The Epidemiology of Pork-Borne Parasitic Zoonoses in Rural Communities in the Central Highlands of Vietnam

THE EPIDEMIOLOGY OF PORK-BORNE PARASITIC ZOONOSES IN RURAL COMMUNITIES IN THE CENTRAL HIGHLANDS OF VIETNAM

Dinh Ngoc Nguyen http://orcid.org/0000-0002-2028-2186

Submitted in total fulfilment of the requirements for the degree of Doctor of Philosophy

May 2018 Faculty of Veterinary and Agricultural Sciences The University of Melbourne ABSTRACT

Background Food-borne parasitic zoonoses impact negatively on humans in terms of health and loss of productivity as well as economically, through loss of market access and trade. Taenia and Trichinella are two of the most important meat-borne zoonoses globally. In the Central Highlands of Vietnam, living standards are poor; open defaecation via use of outdoor latrines is commonly practiced and livestock access to these latrine areas is, for the most part, unrestricted. Moreover, the consumption of undercooked or raw pork dishes is common traditional practice. These factors combined are highly conducive for the transmission of several pork-borne parasitic zoonoses, including taeniasis, Taenia solium cysticercosis and trichinellosis. In this context, developing and validating diagnostic assays to study the epidemiology of these pork-borne parasites within these communities allows risk factors for exposure and infection to be ascertained and this, in turn, allows community-based control programs to be better targeted.

Methodology/Principle findings This project was conducted as a cross-sectional study in Buon Don, Krong Nang and M’Drak district in Dak Lak province in the Central Highlands of Vietnam. A combination of parasitological, serological and molecular diagnostic methods were utilised to determine the levels of infection of Taenia spp. in humans and Trichinella spp. in pigs, and seroexposure to T. solium in humans and pigs. The data were analysed in relation to host, management and environmental risk factors using frequentist and Bayesian multivariable analyses.

In Chapter 2, a systematic review on previous research on taeniasis, T. solium cysticercosis and trichinellosis was performed and showed that little to no data are available on the prevalence of these food-borne zoonoses in the central and southern areas of the country. Moreover, utilisation of tests with either low sensitivity or poor specificity for the diagnosis of taeniasis in humans and T. solium cysticercosis in humans and pigs produced either an over- or underestimation of their TP. To address this, a Bayesian model was used to estimate the TP of taeniasis and Taenia solium

i cysticercosis and demonstrated a true prevalences of up to 25% and 24%, respectively for specific rural ‘hotspots’.

In Chapter 3, a multiplex Taq-Man probe-based real-time PCR for the species-specific detection of three Taenia species in human stool was developed and its diagnostic parameters compared with the Kato-Katz thick smear and coproantigen ELISA assay using field samples collected as part of a cross-sectional survey in Dak Lak province. The overall AP of human taeniasis by the multiplex qPCR was 6.7% (95% CI 4.4 to 10). The sympatric existence of Taenia solium, Taenia asiatica and Taenia saginata in Dak Lak province was confirmed for the first time, as was the extension of the known distribution of T. asiatica to southern Vietnam. Risk factors associated with Taenia spp. infection included the routine consumption of undercooked beef and pork, routine consumption of pork tongue, and personal observation of Taenia proglottids in stool.

Chapter 4 describes a cross-sectional study on the seroprevalence of human T. solium cysticercosis in Dak Lak determined by two diagnostic assays, the apDia-ELISA, a commercial antigen detection assay and the LLGP-EITB antibody detection assay. The seroprevalence of exposure to T. solium in humans was 5% (95% CI 3 to 8), while the apDia-ELISA, detected circulating T. solium antigens in 7.3% (95% CI 4.9 to 10.8) of the sampled population. Consuming raw vegetables, drinking water sourced from lakes, streams or ponds, and outdoor defaecation were associated with increased T. solium exposure risk in humans.

In Chapter 5, the seroexposure of pigs to T. solium cysticercosis in Dak Lak was determined using rT24H and native LLGP antigens in series, in an EITB format. The seroprevalence of exposure to T. solium was low at 0.94 (95% CI 0.51 to 1.68). Coprophagy of human faeces and scavenging for food were factors significantly associated with T. solium exposure in pigs.

In Chapter 6, the contribution of all modifiable risk factors for taeniasis, T. solium exposure in humans and pigs is explored and this was estimated to range from 24% to 77%. Spatial analyses demonstrated that human- or pig-T. solium exposure-positive households were more likely to be surrounded by other exposure-positive households in some study locations. This spatial aggregation of human exposure-positive households

ii was hypothesised to be due to a combination of demographic, anthropological and micro environmental factors, whilst scavenging for food and coprophagy of human faeces are likely to drive the aggregation of pig exposure-positive households.

Finally, no Trichinella larvae were detected in the confined and free-roaming pig populations using the artificial digestion technique on tongue, diaphragm and masseter muscle. The estimated TP of Trichinella was low, ranging from 0.09 to 0.80% in domestic pigs. Further diagnostic screening using serology was beyond the scope of this current project but advocated for future research.

Conclusions This thesis has contributed significantly to a better understanding of the epidemiological characteristics of Taenia spp. infection, and seroexposure to T. solium and Trichinella spp. infection among humans and pigs in the Central Highlands of Vietnam. This information will assist local government and residents to develop appropriate methods and strategies to reduce or mitigate the burden of these diseases in Dak Lak.

iii

DECLARATION

This is to certify that

I. the thesis comprises only my original work towards the PhD except where indicated the text II. due acknowledgement has been made in the text to all other materials used III. the thesis is fewer than 100,000 words in length, exclusive of tables, figures, bibliographies and appendices

……………… Dinh Ngoc Nguyen Faculty of Veterinary and Agricultural Sciences The University of Melbourne

iv PREFACE

This study outcomes presented here were achieved by self-effort with the dedicated support from colleagues and students at Tay Nguyen University, local medical staff and veterinarians in Dak Lak, Vietnam that assisted with fieldwork, and from experts in Australia, Belgium and the USA that assisted laboratory and statistical analysis.

The results herein are original work conducted during the duration of my PhD and has not been not published previously except for the references cited in the text. This thesis is composed of a combination of peer reviewed published papers, manuscripts that are currently under review and those in preparation. This thesis content does not contain in any part work that has been submitted to qualify for the award of any other degree or diploma in any university or other tertiary institution.

Peer-reviewed articles published in international scientific journals

Ng-Nguyen D, Stevenson MA, Traub RJ. A systematic review of taeniasis, cysticercosis and trichinellosis in Vietnam. Parasit Vectors 2017;10:150.

Ng-Nguyen D, Stevenson MA, Dorny P, Gabriël S, Vo T Van, Nguyen VAT, Phan TV, Hii SF, Traub RJ. Comparison of a new multiplex real-time PCR with the Kato Katz thick smear and copro-antigen ELISA for the detection and differentiation of Taenia spp. in human stools. PLoS Negl Trop Dis 2017;11:e0005743.

Ng-Nguyen D, Stevenson MA, Breen K, Phan T Van, Nguyen V-AT, Vo T Van, et al. The epidemiology of Taenia spp. infection and Taenia solium cysticerci exposure in humans in the Central Highlands of Vietnam. BMC Infect Dis 2018;18:527.

Ng-Nguyen D, Noh J, Breen K, Stevenson MA, Handali S, Traub RJ. The epidemiology of porcine Taenia solium cysticercosis in communities of the Central Highlands in Vietnam. Parasit Vectors. 2018;11:360.

Ng-Nguyen D, Traub RJ, Nguyen V-AT, Breen K, Stevenson MA. Spatial distribution of Taenia solium exposure in humans and pigs in the Central Highlands of Vietnam. PLoS Negl Trop Dis 2018;12:e0006810.

Conference proceedings and seminars given

Ng-Nguyen D, 2017. The epidemiology of pork-borne parasitic zoonoses in rural communities in the central highlands of vietnam. PhD Oration, Faculty of Veterinary and Agricultural Sciences, The University of Melbourne, Melbourne, Australia, 30 November 2017.

Ng-Nguyen D, Stevenson MA, Handali S, Noh J, Traub RJ, 2017. Individual- and environmental-level risk factors of Taenia solium cysticercosis in the Central Highlands of Vietnam. Annual Conference of the Australian Society for Parasitology, Leura, Australia, 26-29 June 2017.

Ng-Nguyen D, Stevenson MA, Traub RJ. Individual- and environmental-level risk factors for taeniasis and taenia solium cysticercosis in the central highlands of Vietnam. International Congress for Tropical Medicine and Malaria, Brisbane, Australia, 18-22 September 2016.

Ng-Nguyen D, 2015. The epidemiology of pork-borne parasitic zoonoses in rural communities in the central highlands of Vietnam. Confirmation of Candidature, Faculty of Veterinary and Agricultural Sciences, The University of Melbourne, Melbourne, Australia, 24 February 2015.

ACKNOWLEDGEMENTS

This thesis is dedicated my parents, Nguyen Manh Dong and Pham Thi Nhung, and Aunt, Nguyen Thi Boc, who always provided their endless love and blessings throughout my life. This thesis is especially dedicated to my beloved wife, Nguyen Thi Van-Anh, for her constant encouragement, limitless giving, great sacrifice and help and support during my PhD study; to my kids, Nguyen Ngoc Dang Khoa and Nguyen Ngoc Thao Vy, for their endurance owing to their lack of a Father’s care during my period away from home.

It is with deepest gratitude that I acknowledge my supervisors, Prof. Rebecca Justine Traub and Prof. Mark Anthony Stevenson. This study would not have been possible without their valuable guidance and expert advice. My profound gratitude is especially extended to Prof. Rebecca Justine Traub for her patience, motivation and inspiration for me to commence my first steps towards a career in academia. The results of this PhD thesis would not have been obtained without financial support from my dear supervisor Prof. Rebecca Justine Traub who funded both fieldwork and laboratory costs.

My dreams of undertaking a PhD in Australia would not have come true without an Australia Awards Scholarships (AAS) bestowed on me by the Department of Foreign Affairs and Trade, Australian Government. I would like to express my sincere gratitude to AAS for the valuable support before and during my studies and in future.

I would like to take this opportunity to say warmly thanks to Tay Nguyen University colleagues and students, the local medical and veterinary staff, the Institute of Biotechnology and Environmental Sciences, and the Faculty of Animal Sciences and Veterinary Medicine of Tay Nguyen University, Vietnam for their kind support during my fieldwork.

I would like to express my sincere gratitude to my dear friends, Thomas Teoh and Hii Sze Fui for their mental support and company during my period of stay in Australia. My special thanks to Hii Sze Fui for her valuable advice on laboratory work.

vii

TABLE OF CONTENTS

ABSTRACT ...... i PREFACE ...... v ACKNOWLEDGEMENTS ...... vii TABLE OF CONTENTS ...... viii LIST OF FIGURES ...... xiv LIST OF TABLES ...... xvi LIST OF ABBREVIATIONS ...... xviii

Chapter 1 - Introduction and literature review

1.1 Introduction ...... 2 1.1.1 Background ...... 2 1.1.2 Research hypotheses, goals and objectives ...... 4 1.1.3 Contribution of this thesis ...... 5 1.1.4 Research and thesis structure ...... 6 1.2 Human tapeworm Taenia ...... 10 1.2.1 Morphological classification ...... 10 1.2.2 Molecular classification ...... 10 1.2.3 Life cycle of human tapeworms ...... 11 1.2.3.1 Taenia solium ...... 12 1.2.3.2 Taenia asiatica ...... 12 1.2.3.3 Taenia saginata ...... 13 1.2.4 Prevalence and geographic distribution ...... 14 1.2.5 Health and economic impact of taeniasis and cysticercosis ...... 19 1.2.6 Diagnosis of cysticercosis in animals ...... 20 1.2.6.1 Post-mortem carcass inspection...... 20 1.2.6.2 Ante-mortem examination...... 21 1.2.6.3 Antibody-ELISA ...... 21 1.2.6.4 Antigen-ELISA ...... 22 1.2.6.5 Enzyme-linked Immunoelectrotransfer Blot ...... 23 1.2.7 Diagnosis of cysticercosis in humans ...... 24 1.2.7.1 Imaging techniques ...... 24 1.2.7.2 Antibody-ELISA ...... 25 1.2.7.3 Antigen-ELISA ...... 26

viii

1.2.7.4 Enzyme-Linked Immunoelectrotransfer Blot ...... 26 1.2.8 Diagnosis of taeniasis...... 27 1.2.8.1 Faecal examination ...... 27 1.2.8.2 Polymerase Chain Reactions ...... 28 1.2.8.3 Immunological technique ...... 30 1.3 Trichinella spp...... 31 1.3.1 Taxonomy ...... 31 1.3.2 Life cycle...... 32 1.3.3 Clinical signs in humans ...... 33 1.3.4 Prevalence and geographic distribution ...... 34 1.3.5 Impact of Trichinella and trichinellosis ...... 36 1.3.6 Diagnosis of Trichinella in pork ...... 37 1.3.6.1 Artificial digestion methods ...... 37 1.3.6.2 Molecular diagnosis ...... 38 1.3.6.3 Enzyme-linked Immunosorbent Assay ...... 39 1.3.6.4 Enzyme-linked Electroimmunotransfer Blot or Western Blot ...... 40 1.4 References ...... 41

Chapter 2 - A systematic review of taeniasis, cysticercosis and trichinellosis in Vietnam

2.1 Abstract ...... 77 2.2 Background ...... 78 2.3 Methods ...... 80 2.3.1 Search protocol ...... 80 2.3.2 Study selection ...... 80 2.3.3 Data collection ...... 81 2.3.4 Statistical analysis ...... 81 2.4 Results ...... 82 2.4.1 Records used for quality and quantity analysis ...... 82 2.4.2 Human taeniasis ...... 83 2.4.3 Human cysticercosis ...... 84 2.4.4 Porcine cysticercosis ...... 86 2.4.5 Trichinellosis ...... 86 2.5 Discussion ...... 87 2.6 Conclusions ...... 89 2.7 References ...... 91

Chapter 3 - Comparison of a new multiplex real-time PCR with the Kato Katz thick smear and copro-antigen ELISA for the detection and differentiation of Taenia spp. in human stools

3.1 Abstract ...... 111 3.1.1 Background ...... 111 3.1.2 Methodology/Principal findings ...... 111 3.1.3 Conclusions ...... 112 3.1.4 Author summary ...... 112 3.2 Introduction ...... 113 3.3 Materials and methods ...... 115 3.3.1 Multiplex real-time PCR ...... 115 3.3.2 Coproantigen enzyme-linked-immunosorbent assay ...... 118 3.3.3 Microscopy examination ...... 119 3.3.4 Study site and sampling ...... 119 3.3.5 Ethics ...... 121 3.3.6 Statistical analysis ...... 121 3.4 Results ...... 122 3.4.1 Optimization and validation of the multiplex qPCR ...... 122 3.4.2 Field study ...... 122 3.4.3 Comparison of T3qPCR with microscopy ...... 123 3.4.4 Comparison of T3qPCR with cAg-ELISA ...... 123 3.5 Discussion ...... 124 3.6 References ...... 127

Chapter 4 - The epidemiology of Taenia spp. infection and Taenia solium cysticerci exposure in humans in the Central Highlands of Vietnam

4.1 Abstract ...... 139 4.1.1 Background ...... 139 4.1.2 Methods ...... 139 4.1.3 Results ...... 139 4.1.4 Conclusions ...... 140 4.2 Background ...... 140 4.3 Methods ...... 141 4.3.1 Study site and sampling ...... 141

4.3.2 Household questionnaire ...... 142 4.3.3 Laboratory procedures ...... 143 4.3.3.1 Detection of antibodies against T. solium cysticerci ...... 143 4.3.3.2 Detection of Taenia spp. infection ...... 143 4.3.4 Statistical methods ...... 143 4.4 Results ...... 146 4.4.1 General description of study population and data structure ...... 146 4.4.2 Risk factors for T. solium cysticerci exposure ...... 146 4.4.3 Risk factors for Taenia spp. infection ...... 147 4.5 Discussion ...... 148 4.6 Conclusions ...... 151 4.7 Declarations ...... 151 4.7.1 Ethics approval and consent to participate ...... 151 4.7.2 Funding ...... 152 4.7.3 Authors' contributions ...... 152 4.7.4 Acknowledgements ...... 152 4.8 References ...... 153

Chapter 5 - The epidemiology of porcine Taenia solium cysticercosis in communities of the Central Highlands in Vietnam

5.1 Abstract ...... 163 5.1.1 Back ground ...... 163 5.1.2 Methods ...... 163 5.1.3 Results ...... 163 5.1.4 Conclusions ...... 164 5.2 Background ...... 164 5.3 Materials and Methods ...... 165 5.3.1 Study site and sampling ...... 165 5.3.2 Sample size estimation ...... 166 5.3.3 Laboratory procedures ...... 166 5.3.4 Statistical analysis ...... 168 5.4 Results ...... 170 5.4.1 General description of study population ...... 170 5.4.2 Risk factors for porcine T. solium cysticercosis ...... 171 5.5 Discussion ...... 171 5.6 Conclusions ...... 173

5.7 Ethics approval and consent to participate ...... 174 5.8 References ...... 175

Chapter 6 - Spatial distribution of Taenia solium exposure in humans and pigs in the Central Highlands of Vietnam

6.1 Abstract ...... 185 6.1.1 Background ...... 185 6.1.2 Methodology/Principal findings ...... 185 6.1.3 Conclusions ...... 186 6.1.4 Author summary ...... 186 6.2 Introduction ...... 187 6.3 Materials and methods ...... 189 6.3.1 Ethics statement ...... 189 6.3.2 Field procedures ...... 189 6.3.3 Taenia solium carrier identification ...... 191 6.3.4 Human T. solium exposure identification ...... 191 6.3.5 Pig T. solium exposure identification ...... 191 6.3.6 Spatial data analyses ...... 191 6.4 Results ...... 193 6.4.1 General data description ...... 193 6.4.2 Distribution of pig T. solium exposure households...... 194 6.4.3 Distribution of human T. solium exposure households ...... 194 6.4.4 Distribution of Taenia-positive households ...... 194 6.5 Discussion ...... 195 6.6 References ...... 199

Chapter 7 - The prevalence of Trichinella in pigs in the Central Highlands of Vietnam

7.1 Introduction ...... 212 7.2 Materials and methods ...... 213 7.2.1 Study sites and sampling ...... 213 7.2.2 Artificial digestion ...... 214 7.2.3 Statistical analysis ...... 215 7.3 Results ...... 216

xii

7.4 Discussion ...... 216 7.5 Conclusions ...... 217 7.6 References ...... 218

Chapter 8 - General discussion

8.1 Aims of the study revisited ...... 224 8.2 Limitation of this thesis ...... 225 8.3 Application of new and multiple diagnostic tests ...... 227 8.3.1 Diagnosis of taeniasis...... 227 8.3.2 Diagnosis of Taenia solium cysticercosis in humans...... 228 8.3.3 Diagnosis of Taenia solium cysticercosis in pigs ...... 228 8.4 Epidemiological characteristics of Taenia in Vietnam...... 229 8.5 What are the distinctive characteristics of Dak Lak? ...... 232 8.6 Trichinella study - The first attempt ...... 233 8.7 Recommendations for controlling of the pork-borne zoonoses...... 234 8.7.1 Community education ...... 235 8.7.2 Veterinary measures ...... 236 8.7.3 Medical measures ...... 237 8.7.4 Other measures to ensure the success of intervention programs ...... 239 8.8 Future directions ...... 240 8.9 References ...... 243

xiii

LIST OF FIGURES

Fig. 1.1 Thesis structure ...... 9

Fig. 1.2 The life cycle of three human Taenia tapeworms ...... 14

Fig. 1.3 Trichinella life cycle ...... 33

Fig. 2.1 Flow diagram of searching strategy...... 104

Fig. 2.2 Studies of the prevalence of taeniasis in Vietnam, 1999 to present...... 105

Fig. 2.3 Studies of the prevalence of human T. solium cysticercosis in Vietnam, 1992 to present...... 106

Fig. 2.4 Studies of the prevalence of T. solium cysticercosis in pigs in Vietnam, 1994 to present...... 107

Fig. 2.5 Map of trichinellosis outbreaks in Vietnam...... 108

Fig. 2.6 Distribution of human T. solium cysticercosis and taeniasis...... 109

Fig. 3.1 Map of study site. Map of Dak Lak province showing districts where sampling was carried out...... 135

Fig. 3.2 Ct value comparison of multiplex to singleplex qPCR ...... 136

Fig. 3.3 Infection status of human Taenia species ...... 137

Fig. 6.1 Map of Dak Lak province showing the location of study sites...... 204

Fig. 6.2 A picture showing a typical household in rural areas in Dak Lak province. .. 205

Fig. 6.3 Location of households for human (a and b) and pig (c and d) T. solium exposure in Ea Noul (a and c) and Ea Wer (b and d) villages in Buon Don district .... 206

xiv

Fig. 6.4 Location of households for human (a and b) and pig (c and d) T. solium exposure in Ea Ho (a and c) and Dlieya (b and d) villages in Krong Nang district ..... 207

Fig. 6.5 Location of households for human (a and b) and pig (c and d) T. solium exposure in Krong Jing (a and c) and Cu Mta (b and d) villages in M’Drak district ... 208

Fig. 6.6 Estimated K-function differences for human and pig T. solium exposure- and Taenia-positive households in three studied districts ...... 209

Fig. 6.7 Location of T. solium taeniasis- and exposure-positive households ...... 210

Fig. 7.1 Estimated true prevalence of Trichinella with assuming test Sp of 100%...... 222

Fig. 8.1 The pattern for transmission and circulation of taeniasis and T. solium exposure in humans and pigs in communities of Dak Lak province ...... 250

LIST OF TABLES

Table 1.1 Distribution of Trichinella taxa ...... 34

Table 2.1 Research on human taeniasis and estimated true prevalence in Vietnam .... 100

Table 2.2 Hospital and community-based survey on human cysticercosis and estimated true prevalence of cysticercosis in Vietnam ...... 101

Table 2.3 Research on porcine cysticercosis and estimated true prevalence in Vietnam ...... 102

Table 2.4 Trichinellosis outbreaks in Vietnam since 1970 ...... 103

Table 3.1 Primers and probes for multiplex qPCR ...... 133

Table 3.2 gBlocks positive controls ...... 133

Table 3.3 Parameters of prior information ...... 133

Table 3.4 Results of three diagnostic tests for taeniasis ...... 134

Table 3.5 The agreement statistics of T3qPCR, cAg-ELISA and KK for the detection of Taenia spp. in stool ...... 134

Table 3.6 The estimated characteristics of three diagnostic tests for the detection of Taenia spp. in stool ...... 134

Table 4.1 Structure of the data from 342 study participants from six villages in M’Drak, Buon Don and Krong Nang districts ...... 158

Table 4.2 General description of household data ...... 158

Table 4.3 Demographic data for individual study participants ...... 159

xvi

Table 4.4 Risk factors associated with T. solium cysticerci exposure identified by a LLGP-EITB assay ...... 160

Table 4.5 Logistic regression model showing the effect of sex, dietary and observation of proglottids on the odds of a taeniasis case ...... 161

Table 5.1 Details of information about general and pig information ...... 180

Table 5.2 Structure of the data from 1281 study pigs from six villages in M’Drak, Buon Don and Krong Nang districts ...... 180

Table 5.3 General description of household data ...... 181

Table 5.4 General description of pig data...... 182

Table 5.5 Risk factors associated with T. solium cysticercosis positive in pigs...... 183

Table 6.1 Spatial characteristics of T. solium exposure and taeniasis. Data structure of household and individual sampling ...... 203

Table 6.2 Spatial characteristics of T. solium exposure and taeniasis. Pig management in the study districts of Dak Lak province ...... 203

Table 7.1 Estimated true prevalence of Trichinella with assuming test Sp of 100% ... 222

xvii

LIST OF ABBREVIATIONS

µL microliter Ab-ELISA antibody-ELISA AD artificial digestion Ag-ELISA antigen-ELISA AP apparent prevalence apDia-ELISA commercial ApDia cysticercosis antigen-ELISA bp base pair cAg-ELISA copro-antigen ELISA CI confidence interval Cob cytochrome oxidase-b Cox-1 cytochrome c oxidase subunit I CrI credible interval CT computed tomography CT computed tomography DALYs disability-adjusted life years EITB enzyme-linked immunoelectrotransfer blot ELISA enzyme-linked immunosorbent assay ESV expansion segment V FAO food and agriculture organization FAST-ELISA falcon assay screening test ELISA Fg femtogram GDP gross domestic product Ig immunoglobulin ITRC international Trichinella reference centre ITS-2 internal transcribed space-2 kDa kilo Danton KK kato-katz thick smear LAMP loop-mediated isothermal amplification LLGP lentil-lectin glycoprotein

xviii

LLGP-EITB lentil-lectin glycoprotein EITB LPG larvae per gram MC Markov chain MCMC Markov chain Monte Carlo mg milligram MoAb monoclonal antibody MRI magnetic resonance imaging MRI magnetic resonance imaging NAHD nicotine adenine dinucleotide dehydrogenase NCC neurocysticercosis OD optical density OIE world organisation for animal health PCR polymerase chain reaction PCR-RFLP PCR-restriction fragment length polymorphism preferred reporting items for systematic reviews and meta- PRISMA analyses qPCR real-time PCR rDNA ribosomal deoxyribonucleic acid rRNA ribosomal ribonucleic acid

SDS-PAGE sodium dodecyl sulphate polyacrylamide gel electrophoresis

Se sensitivity Sp specificity T3qPCR multiplex real-time PCR TP true prevalence tRNA transfer ribonucleic acid WB western blot WHO world health organization

xix

CHAPTER 1

Introduction and literature review

1.1 Introduction

1.1.1 Background

Zoonotic diseases comprise 61% of human pathogens [1] and approximately 25% of the world’s population suffer some form of parasitic infection [2]. Zoonotic diseases negatively impact on public health systems, productivity of livestock, food-safety, the economy and ability to trade. In 2012, a multicriteria-based ranking for risk management of food-borne parasites conducted by the WHO/FAO ranked Taenia solium as the most important, and Trichinella spiralis as the seventh most important food-borne parasites in terms of public health, socioeconomic and trade impact [3]. The number of individuals suffering from human cysticercosis and Trichinella globally have been estimated at over 370,000 and 4,500 in 2010, respectively [4].

There are three species of food-borne Taenia spp. that are zoonotic namely, Taenia solium, Taenia asiatica and Taenia saginata [5, 6]. Humans act as definitive hosts for all three-tapeworm species via the ingestion of undercooked/raw meat or the visceral organs of pork, resulting in taeniasis [7]. Both T. solium and T. asiatica utilise swine as intermediate hosts; whereas, the intermediate hosts of T. saginata are cattle [8]. Humans may also act as the intermediate hosts for T. solium through accidental ingestion of T. solium eggs in contaminated food and/or water or through autoinfection. Poor hygiene and environmental sanitary conditions, in particular open defaecation, results in cysticercosis in cattle and pigs as a result of consuming water and/or food contaminated with Taenia eggs, passed by human tapeworm carriers [9].

In humans, the symptoms of taeniasis are subtle and include abdominal distension, abdominal pain, digestive disorders and anal pruritus [10–12]. T. solium cysticercosis in humans, on the other hand, may result in epilepsy, paralysis, dementia, chronic headache, blindness, muscular pain, subcutaneous nodules and even death [10, 13]. In contrast to human cysticercosis, there are few obvious clinical signs of cysticercosis in pigs [14]. Taenia infection not only negatively impacts on human health and well-being but also causes economic losses due to costs associated with treatment in humans, loss of productivity and in livestock, carcass condemnation. In West Cameroon, cysticercosis in humans and pigs results in an estimated loss of over 10 million Euros

2 per year [15]. In four provinces in Laos, the pork industry estimated a loss in value of US $55,000 to 96,000 when 20% of carcasses was condemned due to the presence of cysticerci over a 21-month period [16].

Currently, T. solium is endemic in developing countries of Asia, Africa, and Latin America [17], particularly in areas where poverty is widespread, pigs are allowed to freely roam and consumption of raw or under cooked pork is common. Taenia infection has been reported to be endemic in Southeast Asia [18]. The prevalence of the infection in humans and pigs in the region vary both between and within each country [9, 18]. In Vietnam, the prevalence of taeniasis and cysticercosis in humans ranges from 0.2 - 12% and 1 - 7.2%, respectively; cysticercosis in pigs ranges from 0.03 to 2.17% [10, 19]. The primary risk factors for human cysticercosis and taeniasis infection include consumption of raw or undercook meat, poor sanitation, lack of meat inspection, use of un-treatment human waste for fertilising crops and non-confinement of pigs [10, 20, 21]. The prevalence data, however, should be interpreted with caution because of current limitations in diagnostic tests utilised. For example, Taenia eggs in human faeces are morphologically identical, making molecular-based diagnosis the only logistically feasible option for species-level diagnosis in epidemiological studies. Moreover, although there are several reports on the seroprevalence of T. solium cysticercosis in humans, prevalence data for this zoonosis in swine are sparse. T. asiatica has been reported by Willingham et al. (2003) [20] and Somers et al. (2007) [22] as the most common zoonotic tapeworm in pigs in Vietnam; however, current ELISA-based serological methods cannot distinguish this species from T. hydatigena and T. solium cysticercosis.

Despite the fact that Vietnam is located in a region where Taenia tapeworms are endemic, almost all of the studies cited above have been carried out in the north of Vietnam. Currently, information on this important zoonosis in the Central Highlands of Vietnam is lacking. Forty-seven resource-poor ethnic groups [23] that possess a multitude of risk factors for exposure to this disease live in Central Highland communities, advocating the pursuit of this study.

Similar to Taenia spp., Trichinella spp. is one of the most widespread meat-borne zoonotic parasites, a helminth that not only is a public health risk but also of economic concern to the pig industry and producers in terms of market access [24, 25]. The genus Trichinella is divided into two groups, encapsulated and non-encapsulated forms consisting of nine species including Trichinella spiralis, Trichinella nativa, Trichinella britovi, Trichinella pseudospiralis, Trichinella murrelli, Trichinella nelsoni, Trichinella papuae, Trichinella zimbabwensis and Trichinella patagonienis, and three genotypes of Trichinella T6, T8, and T9 [26]. Trichinella infection in domestic and wild animals does not cause clinical signs [27]; however, the pathology in humans varies ranging from asymptomatic infections to death depending on the stage of infection and number of larvae ingested [28]. In humans symptoms may consist of diarrhoea and abdominal pain in the initial intestinal phase; whereas, high fever, myalgia, periorbital oedema, splinter haemorrhages and rash may occur during the parenteral phase [24, 29]. Heavy infections and complications can lead to cardiac problems and/or central nervous system involvement, and occasionally death [30]. While some developed countries lack human cases, other countries in Eastern Europe and Asia report hundreds or thousands of cases annually [24].

Since the first report of trichinellosis in humans in Vietnam in 1968 [31], there have been five reported outbreaks, involving 114 individuals and eight fatalities [32]. The reported outbreaks took place in the North and North West of the country [33]. In conjunction with outbreaks in humans, 3.2% of wild boars and 20% of domestic pigs in the region were identified seropositive for Trichinella [34, 35]. Although the Central Highlands of Vietnam have similar risk factors that expose the population to trichinellosis, there is a complete lack of data on this zoonosis in this region.

