219 (9.3%) 154 (9.2%) 17 (8.7%) Ghurkauli 282 (11.9%) 215 (12.9%) 22 (11.3%) Jabdi 225 (9.5%) 174 (10.4%) 19 (9.7%) Netraganj 240 (10.2%) 162 (9.7%) 15 (7.7%) Lalbandi 231 (9.8%) 162 (9.7%) 20 (10.3%) Raniganj 144 (6.1%) 89 (5.3%) 11 (5.6%) Caste Brahmin 236 (10.0%) 170 (10.2%) 10 (5.1%) Chhetri 298 (12.6%) 200 (12.0%) 16 (8.2%) Vaishya 1296 (54.8%) 946 (56.7%) 98 (50.3%) Shudra 238 (10.1%) 165 (9.9%) 29 (14.9%) Muslim 174 (7.4%) 115 (6.9%) 24 (12.3%) Other/Unknown 121 (5.1%) 73 (4.4%) 18 (9.2%) Main drinking water source Tubewell/Tap 2024 (85.7%) 1435 (86.0%) 175 (90.2%) Protected Ring Well 338 (14.3%) 233 (14.0%) 19 (9.8%) No latrine (open defecation) 1115 (47.2%) 785 (47.0%) 100 (51.3%) Washes hands before eating 2352 (99.6%) 1661 (99.5%) 195 (100.0%) Strict vegetarian 221 (9.4%) 163 (9.8%) 6 (3.1%) Eats meat 2142 (90.6%) 1506 (90.2%) 189 (96.9%) Eats pig meat 585 (24.9%) 403 (24.2%) 62 (31.8%) Eats buffalo meat 916 (38.8%) 648 (38.8%) 98 (50.3%) Eats goat/mutton 1989 (84.2%) 1387 (83.1%) 173 (88.7%) Eats fowl 1968 (83.3%) 1380 (82.7%) 177 (90.8%) Eats fish/shellfish 2012 (85.2%) 1413 (84.7%) 182 (93.3%) Stays home all day 633 (26.8%) 464 (27.8%) 42 (21.5%) Jaundiced contact (past month) 51 (2.2%) 44 (2.6%) 8 (4.1%) Table 5. Baseline characteristics of full cohort, all women providing baseline blood samples, and women seropositive for anti-HEV IgG at baseline. VDC=village development committee.

100

Figure 8. Baseline seroprevalence of anti-HEV IgG antibodies by selected exposures. Unadjusted odds ratios are shown at the top of each bar; bars labeled “OR:” represent reference categories.

101

Figure 9. Geographic variation in anti-HEV IgG seroprevalence and distribution of selected covariates within the study area. Top left: Study area map showing VDCs (village development committees) and wards. Top right: distribution of participants by caste and VDC. Center: Baseline anti-HEV IgG seroprevalence by VDC (left) and ward (right). Bottom: Prevalence of open defecation (no household latrine) by VDC (left) and ward (right). There was no association at the individual, ward, or VDC level of latrine access with anti-HEV IgG seropositivity.

102

Variable Unadjusted OR Unadj. Adjusted OR Adj. (95% CI) P-Value (95% CI) P-Value Age (Yrs) 1.08 (1.05-1.11) 0.000 1.07 (1.04-1.11) 0.000 Literate No ------Yes 0.48 (0.35-0.66) 0.000 0.87 (0.59-1.28) 0.469 Stays Home No ------Yes 0.68 (0.48-0.98) 0.039 0.69 (0.43-1.11) 0.124 VDC Hariyon ------Sasapur 0.70 (0.37-1.30) 0.259 0.87 (0.40-1.85) 0.708 Dhungrekhola 0.44 (0.25-0.77) 0.004 0.44 (0.22-0.89) 0.023 Karmaiya 0.64 (0.36-1.14) 0.134 1.09 (0.54-2.17) 0.811 Ghurkauli 0.59 (0.35-1.00) 0.048 0.66 (0.36-1.23) 0.192 Jabdi 0.64 (0.37-1.10) 0.108 0.85 (0.40-1.71) 0.662 Netraganj 0.53 (0.29-0.96) 0.038 0.83 (0.40-1.71) 0.610 Lalbandi 0.73 (0.42-1.26) 0.259 1.16 (0.63-2.12) 0.635 Raniganj 0.73 (0.37-1.46) 0.374 1.26 (0.58-2.74) 0.558 Caste Brahmin ------Chhetri 1.39 (0.61-3.15) 0.429 1.38 (0.59-3.22) 0.452 Vaishya 1.85 (0.94-3.62) 0.073 1.92 (0.93-3.97) 0.080 Shudra 3.41 (1.60-7.25) 0.001 3.32 (1.43-7.68) 0.005 Muslim 4.22 (1.93-9.22) 0.000 3.10 (1.23-7.79) 0.016 Other/Unknown 5.24 (2.28-12.03) 0.000 29.96 (3.94-227.64) 0.001 Land Owned (Khattas) 0.97 (0.95-0.98) 0.000 0.98 (0.96-0.99) 0.003 % in Ward Lacking Latrine 1.00 (1.00-1.01) 0.673 1.00 (0.99-1.01) 0.711 Water Source Tubewell/Tap ------Ring Well 0.64 (0.39-1.05) 0.077 0.86 (0.46-1.61) 0.627 Purifies Water No ------n/a n/a Yes 0.70 (0.21-2.32) 0.564 n/a n/a Washes Before Eating No ------n/a n/a Yes n/a n/a n/a n/a Eats Meat No ------Yes 3.76 (1.64-8.61) 0.002 3.03 (1.19-7.73) 0.020 Table 6. Crude and adjusted odds ratios for the association of selected covariates with baseline anti- HEV IgG seropositivity. Dashes indicate reference group for categorical variables. Adjusted odds ratios were derived from a 1556- observation-based logistic regression model that included all other covariates as shown in the table, with the exception of “purifies water” and “washes [hands] before eating”, which were removed due to the extremely small number of women who purified water or reported not washing hands prior to eating. In univariate analysis, water purification was associated with reduced odds of seropositivity, but this association was not statistically significant; all of the 8 women who reported not washing hands before eating were seronegative. Age (years) and land ownership (# of household-owned khattas of khet and/or bari land) were modeled as a linear variables because these fit better than categorical variables with a range of cut-points. 103

Figure 10. Patterns of baseline anti-HEV IgG seroprevalence by meat consumption habits and frequency in pregnant women in Sarlahi District, Nepal. Top: Baseline anti-HEV IgG seroprevalence among vegetarians, meat eaters (generally), and eaters of specific types of meat. The number of women who reported consumption of each type of meat is listed in parentheses. Note that among meat-eaters, the categories are not mutually-exclusive (i.e., a woman who ate both fish and goat would be counted in the denominator of both of those two bars, as well as in the “any meat” category). Bottom: Baseline seroprevalence of anti-HEV IgG by frequency of exposure to specific meats in the past 30 days. Vegetarians are included in the denominator of the zero-frequency categories. Both when grouped as shown, and when treated as linear variables, a test for trend was statistically-significant only for frequency of pig consumption.

104

REFERENCES

1 Labrique AB, Sikder SS, Krain LJ, West KP, Jr., Christian P, Rashid M, et al. (2012) Hepatitis E, a Vaccine-Preventable Cause of Maternal Death. Emerg Infect Dis 18(9): 1401-1404. [doi:10.3201/eid1809.120241]

2 Gurley ES, Halder AK, Streatfield PK, Sazzad HM, Huda TM, Hossain MJ, et al. (2012) Estimating the burden of maternal and neonatal deaths associated with jaundice in Bangladesh: possible role of hepatitis E infection. Am J Public Health 102(12): 2248-2254. [doi:10.2105/ajph.2012.300749]

3 Hillis A, Shrestha S, & Saha N. (1973) An epidemic of infectious hepatitis in the Kathmandu Valley. J Nepal Med Assoc 11(5 & 6): 145-154.

4 Kane MA, Bradley DW, Shrestha SM, Maynard JE, Cook EH, Mishra RP, et al. (1984) Epidemic non-A, non-B hepatitis in Nepal. Recovery of a possible etiologic agent and transmission studies in marmosets. JAMA 252(22): 3140-3145.

5 Durkin M. (1988) Ayurvedic treatment for jaundice in Nepal. Soc Sci Med 27(5): 491-495.

6 Clayson ET, Vaughn DW, Innis BL, Shrestha MP, Pandey R, & Malla DB. (1998) Association of hepatitis E virus with an outbreak of hepatitis at a military training camp in Nepal. J Med Virol 54(3): 178-182. [doi:10.1002/(SICI)1096-9071(199803)54:3<178::AID-JMV6>3.0.CO;2-2]

7 Sharma SP. (2006) Hepatitis E and cholera outbreak in Kathmandu. CMAJ 175(8): 860. [doi:10.1503/cmaj.061166]

8 Shrestha SM. (2006) Hepatitis E in Nepal. Kathmandu Univ Med J 4(4): 530-544.

9 ProMED-Mail. (2006) Hepatitis - Nepal (Birgunj). [http://promedmail.org/post/20060503.1277 Accessed 10 Febuary 2016].

10 Shrestha A, Lama TK, Karki S, Sigdel DR, Rai U, Rauniyar SK, et al. (2015) Hepatitis E Epidemic, Biratnagar, Nepal, 2014. Emerg Infect Dis 21(4): 711. [doi:10.3201/eid2104.141512]

11 Asher LV, Innis BL, Shrestha MP, Ticehurst J, & Baze WB. (1990) Virus-like particles in the liver of a patient with fulminant hepatitis and antibody to hepatitis E virus. J Med Virol 31(3): 229-233.

12 Bista BK, & Rana A. (2006) Acute hepatitis E in pregnancy--study of 16 cases. JNMA J Nepal Med Assoc 45(161): 182-185.

13 Shrestha P, Bhandari D, Sharma D, & Bhandari BP. (2009) A study of viral hepatitis during pregnancy in Nepal Medical College Teaching Hospital. Nepal Med Coll J 11(3): 192-194.

105

14 Rana A, Pradhan N, Manandhar B, Bitsta KD, Adhikari S, Gurung G, et al. (2009) Maternal mortality over the last decade: a changing pattern of death due to alarming rise in hepatitis in the latter five-year period. J Obstet Gynaecol Res 35(2): 243-251.

15 Izopet J, Labrique AB, Basnyat B, Dalton HR, Kmush B, Heaney CD, et al. (2015) Hepatitis E virus seroprevalence in three hyperendemic areas: Nepal, Bangladesh and southwest France. J Clin Virol 70(39-42. [doi:10.1016/j.jcv.2015.06.103]

16 Abe K, Li TC, Ding X, Win KM, Shrestha PK, Quang VX, et al. (2006) International collaborative survey on epidemiology of hepatitis E virus in 11 countries. Southeast Asian J Trop Med Public Health 37(1): 90-95.

17 Bhattarai A. (2007) Hepatitis E vaccine. N Engl J Med 356(23): 2421-2422; author reply 2421-2422.

18 Stevenson P. (2000) Nepal calls the shots in hepatitis E virus vaccine trial. Lancet 355(9215): 1623.

19 Basnyat B. (2010) Neglected hepatitis E and typhoid vaccines. Lancet 376(9744): 869. [doi:10.1016/S0140-6736(10)61403-1]

20 Shrestha MP, Scott RM, Joshi DM, Mammen MP, Jr., Thapa GB, Thapa N, et al. (2007) Safety and efficacy of a recombinant hepatitis E vaccine. N Engl J Med 356(9): 895-903. [doi:10.1056/NEJMoa061847]

21 Holmberg SD. (2010) Hepatitis E vaccine: not a moment too soon. Lancet 376(9744): 849-851. [doi:10.1016/S0140-6736(10)61260-3]

22 Longer CF, Shrestha MP, MacArthy PO, Myint KS, Hoke Jr CH, Ticehurst J, et al. (1994) Epidemiology of hepatitis E virus (HEV): a cohort study in Kathmandu, Nepal, in Viral Hepatitis and Liver Disease, eds Nishioka K, Suzuki H, Mishiro S, & Oda T. Springer)

23 Clayson ET, Shrestha MP, Vaughn DW, Snitbhan R, Shrestha KB, Longer CF, et al. (1997) Rates of hepatitis E virus infection and disease among adolescents and adults in Kathmandu, Nepal. J Infect Dis 176(3): 763-766.

24 Zhu FC, Zhang J, Zhang XF, Zhou C, Wang ZZ, Huang SJ, et al. (2010) Efficacy and safety of a recombinant hepatitis E vaccine in healthy adults: a large-scale, randomised, double-blind placebo- controlled, phase 3 trial. Lancet 376(9744): 895-902. [doi:10.1016/S0140-6736(10)61030-6]

25 Pagliusi S. (2012) World's first hepatitis E vaccine approved in China. WHO Global Immunization News 2012(Jan): 14. [www.who.int/immunization/GIN_January_2012.pdf Accessed May 7, 2012.]

106

26 Yang LE. (2012) China approves hepatitis E vaccine. Xinhua English News January 11, 2012. [http://news.xinhuanet.com/english/china/2012-01/11/c_131355108.htm Accessed January 29, 2012].

27 Nelson KE, Shih JW, Zhang J, Zhao Q, Xia N, Ticehurst JR, et al. (2014) Hepatitis E vaccine to prevent morbidity and mortality during epidemics. Open Forum Infect Dis 1(3): ofu098. [doi:10.1093/ofid/ofu098]

28 Wu T, Zhu FC, Huang SJ, Zhang XF, Wang ZZ, Zhang J, et al. (2011) Safety of the hepatitis E vaccine for pregnant women: A preliminary analysis. Hepatology 55(6): 2038. [doi:10.1002/hep.25522]

29 Wu T, Li SW, Zhang J, Ng MH, Xia NS, & Zhao Q. (2012) Hepatitis E vaccine development: A 14 year odyssey. Hum Vaccin Immunother 8(6). [doi:10.4161/hv.20042]

30 Nepal Central Bureau of Statistics. (2012) National Population and Housing Census 2011 (Village Development Committee/Municipality). Vol. 2 (Kathmandu, Nepal) [http://cbs.gov.np/nada/index.php/catalog/54/download/467 Accessed 06 June 2016].

31 Kathmandu Post. (2014) Govt announces 72 new municipalities Kathmandu Post 09 May 2014. [http://kathmandupost.ekantipur.com/printedition/news/2014-05-08/govt-announces-72-new- municipalities.html Accessed 19 May 2016].

32 Seriwatana J, Shrestha MP, Scott RM, Tsarev SA, Vaughn DW, Myint KS, et al. (2002) Clinical and epidemiological relevance of quantitating hepatitis E virus-specific immunoglobulin M. Clin Diagn Lab Immunol 9(5): 1072-1078.

33 Tsarev SA, Binn LN, Gomatos PJ, Arthur RR, Monier MK, van Cuyck-Gandre H, et al. (1999) Phylogenetic analysis of hepatitis E virus isolates from Egypt. J Med Virol 57(1): 68-74.

34 Labrique AB, Zaman K, Hossain Z, Saha P, Yunus M, Hossain A, et al. (2010) Epidemiology and risk factors of incident hepatitis E virus infections in rural Bangladesh. Am J Epidemiol 172(8): 952- 961. [doi:10.1093/aje/kwq225]

35 Labrique AB, Zaman K, Hossain Z, Saha P, Yunus M, Hossain A, et al. (2009) Population seroprevalence of hepatitis E virus antibodies in rural Bangladesh. Am J Trop Med Hyg 81(5): 875- 881. [doi:10.4269/ajtmh.2009.09-0352]

36 Lwanga SK, & Lemeshow S. (1991) Sample size determination in health studies: a practical manual. Sample size determination in health studies: a practical manual, 80 p. (World Health Organization: Geneva).

107

37 World Health Organization. (2016) WHO vaccine-preventable diseases: monitoring system. 2016 global summary: Nepal. [http://apps.who.int/immunization_monitoring/globalsummary/schedules?sc[c][]=NPL&sc[d]=&sc[ v][]=BCG&sc[v][]=CHOLERA&sc[v][]=DIP&sc[v][]=DT&sc[v][]=DTAP&sc[v][]=DTAPHEPBIPV&sc[v][ ]=DTAPHEPIPV&sc[v][]=DTAPHIB&sc[v][]=DTAPHIBHEP&sc[v][]=DTAPHIBHEPIPV&sc[v][]=DTAP HIBIPV&sc[v][]=DTAPIPV&sc[v][]=DTIPV&sc[v][]=DTPHIBHEP&sc[v][]=DTWP&sc[v][]=DTWPHEP& sc[v][]=DTWPHIB&sc[v][]=DTWPHIB&sc[v][]=DTWPHIBHEPB&sc[v][]=DTWPHIBHEPBIPV&sc[v][] =DTWPHIBIPV&sc[v][]=DTWPIPV&sc[v][]=HEPA&sc[v][]=HEPAHEPB&sc[v][]=HEPB&sc[v][]=HEPB _ADULT&sc[v][]=HEPB_PEDIATRIC&sc[v][]=HFRS&sc[v][]=HIB&sc[v][]=HIB&sc[v][]=HIBMENC&sc[ v][]=HPV&sc[v][]=INFLUENZA&sc[v][]=INFLUENZA_ADULT&sc[v][]=INFLUENZA_PEDIATRIC&sc[v] []=IPV&sc[v][]=JAPENC&sc[v][]=JE_INACTD&sc[v][]=JE_LIVEATD&sc[v][]=MEASLES&sc[v][]=MENA &sc[v][]=MENAC&sc[v][]=MENACW&sc[v][]=MENACWY&sc[v][]=MENBC&sc[v][]=MENC_CONJ&sc[v ][]=MM&sc[v][]=MMR&sc[v][]=MMRV&sc[v][]=MR&sc[v][]=MUMPS&sc[v][]=OPV&sc[v][]=PNEUMO _CONJ&sc[v][]=PNEUMO_PS&sc[v][]=RABIES&sc[v][]=ROTAVIRUS&sc[v][]=RUBELLA&sc[v][]=TBE& sc[v][]=TD&sc[v][]=TDAP&sc[v][]=TDAP&sc[v][]=TDAPIPV&sc[v][]=TDIPV&sc[v][]=TT&sc[v][]=TYP HOID&sc[v][]=TYPHOIDHEPA&sc[v][]=VARICELLA&sc[v][]=VITA&sc[v][]=VITAMINA&sc[v][]=YF&s c[v][]=ZOSTER&sc[OK]=OK Accessed 6 June 2016.]

38 Labrique AB. (2007) The epidemiology of hepatitis E virus infections in Bangladesh PhD thesis, Johns Hopkins University, (2007).

39 Tielsch JM, Steinhoff MC, Shrestha L, Englund JA, Katz J, Khatry SK, et al. (2012) Nepal Maternal Influenza Vaccine Trial: Mother's Gift Trial Manual of Operations. 129.

40 Pradhan EK, Katz J, LeClerq SC, & West KP, Jr. (1994) Data management for large community trials in Nepal. Control Clin Trials 15(3): 220-234.

41 Wee DW, Mullany LC, Katz J, Khatry SK, LeClerq SC, & Tielsch JM. (2012) Retraction: Pregnancy- related mortality in southern Nepal between 2001 and 2006: independent estimates from a prospective, population-based cohort and a direct sisterhood survey. Am J Epidemiol 175(5): 478.

42 Begum N, Devi SG, Husain SA, & Kar P. (2009) Seroprevalence of subclinical HEV infection in pregnant women from north India: a hospital based study. Indian J Med Res 130(6): 709-713.

43 Arankalle VA, Tsarev SA, Chadha MS, Alling DW, Emerson SU, Banerjee K, et al. (1995) Age- specific prevalence of antibodies to hepatitis A and E viruses in Pune, India, 1982 and 1992. J Infect Dis 171(2): 447-450.

108

44 Kmush BL, Labrique A, Li W, Klein SL, Schulze K, Shaikh S, et al. (2016) The Association of Cytokines and Micronutrients with Hepatitis E Virus Infection During Pregnancy and the Postpartum Period in Rural Bangladesh. Am J Trop Med Hyg 94(1): 203-211. [doi:10.4269/ajtmh.15-0238]

45 Hossain MZ, Sikder SS, Zaman K, Saha P, Yunus M, Nelson KE, et al. (2013) Screening utility, local perceptions, and care-seeking for reported jaundeesh among respondents lacking signs of icterus in rural Bangladesh. J Health Popul Nutr 31(3): 367-375.