1.1.2 Research hypotheses, goals and objectives

The overall hypothesis of this study is that multiple risk-factors similar to those present in other parts of Vietnam are also conductive to the endemic presence of Taenia and Trichinella-related pork-borne zoonoses in Dak Lak. The goal of this present study is to identify the epidemiological characteristics of Taenia sp. and Trichinella sp. infection in humans

4 and pigs in rural communities of Dak Lak in the Central Highlands of Vietnam. This PhD study was driven by specific objectives as follows:

Objective 1: To estimate the prevalence of T. solium, T. saginata and T. asiatica infection in human stools.

Objective 2: To estimate the seroprevalence of T. solium exposure in humans and to identify risk factors for taeniasis and exposure to T. solium in humans.

Objective 3: To determine the seroprevalence of T. solium in pigs and to identify risk factors for exposure to T. solium in domestic pigs.

Objective 4: To document the spatial distribution of T. solium taeniasis and cysticerci exposed pigs and humans and ascertain the spatial relationship between them.

Objective 5: To determine the prevalence and associated risk factors for Trichinella infection in domestic pigs.

1.1.3 Contribution of this thesis

This cross-sectional study identified and confirmed the sympatric existence of all three human tapeworms, T. solium, T. saginata and T. asiatica in Dak Lak province in the Central Highlands of Vietnam for the first time, through the development and validation of a novel Taq-Man probe-based real-time PCR. This information can be used to inform veterinarians how to better target meat inspection predilection sites for all three species of cysticerci in the carcasses of pigs and cattle. The information can also be used by public health sectors to discourage the dietary practice of ingesting raw or undercooked beef and pork, including pig liver.

The qPCR developed during this study provides highly sensitive and specific detection and discrimination of all three tapeworm species using a high throughput analysis, making it suitable for large-scale community surveys. The qPCR can be translated to both medical and veterinary surveys for species-specific diagnosis of taeniasis in humans and like-wise, cysticercosis lesions in pigs and cattle.

Risk factors for Taenia spp. infection and T. solium exposure risk in humans were quantified and the distribution of T. solium exposure was spatially characterised. Consequently, this study has generated evidence-based suggestions for future implementation of cost-effective and targeted intervention programs. This should allow better allocation of resources for the management of Taenia transmission in the region.

This study represents the first study on Trichinella in domestic pigs deployed in rural communities of Dak Lak.

Overall, the results presented in this thesis provide a quantitative assessment of the prevalence of pork-borne zoonoses in the Central Highland communities of Vietnam and risk factors for pork-borne zoonoses. This information can be used to advocate and propose sustainable strategies for disease control.

1.1.4 Research and thesis structure

This thesis consists of an introduction and literature review, six chapters describing original research findings and a general discussion (Fig. 1.1). The details of each chapter are listed below.

Chapter 1 contains the introduction and literature review and outlines the major themes of the thesis. The chapter provide the background information on the classification, lifecycle, prevalence and distribution, and diagnosis methods of Taenia and Trichinella.

Chapter 2 is a systematic review that provides detail on research regarding taeniasis, cysticercosis and trichinellosis in Vietnam. The hypothesis is that the low apparent prevalence (AP) of T. solium cysticercosis in pigs is due to use of diagnostic tests with relatively low Se (carcass examination) and/ or the internal validity of the study’s findings was poor because of selection bias. This chapter describes the study locations and provided estimates of the true prevalence (TP) of taeniasis and cysticercosis in humans and pigs. The systematic review highlighted the limitation of research on taeniasis and cysticercosis that could affect the TP. The results of this chapter supplement more information to clarify other chapters in this thesis.

Chapter 3 describes the utilisation and comparison of three different methods for diagnosis of taeniasis in human. We hypothesize that the utilization of multiplex real- time PCR for the specific and sensitive detection of all three human species of Taenia in human faeces would provide a superior diagnostic test in terms of accuracy compared to the conventional Kato-Katz thick smear and coproantigen ELISA (cAg-ELISA). Objective 1 of this thesis is addressed in this chapter. This chapter reports the AP of taeniasis in Dak Lak province and the utilisation of a novel self-developed qPCR for identification and differentiation of three tapeworm species in human stools.

Chapter 4 describes the epidemiological characteristics of Taenia spp. infection and T. solium exposure risk in humans in the communities of Dak Lak province. Risk factors for Taenia spp. infection, and T. solium exposure were quantified. Our hypothesis is that multiple risk factors are significantly associated with the prevalence of taeniasis and T. solium exposure in humans in Dak Lak and these reflect those of other communities in the region, namely, the consumption of raw pork or beef dishes and outdoor defaecation. Objective 2 of this thesis was addressed in this chapter.

Similar to Chapter 4, Chapter 5 describes the epidemiological characteristics of T. solium exposure in pigs. We hypothesize that due to the free-roaming practices of pigs and promiscuous defecation practices of humans, the exposure to T. solium cysticercosis in pigs is high in Dak Lak. Objective 3 of this thesis is addressed in this chapter.

Chapter 6 describes the spatial distribution of T. solium exposure in humans and pigs at fine geographic scales, addressing Objective 4. Our research hypothesis is that the distribution of T. solium exposure in humans and pigs is spatially clustered within communities in Dak Lak. The results of this chapter will assist local government authorities to allocate resource to those places where they are most required.

Chapter 7 describes the first attempt to document the prevalence of Trichinella in domestic pigs in the communities of Dak Lak province. In this chapter, it is hypothesized that Trichinella would circulate in i) back-yard pig populations in Dak Lak at a seroprevalence of between 0.1-5% owing to their non-confinement and ii) commercially-raised pigs at a lower prevalence (0.1-0.2%) owing to poor rodent control practices.

Chapter 8 provides a discussion of the key findings from Chapters 2 to 7. In this chapter, all findings are placed in the context of other research. The chapter makes a statement of general conclusions and identifies future research priorities.

Fig. 1.1 Thesis structure

1.2 Human tapeworm Taenia

1.2.1 Morphological classification

Traditionally, humans were known definitive hosts for two species of Taenia, T. solium and T. saginata. These tapeworms are distinguishable by the clear existence of an armed rostellum, presence of three ovarian lobes and the absence of a vaginal sphincter for T. solium [7, 36, 37].

In 1993, Eom and Rim [38] provided detailed morphological description to propose existence of a third human tapeworm, T. asiatica. The tapeworm can be distinguished from T. saginata by the presence of a rostellum on its scolex, a posterior protuberance in gravid proglottids and uterine twigs. Metacestodes of T. asiatica can also be distinguished from T. saginata by the presence of wart-like formations on the surface of the cysticercus, instead of rugae [38]. The number of uterus branches in a gravid proglottid is a criterion to distinguish the three human tapeworms, in which proglottids with 8 (5 – 11) branches in each side from the central uterus is identified as T. solium, while T. saginata and T. asiatica have 23 (4 – 32) and 20 (11 – 32), respectively [5]. The debate on the taxonomic classification of T. asiatica is still not resolved, however, distinct differences in biology and life cycles of the two subspecies, favours T. asiatica being classified as a separate species to T. saginata [39].

1.2.2 Molecular classification

The complete mitochondrial genome of T. asiatica, T. saginata and T. solium, comprising 12 protein-coding, 2 ribosomal ribonucleic acid (rRNA) and 22 transfer ribonucleic acid (tRNA) genes, is 13,703 base pairs (bp), 13,670 bp, and 13,709 bp in length, respectively [40–42]. There is a difference of 4.6% in overall nucleotide composition between T. asiatica and T. saginata, and 11% between T. saginata and T. solium [40]. At the cytochrome oxidase I gene (cox-1), nucleotide variations of 4.6% exist between T. saginata and T. solium. At the cytochrome oxidase-b (cob) gene, a 4.1% nucleotide variation exist between T. saginata and T. asiatica and a nucleotide variation of 12.9% between T. solium and T. asiatica [41, 43]. While a 61.5% sequence similarity of the of 18 kilodalton (kDa) protein-encoding gene exists between T. solium

10 and T. asiatica, this similarity is increased to 5% between T. saginata and T. asiatica [44]. At the 18S rRNDA gene, 99.2% sequence similarity exists between T. asiatica and T. saginata [45]. The aforementioned sequence analyses support the placement of T. asiatica as closely related to T. saginata compared with T. solium [39].

Bowles & McManus (1994) [46] indicated that the 28S ribosomal deoxyribonucleic acid (rDNA) sequence of T. asiatica and T. saginata are identical and it was proposed using allele analysis, that T. asiatica and T. saginata occasionally hybridize supporting them as closely related subspecies [47–49]. However, phylogenetic analysis of the three human Taenia based on their ribosomal internal transcribed space-2 regions (ITS-2) suggest that T. asiatica is distinct species [50]. Jeon et al., (2009) [51] provided similar conclusions to Eom et al. (2002) [50] demonstrating that a DNA sequence divergence of 4.6% between mitochondrial cox-1 and cob gene between T. saginata and T. asiatica is consistent with classification of these closely related tapeworms as two separate species.

1.2.3 Life cycle of human tapeworms

Southeast Asian countries are endemic for all three human Taenia species, T. solium (pork tapeworm), T. saginata (beef tapeworm) and T. asiatica [52–54]. The larval stages of these tapeworms are also referred to as Cysticercus cellulosae, Cysticercus bovis and Cysticercus viscerotropica, respectively [55], but rarely used. Humans act as definitive host for all three-tapeworm species via the ingestion of undercooked meat [7, 10, 46] . Both T. solium and T. asiatica utilise swine as intermediate hosts; whereas, T. saginata utilises cattle as intermediate hosts [8]. While adult Taenia parasitise the intestine of humans resulting in taeniasis, humans may also act as intermediate hosts of T. solium following accidental ingestion of eggs of this parasite. Cysticerci most commonly develop in muscle, subcutaneous tissue, within the eye and in particular the brain, with the latter termed neurocysticercosis (NCC) [56]. The survival time of Taenia eggs in the environment vary and depend on temperature and moisture conditions. In the lower or moderate temperature ranges and in humid climates, the survival time of eggs can last months [57]. For example, T. saginata eggs are able of surviving for up to nine months in summer and autumn, but only up to five months in winter and spring in southeastern Europe (Rukhova, (1963) cited in Murrell et al., 2005) [57].

1.2.3.1 Taenia solium

Ingestion of cysticerci of T. solium through the consumption of under cooked or raw pork results in human taeniasis [53]. The predilection sites of cysticerci in pigs are skeletal muscle, tongue, lungs, heart and diaphragm [58]. Following ingestion of cysts by humans, the scolex of the metacestode evaginates and attaches to the small intestinal wall by two rows of hooks and four suckers. It then proceeds to develop into a fully mature tapeworm that produces and expels gravid proglottids via the stool after approximately eight weeks. Each gravid segment may release approximately 40,000 eggs [36] into the environment. The adult tapeworm can survive approximately 25 years in the human intestine without producing any symptoms [59]. The eggs of mature tapeworms excreted in faeces pollute the environment, contaminating vegetables, and water. Humans may also serve as the intermediate hosts of T. solium. This occurs through accidental ingestion of T. solium eggs through autoinfection caused by poor hygiene, or via consumption of contaminated raw vegetables or water [9].

Once the eggs are ingested by humans or pigs, the oncospheres are released and penetrate through the intestinal wall and travel via the circulatory system to various sites including the subcutaneous tissue, heart, skeletal muscle, brain, and eyes and cause cysticercosis [46], where they may develop into cysticerci [59].

1.2.3.2 Taenia asiatica

Similar to T. solium, T. asiatica use swine and humans as intermediate and definitive hosts, respectively; however differs from T. solium in predilection sites occupied by cysticerci, and has a shorter life cycle than other human tapeworms [5, 39, 60–62]. In pigs, T. asiatica cysticerci are found mainly in the liver. Other organs may also the harbour cysts and these include the lungs, omentum, serosa and mesentery, but not muscle [60, 62–64]. Singh et al. (2016) [65] however, reported a case of T. asiatica cysticerci found in the muscle of a naturally infected pig in India, suggesting that humans are at risk of acquiring infection with T. asiatica taeniasis through the consumption of meat [8]. The life expectancy of the adult worms is reported as approximately 30 years in some cases [12]. Till now, clear evidence to conclude

12 whether humans may also serve as the intermediate hosts of T. asiatica are ambiguous [8, 39].

Preliminary experimental studies exploring the life cycle and biology of T. asiatica in aim of differentiating it from T. saginata found that T. asiatica cysticerci could successfully infect internal organs of both pigs and cattle. Cysticerci were found mainly in the parenchyma and surfaces of the pig’s liver, other predilection sites included the omentum, lung and serosa. In cattle, cysticerci were found in the liver; however, the development of these cysts was less favourable than those in pigs [60]. In contrast, experimental infection of T. saginata demonstrated that T. saginata could only infect muscle and viscera of cattle, but not pigs [60].

1.2.3.3 Taenia saginata

In contrast to the two other human tapeworms, which use swine are their intermediate host, T. saginata utilises cattle as its intermediate hosts. When eggs are ingested by cattle, the embryo hatches via exposure to digestive juices and penetrates the circulatory system, migrates to muscular tissues and develop into cysticerci. The predilection sites of cysticerci include skeletal and cardiac muscle. Humans acquire T. saginata tapeworm by eating infected raw or undercook beef dishes. The first excretion of gravid proglottids takes place after three months. Each mature proglottid consists of 50,000 - 80,000 eggs which are passively or actively excreted via the faeces within proglottids [59]. The eggs are able to persist for several weeks or months in the environment until digested by cattle (Rukhova, (1963) cited in Murrell et al., 2005) [57].

Fig. 1.2 The life cycle of three human Taenia tapeworms (Source: Wandra et al. 2013) 1.2.4 Prevalence and geographic distribution

The prevalence and geographic distribution of Taenia cysticercosis and taeniasis depends largely on standards of hygiene and sanitation, socio-economic standards, dietary culture and pig husbandry practices (FAO/WHO, 2014). T. solium and T. saginata are distributed worldwide; however, mainly in developing countries of Latin America, Africa and Asia, while the distribution of T. asiatica is restricted to Asia [17, 39, 66]. In 1947, Stoll estimated that in total approximately 41.5 million individuals were T. saginata and T. solium tapeworm carriers at globally [67]. This figure increased to 50 million people and 50,000 deaths due to T. solium annually in1992 [68]. Approximately 300,000 individuals were estimated to be infected with T. solium globally between 2010 and 2015, resulting in over 28,100 deaths [4].

Taenia solium

Taenia solium is distributed globally, however the parasite, has been controlled effectively in most countries of Europe, North America, Australia and New Zealand [56]. Infection with T. solium commonly occurs in resource-poor communities where poor sanitation, non-confinement of pigs, outdoor human defaecation and the consumption of raw/undercooked pork are abundantly practiced [69].

In Africa, the seroprevalence of human cysticercosis has been reported to range widely from 0.7 to 45% [70]. Coral-Almeida et al. (2015) [71] performed a systematic review and estimated the prevalence of circulating antigens of T. solium as 7.3% and antibodies at over 17% for the period ranging 1988 to 2014. In addition, the number of T. solium tapeworm carriers in Africa was reported to vary from zero to 14% depending on the studied area [71]. In Tanzania, Mwanjali et al. (2013) [70] reported the seroprevalence of T. solium cysticercosis in humans as high as 45%. In other parts of the continent, lower figures of between 0.7 to 7% in central Africa have been reported, with sporadic cases confirmed in Ghana, Burkina Faso, Ivory Coast, and Senegal [72]. Porcine cysticercosis reported by method of meat inspection ranged from 0.6 to 39% in central and western Africa [72], whereas, in the east of Africa a seroprevalence of between a 8.5% to 35% was recorded for porcine cysticercosis [73–75]. In South Africa, the seroprevalence of porcine cysticercosis was reported to range from 14 to 34% [76].

The prevalence of taeniasis and porcine and human cysticercosis varies greatly within and between countries of Asia and Latin America. It is estimated that a population of 400,000 humans across 15 Latin American countries were infected with T. solium cysticercosis in 2008 [77]. In Mexico, the prevalence of taeniasis, porcine and human cysticercosis was reported at 0.2 - 2.4%, 2.5 - 35% and 3.7 - 12.2%, respectively [78]. Bolivia recorded the highest seroprevalence of porcine and human cysticercosis at roughly 39% and 23%, respectively, while in Guatemala 9% of pigs and 13.5% of humans were found seropositive [78]. On a review carried out by Braae et al. 2017 [79] within the period 1986 to 2017, T. solium was present in 16 countries of central America and the Caribbean basin and the existence of porcine cysticercosis was confirmed in 11 administrative regions across six countries of Colombia, Guatemala, Honduras, Mexico, Nicaragua and Venezuela.

T. solium has been reported to be present in at least 15 Asian countries [53]. High rates of T. solium infection was recoded in Nepal in which 32.5% of infected pigs were detected using the lingual palpation method and 10 - 50% of Nepalese people were

15 reported to suffer from taeniasis, but this figure could also include T. saginata cases. Other countries include Indonesia, China and India, where the prevalence of taeniasis and cysticercosis range from 0.06 to 23% and 0.02 to 9.3%, respectively [80]. T. solium cysticercosis in humans and pigs in South East Asian countries of Vietnam, Lao PDR, Thailand, Malaysia, the Philippines, Cambodia, Indonesia and Myanmar vary between and within countries depending on the dietary, culture, religious and economic status. Among these countries, a hyperendemic focus of T. solium was found in West Papua, Indonesia, in which 77% of 35 pigs slaughtered were found to be positive to T. solium cysticercosis (Maitindom (2008) cited in Willingham III et al., 2010) [9] and 50.6% of 106 [81] individuals in Irian Jaya and 22 - 29% of 2,931 people in the Central Highlands of Papua [82] were reported exposed to T. solium antigens in 2007. Hence, the prevalence of human and porcine cysticercosis in Papua is considered one of the highest in the world. Since then, this figure has dropped significantly, but the prevalence of cysticercosis still remains high at 8.3% in humans and 19% in pigs, recorded in 2012 [83]. In 2011, an investigation carried out in four provinces of Lao PDR showed that over 46% people were infected with taeniasis and roughly 67% villagers with T. solium cysticercosis [84].

In contrast to Asia, Sub Saharan Africa and Latin America, human T. solium infection is rare in European countries, but still considered a public health concern due to returned overseas travellers and migrant activity [85–87]. According to the review by Fabiani & Bruschi (2013) [85], 176 cases of NCC were reported in 17 countries within Europe between 1976 to 2011, for which the majority were sourced to returning overseas travellers or migrants from endemic regions. In a systematic review paper, Zammarchi et al. (2013) [88] reported that between 1990 and 2011, of 846 cysticercosis-positive individuals, 522 were identified as autochthonously infected and 68 individuals were (assumed) T. solium tapeworm carriers, however the species of Taenia were not confirmed.

Taenia asiatica

T. asiatica is endemic and restricted to Asia. Following its first discovery and documentation in Taiwan [5] the parasite is now recognised to be present in almost all

Asian countries including Korea [89], Japan [90], Indonesia [91, 92], The Philippines [61], Vietnam [22], China [93], Thailand [94, 95], Myanmar [96], Nepal [97] and northern India [65]. The presence of T. asiatica is associated with the consumption of raw or undercooked visceral organs of pork. Fan (1988, 1989) suspected 17% of indigenous Taiwanese and 7% of Koreans were infected with T. asiatica instead of T. saginata, as previously presumed since consumption of raw or undercooked visceral organs of pork was associated with infection, as opposed to beef [5, 89]. In 1991, Zarlenga et al. [98, 99] confirmed the presence of T. asiatica in Taiwan and Korea using molecular methods. Since then, numerous studies using molecular methods indicated the existence of the parasite in many Asian countries. According to Jeon et al. (2008) [100] 51 out of 68 samples collected in Taiwan between 1953 and 2005 were re- identified as T. asiatica instead of T. saginata. In Indonesia, Thailand and the Philippines, Fan and his colleagues suspected the presence T. asiatica based on morphological criteria of the adult [61, 92, 101]. Subsequent molecular analysis classified six of 240 samples as T. asiatica. In Indonesia, approximately 23% of pigs in Bali harboured the cysticerci of T. asiatica in pork liver and 2.5% of individuals were shown to harbour T. asiatica taeniasis [83, 102]. Among 24 samples collected from 2002 to 2005 in Kanchanaburi province of Thailand, six were identified T. asiatica and the parasite shown to occur sympatrically with T. saginata and T. solium [103]. Research in China identified the presence of T. asiatica in four southern provinces [50]. Yamasaki (2013) [90] reported sporadic cases of T. asiatica taeniasis in local residents and immigrants in Japan. In Vietnam, T. asiatica was reported for the first time in 2001 from a 14 year old patient in Ha Noi [20]. Through identification of 65 proglottids collected in 14 provinces of Vietnam’s north between 2002 and 2003, Somers et al. (2007) [22] demonstrated the sympatric existence of all three tapeworm species in this region, in which T. asiatica accounted for the highest prevalence of proglottids at 55.4%, followed by T. saginata and T. solium. The presence of T. asiatica in Cambodia, Lao PDR and Malaysia remains uncertain [39, 84, 104].

Taenia saginata

The beef tapeworm, T. saginata, is known as a globally distributed parasite with a more ubiquitous distribution than other two human tapeworms, T. solium and T. asiatica [56].

The tapeworm is reported to be present in countries that T. solium was effectively controlled including Europe [87, 105], New Zealand [106] and Australia [107] although at significantly lower prevalence than in Asia, Latin America and Africa [108, 109]. The prevalence of T. saginata taeniasis in humans varies from over 10% in highly endemic zones, for example in Ethiopia, Kenya, and Democratic Republic of Congo to under 0.1% in zones where the parasite is rare or non-existent [57]. Central and East Africa represent highly endemic regions; other regions within Europe, South East Asia, India, Japan and South America report a moderate prevalence [108, 109].

The prevalence of bovine cysticercosis differs significantly, ranging from 0.03% in North America and Europe to as high as 10 - 80% in Africa and Latin America [109]. Based on carcass inspection data, the proportion of cattle infected with T. saginata cysticercosis ranges between 0.01 and 6.8% within European countries [110]. In France in 2010 T. saginata cysticercosis was recorded at 1.23% [111], but the TP was estimated between three and ten times higher [112] using immunodiagnostic methods, owing to the low Se of organoleptic inspection methods. In Belgium, the prevalence of bovine cysticercosis was estimated at around 0.22% based on meat inspection, however this figure increased to a seroprevalence of 36% using a combination of techniques (dissection, Ag-ELISA and Ab-ELISA) [113]. In Asia, the data is disparate between countries and highly dependent on their economic standing. In high-income economies in East Asia, the prevalence of T. saginata cysticercosis is negligible (0.001% in Japan; four cases reported in Korea in 2013 [114]). In contrast prevalence of up to 43% in the Philippines and of between 2.2-43% have been reported in Indonesia [102, 115, 116]. T. saginata cysticercosis is reported extremely low or scarce in India perhaps owing to limited beef consumption by Hindus that comprise the majority of the population [117]. A slaughterhouse based survey in Brazil between 2010 and 2015 indicated the prevalence of bovine cysticercosis at 0.6% [118]; USA, Australia, South America and some Western Pacific countries maintains a low prevalence of taeniasis of less than 0.5% [36].

1.2.5 Health and economic impact of taeniasis and cysticercosis

Taeniasis and cysticercosis in humans and pigs have a negative impact not only on human health but also the economy. In humans, costs are associated with loss of working days, treatment and implementation of control programs. In pigs, costs are associated with implementing compliant housing and management practices to minimise risk of transmission to domestic animals and due to organ/carcase condemnation at slaughter. [119].

In Eastern Cape province of South Africa, the agricultural sector alone was estimated to have lost US$ 5 million in 2004, with the overall figure lost varying from US$ 18.6 million to 34.2 million [120]. A study in West Cameroon estimated the total annual cost due to T. solium cysticercosis was over EUR 10 million. These included direct losses associated with hospital costs and indirect losses associated with lost productivity. Human cysticercosis accounted for 95.3% of the losses and 4.7% was attributed to losses within the pig husbandry sector [15]. In Tanzania, T. solium was estimated to cost the agricultural sector approximately US$ 3 million owing to porcine cysticercosis and US$ 5 million for the treatment of NCC in humans [121]. In the United States of America, over 18,500 NCC cases associated with hospital charges of over US$ 908 million were reported between the period 2003 to 2012 [118]. In India, the average cost per case of NCC-related epilepsy amounted to US$ 344 per year (equivalent to 88% of average income per capita), and the total cost for the estimated 5 million people affected was equivalent to 0.5% of the gross national product [122]. Singh and colleagues estimated that Indian spent over Rupees 12 million for treatment associated NCC in 2011 [123]. A patient in Mexico was estimated to endure costs up to US$ 2,500 for hospitalization and surgery for cysticercosis treatment [124]. Treatment and working days lost due to taeniasis was estimated to cost approximately 32.6 million annually in the mountainous areas in Taiwan, Cheju Island in Korea and Samosir Island in Indonesia [125].

Within the meat industry, South America loses US$ 43 million annually owing to bovine and porcine cysticercosis [126]. In four provinces in Laos between US $55,000 to 96,000 is estimated lost assuming a carcass condemnation rate of 20% [16]. A

19 calculation based on published data of Abdussalam (1975) and Schenone (1975) by Murrell (1991) [127] reveals an annual combined cost caused by bovine and porcine cysticercosis in Latin American countries of US $428 million. In Africa, bovine cysticercosis was estimated to cost the continent US $1.8 billion, in which four and two million were estimated for Kenya and Botswana, respectively. In England an estimated £100 per carcass condemned, or £4.0 million annually is attributed to cysticercosis (Gracey et al. 1999 cited in Abuseir et al., 2006) [128].

Not only do taeniasis and cysticercosis result in significant losses in term of economic impact, but these diseases also result in human morbidity and suffering. Individuals with taeniasis may endure symptoms of abdominal distension, anal pruritus and disorders of digestion, wakefulness, or hypotension [10]. According to Komba et al. (2013) [73], Trung et al. (2013) [21] and Praet et al. (2009) [15], human cysticercosis causes headache, loss of memory, seizures, paralysis and muscular pain. In 2010, the WHO estimated that T. solium cysticercosis caused 2.78 million disability-adjusted life years (DALYs) lost, that 370,000 individuals were infected and over 28,000 deaths occurred globally [4]. In India, NCC resulted a total of 2.10 million DALYs per year [123]. DALYs associated with NCC in Mexico was estimated at 0.25 per 1000 personal-years in 90% of the population with NCC [129].

1.2.6 Diagnosis of cysticercosis in animals

1.2.6.1 Post-mortem carcass inspection

Carcass inspection for cysticerci is based on organoleptic inspection at slaughter and usually involves partial incisions and examination of organs at predilection sites such as the neck, shoulder, masseters, tongue, diaphragm, heart, omentum, serosa and liver [58]. Post-mortem inspection is considered a gold standard for the diagnosis of porcine cysticercosis in pigs [130]; however, the efficiency of the method relies not only on the skill of inspectors, but also on the cyst burden [57]. Although the method is highly specific (100%), the Se is notoriously poor and as low as 5.8 to 22% [130, 131]. The method is applied for the ante-mortem detection of T. solium in pigs in the majority of developing countries owing to its negligible cost and practicality.

1.2.6.2 Ante-mortem examination

Ante-mortem diagnosis includes ocular inspection and tongue palpation. The method is routinely performed in pigs. The ocular inspection is aimed to visualise metacestodes in sub-conjunctival or periocular regions. The method has been reported to have a low Se. Ante-mortem tongue palpation is a simple, low cost method of screening for foci or T. solium cysticerci in pigs. Although the method is highly specific (100%) [130, 132], it is limited by its low Se of between 50 - 70% for cyst burdens of greater than 100 [58, 133, 134]. It may not be possible to detect cysts in pigs harbouring fewer than 100 cysts [135]. Using a Bayesian approach, Dorny et al. (2004) [130] estimated the Se of tongue palpation to be even lower (21%) for greater than 100 cyst. In addition, tongue examination is time consuming and the inspector risks getting bitten [58, 130, 132].

1.2.6.3 Antibody-ELISA

Antibody-ELISA detect immunoglobulin (Ig) G against cysticerci circulating in pigs with T. solium cysticercosis with high Se, but is limited by cross-reactivity if the animal is infected with closely related species of Taenia and highly dependent on the type of antigen used [136]. Antigens included in Ab-ELISA format include solubilised metacestode and scolex, cyst fluid [137] and excretory/secretory (ES) products [138, 139] derived from Taenia spp. The Se and Sp of the Ab-ELISA for the detection of T. solium in free roaming pigs using ES antigen is 92% and 100%, respectively [138]; using Taenia crassiceps metacestode fluid, the Se of the Ab-ELISA assay increases significantly to between 96% and 100% [140, 141], however this results in lowered Sp and cross reactions with sera from pigs infected with other parasites including Ascaris suum, T. hydatigena, Echinococcus granulosus, Fasiolopsis buski [139, 141–144]. The utilisation of crude antigen also results in cross-reactions in pigs infected with T. hydatigena cysticercosis [136]. The Sp of Ab-ELISA has been improved by using purified antigens. Ito et al. (1999) [145] used T. solium antigen glycoprotein fragments from cyst fluid and intact cysts, purified by isoelectric-focusing electrophoresis and to detect all eight pigs naturally infected with T. solium cysticercosis. Ab-ELISA using isoelectric focusing- purified glycoproteins was demonstrated not to cross react with T. hydatigena cysticerci antigens in pigs with a Se and Sp similar to those of western blot (WB) [146]. During the

21 last 10 years, a newly developed generation of recombinant or synthetic antigens derived from the 8 kDa family [147] and 50 kDa proteins [148] has been cloned including rGP13, sTs14, sTs18, rGP21, rT24H, rGP42, rGP50 [147, 148, 157–159, 149–156]. Of these antigens, the sTs18, rT24H and rGP50 were trailed and evaluated in different ELISA platforms including Fast-ELISA and Quick-ELISA for the detection of porcine cysticercosis and showed good performance with a Se and Sp of between from 74% to 97%, and 90% to 96%, respectively [149, 150]. Currently, the ability of these Ab-ELISAs to cross-react with pigs infected with T. asiatica or T. hydatigena remains uncertain. It is assumed that antibody detection methods for porcine cysticercosis are not specific for T. asiatica and cross-reactivity of T. asiatica was observed even with the most specific T. solium serodiagnostic test [54].

1.2.6.4 Antigen-ELISA

The Ab-ELISA, though a reliable tool for serological detection of cysticercosis, has its limitations. These includes cases in which pigs have passive antibody transfer via colostrum [160] or have developed persistent antibody following temporary exposure to the parasite, which does not necessarily equate to current infection [161, 162]. Consequently, the Ab-ELISA assay results in falsely elevated seroprevalence levels (higher than TP). There have been several Ag-ELISAs developed for the detection of T. solium cysticercosis using monoclonal antibodies to overcome the limitations of Ab- ELISA [162–165]. The monoclonal antibodies detect antigenic determinants of glycoprotein antigens on the surface of and in the secretion of cysticerci [162].

Harrison et al. (1989) [162] introduced mouse monoclonal antibody (MoAb) namely HP10, with the purpose of detecting antibody against T. saginata in cattle as well as T. solium in humans, produced by using glycoprotein extracted from T. saginata metacestodes in immunised mice. The application of MoAbs in detecting antigen of T. saginata in serum of cattle is applicable to targeting T. solium antigen in humans and pigs [162, 163]. Another MoAbs called B158C11 and B60H8, originally developed by Brandt et al. (1992) [163] to detect T. saginata cysticerci in cattle and modified by Van Kerckhoven et al. (1998) [165] and Dorny et al. (2000) [112], was also utilised to detect Taenia cysticerci in pigs and humans. A sero-epidemiological survey of porcine

22 cysticercosis in West Cameron using MoAbs B158C11 and B60H8 showed that that 11% of pigs were positive for T. solium by Ag-ELISA, while 21.8% were positive by Ab-ELISA [133]. The Ag-ELISA has been applied widely for T. solium cysticercosis sero-surveys in pigs and humans [70, 73–75, 133, 164, 166–168]. These Ag-ELISAs are noteworthy owing to low levels of cross-reaction when applied to both human and pig sera infected with a wide range of non-taeniid helminth parasites [162, 166]. However, the Ag-ELISA does cross react with animals infected with T. hydatigena [136], and may thus have limited use in pigs in regions where this tapeworm is endemic.

1.2.6.5 Enzyme-linked Immunoelectrotransfer Blot

EITB is the most specific assay for the detection of antibody against T. solium cysticerci [136, 169] and used widely for the purpose of diagnosing exposure to T. solium cysticerci in humans and pigs. A handful of innovative antigen production methods have been applied to diagnose T. solium cysticercosis in humans and pigs ranging from using crude [144] to purified, recombined and synthetic antigen such as Rec- Ag1V1/Ag2 [170, 171], TsRS1, Ts14 and Ts18, rT24H and rGP50 [147, 148, 152, 158, 172]. The TsRS1, Ts14 and Ts18, rT24H and rGP50 were synthesized, and characterised as reliable diagnostic antigens, which are closely related to the family of the 8 kDa protein [147, 148, 152]. The purified glycoprotein antigens not only offer Se and the Sp of 99% for detection of cysticercosis in humans and pigs but also do not depend strictly on expensive apparatus and provide capability of transferring easily between laboratories [158].

Pathak et al. (1994) [144] used crude-somatic antigens extracted from T. solium cysts of pigs in India for specific detection of exposure to cysticerci. The Se and Sp of this assay was 90% and 100%, respectively, based on the detection of four polypeptides of sizes (8, 11, 16 and 23 kDa). There was a lack of cross-reactivity when this antigen was appraised with pigs infected with Trichinella spiralis, Echinocosus granulosus and Fasciolosis buski but the study did not assess the potential for cross-reactivity with T. hydatigena. Sato et al. (2006) [171] found two genotypes of T. solium (African/American and Asian) generating different immuno-responsive bands against glycoprotein antigens in sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). This provided the

23 potential ability to differentiate between species of cestodes including human tapeworms and T. hydatigena.

By utilising glycoproteins, which are purified by lentil lectin-purified chromatography, Tsang et al. (1989) [169] developed a highly specific EITB assay for T. solium cysticercosis in humans with a reported Se of 98%; however, this was reduced to 28% [173] when a single cyst was present in the brain. In 1998, Sciutto et al. [134] found that in regions where pigs have low T. solium cyst burdens, both EITB and ELISA were limited by their Se. In the past, EITB assays have been limited by the requirement for specialised equipment not always available in developing countries and the generation of sufficient antigen from fresh viable cysts [174–176]. To overcome this, recombinant/synthesis antigens, including TsRS1, Ts14 and Ts18, rT24H and rGP50, have been developed [158].

1.2.7 Diagnosis of cysticercosis in humans

1.2.7.1 Imaging techniques

Imaging techniques utilised for cysticercosis diagnosis in humans currently include radiology, ultrasound, computed tomography (CT), and magnetic resonance imaging (MRI). Images of parasite residing in human’s organs can be visualized and identified [177, 178].

There are four transforming stages of tapeworm once the egg is accidently ingested by intermediate hosts. Oncospheres hatch and once they invade the brain parenchyma, muscle and other tissues [177], they develop through four stages: vesicular, colloidal, granular-nodular, and calcified forms [179]. Hence, depending on each phase the pathological finding may differ. The most sensitive and specific imaging techniques are the CT scan and MRI which offer the ability of identifying all four phases. The Se and Sp of the CT scan provides over 95% Se for the diagnosis of NCC; however, it has lower Se for detecting ventricular or cisternal forms [180]. While the CT scan is the preferred method for detection of cysticerci at the stage of calcification, the MRI is best for detecting non-calcified cysticerci [177], but its high cost and scarce availability form drawbacks of the technology [180]. Radiography shows it usefulness in determining the

24 size, location and number of cysticerci [180]; meanwhile, ultrasound is useful for the diagnosis of extraneural cysticercosis [181, 182].

1.2.7.2 Antibody-ELISA

Individuals with T. solium cysticerci infection stimulate a predominantly IgG response; with some individuals retaining IgM, IgA or IgE class antibody responses [177, 183]. The Ab-ELISA aims to detect these immunoglobulin classes. As with diagnosis in swine, the efficiency of antibody ELISA for the detection of T. solium cysticercosis in humans is dependent on the quality of antigens that are used for the assay [184]. A major advantage of diagnosing T. solium cysticercosis in humans is the inability of other closely related Taenia metacestodes to infect humans (with exception of T. multiceps), and this obviates the problems of cross-reactivity observed in pigs. Antigens derived from cyst fluid have demonstrated more specific detection than other antigens such as antigens derived from the whole parasite and scolex [185–187]. ES antigens extracted from Taenia cysticerci cultured in artificial broth have also been utilised in Ab-ELISA; however, cross-react to E. granulosus hydatids may occur [139, 141, 188]. Purified glycoprotein either from cyst fluid or intact cysts using isoelectric-focusing electrophoresis offers 100% Se and Sp for the confirmation of T. solium NCC in humans. Three major antigenic bands ranging from 10 to 26 kDa corresponding to the glycoprotein components of Taenia cyst fluid or intact cysts are observed to be highly specific in humans infected with T. solium cysticerci [189]. As discussed before, limitations of EITB make the technique challenging and costly to apply in developing countries but to overcome this, recombinant or synthetic antigens can be applied in ELISA assays, which offer the advantage of high Sp and Se, ease of reproduction between laboratories coupled with minimal cost and need for equipment. The common recombinant antigens used for ELISA assay in confirmation of NCC are a component of lentil lectin-bound glycoproteins [147, 148, 152, 158, 170, 172]. The recombinant or synthetic antigens TSRS1, sTs14, sTs18, rT24H and rGP50, components of lentil lectin-bound glycoproteins, have displayed a Se and Sp in the vicinity of 77 to 97% and from 85 to 99%, respectively [151, 154, 156]. Ito et al. (1998) [189] used T. solium antigen glycoprotein fragments from Taenia cysts fluid and intact cysts purified by isoelectric-focusing electrophoresis for detecting NCC in humans with a Sp and Se of 100%.

1.2.7.3 Antigen-ELISA

A positive Ab-ELISA does not necessarily represent an existing infection with viable cysticerci in the host since antibody detection ELISAs may detect antibodies of T. solium in cases where unviable cysts exist or when antibodies are stimulated during transient infection or continue to circulate following elimination of cysts as a response to drugs or immune clearance [160, 162, 190, 191]. Approximately 10% of the population in endemic communities possess circulating antibodies against T. solium metacestodes, resulting in an overestimate of TP [192]. The antigen detection ELISA can be performed using serum or cerebrospinal fluid in both humans and animals [193], in which cerebrospinal fluid offers a better performance in comparison to that of serum as cysticerci normally localizes in brain [136]. Harrison et al. (1989) [162] introduced a monoclonal antibody (HP10) immunized to mouse that can target glycoproteins of T. saginata circulating in serum of humans with T. solium. The Se and Sp of the ELISA assay when utilising the HP10 antibody to confirm 46 cases of NCC was 85% and 94%, respectively [190]. A similar Se of 85% and Sp of 92% was found in another study [194]. Other monoclonal antibodies namely 158C11 and 60H8 were developed to detect ES antigens from T. saginata cysticerci by Brandt et al. (1992) [163] and modified by Van Kerckhoven et al. (1998) [165] and Dorny et al. (2000) [112] and is currently being deployed in epidemiology studies of human cysticercosis [166, 195].

1.2.7.4 Enzyme-Linked Immunoelectrotransfer Blot

EITB is considered the reference standard assay for the serological diagnosis of cysticercosis in humans with a reported Se of 98% and Sp of 100% [169]; however, the Se and Sp may drop dramatically in cases with single viable cyst, calcified cysts or degeneration cysts [115, 158]. The immunoblot is used to detect antibodies to T. solium cysticerci in humans and pigs with the utilisation of purified glycoproteins from raw cysticerci extracted through chromatography with lentil-lectin [196]. The utilisation of lentil-lectin purified glycoprotein (LLGP) from raw cysticerci raises some drawbacks depending on the source of pigs infected used for generation of the T. solium antigens and relies on expensive apparatus and expertise. Hence, the utilisation of recombinant or synthetic antigens derived from the seven diagnostic LLGP including GP50, GP42-39,

GP24, GP21, GP18, GP14, and GP13 [158] was used to overcome the aforementioned drawbacks. The LLGP-EITB provides a Se of 95% and Sp of 100% when using serum or cerebrospinal fluid samples from people with multiple intracranial lesions [158] and demonstrates no cross reactivity to serum from people with hydatid disease (Echinococcus granulosus and E. multilocularis) and other heterologous infections [169]. When compared with CT and/or MRI, the EITB demonstrated a high Se (94%) in confirmation of NCC from patients with two or more lesions; but dropped to 28% in cases with one lesion in the brain [173].

Other antigens used for the EITB assay include purified glycoproteins from T. solium cysticerci using preparative isoelectric-focusing electrophoresis (pH 9.2-9.6) [189]. The antigen was introduced for the EITB assay in 1998 by Ito and his colleagues [189]. The purified glycoproteins, containing three major bands ranging from 10 kDa to 26 kDa, offer the Sp and Se of 100% in the detection of human NCC. Furthermore, these glycoproteins can be applied to ELISA format. Similarly, Yang et al. (1998) [197] demonstrated the 10 kDa native antigen, which is purified form T. solium cyst fluid, offered 90% Sp and 85% Se; whereas the recombinant form of these glycoprotein provided a Se and Sp of 97% and 98% respectively [198].

1.2.8 Diagnosis of taeniasis

1.2.8.1 Faecal examination

Faecal examination for the detection of Taenia carriers includes the utilisation of microscopy based examination and self-observation of exiting proglottids [199]. The expulsion of proglottids is commonly observed in T. saginata and T. asiatica carriers in which the tapeworms frequently shed gravid proglottids, whereas, the proglottids of T. solium are expelled passively and infrequently. T. saginata/T. asiatica tapeworm carriers are consequently aware their infection, but not this is not necessarily the case for T. solium carriers. Questioning the history of proglottid expulsion can hence be used as a complementary tool for the diagnosis of taeniasis [200]. The feasibility of this detection method was confirmed for all taeniid infections, but highly reliable for identification of carriers of T. saginata and T. asiatica [201, 202]. The reliability of the anamnesis of proglottid expulsion differs widely between reports and as well as for

27 species of Taenia infection. For example, the accuracy of the anamnesis method for T. saginata infection was 100%, whilst only 67% for T. solium in Karangasem villages of Bali (Swastika et al., 2017) [201] and under 50% tapeworm carriers were identified based on self-report of proglottid shedding in Honduras, but up to 80% in Sichuan, China [202, 203].

Microscopy-based examination can be used to detect Taenia eggs from stool samples, namely KK thick smears, formalin-ether sedimentation and faecal flotation techniques; however, these method have a low Se owing to the sporadic excretion of taeniid eggs as well as the non-homogenous distribution of eggs in faeces [204–206]. According to Hall et al. (1981) [207], Deplazes et al. (1991) [208] and Allan et al. (1996) [205], ether sedimentation and formalin-ether concentration techniques are able to detect between 62 - 68% of T. saginata and 38% of T. solium taeniasis infections. To increase Se, repeated microscopic examination may be required [207]. Another drawback of the technique is that it is not possible to differentiate the species of Taenia via the morphology of their egg [209], consequently the Sp of microscopic examination is poor [210]. Microscopy- based examination is however useful in remote communities for screening Taenia infection in the general population [201]. Morphological identification of gravid proglottid and scolex using light microscopy is possible for Taenia species differentiation, however this method is reported to be of low Sp for T. saginata and T. asiatica and relies on the successful collection of the proglottid and scolex [211].

1.2.8.2 Polymerase Chain Reactions

Several molecular-based assays have been utilised to diagnose and differentiate species of Taenia in humans and their metacestode stages in animals. Human taeniasis and cysticercosis in humans and pigs can be diagnosed by applying PCR to extracted DNA from i) expelled proglottids from treated carriers [50, 212, 213]; ii) tapeworm eggs from faecal samples or iii) cysticerci from humans and animals with cysticercosis [214]. To date, PCR assays include the use of qPCR using sequence-specific DNA probes [215– 219], singleplex [220] and multiplex conventional PCR [46, 214, 221], PCR-Restriction Fragment Length Polymorphism (RFLP) [211, 213, 222], DNA sequencing of Taenia- specific PCR products [104], and pyrosequencing assays [209]. PCR assays have

28 primarily targeted the Cox-1, Cob and Nicotine Adenine Dinucleotide Dehydrogenase (NAHD) regions of mitochondrial DNA and/or the ITS-1/ITS-2, 5.8S and 28S regions of rDNA of Taenia spp.

PCR has primarily been reserved for research purposes due to the expense involved [223, 224]. Ramahefarisoa et al. (2010) [225] demonstrated that nested PCR assay based on primary primers introduced by Theis et al. (1996) [226] targeting the Cox-1 gene provided a high Sp of 100% but a significantly low Se of up to 65% for the detection of taeniid eggs in faeces. Using a Bayesian approach, Praet et al. (2013) [227] estimated PCR to have a Sp of 99% and Se of 82.7% compared to coprology and coproantigen ELISA (cAg-ELISA). Thus, PCR assays were demonstrated as useful for confirmation and for species-level identification of eggs present in stool of coproantigen ELISA-positive individuals [223]. The multiplex PCR assay developed by Jeon, Chai, et al. (2009) [51] can detect a minimum of 17 eggs per a gram of faeces. A nested PCR developed by Mayta et al. (2008) [220] for specific diagnosis of T. solium taeniasis was proven 100% specific and capable of detecting a minimum of 10 T. solium eggs per 250 mg faeces or a target genomic DNA concentration of 100 fg for Taenia.

For the purpose of species-specific differentiation of the three human tapeworms, Yamasaki et al. (2004) [214] designed a multiplex PCR targeting the Cox-1 region of Taenia species. The assay successfully detected as little as 20 T. saginata eggs/gram of stool. Jeon et al. (2009) [51] developed a multiplex-PCR with primers, Ta4978F, Ts5058F, Tso7421F and Rev7951 targeting valine tRNA and NADH dehydrogenase subunit 1 and 2 genes. The assay was capable of detecting as low as 10 eggs of Taenia per gram of faeces.

PCRs are considered superior tools for the detection of Taenia eggs in faecal samples owing to their high diagnostic Se and the ability to distinguish Taenia species. Currently, the only conventional multiplex-PCR [51, 214] and PCR-RFLP [22] able to differentiate all human Taenia tapeworms is unsuited for large-scale community surveys owing to labour intensive process of restriction digest and gel electrophoresis. A better suited real-time Taq-Man probe-based PCR [116] has been developed but it is only capable of discriminating T. solium and T. saginata.

1.2.8.3 Immunological technique

Immunological techniques can be also utilised for diagnosis of human taeniasis [196]. The diagnosis can be performed in ELISA or EITB platforms. The ELISA utilises polyclonal antibodies to detect the presence of antigen excreted from adult Taenia in faecal samples (cAg-ELISA) [227]. The cAg-ELISA has been proven to offer a higher Se in comparison to coprological detection of taeniid eggs in faeces with one study demonstrating a 2.6 times higher Se than microscopy for the detection of human taeniasis [205]. The assay was not able to discriminate between the three human tapeworm species but was useful for the evaluation of mass treatment efficacy [228] or the effect health education programs [228] and epidemiological surveys [229]. In 2009, a modified assay boasting a highly species-specific cAg-ELISA was developed to detect T. solium taeniasis with a test Sp of 100% and Se of 96.4% [230]. The Se was maintained using a capture antibody of rabbit IgG against T. solium adult whole worm somatic extracts, whereas species specificity was achieved by utilisation of an enzyme-conjugated rabbit IgG against T. solium adult ES antigens. In comparison to microscopy with a Se of 38%, the coproantigen testing offered the ability of detecting 98% of diagnosed cases, and was useful for large-scaled epidemiological surveillance for taeniasis in large field studies [231].

Besides utilisation of ELISAs, EITB assays to detect antibody circulating in serum of tapeworm carriers has been developed and gained the potential to overcome the limitations of coproantigen assay, offering species-specific detection of taeniasis and avoiding the potential biohazard of collecting and handling faecal material [232]. In 1999, Wilkins et al. [233] developed an immunoblot assay using ES antigens derived from T. solium to detect T. solium carriers by testing sera from individuals with taeniasis and cysticercosis. The antibodies of individuals with taeniasis responded to antigens with molecular weights of 32.7 kDa and 37.8 kDa; the assay did not cross-react to antibodies from cysticercosis positive persons and individuals with Echinococcus, Hymenolepis nana, Ascaris, or those with filariasis and schistosomiasis. The EITB assay was also 100% specific but was not evaluated for patients positive for T. asiatica infection; hence the potential for false positive results due to T. asiatica may exist [54]. In 2009, Jeon & Eom (2009) [234] designed a EITB assay to evaluate the

30 immunological response of antibodies from T. asiatica taeniasis to antigens delivered from T. asiatica, T. saginata and T. solium. The result showed T. asiatica taeniasis antibody responded specifically to antigens of T. asiatica for a protein with a molecular weight of 21.5 kDa, offering 100% Sp for the diagnosis of T. asiatica infection. These mentioned studies offer a potential tool for species-specific diagnosis of taeniasis using a combination of species specific antigens for all three species of Taenia, or diagnosis of cysticercosis and taeniasis at once when used in combination with other immunodiagnostic test for cysticercosis [232].

1.3 Trichinella spp.

1.3.1 Taxonomy

Trichinella belongs to the Phylum Nematoda, Class Adenophorea, Order Trichurida and Family Trichinellidae [235]. All species of Trichinella are classified as zoonotic, however, they differ considerably in their biological characteristics such as host- specificity, epidemiological cycles (sylvatic versus domestic) and ability of larvae to remain viable in carcases at various temperatures.

After the first discovery of T. spiralis, new species of Trichinella have been added corresponding to research in different geographies [26]. In 1972, T. nativa, T. pseudospiralis, and T. nelsoni were discovered by Garkavi (1972), and in 1992 a new species, T. britovi, and three new genotypes of Trichinella, namely T5 (T. murelli), T6 and T8 were described (Pozio et al., 1992). Between 1999 and 2008 Trichinella T9 (Nagano et al., 1999), T. papuae (Pozio et al., 1999), T. zimbabwensis (Pozio et al., 2002), and T. patagoniensis (Krivokapich et al., 2008) were described (cited in Pozio & Zarlenga, 2013) [26]. All descriptions were provided on genetic basis as larvae are indistinguishable.

Currently, the nine species and three genotypes of Trichinella are divided into two groups, encapsulated and non-encapsulated [24, 27, 236]. The species belonging to the encapsulated group include T. spiralis, T. nativa, T. britovi, T. murrelli, T. nelsoni, T. patagoniensis and genotypes Trichinella T6, T8, T9 circulate only in mammals. The non-encapsulated species, T. pseudospiralis, infect birds and mammals, while T. papuae

31 and T. zimbabwensis parasitise mammals and reptiles [27]. Because of the diversity and divergence between Trichinella species and genotypes, it is predicted that there potentially may be more species described in future [237].

1.3.2 Life cycle

Trichinella spp. have a simple, direct life cycle which is typically maintained in nature by definitive hosts through predation and scavenging through the consumption of raw meat containing first stage larvae (L1) of Trichinella [24]. Once ingested, the larvae are released, moult four times and transform into adult worms in the intestine [238, 239]. The adult worms are viviparous and following mating produce L1 larvae. Following this, the adults are expelled from the small intestine by host immune system [240, 241] within four to six weeks. The newborn L1 larvae enter circulation and encyst (encapsulated) or become dormant (non-encapsulated) in the hosts’ muscle and survive for many years waiting to be indigested by the next host; the infective larvae of T. spiralis can survive in humans for up to 40 years and over 20 years in the muscle of polar bears [238].

Viability of larvae in muscle following death of the animal also varies according to the species of Trichinella and host tissue. For example, 50 - 60% of larvae of T. spiralis L1 survive for 27 months in the muscle of bears stored at -6.50C to -200C, but only for four months in wolverine muscle [242]. On the other hand, the non-encapsulated species, T. papuae, has a survival time of nine days in pig carcasses and two days in carcasses at ambient temperatures associated with the tropics, of 350C [243]. T. nativa, a species distributed in North Pole and sub-arctic region, has a high tolerance to extremely cold temperatures, with a survival time of seven days at 210C [244], and up to four years at - 180C [245] in carnivore meat.

There are two epidemiological cycles of Trichinella, domestic and sylvatic. The domestic cycle occurs between domestic animals, particularly domestic pigs and rodents. Of all the species and genotypes, T. spiralis is the most widely distributed and most suitably adapted to domestic pigs, and therefore the most common cause of human trichinellosis around the world [246]; however, other species such as T. nativa, T.

32 murelli, T. nelson, T. pseudospiralis and T. papuae in infected meat from wild and game animals are also infectious to humans [247].

Fig. 1.3 Trichinella life cycle (Source: DPDx-Laboratory Identification of Parasitic Diseases of Public Health Concern, CDC) [248])

1.3.3 Clinical signs in humans

The symptoms of trichinellosis in humans differ according to stages of invasion, migration and encystment [249]. Symptoms of nausea and diarrhoea can be observed during invasion of larvae into the circulatory system via the intestinal wall. Subsequent larval migration results in pyrexia and petechia, nailbed haemorrhages and periorbital oedema, as a consequence of vasculitis [250]. The final stage, encystment in striated muscle cells, produces signs of myalgia owing to myositis [247]. The severity of

33 clinical signs is directly proportional to the larval burden ingested. Heavy infections and complications can lead to cardiac problems and/or central nervous system involvement, and occasionally death. In contrast to human trichinellosis, Trichinella infection of domestic and wild animals is generally asymptomatic [24].

1.3.4 Prevalence and geographic distribution

With the exception of Antarctica, Trichinella has a cosmopolitan distribution [27] and is known to infect more than 150 animal species, including humans [251] making it an important food-borne zoonosis. The parasite is the cause of human trichinellosis which has been documented in 55 countries around the world [252].

Of the twelve known taxa, T. spiralis and T. pseudospiralis are spread globally as they are propagated by humans through domestic porcine migration and via the migration of birds, respectively [3]; whereas, other taxa are almost always confined in certain geographical regions owing to their sylvatic hosts and associated cycles and climactic specificity.

Table 1.1 Distribution of Trichinella taxa

Trichinella taxa Geographical distribution T. spiralis (T1) Globally T. nativa (T2) North Pole and sub-arctic region T. britovi (T3) Europe, western Asia, North and West Africa T. pseudospiralis (T4) Globally T. murrelli (T5) Southern Canada, the US, and northern Mexico T. nelsoni (T7) Eastern and southern Africa Trichinella T6 North Pole, and sub-Arctic region of North America Trichinella T8 Southwest Africa Trichinella T9 Japan T. papuae (T10) Southeast Asia and Pacific regions incl. northern Australia T. zimbabwensis (T11) South Africa T. patagonienis (T12) South America

Adjusted from Feidas et al. (2014) [253]; Pozio (2013) [254]; Pozio & Murrell (2006) [235]; Pozio & Rosa (2009) [27] and Pozio & Zarlenga (2013) [26]

In Africa, Trichinella spp. is present in 12 [255] or 13 [252] countries. T. zimbabwensis, T. britovi, T. nelsoni, T. spiralis and Trichinella T8 have been recorded in wild animals

34 such as wild boars, jackals, hyenas, rats, crocodiles, warthogs, lions, etc. and domestic pigs [252, 256]. In humans, several outbreaks were reported in Congo DR, Egypt, Algeria, Kenya and Senegal in either local resident or tourists [238].

Seven countries and territories in America have confirmed the existence of Trichinella in domestic and wild animals, and several outbreaks in humans have been reported (ITRC, 2017). In Argentina, three species, T. spiralis, T. patagoniensis and T. pseudospiralis are recorded [257, 258]. The latest outbreak in humans was in 2010 and affected 64 individuals caused by T. spiralis [259]. Similarly, T. spiralis has resulted in trichinellosis in 631 Chileans and resulted in the death of four people (Pozio, 2013). In Mexico 2.36% of women in rural areas were found seropositive to Trichinella spp. antibodies [260]. Among 113 indigenous people living in southern Saskatchewan (Canada), the percentage of individuals exposed to Trichinella spp. was 2.7% [261]. In the United States between 1975 and 2002, 1,680 people were infected with Trichinella owing to consumption of primarily raw pork products. The latest report on trichinellosis in humans occurred in Alaska between 2016 and 2017 with two outbreaks involving 10 people after the consumption of raw/undercooked walrus meat [262]

In Europe, the distribution of T. spiralis is characterised with sporadic foci in Germany, Spain and eastern countries of Bulgaria, Croatia, Lithuania, Poland, Romania and Serbia and within restricted foci in Austria, Czech Republic, Estonia, Finland, Hungary, Ireland, Latvia and the Slovak Republic [263]. Human outbreaks of trichinellosis have continued to be reported sporadically in other parts of Europe. During the 1980’s, the incidence of trichinellosis was 0.55 per 100,000 inhabitants in Belarus [264]. In Poland the incidence has decline considerably in the last decade, to only 0.06/100,000 in 2011 [265]. In Bosnia-Herzegovina over 51 outbreaks of trichinellosis involving 775 people were recorded between 1996 and 2006, and in France, 20 outbreaks of trichinellosis were recorded between 1975 and 2006 [266, 267]. Although the incidence of trichinellosis in humans in Europe has declined considerably in the last three decades, the parasite cycles endemically within sylvatic hosts. The sylvatic species of Trichinella continue to provide a food safety concern and risk of spill-over to the domestic cycle, with a high percentage of foxes positive for T. britovi, T. pseudospiralis, T. spiralis and T. nativa in recent years in Poland, Germany and Romania [268]. The seroprevalence of

Trichinella spp. in wild boar was reported to range from 17 to 42% in Estonia, meanwhile there was no evidence of the parasite circulating in domestic pigs [269]. In Italy, T. spiralis was believed not to be circulating in domestic or sylvatic cycles but in 2016 this zoonotic pathogen was detected for the first time in a red fox [270].

According to the ITRC, (2017) [255], 20 countries in Asia and three in the Pacific region have recorded the presence of Trichinella spp. either in domestic or sylvatic cycles. Of these countries, the parasite has been reported to prevail in domestic animals and wildlife mainly in Southeast Asia, China and India. In India, sporadic cases of trichinellosis in humans and Trichinella spp. in rats as well as within pigs within domestic cycles have been reported from 1977 to 2010 [271–274]. In China, outbreaks of trichinellosis with cases of death are recorded yearly; those outbreaks occur mainly in ten provinces where T. spiralis is known to circulate in 50% of the pig populations tested [252]. Currently, the species of T. spiralis, T. nativa, T. pseudospiralis, and T. papuae have been confirmed to be distributed in China [275] in 15 different animals [276]. In Southeast China the seroprevalence of Trichinella spp. in man was 3.81% among 946 people tested [277].

In Southeast Asia, the parasite is present in seven out of 11 countries (ITRC, 2017) [255]. Several outbreaks of Trichinella spp. each year are recorded, especially in remote areas of Lao PDR, North Thailand, Cambodia, and Northwest Vietnam [33, 278–280]. The disease occurs due to the consumption of raw and/or improperly cooked meat dishes, in particular wild boar, especially during events held during New Years festivities, funerals or weddings [34]. Currently, three species of Trichinella are recorded to exist in the region namely, T. psuedospiralis, T. papuae and T. spiralis [34, 281]. Free roaming pigs contribute to the maintenance of domestic cycle of the parasite in the region [34]. In northern Lao PDR, the seroprevalence of human trichinellosis by antigen-ELISA was extremely high with over 19% reported exposed, and the prevalence of T. spiralis in pigs was 2.1% [282].

1.3.5 Impact of Trichinella and trichinellosis

The World Health Organization (WHO), and The Food and Agriculture Organization (FAO) ranked T. spiralis and Trichinella spp. as one of the top 24-foodborne parasites

36 with the greatest impact globally based on a multi-criteria ranking system in term of socio-economic impact, trade and public health [3].

Trichinella was estimated to affect around 11 million people globally in 1998 [283], and was implicated in almost 66,000 infections and 42 deaths during the period 1986 to 2009 [3]. Trichinella was estimated to be the cause of approximately 4,500 illness and four deaths with a total of 550 DALYs lost [4] in 2010 and 76 DALYs lost per billion persons yearly globally [284]. Trichinellosis outbreaks in humans take place annually with an estimated 2,500 infections [285], particularly in the areas where inhabitants have the habit of consuming raw and undercooked pork or game meat [33, 252, 278]. Dupouy-Camet (2000) [283] estimated around 3000 individuals were infected owing to consumption of horse meat between 1975 and 2000 in France and Italy. In China, an estimated 40 million people are at risk of Trichinella infection [251] and approximately 2.2 million CNY is spent for control and inspection of Trichinella spp. each year [286]. Trichinella infection results in a reduction of profit for pig farmers and meat manufacturers because the parasite exists permanently in the sylvatic food chain, posing a threat of spill-over into the domestic chain. In the European Union, an estimation of hundreds of millions of Euros each year is spent towards its surveillance and control in the form of meat inspection and piggery certification [238] and aproximately US$ 570 million for Trichinella inspection for pork alone [24]. Similarly, in the United States an effort to eradicate Trichinella spp. between 1968 and 1980, cost the meat industry an estimated 15% of revenues but with a resulting benefit of around US$ 400 million each year since, owing to increase in exports, consumer confidence, consumption and reduction of inspection costs (US Department of Energy, 1983 cited in Murrell, (1991) [127].