46 Ramachandran J, Ramakrishna B, Eapen C, Abraham P, Zachariah U, Jayram A, et al. (2008) Subacute hepatic failure due to hepatitis E. J Gastroenterol Hepatol 23(6): 879. [doi:10.1111/j.1440- 1746.2007.05205.x]

47 Lee NM, & Brady CW. (2009) Liver disease in pregnancy. World J Gastroenterol 15(8): 897-906. [doi:10.3748/wjg.15.897]

48 Surya IG, Kornia K, Suwardewa TG, Mulyanto, Tsuda F, & Mishiro S. (2005) Serological markers of hepatitis B, C, and E viruses and human immunodeficiency virus type-1 infections in pregnant women in Bali, Indonesia. J Med Virol 75(4): 499-503. [doi:10.1002/jmv.20314]

49 Corwin AL, Tien NT, Bounlu K, Winarno J, Putri MP, Laras K, et al. (1999) The unique riverine ecology of hepatitis E virus transmission in South-East Asia. Trans R Soc Trop Med Hyg 93(3): 255- 260.

50 Sedyaningsih-Mamahit ER, Larasati RP, Laras K, Sidemen A, Sukri N, Sabaruddin N, et al. (2002) First documented outbreak of hepatitis E virus transmission in Java, Indonesia. Trans R Soc Trop Med Hyg 96(4): 398-404.

51 Corwin A, Jarot K, Lubis I, Nasution K, Suparmawo S, Sumardiati A, et al. (1995) Two years' investigation of epidemic hepatitis E virus transmission in West Kalimantan (Borneo), Indonesia. Trans R Soc Trop Med Hyg 89(3): 262-265.

52 Gouvea V, Snellings N, Cohen SJ, Warren RL, Myint KS, Shrestha MP, et al. (1997) Hepatitis E virus in Nepal: similarities with the Burmese and Indian variants. Virus Res 52(1): 87-96.[pii: S0168- 1702(97)00112-3]

53 Gouvea V, Snellings N, Popek MJ, Longer CF, & Innis BL. (1998) Hepatitis E virus: complete genome sequence and phylogenetic analysis of a Nepali isolate. Virus Res 57(1): 21-26. [pii: S0168- 1702(98)00079-3]

54 Shrestha SM, Shrestha S, Tsuda F, Nishizawa T, Gotanda Y, Takeda N, et al. (2003) Molecular investigation of hepatitis E virus infection in patients with acute hepatitis in Kathmandu, Nepal. J Med Virol 69(2): 207-214. [doi:10.1002/jmv.10276] 109

55 Shrestha SM, Shrestha S, Tsuda F, Nishizawa T, Takahashi M, Gotanda Y, et al. (2004) Genetic changes in hepatitis E virus of subtype 1a in patients with sporadic acute hepatitis E in Kathmandu, Nepal, from 1997 to 2002. J Gen Virol 85(Pt 1): 97-104. [doi:10.1099/vir.0.19571-0]

56 Arankalle VA, Chobe LP, Joshi MV, Chadha MS, Kundu B, & Walimbe AM. (2002) Human and swine hepatitis E viruses from Western India belong to different genotypes. J Hepatol 36(3): 417-425.

57 Widasari DI, Yano Y, Utsumi T, Heriyanto DS, Anggorowati N, Rinonce HT, et al. (2013) Hepatitis E Virus Infection in Two Different Community Settings with Identification of Swine HEV Genotype 3 in Indonesia. Microbiol Immunol. [doi:10.1111/1348-0421.12083]

58 Clayson ET, Innis BL, Myint KS, Narupiti S, Vaughn DW, Giri S, et al. (1995) Detection of hepatitis E virus infections among domestic swine in the Kathmandu Valley of Nepal. Am J Trop Med Hyg 53(3): 228-232.

59 Shukla P, Chauhan UK, Naik S, Anderson D, & Aggarwal R. (2007) Hepatitis E virus infection among animals in northern India: an unlikely source of human disease. J Viral Hepat 14(5): 310-317. [doi:10.1111/j.1365-2893.2006.00815.x]

60 El-Tras WF, Tayel AA, & El-Kady NN. (2013) Seroprevalence of hepatitis E virus in humans and geographically matched food animals in Egypt. Zoonoses Public Health 60(3): 244-251. [doi:10.1111/j.1863-2378.2012.01516.x]

61 Xu F, Pan Y, Baloch AR, Tian L, Wang M, Na W, et al. (2014) Hepatitis E virus genotype 4 in yak, northwestern China. Emerg Infect Dis 20(12): 2182-2184. [doi:10.3201/eid2012.131599]

62 Wu J, Si F, Jiang C, Li T, & Jin M. (2015) Molecular detection of hepatitis E virus in sheep from southern Xinjiang, China. Virus Genes 50(3): 410-417. [doi:10.1007/s11262-015-1194-9]

63 Huang F, Li Y, Yu W, Jing S, Wang J, Long F, et al. (2016) Excretion of infectious hepatitis E virus into milk in cows imposes high risks of zoonosis. Hepatology. [doi:10.1002/hep.28668]

64 Chobe LP, Lole KS, & Arankalle VA. (2006) Full genome sequence and analysis of Indian swine hepatitis E virus isolate of genotype 4. Vet Microbiol 114(3-4): 240-251. [doi:10.1016/j.vetmic.2005.12.007]

65 Arankalle VA, Chobe LP, & Chadha MS. (2006) Type-IV Indian swine HEV infects rhesus monkeys. J Viral Hepat 13(11): 742-745. [doi:10.1111/j.1365-2893.2006.00759.x]

110

66 Rolfe KJ, Curran MD, Mangrolia N, Gelson W, Alexander GJ, L'Estrange M, et al. (2010) First case of genotype 4 human hepatitis E virus infection acquired in India. J Clin Virol 48(1): 58-61. [doi:10.1016/j.jcv.2010.02.004]

67 Vivek R, & Kang G. (2011) Hepatitis E virus infections in swine and swine handlers in Vellore, Southern India. Am J Trop Med Hyg 84(4): 647-649. [doi:10.4269/ajtmh.2011.10-0456]

68 Song YJ, Jeong HJ, Kim YJ, Lee SW, Lee JB, Park SY, et al. (2010) Analysis of complete genome sequences of swine hepatitis E virus and possible risk factors for transmission of HEV to humans in Korea. J Med Virol 82(4): 583-591. [doi:10.1002/jmv.21730]

69 Gao S, Li D, Zha E, Zhou T, Wang S, & Yue X. (2015) Surveillance of hepatitis E virus contamination in shellfish in China. Int J Environ Res Public Health 12(2): 2026-2036. [doi:10.3390/ijerph120202026]

70 Diez-Valcarce M, Kokkinos P, Soderberg K, Bouwknegt M, Willems K, de Roda-Husman AM, et al. (2012) Occurrence of human enteric viruses in commercial mussels at retail level in three European countries. Food Environ Virol 4(2): 73-80. [doi:10.1007/s12560-012-9078-9]

71 Donia D, Dell'Amico MC, Petrinca AR, Martinucci I, Mazzei M, Tolari F, et al. (2012) Presence of hepatitis E RNA in mussels used as bio-monitors of viral marine pollution. J Virol Methods 186(1-2): 198-202. [doi:10.1016/j.jviromet.2012.06.007]

72 Campbell RK, Talegawkar SA, Christian P, LeClerq SC, Khatry SK, Wu LS, et al. (2014) Seasonal dietary intakes and socioeconomic status among women in the Terai of Nepal. J Health Popul Nutr 32(2): 198-216.[http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4216957/pdf/jhpn0032-0198.pdf]

73 Adhikari C, Acharya G, & Kristjanson P. (2011) Summary of Baseline Household Survey Results: Sarlahi, Nepal. 27pp. (Copenhagen: CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS)) [https://cgspace.cgiar.org/bitstream/handle/10568/21097/ccafs_hbs_sarlahi_nepal.pdf?sequence= 2&isAllowed=y Accessed 26 June 2016]

74 Bennett L, Dahal DR, & Govindasamy P. (2008) Caste, Ethnic and Regional Identity in Nepal: Further Analysis of the 2006 Nepal Demographic and Health Survey. 45pp. (Calverton, MD: Macro International Inc. for USAID Measure DHS Program.) [http://pdf.usaid.gov/pdf_docs/PNADM638.pdf Accessed 27 June 2016]

75 Dahal DR. (2014) Social Composition of the Population: Caste/Ethnicity and Religion in Nepal, in Population Monograph of Nepal. 438pp. (Kathmandu: Nepal Central Bureau of Statistics)

111

[http://un.org.np/sites/default/files/Population%20Monograph%20Volume%202.pdf Accessed 20 June 2016.]

76 Ippagunta SK, Naik S, Sharma B, & Aggarwal R. (2007) Presence of hepatitis E virus in sewage in Northern India: frequency and seasonal pattern. J Med Virol 79(12): 1827-1831. [doi:10.1002/jmv.21017]

77 Nepal Central Bureau of Statistics. (2012) National Population and Housing Census 2011 (Village Development Committee/Municipality). Vol. 2 (Kathmandu, Nepal) [http://cbs.gov.np/nada/index.php/catalog/54/download/467 Accessed 06 June 2016].

78 Nepal Central Bureau of Statistics. (2012) National Population Census 2011: Household and Population by Sex, Ward Level: Sarlahi. (Kathmandu) [http://cbs.gov.np/image/data/Population/Ward%20Level/19Sarlahi_WardLevel.pdf Accessed 21 June 2016]

79 Labrique AB, Kuniholm M, Engle RE, Rashid M, Ali H, Nelson KE (2009) Seroprevalence of Antibodies to Hepatitis E Virus (HEV) and HEV Incidence in Pregnant Women in Northwest Rural Bangladesh. 13th International Symposium on Viral Hepatitis and Liver Disease, Washington, DC, March 20–24, 2009.

80 Wenzel JJ, Preiss J, Schemmerer M, Huber B, & Jilg W. (2013) Test performance characteristics of Anti-HEV IgG assays strongly influence hepatitis E seroprevalence estimates. J Infect Dis 207(3): 497- 500. [doi:10.1093/infdis/jis688]

81 Caron M, & Kazanji M. (2008) Hepatitis E virus is highly prevalent among pregnant women in Gabon, central Africa, with different patterns between rural and urban areas. Virol J 5(158). [doi:10.1186/1743-422x-5-158]

82 Cevrioglu AS, Altindis M, Tanir HM, & Aksoy F. (2004) Investigation of the incidence of hepatitis E virus among pregnant women in Turkey. J Obstet Gynaecol Res 30(1): 48-52.

83 Cong W, Sui JC, Zhang XY, Qian AD, Chen J, & Zhu XQ. (2015) Seroprevalence of hepatitis E virus among pregnant women and control subjects in China. J Med Virol 87(3): 446-450. [doi:10.1002/jmv.24058]

84 Ippagunta SK, Naik S, Sharma B, & Aggarwal R. (2007) Presence of hepatitis E virus in sewage in Northern India: frequency and seasonal pattern. J Med Virol 79(12): 1827-1831. [doi:10.1002/jmv.21017]

112

85 Vivek R, Zachariah UG, Ramachandran J, Eapen CE, Rajan DP, & Kang G. (2013) Characterization of hepatitis E virus from sporadic hepatitis cases and sewage samples from Vellore, south India. Trans R Soc Trop Med Hyg 107(6): 363-367. [doi:10.1093/trstmh/trt030]

86 Jiang T, Christian P, Khatry SK, Wu L, & West KP, Jr. (2005) Micronutrient deficiencies in early pregnancy are common, concurrent, and vary by season among rural Nepali pregnant women. J Nutr 135(5): 1106-1112.

87 Schulze KJ, West KP, Jr., Gautschi LA, Dreyfuss ML, LeClerq SC, Dahal BR, et al. (2003) Seasonality in urinary and household salt iodine content among pregnant and lactating women of the plains of Nepal. Eur J Clin Nutr 57(8): 969-976. [doi:10.1038/sj.ejcn.1601632]

113

CASE REPORT: HEPATITIS E VIRUS (HEV) IN AN ACUTELY-ILL PREGNANT WOMAN IN

NEPAL, A PRELIMINARY PHYLOGENETIC ANALYSIS, AND A CAUTIONARY TALE

Lisa J. Krain

ABSTRACT

Hepatitis E virus (HEV) is a major cause of morbidity and mortality in South Asia, especially among pregnant women. Few cases have been described outside of hospital settings, and little is known about the strains of HEV causing sporadic disease in Nepal, particularly outside of the

Kathmandu Valley. Here we describe a case of sporadic, self-limiting hepatitis E in a late-second- trimester pregnant woman in southern Nepal. To help characterize the strains causing illness in pregnancy, we isolated hepatitis E virus (HEV) RNA at two time-points, one prior to the onset of symptoms and one during the acute symptomatic phase. Phylogenetic analysis of the isolates, based on a 291-nucleotide sequence flanking the junction of open reading frames 1, 2, and 3, placed this virus in a cluster with a genotype 1d strain from Morocco, only more distantly related to other

HEV isolates from the Indian subcontinent. This solution, while intriguing, is almost certainly spurious, as genotype 1d HEV has not previously been reported outside of North Africa. Further analysis of longer, more informative regions of the genome may help clarify the evolutionary relationship of this HEV strain to others circulating in the region.

114

BACKGROUND

Hepatitis E virus (HEV) is a pathogen with global reach. HEV genotypes 1 and 2, associated with human disease, circulate in areas where sanitation and access to clean drinking water are inadequate. Genotypes 3 and 4 are common in swine worldwide; they are increasingly recognized as a cause of illness in industrialized countries, often transmitted via undercooked meat.

In South Asia, both sporadic cases and waterborne outbreaks caused by HEV genotype 1 are common. The first documented outbreak in Nepal, in 1973, affected ~10,000 people in the

Kathmandu Valley.1 Through the 1970s and 1980s, Kathmandu Valley was hit by recurring epidemics.2,3 In 2014, an outbreak in the Terai (lowland) of Nepal gripped the city of Biratnagar for several months, sickened ~7000, and killed more than 15 people, including at least two pregnant women.4-6

Analyses of historic hepatitis E epidemics have highlighted the elevated risk of severe illness and death in pregnant women.2,7 More recently, studies in Bangladesh have demonstrated the contribution of sporadic hepatitis E to maternal morbidity and mortality in South Asia.8,9 Even in women who survive the illness, HEV is also associated with adverse pregnancy outcomes, including premature delivery and fetal loss.10-12

Despite the heavy public health and economic impact13 of hepatitis E in Nepal, little is known about the strains of HEV causing illness in the country, particularly outside of the

Kathmandu Valley. The complete genome sequence of only one HEV isolate from Nepal, collected from a pregnant woman in the Kathmandu Valley in 1992, has been published to date.14

In this report, we describe a case of sporadic hepatitis E in a pregnant woman in Sarlahi

District, in the Terai of Nepal, and discuss a preliminary phylogenetic analysis of the HEV strain with which she was infected. 115

METHODS

Study Overview

We conducted a community-based, observational cohort study of ~1700 pregnant women in nine villages along the East-West Highway in Sarlahi District, Nepal, [Figure 11] from October

2012 – April 2014. The study protocol received ethical review and approval from the Nepal Health

Research Council (Reg. #92/2011), Johns Hopkins University (IRB #00002458), and the Walter

Reed Army Institute of Research (WRAIR #1973).

The cohort study was conducted in conjunction with an ongoing influenza vaccine trial15 carried out under the auspices of the Nepal Nutrition Intervention Project – Sarlahi (NNIPS).

Married women in the nine villages were visited monthly by field staff. Women who had missed menstrual periods were offered pregnancy tests and, if pregnant, were recruited into the study with their informed consent. Questionnaires covering household, demographic, educational, occupational, and behavioral factors were administered at the time of enrollment. Blood samples were collected from all participants at enrollment, within one week of delivery, and at three months postpartum. Pregnancy and birth outcomes were recorded by a dedicated team from NNIPS.

Identification of Acute Hepatitis Cases

Women were visited monthly during pregnancy by trained team leader-interviewers (TLIs) and asked questions about symptoms of illness. Those women meeting minimum threshold criteria for symptoms of acute hepatitis (e.g., jaundice and/or a combination of fever, nausea/vomiting, dark urine/pale stools, weakness, and dysosmia) were classified as potential cases of viral hepatitis

[cf. previous chapter, Table 3] and visited by a team of auxiliary nurse-midwives (ANMs). At these visits, the ANMs inquired about symptom onset, duration, and management; drew an additional

116

blood sample; performed a rapid HEV IgM assay (Wantai BP Enterprise Co., Ltd., Beijing, China) in the field; and provided referral information for local hospitals.

Laboratory Analyses

All blood samples were processed in the field lab at NNIPS’ headquarters in Hariyon, Sarlahi

District, Nepal. Sera were aliquoted and shipped in liquid nitrogen to Kathmandu, where they were tested for antibodies to hepatitis E virus and other hepatitides at the Walter Reed Research Unit –

Nepal (WARUN).

WARUN attempted to isolate RNA from anti-HEV IgM-positive samples, and isolates were forwarded to the Armed Forces Research Institute of Medical Science (AFRIMS) in Bangkok,

Thailand, for molecular analysis.

Molecular and Phylogenetic Analysis

AFRIMS amplified and sequenced a 291-nucleotide segment of HEV RNA, flanking the junction of open reading frames (ORFs) 1, 2, and 3 of the HEV genome. This region corresponds to nucleotides 5011-5301 of the Burma strain (GenBank accession # M73218).

We queried GenBank using the Sarlahi sequence data to identify highly homologous sequences. We also obtained sequences representing a variety of genotype 1 HEV isolates from

Asia, as well as genotypes 2, 3, 4, and avian HEV, from the database.

To conduct phylogenetic analysis, we used CLUSTALW16 in MEGA software (v.6)17 to align the sequences. We generated phylogenetic trees using the maximum-likelihood method18 with

1000 bootstrap replicates.

117

CASE DESCRIPTION

PG, a primagravida in her mid-20s living in a village in northern Sarlahi District, was a participant in the observational cohort study. [Figure 12] Based on the self-reported date of her last menstrual period, she was estimated to be ~25 weeks pregnant at the time of enrollment. At enrollment, she reported no recent (30-day) history of fever, pain, jaundice, vomiting, or anorexia.

A routine baseline blood sample was collected 2 weeks after initial enrollment.

At her first monthly symptom follow-up visit, 30 days after enrollment, PG reported jaundice, anorexia, and dark urine of several weeks’ duration. This triggered a visit by auxiliary nurse midwives (ANMs) one day later. With PG’s consent, the ANMs collected a blood sample, and they confirmed acute hepatitis E in the field via rapid anti-HEV IgM assay. They provided PG with information on hepatitis E and local medical facilities, as well as a printed card noting the field diagnosis of hepatitis E.

PG recovered fully from hepatitis E without additional complications. At her second monthly follow-up visit, 60 days after enrollment, PG reported no lingering symptoms of hepatitis

E, though she had lost 1.7 kg of body weight during her illness. She regained the weight by the third monthly follow-up visit.

She went on to give birth to a full-term, healthy baby boy weighing 3560 g and measuring

51.3 cm in length. PG could not be contacted to provide a routine post-delivery blood sample, but she provided one additional scheduled blood sample three months post-partum. Both mother and infant remained healthy through the end of the study follow-up period, 6 months after delivery.

 PG is a pseudonym used to protect the participant’s privacy.

118

FURTHER INVESTIGATION

Exposure History

PG’s responses to the questionnaire administered at enrollment indicated that she spent most of her time in or near her home: performing housework, tending crops, and gathering fuel for the stove. She reported no recent history of travel (within 30 days prior to enrollment) outside of her village.

The household’s primary source of water for drinking, cooking, bathing, and washing dishes and clothing was a shared tubewell, and some drinking water was stored in a communal covered container shared with other families. PG also frequently (9 times in the 30 days prior to enrollment) consumed fish, shellfish, or snails, likely sourced from local rivers. PG stated that she had not eaten any foods away from home, nor consumed prepared foods bought from the haat bazaar (market) or from street vendors, within the 14 days prior to enrollment.

Her household, like many others in the area, lacked a latrine; she practiced open defecation.

PG reported regular hand-washing with soap and water following defecation and prior to handling food. No other household members, including another pregnant relative, were reported to have symptoms prior to or concurrent with PG; however, we could not collect blood from other family members to determine whether or not they may have had evidence of asymptomatic infection.

Serology

In the laboratory, all three available samples from PG (baseline at 27 weeks gestation, symptomatic case visit at 29 weeks, and 3 months postpartum) tested positive for anti-HEV IgG, while only the baseline and case visit samples had detectable levels of anti-HEV IgM. HEV RNA was isolated from both the baseline sample and from the case visit sample, collected 2 weeks apart. The

119

woman had antibodies to HAV IgG and HBc at enrollment, but was negative for anti-HAV IgM,

HBSAg, and anti-HCV IgG and IgM. [Table 7]

Molecular Analysis and Phylogenetics

Preliminary molecular sequencing and phylogenetic analyses were performed by the Armed

Forces Research Institute of Medical Sciences (AFRIMS) in Bangkok, Thailand.

Two 291nt sequences were isolated from samples taken two weeks apart. These two sequences, flanking the junction region of HEV open reading frames (ORFs) 1, 2, and 3, were 99.7% identical, differing only in a C-T polymorphism at position 286/291. The Sarlahi sequences were

98-99% identical to the closest matches in GenBank.

Phylogenetic analysis of these 291nt sequences performed by AFRIMS placed them in a cluster with a genotype 1d HEV isolated from Morocco in 1994.19 We were able to independently replicate this clustering using the 291nt sequences provided by AFRIMS, but this solution had poor bootstrap support. [Figure 13] Additional trees produced using the neighbor-joining method, restricted to HEV genotypes 1-4, and with different subsets of sequences included, also showed this clustering, though with different overall topologies.

DISCUSSION

We have described a sporadic, uncomplicated case of acute hepatitis E in a 2nd-trimester pregnant woman. Based on the woman’s lack of travel, the infection appears to have been acquired locally in the Terai of Nepal during the winter season. She may have been exposed by drinking untreated water, or by eating fish, shellfish, or snails from local waters contaminated with human waste. She may also have come into contact with infectious human waste in the fields near her home, where open defecation is routinely practiced. It is worth noting, however, that none of these exposures is remarkable; in the larger cohort of ~2300 pregnant women from Sarlahi District, fully 120

98% regularly drank untreated well water, 47% lacked access to a latrine at home, and 94% routinely consumed locally-caught fish or shellfish.

The HEV genomic sequence isolated from the woman’s serum is of genotype 1. This viral strain appears to be distinct from the HEV strain that caused an outbreak in Biratnagar, in the eastern Terai of Nepal, approximately one year later. [Phylogenetic results including the Biratnagar sequences have been omitted by agreement with Akbar et al., who provided an unpublished full- length sequence from Biratnagar for comparison.]

Preliminary phylogenetic analyses of a small segment of the HEV genome from Sarlahi suggested a closer evolutionary relationship with hepatitis E viruses isolated in Morocco and

Algeria in the 1970s-1980s (sometimes referred to as subgenotype 1d) than with any of the HEV strains known to be circulating in Asia. However, in the absence of corroborating evidence, it would be imprudent to accept this conclusion. How such a geographically-divergent strain might materialize in the Terai of Nepal is unclear; the possibilities would appear to be direct or indirect import by a foreign visitor or a returning citizen. International tourism is Nepal’s primary source of income, with ~800,000 foreigners arriving annually at Tribhuvan International Airport in 2012 and in 2013, but tourism from Africa is minimal.20 Remittances from Nepali migrant laborers (often in the Middle East) make up the second-largest (and an increasing) share of Nepal’s gross domestic product,21 but a North African HEV strain would certainly be an unusual “remittance”! Dozens of

Nepalis from communities in the Terai22 join the ~1.6 million Muslims worldwide, including some from North Africa, who go on hajj (pilgrimage) to Mecca each year, where crowded and often unsanitary conditions increase the risk of enteric disease.23,24 In any of these scenarios, however, the plausibility of someone being exposed to a North African HEV strain and importing it into Nepal appears vanishingly small, given the chain of contingencies that such an encounter would require.

121

Another possible explanation for this result is contamination of the sample in the laboratory. Relatively few North African isolates exist, and we are unaware of such specimens having been analyzed in AFRIMS’ Bangkok laboratory, but we cannot rule out the possibility.

Perhaps the most likely explanation for the unexpected phylogenetic result we obtained is related to the structure of the HEV genome itself in the sequenced region. The junction region of the three open reading frames is a highly-conserved region of the genome.25 On the one hand, this means that we would expect few differences between strains of the same genotype, even from different regions. On the other hand, it means than any (convergent?) mutations within this region may be given undue weight by a phylogenetic tree-generating algorithm. Two uncommon polymorphisms are shared by both the Sarlahi specimens and the Morocco strain: a G < A mutation at position 48/291 (also present in a Chinese hepatitis E patient from Xinjiang, GenBank accession

#D11093) and a T < C mutation at position 200/291 (also present in a 2007 isolate from an acute liver failure patient in Pune, India, GenBank accession #JF443726, which is 98-99% identical to the

Sarlahi specimens and the closest match to both Sarlahi specimens in GenBank). [Figure 14] These two mutations, along with limited variation in the sequenced segment overall, may explain the surprising clustering we observed.

Sequencing of HEV genomic material from PG’s remaining specimens (which were sent directly from Nepal to the United States and have not been opened since they were aliquoted) would help clarify the phylogenetic position of the Sarlahi HEV strain. Sequencing by an independent laboratory would help ensure that our results are not attributable to contamination of the Sarlahi sample with a North African strain. Obtaining better coverage of the genome will provide better resolution of the phylogeny and help guard against statistical flukes like the one probably responsible for the tree we obtained.

122

The pathogenesis of hepatitis E virus in pregnancy – and in general – has been challenging to unravel. Hormonal and immunological changes,26 viral replication in the placenta,27 and other pregnancy-specific factors may predispose some women to poorer outcomes, though disentangling cause and effect can be difficult.28-32 Immunogenetics,33,34 nutritional status,35 toxic exposures,36 and coinfections,37 among others, seem to influence susceptibility to HEV infection and illness.

Genomic variability over time and space in circulating virus strains may also play a role, likely in interaction with these myriad host and environmental factors. Recent studies have begun looking at the association of viral genomic variability with the development and course of hepatitis E in pregnant and nonpregnant individuals,38-40 but limited geographic coverage and small sample sizes afford poor resolution and ability to detect clear signals.

This case, which occurred against a background of relatively low HEV seroprevalence and low hepatitis E incidence in the Terai of Nepal, ultimately resolved with a good outcome for both PG and her infant. Many women who become infected with hepatitis E during pregnancy are not so fortunate. Improvements in access to clean water and sanitary facilities (and possibly vaccines) will ultimately reduce the risk of water-borne diseases like hepatitis E in developing countries.

Meanwhile, efforts to characterize the diversity of hepatitis E viral strains may help shed light on the evolution, geographic transmission patterns, and pathogenesis of this formidable viral foe.

ACKNOWLEDGMENTS

The work was funded by grant #185081 from the Bill and Melinda Gates Foundation. We are grateful to everyone at NNIPS, especially the flu study field workers in Sarlahi and the data center staff in Kathmandu, for their hard work in support of the hepatitis study. We thank WARUN in Kathmandu, Nepal, and AFRIMS in Bangkok, Thailand, for providing laboratory support. We thank John Ticehurst for his helpful comments regarding HEV genotyping. Finally, we express our

123

sincere appreciation to PG and the ~2300 other pregnant women in Sarlahi who participated in the study.

124

FIGURES AND TABLE

Figure 11. Location of the study area in Sarlahi District, Nepal. Left: Location of Sarlahi District in the Terai (southern lowlands) of Nepal. Center: Study area within Sarlahi District. Right: Nine participating village development committees located along the East-West Highway in Sarlahi.

125

Figure 12. Timeline of PG's hepatitis symptoms and serologic results relative to her pregnancy and study participation. Abbreviations: + = positive for biomarker, - = negative for biomarker, IgG = anti-HEV immunoglobulin G, IgM = anti-HEV immunoglobulin M, RNA = HEV RNA detected by polymerase chain reaction. Trimester was estimated from PG’s self-reported last menstrual period.

126

Assay Anti- Anti- Anti- Anti- Anti- Anti- Anti- HEV HAV HAV HBsAg HCV HCV HEV HEV HBc RNA Specimen IgG IgM IgG IgM IgG IgM Baseline + nt + nt – nt + + + (~27 weeks) Symptomatic Case Visit nt – nt – nt – + + + (~29 weeks) 3 Months nt nt + – – nt + – nt Postpartum Table 7. Serologic results of viral hepatitis assays for participant PG at three time-points. PG had evidence of prior infection with HAV and HBV at baseline, but was negative for markers of acute infection with hepatitis viruses other than hepatitis E virus. Notation: + = positive test result, – = negative test result, nt=not tested, IgG = immunoglobulin G, IgM = immunoglobulin M, HAV = hepatitis A virus, HBc = hepatitis B virus core antigen, HBsAg = hepatitis B virus surface antigen, HCV = hepatitis C virus, HEV = hepatitis E virus. Serologic assays used were Wantai HAV IgG ELISA, HAV IgM ELISA, AiD HBsAg ELISA, AiD HCV ELISA, HEV IgG ELISA, and HEV IgM ELISA (Beijing Wantai BP Enterprise Co., Beijing, China). HEV RNA was detected by RT-PCR. Gestational age in weeks, estimated from PG’s self-reported last menstrual period, shown for baseline and acute sympatomatic case visit specimens.

127

Figure 13. Preliminary phylogenetic tree of hepatitis E virus (HEV) isolated from pregnant woman in Sarlahi District, Nepal, based on 189 nucleotides at the junction of open reading frames 1, 2, and 3. Tree produced with MEGA v.6.06 software using the maximum-likelihood method.41 Caveat: This tree topology may be “maximally-likely” given the sequences included and the assumptions built into the algorithm, but it is also likely illusory! 128

Figure 14. Alignment of 291-nucleotide sequences showing uncommon variants shared by Sarlahi and Morocco hepatitis E virus (HEV) strains at positions 48 and 200. Sequence alignment produced using ClustalW42 within MEGA v6.06 software.41

129

REFERENCES

1 Hillis A, Shrestha S, & Saha N. (1973) An epidemic of infectious hepatitis in the Kathmandu Valley. J Nepal Med Assoc 11(5 & 6): 145-154.

2 Teo CG. (2012) Fatal outbreaks of jaundice in pregnancy and the epidemic history of hepatitis E. Epidemiol Infect 140(5): 767-787. [doi:10.1017/s0950268811002925]

3 Shrestha SM. (2006) Hepatitis E in Nepal. Kathmandu Univ Med J 4(4): 530-544.

4 Shrestha A, Lama TK, Karki S, Sigdel DR, Rai U, Rauniyar SK, et al. (2015) Hepatitis E Epidemic, Biratnagar, Nepal, 2014. Emerg Infect Dis 21(4): 711. [doi:10.3201/eid2104.141512]

5 Kathmandu Post. (2014) Jaundice toll hits a dozen. Kathmandu Post 21 May 2014. [http://kathmandupost.ekantipur.com/news/2014-05-21/jaundice-toll-hits-a-dozen.html Accessed 17 February 2016].

6 Kattar P. (2015) After Nepal quakes, worries in the water. [Video] 00:06:34 [http://www.nytimes.com/video/health/100000003883905/nepals-waterborne-worries.html Accessed 22 February 2016].

7 Nayak NC, Panda SK, Datta R, Zuckerman AJ, Guha DK, Madanagopalan N, et al. (1989) Aetiology and outcome of acute viral hepatitis in pregnancy. J Gastroenterol Hepatol 4(4): 345-352.

8 Labrique AB, Sikder SS, Krain LJ, West KP, Jr., Christian P, Rashid M, et al. (2012) Hepatitis E, a Vaccine-Preventable Cause of Maternal Death. Emerg Infect Dis 18(9): 1401-1404. [doi:10.3201/eid1809.120241]

9 Gurley ES, Halder AK, Streatfield PK, Sazzad HM, Huda TM, Hossain MJ, et al. (2012) Estimating the burden of maternal and neonatal deaths associated with jaundice in Bangladesh: possible role of hepatitis E infection. Am J Public Health 102(12): 2248-2254. [doi:10.2105/ajph.2012.300749]

10 Patra S, Kumar A, Trivedi SS, Puri M, & Sarin SK. (2007) Maternal and fetal outcomes in pregnant women with acute hepatitis E virus infection. Ann Intern Med 147(1): 28-33. [pii:147/1/28]

11 Rasheeda CA, Navaneethan U, & Jayanthi V. (2008) Liver disease in pregnancy and its influence on maternal and fetal mortality: a prospective study from Chennai, Southern India. Eur J Gastroenterol Hepatol 20(4): 362-364. [doi:10.1097/MEG.0b013e3282f246d6]

12 Krain LJ, Atwell JE, Nelson KE, & Labrique AB. (2014) Fetal and neonatal health consequences of vertically transmitted hepatitis E virus infection. Am J Trop Med Hyg 90(2): 365-370. [doi:10.4269/ajtmh.13-0265]

13 Clark KL, Howell RM, Scott RM, Vaughn DW, Shrestha MP, Longer CF, et al. (1999) The socioeconomic impact of hepatitis E in Nepal. Am J Trop Med Hyg 61(3): 505-510.

14 Gouvea V, Snellings N, Popek MJ, Longer CF, & Innis BL. (1998) Hepatitis E virus: complete genome sequence and phylogenetic analysis of a Nepali isolate. Virus Res 57(1): 21-26.

15 Tielsch JM, Steinhoff M, Katz J, Englund JA, Kuypers J, Khatry SK, et al. (2015) Designs of two randomized, community-based trials to assess the impact of influenza immunization during pregnancy on respiratory illness among pregnant women and their infants and reproductive outcomes in rural Nepal. BMC Pregnancy Childbirth 15(40. [doi:10.1186/s12884-015-0470-y] 130

16 Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, et al. (2007) Clustal W and Clustal X version 2.0. Bioinformatics 23(21): 2947-2948. [doi:10.1093/bioinformatics/btm404]

17 Tamura K, Stecher G, Peterson D, Filipski A, & Kumar S. (2013) MEGA6: Molecular Evolutionary Genetics Analysis version 6.0. Mol Biol Evol 30(12): 2725-2729. [doi:10.1093/molbev/mst197]

18 Tamura K, Peterson D, Peterson N, Stecher G, Nei M, & Kumar S. (2011) MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28(10): 2731-2739. [doi:10.1093/molbev/msr121]

19 Benjelloun S, Bahbouhi B, Bouchrit N, Cherkaoui L, Hda N, Mahjour J, et al. (1997) Seroepidemiological study of an acute hepatitis E outbreak in Morocco. Res Virol 148(4): 279-287.

20 United Nations World Tourism Organization. (2015) Compendium of tourism specifics: Nepal country-specific basic indicators 2009 - 2013 [http://www.e- unwto.org/doi/pdf/10.5555/unwtotfb0524010020092013201504 Accessed 06 June 2016].

21 World Bank Databank. (2016) Personal remittances, received (% of GDP) [http://data.worldbank.org/indicator/BX.TRF.PWKR.DT.GD.ZS/countries/1W-NP?display=default Accessed 06 June 2016].

22 Nepal National Muslim Commission. (2014) Central Hajj Committee - List of Hajjaj. [http://www.nmc.gov.np/site/cms/14 Accessed 07 June 2016.]

23 Rafiq SM, Rashid H, Haworth E, & Booy R. (2009) Hazards of hepatitis at the Hajj. Travel Med Infect Dis 7(4): 239-246. [doi:10.1016/j.tmaid.2008.09.008]

24 Erdem H, Ak O, Elaldi N, Demirdal T, Hargreaves S, Nemli SA, et al. (2016) Infections in travellers returning to Turkey from the Arabian peninsula: a retrospective cross-sectional multicenter study. Eur J Clin Microbiol Infect Dis 35(6): 903-910. [doi:10.1007/s10096-016-2614-z]

25 Meng J, Cong M, Dai X, Pillot J, Purdy MA, Fields HA, et al. (1999) Primary structure of open reading frame 2 and 3 of the hepatitis E virus isolated from Morocco. J Med Virol 57(2): 126-133.

26 Bose PD, Das BC, Kumar A, Gondal R, Kumar D, & Kar P. (2011) High viral load and deregulation of the progesterone receptor signaling pathway: association with hepatitis E-related poor pregnancy outcome. J Hepatol 54(6): 1107-1113. [doi:10.1016/j.jhep.2010.08.037]

27 Bose PD, Das BC, Hazam RK, Kumar A, Medhi S, & Kar P. (2014) Evidence of extrahepatic replication of Hepatitis E virus in human placenta. J Gen Virol. [doi:10.1099/vir.0.063602-0]

28 Krain LJ, Nelson KE, & Labrique AB. (2014) Host Immune Status and Response to Hepatitis E Virus Infection. Clin Microbiol Rev 27(1): 139-165. [doi:doi:10.1128/CMR.00062-13]

29 Renou C, Pariente A, Nicand E, & Pavio N. (2008) Pathogenesis of Hepatitis E in pregnancy. Liver Int 28(10): 1465; author reply 1466. [doi:10.1111/j.1478-3231.2008.01885.x]

30 Robinson DP, & Klein SL. (2012) Pregnancy and pregnancy-associated hormones alter immune responses and disease pathogenesis. Horm Behav 62(3): 263-271. [doi:10.1016/j.yhbeh.2012.02.023]

31 Navaneethan U. (2009) Seroprevalence of hepatitis E infection in pregnancy - more questions than answers. Indian J Med Res 130(6): 677-679. 131

32 Navaneethan U, Al Mohajer M, & Shata MT. (2008) Hepatitis E and pregnancy: understanding the pathogenesis. Liver Int 28(9): 1190-1199. [doi:10.1111/j.1478-3231.2008.01840.x]

33 Devi SG, Kumar A, Kar P, Husain SA, & Sharma S. (2014) Association of pregnancy outcome with cytokine gene polymorphisms in HEV infection during pregnancy. J Med Virol 86(8): 1366-1376. [doi:10.1002/jmv.23925]

34 Mishra N & Arankalle VA. (2011) Association of polymorphisms in the promoter regions of TNF- alpha (-308) with susceptibility to hepatitis E virus and TNF-alpha (-1031) and IFN-gamma (+874) genes with clinical outcome of hepatitis E infection in India. J Hepatol 55(6): 1227-1234. [doi:10.1016/j.jhep.2011.03.023]

35 Kmush BL, Labrique A, Li W, Klein SL, Schulze K, Shaikh S, et al. (2016) The Association of Cytokines and Micronutrients with Hepatitis E Virus Infection During Pregnancy and the Postpartum Period in Rural Bangladesh. Am J Trop Med Hyg 94(1): 203-211. [doi:10.4269/ajtmh.15-0238]

36 Heaney CD, Kmush B, Navas-Acien A, Francesconi K, Gossler W, Schulze K, et al. (2015) Arsenic exposure and hepatitis E virus infection during pregnancy. Environ Res 142(273-280. [doi:10.1016/j.envres.2015.07.004]

37 Jacobs C, Chiluba C, Phiri C, Lisulo MM, Chomba M, Hill PC, et al. (2014) Seroepidemiology of hepatitis E virus infection in an urban population in Zambia: strong association with HIV and environmental enteropathy. J Infect Dis 209(5): 652-657. [doi:10.1093/infdis/jit409]

38 Borkakoti J, Ahmed G, Hussain SA, Rai A, & Kar P. (2014) Novel molecular alterations in the ORF 2 capsid gene of hepatitis E virus in patients with acute liver failure in North India. Arch Virol 159(12): 3391-3394. [doi:10.1007/s00705-014-2198-9]

39 Kumar S, Pujhari SK, Chawla YK, Chakraborti A, & Ratho RK. (2011) Molecular detection and sequence analysis of hepatitis E virus in patients with viral hepatitis from North India. Diagn Microbiol Infect Dis 71(2): 110-117. [doi:10.1016/j.diagmicrobio.2011.06.019]

40 Mishra N, Walimbe AM, & Arankalle VA. (2013) Hepatitis E virus from India exhibits significant amino acid mutations in fulminant hepatic failure patients. Virus Genes 46(1): 47-53. [doi:10.1007/s11262-012-0833-7]

41 Tamura K, Stecher G, Peterson D, Filipski A, & Kumar S. (2013) MEGA6: Molecular Evolutionary Genetics Analysis version 6.0. Mol Biol Evol 30(12): 2725-2729. [doi:10.1093/molbev/mst197]

42 Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, et al. (2007) Clustal W and Clustal X version 2.0. Bioinformatics 23(21): 2947-2948. [doi:10.1093/bioinformatics/btm404]

132

THE 2014 BIRATNAGAR, NEPAL HEPATITIS E OUTBREAK IN CONTEXT

Lisa J. Krain, a Birendra Prasad Gupta, c John R. Ticehurst, a

Alain B. Labrique, ba and Kenrad E. Nelson a

a Department of Epidemiology and b Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, United States; c Central Department of Biotechnology, Tribhuvan University, Kirtipur, Nepal ABSTRACT

A large outbreak of hepatitis E occurred in Biratnagar, a metropolitan area in southern

Nepal, from April-July 2014. We tested human serum samples collected from healthy adult populations in Biratnagar and nearby Dharan a few months prior to the outbreak, as well as serum samples collected during the outbreak (week of April 27 - May 3, 2014) from individuals presenting with hepatitis-like symptoms at one public hospital, one non-profit teaching hospital, and one private hospital in Biratnagar. We found a low population seroprevalence of IgG antibodies to hepatitis E virus (HEV) in healthy adults (college faculty, students, and staff) in Biratnagar (~5.6%) and Dharan (~12.3%) in January 2014, prior to the outbreak. Among 450 ambulatory patients presenting at Biratnagar hospitals during the epidemic, 42.2% (190/450) were IgG+, 14.7%

(66/450) were IgM+, and 2.9% (13/450) were HEV Ag+. Seroprevalence of acute infection markers

(anti-HEV IgM and HEV Ag) varied geographically and by hospital. Neither IgM nor Ag positivity was uniquely associated with the presence of RNA. HEV RNA recovered from epidemic patients was of genotype 1, but subgenotype could not be consistently determined. The outbreak underscores the need for improved access to water, sanitation, and hygiene in the Terai – where population susceptibility to HEV is high – and the need for improved epidemiologic surveillance and response.