1.3.6 Diagnosis of Trichinella in pork

1.3.6.1 Artificial digestion methods

The artificial digestion (AD) method aids release of Trichinella larvae in muscle samples from infected animals through the digestive action of hydrochloric acid and pepsin solution [287]. In the past, there were six methods of digestion approved by the EU regulation but to date only four methods have been accepted, including the magnetic stirrer, stomacher sedimentation technique, Trichomatic 35' automated digestion method

37 and stomacher filtration technique [288–290]. Of these methods, the magnetic stirrer method is considering the gold standard [291] with a detection rate of 78.47% when the digested duration is for 2 hours [292]; however, the method is still limited in term of low Se in compared to the sequential sieving step [293]. Depending on different kinds of muscle types and animals, the digestibility of AD tests is dissimilar [287]. For example, the duration for digesting diaphragm of pigs takes maximum 30 min at 440C to 460C degree; nevertheless, it may take two to three times longer to digest the masseter of horse [291]. Furthermore, the efficiency of AD depends widely on different conditions of ambient temperature and sizes of mesh. According to Li et al. (2010) [294] the rate of larvae recovery was 98.42%, 90.59%, and 81.63% when performing the method with sieves of 425 µm 250 µm, and 180 µm at 40C degree, but at 400C degrees lower larval recovery rates of 95.12%, 78.60%, and 44.16% respectively, were achieved. AD method shows the capability of detecting both, encapsulated and non- encapsulated forms of Trichinella spp. [287]. The AD test for 1 gram of samples is only reliable when there are more than three larvae in a gram of tissue, so that the test might only be an optimal monitoring measure to protect humans from Trichinella spp. [295] when larvae burdens are greater than 1 larvae per gram of meat. The process of AD method is labour intensive, highly reliant on the examiner’s skills and the transportation or storage of muscle samples prior to testing may be logistically difficult in remote regions [296].

1.3.6.2 Molecular diagnosis

Molecular tools have been proven to be invaluable for taxonomic classification and differentiation of Trichinella larvae [297]. Isoenzyme electrophoresis, conventional, real-time and multiplex PCRs and loop-mediated isothermal amplification (LAMP) have been utilised for the characterisation of Trichinella larvae [298–301]. Most of these molecular assays have been based on the detection, amplification and characterisation of the small and large subunit rRNA, 5S rRNA, ITS region, the expansion segment V (ESV) of rRNA and Cox-1 of mDNA genes of Trichinella [299– 301].

For the diagnosis of Trichinella larvae, Nagano et al. (1999) [300] developed a PCR- RFLP based on the Cox-1 gene with the purpose of distinguish nine genotypes; Genotype-specific primers, based upon the ESV, have been also been developed and provide a potential method for differentiation of Trichinella species and genotypes [302, 303]. Both assays, however, had the limitation of being unable to differentiate T. nativa from Trichinella T6 as well as T. britovi from Trichinella T8. To overcome this challenge, a multiplex-PCR that utilised ESV, ITS-1, and ITS-2 primers was developed by Zarlenga et al. (1999) [301] resulting in a discrimination of nine genotypes namely, Trichinella T6, T. spiralis, T. nativa, T. pseudospiralis, T. murrelli, T. nelsoni, T. zimbabwensis, T. papuae, and T. britovi. The multiplex-PCR generates similar amplicon size products for T. britovi, Trichinella T8 and T9, as well as between T. nativa and Trichinella T12 [27]. These genotypes can then be differentiated from each other by PCR-RFLP, with the exception of Trichinella T8 from T9. Uppermost, the multiplex-PCR introduced by Zarlenga et al. (1999) [301] and recommended by the Community Reference Laboratory [304] currently is the most utilised tool for differentiating Trichinella species and genotypes.

Among conventional PCR and LAMP assays, real-time PCRs have proven the most sensitive in comparison with the others in the ability of detecting Trichinella spp. at a threshold of 100 fg/µL and 0.001 larvae [299]. Similarly, when spiked with larvae of T. pseudospiralis and T. papuae, the assay was capable of detecting as low as 0.1 larvae per gram using 10 gram meat [298, 305]. Because of its high Se, real-time PCR is a potentially suitable tool for monitoring Trichinella spp. in the meat industry, however the considerably high costs associated with DNA extraction [305] limit its translation to commercial enterprise.

1.3.6.3 Enzyme-linked Immunosorbent Assay

Serological diagnosis methods such as ELISAs are not recommended by the International Commission on Trichinellosis as replaceable tools for the AD method for the surveillance of Trichinella spp. in individual carcasses for the purpose of meat inspection [290], however, the assays are acceptable for epidemiological surveillance in areas endemic for Trichinella spp. [287, 306].

Of the serodiagnostic tools, ELISAs using ES antigens show advantage in term of speed, cost efficiency, and Se-Sp balance [306]. The Se and Sp of the assay of 97.1 – 97.8% and 99.5 – 99.8%, respectively has a superior rate of detection of infection compared to the AD test [295]. In other reports, the Se and Sp of ELISA assays using ES antigens for detecting Trichinella spp. in pigs range from 93.1 to 99.2% and 90.6 to 99.4% respectively; the threshold detection of the assay has been reported at 1 larva per 100g of tissue [306]. Commercial ELISA using ES antigen are utilised widely; however, results have to be interpreted with caution due to the high frequently of false positive results [306]. However, overall, the sensitive and Sp of commercial ELISA was appraised to be better than or equal to the in-house ELISA [295, 296, 307] with a limit of detection of larvae at 0.025 larvae per gram of tissue.

1.3.6.4 Enzyme-linked Electroimmunotransfer Blot or Western Blot

ELISA shows its advantages in epidemiological surveillance; nevertheless, cross-reactions with other nematode parasites [308] results in a high rate of false-positive results [309]. Therefore, the results should be confirmed by WB. Gomez-Morales et al. (2014) [310] singled out that ELISA assays exacerbate positive results compared to WB; 315 out of 1,462 wild board meat juice samples were positive with Trichinella via ELISA, while only 32 out of 315 ELISA-positive samples were confirmed by WB. WB provides a Sp of 100% and is useful when applied in series with ELISA (i.e. used to confirm ELISA positive results) [309].

1.4 References

[1] Taylor LH, Latham SM, Woolhouse MEJ. Risk factors for human disease emergence. Philos Trans R Soc B Biol Sci 2001;356:983–9.

[2] Youssef AI, Uga S. Review of parasitic zoonoses in egypt. Trop Med Heal 2014;42:3–14.

[3] FAO/WHO. Multicriterial-based Ranking for Risk Management of Food-borne Parasites. Microbiologycal Risk Assessment Series No 23. vol. 23. Rome: WHO press; 2014.

[4] Torgerson PR, Devleesschauwer B, Praet N, Speybroeck N, Willingham AL, Kasuga F, et al. World Health Organization estimates of the global and regional disease burden of 11 foodborne parasitic diseases, 2010: A data synthesis. PLoS Med 2015;12:e1001920.

[5] Fan PC. Taiwan Taenia and taeniasis. Parasitol Today 1988;4:86–8.

[6] Jeon HK, Eom KS. Molecular approaches to Taenia asiatica. Korean J Parasitol 2013;51:1–8.

[7] Ito A, Nakao M, Wandra T. Human taeniasis and cysticercosis in Asia. Lancet 2003;362:1918–20.

[8] Galán-Puchades MT, Fuentes M V. Updating Taenia asiatica in humans and pigs. Parasitol Res 2016;115:4423–5.

[9] Willingham III AL, Wu H-W, Conlan J, Satrija F. Chapter 9 - Combating Taenia solium Cysticercosis in Southeast Asia: An Opportunity for Improving Human Health and Livestock Production. In: Xiao-Nong Zhou RBRO, Jürg U, editors. Adv. Parasitol., vol. 72, Academic Press; 2010, p. 235–66.

[10] Van De N, Le TH, Lien PTH, Eom KS. Current status of taeniasis and cysticercosis in Vietnam. Korean J Parasitol 2014;52:125–9.

[11] Gonzales I, Rivera JT, Garcia HH, Cysticercosis Working Group in Peru. Pathogenesis of Taenia solium taeniasis and cysticercosis. Parasite Immunol 2016;38:136–46.

[12] Fan PC, Chung WC, Lin CY, Chan CH. Clinical manifestations of taeniasis in Taiwan aborigines. J Helminthol 1992;66:118–23.

[13] Carabin H, Ndimubanzi PC, Budke CM, Nguyen H, Qian Y, Cowan LD, et al. Clinical manifestations associated with neurocysticercosis: a systematic review. PLoS Negl Trop Dis 2011;5:e1152.

[14] Fleury A, Trejo A, Cisneros H, Garcia-Navarrete R, Villalobos N, Hernandez M, et al. Taenia solium: development of an experimental model of porcine neurocysticercosis. PLoS Negl Trop Dis 2015;9:e0003980.

[15] Praet N, Speybroeck N, Manzanedo R, Berkvens D, Nforninwe DN, Zoli A, et al. The disease burden of Taenia solium cysticercosis in Cameroon. PLoS Negl Trop Dis 2009;3:e406.

[16] Choudhury AA, Conlan J V, Racloz VN, Reid SA, Blacksell SD, Fenwick SG, et al. The economic impact of pig-associated parasitic zoonosis in Northern Lao PDR. Ecohealth 2013;10:54–62.

[17] Donadeu M, Lightowlers MW, Fahrion AS, Kessels J, Donadeu M, Lightowlers MW, et al. Taenia solium: WHO endemicity map update Taenia solium: carte d’endémicité actualisée de l’OMS pour 2016 2016;4:595–9.

[18] Wu HW, Ito A, Ai L, Zhou XN, Acosta LP, Lee Willingham A. Cysticercosis/taeniasis endemicity in southeast Asia: Current status and control measures. Acta Trop 2015;165:121–32.

[19] Somers R, Dorny P, Nguyen VK, Dang TCT, Goddeeris B, Craig PS, et al. Taenia solium taeniasis and cysticercosis in three communities in north Vietnam. Trop Med Int Heal 2006;11:65–72.

[20] Willingham AL, Van De N, Doanh NQ, Cong D, Dung T V, Dorny P, et al. Current status of cysticercosis in Vietnam. Southeast Asian J Trop Med Public Health 2003;34:35–50.

[21] Trung DD, Praet N, Cam TDT, Lam BVT, Manh HN, Gabriël S, et al. Assessing the burden of human cysticercosis in Vietnam. Trop Med Int Heal 2013;18:352–6.

[22] Somers R, Dorny P, Geysen D, Nguyen LA, Thach DC, Vercruysse J, et al. Human tapeworms in north Vietnam. Trans R Soc Trop Med Hyg 2007;101:275–7.

[23] Dak Lak Provincial People’s Committee. An overview about Dak Lak province 2014. http://daklak.gov.vn/portal/page/portal/daklak/tong-quan- daklak?p_page_id=13371692&pers_id=&folder_id=15937106&item_id=159376 43&p_details=1.

[24] Gottstein B, Pozio E, Nockler K. Epidemiology, diagnosis, treatment, and control of trichinellosis. Clin Microbiol Rev 2009;22:127–45.

[25] Li LZ, Wang ZQ, Jiang P, Zhang X, Ren HJ, Cui J. Molecular Identification of a Trichinella Isolate from a Naturally Infected Pig in Tibet, China. Korean J Parasitol 2011;49:381–4.

[26] Pozio E, Zarlenga DS. New pieces of the Trichinella puzzle. Int J Parasitol 2013;43:983–97.

[27] Pozio E, Rosa G la. Trichinella. In: Liu D, editor. Mol. Detect. Foodborne Pathog. 1st ed., CRC Press; 2009, p. 853.

[28] Hueffer K, Parkinson AJ, Gerlach R, Berner J. Zoonotic infections in Alaska: Disease prevalence, potential impact of climate change and recommended actions for earlier disease detection, research, prevention and control. Int J Circumpolar Health 2013;72:19562.

[29] Clausen MR, Meyer CN, Krantz T, Moser C, Gomme G, Kayser L, et al. Trichinella infection and clinical disease. QJM 1996;89:631–6.

[30] Pozio E, Gomez Morales MA, Dupouy-Camet J. Clinical aspects, diagnosis and treatment of trichinellosis. Expert Rev Anti Infect Ther 2003;1:471–82.

[31] Phan L. Parasite and Veterinary Parasitic Diseases. Hanoi: Hanoi Agriculture Publishing House; 1997.

[32] Ng-Nguyen D, Stevenson MA, Traub RJ. A systematic review of taeniasis, cysticercosis and trichinellosis in Vietnam. Parasit Vectors 2017;10:150.

[33] Van De N, Trung N V, Ha NH, Nga VT, Ha NM, Thuy PT, et al. An outbreak of trichinosis with molecular identification of Trichinella sp. in Vietnam. Korean J Parasitol 2012;50:339–43.

[34] Vu Thi N, Nguyen VD, Praet N, Claes L, Gabriël S, Huyen NT, et al. Trichinella infection in wild boars and synanthropic rats in northwest Vietnam. Vet Parasitol 2014;200:207–11.

[35] Vu Thi N, Dorny P, La Rosa G, To Long T, Nguyen Van C, Pozio E. High prevalence of anti-Trichinella IgG in domestic pigs of the Son La province, Vietnam. Vet Parasitol 2010;168:136–40.

[36] Pawlowski ZS. Taenia solium: Basic Biology and Transmission. In: Singh G, Prabhakar S, editors. Taenia solium Cysticercosis from Basic to Cinical Sci., Wallingford, Oxfordshire, UK: CABI Publishing; 2002, p. 3–11.

[37] Verster A. Redescription of Taenia solium Linnaeus, 1758 and Taenia saginata Goeze, 1782. Zeitschrift Für Parasitenkd 1967;29:313–28.

[38] Eom KS, Rim HJ. Morphologic descriptions of Taenia asiatica sp. n. Korean J Parasitol 1993;31:1–6.

[39] Ale A, Victor B, Praet N, Gabriël S, Speybroeck N, Dorny P, et al. Epidemiology and genetic diversity of Taenia asiatica: A systematic review. Parasites and Vectors 2014;7:45.

[40] Jeon HK, Kim KH, Eom KS. Complete sequence of the mitochondrial genome of Taenia saginata: Comparison with T. solium and T. asiatica. Parasitol Int 2007;56:243–6.

[41] Jeon HK, Lee KH, Kim KH, Hwang UW, Eom KS. Complete sequence and structure of the mitochondrial genome of the human tapeworm, Taenia asiatica (Platyhelminthes; Cestoda). Parasitology 2005;130:717–26.

[42] Nakao M, Sako Y, Ito A. The mitochondrial genome of the tapeworm Taenia solium: a finding of the abbreviated stop codon U. J Parasitol 2003;89:633–5.

[43] Jeon HK, Eom KS. Taenia asiatica and Taenia saginata: Genetic divergence estimated from their mitochondrial genomes. Exp Parasitol 2006;113:58–61.

[44] Gonzalez LM, Ramiro R, Garcia L, Parkhouse RM, McManus DP, Garate T. Genetic variability of the 18 kDa/HP6 protective antigen in Taenia saginata and Taenia asiatica: implications for vaccine development. Mol Biochem Parasitol 2011;176:131–4.

[45] Yan H, Lou Z, Li L, Ni X, Guo A, Li H, et al. The nuclear 18S ribosomal RNA gene as a source of phylogenetic information in the genus Taenia. Parasitol Res 2013;112:1343–7.

[46] Bowles J, McManus DP. Genetic characterization of the Asian Taenia, a newly described taeniid cestode of humans. Am J Trop Med Hyg 1994;50:33–44.

[47] Okamoto M, Nakao M, Blair D, Anantaphruti MT, Waikagul J, Ito A. Evidence of hybridization between Taenia saginata and Taenia asiatica. Parasitol Int 2010;59:70–4.

[48] Yamane K, Yanagida T, Li T, Chen X, Dekumyoy P, Waikagul J, et al. Genotypic relationships between Taenia saginata, Taenia asiatica and their hybrids. Parasitology 2013;140:1595–601.

[49] Yamane K, Suzuki Y, Tachi E, Li T, Chen X, Nakao M, et al. Recent hybridization between Taenia asiatica and Taenia saginata. Parasitol Int 2012;61:351–5.

[50] Eom KS, Jeon HK, Kong Y, Hwang UW, Yang Y, Li X, et al. Identification of Taenia asiatica in China: molecular, morphological, and epidemiological analysis of a Luzhai Isolate. J Parasitol 2002;88:758–64.

[51] Jeon HK, Chai JY, Kong Y, Waikagul J, Insisiengmay B, Rim HJ, et al. Differential diagnosis of Taenia asiatica using multiplex PCR. Exp Parasitol 2009;121:151–6.

[52] Aung AK, Spelman DW. Taenia solium taeniasis and cysticercosis in southeast Asia. Am J Trop Med Hyg 2016;94:947–54.

[53] Ito A, Wandra T, Li T, Dekumyoy P, Nkouawa A, Okamoto M, et al. The present situation of human taeniases and cysticercosis in Asia. Recent Pat Antiinfect Drug Discov 2015;9:173–85.

[54] Galán-Puchades MT, Fuentes M V. Taenia asiatica: The most neglected human Taenia and the possibility of cysticercosis. Korean J Parasitol 2013;51:51–4.

[55] Eom KS. What is Asian Taenia? Parasitol Int 2006;55:S137–41.

[56] Okello A, Thomas LF. Human taeniasis: Current insights into prevention and management strategies in endemic countries. Risk Manag Healthc Policy 2017;10:107–16.

[57] Murrell KD, Dorny P, Flisser A, Geerts S, Kyvsgaard NC, McManus DP, et al. WHO/FAO/OIE Guidelines for the Surveillance, Prevention and Control of Taeniosis/Cysticercosis. Paris: OIE, WHO and FAO; 2005.

[58] Boa ME, Kassuku AA, Willingham IAL, Keyyu JD, Phiri IK, Nansen P. Distribution and density of cysticerci of Taenia solium by muscle groups and organs in naturally infected local finished pigs in Tanzania. Vet Parasitol 2002;106:155–64.

[59] Flisser A, Correa D, Avilla G, Marvilla P. Biology of Taenia solium, Taenia saginata and Taenia saginata asiatica. In: Murrell KD, editor. WHO/FAO/OIE Guidel. Surveillance, Prev. Control Taeniosis/Cysticercosis, Paris: OIE, WHO and FAO; 2005, p. 1–9.

[60] Eom KS, Rim HJ, Geerts S. Experimental infection of pigs and cattle with eggs of Asian Taenia saginata with special reference to its extrahepatic viscerotropism. Korean J Parasitol 1992;30:269–75.

[61] Fan PC, Lin CY, Chung WC. Experimental infection of Philippine Taenia in domestic animals. Int J Parasitol 1992;22:235–8.

[62] Fan PC, Chung WC, Lin CY, Wu CC. The pig as an intermediate host for Taiwan Taenia infection. J Helminthol 1990;64:223–31.

[63] Fan PC, Soh CT, Kosin E. Pig as a favorable intermediate host of a possible new species of Taenia in Asia. Yonsei Reports Trop Med 1990;21:39–56.

[64] Chung WC, Lin CY, Fan PC. Parasitologists ectopic locations of Taenia saginata asiatica cysticerci in the abdominal cavity of domestic pig and monkey. J Parasitol 1996;82:1032–4.

[65] Singh SK, Prasad KN, Singh AK, Gupta KK, Chauhan RS, Singh A, et al. Identification of species and genetic variation in Taenia isolates from human and swine of North India. Parasitol Res 2016;115:1–5.

[66] Eom KS, Jeon HK, Rim HJ. Geographical distribution of Taenia asiatica and related species. Korean J Parasitol 2009;47:S115–24.

[67] WHO. Research needs in taeniasis cysticercosis (Memorandum). Bull World Health Organ 1976;53:67–73.

[68] International Task Force for Disease Eradication. Update: International Task Force for Disease Eradication, 1992. vol. 41. 1992.

[69] Ngowi H, Ozbolt I, Millogo A, Dermauw V, Somé T, Spicer P, et al. Development of a health education intervention strategy using an implementation research method to control taeniasis and cysticercosis in Burkina Faso. Infect Dis Poverty 2017;6:95.

[70] Mwanjali G, Kihamia C, Kakoko D V, Lekule F, Ngowi H, Johansen M V, et al. Prevalence and risk factors associated with human Taenia solium infections in Mbozi District, Mbeya Region, Tanzania. PLoS NeBioMed Res Int Trop Dis 2013;7:e2102.

[71] Coral-Almeida M, Gabriël S, Abatih EN, Praet N, Benitez W, Dorny P. Taenia solium human cysticercosis: A systematic review of sero-epidemiological data from endemic zones around the world. PLoS Negl Trop Dis 2015;9:e0003919.

[72] Geerts S, Zoli A, Nguekam JP, Brandt J, Dorny P. The taeniasis-cysticercosis complex in West and Central Africa. Southeast Asian J Trop Med Public Health 2004;35:262–5.

[73] Komba EVG, Kimbi EC, Ngowi HA, Kimera SI, Mlangwa JE, Lekule FP, et al. Prevalence of porcine cysticercosis and associated risk factors in smallholder pig production systems in Mbeya region, southern highlands of Tanzania. Vet Parasitol 2013;198:284–91.

[74] Pondja A, Neves L, Mlangwa J, Afonso S, Fafetine J, Willingham AL, et al. Prevalence and risk factors of porcine cysticercosis in Angónia District, Mozambique. PLoS Negl Trop Dis 2010;4:e594.

[75] Waiswa C, Fèvre EM, Nsadha Z, Sikasunge CS, Willingham AL. Porcine cysticercosis in southeast Uganda: seroprevalence in Kamuli and Kaliro districts. J Parasitol Res 2009;2009:375493.

[76] Tsotetsi-Khambule AM, Njiro S, Katsande TC, Thekisoe OMM, Harrison LJS. Sero-prevalence of Taenia spp. infections in cattle and pigs in rural farming communities in Free State and Gauteng provinces, South Africa. Acta Trop 2017;172:91–6.

[77] Hotez PJ, Bottazzi ME, Franco-Paredes C, Ault SK, Periago MR. The neglected tropical diseases of Latin America and the Caribbean: A review of disease burden and distribution and a roadmap for control and elimination. PLoS Negl Trop Dis 2008;2:e300.

[78] Flisser A, Sarti E, Lightowlers M, Schantz P. Neurocysticercosis: Regional status, epidemiology, impact and control measures in the Americas. Acta Trop 2003;87:43–51.

[79] Braae UC, Devleesschauwer B, Sithole F, Wang Z, Willingham AL. Mapping occurrence of Taenia solium taeniosis/cysticercosis and areas at risk of porcine cysticercosis in Central America and the Caribbean basin. Parasit Vectors 2017;10:424.

[80] Rajshekhar V, Joshi DD, Doanh NQ, van De N, Xiaonong Z. Taenia solium taeniosis/cysticercosis in Asia: epidemiology, impact and issues. Acta Trop 2003;87:53–60.

[81] Subahar R, Hamid A, Purba W, Wandra T, Karma C, Sako Y, et al. Taenia solium infection in Irian Jaya (West Papua), Indonesia: a pilot serological survey of human and porcine cysticercosis in Jayawijaya District. Trans R Soc Trop Med Hyg 2001;95:388–90.

[82] Salim L, Ang A, Handali S, Tsang VCW, Cysticercosis Working Grp P. Seroepidemiologic survey of cysticercosis-taeniasis in four Central Highland districts of Papua, Indonesia. Am J Trop Med Hyg 2009;80:384–8.

[83] Wandra T, Ito A, Swastika K, Dharmawan NS, Sako Y, Okamoto M. Taeniases and cysticercosis in Indonesia: past and present situations. Parasitology 2013;140:1608–16.

[84] Okello A, Ash A, Keokhamphet C, Hobbs E, Khamlome B, Dorny P, et al. Investigating a hyper-endemic focus of Taenia solium in northern Lao PDR. Parasit Vectors 2014;7:134.

[85] Fabiani S, Bruschi F. Neurocysticercosis in Europe: Still a public health concern not only for imported cases. Acta Trop 2013;128:18–26.

[86] Laranjo-González M, Devleesschauwer B, Trevisan C, Allepuz A, Sotiraki S, Abraham A, et al. Epidemiology of taeniosis/cysticercosis in Europe, a systematic review: Western Europe. Parasit Vectors 2017;10:349.

[87] Devleesschauwer B, Allepuz A, Dermauw V, Johansen M V., Laranjo-González M, Smit GSA, et al. Taenia solium in Europe: Still endemic? Acta Trop 2017;165:96–9.

[88] Zammarchi L, Strohmeyer M, Bartalesi F, Bruno E, Muñoz J, Buonfrate D, et al. Epidemiology and management of cysticercosis and Taenia solium taeniasis in Europe, systematic review 1990-2011. PLoS One 2013;8:e69537.

[89] Fan PC, Soh CT. The Korean Taenia and taeniasis. Yonsei Reports Trop Med 1989;20:64–7.

[90] Yamasaki H. Current status and perspectives of cysticercosis and taeniasis in Japan. Korean J Parasitol 2013;51:19–29.

[91] Fan PC, Kosman ML, Kosin E, Depary AA, Napitupulu T. Indonesia Taenia and taeniasis. Yonsei Reports Trop Med 1990;21:33–7.

[92] Fan PC, Lin CY, Kosman ML, Kosin E. Experimental infection of Indonesia Taenia (Samosir strain) in domestic animals. Int J Parasitol 1989;19:809–12.

[93] Zhang L, Tao H, Zhang B, Wang H, Wang Y, Li Z, et al. First discovery of Taenia saginata asiatica infection in Yunnan province. Chinese J Parasitol Parasit Dis 1999;17:95–6.

[94] McCleery EJ, Prapas P, Pornsawan P, Sasisophin C, Saruda T, Papaspong J, et al. Taeniasis among refugees living on Thailand-Myanmar boder, 2012. Cent Dis Control Prev 2015;21:1824–6.

[95] Anantaphruti MT, Okamoto M, Yoonuan T, Saguankiat S, Kusolsuk T, Sato M, et al. Molecular and serological survey on taeniasis and cysticercosis in Kanchanaburi Province, Thailand. Parasitol Int 2010;59:326–30.

[96] Fan PC, Lin CY, Chen LM. Experimental infection and morphology of Taenia saginata (Burma strain) in domestic animals. Ann Trop Med Parasitol 1992;86:317–8.

[97] Devleesschauwer B, Aryal A, Joshi DD, Rijal S, Sherchand JB, Praet N, et al. Epidemiology of Taenia solium in Nepal: is it influenced by the social characteristics of the population and the presence of Taenia asiatica? Trop Med Int Heal 2012;17:1019–22.

[98] Zarlenga DS, McManus DP, Fan PC, Cross JH. Characterization and detection of a newly described Asian taeniid using cloned ribosomal DNA fragments and sequence amplification by the polymerase chain reaction. Exp Parasitol 1991;72:174–83.

[99] Zarlenga DS. The differentiation of a newly described Asian taeniid from Taenia saginata using enzymatically amplified non-transcribed ribosomal DNA repeat sequences. Southeast Asian J Trop Med Public Health 1991;22 Suppl:251–5.

[100] Jeon HK, Kim KH, Chai JY, Yang HJ, Rim HJ, Eom KS. Sympatric distribution of three human Taenia tapeworms collected between 1935 and 2005 in Korea. Korean J Parasitol 2008;46:235–41.

[101] Wandra T, Sudewi AA, Swastika IK, Sutisna P, Dharmawan NS, Yulfi H, et al. Taeniasis/cysticercosis in Bali, Indonesia. Southeast Asian J Trop Med Public Heal 2011;42:793–802.

[102] Anantaphruti MT, Yamasaki H, Nakao M, Waikagul J, Watthanakulpanich D, Nuamtanong S, et al. Sympatric Occurrence of Taenia solium, T. saginata, and T. asiatica, Thailand. Emerg Infect Dis 2007;13:1413–6.

[103] Jeon HK, Yong TS, Sohn WM, Chai JY, Hong SJ, Han ET, et al. Molecular identification of Taenia tapeworms by Cox1 gene in Koh Kong, Cambodia. Korean J Parasitol 2011;49:195–7.

[104] Dorny P, Praet N. Taenia saginata in Europe. Vet Parasitol 2007;149:22–4.

[105] New Zealand Public Health Surveillance Report. New Zealand Public Health Surveillance Report. March 2009: Covering October – December 2008. vol. 7. 2009.

[106] Jenkins DJ, Brown GK, Traub RJ. “Cysticercosis storm” in feedlot cattle in north-west New South Wales. Aust Vet J 2013;91:89–93.

[107] Eom KS, Rim HJ. Epidemiological understanding of Taenia tapeworm infections with special reference to Taenia asiatica in Korea. Korean J Parasitol 2001;39:267–83.

[108] Lloyd S. Cysticercosis and Taeniosis Taenia saginata, Taenia solium, and Asian Taenia. In: Palmer SRS, LordSimpson DIH, editors. Zoonoses Biol. Clin. Pract. Public Heal. Control, Oxford: Oxford University Press; 1998, p. 635–49.

[109] European Commission-Health and Consumer Protection Directorate General. Opinion on the control of taeniosis/cysticercosis in man and animals, adopted on 27–28 September 2000. https://ec.europa.eu/food/sites/food/files/safety/docs/sci- com_scv_out36_en.pdf (accessed March 8, 2018).

[110] Dupuy C, Morlot C, Gilot-Fromont E, Mas M, Grandmontagne C, Gilli-Dunoyer P, et al. Prevalence of Taenia saginata cysticercosis in French cattle in 2010. Vet Parasitol 2014;203:65–72.

[111] Dorny P, Vercammen F, Brandt J, Vansteenkiste W, Berkvens D, Geerts S. Sero- epidemiological study of Taenia saginata cysticercosis in Belgian cattle. Vet Parasitol 2000;88:43–9.

[112] Jansen F, Dorny P, Berkvens D, Van Hul A, Van den Broeck N, Makay C, et al. High prevalence of bovine cysticercosis found during evaluation of different post-mortem detection techniques in Belgian slaughterhouses. Vet Parasitol 2017;244:1–6.

[113] Cho J, Jung BK, Lim H, Kim MJ, Yooyen T, Lee D, et al. Four Cases of Taenia saginata Infection with an Analysis of Cox1 Gene. Korean J Parasitol 2014;52:79–83.

[114] Singh G, Kaushal V, Ram S, Kaushal RK, Dhanuka AK, Khurana S. Cysticercus immunoblot assay in patients with single, small enhancing lesions and multilesional neurocysticercosis. J Assoc Physicians India 1999;47:476–9.

[115] Gordon CA, McManus DP, Acosta LP, Olveda RM, Williams GM, Ross AG, et al. Multiplex real-time PCR monitoring of intestinal helminths in humans reveals widespread polyparasitism in northern Samar, the Philippines. Int J Parasitol 2015;45:477–83.

[116] Bhatia BB. Current status of food-borne parasitic zoonoses in India. Southeast Asian J Trop Med Public Health 1991;22:36–41.

[117] O’Neal SE, Flecker RH. Hospitalization Frequency and Charges for Neurocysticercosis, United States, 2003–2012. Emerg Infect Dis J 2015;21:969.

[118] Gabriël S, Dorny P, Mwape KE, Trevisan C, Braae UC, Magnussen P, et al. Control of Taenia solium taeniasis/cysticercosis: The best way forward for sub- Saharan Africa? Acta Trop 2017;165:252–60.

[119] Carabin H, Krecek RC, Cowan LD, Michael L, Foyaca-Sibat H, Nash T, et al. Estimation of the cost of Taenia solium cysticercosis in Eastern Cape Province, South Africa. Trop Med Int Heal 2006;11:906–16.