133

BACKGROUND

Hepatitis E virus (HEV) is spread primarily through fecally-contaminated water and food in developing countries. The virus has an incubation period of roughly 3-6 weeks. HEV infections can be asymptomatic or subclinical, but they can also result in acute hepatitis E, with classic viral hepatitis symptoms including jaundice, vomiting, and elevated liver enzymes. Acute hepatitis E is usually self-limiting and without sequelae, with full recovery within one to two months of symptom onset. Severe hepatitis E, however, can lead to acute liver failure and death.1,2 In contrast to hepatitis A virus, which is similar to HEV in mode of transmission and clinical presentation, historical data indicate that HEV tends to cause more clinical illness in adults than in young children. Pregnant women, as well as individuals with pre-existing liver disease, are particularly susceptible to adverse outcomes and death due to hepatitis E. A 3-dose recombinant HEV vaccine has been commercially available in China since 2012, but it has not been prequalified by the World

Health Organization for use elsewhere.3-6

Nepal and India are considered endemic zones for hepatitis E virus in humans. Major epidemics in Kathmandu,7-10 Kanpur,11,12 Delhi,13,14 and other cities15,16 have sickened >150,000 people over the last century,17 and every year smaller outbreaks and sporadic cases are reported.18,19 Despite this, hepatitis E epidemics have not been well-documented in the populous

Terai (lowland region) of Nepal, which borders India. In 2005-2006, in the midst of the country’s

Maoist uprising, outbreaks of jaundice – assumed to be hepatitis E – occurred in Birgunj, Parsa

District.10,20 The April-June 2014 outbreak in Biratnagar is the first confirmed and largest known hepatitis E epidemic in the Terai to date.

Biratnagar is a city of over 200,000 people21 located in Morang District, Nepal. It is the largest city in the Terai and an important national center of commerce and education. The city

134

straddles the Koshi Highway, a paved north-south road that serves as a major truck route to and from India, which borders Biratnagar to the south-southwest. [Figure 15]

Development of a large-scale, piped water supply system in Biratnagar began in the 1970s, with aid from the United Nations and the World Bank.22 The water supply system was expanded in several phases.23 The Nepal Water Supply Corporation – Biratnagar Branch (NWSC-B) serves a majority of households in the city.24 Municipal water is pumped from deep wells to overhead tanks, minimally treated with chlorine, and supplied to the population for ~12 hours/day on a schedule.24-

26 The Mangadh Water Supply and Sanitation Users Committee, a community-run organization, supplies treated water to wards in northern Biratnagar and environs.27 Many residents also use shallower (20-30m deep) local tubewells to supplement the municipal supply.24

Biratnagar still lacks a system of closed sewers and a wastewater treatment facility, and many residents still do not have access to sanitary latrines.28,29 Numerous organizations have worked to improve sanitation in the city via latrine construction in poorer areas,30,31 hygiene eduation, and even local filming of an anti-open-defecation television comedy, ”Sugandhapur”

(“Fragrant Village”).32,33 However, major municipal and regional infrastructure development projects have been hindered by financial and political conditions.24 Major water supply pipes dating to the first wave of development in the 1970s have not aged well. Many of these deteriorating water supply pipes lie within open waste water channels. Surveys conducted between 2008 and

2012 found microbial contamination, including fecal coliform bacteria, in the majority of tapwater samples.31,34,35 Wastewater runoff from the city contaminates surface water supplies and shallow tubewells.24,36,37 In November 2013, the Asian Development Bank awarded a US$24 million contract for the long-awaited construction of sewerage and storm water drainage systems and a wastewater treatment plant in Biratnagar.29

135

Against this backdrop, in mid-April 2014, people with jaundice began appearing in increasing numbers at Biratnagar health facilities, and in late April the epidemic began receiving media and government attention.38 Hospitals were deluged with hepatitis patients and lacked the beds and supplies needed to attend to all of the ill.39 The radius of reported cases grew to encompass areas outside the municipality,40 the district prison in Biratnagar was affected41 and schools were closed.40,42 Cases peaked by early May43 and diminished considerably by the arrival of the monsoon season in late June. [Figure 16] A total of 2466 hepatitis cases and 12 deaths were officially recorded,44 but as many as ~7000 may have been ill.43 At least two pregnant women, both from outlying villages, were reported to have died from hepatitis E during the outbreak,45,46 contrary to earlier reports.43

We undertook this study to better characterize the 2014 Biratnagar hepatitis E epidemic in terms of prior population susceptibility, burden and geographic distribution of cases presenting at area hospitals, correspondence between different markers of infection, and molecular characteristics of the virus.

METHODS

The research was approved by the Nepal Health Research Council (Reg. No. 153/2014) and the Johns Hopkins School of Public Health Institutional Review Board (IRB#00005879).

Study Populations

Pre-outbreak samples. As part of an independent study conducted to establish standard biochemical values for lab tests in a healthy Nepali population (NHRC Reg. No. 57/2014), serum samples were collected in January 2014 from 110 consenting adults from three sites in Biratnagar,

Morang District, and from 88 adults in Dharan, Sunsari District (pop. ~120,00021). De-identified

136

sera, along with basic demographic data (age and sex) and results of liver function tests (LFTs), were released to the authors for use in this study.

Epidemic-period samples. Serum samples were collected during the outbreak (week of April

27 - May 3, 2014) from one government hospital (Koshi Zonal Hospital), one non-profit teaching hospital (Nobel Medical College Teaching Hospital), and one private hospital (Biratnagar Aspataal,

Pvt.) in Biratnagar. Participants were selected sequentially from among ambulatory patients referred for liver function tests by hospital clinicians. Consenting patients’ sera were collected from the hospital laboratories following completion of LFTs. LFT results, symptom duration, and demographic data (age, sex, neighborhood) were abstracted from medical records and indexed by specimen number to protect participants’ privacy.

Sample Storage and Shipment

Samples collected during the epidemic were stored in hospital laboratory freezers after completion of liver function testing. The samples were transported on ice from Biratnagar to the

Central Department of Biotechnology of Tribhuvan University, Kathmandu, Nepal, where they were stored frozen at 0°C. Pre-epidemic samples were transported from Biratnagar and Dharan to

Kathmandu on ice and frozen at 0°C after biochemical evaluation. Both sets of samples were shipped on dry ice to the Johns Hopkins Bloomberg School of Public Health in Baltimore, Maryland,

USA, and stored at 0°C until processed.

Serologic Testing

We tested 450 epidemic-period specimens and 180 pre-epidemic specimens using highly sensitive and specific anti-HEV IgG, anti-HEV IgM, and HEV Ag assay kits (Wantai Biological

Pharmacy Enterprise, Beijing, China). The assays were carried out according to manufacturer instructions, except that we substituted two each of 1:16 and 1:32 dilutions of the manufacturer-

137

provided positive control for the undiluted positive control, and we did not repeat the tests. Plates were processed using an automated ELx50 plate washer (BioTek, Winooski, Vermont, USA) and optical densities were read using a VERSAmax ELISA microplate reader with SoftMax software

(Molecular Devices, Sunnyvale, California, USA).

Molecular and Phylogenetic Analyses

A subset of 27 epidemic-period specimens with positive antigen tests and/or high anti-HEV

IgM titers, as well as two IgM-positive pre-epidemic specimens, were selected for molecular analysis.

Viral nucleic acids were purified from 200µL of serum plus internal-control bacteriophage

MS247,48 using the MagNA Pure LC 2.0 robotic instrument with the MagNA Pure LC Total Nucleic

Acid Kit - High Performance (Roche Diagnostics, Indianapolis, Indiana, USA). HEV RNA was then amplified from total nucleic acid eluates using qRT-PCR as well as RT- and nested PCR protocols.

We amplified segments of ORF1 using two sets of primers (as described by Wang et al.49 and Dong et al.50), and we used two sets of primers to amplify segments of ORF2 (as described by Wang et al.49 and John R. Ticehurst [pers. comm.]).

Amplicons were sent to the Johns Hopkins Genetic Resources Core Facility (GRCF), where they were sequenced on an Applied Biosystems 3730 DNA Analyzer (Life Technologies Corp., Grand

Island, NY, USA) using the Sanger method.50,51 Contiguous sequences from ORF1 and ORF2 of the

HEV genome were assembled from sequence fragments returned by the GRCF using CodonCode

Aligner (CodonCode Corp., Centerville, MA, USA).

BLASTN v.2.3.052 was used to identify highly homologous sequences to the Biratnagar epidemic strains in GenBank. We also searched GenBank using keywords to locate and obtain sequence data for other Nepali and South Asian HEV isolates of HEV, as well as more distant HEV

138

isolates, including representatives of the four genotypes known to affect humans. We used MEGA v.6 software53 to align the sequences and to carry out phylogenetic analyses using the maximum- likelihood method.

Data Analysis

We produced descriptive statistics using age, sex, geographic information, and laboratory values for liver function tests extracted from participant records, as well as results of our serologic and molecular analyses. Demographic, laboratory, and serologic results were analyzed using Stata

11 (StataCorp LP, College Station, TX). Geographic locations for pre-epidemic participants were limited to city name (Biratnagar or Dharan). Locations within Biratnagar for epidemic-period participants were given as neighborhood or street/intersection names in patient records. Patients from each neighborhood were randomly assigned to locations within polygons approximating the neighborhood boundaries, both due to the low resolution of residence data and also to preserve participant confidentiality. The resulting geographic coordinates were mapped using ArcMap 10.3

(ESRI, Redlands, CA) for illustrative purposes.

RESULTS

Demographic and serologic results for both the pre-epidemic healthy adults and the outbreak patients are summarized in Table 8. Comparisons of liver enzyme levels by serostatus are summarized in Table 9.

Pre-Epidemic Healthy Adults

Among healthy adults sampled in January 2014, the seroprevalence of anti-HEV IgG was

5.6% (6/107) in Biratnagar and 12.3% (9/73) in Dharan. The six IgG+ adults in Biratnagar were substantially older, on average, than the IgG- population, while in Dharan, the age distributions of

139

IgG+ and IgG- populations were similar. [Figure 17] There were two IgM+ specimens obtained from healthy adults in Dharan (2.7%); both were anti-HEV IgG+ but HEV Ag-.

Biratnagar Outbreak Patients

Among the 450 patients seen at three Biratnagar hospitals during the week of April 27-May

3: 190 (42.2%) were seropositive for anti-HEV IgG, 66 (14.7%) were seropositive for anti-HEV IgM, and 13 (2.9%) were seropositive for HEV Ag. A total of 72 (16%) patients were seropositive for at least one marker of acute infection (IgM and/or Ag). All but four of these 72 specimens were seropositive for anti-HEV IgG; these four were Ag+ and IgM-. Only about half (7, or 54%) of the 13

HEV Ag+ individuals also had detectable anti-HEV IgM. [Figure 18]

Demographics. The age distribution of outbreak patients presenting to the three hospitals was similar to that of prior hepatitis E outbreaks in South Asia, with the largest numbers of patients in the 11-40 year age groups, and relatively fewer young children or older adults. [Figure 19] The seroprevalence of HEV infection markers among the patients tested did not vary markedly by age stratum. IgG antibodies were present in 42.2% of outbreak patients overall (190/450), peaking in the 21-30 year old age group, and lowest among patients above age 60. IgM antibodies were detected in 14.2% of patients (64/450) and HEV Ag was detected in 2.7% of patients (12/450) overall. More male (54.4%) than female (45.6%) patients were seen at the three hospitals, though the sex ratio varied markedly by hospital.[ Table 9]

All serologic markers of HEV infection were slightly more common among female patients

(IgG: 42.6%, IgM: 16.7%, Ag: 2.9%) than among male patients (IgG: 41.9%, IgM: 13.0%, Ag: 2.8%).

Neither these differences in seroprevalence by sex, nor in seroprevalence by sex when stratified by age group, were statistically-significant by Pearson’s χ2 (IgG, IgM) and Fisher’s exact test (Ag).

140

Geography. Of the 450 outbreak patients, 405 provided the neighborhood or street name of their residence in Biratnagar. The other 45 (all at Koshi Zonal Hospital) were simply recorded as medical staff or staff relatives. We were unable to uniquely match residential addresses to city wards because some of the streets and neighborhoods lie along or overlap ward boundaries.

Confirmed acute (anti-HEV IgM+ and/or HEV Ag+) cases were concentrated primarily in the densely-populated areas along and east of the Koshi Highway and north of the Devkota water tank.

[Figure 20] In addition to the higher absolute numbers of cases in this northeast quadrant of the city, the seroprevalence of acute markers among patients from these areas was also elevated relative to other areas of the city. Seropositivity for HEV markers varied considerably by hospital,[Table 10] and the three hospitals’ patient populations came from different residential areas. At Koshi Zonal Hospital (n=153), where there were no IgM+ or Ag+ specimens, only 13.1% of patients reported living in the northeast quadrant of the city, while 57% of patients came from the less-densely-populated western and southern parts of the municipality, and another 29.4% were medical staff or staff relatives whose home address was unknown. In contrast, the northeast quadrant of the city was home to 63.5% of patients at Biratnagar Aspataal and 89.7% of patients at

Nobel Medical College Teaching Hospital, where acute HEV markers were found in 16.2% (12/74) and 26.9% (60/223) of patients, respectively.

Clinical and biochemical measures. Outbreak patients reported median symptom duration of

4 days (range 1 – 9+ days); this did not vary by serostatus. All measured liver enzymes were significantly elevated in outbreak patients as a group relative to the healthy adult reference population, regardless of serostatus.[Table 9] There were no significant differences in liver enzyme levels between seronegative outbreak patients and outbreak patients seropositive only for anti-

HEV IgG, with seronegative patients actually having slightly higher values. Outbreak patients with acute HEV infection markers (anti-HEV IgM and/or HEV Ag) had slightly higher values than

141

outbreak patients who were IgG+ only or seronegative for all markers, but only SGOT was statistically significantly higher.

Molecular and Phylogenetic Characterization of HEV Isolates

No HEV RNA was detected in two IgM+ pre-epidemic samples from Dharan. Of 27 epidemic specimens with HEV antigen and/or high anti-HEV IgM titers, 12 had detectable HEV RNA. Five of these were Ag+ specimens. Interestingly, all 12 of the RNA-positive specimens were also seropositive for anti-HEV IgG.[Table 10] Eight of the 12 specimens had sufficient RNA to attempt to isolate material for sequencing, of which three yielded expected products with at least one ORF1 primer and six yielded expected products with at least one ORF2 primer.

We were able to assemble a 530 nucleotide (nt) ORF1 sequence for three samples and a

191nt ORF2 sequence for six samples, corresponding to nt 80-609 and 6369-6559, respectively, of the Burma (B1) strain of HEV (GenBank Accession #M73218).54 Over the length of the 530nt ORF1 segment, the three Biratnagar isolates were 99% identical to each other, but shared <95% identity with the closest matches in GenBank. The shorter ORF2 sequence, 100% identical among the six

Biratnagar isolates, was 96% identical to the closest match in the database.

Phylogenetic analysis of the 540nt ORF1 sequence [Figure 21] suggested that the Biratnagar

HEV isolates share a common ancestor with HEV strains from India, China, and Pakistan identified elsewhere as genotypes 1b and 1c.55 However, phylogenetic analysis of the 191nt ORF2 sequence placed the Biratnagar HEV isolates in a cluster with HEV strains identified in other publications as subgenotype 1a, including the Burma strain and a strain isolated during an outbreak in Madras,

India (GenBank accession #X99441), while still placing the previously-identified 1b and 1c strains as outgroups to the 1a strains, albeit with low bootstrap support. [Figure 22]

142

DISCUSSION

The low (5.6%) seroprevalence of anti-HEV antibodies in an adult reference population in

Biratnagar in January 2014 stands in marked contrast to the higher seroprevalence typically observed in adults in Kathmandu (24.6%-66%)56-58 and elsewhere in South Asia (33.7% in pregnant women in New Delhi,59 ~25-40% in Pune,60 29.5% and 14.9% in urban and rural Vellore, and 22.5% in rural Bangladesh61); however, serosurveys of other Terai populations have also yielded lower –than-anticipated seroprevalence (12.3% in adults in Dharan, and 11.7% in pregnant women in Sarlahi District [cf cohort study chapter]). Assuming that the reference population is representative of adults in Biratnagar, the low seroprevalence suggests relatively low exposure to

HEV in Biratnagar prior to the outbreak, and corresponding limited immunity and high susceptibility to hepatitis E.

Among 450 ambulatory, symptomatic patients seen at three Biratnagar hospitals during the week of April 27-May 3, we detected anti-HEV IgG in 42.2%, IgM in 14.7%, and HEV Ag in 2.9%, far lower than the 94%+ seroprevalence of the same three HEV markers detected in 48 patients at

Koshi Zonal Hospital by Shrestha et al.10 It is likely that this reflects differences in selection criteria; our patient population was ambulatory, and we did not limit our sample to individuals with frank jaundice.

The lack of precise residence information for each case precluded formal geospatial analysis. Among confirmed acute cases (IgM+ and/or Ag+), 86% came from neighborhoods in the northeast quadrant of the city, with the remainder from a single neighborhood extending west a few blocks from the Koshi Highway. In our sample, acute cases were most concentrated in the vicinity of the Tinpaini water tank, which failed to meet water quality standards in the midst of the outbreak.62 However, as data on individuals’ water sources at home or at their workplaces were unavailable, we cannot conclusively link the outbreak to this water tank. Furthermore, during the 143

outbreak, the water tank at Devkota Chowk was also reported to have dispensed contaminated water.63 Our sample included few patients and no confirmed acute cases from areas south of

Devkota Chowk, including Ward 11, but news media reported cases in the prison, as well as several deaths from hepatitis E, in that area.41,64

The outbreak patients, including those seronegative for all markers of HEV, had elevated liver enzyme levels relative to the healthy reference population. Our failure to detect markers of

HEV infection in many of these patients with elevated liver enzymes may reflect natural differences in the timing or quantity of appearance of anti-HEV antibodies across individuals (i.e., they may have been infected, but without having mounted a detectable immune response and without circulating HEV antigen), or it may reflect the presence of other hepatotrophic pathogens. Some news reports claimed that hepatitis A was co-circulating in Biratnagar,38 but we were unable to test for other hepatitides. We also cannot rule out dengue,65 malaria,66 or leptospirosis67 as possible causes of illness in some patients, but these are unlikely to have been major causes of illness during the hot, dry season.

Our phylogenetic analysis of a 191nt portion of HEV ORF2 was broadly consistent with the tree published by Shrestha et al., which identified the Biratnagar outbreak strains as subgenotype

1a on the basis of a non-overlapping 412nt ORF2 segment.43 However, our phylogenetic analysis of a 540nt segment of HEV ORF1 yielded a different tree topology, one which suggests that the

Biratnagar outbreak strain is more closely related to strains from China, India, and Pakistan

(classified elsewhere as subgenotypes 1b and 1c) than to the Burma strain. Cases of HEV in Nepal have historically been attributed to subgenotypes 1a and 1c, and 1a-1c coinfections have been identified.68 Whole-genome molecular analysis of the Biratnagar outbreak strain could help clarify whether our results reflect a natural recombination event between different HEV strains (as has

144

been inferred in other cases69-72), distinct evolutionary trajectories of different portions of the genome, or simply poor resolution of tree topology at the subgenotype level.73

The hepatitis E outbreak in Biratnagar, which sickened thousands and killed more than a dozen people, resulted from widespread exposure to fecally-contaminated water. The epidemic curve suggests a relatively abrupt population exposure to HEV in Biratnagar, which the head of

Nepal’s Epidemic and Disease Control Division (EDCD) attributed to rupture of a waste water pipe during excavations for a road widening project.74,75

In the wake of the 2014 outbreak, the Nepali government pledged millions of rupees (NR) to fund replacement of the water supply pipes in Biratnagar. Half a year later, however, Biratnagar residents’ concerns about the safety of the drinking water supply remained unallayed.76 The large

March 2015 earthquakes in Nepal, along with political unrest, strikes, and months-long border blockades in the Terai following the drafting of Nepal’s new Constitution further delayed progress toward ensuring access to safe, clean drinking water in Biratnagar.

More recent samples collected from people in Biratnagar and Dharan may help establish how the population seroprevalence of antibodies to HEV has changed in the wake of Biratnagar’s hepatitis E outbreak. Human health surveillance, water-quality monitoring, and infrastructure improvement and maintenance are ongoing needs made more challenging by the same economic and political realities that also make them more urgent. We echo the calls made by EDCD head Dr.

Baburam Marasini77 and others78,79 for improved cross-sector coordination of epidemiologic prevention activities, surveillance, and response in Nepal.

ACKNOWLEDGMENTS

We thank Manish Das (Nobel Medical College Teaching Hospital), contacts at Koshi Zonal

Hospital and Biratnagar Aspataal Pvt. , and Ram Vinod Mahato (Institute of Science and Technology, 145

Tribhuvan University / Nepalese Association for Clinical Chemistry) for providing access to serum specimens. This study was funded through a grant from the Bill and Melinda Gates Foundation.

146

FIGURES AND TABLES

Figure 15. Map of Nepal with inset showing locations of Biratnagar (Morang District) and Dharan (Sunsari District) along the Koshi Highway (H8) in southeastern Nepal. Base imagery from NASA satellite image VE-IMG-4753.

147

Jaundice Patients Visiting Two Biratnagar Hospitals by Week, April-June 2014 80

60 50 40 30

20 Cases Cases SeenHospital at 10 0

Date

Figure 16. Epidemiologic curve showing course of 2014 Biratnagar hepatitis E outbreak. Curve graphed from data published by Shrestha et al., EID, 2015.43

148

Pre-Outbreak Healthy Biratnagar Outbreak Patients Adults (January 2014) (April 27 - May 4, 2014) Biratnagar Dharan Biratnagar Koshi Zonal Nobel MCT All Hospitals Aspataal Hospital Hospital Combined Total 107 73 74 153 223 450 n= Male 62 (57.9%) 41 (56.2%) 41 (55.4%) 106 (69.3%) 98 (43.9%) 245 (54.4%) Female 45 (42.1%) 32 (43.8%) 33 (44.6%) 47 (30.7%) 125 (56.1%) 205 (45.6%) Age: Years 35.8 (15.1) 35.1 (13.4) 30.9 (15.9) 29.9 (15.2) 30.6 (16.2) 30.4 (15.8) mean (sd) 0-10 0 0 4 (5.4%) 15 (9.8%) 14 (6.3%) 31 (6.9%) 11-20 25 (23.4%) 13 (17.8%) 20 (27.0%) 37 (24.2%) 58 (26.0%) 115 (25.6%) 21-30 23 (21.5%) 20 (27.4%) 14 (18.9%) 30 (19.6%) 56 (25.1%) 101 (22.4%) 31-40 15 (14.0%) 14 (19.2%) 20 (27.0%) 38 (24.8%) 51 (22.9%) 109 (24.2%) 41-50 23 (21.5%) 16 (21.9%) 5 (6.8%) 19 (12.4%) 13 (5.8%) 37 (8.2%) 51-60 14 (13.1%) 7 (9.6%) 6 (8.1%) 11 (7.2%) 18 (8.1%) 36 (8.0%) >60 7 (6.5%) 3 (4.1%) 5 (6.8%) 3 (2.0%) 13 (5.8%) 21 (4.7%) IgG+ 6 (5.6%) 9 (12.3%) 43 (58.1%) 60 (39.2%) 87 (39.0%) 190 (42.2%) IgM+ 0 2 (2.7%) 12 (16.2%) 0 52 (23.3%) 66 (14.7%) Ag+ 0 0 0 0 13 (5.4%) 13 (2.9%) Table 8. Demographic and Serologic Characteristics of Pre-Outbreak (Healthy Adults) and Outbreak Population Sample. Abbreviations: Nobel MCT Hopsital – Nobel Medical College Teaching Hopsital, n – number, sd – standard deviation, IgG+ – anti-HEV immunoglobulin G positive, IgM+ – anti-HEV immunoglobulin M positive, Ag+ – HEV antigen positive. Wantai HEV IgG, IgM, and Ag assays were used (Beijing Wantai BP Enterprise Co., Beijing, China).

149

Figure 17. Boxplots showing age distribution of pre-epidemic healthy adults by city and anti-HEV IgG serostatus. Box shows median and interquartile range; extended lines show full range. The mean ages of IgG+ healthy adults in Biratnagar (52.7 years) vs. IgG- adults in Biratnagar (34.8 years) differed statistically significantly by the Wilcoxon rank-sum test (p=0.01). The mean ages of IgG+ and IgG- healthy adults in Dharan (31.3 years and 35.6 years) did not statistically significantly differ (p=0.31).

150

Figure 18. Venn diagram of Biratnagar epidemic patients who were seropositive for at least one serologic marker of past or acute infection with hepatitis E virus (HEV). Abbreviations: anti-HEV immunoglobulin G (IgG), anti-HEV immunoglobulin M (IgM), and HEV antigen (Ag). Ellipses generated with eulerAPE software (http://www.eulerdiagrams.org/eulerAPE/).

151

Presence and Seroprevalence of Hepatitis E Virus Infection Biomarkers Among Outbreak Patients, by Age Group 120

n=115 n=109 Total 100 n=101 Tested IgG+

80 IgM+

Ag+

60 53%

38% # of Specimens of # 39% 40

n=37 n=36 n=31 22% 20 15% 47% 39% 35% n=21 9% 17% 33% 16% 4% 3% 3% 8% 6% 14% 0% 0% 5% 0 0-10 11-20 21-30 31-40 41-50 51-60 61+ Age (Years)

Figure 19. Age distribution of Biratnagar outbreak patients and age-specific seroprevalence of hepatitis E virus (HEV) infection biomarkers. Bar heights indicate absolute numbers of specimens; seroprevalence of each biomarker in each age group is indicated as a percent above the bar. Wantai HEV IgG, IgM, and Antigen Assay Kits (Wantai Biologic Pharmacy Enterprise, Beijing, China) were used for serologic testing. Anti-HEV IgG seroprevalence was highest in the 21-30 year old age group (53%) and lowest in patients above age 60 (33%), but the seroprevalence was not statistically-significantly different across age groups by Pearson’s . There was no clear age-related pattern of anti-HEV IgM seroprevalence, which ranged from a low of 8% among 41-50 year olds to a high of 22% among 21-30 year olds. HEV antigen was detected in ≤6% of patients in each age group. Abbreviations: HEV = hepatitis E virus; IgG = anti-HEV immunoglobulin G; IgM = anti-HEV immunoglobulin M; Ag = HEV antigen.

152

Figure 20. Map of Biratnagar, Morang District, Nepal, showing geographic distribution and hepatitis E serostatus of 405 outbreak patients from three hospitals. Dots represent individual participants and positive serologic test results (anti-HEV IgG, anti-HEV IgM, and/or HEV Ag), geocoded and randomly distributed within neighborhood boundaries to protect participant confidentiality. Individuals seropositive for multiple biomarkers are represented once for each positive test. Excluded were 45 participants who gave no residential address (staff and staff relatives from Koshi Zonal Hospital); seroprevalence of IgG was 35.6% in this group and all were IgM- and Ag-. Numbered divisions are city wards. Red stars represent hospitals in this study (NMCTH= Nobel Medical College Teaching Hospital; BA = Biratnagar Aspataal, Pvt.; KZH = Koshi Zonal Hospital). Blue rectangles represent major water supply tanks. Increasing population density is represented by darker shading [pop. density map layer courtesy of GENESIS Consultancy Ltd.]. 153

Healthy Adults Outbreak Patients Seronegative or Serostatus All IgM+ and/or Ag+ IgG+ only Number 180 378 72 Male 103 (57.2%) 211 (55.8%) 35 (48.6%) Female 77 (42.8%) 167 (44.2%) 37 (51.4%) Age, Years 35.5 (14.4) 30.7 (15.8) 29.2 (15.5) Symptom Duration, Days n/a 4 4 Liver Function Tests Bilirubin (Total) 0.6 (0.3) 1.8 (1.3) 2.0 (1.6) Bilirubin (Direct) 0.3 (0.2) 0.7 (0.7) 0.8 (0.7) Aspartate Transaminase 29.9 (17.0) 49.6 (27.4) 67.0 (52.1) (AST / SGOT) Alanine Transaminase 20.7 (12.2) 53.0 (33.1) 65.2 (46.6) (ALT / SGPT) Alkaline Phosphatase 89.8 (30.6) 261.7 (126.1) 267.4 (99.2) (ALP) Table 9. Clinical and biochemical comparison of healthy adults with outbreak patients by hepatitis E virus (HEV) serostatus. Values for age and liver function tests are given as mean (standard deviation). Symptom duration is given as median because durations of 9 days or more were not recorded separately.

154

Demographic, Serologic, and Molecular Characteristics of Hepatitis E Virus (HEV) RNA-Positive Outbreak Specimens Anti-HEV Anti-HEV HEV ORF1 ORF2 Specimen Age Sex IgG IgM Antigen RNA RNA A 26 F + + – – – B 20 M + + – – – C 53 F + weak + + – + 40 M + + – 19 M + + – 23 F + + weak + D 30 F + + + + + E 16 M + + + – + 13 M + – + F 16 M + + + + + G 39 M + + + + + H 17 M + + – – + Table 10. Demographic, serologic, and molecular characteristics of outbreak specimens positive for HEV RNA by RT-PCR. Key: + = positive result; – = negative result. Specimens assigned letter identifiers were those with sufficient material for nested PCR.

155

Figure 21. Phylogenetic tree of 540-nt partial hepatitis E virus (HEV) open reading frame 1 (ORF1) sequences. Genotype/subgenotypes following Lu et al.55 are indicated at right; see comments in text. Biratnagar outbreak strains cluster within genotype 1, along with (but outside of) subgenotypes 1b and 1c. Nucleotides used in this analysis correspond to nt80-609 of the Burma reference strain (GenBank accession number M73218), including portions of the 5’ untranslated region and the methyltransferase gene. Label format is GenBank accession number | country where specimen was obtained | strain designation (if available) | original host species. Bootstrap values from 1000 replicates shown at nodes. Consensus tree generated using maximum likelihood method in MEGA v.6.06 software.53

156

Figure 22. Phylogenetic tree of 191-nt partial hepatitis E virus (HEV) open reading frame 2 (ORF2) sequences. Nucleotides used in this analysis correspond to nt6369-6559 of the Burma reference strain (GenBank accession number M73218). Genotype/subgenotypes following Lu et al.55 are indicated at right; see comments in text. Label format is GenBank accession number | country where specimen was obtained | strain designation (if available) | original host species. Bootstrap values from 1000 replicates shown at nodes. Consensus tree generated using maximum likelihood method in MEGA v.6.06 software.53

157

REFERENCES

1 Aggarwal RA. (2013) Hepatitis E: clinical presentation in disease-endemic areas and diagnosis. Semin Liver Dis 33(1): 30-40. [doi:10.1055/s-0033-1338112]

2 Aggarwal R. (2011) Clinical presentation of hepatitis E. Virus Res 161(1): 15-22. [doi:10.1016/j.virusres.2011.03.017]

3 Nelson KE, Shih JW, Zhang J, Zhao Q, Xia N, Ticehurst JR, et al. (2014) Hepatitis E vaccine to prevent morbidity and mortality during epidemics. Open Forum Infect Dis 1(3): ofu098. [doi:10.1093/ofid/ofu098]

4 Zhu FC, Zhang J, Zhang XF, Zhou C, Wang ZZ, Huang SJ, et al. (2010) Efficacy and safety of a recombinant hepatitis E vaccine in healthy adults: a large-scale, randomised, double-blind placebo- controlled, phase 3 trial. Lancet 376(9744): 895-902. [doi:10.1016/S0140-6736(10)61030-6]

5 World Health Organization. (2015) Hepatitis E vaccine: WHO position paper, May 2015. Wkly Epidemiol Rec 90(18): 185-200.[http://www.who.int/entity/wer/2015/wer9018.pdf]

6 Pagliusi S. (2012) World's first hepatitis E vaccine approved in China. WHO Global Immunization News 2012(Jan): 14.[www.who.int/immunization/GIN_January_2012.pdf]

7 Kane MA, Bradley DW, Shrestha SM, Maynard JE, Cook EH, Mishra RP, et al. (1984) Epidemic non-A, non-B hepatitis in Nepal. Recovery of a possible etiologic agent and transmission studies in marmosets. JAMA 252(22): 3140-3145.

8 Hillis A, Shrestha S, & Saha N. (1973) An epidemic of infectious hepatitis in the Kathmandu Valley. J Nepal Med Assoc 11(5 & 6): 145-154.

9 Clayson ET, Vaughn DW, Innis BL, Shrestha MP, Pandey R, & Malla DB. (1998) Association of hepatitis E virus with an outbreak of hepatitis at a military training camp in Nepal. J Med Virol 54(3): 178-182. [doi:10.1002/(SICI)1096-9071(199803)54:3<178::AID-JMV6>3.0.CO;2-2]

10 Shrestha SM. (2006) Hepatitis E in Nepal. Kathmandu Univ Med J 4(4): 530-544.

11 Naik SR, Aggarwal R, Salunke PN, & Mehrotra NN. (1992) A large waterborne viral hepatitis E epidemic in Kanpur, India. Bull World Health Organ 70(5): 597-604.

12 Ray R, Aggarwal R, Salunke PN, Mehrotra NN, Talwar GP, & Naik SR. (1991) Hepatitis E virus genome in stools of hepatitis patients during large epidemic in north India. Lancet 338(8770): 783- 784.[http://www.sciencedirect.com/science/article/pii/014067369190667E]

13 Viswanathan R. (1957) Certain epidemiological features of infectious hepatitis during the Delhi epidemic, 1955-1956, in Hepatitis Frontiers (Henry Ford Hospital International Symposium), eds Hartman FW, LoGrippo GA, Mateer JG, & Barron J. 207-210. (London: Churchill)

14 Arankalle VA, Chadha MS, Tsarev SA, Emerson SU, Risbud AR, Banerjee K, et al. (1994) Seroepidemiology of water-borne hepatitis in India and evidence for a third enterically-transmitted hepatitis agent. Proc Natl Acad Sci U S A 91(8): 3428-3432.

15 Dilawari JB, Singh K, Chawla YK, Ramesh GN, Chauhan A, Bhusnurmath SR, et al. (1994) Hepatitis E virus: epidemiological, clinical and serological studies of north Indian epidemic. Indian J Gastroenterol 13(2): 44-48. 158

16 Awsathi S, Rawat V, Rawat CM, Semwal V, & Bartwal SJ. (2014) Epidemiological investigation of the jaundice outbreak in Lalkuan, Nainital District, Uttarakhand. Indian J Community Med 39(2): 94- 97. [doi:10.4103/0970-0218.132725]

17 Teo CG. (2012) Fatal outbreaks of jaundice in pregnancy and the epidemic history of hepatitis E. Epidemiol Infect 140(5): 767-787. [doi:10.1017/s0950268811002925]

18 Sharma SP. (2006) Hepatitis E and cholera outbreak in Kathmandu. CMAJ 175(8): 860. [doi:10.1503/cmaj.061166]

19 Majumdar M, Singh MP, Goyal K, Chawla Y, & Ratho RK. (2015) Detailed investigation of ongoing subclinical hepatitis E virus infections; occurring in outbreak settings of North India. Liver Int 35(3): 826-833. [doi:10.1111/liv.12568]

20 ProMED-Mail. (2006) Hepatitis - Nepal (Birgunj). [http://promedmail.org/post/20060503.1277 Accessed 10 Febuary 2016].

21 Nepal Central Bureau of Statistics. (2014) Statistical year book of Nepal - 2013. (Kathmandu, Nepal: Nepal Central Bureau of Statistics) [http://cbs.gov.np/image/data/Publication/Statistical%20Year%20book%202013_SS/Statistical- Year-book-2013_SS.pdf Accessed 06 March 2016].

22 World Bank. (1977) Nepal - Second Water Supply and Sewerage Project. 107pp. (Washington, DC: World Bank) [http://documents.worldbank.org/curated/en/1977/04/727361/nepal-second-water- supply-sewerage-project Accessed 10 February 2016]

23 World Bank. (1989) Nepal - Third Water Supply and Sewerage Project. 58pp. (Washington, DC: World Bank) [http://documents.worldbank.org/curated/en/1989/09/738369/nepal-third-water- supply-sewerage-project Accessed 10 February 2016]

24 Shrestha V. (2012) Assessment of Nepal Water Supply Corporation - Biratnagar Branch: functionality of water supply services. 31pp. (Jalsrot Vikas Sanstha (JVS) Nepal) [http://www.waterintegritynetwork.net/wp- content/uploads/2015/03/JVS_WIN_Assessment_NepalWaterSupplyCorporation.pdf Accessed 21 December 2015]

25 Nepal Ministry of Urban Development. (2013) Promoting financially sustainable regulatory framework for water tariff in South Asia (Nepal). (Asian Development Bank ) [http://www.adb.org/sites/default/files/project-document/79005/45373-001-reg-tacr-01.pdf Accessed

26 SEAM Nepal Project. (2009) A Report on Water Quality Monitoring Situation Analysis & Need Assessment in SEAM-N Project Area. 37pp. [http://www.enviro- nepal.com/upload/documents/c520c889e4WQM%20situation%20and%20need%20analysis- SEAM-N.pdf Accessed 17 February 2016]

27 Nepal Ministry of Urban Development. (2014) Water service providers data book, 2069-2070 (2012-2013).) [http://www.seiu.gov.np/index.php/documents/download- file?path=Publications/final+Nepal+WSP+Data+Book+2012+13.pdf Accessed 17 Febuary 2016].

28 Nepal Central Bureau of Statistics. (2014) Environment statistics of Nepal 2013. 205pp. (Kathmandu: Nepal Central Bureau of Statistics) 159

[http://cbs.gov.np/image/data/2015/Environment%20Statistics%20of%20Nepal%202013.pdf Accessed 12 February 2015]

29 Asian Development Bank. (2015) Procurement Plan. 6pp. (Tokyo: Asian Development Bank) [http://www.adb.org/sites/default/files/project-document/63406/36188-023-pp.pdf Accessed 12 February 2016]

30 WaterAid Nepal. (2011) WaterAid update on "BIWASH" project in Biratnagar, Nepal. [http://www.youtube.com/watch?v=BWCMli6SzWE Accessed 10 February 2016].

31 Asian Development Bank. (2010) Main report. Report No. TA7182-NEP. 225pp. (Tokyo: Asian Development Bank) [http://www.adb.org/sites/default/files/project-document/67547/36188-01- nep-tacr-01.pdf Accessed 10 February 2016]

32 NGO Forum. (2010) Sanitation awareness through Sugandhapur. [http://www.ngoforum.net/index2.php?option=com_content&task=view&id=9011&pop=1&page=0 &Itemid=1 Accessed 10 February 2016].