[120] Trevisan C, Devleesschauwer B, Schmidt V, Winkler AS, Harrison W, Johansen MV. The societal cost of Taenia solium cysticercosis in Tanzania. Acta Trop 2017;165:141–54.

[121] WHO. Atlas: Epilepsy Care in the World. Geneva: WHO press; 2005.

[122] Singh BB, Khatkar MS, Gill JPS, Dhand NK. Estimation of the health and economic burden of neurocysticercosis in India. Acta Trop 2017;165:161–9.

[123] Bhattarai R, Carabin H, Proaño J V., Flores-Rivera J, Corona T, Flisser A, et al. Cost of neurocysticercosis patients treated in two referral hospitals in Mexico City, Mexico. Trop Med Int Heal 2015;20:1108–19.

[124] Fan PC. Annual economic loss caused by Taenia saginata asiatica Taeniasis in East Asia. Parositotogy Today 1997;13:194–6.

[125] Abdussalam M. The problem of taeniasis-cysticercosis. Sci. Publ. - Pan Am. Heal. Organ., Washington, D.C.(USA): PAHO; 1975.

[126] Murrell KD. Economic losses resulting from food-borne parasitic zoonoses. Southeast Asian J Trop Med Public Health 1991;22:377–81.

[127] Abuseir S, Epe C, Schnieder T, Klein G, Kühne M. Visual diagnosis of Taenia saginata cysticercosis during meat inspection: is it unequivocal? Parasitol Res 2006;99:405–9.

[128] Bhattarai R, Budke CM, Carabin H, Proaño J V., Flores-Rivera J, Corona T, et al. Estimating the non-monetary burden of neurocysticercosis in mexico. PLoS Negl Trop Dis 2012;6:e1521.

[129] Dorny P, Phiri IK, Vercruysse J, Gabriel S, Willingham IAL, Brandt J, et al. A Bayesian approach for estimating values for prevalence and diagnostic test characteristics of porcine cysticercosis. Int J Parasitol 2004;34:569–76.

[130] Goussanou JSE, Korsak N, Saegerman C, Youssao AKI, Azagoun E, Farougou S, et al. Assessment of routine inspection method for diagnostic of porcine cysticercosis in South East Benin by using meat inspection records and Ag- ELISA test. Int J Anim Vet Adv 2014;6:80–6.

[131] Gonzalez AE, Cama V, Gilman RH, Tsang VCW, Pilcher JB, Chavera A, et al. Prevalence and comparison of serologic assays, necropsy, and tongue examination for the diagnosis of porcine cysticercosis in Peru. Am J Trop Med Hyg 1990;43:194–9.

[132] Pouedet MSR, Zoli AP, Nguekam, Vondou L, Assana E, Speybroeck N, et al. Epidemiological survey of swine cysticercosis in two rural communities of West- Cameroon. Vet Parasitol 2002;106:45–54.

[133] Sciutto E, Martinez JJ, Villalobos NM, Hernandez M, Jose M V, Beltran C, et al. Limitations of current diagnostic procedures for the diagnosis of Taenia solium cysticercosis in rural pigs. Vet Parasitol 1998;79:299–313.

[134] Nguekam A, Zoli AP, Vondou L, Pouedet SMR, Assana E, Dorny P, et al. Kinetics of circulating antigens in pigs experimentally infected with Taenia solium eggs. Vet Parasitol 2003;111:323–32.

[135] Dorny P, Brandt J, Zoli A, Geerts S. Immunodiagnostic tools for human and porcine cysticercosis. Acta Trop 2003;87:79–86.

[136] Larralde C, Laclette JP, Owen CS, Madrazo I, Sandoval M, Bojalil R, et al. Reliable serology of Taenia solium cysticercosis with antigens from cyst vesicular fluid - ELISA and hemagglutination tests. Am J Trop Med Hyg 1986;35:965–73.

[137] D’Souza PE, Hafeez M. Detection of Taenia solium cysticercosis in pigs by ELISA with an excretory-secretory antigen. Vet Res Commun 1999;23:293–8.

[138] Ko RC, Ng TF. Evaluation of excretory/secretory products of larval Taenia solium as diagnostic antigens for porcine and human cysticercosis. J Helminthol 1998;72:147–54.

[139] Nunes CM, Biondi GF, Heinemann MB, Richtzenhain LJ. Comparative evaluation of an indirect ELISA test for diagnosis of swine cysticercosis employing antigen from Taenia solium and Taenia crassiceps metacestodes. Vet Parasitol 2000;93:135–40.

[140] Pinto PSA, Vaz AJ, Germano PML, Nakamura PM. Performance of the ELISA test for swine cysticercosis using antigens of Taenia solium and Taenia crassiceps cysticerci. Vet Parasitol 2000;88:127–30.

[141] Cheng RWK, Ko RC. Cross-reactions between crude antigens of larval Taenia solium (Cysticercus cellulosae) and other helminths of pigs. Vet Parasitol 1991;39:161–70.

[142] Kumar D, Gaur SNS. Serodiagnosis of porcine cysticercosis by enzyme-linked- immunosorbent-assay (ELISA) using fractionated antigens. Vet Parasitol 1987;24:195–202.

[143] Pathak KML, Allan JC, Ersfeld K, Craig PS. A western blot and ELISA assay for the diagnosis of Taenia solium infection in pigs. Vet Parasitol 1994;53:209–17.

[144] Ito A, Plancarte A, Nakao M, Nakaya K, Ikejima T, Piao ZX, et al. ELISA and immunoblot using purified glycoproteins for serodiagnosis of cysticercosis in pigs naturally infected with Taenia solium. J Helminthol 1999;73:363–5.

[145] Sato MO, Yamasaki H, Sako Y, Nakao M, Nakaya K, Plancarte A, et al. Evaluation of tongue inspection and serology for diagnosis of Taenia solium cysticercosis in swine: usefulness of ELISA using purified glycoproteins and recombinant antigen. Vet Parasitol 2003;111:309–22.

[146] Hancock K, Khan A, Williams FB, Yushak ML, Pattabhi S, Noh J, et al. Characterization of the 8-kilodalton antigens of Taenia solium metacestodes and evaluation of their use in an enzyme-linked immunosorbent assay for serodiagnosis. J Clin Microbiol 2003;41:2577–86.

[147] Hancock K, Pattabhi S, Greene RM, Yushak ML, Williams F, Khan A, et al. Characterization and cloning of GP50, a Taenia solium antigen diagnostic for cysticercosis. Mol Biochem Parasitol 2004;133:115–24.

[148] Handali S, Gonzalez AE, Hancock K, Garcia HH, Roberts JM, Gilma RH, et al. Porcine antibody responses to Taenia solium antigens rGP50 and sTS18var1. Am J Trop Med Hyg 2004;71:322–6.

[149] Handali S, Pattabhi S, Lee Y-M, Silva-Ibanez M, Kovalenko VA, Levin AE, et al. Development and evaluation of porcine cysticercosis QuickELISA in Triturus EIA Analyzer. J Immunoass Immunochem 2009;31:60–70.

[150] Bueno EC, Scheel CM, Vaz AJ, Machado LR, Livramento JA, Takayanagui OM, et al. Application of synthetic 8-kD and recombinant GP50 antigens in the diagnosis of neurocysticercosis by enzyme-linked immunosorbent assay. Am J Trop Med Hyg 2005;72:278–83.

[151] Hancock K, Pattabhi S, Whitfield FW, Yushak ML, Lane WS, Garcia HH, et al. Characterization and cloning of T24, a Taenia solium antigen diagnostic for cysticercosis. Mol Biochem Parasitol 2006;147:109–17.

[152] Handali S, Klarman M, Gaspard AN, Dong XF, LaBorde R, Noh J, et al. Development and evaluation of a magnetic immunochromatographic test to detect Taenia solium, which causes taeniasis and neurocysticercosis in humans. Clin Vaccine Immunol 2010;17:631–7.

[153] Handali S, Klarman M, Gaspard AN, Noh J, Lee YM, Rodriguez S, et al. Multiantigen print immunoassay for comparison of diagnostic antigens for Taenia solium cysticercosis and taeniasis. Clin Vaccine Immunol 2010;17:68–72.

[154] Lee YM, Handali S, Hancock K, Pattabhi S, Kovalenko VA, Levin A, et al. Serologic diagnosis of human Taenia solium cysticercosis by using recombinant and synthetic antigens in QuickELISATM. Am J Trop Med Hyg 2011;84:587–93.

[155] Gilman RH, Tsang VCW, Cysticercosis THE, Group W. Serodiagnosis Of neurocysticercosis using synthetic 8-KD proteins: Comparison of assay formats 2005;73:771–6.

[156] O’Neal SE, Moyano LM, Ayvar V, Rodriguez S, Gavidia C, Wilkins PP, et al. Ring-screening to control endemic transmission of Taenia solium. PLoS Negl Trop Dis 2014;8:e3125.

[157] Noh J, Rodriguez S, Lee YM, Handali S, Gonzalez AE, Gilman RH, et al. Recombinant protein- and synthetic peptide-based immunoblot test for diagnosis of neurocysticercosis. J Clin Microbiol 2014;52:1429–34.

[158] Porphyre V, Betson M, Rabezanahary H, Mboussou Y, Zafindraibe NJ, Rasamoelina-Andriamanivo H, et al. Taenia solium porcine cysticercosis in Madagascar: Comparison of immuno-diagnostic techniques and estimation of the prevalence in pork carcasses traded in Antananarivo city. Vet Parasitol 2016;219:77–83.

[159] Garcia HH, Gonzalez AE, Gilman RH, Palacios LG, Jimenez I, Rodriguez S, et al. Short report: transient antibody response in Taenia solium infection in field conditions-a major contributor to high seroprevalence. Am J Trop Med Hyg 2001;65:31–2.

[160] Garcia HH. Serological diagnosis and follow-up of severe neurocysticercosis using HP10 antigen detection. Nat Clin Pract Neurol 2007;3:488–9.

[161] Harrison LJ, Joshua GWP, Wright SH, Parkhouse RME. Specific detection of circulating surface secreted glycoproteins of viable cysticerci in Taenia saginata cysticercosis. Parasite Immunol 1989;11:351–70.

[162] Brandt JRA, Geerts S, Deken R de, Kumar V, Ceulemans F, Brijs L, et al. A monoclonal antibody-based ELISA for the detection of circulating excretory- secretory antigens in Taenia saginata cysticercosis. Int J Parasitol 1992;22:471–7.

[163] Dorny P, Phiri I, Gabriel S, Speybroeck N, Vercruysse J. A sero-epidemiological study of bovine cysticercosis in Zambia. Vet Parasitol 2002;104:211–5.

[164] Van Kerckhoven I, Vansteenkiste W, Claes M, Geerts S, Brandt J. Improved detection of circulating antigen in cattle infected with Taenia saginata metacestodes. Vet Parasitol 1998;76:269–74.

[165] Erhart A, Dorny P, Van De N, Vien H V, Thach DC, Toan ND, et al. Taenia solium cysticercosis in a village in northern Viet Nam: Sero-prevalence study using an ELISA for detecting circulating antigen. Trans R Soc Trop Med Hyg 2002;96:270–2.

[166] Mohan VR, Tharmalingam J, Muliyil J, Oommen A, Dorny P, Vercruysse J, et al. Prevalence of porcine cysticercosis in Vellore, South India. Trans R Soc Trop Med Hyg 2013;107:62–4.

[167] Ngowi H, Kassuku A, Carabin H, Mlangwa J, Mlozi M, Mbilinyi B, et al. Spatial clustering of porcine cysticercosis in Mbulu district, northern Tanzania. PLoS Negl Trop Dis 2010;4:e652.

[168] Tsang VCW, Brand JA, Boyer AE. An enzyme-linked immunoelectrotransfer blot assay and glycoprotein antigens for diagnosing human cysticercosis (Taenia solium). Oxford Journals 1989;159:50–9.

[169] Sako Y, Nakao M, Ikejima T, Piao XZ, Nakaya K, Ito A. Molecular characterization and diagnostic value of Taenia solium low-molecular-weight antigen genes. J Clin Microbiol 2000;38:4439–44.

[170] Sato MO, Sako Y, Nakao M, Yamasaki H, Nakaya K, Ito A. Evaluation of purified Taenia solium glycoproteins and recombinant antigens in the serologic detection of human and swine cysticercosis. J Infect Dis 2006;194:1783–90.

[171] Greene RM, Hancock K, Wilkins PP, Tsang VC. Taenia solium: Molecular cloning and serologic evaluation of 14- and 18-kDa related, diagnostic antigens. J Parasitol 2000;86:1001–7.

[172] Wilson M, Bryan RT, Fried JA, Ware DA, Schantz PM, Pilcher JB, et al. Clinical evaluation of the cysticercosis enzyme-linked immunoelectrotransfer blot in patients with neurocysticercosis. J Infect Dis 1991;164:1007.

[173] Pal DK, Carpio A, Sander JW. Neurocysticercosis and epilepsy in developing countries. J Neurol Neurosurg Psychiatry 2000;68:137–43.

[174] Rodriguez-Canul R, Allan JC, Dominguez JL, Villegas S, Cob L, Rodriguez RI, et al. Application of an immunoassay to determine risk factors associated with porcine cysticercosis in rural areas of Yucatan, Mexico. Vet Parasitol 1998;79:165–80.

[175] Rodriguez-Canul R, Allan JC, Fletes C, Sutisna IP, Kapti IN, Craig PS. Comparative evaluation of purified Taenia solium glycoproteins and crude metacestode extracts by immunoblotting for the serodiagnosis of human T. solium cysticercosis. Clin Diagn Lab Immunol 1997;4:579–82.

[176] Carpio A, Escobar A, Hauser WA. Cysticercosis and Epilepsy: A Critical Review. Epilepsia 1998;39:1025–40.

[177] Kalra V, Seth R, Mishra D. Extraneural cysticercosis - presenting as painless cervical swellings. J Trop Pediatr 2006;52:141–3.

[178] Escobar A, Weidenheim KM. The Pathology of Neurocysticercosis. Wallingford, UK: CABI Publishing; 2002.

[179] García HH, Gonzalez AE, Evans CAW, Gilman RH. Taenia solium cysticercosis. Lancet 2003;362:547–56.

[180] Jankharia BG, Chavhan GB, Krishnan P, Jankharia B. MRI and ultrasound in solitary muscular and soft tissue cysticercosis. Skeletal Radiol 2005;34:722–6.

[181] Sundaram PM, Jayakumar N, Noronha V. Extraocular muscle cysticercosis - a clinical challenge to the ophthalmologists. Orbit 2004;23:255–62.

[182] Goodman KA, Ballagh SA, Carpio A. Case-control study of seropositivity for cysticercosis in Cuenca, Ecuador. Am J Trop Med Hyg 1999;60:70–4.

[183] Pardini AX, Peralta RH, Vaz AJ, Machado L dos R, Peralta JM. Use of Taenia crassiceps cysticercus antigen preparations for detection of antibodies in cerebrospinal fluid samples from patients with neurocysticercosis (Taenia solium). Clin Diagn Lab Immunol 2002;9:190–3.

[184] Arruda GC, da Silva ADT, Quagliato EMAB, Maretti MA, Rossi CL. Evaluation of Taenia solium and Taenia crassiceps cysticercal antigens for the serodiagnosis of neurocysticercosis. Trop Med Int Heal TM IH 2005;10:1005–12.

[185] Bueno EC, Snege M, Vaz AJ, Leser PG. Serodiagnosis of human cysticercosis by using antigens from vesicular fluid of Taenia crassiceps cysticerci. Clin Diagn Lab Immunol 2001;8:1140–4.

[186] Maria M.I I, Regina Helena S P, José. A L, Sumie H-S, José M P, Adelaide J V. Serodiagnosis of neurocysticercosis in patients with epileptic seizure using ELISA and immunoblot assay. Rev Inst Med Trop Sao Paulo 2006;48:343.

[187] Montenegro T, Gilman RH, Castillo R, Tsang V, Brandt J, Guevara A, et al. The diagnostic importance of species specific and cross-reactive components of Taenia solium, Echinococcus granulosus, and Hymenolepis nana. Rev Inst Med Trop Sao Paulo 1994;36:327–34.

[188] Ito A, Plancarte A, Ma L, Kong Y, Flisser A, Cho SY, et al. Novel antigens for neurocysticercosis: Simple method for preparation and evaluation for serodiagnosis. Am J Trop Med Hyg 1998;59:291–4.

[189] Fleury A, Hernandez M, Avila M, Cardenas G, Bobes RJ, Huerta M, et al. Detection of HP10 antigen in serum for diagnosis and followup of

subarachnoidal and intraventricular human neurocysticercosis. J Neurol Neurosurg Psychiatry 2007;78:970–4.

[190] Garcia HH, Gilman RH, Catacora M, Verastegui M, Gonzalez AE, Tsang VCW. Serologic evolution of neurocysticercosis patients after antiparasitic therapy. J Infect Dis 1997;175:486.

[191] Bern C, Garcia HH, Evans C, Gonzalez AE, Verastegui M, Tsang VCW, et al. Magnitude of the disease burden from neurocysticercosis in a developing country. Clin Infect Dis 1999;29:1203–9.

[192] Garcia HH, Harrison LJ, Parkhouse RM, Montenegro T, Martinez SM, Tsang VC, et al. A specific antigen-detection ELISA for the diagnosis of human neurocysticercosis. The Cysticercosis Working Group in Peru. Trans R Soc Trop Med Hyg 1998;92:411–4.

[193] Garcia HH, Parkhouse RM, Gilman RH, Montenegro T, Bernal T, Martinez SM, et al. Serum antigen detection in the diagnosis, treatment, and follow-up of neurocysticercosis patients. Trans R Soc Trop Med Hyg 2000;94:673–6.

[194] Prado-Jean A, Kanobana K, Druet-Cabanac M, Nsengyiumva G, Dorny P, Preux PM, et al. Combined use of an antigen and antibody detection enzyme-linked immunosorbent assay for cysticercosis as tools in an epidemiological study of epilepsy in Burundi. Trop Med Int Heal 2007;12:895–901.

[195] Deckers N, Dorny P. Immunodiagnosis of Taenia solium taeniosis/cysticercosis. Trends Parasitol 2010;26:137–44.

[196] Yang HJ, Chung JY, Yun DH, Kong Y, Ito A, Ma L, et al. Immunoblot analysis of a 10 kDa antigen in cyst fluid of Taenia solium metacestodes. Parasite Immunol 1998;20:483–8.

[197] Chung JY, Bahk YY, Huh S, Kang SY, Kong Y, Cho SY. A recombinant 10-kDa protein of Taenia solium metacestodes specific to active neurocysticercosis. J Infect Dis 1999;180:1307–15.

[198] Ito A, Li T, Chen X, Long C, Yanagida T, Nakao M, et al. Mini review on chemotherapy of taeniasis and cysticercosis due to Taenia solium in Asia, and a case report with 20 tapeworms in China. Trop Biomed 2013;30:164–73.

[199] Raoul F, Li T, Sako Y, Chen X, Long C, Yanagida T, et al. Advances in diagnosis and spatial analysis of cysticercosis and taeniasis. Parasitology 2013;140:1578–88.

[200] Swastika K, Wandra T, Dharmawan NS, Sudarmaja IM, Saragih JM, Diarthini LPE, et al. Taeniasis caused by Taenia saginata in Gianyar town and Taenia solium in Karangasem villages of Bali, Indonesia, 2011–2016: How to detect tapeworm carriers, anamnesis or microscopy? Acta Trop 2017;174:19–23.

[201] Li T, Ito A, Chen X, Long C, Okamoto M, Raoul F, et al. Usefulness of pumpkin seeds combined with areca nut extract in community-based treatment of human taeniasis in northwest Sichuan Province, China. Acta Trop 2012;124:152–7.

[202] De Kaminsky RG. Albendazole treatment in human taeniasis. Trans R Soc Trop Med Hyg 1991;85:648–50.

[203] Allan JC, Velasquez-Tohom M, Fletes C, Torres-Alvarez R, Lopez-Virula G, Yurrita P, et al. Mass chemotherapy for intestinal Taenia solium infection: effect on prevalence in humans and pigs. Trans R Soc Trop Med Hyg 1997;91:595–8.

[204] Allan JC, Velasquez-Tohom M, Torres-Alvarez R, Yurrita P, Garcia-Noval J. Field trial of the coproantigen-based diagnosis of Taenia solium taeniasis by enzyme-linked immunosorbent assay. Am J Trop Med Hyg 1996;54:352–6.

[205] Garcia HH, Gilman RH, Gonzalez AE, Verastegui M, Rodriguez S, Gavidia C, et al. Hyperendemic human and porcine Taenia solium infection in Peru. Am J Trop Med Hyg 2003;68:268–75.

[206] Hall A, Latham MC, Crompton DW, Stephenson LS. Taenia saginata (Cestoda) in western Kenya: The reliability of faecal examinations in diagnosis. Parasitology 1981;83:91–101.

[207] Deplazes P, Eckert J, Pawlowski ZS, Machowska L, Gottstein B. An enzyme- linked immunosorbent assay for diagnostic detection of Taenia saginata coproantigens in humans. Trans R Soc Trop Med Hyg 1991;85:391–6.

[208] Thanchomnang T, Tantrawatpan C, Intapan PM, Sanpool O, Janwan P, Lulitanond V, et al. Rapid molecular identification of human taeniid cestodes by pyrosequencing approach. PLoS One 2014;9:e100611.

[209] Chandra Parija S, Korol Ponnambath D. Laboratory diagnosis of Taenia asiatica in humans and animals. Trop Parasitol 2013;3:120–4.

[210] Mayta H, Talley A, Gilman RH, Jimenez J, Verastegui M, Ruiz M, et al. Differentiating Taenia solium and Taenia saginata infections by simple hematoxylin-eosin staining and PCR-restriction enzyme analysis. J Clin Microbiol 2000;38:133–7.

[211] González LM, Montero E, Puente S, López-Velez R, Hernández M, Sciutto E, et al. PCR tools for the differential diagnosis of Taenia saginata and Taenia solium taeniasis/cysticercosis from different geographical locations. Diagn Microbiol Infect Dis 2002;42:243–9.

[212] Rodriguez-Hidalgo R, Geysen D, Benítez-Ortiz W, Geerts S, Brandt J. Comparison of conventional techniques to differentiate between Taenia solium and Taenia saginata and an improved polymerase chain reaction-restriction fragment length polymorphism assay using a mitochondrial 12S rDNA fragment. J Parasitol 2002;88:1007–11.

[213] Yamasaki H, Allan JC, Sato MO, Nakao M, Sako Y, Nakaya K, et al. DNA differential diagnosis of taeniasis and cysticercosis by multiplex PCR. J Clin Microbiol 2004;42:548–53.

[214] Chapman A, Vallejo V, Mossie KG, Ortiz D, Agabian N, Flisser A. Isolation and characterization of species-specific DNA probes from Taenia solium and Taenia saginata and their use in an egg detection assay. J Clin Microbiol 1995;33:1283–8.

[215] González LM, Bailo B, Ferrer E, García MD, Harrison LJ, Parkhouse MR, et al. Characterization of the Taenia spp HDP2 sequence and development of a novel PCR-based assay for discrimination of Taenia saginata from Taenia asiatica. Parasit Vectors 2010;3:51–7.

[216] González LM, Montero E, Harrison LJ, Parkhouse RM, Garate T. Differential diagnosis of Taenia saginata and Taenia solium infection by PCR. J Clin Microbiol 2000;38:737–44.

[217] Harrison LJ, Delgado J, Parkhouse RM. Differential diagnosis of Taenia saginata and Taenia solium with DNA probes. Parasitology 1990;100:459–61.

[218] Rishi AK, McManus DP. Molecular cloning of Taenia solium genomic DNA and characterization of taeniid cestodes by DNA analysis. Parasitology 1988;97:161–76.

[219] Mayta H, Gilman RH, Prendergast E, Castillo JP, Tinoco YO, Garcia HH, et al. Nested PCR for specific diagnosis of Taenia solium taeniasis. J Clin Microbiol 2008;46:286–9.

[220] Jeon HK, Kim KH, Huh S, Chai JY, Min DY, Rim HJ, et al. Morphologic and genetic identification of Diphyllobothrium nihonkaiense in Korea. Korean J Parasitol 2009;47:369–75.

[221] McManus DP. Molecular discrimination of taeniid cestodes. Parasitol Int 2006;55:S31–7.

[222] Mathis A, Deplazes P. Copro-DNA tests for diagnosis of animal taeniid cestodes. Parasitol Int 2006;55:S87–90.

[223] Nunes CM, Lima LG, Manoel CS, Pereira RN, Nakano MM, Garcia JF. Taenia saginata: polymerase chain reaction for taeniasis diagnosis in human fecal samples. Exp Parasitol 2003;104:67–9.

[224] Ramahefarisoa RM, Rakotondrazaka M, Jambou R, Carod JF. Comparison of ELISA and PCR assays for the diagnosis of porcine cysticercosis. Vet Parasitol 2010;173:336–9.

[225] Theis JH, Cleary M, Syvanen M, Gilson A, Swift P, Banks J, et al. DNA- confirmed Taenia solium cysticercosis in black bears (Ursus americanus) from California. Am J Trop Med Hyg 1996;55:456–8.

[226] Praet N, Verweij JJ, Mwape KE, Phiri IK, Muma JB, Zulu G, et al. Bayesian modelling to estimate the test characteristics of coprology, coproantigen ELISA and a novel real-time PCR for the diagnosis of taeniasis. Trop Med Int Heal 2013;18:608–14.

[227] Sarti E, Flisser A, Schantz PM, Gleizer M, Loya M, Plancarte A, et al. Development and evaluation of a health education intervention against Taenia solium in a rural community in Mexico. Am J Trop Med Hyg 1997;56:127–32.

[228] Allan JC, Craig PS. Coproantigens in taeniasis and echinococcosis. Parasitol Int 2006;55:S75–80.

[229] Guezala MC, Rodriguez S, Zamora H, Garcia HH, Gonzalez AE, Tembo A, et al. Development of a species-specific coproantigen ELISA for human Taenia solium taeniasis. Am J Trop Med Hyg 2009;81:433–7.

[230] Allan JC, Wilkins PP, Tsang VCW, Craig PS. Immunodiagnostic tools for taeniasis. Acta Trop 2003;87:473–7.

[231] Allan JC, Wilkins PP, Tsang VCW, Craig PS. Review article: Immunodiagnostic tools for taeniasis. Acta Trop 2003;87:87–93.

[232] Wilkins PP, Allan JC, Verastegui M, Acosta M, Eason AG, Hugo Garcia H, et al. Development of a serologic assay to detect Taenia solium taeniasis. Am J Trop Med Hyg 1999;60:199–204.

[233] Jeon HK, Eom KS. Immunoblot patterns of Taenia asiatica taeniasis. Korean J Parasitol 2009;47:73–7.

[234] Pozio E, Murrell KD. Systematics and epidemiology of Trichinella. Adv Parasitol 2006;63:367–439.

[235] Korhonen PK, Pozio E, La Rosa G, Chang BCH, Koehler A V., Hoberg EP, et al. Phylogenomic and biogeographic reconstruction of the Trichinella complex. Nat Commun 2016;7:10513.

[236] Pastusiak K. Methods and tools for parasite differentiation within the genus Trichinella. Wiad Parazytol 2006;52:165–73.

[237] FAO/WHO/OIE. Guidelines for the Surveillance, Management, Prevention and Control of Trichinellosis. Paris: FAO, WHO, OIE; 2007.

[238] La Rosa G, Tasciotti L, Pozio E. DNA repetitive probes for the characterization and identification of Trichinella parasites. Int. Conf. Trichinellosis, Istituto Superiore di Sanità; 1993.

[239] Campbell WC. Trichinella and Trichinosis. 8th ed. New York: Plenum Press; 1983.

[240] Capo V, Despommier DD. Clinical aspects of infection with Trichinella spp. Clin Microbiol Rev 1996;9:47–54.

[241] Worley DE, Seesee FM, Espinosa RH, Sterner MC. Survival of sylvatic Trichinella spiralis isolates in frozen tissue and processed meat products. J Am Vet Med Assoc 1986;189:1047–9.

[242] Owen IL, Reid SA. Survival of Trichinella papuae muscle larvae in a pig carcass maintained under simulated natural conditions in Papua New Guinea. J Helminthol 2007;81:429–32.

[243] Von Koller J, Kapel CM, Enemark HL, Hindsbo O. Infectivity of Trichinella spp. recovered from decaying mouse and fox muscle tissue. Parasite 2001;8:S209-12.

[244] Kapel CM, Pozio E, Sacchi L, Prestrud P. Freeze tolerance, morphology, and RAPD-PCR identification of Trichinella nativa in naturally infected arctic foxes. J Parasitol 1999;85:144–7.

[245] Dupouy-Camet J. Presidential address of ICT12 Conference: ‘‘Trichinella and trichinellosis - A never ending story. Vet Parasitol 2009;159:194–196.

[246] Robertson LJ, Sprong H, Ortega Y, van der Giessen JWB, Fayer R. Response to galán-puchades and fuentes: Taenia asiatica: Neglected - but not forgotten - and almost certainly being quietly globalised. Trends Parasitol 2014;30:56–7.

[247] CDC. Trichinella life cycle. DPDx-Laboratory Identif Parasit Dis Public Heal Concern 2018. http://www.cdc.gov/parasites/trichinellosis/biology.html (accessed March 8, 2018).

[248] Rostami A, Gamble HR, Dupouy-Camet J, Khazan H, Bruschi F. Meat sources of infection for outbreaks of human trichinellosis. Food Microbiol 2017;64:65–71.

[249] Faber M, Schink S, Mayer-Scholl A, Ziesch C, Schonfelder R, Wichmann- Schauer H, et al. Outbreak of trichinellosis due to wild boar meat and evaluation of the effectiveness of post exposure prophylaxis, Germany, 2013. Clin Infect Dis 2015;60:e98–104.

[250] Bai X, Hu X, Liu X, Tang B, Liu M. Current research of trichinellosis in China. Front Microbiol 2017;8:1–7.

[251] Pozio E. World distribution of Trichinella spp. infections in animals and humans. Vet Parasitol 2007;149:3–21.

[252] Feidas H, Kouam MK, Kantzoura V, Theodoropoulos G. Global geographic distribution of Trichinella species and genotypes. Infect Genet Evol 2014;26:255–66.

[253] Pozio E. The opportunistic nature of Trichinella—Exploitation of new geographies and habitats. Vet Parasitol 2013;194:128–32.

[254] ITRC. Epidemiological database. Int Trichinella Ref Cent 2017. https://www.iss.it/site/Trichinella/scripts/epid.asp (accessed October 20, 2017).

[255] Mukaratirwa S, Grange L la, Pfukenyi DM. Trichinella infections in animals and humans in sub-Saharan Africa: a review. Acta Trop 2013;125:82–9.

[256] Krivokapich SJ, Gonzalez Prous CL, Gatti GM, Saldía L. First finding of Trichinella pseudospiralis in the Neotropical region. Vet Parasitol 2015;208:268–71.

[257] Krivokapich SJ, Pozio E, Gatti GM, Prous CLG, Ribicich M, Marucci G, et al. Trichinella patagoniensis n. sp. (Nematoda), a new encapsulated species infecting carnivorous mammals in South America. Int J Parasitol 2012;42:903–10.

[258] Calcagno MA, Bourlot I, Taus R, Saracino MP, Venturiello SM. Description of an outbreak of human trichinellosis in an area of Argentina historically regarded as Trichinella-free: the importance of surveillance studies. Vet Parasitol 2014;200:251–6.

[259] De-la-Rosa JL, Aranda JG, Padilla E, Correa D. Prevalence and risk factors associated with serum antibodies against Trichinella spiralis 1998;28:317–21.

[260] Schurer JM, Ndao M, Quewezance H, Elmore SA, Jenkins EJ. People, pets, and parasites: one health surveillance in southeastern Saskatchewan. Am J Trop Med Hyg 2014;90:1184–90.

[261] Springer YP, Casillas S, Helfrich K, Mocan D, Smith M, Arriaga G, et al. Two outbreaks of trichinellosis linked to consumption of Walrus meat - Alaska, 2016- 2017. Morb Mortal Wkly Rep 2017;66:692–6.

[262] Pozio E, Rinaldi L, Marucci G, Musella V, Galati F, Cringoli G, et al. Hosts and habitats of Trichinella spiralis and Trichinella britovi in Europe. Int J Parasitol 2009;39:71–9.

[263] Skripova L V, Kovchur VN. Trichinellosis in Byelorussia. Wiad Parazytol 1994;40:389–91.

[264] Sadkowska-Todys M, Golab E. Trichinellosis in Poland in 2011. Przegl Epidemiol 2013;67:259–61.

[265] De Bruyne A, Ancelle T, Vallee I, Boireau P, Dupouy-Camet J. Human trichinellosis acquired from wild boar meat: a continuing parasitic risk in France. Euro Surveill 2006;11:e060914.5.

[266] Ravlija J, Puvacic S, Puvacic Z, Curic I. Epidemiological characteristics of trichinosis in Federation of Bosnia and Herzegovina. Med Arh 2006;60:251–4.

[267] Oltean M, Kalmar Z, Kiss BJ, Marinov M, Vasile A, Sandor AD, et al. European mustelids occupying pristine wetlands in the Danube Delta are infected with Trichinella likely derived from domesticated swine. J Wildl Dis 2014;50:972–5.

[268] Karssin A, Velstrom K, Gomez-Morales MA, Saar T, Jokelainen P, Lassen B. Cross-sectional study of anti-Trichinella antibody prevalence in domestic pigs and hunted wild boars in Estonia. Vector Borne Zoonotic Dis 2016;16:604–10.

[269] Garbarino C, Interisano M, Chiatante A, Marucci G, Merli E, Arrigoni N, et al. Trichinella spiralis a new alien parasite in Italy and the increased risk of infection for domestic and wild swine. Vet Parasitol 2017;246:1–4.

[270] Bhawna S, Butola KS, Bharti A, Yogesh K, Vijay S. Human trichinosis in remotes of Uttarakhand, India. Indian J Med Sci 2010;64:104–10.

[271] Chatterjee T, Jagani R, Sabhiki AK, Yogesh C. Trichinosis. Med J Armed Forces India 2000;56:161–2.

[272] Niphadkar SM. Ecological study of helminths of three species of rats in Bombay in relation to veterinary and public health. 1st Natl. Congr. Parasitol. Baroda, 1977, p. 45–6.

[273] Niphadkar SM, Pardhan MH, Deshpande VS. Rediscovery of Trichinella spiralis (Owen, 1835) in domestic pigs in India. Curr Sci 1979;48:372–3.

[274] Lo YC, Hung CC, Lai CS, Wu Z, Nagano I, Maeda T, et al. Human trichinosis after consumption of soft-shelled turtles, Taiwan. Emerg Infect Dis 2009;15:2056–8.

[275] Liu M, Boireau P. Trichinellosis in China: epidemiology and control. Trends Parasitol 2002;18:553–6.

[276] Luo X, Hong N. Seroprevalence of antibodies against three food safety-related porcine pathogens in Guizhou Province, Southwestern China. J Anim Vet Adv 2014;13:668–70.

[277] Barennes H, Sayasone S, Odermatt P, De Bruyne A, Hongsakhone S, Newton PN, et al. A major trichinellosis outbreak suggesting a high endemicity of Trichinella infection in northern Laos. Am J Trop Med Hyg 2008;78:40–4.

[278] Cui J, Wang ZQ, Xu BL. The epidemiology of human trichinellosis in China during 2004-2009. Acta Trop 2011;118:1–5.

[279] Sayasone S, Odermatt P, Vongphrachanh P, Keoluangkot V, Dupouy-Camet J, Newton PN, et al. A trichinellosis outbreak in Borikhamxay Province, Lao PDR. Trans R Soc Trop Med Hyg 2006;100:1126–9.

[280] Jongwutiwes S, Nutaros C, Phisai K, Padet S, Putaporntip C, Tamburrini A, et al. First outbreak of human trichinellosis caused by Trichinella pseudospiralis. Clin Infect Dis 1998;26:111–5.

[281] Conlan J V, Vongxay K, Khamlome B, Gomez-Morales MA, Pozio E, Blacksell SD, et al. Patterns and risks of Trichinella infection in humans and pigs in northern Laos. PLoS Negl Trop Dis 2014;8:e3034.

[282] Dupouy-Camet J. Trichinellosis: a worldwide zoonosis. Vet Parasitol 2000;93:191–200.

[283] Devleesschauwer B, Praet N, Speybroeck N, Torgerson PR, Haagsma JA, De Smet K, et al. The low global burden of trichinellosis: Evidence and implications. Int J Parasitol 2015;45:95–9.

[284] Murrell KD, Pozio E. Worldwide occurrence and impact of human trichinellosis, 1986-2009. Emerg Infect Dis 2011;17:2194–202.

[285] Jen JJS, Chen J. Chapter 9: Foodborne Parasitic Diseases in China. In: Hoboken NJ, editor. Food Saf. China Sci. Technol. Manag. Regul., Chennai, India: JohnWiley & Sons; 2017, p. 127–46.

[286] OIE. Manual Of Diagnostic Tests And Vaccines For Terrestrial Animals (mammals, birds and bees). vol. 1. 6th ed. Paris: OIE; 2004.

[287] Webster P, Maddox-Hyttel C, Nockler K, Malakauskas A, van der Giessen J, Pozio E, et al. Meat inspection for Trichinella in pork, horsemeat and game within the EU: available technology and its present implementation. Euro Surveill 2006;11:50–5.

[288] Gamble HR. Detection of trichinellosis in pigs by artificial digestion and enzyme immunoassay. J Food Prot 1996;59:295–8.

[289] Gamble HR, Bessonov AS, Cuperlovic K, Gajadhar AA, van Knapen F, Noeckler K, et al. International Commission on Trichinellosis: Recommendations on methods for the control of Trichinella in domestic and wild animals intended for human consumption. Vet Parasitol 2000;93:393–408.

[290] Kapel CM, Webster P, Gamble HR. Muscle distribution of sylvatic and domestic Trichinella larvae in production animals and wildlife. Vet Parasitol 2005;132:101–5.

[291] Wang GY, Du JF, Dun GQ, Sun WL, Wang JX. Evaluation of artificial digestion method on inspection of meat for Trichinella spiralis contamination and influence of the method on muscle larvae recovery. Zhongguo Xue Xi Chong Bing Fang Zhi Za Zhi 2011;23:211–3.

[292] Franssen F, Deksne G, Esite Z, Havelaar A, Swart A, Giessen J van der. Trend analysis of Trichinella in a red fox population from a low endemic area using a validated artificial digestion and sequential sieving technique. Vet Res 2014;45:120.

[293] Li F, Cui J, Wang ZQ, Jiang P. Sensitivity and optimization of artificial digestion in the inspection of meat for Trichinella spiralis. Foodborne Pathog Dis 2010;7:879–85.

[294] Frey CF, Buholzer P, Beck R, Marinculic A, Raeber AJ, Gottstein B, et al. Evaluation of a new commercial enzyme-linked immunosorbent assay for the detection of porcine antibodies against Trichinella spp. J Vet Diagn Invest 2009;21:692–7.

[295] Cuttell L, Gomez-Morales MA, Cookson B, Adams PJ, Reid SA, Vanderlinde PB, et al. Evaluation of ELISA coupled with Western blot as a surveillance tool for Trichinella infection in wild boar (Sus scrofa). Vet Parasitol 2014;199:179–90.

[296] Zarlenga DS, La Rosa G. Molecular and biochemical methods for parasite differentiation within the genus Trichinella. Vet Parasitol 2000;93:279–92.

[297] Guenther S, Nockler K, von Nickisch-Rosenegk M, Landgraf M, Ewers C, Wieler LH, et al. Detection of Trichinella spiralis, T. britovi and T. pseudospiralis in muscle tissue with real-time PCR. J Microbiol Methods 2008;75:287–92.

[298] Lin Z, Cao J, Zhang H, Zhou Y, Deng M, Li G, et al. Comparison of three molecular detection methods for detection of Trichinella in infected pigs. Parasitol Res 2013;112:2087–93.

[299] Nagano I, Wu Z, Matsuo A, Pozio E, Takahashi Y. Identification of Trichinella isolates by polymerase chain reaction-restriction fragment length polymorphism of the mitochondrial cytochrome c-oxidase subunit I gene. Int J Parasitol 1999;29:1113–20.

[300] Zarlenga DS, Chute MB, Martin A, Kapel CM. A multiplex PCR for unequivocal differentiation of all encapsulated and non-encapsulated genotypes of Trichinella. Int J Parasitol 1999;29:1859–67.

[301] Appleyard GD, Zarlenga D, Pozio E, Gajadhar AA. Differentiation of Trichinella genotypes by polymerase chain reaction using sequence-specific primers. J Parasitol 1999;85:556–9.

[302] Zarlenga DS, Dame JB. The identification and characterization of a break within the large subunit ribosomal RNA of Trichinella spiralis: comparison of gap sequences within the genus. Mol Biochem Parasitol 1992;51:281–9.

[303] Mayer-Scholl A, Broglia A, Reckinger S, Nöckler K. Polymerase chain reaction - Restriction fragment length polymorphism analysis for the differentiation of Trichinella nativa and Trichinella britovi. Vet Parasitol 2014;203:247–9.

[304] Cuttell L, Corley SW, Gray CP, Vanderlinde PB, Jackson LA, Traub RJ. Real- time PCR as a surveillance tool for the detection of Trichinella infection in muscle samples from wildlife. Vet Parasitol 2012;188:285–93.

[305] Gamble HR, Pozio E, Bruschi F, Nockler K, Kapel CM, Gajadhar AA. International Commission on Trichinellosis: recommendations on the use of serological tests for the detection of Trichinella infection in animals and man. Parasite 2004;11:3–13.

[306] Noeckler K, Hamidi A, Fries R, Heidrich J, Beck R, Marinculic A. Influence of methods for Trichinella detection in pigs from endemic and non-endemic European region. J Vet Med B 2004;51:297–301.

[307] De-la-Rosa JL, Alcantara P, Correa D. Investigation of cross-reactions against Trichinella spiralis antigens by enzyme-linked immunosorbent assay and enzyme-linked immunoelectrotransfer blot assay in patients with various diseases. Clin Diagn Lab Immunol 1995;2:122–4.

[308] Gomez-Morales MA, Ludovisi A, Amati M, Blaga R, Zivojinovic M, Ribicich M, et al. A distinctive Western blot pattern to recognize Trichinella infections in humans and pigs. Int J Parasitol 2012;42:1017–23.

[309] Gomez-Morales MA, Ludovisi A, Amati M, Bandino E, Capelli G, Corrias F, et al. Indirect versus direct detection methods of Trichinella spp. infection in wild boar (Sus scrofa). Parasites and Vectors 2014;7:171.

CHAPTER 2

A systematic review of taeniasis, cysticercosis and trichinellosis in Vietnam

Presented as published:

Ng-Nguyen D, Stevenson MA, Traub RJ. A systematic review of taeniasis, cysticercosis and trichinellosis in Vietnam. Parasit Vectors 2017;10:150.

2.1 Abstract

Taeniasis, cysticercosis and trichinellosis have been ranked as the most important food- borne parasites of humans in terms of public health, socioeconomic and trade impact. Despite this, information on these food-borne zoonoses in Vietnam is scarce and fragmented, and many local reports remain inaccessible to the international research community. This study aims to conduct comprehensive literature searches to report on the incidence and estimate the TP of taeniasis in humans and T. solium cysticercosis in humans and pigs in Vietnam utilising Bayesian models; in addition, to report the incidence and the distribution of trichinellosis.

A Bayesian approach was used to estimate the TP of taeniasis and cysticercosis based on published diagnostic test characteristics used in each published cross-sectional survey. The utilisation of coproscopic-based examination of Taenia eggs in stool, although highly specific for genus-level detection, has poor Se and led to an underestimation of the prevalence of human taeniasis. Similarly, post-mortem-based surveys of T. solium cysticercosis in pigs also led to the underestimation of prevalence of porcine cysticercosis. On the other hand, the low Sp of immunodiagnostic methods, in particular Ab-ELISA, led to a likely overestimation of T. solium cysticercosis in humans. Due to the use of imperfect diagnosis tests combined with poor descriptions of sampling methods, our ability to draw solid conclusions from these data is limited. We estimate that the TP of taeniasis and T. solium cysticercosis in rural 'hotspots', is as high as 13% for each, in humans. Taeniasis and T. solium cysticercosis occurs in 60 of the 63 provinces of Vietnam. Most of the information relating to the distribution and prevalence of porcine cysticercosis is limited to commercial abattoir surveys. In Vietnam, Taenia asiatica appears to be confined to the north where it occurs sympatrically with T. solium and Taenia saginata. The status of T. asiatica in Central and South Vietnam remains unascertained. To date, five outbreaks of trichinellosis have been reported in the north and northwest of Vietnam, affecting a total of 114 people and responsible for eight fatalities. In the same region, studies of free-roaming pigs showed evidence of high levels of exposure to Trichinella and, in cases where larvae were recovered, the species present were identified as Trichinella spiralis. Based on five studies, the main risk factors for pork-borne zoonoses in Vietnam include the

77 consumption of undercooked/raw meat and vegetables and the use of night-soil for fertilization of local produce. This systematic review draws attention to the importance of these pork-borne zoonoses.

2.2 Background

Taeniasis, cysticercosis and trichinellosis have been ranked as the most important food- borne parasites of humans in terms of public health, socioeconomic and trade impact [1]. In 2010 it was estimated that approximately 300,000 individuals were infected with T. solium cysticercosis globally, resulting in over 28,000 deaths [2]. Between 2.5 to 5 million people are estimated to harbour adult tapeworms of T. solium [3–5]. Humans act as the definitive hosts for all three-tapeworm species (Taenia solium, Taenia saginata and Taenia asiatica) via the ingestion of undercooked/raw meat and/or offal [6–8]. Swine are the intermediate hosts of T. solium and T. asiatica whereas cattle are the intermediate host for T. saginata [9]. Humans may also become infected with cysticerci of T. solium. In humans, the symptoms of taeniasis are subtle and mild and include abdominal distension, abdominal pain, digestive disorders and anal pruritis [7, 10]. The signs of T. solium NCC in humans, on the other hand, are distinctive and include seizures, paralysis, dementia, chronic headache, blindness or even death [11, 12].

Trichinellosis is reported to be present in 55 countries [13] with 66,000 individuals estimated to have been infected during the period 1986 to 2009 [1]. In 2010 it was estimated that globally there were around 4,400 cases of trichinellosis reported, with four deaths [2]. Parasites of the genus Trichinella comprise nine species and four genotypes divided into two groups, encapsulated and non-encapsulated [14]. There are two epidemiological cycles of Trichinella: domestic and sylvatic. The domestic cycle occurs between domestic animals, particularly domestic pigs and rodents. Of all Trichinella species and genotypes, T. spiralis is the most widely distributed and most adapted to domestic pigs, accounting for the most cases of human trichinellosis in Asia, however other sylvatic species such as T. pseudospiralis and T. papuae have also been associated with human outbreaks in the region [15, 16]. Typically, trichinellosis results in nausea, diarrhoea, high fever, petechial and nailbed hemorrhages, periorbital oedema and myalgia [17]. The severity of symptoms in humans are directly proportional to the number of larvae ingested.

Taeniasis, T. solium cysticercosis [18] and trichinellosis [13] have been reported as endemic in Southeast Asia, including Vietnam. Vietnam has a population of around 90 million people that reside in 63 provinces. There are 54 ethnic groups and approximately 67% of the population live in rural areas [19, 20]. The standard of living in most rural communities is poor. Open defaecation using outdoor latrines is common practice and livestock access to these latrine areas is, for the most part, unrestricted. Vietnam is grouped as a lower middle-income country with a Gross Domestic Product (GDP) per capita of USD 2,052, ranking it 116th out of 188 countries listed in the United Nations Human Development Index [21, 22]. There are only 79 physicians per 100,000 people; however, this figure is likely to be much lower in remote regions [20]. Moreover, consumption of raw or undercooked blood, meat and organs from domesticated and wild pigs and cattle and raw vegetables is common practice, as are the practices of using night- soil (human faeces) and waste water to fertilise and irrigate crops [23, 24]. These factors combined, are highly conducive for the transmission of taeniasis, T. solium cysticercosis and trichinellosis.

Vietnam’s domestic livestock sector comprises approximately 8 million cattle and buffalo, and 27 million pigs, which produce nearly 3,800 million tons of meat products annually [20]. Two primary types of pig- and cattle-husbandry practices exist: commercial farming and backyard husbandry. In rural regions, backyard husbandry practices dominate [23], although the trend is rapidly changing to expand to commercial farming practices aimed at increasing productivity by using hybrid rather than local/traditional breeds. The local/traditional breeds are, however, still raised in many remote and rural areas to use as a supply of food for their owners or nearby communities [25]. The practice of non-confinement of pigs and cattle is common in rural regions [25] with slaughter activities commonly carried out in backyards without official meat inspection. Meat inspection is only carried out at abattoirs or slaughter- points, which operate at the district level and/or clusters of large villages.

This review provides updated literature on the incidence and distribution, and critically evaluates both available and previously unavailable local reports on taeniasis, T. solium cysticercosis and trichinellosis in Vietnam. As such, it aims to inform the distribution of reported cases of trichinellosis, and estimate the TP of taeniasis in humans, and T.

79 solium cysticercosis in humans and pigs in Vietnam utilising Bayesian models thus drawing attention to the importance of these pork-borne zoonoses.

2.3 Methods

We selected published articles reporting on the prevalence, incidence and/or occurrence of Taenia and Trichinella in humans and pigs in Vietnam. The results of the search were not restricted by time, journal, or status of publication. The protocol of this systematic review followed guidelines of the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) [26].

2.3.1 Search protocol

Information from published articles for this study was sourced in two ways: (i) through international and Vietnamese online database searches, (ii) through Vietnamese publications obtained from hospitals, government, and research institution and university libraries. Online database searches were performed using PubMed, CABI abstract, Web of Science, MEDLINE (Web of Knowledge), and Scopus. English search terms and keywords in the Boolean operators were used as follows: (Taenia solium OR Taenia saginata OR Taenia asiatica OR human tapeworms OR cysticercosis OR NCC OR taeniasis OR taeniosis OR Trichinella OR trichinellosis) AND (epidemiology OR prevalence) AND (human OR porcine OR swine OR pig OR bovine OR cattle OR cow OR calf) AND Vietnam. Scientific names and Vietnamese synonym names of Taenia and Trichinella were used to search databases of the National Library of Vietnam (NLV) and the National Agency for Science and Technology Information of Vietnam (NASATI, http://db.vista.gov.vn). Published articles written in Vietnamese were also sourced through the public search engine (Google.com.vn) using Vietnamese keywords for Taenia or Trichinella.

2.3.2 Study selection

After independent identification, duplicate articles were removed. The titles and abstracts of articles were then critically evaluated and excluded in cases where they did not address the prevalence, and/or incidence of taeniasis, T. solium cysticercosis and/or Trichinella in humans and/or pigs. Full-text records were excluded if they were review

80 articles; or two articles using the same data; or were not written in English or Vietnamese; or not addressing Taenia and/or Trichinella in humans and pigs in Vietnam. Remaining full-text articles were included for this review. Risk of bias was reduced by only including published reports and excluding studies in which sampling was not comprehensive, such as: source of samples not identified, or where the sample size was not clear for inclusion in the Bayesian modelling (see Statistical analysis).

2.3.3 Data collection

The following data were retrieved from each study: the title, authors, year of publication, time, sample source/place, sample size, number of positive samples, diagnostic method utilised for prevalence/incidence determination and factors effecting prevalence, incidence and distribution of positively identified samples.

2.3.4 Statistical analysis

In this review the prevalence of taeniasis and cysticercosis, as documented in each of the selected published articles is reported as ‘apparent prevalence’, AP. That is, for a given number of individuals sampled and tested using a given diagnostic method, AP equals the number of test-positive individuals divided by the total number of individuals tested. Given differences in the diagnostic test protocols used in each of the cross- sectional studies reported in this review, AP estimates have been adjusted to TP, estimates (taking into account imperfect diagnostic Se and Sp using the general approach described by Rogan & Gladen (1978) [27], and modified for the low prevalence situation using Bayesian methods as described by Messam, Branscum, Collins, & Gardner, (2008) [28].

In brief, the Bayesian approach accounts for uncertainty in the values of TP and diagnostic Se and Sp of the testing protocol. If 푥 equals the number of individuals testing positive using a diagnostic test of 푆푒 and 푆푝, the distribution of the number of test positive individuals is 푥 | (푇푃, 푆푒, 푆푝) ~ 푏푖푛표푚푖푎푙(푛, 퐴푃) where

퐴푃 = 푇푃 × 푆푒 +(1− 푇푃)(1 − 푆푝).

To estimate the TP of taeniasis and cysticercosis for each study beta prior distributions for 푆푒 and 푆푝 were used. We assumed that the TP of taeniasis and cysticercosis was not equal to zero and that all prevalences were equally likely, which translates to a 푇푃 ~ 푏푒푡푎(1,1) distribution. Markov chain Monte Carlo (MCMC) methods were used to derive posterior estimates of 푇푃 using WinBUGS [29, 30]. In WinBUGS, the MCMC sampler was run for 200,000 iterations and the first 1,000 ‘burn in’ samples were discarded. The posterior distribution of 푇푃 was obtained by running sufficient iterations to ensure that the Monte Carlo standard error of the posterior means were at least one order of magnitude smaller than their posterior standard deviation [31] The point estimate and 95% credible interval (CrI) for 푇푃 is reported as the median and 0.025 and 0.975 quantiles of the posterior distribution of 푇푃.

2.4 Results

2.4.1 Records used for quality and quantity analysis

The protocol for selecting the published records is shown in Fig. 2.1. There were initially 75 records retrieved from online data, the internet and other resources. Of these records, 40 records were eligible for qualitative and quantitative (23 records) analysis. Of 40 retrieved records, 13 studies were published in international journals, 24 studies were published in national journals and three records were circulated within professional institutes. The numbers of record studied on taeniasis, and both taeniasis and human cysticercosis were 12 and 10, respectively; whereas there were eight papers report on porcine cysticercosis. Research on Trichinella and trichinellosis were identified in seven records, in which three were studies of humans, three were studies of pigs and one was a study of both of humans and pigs. Our systematic search identified only one (Hanoi-based) study reporting on the prevalence of bovine cysticercosis between 2002 and 2003.

Tables 2.1 - 2.3 provide a summary of each of the studies cited in this review. For each study the sampling protocol (if known) is reported as well as the diagnostic test used, the number of individuals sampled, the number of individuals testing positive, and the apparent and TP estimates. For both prevalence estimates 95% CrI for the true population values are reported.

2.4.2 Human taeniasis

Data on the AP and estimated TP of human taeniasis is summarized in Table 2.1. Data on the prevalence of taeniasis in Vietnam varied markedly across study sites depending on dietary habits, pig husbandry practices and the socio-economic status of study participants. Human taeniasis was reported in 50 of the 63 provinces of Vietnam [33], with the average province-level AP estimates from 0.11% in Hoa Binh in the north to 10% in Kon Tum, in the Central Highlands (Table 2.1).

In North Vietnam, the AP of taeniasis between 2002 and 2012 ranged from 0.11% to 8.65% (Table 2.1). In the North of Vietnam, high foci of taeniasis were found in the provinces of Yen Bai (9.0%) [32], Bac Ninh (12.68%) [33] and Phu Tho (8.65%) [34]. In Ha Giang, located in Northeast Vietnam, 5 out of 84 (6.0%) people were positive for Taenia spp. [35]. In contrast, a retrospective study of 6,570 patients presenting to Thai Binh Medical University Hospital with digestive disorders between 2008 and 2010 found that only 11 (0.17%) people were confirmed positive for taeniasis [36].

In 2005, the Institute of Malariology, Parasitology and Entomology Quy Nhon reported 0.33% of the population to be positive for taeniasis, based on a large-scale helminth survey using the Kato-Katz (KK) method. This study involved 35,651 participants in 14 provinces spanning the Central and Central Highland regions of Vietnam [37]. The AP of taeniasis in Chuong [38] and Van Tuan [39] in 2011 and 2014 was relatively high (8%; 95% CI 7 to 9 and 10%; 95% CI 8 to 13, respectively). It should be noted that sampling in this study was targeted towards those individuals known to habitually consume raw or undercooked beef. Although the sampling protocol used in this study was likely to result in an estimate of the prevalence of taeniasis that was greater than that of the general population, the TP among habitual raw or undercooked beef consumers was 11.0% (95% CrI 7.0 to 21) and 13.5% (95% CrI 8.0 to 21) in 2011 and 2014, respectively (Table 2.1). To the best of our knowledge, to date there have been no reports on the prevalence of taeniasis in South Vietnam (Fig. 2.2).

Consumption of raw and/or undercooked pork or beef in traditional dishes such as ‘Nem Chua’, undercooked liver, undercooked-beef/pork noodle soup, grilled pork/beef and visceral-raw blood [32, 34, 39] have been identified as major risk factors for taeniasis in

83 the above-mentioned provinces. For all epidemiological studies of taeniasis carried out across Vietnam, it has consistently been shown that males account for a higher proportion of taeniasis cases compared with females. In addition, adults of working age have been shown to have a higher risk of infection compared with other age groups [34, 38–42].

Before the first report of T. asiatica in the North of Vietnam in 2001 [43] the relative proportions of T. saginata to T. solium cases were 80% and 22%, respectively, based on morphological identification of purged proglottids in stools [44]. Later, Doanh et al. (2006) [33] confirmed the presence of T. asiatica in six samples from Bac Ninh province using molecular diagnostic techniques. Somers et al. (2007) [45] showed that T. asiatica dominated taeniasis cases (55%) in the northern provinces of Vietnam, followed by T. saginata (38.5%) and T. solium (6.2%). Between 2005 and 2006 Taenia proglottids from 65 individuals were identified from 19 provinces in the North. This study confirmed the presence of T. asiatica and T. saginata but not T. solium [42]. Huong (2006) [35] confirmed the presence of T. asiatica in three out of five samples collected in Ha Giang province. De and Hoa (2006) [46] identified T. saginata proglottids in eight patients in the Central provinces, the Central Highlands and the South of Vietnam using molecular diagnostic techniques. No T. solium or T. asiatica proglottids were identified. The latter study is the only report describing the species of human tapeworms occurring in Central and South Vietnam. A limited number of studies indicate that T. asiatica is restricted to the north of the country.

2.4.3 Human cysticercosis

A review of the literature indicates that human cysticercosis has been reported in 55 of the 63 provinces of Vietnam. Data on the AP and estimated TP of T. solium cysticercosis in humans is summarized in Table 2.2. The prevalence of infection varies widely between study areas ranging from no cases found in the central province of Hai Duong to 13% (95% CI 0.80 to 22) in Ha Giang province in the north.

In the six-year period from 2006 to 2011, an estimated 250 to 400 patients from 34 provinces in North Vietnam were hospitalized and treated for T. solium cysticercosis by the National Institute of Malariology, Parasitology and Entomology (NIMPE) annually.

The majority of these patients were from Hai Phong, Thanh Hoa, Ha Noi, Bac Giang and Bac Ninh provinces [32]. Between 1999 and 2000, the results of three village-based surveys conducted in Bac Ninh province showed an AP of cysticercosis of 5.0% (range from 2.2 to 7.2%) [47]. A cross-sectional study carried out in the same province by Erhart et al. (2002) [48] in 1999 showed that 12 (5.7%) out of 210 individuals in Ty Dien village had circulating T. solium cysticercosis antigen and nine were confirmed as having NCC using computed tomography scanning. A short report on the situation of taeniasis and cysticercosis in Vietnam by De (2004) stated that of 4,017 people diagnosed with helminth infection by NIMPE, 633 were also positive for cysticercosis. No further details of area-specific prevalence or methods of diagnosis were provided in this report.

In the South of Vietnam, a sero-epidemiological survey carried out in hospitals located in Ho Chi Minh City from 1992 to 2000 involving 3,814 people of all ages mainly originating from South Vietnam found that 4.3% were positive for cysticercosis. Patients that were residents of Ho Chi Minh City and the province of An Giang accounted for the highest proportions of study subjects that were cysticercus positive i.e. 20% and 14%, respectively [49]. The study of Anh Tuan et al. (2001) [49] is the only study documenting the prevalence of T. solium cysticercosis in the South of Vietnam, and to the best of our knowledge, there are no reports of the prevalence of T. solium cysticercosis for Central Vietnam (Fig. 2.3). A single case report describes T. solium NCC in three ethnic minority patients from Quang Tri and Quang Ngai province hospitalized at Hue Central Hospital in 2012. Each had seizures and headaches [50].

NIMPE (2012) reported that among 2,687 T. solium cysticercosis cases, the proportion of males was significantly greater than that of the proportion females [32]. In addition, study subjects that were between 30 and 60 years of age were over-represented. Anh Tuan et al. (2001) [49] reported that in Ho Chi Minh City, males comprised 56% and adults comprised 86% of the 163 study subjects that were positive to an Ab-ELISA for T. solium cysticercosis. The consumption of raw vegetables, drinking unboiled water, not washing hands before eating and outdoor defecation were found to be risk factors for T. solium cysticercosis in this study. Undoubtedly, utilisation of night-soil for fertilizing crops is a major contributing factor to T. solium cysticercosis and this practice

85 has been reported in both the North and South of Vietnam.

2.4.4 Porcine cysticercosis

Data on the prevalence of porcine cysticercosis at the national level is lacking, fragmented and/or out of date. Surveys on porcine cysticercosis were carried out primarily in the north and mostly in Hanoi slaughterhouses (Fig. 2.4). Data on the AP and estimated TP of porcine cysticercosis is summarized in Table 2.3. There were no cases of cysticercosis among approximately 8,000 carcasses examined at slaughterhouses in Nam Dinh and Ha Nam, Hai Duong and Hung Yen provinces [51] (Table 2.3). Between 1999 and 2001, 0.06% of 198,887 pig carcasses examined in five provinces in the North and one province in Central Vietnam were identified as infected [52]. In a community-based survey carried out between 1999 to 2000 in five villages in Bac Kan and Bac Ninh province (known to be highly endemic foci for T. solium cysticercosis) an apparent cysticercosis seroprevalence of 9.91% (95% CI 7.10 to 14) was reported in pigs [47] using Ag-ELISA. In a survey from 2002 to 2003 on porcine cysticercosis was carried out in abattoirs and markets in Hanoi, in which only two out of 143,868 carcasses examined were found to be infected with T. solium [53].