33 MaHa Jodi. (2010) Sugandhapur [TV Serial]. [http://mahajodi.com.np/sugandhapur Accessed 10 February 2016].

34 Asian Development Bank. (2008) Biratnagar survey data. 151pp. (Tokyo: Asian Development Bank) [http://www.adb.org/sites/default/files/project-document/75212/36188-01-nep-tacr- 11.pdf Accessed 18 February 2016]

35 Rai SK, Ono K, Yanagida JI, Ishiyama-Imura S, Kurokawa M, & Rai CK. (2012) A large-scale study of bacterial contamination of drinking water and its public health impact in Nepal. Nepal Med Coll J 14(3): 234-240.

36 Nepal Department of Urban Development and Building Construction. (2010) Initial environmental examination, Biratnagar Subproject. 100pp. (Asian Development Bank) [http://www.adb.org/sites/default/files/linked-documents/36188-02-nep-iee-01.pdf Accessed

37 Asian Development Bank. (2013) Sewerage and drainage network, waste water treatment plant, and road and lanes improvement subproject, Biratnagar Municipality,Nepal. 128pp. (Tokyo: Asian Development Bank) [http://www.munbiratnagar.gov.np/userfiles/IEE%20REPORT%20FINAL- Biratnagar.pdf Accessed 15 February 2016]

38 Subedi B. (2014) Viral fevers spreading in Biratnagar. Republica 27 April 2014. [http://www.myrepublica.com/portal/index.php?action=news_details&news_id=73649#sthash.vvH wkAg2.dpuf Accessed 17 February 2016].

39 Jha AK. (2014) EDCD team in Biratnagar. Kathmandu Post 9 May 2014. [http://kathmandupost.ekantipur.com/printedition/news/2014-05-08/edcd-team-in- biratnagar.html Accessed 17 February 2016].

40 Kathmandu Post. (2014) Disease reaches rural areas. Kathmandu Post 13 May 2014. [http://kathmandupost.ekantipur.com/printedition/news/2014-05-12/disease-reaches-rural- areas.html Accessed 17 February 2016].

41 Kathmandu Post. (2014) Jaundice outbreak spreads to district prison. Kathmandu Post 14 May 2014. [http://kathmandupost.ekantipur.com/printedition/news/2014-05-13/jaundice-outbreak- spreads-to-district-prison.html Accessed 17 February 2016]. 160

42 Kathmandu Post. (2014) Jaundice outbreak: Biratnagar schools shut for 5 days. Kathmandu Post 12 May 2014. [http://kathmandupost.ekantipur.com/news/2014-05-12/jaundice-outbreak-biratnagar- schools-shut-for-5-days.html Accessed 17 February 2016].

43 Shrestha A, Lama TK, Karki S, Sigdel DR, Rai U, Rauniyar SK, et al. (2015) Hepatitis E Epidemic, Biratnagar, Nepal, 2014. Emerg Infect Dis 21(4): 711. [doi:10.3201/eid2104.141512]

44 Nepal Department of Health Services. (2015) Annual report 2070/71 (2013/14). 425pp. (Kathmandu: Nepal Department of Health Statistics) [http://dohs.gov.np/wp- content/uploads/2014/04/Annual_Report_2070_71.pdf Accessed 10 March 2016]

45 Kattar P. (2015) After Nepal quakes, worries in the water. [Video] 00:06:34 [http://www.nytimes.com/video/health/100000003883905/nepals-waterborne-worries.html Accessed 22 February 2016].

46 Kathmandu Post. (2014) Jaundice toll hits a dozen. Kathmandu Post 21 May 2014. [http://kathmandupost.ekantipur.com/news/2014-05-21/jaundice-toll-hits-a-dozen.html Accessed 17 February 2016].

47 Dreier J, Stormer M, & Kleesiek K. (2005) Use of bacteriophage MS2 as an internal control in viral reverse transcription-PCR assays. J Clin Microbiol 43(9): 4551-4557. [doi:10.1128/jcm.43.9.4551- 4557.2005]

48 Rolfe KJ, Parmar S, Mururi D, Wreghitt TG, Jalal H, Zhang H, et al. (2007) An internally controlled, one-step, real-time RT-PCR assay for norovirus detection and genogrouping. J Clin Virol 39(4): 318- 321. [doi:10.1016/j.jcv.2007.05.005]

49 Wang Y, Ling R, Erker JC, Zhang H, Li H, Desai S, et al. (1999) A divergent genotype of hepatitis E virus in Chinese patients with acute hepatitis. J Gen Virol 80 ( Pt 1)(169-177. [doi:10.1099/0022- 1317-80-1-169]

50 Sanger F, & Coulson AR. (1975) A rapid method for determining sequences in DNA by primed synthesis with DNA polymerase. J Mol Biol 94(3): 441-448.

51 Sanger F, Nicklen S, & Coulson AR. (1977) DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci U S A 74(12): 5463-5467.

52 Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, et al. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 25(17): 3389- 3402.

53 Tamura K, Stecher G, Peterson D, Filipski A, & Kumar S. (2013) MEGA6: Molecular Evolutionary Genetics Analysis version 6.0. Mol Biol Evol 30(12): 2725-2729. [doi:10.1093/molbev/mst197]

54 Tam AW, Smith MM, Guerra ME, Huang CC, Bradley DW, Fry KE, et al. (1991) Hepatitis E virus (HEV): molecular cloning and sequencing of the full-length viral genome. Virology 185(1): 120-131.

55 Lu L, Li C, & Hagedorn CH. (2006) Phylogenetic analysis of global hepatitis E virus sequences: genetic diversity, subtypes and zoonosis. Rev Med Virol 16(1): 5-36. [doi:10.1002/rmv.482]

56 Clayson ET, Shrestha MP, Vaughn DW, Snitbhan R, Shrestha KB, Longer CF, et al. (1997) Rates of hepatitis E virus infection and disease among adolescents and adults in Kathmandu, Nepal. J Infect Dis 176(3): 763-766. 161

57 Abe K, Li TC, Ding X, Win KM, Shrestha PK, Quang VX, et al. (2006) International collaborative survey on epidemiology of hepatitis E virus in 11 countries. Southeast Asian J Trop Med Public Health 37(1): 90-95.[http://www.ncbi.nlm.nih.gov/pubmed/16771218]

58 Izopet J, Labrique AB, Basnyat B, Dalton HR, Kmush B, Heaney CD, et al. (2015) Hepatitis E virus seroprevalence in three hyperendemic areas: Nepal, Bangladesh and southwest France. J Clin Virol 70(39-42. [doi:10.1016/j.jcv.2015.06.103]

59 Begum N, Devi SG, Husain SA, & Kar P. (2009) Seroprevalence of subclinical HEV infection in pregnant women from north India: a hospital based study. Indian J Med Res 130(6): 709-713.

60 Arankalle VA, Tsarev SA, Chadha MS, Alling DW, Emerson SU, Banerjee K, et al. (1995) Age- specific prevalence of antibodies to hepatitis A and E viruses in Pune, India, 1982 and 1992. J Infect Dis 171(2): 447-450.

61 Labrique AB, Zaman K, Hossain Z, Saha P, Yunus M, Hossain A, et al. (2009) Population seroprevalence of hepatitis E virus antibodies in rural Bangladesh. Am J Trop Med Hyg 81(5): 875- 881. [doi:10.4269/ajtmh.2009.09-0352]

62 Nepal Water Supply Corporation. (2014) Water analysis report, Biratnagar (24 Baisakh 2071). [http://nwsc.gov.np/download/Water_analysis_report_Biratnagar_24Baisakh_2071.pdf Accessed 12 February 2016]

63 Himalayan Times. (2014) Jaundice outbreak looms in Biratnagar. Himalayan Times 9 May 2014 [Accessed 25 March 2016].

64 Subedi B. (2014) Jaundice outbreak claims 10 lives in Morang. Republica 12 May 2014. [http://www.myrepublica.com/portal/index.php?action=news_details&news_id=74549 Accessed 16 February 2016.]

65 Sah OP, Subedi S, Morita K, Inone S, Kurane I, & Pandey BD. (2009) Serological study of dengue virus infection in Terai region, Nepal. Nepal Med Coll J 11(2): 104-106.

66 Dhimal M, Ahrens B, & Kuch U. (2014) Malaria control in Nepal 1963–2012: challenges on the path towards elimination. Malar J 13(241): 10.1186.

67 Myint KS, Murray CK, Scott RM, Shrestha MP, Mammen MP, Jr., Shrestha SK, et al. (2010) Incidence of leptospirosis in a select population in Nepal. Trans R Soc Trop Med Hyg 104(8): 551-555. [doi:10.1016/j.trstmh.2010.04.001]

68 Shrestha SM, Shrestha S, Tsuda F, Nishizawa T, Gotanda Y, Takeda N, et al. (2003) Molecular investigation of hepatitis E virus infection in patients with acute hepatitis in Kathmandu, Nepal. J Med Virol 69(2): 207-214. [doi:10.1002/jmv.10276]

69 Chen X, Zhang Q, He C, Zhang L, Li J, Zhang W, et al. (2012) Recombination and natural selection in hepatitis E virus genotypes. J Med Virol 84(9): 1396-1407. [doi:10.1002/jmv.23237]

70 van Cuyck H, Fan J, Robertson DL, & Roques P. (2005) Evidence of recombination between divergent hepatitis E viruses. J Virol 79(14): 9306-9314. [doi:10.1128/jvi.79.14.9306-9314.2005]

71 Lhomme S, Abravanel F, Dubois M, Sandres-Saune K, Mansuy JM, Rostaing L, et al. (2014) Characterization of the polyproline region of the hepatitis E virus in immunocompromised patients. J Virol 88(20): 12017-12025. [doi:10.1128/jvi.01625-14] 162

72 Wang H, Zhang W, Ni B, Shen H, Song Y, Wang X, et al. (2010) Recombination analysis reveals a double recombination event in hepatitis E virus. Virol J 7(129. [doi:10.1186/1743-422x-7-129]

73 Oliveira-Filho EF, Konig M, & Thiel HJ. (2013) Genetic variability of HEV isolates: inconsistencies of current classification. Vet Microbiol 165(1-2): 148-154. [doi:10.1016/j.vetmic.2013.01.026]

74 Himalayan Times. (2014) Government plans to combat jaundice. Himalayan Times 4 May 2014 Accessed 25 March 2016].

75 Ghimire LB. (2014) Jogmani-Dharan road widening begins. Kathmandu Post 17 March 2014. [http://kathmandupost.ekantipur.com/news/2014-03-17/jogmani-dharan-road-widening- begins.html Accessed 21 April 2016].

76 Jha AK. (2014) Biratnagar residents complain of contaminated water supply Kathmandu Post 14 December 2014. [http://kathmandupost.ekantipur.com/printedition/news/2014-12-13/biratnagar- residents-complain-of-contaminated-water-supply.html Accessed.

77 Kathmandu Post. (2014) Nipping in the bud. Kathmandu Post August 19, 2014. [http://kathmandupost.ekantipur.com/printedition/news/2014-08-18/nipping-in-the-bud.html Accessed: 17 February 2016]

78 Dixit SM. (2015) Op-Ed: Disease surveillance. Kathmandu Post November 6, 2015. [http://kathmandupost.ekantipur.com/news/2015-11-06/disease-surveillance.html Accessed: 17 February 2016].

79 Akbar SMF. (2016) Comparing HEV outbreaks in Japan, Bangladesh, and Nepal in US – Japan Cooperative Medical Sciences Program 18th International Conference on Emerging Infectious Diseases (EID), Hepatitis Panel.)

163

DISCUSSION

SUMMARY OF ORIGINAL RESEARCH

The research presented in this dissertation has focused broadly on the problem of hepatitis

E in the Terai region of Nepal.

Hepatitis E Virus in Pregnant Women in Sarlahi District, Nepal

This manuscript detailed a community-based, observational, prospective cohort study of pregnant and postpartum women in Sarlahi District, Nepal, conducted from October 2012 to May

2014. The study was situated within an ongoing randomized, clinical trial, taking advantage of established research infrastructure to carry out the work in communities relatively enthusiastic about and accustomed to participation in health research studies.

The study enrolled a total of 2363 pregnant women and followed them through three months postpartum or the end of the study period. These pregnant women collectively contributed

8716 symptom surveys at roughly monthly intervals over the course of the study. Of the 2363 participants, 1669 also provided baseline serologic samples at enrollment, 811 provided samples shortly after delivering, and 659 provided samples three months after the end of their pregnancies.

These samples were tested for markers of infection with hepatitis viruses and used to identify seroconverters.

At baseline, 11.7% of the participants were seropositive for anti-HEV IgG. Greater age, lower caste and land ownership, and meat consumption were associated with seropositivity, while primary drinking water source and latrine access were not. We identified 20 seroconverters over the study period, 9 of whom seroconverted between enrollment and delivery, yielding an incidence of roughly 32.3/1000 person-years during pregnancy and 44.3/1000 person-years from enrollment

164

through 3 months postpartum. We also observed the phenomenon of seroreversion, described in other populations, in several participants, including all 5 of the women who seroconverted during pregnancy and provided a 3 month postpartum blood sample.

During follow-up, we identified 15 cases of hepatitis-like illness based on a symptom score, one of which (discussed in the following chapter) was confirmed to be hepatitis E. The remainder could not be definitively diagnosed as viral hepatitis. There were 2 maternal deaths, one of which was determined by retrospective verbal autopsy to involve jaundice, but we could not confirm hepatitis E and some features of the course of the woman’s illness suggest another cause is more likely.

We did not detect disproportionately poor birth outcomes among seroconverters or in the symptomatic case. This finding, however, is qualified by the fact that many women with early or mid-pregnancy losses were not formally enrolled into the parent influenza vaccine trial. Thus, women with adverse outcomes that could potentially have been associated with HEV infection were systematically excluded. This, and the relatively low number of infections we observed, prevents us from drawing robust and valid conclusions about broader associations of HEV infection with adverse pregnancy outcomes.

Case Report: Hepatitis E Virus (HEV) in an Acutely-Ill Pregnant Woman in Nepal

Although in most ways an unremarkable case of uncomplicated acute hepatitis E, in context, this case was nonetheless notable for several reasons:

. It was the only case of acute hepatitis E detected in our cohort of ~2000 Nepali women

followed during pregnancy and postpartum.

165

. Blood samples provided by the woman two weeks apart were both seropositive for anti-

HEV IgG and IgM and viremic, allowing for real-time hepatitis E diagnosis in the field by

auxiliary nurse-midwives using a rapid IgM ELISA, and for later extraction of viral RNA.

. A collaborating laboratory provided a phylogenetic tree indicating that the virus isolated

from this Nepali woman was infected with HEV of a subgenotype (1d) previous known only

from North Africa. While this tree was independently replicable and the isolate does

represent genotype 1 HEV, we found that reliance on a short sequence from a conserved

segment of the genome likely yielded misleading subgenotype and phylogenetic

information.

. The course of this woman’s illness and its ultimately positive outcome provide an anecdotal

counterpoint to the hospital-based, case-based literature, which is dominated by the most

severe cases – especially in areas of South Asia where access to care is beyond the reach of

much of the population. In communities across South Asia and the developing world, most

hepatitis E goes undiagnosed, but the burden of these milder hepatitis E illnesses is not

without impact on people’s lives. Although many people who develop hepatitis E are not as

fortunate as our study participant and her young son, this case serves as a reminder that the

broader hepatitis E story is still evolving in often-unexpected ways.

The 2014 Biratnagar, Nepal Hepatitis E Outbreak

In April 2014, a hepatitis E epidemic affecting thousands of people struck the city of

Biratnagar, along the Indian border in Morang District, Nepal. We collaborated with a Nepali virologist to obtain serum samples and medical record information for 450 patients presenting at three Biratnagar hospitals (one teaching hospital, one private hospital, and one government hospital) and referred for liver function tests during the outbreak. We also obtained 180 samples from healthy adults collected in January 2014 from Biratnagar and the nearby city of Dharan as part

166

of a separate study to establish reference values for clinical biomarkers. We tested the samples for anti-HEV IgM and IgG and for HEV Ag. Using RT-PCR, we attempted to extract viral RNA from samples positive for HEV Ag or high-titer IgM. We used primers covering two different segments of the hepatitis E virus genome to extract nucleic acids and performed phylogenetic analysis on sequences obtained from these samples.

We found a seroprevalence in healthy adults of 5.6% in Biratnagar and 12. 3% in Dharan prior to the outbreak. Among outbreak patients, 2.2% (190/450) were IgG+, 14.7% (66/450) were

IgM+, and 2.9% (13/450) were HEV Ag+. Seropositivity varied by hospital and by patient neighborhood of origin, but did not vary tremendously by age or by sex. However, the absolute number of patients mirrored frequently-reported patterns during hepatitis E outbreaks, with a greater number of cases being seen in young-to-middle-aged adults and relatively fewer in very young or very old populations. Liver function tests were generally elevated in outbreak patients relative to the healthy adult sample, but did not vary substantially with serostatus.

From among the viremic patients, we obtained 3 sequences using HEV ORF1 primers and 6 sequences using ORF2 primers. We determined the outbreak strains to share high % identity and to belong to genotype 1 as expected, though their relationship to other genotype 1 strains was obscured by variability in the trees produced using the two different genomic segments.

We observed significant differences in both serologic results and patient catchment area by hospital. Lacking detailed information on patients’ water sources, or, failing that, water tank service areas, we could not definitively link the outbreak to a single water source. Furthermore, the geographic distribution of patients for whom specimens were available from the three hospitals did not cover all areas of the city known to be affected by the outbreak.

167

This work adds dimension to accounts of the outbreak and reveals differences in populations seen at each of three Biratnagar hospitals during the epidemic. It suggests that one contributor to the large size of the outbreak may have been low seroprevalence of anti-HEV antibodies and corresponding high susceptibility of the population in Biratnagar.

PUBLIC HEALTH IMPLICATIONS

The literature on hepatitis E virus reviewed in the early chapters of this work describes a pathogen responsible for a large and yet underappreciated share of illnesses and deaths in young adults in developing countries. It highlights the sometimes-dire consequences of hepatitis E observed in pregnant women both during and in the absence of broader outbreaks. It posits various pathways by which the virus, in concert with changes in hormones and immune function and the placental environment, may provoke severe or even fatal outcomes for both mother and infant, and laments the dearth of available primary and secondary prevention strategies in the settings where they are most desperately needed.

In pregnant women in Sarlahi – as in healthy adults from Dharan, Sunsari, and Biratnagar,

Morang – we found a somewhat lower seroprevalence of antibodies to HEV than has typically been reported in the Kathmandu Valley and other rural and urban areas in South Asia. This relatively lower seroprevalence, however, does not imply an absence of infection or even of serious illness.

We estimated, based on serology, an incidence of 32.3 infections / 1000 p*y in pregnant women, and 43.5 / 1000 p*y in pregnant and postpartum women in Sarlahi. As a further caveat, our surveillance period covered only a few seasons, we may have systematically missed women who developed HEV infections earlier in pregnancy, and the number of deaths was sufficiently small in our sample to preclude making firm statements about the proportion of maternal deaths attributable to hepatitis E. Although one pregnant woman who became the single, verified hepatitis

E case did recover and give birth to a healthy, full-term baby boy, even this relatively benign course 168

of illness took a toll on the mother’s health, leading to symptoms lasting at least a month and a decrease in body weight during the late second trimester.

Striking disparities in seroprevalence of HEV in Sarlahi by caste and land ownership suggest that the burden of infection and illness falls disproportionately on certain communities, and in those communities, the impact of hepatitis E may be felt more acutely and the need for prevention and intervention may be far greater.

The outbreak that affected thousands of people in Biratnagar is another clear signal that

HEV poses a serious threat in the Terai. There, a large population with low prior immunity to HEV was hit hard when the virus was introduced into a municipal water system serving densely- populated neighborhoods. The Biratnagar outbreak was attributed, perhaps correctly, to the accidental rupture of a pipe during road construction, which allowed sewage to contaminate the water supply.1

But to view the Biratnagar outbreak as the result of a singular, unfortunate accident is to disregard the context that contributed to its occurrence. Nepal is undertaking a variety of infrastructure development projects, often with the financial backing of other governments or international development agencies. The pace of road widening in Kathmandu has been furious over the past decade, the façades of historic brick buildings sheared clean off to make room for asphalt and utility poles.

Although perhaps less dramatically, life in the Terai has also undergone substantial changes.