In South Vietnam, there is a paucity of information on the prevalence of cysticercosis in pigs (Fig. 2.4). A single study conducted in 1994 showed that 0.90% of 891 pigs from 18 districts located in 12 southern provinces were T. solium cysticercosis positive [54].

2.4.5 Trichinellosis

In Vietnam, the first human case of trichinellosis was identified in 1968 in a group of people consuming pork from Lao PDR [55]. Since then, there have been five reported outbreaks in four provinces (Table 2.4 and Fig. 2.5). In 1970, an outbreak in Mu Cang Trai District, Yen Bai province (in the north) resulted in 30 cases of trichinellosis, including four deaths [56]. In 2002, 22 people in Tuan Giao District, Dien Bien province were confirmed infected of which two people died. In 2008, in Bac Yen District, Son La province an outbreak of trichinellosis arising from consumption of undercooked pork from domestic pigs resulted in 23 cases of trichinellosis and two deaths [57]. Similarly, 24 out of 27 individuals acquired trichinellosis after eating raw

86 pork in a mountainous region of Thanh Hoa province in 2012. Of these, six went on to develop serious symptoms [58]. Following the outbreak of human trichinellosis in Son La province in 2008, 20% of free-roaming pigs were reported seropositive for Trichinella antibodies in the four villages in which the human outbreaks occurred [59]. Of the 206 muscle samples that were seropositive, Trichinella larvae were recovered from 11 samples and identified as T. spiralis using multiplex PCR. The proportion of Trichinella seropositive wild boars and rats in Son La and Dien Bien province was 3.2% (2 positive out of 62 tested) and 2.8% (23 positive out of 820 tested), respectively, and 4% (5 positive out of 125 tested) of dogs were also found to be seropositive to Trichinella [60]. Surprisingly, none of the 261 confined wild boars resident on seven farms or 98 cats in these two provinces were positive [61]. In the same study, T. spiralis larval burdens quantified using multiplex PCR were relatively low at 0.1 to 0.3 larvae/gram of muscle in wild boars and 0.6 larvae/gram of muscle in rats [61]. Until now, the only species of Trichinella isolated in domestic and sylvatic hosts in North Vietnam has been T. spiralis [62]. The habit of eating traditional dishes prepared using under-cooked and/or raw game meat (wild pork) at special events such as the lunar New Year, weddings and funeral events is the reason attributed to Trichinella outbreaks in North Vietnam. Leaving pigs to free roam, and feeding raw and/or left-over food to pigs has been blamed for the transmission of Trichinella in the region where medical personnel often lack knowledge about trichinellosis and its clinical symptoms. Moreover, in this area of Vietnam healthcare services are difficult to access [60]. There are no reports available on the seroprevalence of Trichinella infection in wild boars and domestic swine from other regions in Vietnam.

2.5 Discussion

Our study has shown that taeniasis and T. solium cysticercosis occurs in 60 of the 63 provinces of Vietnam (Fig. 2.6). While data on the prevalence of taeniasis, cysticercosis and trichinellosis are available for North Vietnam, the relatively small number of studies carried out in the centre and south of Vietnam mean that it is difficult to draw definitive conclusions about the prevalence of these conditions in these areas of the country.

Across all of the cited studies utilising microscopy to detect taeniasis, TP estimates were greater than the AP estimates (Table 2.1). For example, the TP of taeniasis in Ha Giang [35] at 8.00% (95% CrI 2.70 to 20) was greater than that of the AP (6.00%, 95% CI 2.60 to 13). Similarly the estimated TP in Chuong (2005) [37] was 0.52% (95%, CrI 0.03 to 6.0) nearly twice that of the AP (Table 2.1). It is highly likely that the prevalence of taeniasis has been grossly underestimated due to the utilisation of microscopy-based examinations for the presence of Taenia spp. eggs using KK and/or Formalin-Ether Concentration techniques. The difference between the reported apparent and estimated TP is likely to be a result of the low Se of the microscopy-based methods to detect light taeniasis infections [63] and intermittent proglottid shedding [64]. It is also noteworthy that microscopic-based techniques are incapable of species-level identification of Taenia spp. in stool as the eggs are morphologically identical.

In contrast to the TP of taeniasis, the TP estimates for human cysticercosis were lower than the AP estimates in across all of the cited studies (Table 2.2). The TP of cysticercosis in the study by Anh Tuan et al. (2001) [49] in South Vietnam was 0.89% (95% CrI 0.16 to 2.53), much lower than the AP of 4.30% (95% CI 3.70 to 4.99). It should be noted that Ab-ELISA with a relatively poor Sp were utilised for this study; thus, overestimation of AP on human cysticercosis in this region may occur since Ab-ELISAs cross-react with Echinococcus granulosus (hydatid disease), Schistosoma spp. Angiostrongylus cantonensis, Spirometra spp., Fasciola spp. and Hymenolepis spp. [65, 66]. On the other hand, Ab-ELISAs provide a measure of T. solium cysticercosis exposure, not T. solium cysticercosis infection. As a result, Ab-ELISAs may also detect cases of transitional antibody exposure to T. solium [67, 68] and/or immunologically persistent antibody [69], hence it may overestimate both the AP and TP of active T. solium cysticercosis in a community.

Ag-ELISA is known to have an acceptable Se of 87% (95% CI 62 to 98) and Sp of 95% (95% CI 90 to 99) for the detection of T. solium cysticercosis active infections in humans [70]; however, the estimated TP of human cysticercosis was lower than that of the AP in all of the cited studies in this review. This can be attributed to the uncertainty of the TP of human cysticercosis across study locations couple with imperfect of the assay (95%). Ag-ELISAs are superior to Ab-ELSIAs in term of detecting active

88 infection of T. solium cysticercosis; however, since both assays are genus-specific, this does not necessarily apply to pigs. In pigs, the assays are known to cross-react with antigens of Taenia hydatigena, a common tapeworm of swine in Vietnam [71]. Therefore, the AP of T. solium cysticercosis in village pigs in Bac Kan and Bac Ninh province of 9.9% (95% CI 6.1 to 15) is likely to be an overestimate. Of 29 pigs positive to Ag-ELISA, only five pigs were found to be infected with cysts of T. solium and ten pigs with cysts of T. hydatigena [52].

From the data presented in this systematic review, it appears that abattoir-based surveys may be biased and are likely to result in an underestimation of the risk of taeniasis and cysticercosis in humans because the majority pigs presented to these abattoirs were sourced from commercial piggeries. On the other hand, village-reared pigs are likely to be slaughtered locally and therefore represent a greater risk of infection for humans. Moreover, post-mortem inspection-based techniques have a low reported Se of 22% for the detection of cysticerci in meat [63] and are likely to lead to a further underestimate of the TP of T. solium cysticercosis.

The prevalence of taeniasis and porcine cysticercosis in Vietnam are most likely to be underestimated whereas the prevalence of human cysticercosis is likely to be overestimated. In relative terms, the prevalence of porcine cysticercosis was low compared with taeniasis. Since microscopic-based diagnosis is genus-specific, it is likely that T. saginata and potentially T. asiatica accounted for a proportion of Taenia infections in humans. Thus, further epidemiological surveys on bovine cysticercosis, and village-based surveys on porcine cysticercosis is necessary to fully unveil the entire epidemiological picture of taeniasis in humans. Gathered risk factors for pork-borne zoonoses includes the consumption of raw/undercooked pork, beef and vegetables and the utilisation of night-soil for fertilization of local produce.

2.6 Conclusions

Although there are detailed data available on the prevalence of food-borne parasitic zoonoses relating to taeniasis, cysticercosis and trichinellosis in the north of Vietnam, by contrast, little to no data are available for the central and southern areas of the country. The utilisation of copro-diagnostic tests for human taeniasis and post-mortem

89 diagnosis of porcine T. solium cysticercosis likely resulted in an underestimation of the TP of infection in these hosts. On the other hand, genus-specific immunodiagnostic tests with imperfect Sp likely resulted in an overestimation of human and porcine cysticercosis. Moreover, sampling methods that were conducted without a clearly defined randomized design limits our ability to make accurate estimates of the TP of these infections in either humans or pigs in each of the three regions of Vietnam. In addition, studies based on slaughterhouse surveillance are believed not to reflect the true risk posed to humans, given the majority of the population live in rural, remote communities.

Future surveillance aimed at conducting random cross-sectional village-based surveys of taeniasis and T. solium cysticercosis in humans and cysticercosis in pigs and cattle using improved molecular and immunodiagnostic methods, will shed further light on the epidemiology of these food-borne zoonoses among rural communities. This information will assist local government and residents to develop appropriate risk mitigation efforts to reduce the burden of these infections for the betterment of market access and public health.

2.7 References

[1] FAO/WHO. Multicriterial-based Ranking for Risk Management of Food-borne Parasites. Microbiologycal Risk Assessment Series No 23. vol. 23. Rome: WHO press; 2014.

[2] Torgerson PR, Devleesschauwer B, Praet N, Speybroeck N, Willingham AL, Kasuga F, et al. World Health Organization estimates of the global and regional disease burden of 11 foodborne parasitic diseases, 2010: A data Synthesis. PLoS Med 2015;12:e1001920.

[3] Murrell KD, Dorny P, Flisser A, Geerts S, Kyvsgaard NC, McManus DP, et al. WHO/FAO/OIE Guidelines for the Surveillance, Prevention and Control of Taeniosis/Cysticercosis. Paris: OIE, WHO and FAO; 2005.

[4] Schantz PM, Cruz M, Sarti E, Pawlowski ZS. Potential eradicability of taeniasis and cysticercosis. Bull PAHO 1993;27:397–403.

[5] WHO. Atlas: Epilepsy Care in the World. Geneva: WHO press; 2005.

[6] Pozio E. World distribution of Trichinella spp. infections in animals and humans. Vet Parasitol 2007;149:3–21.

[7] Bowles J, McManus DP. Genetic characterization of the Asian Taenia, a newly described taeniid cestode of humans. Am J Trop Med Hyg 1994;50:33–44.

[8] Van De N, Le TH, Lien PTH, Eom KS. Current status of taeniasis and cysticercosis in Vietnam. Korean J Parasitol 2014;52:125–9.

[9] Ito A, Nakao M, Wandra T. Human taeniasis and cysticercosis in Asia. Lancet 2003;362:1918–20.

[10] Eom KS, Jeon HK, Rim HJ. Geographical distribution of Taenia asiatica and related species. Korean J Parasitol 2009;47:S115–24.

[11] Gonzales I, Rivera JT, Garcia HH, Cysticercosis Working Group in Peru. Pathogenesis of Taenia solium taeniasis and cysticercosis. Parasite Immunol 2016;38:136–46.

[12] Carabin H, Ndimubanzi PC, Budke CM, Nguyen H, Qian Y, Cowan LD, et al. Clinical manifestations associated with neurocysticercosis: a systematic review. PLoS Negl Trop Dis 2011;5:e1152.

[13] White AC. Neurocysticercosis: updates on epidemiology, pathogenesis, diagnosis, and management. Annu Rev Med 2000;51:187–206.

[14] Pozio E, Hoberg E, La Rosa G, Zarlenga DS. Molecular taxonomy, phylogeny and biogeography of nematodes belonging to the Trichinella genus. Infect Genet Evol 2009;9:606–16.

[15] Pozio E. Taxonomy of Trichinella and the epidemiology of infection in the Southeast Asia and Australian regions. Southeast Asian J Trop Med Public Health 2001;32:129–32.

[16] Takahashi Y, Mingyuan L, Waikagul J. Epidemiology of trichinellosis in Asia and the Pacific Rim. Vet Parasitol 2000;93:227–39.

[17] Robertson LJ, Sprong H, Ortega Y, van der Giessen JWB, Fayer R. Response to galán-puchades and fuentes: Taenia asiatica: Neglected - but not forgotten - and almost certainly being quietly globalised. Trends Parasitol 2014;30:56–7.

[18] Wu HW, Ito A, Ai L, Zhou XN, Acosta LP, Lee Willingham A. Cysticercosis/taeniasis endemicity in southeast Asia: Current status and control measures. Acta Trop 2015;165:121–32.

[19] Hoang Nam. Vietnam image of the community 54 Ethic Groups. Aff Comm Ethn Minor 2016. http://english.ubdt.gov.vn/vietnam-image-of-the-commtnity-of-54- ethnic-groups.htm (accessed September 16, 2016).

[20] General Statistics Office of Vietnam. Statistical Data. Gen Stat Off Vietnam 2017. http://www.gso.gov.vn/ (accessed December 22, 2015).

[21] United Nations. Human Development Index and its components. United Nations Dev Program 2014. http://hdr.undp.org/en/composite/HDI (accessed December 23, 2015).

[22] World Bank. World Development Indicators. World Bank 2014. http://databank.worldbank.org/data/reports.aspx?source=world-development- indicators (accessed December 23, 2015).

[23] Carrique-Mas JJ, Bryant JE. A review of foodborne bacterial and parasitic zoonoses in Vietnam. Ecohealth 2013;10:465–89.

[24] Uga S, Hoa NT, Noda S, Moji K, Cong L, Aoki Y, et al. Parasite egg contamination of vegetables from a suburban market in Hanoi, Vietnam. Nepal Med Coll J 2009;11:75–8.

[25] Somers R, Dorny P, Nguyen VK, Dang TCT, Goddeeris B, Craig PS, et al. Taenia solium taeniasis and cysticercosis in three communities in north Vietnam. Trop Med Int Heal 2006;11:65–72.

[26] Moher D, Liberati A, Tetzlaff J, Altman DG, Oxman AD, Cook DJ, et al. Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. J Clin Epidemiol 2009;62:1006–12.

[27] Rogan W, Gladen B. Estimating prevalence from results of a screening test. American Journal of Epidemiology. Am J Epidemiol 1978;107:71–6.

[28] Messam L, Branscum A, Collins M, Gardner I. Frequentist and Bayesian approaches to prevalence estimation using examples from Johne’s disease. Anim Heal Res Rev 2008;9:1–23.

[29] Branscum AJ, Gardner IA, Johnson WO. Bayesian modeling of animal- and herd-level prevalences. Prev Vet Med 2004;66:101–12.

[30] Lunn DJ, Thomas A, Best N, Spiegelhalter D. WinBUGS - A Bayesian modelling framework: Concepts, structure, and extensibility. Stat Comput 2000;10:325–37.

[31] Wakefield JC, Best NG, Waller L. Bayesian approaches to disease mapping. In: Elliott P, Wakefield J, Best N, Briggs D, editors. Spat. Epidemiol. Methods Appl., Oxford University Press, London; 2000, p. 118.

[32] NIMPE. Taeniasis and Cysticercosis. Rev. Work. helminthiasis Control Act. period 2006-2011 Implement. Work. period 2012-2015, Ministry of Health of Vietnam; 2012.

[33] Doanh NQ, Holland W, Tam PT, De N Van, Hoa LT, Hoan DH. The results of the studies on situation of humans and pigs contracting disease that cestode worm and its larvae cause for. Sci Technol J Agric Rural Dev 2006;4:56–8.

[34] Vien HV, Hung NM, Thach DTC, Thuan LK, Dung DT, Hop NT, et al. Investigate Taenia infection in two communes of Tan Son district, Phu Tho province. J Malar Parasite Dis Control 2012;1:69–74.

[35] Huong NT. Taeniasis and Cysticercosis in a Selected Group of Inhabitants from a Mountainous Province in North Vietnam. Prince Leopold Institute of Tropical Medicine, Antwerpen (Antwerp), Belgium, 2006.

[36] Phuong VTB, Tra HTU, Duyen NT. Research on intestinal parasitic infection on patients in parasitology department in the Hospital of Thai Binh Medical University from 2008 -2010. Y Hoc Thanh Pho Ho Chi Minh 2012;16:7–10.

[37] Van Chuong N. Status of helminthic infection in humans in 14 provinces spanning the Central and Central Highland regions of Vietnam. IMPE-QN 2005. http://www.impe-qn.org.vn/impe-qn/vn/portal/InfoDetail.jsp?area=58&cat=1068& ID=597 (accessed March 1, 2016).

[38] Chuong N Van, Van Tuan B, Kha N Van, Khanh NH. Research on Taenia saginata Infection situation in some areas of Kon Tum province. J Paractical Med 2011;796:156–9.

[39] Van Tuan B, Chuong N Van. Situation of Taenia saginata infestation and effectiveness of some control measures in Dak Mon commune (Dak Glei) and Yaxier commune (Sa Thay), Kon Tum province. J Malar Parasite Dis Control 2014;1:41–8.

[40] Sieu TPM, Ly LTC. Study of some characteristics of taeniasis (Taenia saginata) in Ho Chi Minh city. Y Hoc Thanh Pho Ho Chi Minh 2013;17:126–9.

[41] Ho ST. Study on genotype of pathogen, clinical, subclinical symptoms, treatment efficacy for taeniasis and cysticercosis patients in National Institute of Malariology, Parasitology and Entomology 2007-2010 2013:143. http://luanan.nlv.gov.vn/luanan?a=d&d=TTcFqWrCyuSa2013.1.4&srpos=1&e=------vi-20--1--img-txIN-H%25e1%25bb%2593+S%25e1%25bb%25b9+Tri%25 e1%25bb%2581u------# (accessed March 5, 2016).

[42] Vien HV, Le DD, Manh N., Tan H V., Nguyen DH, Nhung VT. Identication of tapeworm Taenia spp. and cysticercosis in human by Multiplex-PCR. J Malar Parasite Dis Control 2008;1:62–8.

[43] De N V., Hoa LT, Doanh NQ, Ngoc NB, Cong LD. Report on a new species of Taenia (Taenia asiatica) in Hanoi, Vietnam. J Malar Parasite Dis Control 2001;3:80–5.

[44] Willingham AL, Van De N, Doanh NQ, Cong D, Dung T V, Dorny P, et al. Current status of cysticercosis in Vietnam. Southeast Asian J Trop Med Public Health 2003;34:35–50.

[45] Somers R, Dorny P, Geysen D, Nguyen LA, Thach DC, Vercruysse J, et al. Human tapeworms in north Vietnam. Trans R Soc Trop Med Hyg 2007;101:275–7.

[46] De N V., Hoa LT. Molecular Identification of Human Taenia saginata and Taenia asiatica in Vietnam. Vietnam J Med 2006;1:42–50.

[47] Doanh NQ, Kim NT, De N V., Lung NL. Result of survey on taeniasis and cysticercosis humans and pigs in Bac Ninh and Bac Kan provinces. Vet Sci Tech 2002;9:46–9.

[48] Erhart A, Dorny P, Van De N, Vien H V, Thach DC, Toan ND, et al. Taenia solium cysticercosis in a village in northern Viet Nam: Sero-prevalence study using an ELISA for detecting circulating antigen. Trans R Soc Trop Med Hyg 2002;96:270–2.

[49] Anh Tuan P, Dung TTK, Nhi VA. Sero-epidemiological investigation of cysticercosis in the southern provinces. J Malar Parasite Dis Control 2001;4:81– 7.

[50] Trung PT, Doan B Van. Neurocysticercosis: Report of three cases in infectious disease department of Hue Central Hospital. Y Hoc Thanh Pho Ho Chi Minh 2013;17:110–4.

[51] Khue P Van, Luc P. Veterinary Parasitology. Hanoi: Hanoi Agricultural Publishing House; 1996.

[52] Doanh NQ, Holland W, Vercruysse J, De N Van. Results of survey on cysticercosis pig in some nothern provinces in Vietnam. J Malar Parasite Dis Control 2002;6:76–82.

[53] De N Van, Chau L Van, Son DT, Chuyen LT, Hop NT, Vien HV, et al. Taenia solium survey in Hanoi. J Malar Parasite Dis Control 2004;6:93–9.

[54] Huan L Van. Parasitic Helminths in Pigs in Several Southern Provinces and Preventative Measures. National Institution of Veterinary Research, 1994.

[55] Phan L. Parasite and Veterinary Parasitic Diseases. Hanoi: Hanoi Agriculture Publishing House; 1997.

[56] Van De N, Nga VT, Dorny P, Trung NV, Minh PN, Dung DT, et al. Trichinellosis in Vietnam 2015;92:1265–70.

[57] Nguyen Van C, Vu Thi N, Nguyen Trong C. Result of survey and solution for trichinellosis in pigs in Bac Ha district, Son La province. Vet Sci Tech 2012;19:50–6.

[58] Van De N, Trung N V, Ha NH, Nga VT, Ha NM, Thuy PT, et al. An outbreak of trichinosis with molecular identification of Trichinella sp. in Vietnam. Korean J Parasitol 2012;50:339–43.

[59] Vu Thi N, Dorny P, La Rosa G, To Long T, Nguyen Van C, Pozio E. High prevalence of anti-Trichinella IgG in domestic pigs of the Son La province, Vietnam. Vet Parasitol 2010;168:136–40.

[60] Thi N V, De N V, Praet N, Claes L, Gabriël S, Dorny P. Seroprevalence of trichinellosis in domestic animals in northwestern Vietnam. Vet Parasitol 2013;193:200–5.

[61] Vu Thi N, Nguyen VD, Praet N, Claes L, Gabriël S, Huyen NT, et al. Trichinella infection in wild boars and synanthropic rats in northwest Vietnam. Vet Parasitol 2014;200:207–11.

[62] Vu Thi N, Trung DD, Litzroth A, Praet N, Nguyen Thu H, Nguyen Thu H, et al. The hidden burden of trichinellosis in Vietnam: a postoutbreak epidemiological study. Biomed Res Int 2013;2013:1–4.

[63] Dorny P, Phiri IK, Vercruysse J, Gabriel S, Willingham IAL, Brandt J, et al. A Bayesian approach for estimating values for prevalence and diagnostic test characteristics of porcine cysticercosis. Int J Parasitol 2004;34:569–76.

[64] Smyth JD, McManus DP. The Physiology and Biochemistry of Cestodes. New York: Cambridge University Press; 2007.

[65] Rodriguez S, Wilkins P, Dorny P. Immunological and molecular diagnosis of cysticercosis. Pathog Glob Health 2012;106:286–98.

[66] Diaz JF, Verastegui M, Gilman RH, Tsang VC, Pilcher JB, Gallo C, et al. Immunodiagnosis of human cysticercosis (Taenia solium): a field comparison of an antibody-enzyme-linked immunosorbent assay (ELISA), an antigen-ELISA, and an enzyme-linked immunoelectrotransfer blot (EITB) assay in Peru. The Cysticercosis Working Group in. Am J Trop Med Hyg 1992;46:610–5.

[67] Deckers N, Dorny P. Immunodiagnosis of Taenia solium taeniosis/cysticercosis. Trends Parasitol 2010;26:137–44.

[68] Garcia HH, Gonzalez AE, Gilman RH, Palacios LG, Jimenez I, Rodriguez S, et al. Short report: transient antibody response in Taenia solium infection in field conditions-a major contributor to high seroprevalence. Am J Trop Med Hyg 2001;65:31–2.

[69] Garcia HH, Gilman RH, Catacora M, Verastegui M, Gonzalez AE, Tsang VCW. Serologic evolution of neurocysticercosis patients after antiparasitic therapy. J Infect Dis 1997;175:486.

[70] Coral-Almeida M, Gabriël S, Abatih EN, Praet N, Benitez W, Dorny P. Taenia solium human cysticercosis: A systematic review of sero-epidemiological data from endemic zones around the world. PLoS Negl Trop Dis 2015;9:e0003919.

[71] Dorny P, Brandt J, Zoli A, Geerts S. Immunodiagnostic tools for human and porcine cysticercosis. Acta Trop 2003;87:79–86.

[72] Verle P, Kongs A, De N V, Thieu NQ, Depraetere K, Kim HT, et al. Prevalence of intestinal parasitic infections in northern Vietnam. Trop Med Int Heal 2003;8:961–4.

[73] Praet N, Verweij JJ, Mwape KE, Phiri IK, Muma JB, Zulu G, et al. Bayesian modelling to estimate the test characteristics of coprology, coproantigen ELISA and a novel real-time PCR for the diagnosis of taeniasis. Trop Med Int Heal 2013;18:608–14.

[74] Taylor WR, Nguyen K, Nguyen D, Nguyen H, Horby P, Nguyen HL, et al. The spectrum of central nervous system infections in an adult referral hospital in Hanoi, Vietnam. PLoS One 2012;7:e42099.

[75] Trung DD, Praet N, Cam TDT, Lam BVT, Manh HN, Gabriël S, et al. Assessing the burden of human cysticercosis in Vietnam. Trop Med Int Heal 2013;18:352–6.

[76] Luong VH. Parasitic Helminths in Pigs in Several Southern Provinces and Preventative Measures. National Instituion of Veterinary Research, 1994.

[77] Taylor WRJ, Van Tran G, Nguyen TQ, Van Dang D, Nguyen VK, Nguyen CT, et al. Acute febrile myalgia in Vietnam due to trichinellosis following the consumption of raw pork. Clin Infect Dis 2009;49:e79–83.

Table 2.1 Research on human taeniasis and estimated true prevalence in Vietnam

Author Research Research location Regional location Diagnosis Sample Apparent Prior information True prevalence period technique size prevalence (%) [95% CrI] b a (no. of (%) [95% CI] Referred diagnosis Sensitivity Specificity people) technique (%) [95% CI] (%) [95% CI] Verle et al. [72] 1999 Hoa Binh North Formalin-Ether 2,522 0.11 Microscopy-based c 52.5 99.9 0.29 Concentration [0.04 to 0.34] [11.1 to 96.5] [99.5 to 100] [0.00 to 5.38] Doanh et al. [47] 1999 - 2001 Bac Ninh North Faecal examination 597 7.20 9.92 [5.39 to 9.56] [5.06 to 19.29] Somers et al. [45] 2003 - 2004 Bac Kan, Hai Duong, North, cAg-ELISA 606 0.66 cAg-ELISA d 84.5 92.0 0.25 Ha Tinh North Central [0.25 to 1.68] [61.9 to 98] [90 to 93.8] [0.01 to 0.21] Chuong [37] na Binh Thuan, Ninh Thuan, Central, Faecal examination 35,651 0.33 Microscopy-based c 52.5 99.9 0.52 Khanh Hoa, Phu Yen, Central highlands [0.27 to 0.39] [11.1 to 96.5] [99.5 to 100] [0.03 to 6.41] Binh Dinh, Quang Ngai, Quang Nam, Da Nang, Dak Lak, Quang Tri, Quang Binh, Gia Lai, Kon Tum, Thua Thien Hue NIMPE [32] 2006 Phu Tho, Vinh Long, North, Central, Faecal examination 9,547 0.96 1.49 Hung Yen, Cao Bang, South [0.78 to 1.18] [0.14 to 10.79] Ha Tay, Lang Son, Tuyen Quang, Da Nang, Lao Cai, Thanh Hoa, Vinh Phuc, Quang Ngai Huong [35] 2006 Ha Giang North Kato 84 6.00 8.43 [2.56 to 13.18] [2.70 to 19.53] Phuong al et. [36] 2008 - 2010 Thai Binh North Faecal examination 6,570 0.17 0.33 [0.09 to 0.29] [0.17 to 5.07] Chuong et al. [38] 2009 Kon Tum Central Highlands Kato-Katz 1,797 8.18 11.08 [7.00 to 9.53] [6.59 to 20.68] Vien et al. [34] 2009 - 2010 Phu Tho North Kato-Katz 775 8.65 11.67 [6.86 to 10.83] [6.96 to 20.85] Van Tuan 2013 Kon Tum Central Highlands Kato-Katz 731 10.40 13.46 et al. [39] [8.38 to 12.82] [8.42 to 21.44] Abbreviation: na, not applicable; NIMPE, National Institute of Malariology Parasitology and Entomology a Confidence interval b Credible interval c Sensitivity and specificity of microscopy-based technique based on Praet et al. (2013) [73] d Sensitivity and specificity of cAg-ELISAbased on Praet et al. (2013) [73]

100

Table 2.2 Hospital and community-based survey on human cysticercosis and estimated true prevalence of cysticercosis in Vietnam

Reference Research Research location Regional Diagnosis technique Participant Apparent prevalence Prior information True prevalence period location or patient (n) (%) [95% CI] a (%) [95% CrI] b

Referred Sensitivity Specificity diagnosis (%) [95% CI] (%) [95% CI] technique Verle et al. [72] 1999 Hoa Binh North Biopsy of 2,522 1 c na subcutaneous nodules NIMPE [32] 2000–2011 NIMPE National na 2,687 d na na Taylor et al. [74] 2007–2008 National hospital North Cranial radiology 352 0.3 na Anh Tuan et al. [49] 1992–2000 Ho Chi Minh South Ab-ELISA 3,814 4.30 Ab-ELISA e 65.0 63.0 0.89 hospitals [3.70 to 4.99] [0.16 to 2.53] Erhart et al. [48] 1999 Bac Ninh North Ag-ELISA 210 5.71 3.99 [3.29 to 9.72] [1.24 to 8.27] Doanh et al. [47] 1999–2000 Bac Ninh North Ag-ELISA 597 5.02 2.90 [3.54 to 7.08] [0.78 to 5.95] Somers et al. [45] 2003–2004 Bac Kan, Hai Duong, North Ag-ELISA 707 2.40 0.86 Ha Tinh [1.50 to 3.81] [0.10 to 2.67] Huong [35] 2005 Ha Giang North Ag-ELISA 97 13.40 Ag-ELISA f 87.0 95.0 13.27 [0.80 to 21.58] [62 to 98] [90 to 99] [11.08 to 15.76] Trung et al. [75] 2007–2010 Bac Giang, Bac North Ag-ELISA 758 g 6.00 3.74 Ninh, Dien Bien, [5.00 to 8.00] [1.26 to 7.22] Ha Giang,Lai Chau, Lang Son, Tuyen Quang

Abbreviation: na, not applicable a Confidence interval b Credible interval c Case of cysticercosis d Numbers of patients e Sensitivity and specificity of Ab-ELISA based on Diaz et al. (1992) [66] f Sensitivity and specificity of Ag-ELISA based on Coral-Almeida et al. (2015) [70] g Participants have chronic headache and epilepsy

101

Table 2.3 Research on porcine cysticercosis and estimated true prevalence in Vietnam

Author Research Research location Diagnosis Sample Apparent prevalence Prior information True prevalence period technique size (%) [95% CI] a (%) [95% CrI] b (no. of pigs) Referred Sensitivity Specificity diagnosis (%) [95% CI] (%) [95% CI] technique Khue & Luc [51] na Nam Dinh, Ha Nam, Carcass 8,000 0.00 Carcass 22.1 100 na Hai Duong, Hung Yen examination examination c [15 to 27] Doanh et al. [52] 1999–2001 Yen Bai, Lao Cai, Carcass 198,877 0.06 0.14 Nghe An, Bac Kan, examination [0.0 to 0.34] Bac Giang, Hanoi Doanh et al. [47] 1999–2000 Bac Ninh, Bac Kan Ag-ELISA 323 9.91 Ag-ELISA d 86.7 94.7 9.64 [7.10 to 13.65] [62 to 98] [90 to 99.7] [8.06 to 11.43]

De et al. [53] 2002–2003 Hanoi Carcass 143,868 2e Carcass 22.1 100 na examination examination c [15 to 27]

Huan [76] 1994 12 southern provinces Carcass 891 0.90 1.92 examination [0.45 to 1.76] [0.18 to 5.96]

Abbreviation: na, not applicable a Confidence interval b Credible interval c Sensitivity and specificity of carcass examination and Ag-ELISA based on Dorny et al. (2004) [63] d Sensitivity and specificity of Ag-ELISA based on Dorny et al. (2004) [63] e Cases of porcine cysticercosis

102

Table 2.4 Trichinellosis outbreaks in Vietnam since 1970

Outbreak (year) Location (province) Geographical location Trichinella spp. Infected toll (no. of people) Death toll (no. of people) 1970 Yen Bai Northwest na 26 4 2001 Dien Bien Northwest na 22 2 2004 Dien Bien Northwest na 20 na 2008 Son La Northwest T. spiralis (animals) 22 2 2012 Thanh Hoa North Central Coast T. spiralis (humans) 24 na

Abbreviation: na, not applicable Adjusted from: De et al. (2012) [58]; Vu Thi et al. (2013, 2014) [61, 62]; Taylor et al. (2009) [77]

103

Fig. 2.1 Flow diagram of searching strategy. Diagram showing the strategy steps of searching and justification for taeniasis, cysticercosis and trichinellosis in Vietnam.