What was formerly jungle behind the NNIPS headquarters and clinic in Hariyon just a decade ago is now a plain of plowed fields, and the traditional open bamboo-and-mud-walled homes that dot the dirt roads are increasingly punctuated by imposing two- or three-story concrete structures built with money sent back by family members working overseas. Hariyon, whose population now

169

exceeds 20,000, and Lalbandi, at ~15,000,2 were both recently upgraded to “municipality” status as part of an urbanization initiative.3 Neighborhoods in Biratnagar, like Jamungachhi (near the airport) have expanded rapidly, and this growth has challenged the ability of the municipal infrastructure to keep up. The major earthquake in April 2015 4 and the chronic political unrest in the Terai (especially in border cities like Biratnagar)5 surrounding the drafting of Nepal’s first

Constitution disrupted, but did not halt, these trends.

Much of the larger-scale infrastructure development is aimed, nobly, at increasing the access of poorly-connected communities to much-needed services and opportunities. Nonetheless, the circumstances surrounding the hepatitis E outbreak in Biratnagar underscore that the bodies tasked with overseeing the maintenance of even existing infrastructure – the leaking water pipes running through sewage channels adjacent to the newly-paved roads – and with monitoring and responding to infectious disease outbreaks, are already stretched perilously thin.6-8

Increasingly, young Nepalis from the countryside are seeking work in these expanding population centers or in cities abroad. Remittances are a large and growing source of income.9

Among the women in the Sarlahi cohort, nearly 40% (857/2250) reported that at least one member of the household was employed outside of the local area. Of these women reporting one or more migrant laborers, 23% had family working in Kathmandu, 23% had family working elsewhere in

Nepal, 15% had family working in India, 38% reported family working in the Gulf States (often building other countries’ burgeoning infrastructure), 10,11 and 13% had family in Southeast Asia.

This large-scale movement of the population presents a two-way risk for hepatitis E, for those who leave and for those who remain at home. Migrant laborers (mostly men, more often from lower socioeconomic classes) coming from areas with low HEV seroprevalence are likely to lack prior immunity to HEV, and they are therefore at risk of contracting hepatitis E when they move to

170

Kathmandu or Bhojpur or Doha. Many them, especially those in the Gulf States, work in risky professions and, unable to afford more comfortable accommodations, live in extremely crowded conditions with poor sanitation,10,11 further increasing their risk of exposure to enteric pathogens.

Meanwhile, the effects of family members’ migrant labor on the risk of HEV infection among women in the Terai are likely to be mixed. There is evidence that women whose husbands work abroad are less likely than other women to engage in paid employment outside the home, 12 which could potentially reduce their risk of exposure to HEV. Conversely, the work of tending to the family’s own crops outdoors may fall more to women as the result of a family member’s absence, which could increase risk of exposure. Family members who return home to Sarlahi from larger cities may bring with them HEV infections. Intra-household transmission appears to be uncommon with HEV, 13 in contrast to HAV, though it has been reported.14,15 Still, repeated reintroduction of

HEV into the environment by migrants returning from larger cities may help boost its circulation in the Terai, increasing the risk to women who lack prior immunity.

If development outpaces the ability to address the challenges that urbanization often presents for human health – particularly with regard to drinking water treatment and distribution, human waste disposal, and crowding – then more Biratnagar scenarios may be inevitable, particularly in Terai communities that have not historically had the massive outbreaks seen in the capital, and where most of the population may be susceptible. Addressing these challenges to forestall future epidemics of waterborne diseases like hepatitis E will require coordinated efforts on the part of the government, international aid and development organizations, health care providers at all levels, and local communities.

171

DIRECTIONS FOR FUTURE RESEARCH

The work presented herein provides preliminary answers to some of the questions we sought to address, but inevitably provokes others.

Detection of HEV in the Environment and in Animals in the Terai

In Sarlahi, we observed an association of meat-eating with HEV positivity, and a lack of association of drinking water source or latrine access with seropositivity, despite HEV in Nepal being considered a primarily waterborne infection. Our survey data were limited in scope for practical reasons, and as a result, we may have missed some relevant nuances that could help clarify the relationships of each of these variables with socioeceonomic factors and with HEV.

Mapping latrines and well locations relative to cases could improve our understanding of the distribution of infections in this population. In the current study, we had the ability to look at individual-level associations and limited larger-scale associations; however, aggregating data to the administrative ward or VDC level, as we did to approximate community-level exposure, may be a poor representation of natural community configurations. Analysis that does not define communities based on administrative boundaries, but rather uses the spatial distribution of individual-level data to define communities, could provide useful insights.

Environmental testing of groundwater from local wells, as well as sampling of surface water and of fish or snails obtained from local ponds and streams, would complement spatial analyses by providing evidence of the presence, degree, and variability of fecal contamination (and perhaps of

HEV). Determining whether wells of various designs or depths are more likely to be contaminated might help in shaping interventions.

In addition to testing fish and snails, it would be illuminating to test a variety of other local animals – both livestock and wild animals – for evidence of HEV. The presence of HEV in animals in 172

Nepal has previously been investigated among swine in the Kathmandu Valley, where antibodies were present 16 but phylogenetic information was not published, and in sewage runoff from small swine farms where no HEV RNA was detected. [Kerry Morrison et al., unpublished data] A study that purported to show that HEV was prevalent in rats in Kathmandu was later retracted after the samples were found to have been contaminated. 17

Thus, there is a dearth of information on the presence, prevalence, and genotype(s) of HEV strains in Nepali livestock and wild animal species. It is unknown if these animals may represent a potential reservoir for human infection in Nepal. The conflicting results from studies in India make the situation all the more intriguing.

If zoonotic transmission did occur, would these infections be similar to genotype 1 infections in terms of disease outcomes in this population? This could help address several questions, including whether the milder and frequently asymptomatic infections with HEV genotypes 3 and 4 that occur in Europe, North America, and Japan, are primarily a function of general population health status or of reduced virulence in humans of these animal-adapted strains.

Whether such infections would provoke similar antibody responses is also not entirely clear, but is also of interest in explaining the global epidemiology of HEV and the occurrence of outbreaks.

Replication of HEV Findings from Bangladesh on Cytokines, Micronutrients, and Arsenic

The seroprevalence of anti-HEV IgG among pregnant women in Sarlahi is similar to that among pregnant women in rural areas of Bangladesh.18 Although the incidence of infection in

Sarlahi was estimated to be somewhat lower overall,19 the differences may be explained by statistical uncertainty in the estimates, year-to-year differences, or differences in methodology.

Assuming that the population risk of HEV is similar, are there also similar associations of

HEV in Sarlahi with micronutrient levels,20 taking into account potential seasonal confounding?20,21

173

Would we observe similar patterns of cytokine activity prior to infection in women who eventually became seroconverters vs. those who remained uninfected?18

Do toxins like arsenic or other environmental contaminants increase susceptibility to infection, as colleagues have recently posited?22 Bangladesh has historically dealt with major arsenic contamination of the water supply, but in Nepal the issue has received less attention.

Studies from the Terai have shown that arsenic levels in groundwater may exceed safe levels, 23 with perhaps a quarter of all tubewells in the Terai affected.24 In adults aged 15-49 from several

Terai districts, the prevalence of clinically-apparent arsenicosis was generally low (<2%), and lower in women than in men. 25 However, there are focal areas of higher arsenic levels where a greater proportion of the population is exposed,25,26 and health effects may occur at levels lower than those required to produce overt manifestations of toxicity. In the Terai, as in Bangladesh, might people with greater arsenic exposure also be more susceptible to HEV infection? If so, could widespread promotion of locally-suitable arsenic filters for water treatment 27 have the welcome side effect of reducing the incidence of hepatitis E by a small amount, in addition to its other benefits?

On a related note, aflatoxin, produced by fungal contamination of grains with Aspergillus niger, has received little attention in relation to hepatitis E. Aflatoxin is ubiquitous in both Nepal and Bangladesh, and parallel studies conducted in Sarlahi and in Gaibandha show that lifetime exposure to aflatoxin and its metabolites begins before birth.28 Though perhaps best known for its association with liver cancer, aflatoxin has been linked to a wide range of adverse health effects, including immune dysregulation and stunted growth. 28 Chronic hepatitis B viral infection and aflatoxin exposure together have a syngergistic effect on the risk of hepatocellular carcinoma.29

While the pathogenesis and natural history of HEV differ significantly from HBV, it is not unreasonable to wonder whether aflatoxin’s deleterious effects on the liver, on immune function, 174

and on overall health might also increase the risk of infection, serious illness, or death in people exposed to HEV. Given the high levels of human exposure to aflatoxin via food across tropical and subtropical South Asia, where HEV is a perennial menace, it may be worthwhile to investigate whether there is any association of aflatoxin with HEV susceptibility or outcomes.

Characterization of Antibody Responses to HEV During Pregnancy and Beyond

In Sarlahi, we observed a total of 20 seroconversions: 9 during pregnancy (incidence

~33/1000 p*y) , 8 during the first 3 months postpartum (incidence ~99/1000 p*y), and 3 from enrollment in pregnancy to 3 months postpartum (where no intermediate sample was available).

Among the women who seroconverted during pregnancy and provided later samples (n=5), all had seroreverted by 3 months postpartum. In Bangladesh, 39/40 women who were identified as seroconverters based on seropositivity at 3 months postpartum had apparently seroconverted very late in pregnancy or postpartum.18

Even allowing for seasonal coverage and the limited sample size in the Sarlahi study, it is curious that the postpartum incidence of seroconversion is so strikingly high relative to the incidence during pregnancy in both of these studies.

Do these observations reflect a pregnancy-independent pattern of very transient antibody production in response to mild HEV infection, which, on purely probabilistic grounds, would have had more opportunity to disappear over the longer periods of observation during pregnancy than during the 3 months of observation postpartum? Seroreversion has been reported in diverse settings, though with wildly inconsistent rates of occurrence.30

If production of antibodies to HEV is simply very ephemeral, this suggests that we may be

(perhaps vastly) underestimating the specific and cumulative incidence of infections in the population, particularly with longer intervals between samples. Such “missed” asymptomatic

175

infections may be of more than academic interest. Misclassification of infection status may obscure small-to-moderate effects of potentially-modifiable exposure-related variables, reducing opportunities for intervention to prevent infections. In addition, we have not yet adequately assessed the potential effects of mild infections on fetal development and viability, and ephemerality of maternal antibody response does not necessarily imply lack of effect on the fetus.

Shorter observation windows would provide more sensitive detection of infection events; however, balancing this goal with practical and ethical considerations in a community-based trial may prove difficult. Testing a subset of existing, banked late-pregnancy samples from Bangladesh (regardless of postpartum serostatus) could help to clarify whether the postpartum-skewed incidence of seroconversion observed in that cohort is an artifact of the retrospective seroconversion identification strategy or whether it is a real phenomenon. If corroborated in Bangladesh, and if replicated in longitudinal studies of pregnant women conducted in other settings, this would be a fascinating and heretofore unidentified feature of the epidemiology of HEV. Until and unless that occurs, it will remain a curiosity.

176

REFERENCES

1 Himalayan Times. (2014) Government plans to combat jaundice. Himalayan Times 4 May 2014.

2 Nepal Central Bureau of Statistics. (2012) National Population and Housing Census 2011 (Village Development Committee/Municipality). Vol. 2 (Kathmandu, Nepal) [http://cbs.gov.np/nada/index.php/catalog/54/download/467 Accessed 06 June 2016].

3 Kathmandu Post. (2014) Govt announces 72 new municipalities. Kathmandu Post. 09 May 2014. [http://kathmandupost.ekantipur.com/printedition/news/2014-05-08/govt-announces-72-new- municipalities.html Accessed 19 May 2016].

4 United States Geological Survey. (2015) M7.8 - 36km E of Khudi, Nepal. [http://earthquake.usgs.gov/earthquakes/eventpage/us20002926#general Accessed 8 June 2016].

5 Himalayan New Service. (2015) Tarai protest rages on, supplies hit. Himalayan Times. September 29, 2015. [https://thehimalayantimes.com/nepal/tarai-protest-rages-on-supplies-hit/ Accessed 1 July 2016].

6 Shrestha V. (2012) Assessment of Nepal Water Supply Corporation - Biratnagar Branch: functionality of water supply services. 31pp. (Jalsrot Vikas Sanstha (JVS) Nepal) [http://www.waterintegritynetwork.net/wp- content/uploads/2015/03/JVS_WIN_Assessment_NepalWaterSupplyCorporation.pdf Accessed 21 December 2015.]

7 Nepal Department of Urban Development and Building Construction. (2010) Initial environmental examination, Biratnagar Subproject. 100pp. (Asian Development Bank) [http://www.adb.org/sites/default/files/linked-documents/36188-02-nep-iee-01.pdf Accessed 11 February 2016.]

8 Jha AK. (2014) Biratnagar residents complain of contaminated water supply. Kathmandu Post 14 December 2014. [http://kathmandupost.ekantipur.com/printedition/news/2014-12-13/biratnagar- residents-complain-of-contaminated-water-supply.html Accessed 21 April 2016.]

9 World Bank Databank. (2016) Personal remittances, received (% of GDP) [http://data.worldbank.org/indicator/BX.TRF.PWKR.DT.GD.ZS/countries/1W-NP?display=default Accessed 06 June 2016].

10 Joshi S, Simkhada P, & Prescott GJ. (2011) Health problems of Nepalese migrants working in three Gulf countries. BMC Int Health Hum Rights 11(1): 1-10. [doi:10.1186/1472-698x-11-3]

11 Doward J. (2014) Qatar World Cup: 400 Nepalese die on nation's building sites since bid won. The Guardian 16 February 2014. [www.theguardian.com/football/2014/feb/16/qatar-world-cup-400- deaths-nepalese Accessed 1 July 2016].

12 Lokshin M, & Glinskaya E. (2009) The Effect of Male Migration on Employment Patterns of Women in Nepal. The World Bank Economic Review 23(3): 481-507. [doi:10.1093/wber/lhp011]

13 Arankalle VA, Chadha MS, Mehendale SM, & Tungatkar SP. (2000) Epidemic hepatitis E: serological evidence for lack of intrafamilial spread. Indian J Gastroenterol 19(1): 24-28.

177

14 Teshale EH, Grytdal SP, Howard C, Barry V, Kamili S, Drobeniuc J, et al. (2010) Evidence of person-to-person transmission of hepatitis E virus during a large outbreak in Northern Uganda. Clin Infect Dis 50(7): 1006-1010. [doi:10.1086/651077]

15 Khuroo MS, & Dar MY. (1992) Hepatitis E: evidence for person-to-person transmission and inability of low dose immune serum globulin from an Indian source to prevent it. Indian J Gastroenterol 11(3): 113-116.

16 Clayson ET, Innis BL, Myint KS, Narupiti S, Vaughn DW, Giri S, et al. (1995) Detection of hepatitis E virus infections among domestic swine in the Kathmandu Valley of Nepal. Am J Trop Med Hyg 53(3): 228-232.

17 He J, Innis BL, Shrestha MP, Clayson ET, Scott RM, Linthicum KJ, et al. (2006) RETRACTED: Evidence that rodents are a reservoir of hepatitis E virus for humans in Nepal. J Clin Microbiol 44(3): 1208. [doi:10.1128/jcm.44.3.1208.2006]

18 Kmush BL, Labrique A, Li W, Klein SL, Schulze K, Shaikh S, et al. (2016) The Association of Cytokines and Micronutrients with Hepatitis E Virus Infection During Pregnancy and the Postpartum Period in Rural Bangladesh. Am J Trop Med Hyg 94(1): 203-211. [doi:10.4269/ajtmh.15-0238]

19 Labrique AB, Klein S, Ali H, Engle RE, Schulze K, Purcell RH, et al. (2012) Seroincidence of Intrapartum Hepatitis E Virus Infections and Micronutrient Status in Pregnant Bangladeshi Women Between 2001 and 2010. Presented at “Hepatitis E in the United States”, National Institutes of Health, Bethesda, Maryland.

20 Campbell RK, Talegawkar SA, Christian P, LeClerq SC, Khatry SK, Wu LS, et al. (2014) Seasonal dietary intakes and socioeconomic status among women in the Terai of Nepal. J Health Popul Nutr 32(2): 198-216.[http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4216957/pdf/jhpn0032-0198.pdf]

21 Jiang T, Christian P, Khatry SK, Wu L, & West KP, Jr. (2005) Micronutrient deficiencies in early pregnancy are common, concurrent, and vary by season among rural Nepali pregnant women. J Nutr 135(5): 1106-1112.

22 Heaney CD, Kmush B, Navas-Acien A, Francesconi K, Gossler W, Schulze K, et al. (2015) Arsenic exposure and hepatitis E virus infection during pregnancy. Environ Res 142(2015):273-280. [doi:10.1016/j.envres.2015.07.004]

23 Yadav IC, Singh S, Devi NL, Mohan D, Pahari M, Tater PS, et al. (2012) Spatial distribution of arsenic in groundwater of southern Nepal. Rev Environ Contam Toxicol 218(125-140. [doi:10.1007/978-1-4614-3137-4_3]

24 Shrestha RR, Shrestha MP, Upadhyay NP, Pradhan R, Khadka R, Maskey A, et al. (2003) Groundwater arsenic contamination, its health impact and mitigation program in Nepal. J Environ Sci Health A Tox Hazard Subst Environ Eng 38(1): 185-200.

25 Maharjan M, Shrestha RR, Ahmad SA, Watanabe C, & Ohtsuka R. (2006) Prevalence of arsenicosis in Terai, Nepal. J Health Popul Nutr 24(2): 246-252.

26 Maden N, Singh A, Smith L, Maharjan M, & Shrestha S. (2011) Factors Associated with Arsenicosis and Arsenic Exposure Status in Nepal: Implications from Community Based Study. J Community Health 36(1): 76-82. [doi:10.1007/s10900-010-9282-1]

178

27 Pokhrel D, Bhandari BS, & Viraraghavan T. (2009) Arsenic contamination of groundwater in the Terai region of Nepal: an overview of health concerns and treatment options. Environ Int 35(1): 157- 161. [doi:10.1016/j.envint.2008.06.003]

28 Groopman JD, Egner PA, Schulze KJ, Wu LS, Merrill R, Mehra S, et al. (2014) Aflatoxin exposure during the first 1000 days of life in rural South Asia assessed by aflatoxin B(1)-lysine albumin biomarkers. Food Chem Toxicol 74(2014):184-189. [doi:10.1016/j.fct.2014.09.016]

29 Ross RK, Yu M, Henderson B, Yuan J-M, Qian G-S, Tu J-T, et al. (1992) Urinary aflatoxin biomarkers and risk of hepatocellular carcinoma. The Lancet 339(8799): 943-946.

30 Krain LJ, Nelson KE, & Labrique AB. (2014) Host Immune Status and Response to Hepatitis E Virus Infection. Clin Microbiol Rev 27(1): 139-165. [doi:doi:10.1128/CMR.00062-13]

179

APPENDIX 1: FLOWCHART OF DATA AND SAMPLE COLLECTION

180

181

APPENDIX 2: SARLAHI COHORT STUDY FORMS AND DATA COLLECTION

LIST OF MAGIFT FORMS AND SCRIPTS USED IN THE HEPATITIS SUBSTUDY

Form Form Name & Purpose / Data Collected Version | Date Version | Date # (English) (Nepali)

Household Consent Script 5 | 12-Apr-11 2 | 12-Apr-11 1 Describes overall household participation in MaGIFT and seeks consent.

Reproductive Age Women’s Consent Script 4 | 15-May-13 4 | 15-May-13

2 Describes pregnancy surveillance, immunization, and follow-up of women 15-40 and seeks consent.

Household Women’s Census Form 6 | 14-Jan-11 —

5A Records household size, members, ID numbers, ages, and women’s current pregnancy status.

New Household Women’s Census Form 5 | 14-Jan-11 — 5B Records information as above for new households established after baseline census .

Household Characteristics Form (HCF) 5 | 10-Dec-10 1 | 28-Mar-11 6 Records household demographics, home construction, socioeconomic indicators.

Pregnancy Surveillance Form (PSF) 3 | 02-May-11 2 | 02-May-11

7 Records dates of pregnancy surveillance and menstrual periods and results of pregnancy tests for women in a single sector of a ward of a village development committee.

Pregnancy Enrollment Form (PEF) 6 | 16-Dec-11 2 | 16-Dec-11

8 Records health and reproductive history, tobacco and alcohol use, anthropometry, and provision of antenatal care package.

182

Maternal Weekly Morbidity Form 7 | 27-Aug-12 2 | 27-Aug-12 9 Records symptoms of flu-like illness and care visits among pregnant women.

Infant Weekly Morbidity Form 8 | 30-Jul-12 3 | 30-Jul-12

10 Records symptoms of flu-like illness or pertussis and care visits among infants <6 months of age.

Vaccine Receipt Eligibility Form – Phase II 1 | 01-May-12 1 | 01-May-12

13A Records pregnant woman’s eligibility to receive vaccine and history of tetanus toxoid injections.

Monthly Pregnancy Follow-Up Form (PFF) 8 | 15-Oct-12 3 | 15-Oct-12

14 Records woman’s vital status, pregnancy status, 30-day history of illness symptoms, alcohol/tobacco use, anthropometry, and hepatitis score (calculated from symptom report).

Maternal Birth Assessment Form (MBAF) 6 | 10-Jan-13 3 | 10-Jan-13

15 Records date and nature of pregnancy outcome, details of delivery, pre- and post-delivery care, maternal complications, and woman’s vital signs.

Infant Birth Assessment Form (IBAF) 6 | 23-Nov-11 2 | 23-Nov-11

16 Records date and time of birth, vital status and sex of infant, infant health, cord care and breastfeeding practices, and infant anthropometry.

Maternal Verbal Autopsy Form (MVAF) 1 | 25-Mar-13 1 | 15-Mar-13

17 Documents death of enrolled woman and records information solicited from relatives or others pertaining to cause of death.

18 Child Verbal Autopsy Form (CVAF) 1 | 31-Jan-11 1 | 31-Jan-11

183

Documents death of enrolled infant or child and records information solicited from relatives or others pertaining to cause of death.

6 Month Child Anthropometry Follow-Up Form 1 | 15-Nov-11 — 19 Records vital status, height, weight, and head circumference of infants at 6 months of age.

Pregnancy Report Form 1 | 03-Mar-11 1 | 28-Mar-11 23 Documents pregnancies reported via pregnancy surveillance procedures or other means.

Pregnancy Tracking Log 10 | 19-Oct-12 —

25 Used to schedule and track individual study milestones, follow-up visits, and pregnancy outcome.

Post-Partum Morbidity Form (MPPF) 4 | 31-Jan-11 1 | 28-Mar-11

26 Records symptoms of illness or complications and chlorhexidine use shortly after delivery.

Birth Report Form 4 | 14-Sep-12 2 | 14-Sep-12 27 Records date of birth and parents’ names upon receipt of notification.

Married New Woman Form 4 | 02-May-11 —

29 Records arrival / household change and consent of married woman aged 15-40 in study area.

Adverse Event Report Form 6 | 30-May-11 2 | 29-May-11 30 Records occurrence, details, care-seeking, and outcome of adverse event.

Weekly Morbidity Script for Women and Other Adults 1 | 04-Dec-10 1 | 04-Dec-10

34 Script to solicit information on symptoms of flu-like illness among enrolled pregnant women.

184

Weekly Morbidity Script for Infants and Other Children 4 | 27-Aug-12 4 | 27-Aug-12

35 Script to solicit information on symptoms of flu-like illness or pertussis and care visits among infants <6 months of age.

3 | 10-Oct-12 Antibody and Hepatitis E Substudy Form (AHSSF) — 4 | 31-Jul-13 51 Documents dates and label numbers for specimens collected as part of hepatitis substudy.

Post-PEF Tracking Form 1 | 23-Mar-12 58 Documents home and maiti (parents’ home) addresses and visit/return dates.

Shipper Tank / Liquid Nitrogen Maintain Log — 62 Used to track depth of liquid nitrogen in tanks to ensure maintenance of cold chain.

Weekly Lab-Work Schedule (WLS) 3 | 05-Oct-12 —

65 Used to schedule visits for specimen collection, document label numbers, and track met status.

Hepatitis E Substudy Pregnant Woman’s Consent 2 | 12-Mar-12 2 | 12-Mar-12 Script 66 Describes substudy, follow-up procedures, and specimen collection to pregnant women enrolled in MaGIFT and seeks consent.

Specimen Shipment Log 4 | 04-Dec-12 — 67 Records specimen numbers and aliquot types being shipped.

Laboratory Specimen Collection Log (Blood & 2 | 05-Oct-12 — 69 Breastmilk) (LSBL)

Records specimen and participant ID numbers for specimens received in field lab.

185

Specimen Management Log (Blood) 2 | 04-Dec-12 — 71 Records aliquots made from blood specimens and tracks storage and destination.

Hepatitis E Symptomatic Pregnant Woman’s 1 | 12-Apr-11 1 | 11-Apr-11 Consent Script 78 Describes, and seeks consent for, questionnaire and blood draw from women with symptoms of hepatitis-like illness.

HEV Exposure Assessment Questionnaire (HEV- 5 | 18-Oct-12 2 | 15-Oct-12 EAQ)

79 Records potential hepatitis E-related risk factors, including hygiene and sanitation practices, water and food sources and storage, occupation, travel history, and animal exposures.

Hepatitis Case Visit Form (HEV-CVF) 5 | 18-Oct-12 1 | 05-Oct-12

80 Records care sought, treatments used and their perceived effectiveness, and beliefs about illness, as well as label number and results of rapid HEV test and referral status.

Hepatitis E Report Card (HERC) 2 | 18-Oct-12 2 | 23-Jul-12

81 Records date of positive rapid HEV test and provides basic information about hepatitis E and when to seek medical care.

Highlighted forms were specifically developed or significantly modified for the hepatitis substudy and are included in this appendix.

186

FORM 14: MONTHLY PREGNANCY FOLLOW-UP FORM (PFF)

187

188

189

190

FORM 51 V.3: ANTIBODY AND HEPATITIS SUBSTUDY FORM (AHSSF)

191

FORM 51 V.4: ANTIBODY AND HEPATITIS SUBSTUDY FORM (AHSSF)

192

FORM 66: HEPATITIS E SUBSTUDY PREGNANT WOMAN’S CONSENT SCRIPT

193

194

FORM 69: LABORATORY SPECIMEN COLLECTION LOG (BLOOD & BREASTMILK) (LSBL)

195

FORM 71: SPECIMEN MANAGEMENT LOG (BLOOD)

196

FORM 78: HEPATITIS E SYMPTOMATIC WOMAN’S CONSENT SCRIPT

197

198

FORM 79: HEV EXPOSURE ASSESSMENT QUESTIONNAIRE (HEV-EAQ)

199

200

201

202

203

204

205

206

207

FORM 80: HEPATITIS CASE VISIT FORM (HEV-CVF)

208

209

210

FORM 81: HEPATITIS E REPORT CARD (HERC)

211

APPENDIX 3: FIELD LABORATORY SAFETY & INFORMATIONAL SIGNS

LABORATORY BIOHAZARD & EMERGENCY CONTACT INFORMATION

(Names and phone numbers of responsible personnel are blurred here for privacy.)

212

CAUTION: LIQUID NITROGEN

A safety problem was identified, wherein staff would occasionally handle socks of cryovials without insulated gloves.

A primary reason for this was that the insulated gloves were typically kept in a different room from the liquid nitrogen dewars due to minimal storage space inside the room (which had previously been a restroom). Solution: Fabric loops were sewn onto the gloves, and a hook was affixed to the wall next to the dewars so that the gloves would be both visible and easily accessible. Safe handling of liquid nitrogen was discussed during a safety training presentation (given by K. Charron and

A. Andrada) attended by all NNIPS staff who might encounter it, including vehicle drivers and the office sweeper.

213

LAB SAFETY

214

WEAR CLOSED-TOE SHOES IN THE LAB

215

SPECIMEN CHART FOR ONGOING NNIPS STUDIES

This wall chart was created because NNIPS was simultaneously conducting three major studies, each with sub-studies, and staff needed a handy reference for the different types of samples that were being processed and their associated codes.

Abbreviations:

FLU= Maternal Global Influenza Immunization Field Trial (MaGIFT), in which the hepatitis study was nested.

WLS = Weekly Lab Schedule; PP= postpartum; PEF= Pregnancy Enrollment Form; ILI = influenza-like illness.

NOMS= Nepal Oil Massage Study, evaluating the effects of using sunflower vs. mustard oil on the health and skin flora of infants, who are traditionally rubbed with oil by their mothers,

Cookstove = Cookstove Study, testing three types of improved cookstoves for acceptability and effects on indoor air pollution and respiratory health.

216

CURRICULUM VITAE

LISA J. KRAIN

born October 31, 1980, Evanston, Illinois, USA

EDUCATION

PhD Candidate, Epidemiology, Johns Hopkins Bloomberg School of Public Health to be completed 7/2016 Dissertation topic: Hepatitis E Virus (HEV) in the Terai of Nepal Advisors: Kenrad Nelson and Alain Labrique Master of Science (ScM), Epidemiology, Johns Hopkins Bloomberg School of Public Health 05/2008 Thesis topic: Treponema pallidum (yaws and syphilis) in Baka and Bantu populations, rural Cameroon Advisors: Nathan Wolfe and Taha Taha Bachelor of Arts (BA), Anthropology, University of California at Berkeley 05/2003 Graduated with high distinction in general scholarship, Phi Beta Kappa Earth & Planetary Science minor

RESEARCH POSITIONS

Graduate Assistant (Researcher), Johns Hopkins School of Public Health, Baltimore, Maryland 04/2011 – present Collaborated on design and implementation of longitudinal study of viral hepatitis among pregnant women in Nepal, funded by the Bill & Melinda Gates Foundation, and several related projects and manuscripts.

Predoctoral Fellow, Johns Hopkins Center for a Livable Future, Baltimore, Maryland 07/2009 – 06/2010 Assessed scope, components, and geographic coverage of methicillin-resistant Staphylococcus aureus (MRSA) surveillance programs under state and federal auspices. Evaluated availability and viability of surveillance data for epidemiologic study of sources, risk factors and transmission patterns of MRSA in community settings.

Graduate Assistant (Repository Manager), Wolfe Lab/Cameroon Program, Baltimore, Maryland 07/2006 – 02/2008 Maintained repository of human and animal blood and tissue samples, shipped specimens, prepared proposals for funding and IRB, developed data and repository management strategies for Zoonotic Emergence Network (now Global Viral Forecasting Initiative), ordered supplies for Baltimore & Yaoundé labs, supervised work-study students.

Graduate Assistant (Researcher), Johns Hopkins School of Public Health, Baltimore, Maryland 10/2005 – 12/2007 Reviewed and summarized literature on phylogeny, hosts, and emergence of viral, bacterial, and protozoan human pathogens. Contributed to “Characteristics of some major human infectious diseases (Table S1)”, and “Details on each of our 25 major infectious diseases (Note S9)”, supplements to Wolfe, ND, Panosian Dunavan C, and Diamond JM, Origins of major human infectious diseases, Nature 447, 279-283, 2007.

Intern, South Coast Air Quality Management District (AQMD), Diamond Bar, California 08/2004 - 10/2004 Integrated several existing data sets to produce airborne pollution estimates for ocean-going vessels at the Port of Long Beach; performed literature search on bioavailability of Cr(VI) in paints and primers; assembled information on rail shipping routes and facilities in Southern California from Internet sources.

Laboratory Assistant, Berkeley Seismological Laboratory, Berkeley, California 10/1999 - 12/2002 Produced digital maps using a variety of geographic information sources and earthquake databases; designed and maintained websites for education and research initiatives; prepared graphics; participated in public earthquake response. 217

PEER-REVIEWED PUBLICATIONS

Krain LJ, Atwell JA, Nelson KE, and Labrique AB. Fetal and neonatal health consequences of vertically-transmitted hepatitis E virus infection. 2014. American Journal of Tropical Medicine and Hygiene 90(2): 365-370. 2014. [doi:10.4269/ajtmh.13-0265]

Krain LJ, Nelson KE, and Labrique AB. Host immune status and response to hepatitis E virus infection. Clinical Microbiology Reviews 27(1): 139-165. 2014. [doi:10.1128/CMR.00062-13]

Kmush B, Wierzba T, Krain L, Nelson K, Labrique AB. Epidemiology of hepatitis E in low-and middle-income countries of Asia and Africa. Seminars in Liver Disease 33(01):15-29. 2013. [doi:10.1055/s-0033-1338111]

Labrique AB, Sikder SS, Krain LJ, West KP Jr., Christian P, Rashid M, Nelson KE. Hepatitis E, a vaccine-preventable cause of maternal death. Emerging Infectious Diseases 18(9). 2012. [doi:10.3201/eid1809.120241]

Peterson AP, Davis, MF, Julian KG, Awantang G, Greene WH, Price LB, Waters A, Doppalapudi A, Krain LJ, Nelson KE, Silbergeld EK, Whitener CJ. Molecular and phenotypic characteristics of healthcare- and community-associated methicillin- resistant Staphylococcus aureus at a rural hospital. PLoS ONE 7(6). 2012. [doi:10.1371/journal.pone.0038354]

Contributed Table S1 and Note S9 to Wolfe ND, Dunavan CP, Diamond J. Origins of major human infectious diseases. Nature 447(7142):279-83. 2007.

PRESENTATIONS AND SCIENTIFIC POSTERS

*Peterson AE, *Davis MF, Krain LJ, Julian KG, Doppalapudi A, Sperry NL, Pitzer M, Greene W, Price LB, Silbergeld EK, Whitener C, Discordance between epidemiologic and molecular classifications of health care- versus community-associated MRSA [poster], Infectious Diseases Society of America, Vancouver, BC, October, 2010.

*Krain LJ and Silbergeld EK, Challenges and opportunities for source tracking of community-associated methicillin-resistant Staphylococcus aureus (MRSA) in the United States [poster], Int’l. Conf. on Emerg. Infect. Dis., Atlanta, GA, July, 2010.

*Krain LJ and Silbergeld EK, MRSA surveillance in the United States: where do we go from here? [oral presentation], Pew Charitable Trusts, Washington, DC, April, 2010.

* indicates presenter(s) for posters and talks

HONORS AND AWARDS

Center for a Livable Future Predoctoral Fellowship 07/2009 – 06/2010 Baltimore Albert Schweitzer Fellowship (for development and implementation of service project) 05/2007 – 05/2008 JHSPH Master’s Tuition Scholarship (for outstanding work in first two terms) 03/2006 – 05/2006 Superior Service Award, Natural Science Division, Pasadena City College (for work as TA) 06/2005 Elected to Phi Beta Kappa at UC Berkeley (Alpha of California Chapter) 05/2003 California Alumni Association Jonathan D. Sauer Scholarship (for natural science student) 08/2001 - 05/2002 California Alumni Association Emerging Leader Scholarship 08/2000 - 05/2001

218

TEACHING EXPERIENCE Teaching Assistant, Johns Hopkins University, Baltimore, Maryland 06/2006–05/2011 Epi 340.651 Emerging Infections (grad) – 4th term 2010-2011 Epi 340.627 Epidemiology of Infectious Diseases (grad, online) – 3rd term 2010-2011 Epi 280.350 Fundamentals of Epidemiology (undergrad) – spring 2009-2010 Epi 340.601 Principles of Epidemiology (grad) – summer 2010 Epi 340.608 Observational Epidemiology (grad, online) – 3rd term 2008-2009 Epi 340.653 Outbreak Investigations (grad) – 4th term 2007-2008 Epi 340.752 Epidemiological Methods 2 (grad) – 2nd term 2007-2008 Epi 340.847 Origins: Cross-Species Transmission and the Emergence of Human Diseases (grad) – summer 2006

Teaching Assistant, Pasadena City College, Pasadena, California 01/2004 – 06/2005 Designed discipline-specific critical-thinking activities in undergraduate courses Physo 2B: “Physiology and Anatomy” & Geol 1: “Physical Geology”; assisted students with quantitative, science reasoning, and writing skills. Reader, University of California at Berkeley, Berkeley, California 08/2000 - 12/2000 Graded exams, gave feedback to undergraduates, and created website for Earth & Planetary Sci. 20: “Earthquakes”.

SERVICE

Peer Review for Journals since 2014 Clinical and Vaccine Immunology, Clinical Microbiology Reviews, Emerging Infectious Diseases, Journal of Clinical Microbiology, Journal of Viral Hepatitis, Pathogens and Global Health

Earthquake Engineering Research Institute (EERI) Nepal Earthquake Reconnaissance Team 5/2015-6/2015 Served as “Virtual Team Collaborator”, compiling information and images from news, social media, and non-governmental organizations on physical and operational impact of the April 2015 earthquake on healthcare facilities in the Kathmandu Valley and in Gorkha District in preparation for recon team’s field survey.

Dyslexia Tutoring Program 11/2007- 8/2011 Completed 20 hrs. training in Orton-Gillingham instruction methods. Worked one-on-one with dyslexic student to improve academic skills and to increase engagement in learning. Baltimore-Area Resources for Adults with Learning Disabilities and/or ADHD (Schweitzer Project) 09/2006-12/2008 Initiated booklet/website project to facilitate access to information about free and low-cost diagnostic, treatment, support, and educational services, and to informally assess barriers to access. General Epidemiology/Epidemiological Methods Journal Club Coordinator 10/2006 - 05/2007 Masters Thesis “Brown Bag” Coordinator 02/2007 - 05/2007 Epidemiology Comprehensive Exam Review Co-Coordinator 02/2006 – 06/2006 Errand and Escort Service, Huntington Memorial Hospital, Pasadena, California 01/2004 – 05/2005 Logged 220+ hours transporting patients, ferrying charts and blood units, and delivering meals and flowers. Dean’s Student Advisory Committee, College of Letters and Science, Berkeley, California 01/2002 - 5/2002 Education and Outreach Coordinator, Cal Disabled Students’ Union, Berkeley, California 08/1999 - 05/2001 Organized presentations about common disabilities for students and staff, put together campus-wide disability awareness day featuring speakers and approx. 50 disability organizations, designed website. Tutor/Teacher's Aide, Washington Elementary School, Berkeley, California 08/1999 - 05/2001 Spent 100+ classroom hours assisting K, 2, 5, and LD resource room students in language and math. Tutor, Volunteers in Teaching Adults (VITA) Literacy Program, Des Plaines, Illinois 06/1998 - 08/1998 219

PREVIOUS RESEARCH AND FIELDWORK EXPERIENCE Conducted museum-based descriptive/stratigraphic study of Neogene Caribbean gastropods (snails) 08/2002 - 12/2002 Assisted with tomographic study of Earth’s upper mantle using cross-correlation of seismic waveforms 05/2000 - 08/2000 Participated in 8-week archaeological excavation at Tel Dor 06/1999 - 08/1999

OTHER EMPLOYMENT

Personal Care Attendant, Alameda County, Berkeley, California 10/1999-12/1999 Assisted college student with cerebral palsy with activities of daily living.

PROFESSIONAL AFFILIATIONS

American Society of Tropical Medicine and Hygiene (ASTMH) American Public Health Association (APHA), Epidemiology Section Geological Society of America (GSA), Geology and Human Health Division American Geophysical Union (AGU), Biogeosciences Division

ADDITIONAL SKILLS AND CERTIFICATIONS

Computing UNIX, Windows, and Mac platforms Experience HTML, CSS, and website design Statistical software: STATA, R Geographic information systems and digital cartography: ArcGIS, GMT Graphics software: PhotoShop, PaintShop Pro Bioinformatics tools: NCBI (BLAST, PubMed, etc.), MEGA6 (sequence alignment/phylogenetics) Reference management: EndNote, RefWorks FreezerWorks repository management software

Certifications JHSPH CITI Human Subjects Research Training (2007, 2009) JHU Animal Care and Use (2006) JHMI Biosafety Laboratory Training (2006, 2013) JHMI Bloodborne Pathogens Safety (2006, 2009, 2013) JHMI DOT/IATA Hazardous Materials Shipping (2006, 2009, 2013) JHMI Fire Safety and Hazard Communication (2009, 2013) National Weather Service Skywarn Spotter – Basics I Course (2007, 2012) National Weather Service Skywarn Spotter – Advanced Course (2013) Dyslexia Tutoring Program – 20-Hour Orton-Gillingham Method Tutor Training Course (2007)

Languages English – native language French – reading, writing, and conversational abilities (once fluent, now rusty) Nepali – beginning reading, writing, and conversational skills Spanish – intermediate reading knowledge, limited conversation and writing Hebrew – basic conversational, reading, and writing knowledge German, Yiddish – limited reading knowledge

Research interests: Epidemiology at the intersection of environmental health and infectious diseases; disease emergence; spatial and molecular epidemiology; natural disasters; ethics. 220