104

Fig. 2.2 Studies of the prevalence of taeniasis in Vietnam, 1999 to present. a Map of Vietnam showing the location of studies described in the text. b Error bar plot showing the known true prevalence of taeniasis (and their 95% confidence intervals) as a function of the northing coordinate of the province in which the study was carried out.

105

Fig. 2.3 Studies of the prevalence of human T. solium cysticercosis in Vietnam, 1992 to present. a Map of Vietnam showing the location of studies described in the text. b Error bar plot showing the known true prevalence of T. solium cysticercosis in humans (and their 95% confidence intervals) as a function of the northing coordinate of the province in which the study was carried out.

106

Fig. 2.4 Studies of the prevalence of T. solium cysticercosis in pigs in Vietnam, 1994 to present. a Map of Vietnam showing the location of studies described in the text. b Error bar plot showing the known true prevalence of T. solium cysticercosis pigs (and their 95% confidence intervals) as a function of the northing coordinate of the province in which the study was carried out.

107

Fig. 2.5 Map of trichinellosis outbreaks in Vietnam. Choropleth map showing provinces in which trichinellosis outbreaks have occurred since 1970.

108

Fig. 2.6 Distribution of human T. solium cysticercosis and taeniasis. Choropleth map of Vietnam showing provinces in which cysticercosis and/or taeniasis have been identified in humans at present.

109

CHAPTER 3

Comparison of a new multiplex real-time PCR with the Kato Katz thick smear and copro-antigen ELISA for the detection and differentiation of Taenia spp. in human stools

Presented as published:

Ng-Nguyen D, Stevenson MA, Dorny P, Gabriël S, Vo T Van, Nguyen VT, et al. Comparison of a new multiplex real-time PCR with the Kato Katz thick smear and copro-antigen ELISA for the detection and differentiation of Taenia spp. in human stools. PLoS Negl Trop Dis 2017;11:e0005743.

110

3.1 Abstract

3.1.1 Background

Taenia solium, the cause of NCC, has significant socioeconomic impacts on communities in developing countries. This disease, along with taeniasis is estimated to infect 2.5 to 5 million people globally. Control of T. solium NCC necessitates accurate diagnosis and treatment of T. solium taeniasis carriers. In areas where all three species of Taenia tapeworms (T. solium, Taenia saginata and Taenia asiatica) occur sympatrically, conventional microscope- and copro-antigen based diagnostic methods are unable to distinguish between these three Taenia species. Molecular diagnostic tools have been developed to overcome this limitation; however, conventional PCR-based techniques remain unsuitable for large-scale deployment in community-based surveys. Moreover, a real-time PCR (qPCR) for the discrimination of all three species of Taenia in human stool does not exist. This study describes the development and validation of a new triplex Taq-Man probe-based qPCR for the detection and discrimination of all three Taenia human tapeworms in human stools collected from communities in the Central Highlands of Vietnam. The diagnostic characteristics of the test are compared with conventional Kato Katz (KK) thick smear and cAg-ELISA method utilising faecal samples from a community based cross-sectional study. Using this new multiplex qPCR we provide an estimate of the TP of taeniasis in the source population for the community based cross-sectional study.

3.1.2 Methodology/Principal findings

Primers and TaqMan probes for the specific amplification of T. solium, T. saginata and T. asiatica were designed and successfully optimized to target the internal transcribed spacer I (ITS-1) gene of T. solium and the cytochrome oxidase subunit I (COX-1) gene of T. saginata and T. asiatica. The newly designed triplex qPCR (T3qPCR) was compared to KK and cAg-ELISA for the detection of Taenia eggs in stool samples collected from 342 individuals in Dak Lak province, Central Highlands of Vietnam. The overall AP of taeniasis in Dak Lak province was 6.7% (95% confidence interval (CI) 4.4 to 10) in which T. solium accounted for 1.17% (95% CI 0.37 to 3.17), according to the T3qPCR. There was sympatric presence of T. solium, T. saginata and T. asiatica. 111

The T3qPCR proved superior to KK and cAg-ELISA for the detection and differentiation of Taenia species in human faeces. Diagnostic sensitivities of 0.94 (95% credible interval (CrI) 0.88 to 0.98), 0.82 (95% CrI 0.58 to 0.95) and 0.52 (95% CrI 0.07 to 0.94), and diagnostic specificities of 0.98 (95% CrI 0.94 to 1.00), 0.91 (95% CrI 0.85 to 0.96) and 0.99 (95% CrI 0.96 to 1.00) were estimated for the diagnosis of taeniasis for the T3qPCR, cAg-ELISA and KK thick smear in this study, respectively.

3.1.3 Conclusions

T3qPCR is not only superior to the KK thick smear and cAg-ELISA in terms of diagnostic Se and Sp, but it also has the advantage of discriminating between species of Taenia eggs in stools.

Application of this newly developed T3qPCR has identified the existence of all three human Taenia tapeworms in Dak Lak province and proves for the first time, the existence of T. asiatica in the Central Highlands and the south of Vietnam.

3.1.4 Author summary

Human Taenia tapeworms comprise three species, Taenia solium, Taenia saginata and Taenia asiatica. Taeniasis is a meat-borne zoonosis transmitted by the consumption of cysticerci in raw or undercooked pork for T. solium and T. asiatica (liver) and in beef for T. saginata. Accidental ingestion of T. solium eggs by humans also results in the formation of cysticerci, often in the brain, referred to as NCC. T. solium NCC is a significant cause of morbidity and mortality owing to epilepsy in many resource-poor communities. In animals, ingestion of eggs passed by humans results in organ and / or carcass condemnation and suboptimal economic outcomes for farmers. The accurate diagnosis of T. solium tapeworm carriers is essential to monitor the success of control programs. In areas where all three species of Taenia tapeworms occur together, conventional diagnostic methods are unable to distinguish between the different species of Taenia. In this study, we develop and apply a T3qPCR capable of detecting and discriminating all three-tapeworm species in stools in a rapid and high-throughput fashion, suitable for large-scale community surveys. The newly developed T3qPCR

112

proved superior to previously developed immunodiagnostic and conventional microscopic-based tests in terms of diagnostic Se, Sp and the ability to identify and distinguish human Taenia species. This qPCR assay facilitated the identification of T. asiatica tapeworms in the Central Highlands of Vietnam, for the first time.

3.2 Introduction

Humans are infected with three species of Taenia - Taenia solium, Taenia saginata and Taenia asiatica [1, 2]. While T. saginata has a global distribution, T. solium is distributed mostly in developing countries of Latin America, sub-Saharan Africa and Asia, and T. asiatica is restricted to certain Asian countries [3, 4]. Taeniasis is estimated to infect 2.5 to 5 million people globally [5–7]. All three-tapeworm species utilise humans as the definitive hosts (adult tapeworm) due to the ingestion of undercooked and/or raw meat or liver [1, 8]. The intermediate hosts of T. solium and T. asiatica are swine whereas the intermediate hosts of T. saginata are cattle (porcine/bovine cysticercosis) [3]. Humans may also become the accidental intermediate hosts of T. solium via the ingestion of food and water contaminated with T. solium eggs, which may result in NCC when the cysts lodge in the central nervous system. The clinical manifestations of T. solium NCC in humans are varied, ranging from being asymptomatic to severe neurological signs and symptoms such as epilepsy, paralysis, dementia, chronic headache, blindness or even death [9–11]. The symptoms of taeniasis in humans, on the other hand, are mostly subtle and mild, and include abdominal distension, abdominal pain, digestive disorders and anal pruritus (mostly for T. saginata and T. asiatica) [12, 13].

Control of NCC necessitates accurate diagnosis and treatment of T. solium taeniasis carriers to break the cycle of transmission to pigs [14]. Although all cases of taeniasis necessitate treatment, species-specific identification from an epidemiological perspective will allow better targeted control programs to be implemented aimed at interrupting the lifecycle specific to each species of Taenia endemic within a community and region. Several diagnostic tools for detecting taeniasis, including microscopy, copro-antigen ELISAs, sero-antibody immunoblot [15–17], and copro- DNA tests have been developed[18, 19]. Of these diagnostic tools, microscopy-based

113

examinations of faecal samples have the limitation of poor diagnostic Se owing to intermittent shedding of proglottids/eggs and low Sp owing to the inability to morphologically discriminate between the eggs of Taenia species [20]. Although self- detection of proglottid shedding is valuable for confirmation of individual (but not community-wide) taeniasis, it is however, not always reliable especially when deep pit latrines are being utilised. Moreover, in cases of infection with T. solium, the proglottids are immobile and may be misidentified as other worms [19]. Differentiation of T. saginata from T. asiatica based on proglottid morphology is onerous and challenging [21]. The cAg-ELISA utilises polyclonal antigens and more sensitive (0.85, 95% CI 0.62 to 0.98) than microscopy-based methods (0.53, 95% CI 0.11 to 0.97) [22], is only capable of identification to a genus level [23–25]. More recently, a cAg-ELISA specific to T. solium was developed [20, 26]. Another drawback of the cAg-ELISA is that it may miss cases of T. asiatica and T. saginata in areas where other human Taenia tapeworms co-exist [19]. Serum antibody detecting immunoblot assays using ES or recombinant antigens have been developed for identification of antibodies to T. solium [15, 17] and T. asiatica taeniasis [16]. Immunoblot assays were highly sensitive and specific when used in Taenia non-endemic areas, however resulted in a high proportion of false positive results in areas endemic for T. saginata [17]. Copro-PCRs are considered superior tools for the detection of Taenia eggs in faecal samples owing to their high diagnostic Se and the ability to distinguish Taenia species. Currently, the only multiplex-PCR [27, 28] and PCR-RFLP [21] that is able to differentiate all human Taenia tapeworms is unsuited for large-scale community surveys owing to labor intensive process of restriction digest and gel electrophoresis. A better suited real-time PCR [29] has been developed but it is only capable of discriminating T. solium and T. saginata. Loop-mediated isothermal amplification (LAMP) [30] has also shown promise as a potential point-of-care diagnostic technique capable of highly sensitive detection (one copy of target gene/reaction or at least five eggs/gram of faeces) and differentiation of Taenia spp. in stool [30], however the technology is still far from point-of-care based and is considered equivocal to PCR [19].

This study describes the development and validation of a new multiplex Taq-Man probe-based qPCR for the detection and discrimination of all three species of Taenia in

114

human stools. The diagnostic characteristics of the test are compared with the test characteristics of conventional KK thick smear and the cAg-ELISA using hyper immune rabbit anti-Taenia IgG polyclonal antibody by applying all three diagnostic tests to detect Taenia spp. in faecal samples collected as part of a community-based cross-sectional study carried out in the Central Highlands of Vietnam.

3.3 Materials and methods

3.3.1 Multiplex real-time PCR

Primers and probes. The internal transcribed spacer I (ITS-1) gene of T. solium (GenBank accession number LC004200) was utilised to design a T. solium specific primer and Taq-Man-based hydrolysis probe for the specific amplification of T. solium. Sequences spanning the cytochrome oxidase subunit I (COX-1) gene were utilised to design a primer pair for specific amplification of T. saginata and T. asiatica concurrently, and two different species-specific Locked Nucleic Acid probes were used for detection of T. saginata, and T. asiatica. Primers and probe of equine herpesvirus 4 (EHV4) used as an internal control were based on a previously published paper [31] (Table 3.1).

Multiplex qPCR conditions. DNA of EHV4 was spiked into T3qPCR prepared reaction mixtures as an internal reaction control. The multiplex qPCR assay was run as a 20 µL reaction containing 10 µL of GoTaq Probe qPCR Master Mix (Promega

Corporation, Madison, WI, USA), 2.44 µL of H2O, 350 nM of each primer of T. solium, T. saginata and T. asiatica, 40 nM of each primer of EHV, 200 nM probe of T. solium, 250 nM each probe of T. saginata and T. asiatica, 100 nM probe of EHV, 1 µL of EHV4 DNA diluted 20 times and 2 µL of DNA template.

The T3qPCR amplification was carried out in a Magnetic Induction Cycler, MIC (Bio Molecular Systems). The cycling conditions consisted of an initial denaturation step at 95 °C for 2 minutes, followed by 40 amplification cycles, each comprising a denaturation step at 95 °C for 30 seconds and annealing at 66 °C for 60 seconds. All samples were tested in duplicate with positive and negative control samples were included in each amplification assay. 115

Multiplex qPCR controls. Synthesized gBlock gene fragments of T. solium, T. saginata and T. asiatica were used as the positive controls for the T3qPCR assays. The gBlocks Gene® Fragments (Integrated DNA Technologies (IDT), Coralville, USA) are double-stranded and sequence-verified DNA molecules that vouch for the fidelity of the T3qPCR assays. The gBlock fragments of T. solium and T. asiatica were synthetically combined together to an amplicon size of 235 bp in length while the T. saginata gene fragment was synthesized as an individual amplicon with a length of 140 bp (Table 3.2).

In addition, gDNA from proglottids sourced from individuals in this study (following treatment and recovery) belonging to T. asiatica and T. saginata as well as T. solium from eggs sourced from the Centers of Disease Control and Prevention (Atlanta, GA, US) were utilised as positive controls for this study. T. asiatica and T. saginata proglottids were morphologically identified and confirmed using a previously published PCR [27] or singleplex PCR followed by DNA sequencing.

The individual qPCR reactions were optimized by subjecting individual gBlocks for each Taenia species to serial 10-fold dilutions ranging from 5 × 10-7 to 5 × 101 pg/µL and then mixed together to assess the diagnostic Se, Sp and efficiency of the multiplex assay. EHV4 DNA templates were also included in optimization of the multiplex qPCR. Normal melt curves and absolute quantification analyses were used to determine the positive status of individual samples. Threshold and fluorescence cut-off level were set up at 0.1 and 16%, respectively, except for fluorescence cut-off level of EHV at 20%, and the first cycle of qPCR amplification was ignored from the analysis. The T3qPCR results were considered negative if cycle threshold (Ct) values were > 35. This value (5 × 10-5 pg/µL) was the limit of our standard curves (additional gBlocks dilutions were undetectable). Each sample was checked for fidelity/inhibition by comparing the Ct- value of the EHV4 in the sample compared to the EHV4 Ct-value in the negative control. An assay was deemed a failure when its EHV4 Ct-value was greater than two cycles different in comparison to the negative control EHV4 Ct-value.

The diagnostic Sp of the T3qPCR was confirmed utilising DNA templates from 12 different human intestinal parasites and one canine tapeworm, including Cryptosporidium parvum, Giardia duodenalis, Haplorchis taichui, Hymenolepis nana, 116

Opisthorchis viverrini, Ancylostoma ceylanicum, Necator americanus, Trichuris trichiura, Ascaris lumbricoides, Strongyloides stercoralis, Blastocystis hominis, Clonorchis sinensis and Taenia hydatigena. These parasitic DNA templates were sourced either from faeces or adult worms, and confirmed using conventional PCR and DNA sequencing.

Multiplex qPCR confirmation. The positive results of T3qPCR were confirmed using either a conventional multiplex PCR targeting T. solium, T. saginata and T. asiatica developed by Jeon et al. (2009) [32] or a singleplex PCR for Taenia sp. following DNA sequencing.

The singleplex PCR utilised self-designed forward 5’-CATCATATGTTTACGGTTGG- 3’ and reverse primer 5’-GACCCTAATGACATAACATAAT-3’ designed based on COX-1 gene and amplifying a gene of 350 bp. The assay comprised 2.5 µL 10×CoralLoad PCR Buffer (Qiagen, Hiden, Germany), 12.5 pmol of each primer, 0.5 U of HotStar Taq DNA Polymerase, 2 µL dNTPs (4 mM), 0.5 µL of MgCl2 (25 mM) and 1 µL of DNA, in a total volume of 25 µL. The cycling conditions consisted of an initial denaturation step at 95 °C for 5 minutes, 52 °C for 1 minute and 72 °C for 2 minutes followed by 40 amplification cycles, each comprising a denaturation step at 95 °C for 30 seconds, annealing at 52 °C for 30 seconds and extension at 72 oC for 30 seconds and final extension step at 72 °C for 4 minutes. The singleplex PCR products were sent to Macrogen Inc. (Korea) for sequencing.

The multiplex PCR assay comprised 2.5 µL 10×CoralLoad PCR Buffer (Qiagen, Hiden, Germany), 12.5 pmol of each forward primer, 37.5 pmol reverse primer, 0.5 U of

HotStar Taq DNA Polymerase, 2 µL dNTPs (4 mM), 0.5 µL of MgCl2 (25 mM) and 1 µL of DNA, in a total volume of 25 µL. The cycling conditions consisted of an initial denaturation step at 95 °C for 5 minutes, 56 °C for 1 minute and 72 °C for 2 minutes followed by 45 amplification cycles, each comprising a denaturation step at 95 °C for 30 seconds, annealing at 56 °C for 30 seconds and extension at 72 °C for 30 seconds and final extension step at 72 °C for 4 minutes.

117

DNA extraction of faecal samples. Samples stored in 5% potassium dichromate (w/v) were washed three times in distilled water and then subjected to DNA extraction. Approximately 250 mg washed stool was subsequently extracted using the PowerSoil DNA Isolation Kit® (Mo Bio, Carlsbad, CA, USA). The protocol of extraction was in accordance with the manufacturer’s instructions, with the exception that provided beads were replaced with 1 g of Silica/Zirconia 0.5 mm beads (Daintree Scientific, Tasmania, Australia). DNA was eluted in a final volume of 100 µL of elution buffer. The quantity and quality of the extracted DNA was quantified using a NanoDrop® ND-1000 spectrophotometer (Thermo Scientific, Wilmington, DE) and results were analyzed using the NanoDrop 1000 software. The DNA extracted was determined suitable when the yield was >= 2 ng/µL and the ratio of the absorbance at 260 and 280 nm and at 260 and 230 nm were > 1.70 and 2.0 - 2.2, respectively. The extracted DNA was stored at −20 °C until use. All samples were subjected to the multiplex qPCR for the detection and differentiation of Taenia spp. as described earlier.

DNA extraction of proglottids. Proglottids were extracted using the DNeasy Blood & Tissue Kit® (Qiagen, Hilden, Germany) according to the manufacturer’s instructions and eluted in 200 µL of AE Buffer.

3.3.2 Coproantigen enzyme-linked-immunosorbent assay

An in-house coproantigen detection ELISA [33] with slight modifications, as described by Mwape et al. (1990) [34], was performed on the stool samples. Briefly, a mixture of an equal amount of Phosphate Buffered Saline (PBS) and stool sample stored in 10% formalin was soaked for one hour with slight shaking and centrifuged at 2000 g for 30 minutes where after supernatant was collected. The Polystyrene ELISA plates (Nunc® Maxisorp, Waltham, MA, USA) were coated with capturing hyper immune rabbit anti- Taenia IgG polyclonal antibody diluted at 2.5 mg/mL in carbonate-bicarbonate buffer (0.06 M, pH 9.6) and incubated for 1 hour at 37 °C. The plates were washed once with PBS in 0.05% Tween 20 (PBS-T20). All wells were blocked by adding PBS-T20+2% New Born Calf Serum and incubated for 1 hour at 37 °C. After washing the stool supernatant of 100 mL was added to all wells and incubated for 1 hour at 37 °C followed by washing five times with PBS-T20. One hundred microliter of biotinylated 118

hyper immune rabbit IgG polyclonal antibody diluted at 2.5 mg/mL in blocking buffer was added into each well. Between two cycles of incubating for 1 hour at 37 °C followed by washing five times, 100 µL of streptavidin-horseradish peroxidase (Jackson Immunoresearch Lab, West Grove, PA, USA) diluted at 1/10,000 was added. Finally, 100 mL of ortho phenylenediamine (OPD) substrate, prepared by dissolving one tablet in 6 mL of distilled water and adding 2.5 mL of hydrogen peroxide, was added to all the wells, and incubated in a dark for 15 minutes at room temperature before stopping the reaction by adding 50 mL of sulphuric acid (4 N) to each well. The plates were read using an automated spectrophotometer at 490 nm with a reference of 655 nm. To determine the test result, the optical density (OD) of each stool sample was compared with the mean of a series of eight reference Taenia negative stool samples plus 3 standard deviations.

3.3.3 Microscopy examination

All faecal samples were microscopically examined for the presence of Taenia eggs using a duplicate KK thick-smear technique [35–38] and examined by trained parasitologists at Tay Nguyen University.

3.3.4 Study site and sampling

The fieldwork was conducted between May to October 2015 in Krong Nang, M’Drak and Buon Don districts in Dak Lak province, Vietnam (Fig. 3.1). M’ Drak is located in the east of Dak Lak province with an average altitude of 400 m to 500 m above sea level and a tropical monsoon climate typical of the Vietnamese Central Coast. Krong Nang is situated to the north of Dak Lak at 800 m above sea level. Buon Don, situated to the west of Dak Lak at approximately 330 m above sea level and had a hot and dry climate. Living standards in each of three districts is poor; open defaecation using outdoor latrines is a common practice and livestock access to these latrine areas is, for the most part, unrestricted. Moreover, consumption of raw or undercooked blood, meat, organ tissues from domesticated and wild pigs and cattle is common. The practice of non- confinement of pigs and cattle is common in rural regions with slaughter activities commonly carried out in backyards without official meat inspection. Meat inspection is

119

only carried out at abattoirs or slaughter-points, which serve the district level and/or clusters of large villages [32].

The objective of this study was to estimate the TP of human taeniasis in each of the three study districts. Based on the previous study of Van De at al. [13], the TP of human taeniasis was assumed to be 7% and there was 95% certainty that this estimate was within 5% of the true population value (i.e. the TP ranged from 2% to 12%). Ignoring the tendency for taeniasis to cluster within households we estimated that a total of 100 stool samples were required. We then assumed the average number of individuals eligible for the study per household was at least three and the between household (cluster) variance in taeniasis prevalence was around 1.14 times greater than the within household (cluster) variance, returning an intra-class correlation coefficient of 0.07 [39]. Our revised sample size, accounting for the tendency of taeniasis to cluster within households, was 114 for each district and 342 for the whole province.

Within each district, two villages were selected at random. Eighteen to forty households were randomly sampled from within each of the selected villages. Within each household a maximum of five individuals over the age of seven years were selected at random and asked if they would like to take part in the study. Those that consented were then asked to provide informed written consent for stool donation. Participants under the age of 18 years were recruited with assent of themselves as well as written consent from their parents or legal guardians. During the participant recruitment phase investigators informed study participants that they could withdraw from the study at any stage.

A labelled stool container was delivered to each individual. Faecal samples collected were divided into three aliquots. Approximately 5 g of each faecal sample was fixed in 5% potassium dichromate (w/v), and transported to the University of Melbourne, Australia for molecular analysis. The second and the third aliquots were stored in 5% formalin, shipped to the Institute of Tropical Medicine, Belgium, and kept at Tay Nguyen University, Vietnam, for cAg-ELISA testing and microscopy examination (KK), respectively.

120

3.3.5 Ethics

This study was reviewed and approved by the Behavioural and Social Sciences Human Ethics Sub-committee, the University of Melbourne (reference number 1443512) and conducted under the supervision of the local Center for Public Health, Dak Lak, Vietnam. Participants positive for taeniasis on KK thick smear were treated immediately by local medical officers with praziquantel (Distocide®, Shinpoong Daewoo Pharma CO.LTD) at 20 mg/kg as a single dose, whereas those positive by cAg-ELISA and qPCR were treated once results were known.

Production of polyclonal antibodies in rabbits used for cAg-ELISA were obtained from the Ethical Committee for animal experiments of the Institute of Tropical Medicine, Antwerp (Reference number DG012S).

3.3.6 Statistical analysis

Statistical analyses were carried out using R [40], Microsoft Excel 2007 and Microsoft Access 2007. Cohen’s kappa test statistic (K) was used for comparison of T3qPCR to KK, and T3qPCR to cAg-ELISA.

The diagnostic Se and Sp of T3qPCR, cAg-ELISA and KK were estimated using a Bayesian approach. The Bayesian approach utilised prior information based on both previous literature [22] and expert opinion (Table 3.3). Prior information for the diagnostic Se and Sp for the T3qPCR was elicited from six independent molecular parasitology diagnosticians. Prior information for the Se and Sp of cAg-ELISA and KK derived from Praet et al. (2013) [22] (Table 3.3). Prior information for the TP of taeniasis in the three study districts was obtained from previous studies carried out in the same region [41].

The model of estimation of diagnostic Se and Sp for three tests applied to individuals from the same population with no gold standard test was based on Branscum et al. (2005) [42] with minor modification of the assumption that the three tests were independent [43]. The posterior distribution of diagnostic Se and Sp were obtained using Markov chain Monte Carlo (MCMC) techniques. The MCMC sampler was run 121

for 100,000 iterations and the first 1,000 ‘burn in’ samples were discarded. The point estimate and 95% credible interval (CrI) for diagnostic Se and Sp were reported as the median and 2.5% and 97.5% quantiles of the posterior distribution of Se and Sp. Parallel chains were run using diverse initial values to ensure that convergence was achieved to the same distribution [44]. Confirmation that the posterior estimates of the monitored parameters had converged to a stable distribution was achieved by plotting cumulative path plots for each variable [45, 46] and quantified using the Raftery and Lewis convergence diagnostic [47, 48].

3.4 Results

3.4.1 Optimization and validation of the multiplex qPCR

Designed primers/probes for the T3qPCR were highly species-specific and failed to cross-amplify non-target parasites including other species of Taenia. Multiplexing the assay had no effects on the Se and efficiency of the qPCR (Fig. 3.2).

3.4.2 Field study

Of 342 stool samples collected in Dak Lak province that were tested using T3qPCR and KK, 23 (6.7%, 95% CI 4.4 to 10) and 7 (2%, 95% CI 0.9 to 4.4) were positive for Taenia spp. For cAg-ELISA using a cut-off of ≥ 0.2 OD, 21 (6.1%, 95% CI 3.9 to 9.4) samples were classified positive for Taenia spp., whereas 3 (0.9%, 95% CI 0.2 to 2.7) samples were classified positive for T. solium using a higher cut-off of OD ≥ 0.55. The summarized test results are provided in Table 3.4.

Seven individuals that were positive for taeniasis by KK that were administered praziquantel, eliminated at least one adult Taenia worm. In all cases, the species of Taenia identified was consistent with that detected by the T3qPCR, including one individual that harboured co-infection with T. saginata and T. asiatica. Amplification and DNA sequencing of the individual adult tapeworms recovered from this individual showed that one of them had 99% similarity with GenBank sequence LM146668 of T. asiatica and the other to have 100% similarity with GenBank sequence JN986712 of T. saginata.

122

Of the 23 samples positive for Taenia spp. using the T3qPCR, single infections of T. solium and T. saginata were identified in 3 (0.87%) and 14 (4.09%) individuals, respectively. No cases of a single infection with T. asiatica were identified. Mixed infections of T. solium and T. saginata occurred in one individual, and of T. saginata (Fig. 3.3)

3.4.3 Comparison of T3qPCR with microscopy

The T3qPCR was more sensitive for detection of Taenia spp. eggs in human stools compared with KK. Twenty three of 342 individuals (6.7%, 95% CI 4.4 to 10) were positive for Taenia egg using T3qPCR compared with seven of 342 individuals (2.05%, 95% CI 0.9 to 4.4) positive using KK (Table 3.4). All the KK positive samples (7 of 342) were confirmed T. saginata and/or T. asiatica using the T3qPCR. No cases of T. solium detected by qPCR were positive by KK. Cohen’s Kappa test showed moderate agreement between the two methods (Table 3.5). The estimated diagnostic sensitivities of T3qPCR and KK for detecting Taenia spp. were 0.94 (95% CrI 0.88 to 0.98) and 0.52 (95% CrI 0.07 to 0.94), respectively, while the diagnostic Sp of the two tests were 0.98 (95% CrI 0.95 to 1.00), and 0.99 (95% CrI 0.96 to 1.00) (Table 3.6), respectively.

3.4.4 Comparison of T3qPCR with cAg-ELISA

A cut-off value for the cAg-ELISA of 0.2 was obtained by calculating the mean OD value of a series of 8-reference Taenia negative stool samples ± 3 SD. The cAg-ELISA identified 21 (6.14%) samples as positive for taeniasis however failed to detect 15 of 342 samples positive by T3qPCR. The agreement between the two diagnostic tests was fair with a Kappa value of 0.32 (95% CI 0.21 to 0.42) (Table 3.5).

A cut-off value of ≥ 0.55 OD on the cAg-ELISA was assumed to be more specific for the detection of T. solium antigens. Three (0.87%) of 342 individuals were positive for T. solium by this method in agreement with T3qPCR resulting in an AP of 1.16% (95% CI 0.0 to 0.03). The cAg-ELISA failed to detect any additional cases of T. solium detected by T3qPCR. There was very good agreement between the two diagnostic tests with Kappa of 0.86 (95% CI 0.75 to 0.96) (Table 3.5). The diagnostic Se and Sp of the

123

cAg-ELISA was estimated to be 0.82 (95% CrI 0.58 to 0.95) and 0.91 (95% CrI 0.85 to 0.96), respectively (Table 3.6).

3.5 Discussion

The T3qPCR developed in this study demonstrated superior ability to both the KK thick smear and cAg-ELISA for the specific detection and discrimination of all three taeniid tapeworms found in humans, T. solium, T. asiatica and T. saginata. The assay proved highly specific with no cross-reactivity observed for 13 different intestinal parasites as well as with each other. The cross-reactivity of the T3qPCR with Echinococcus multilocularis and Echinococcus granulosus was not evaluated as DNA of these parasites are not expected to shed in stools. The innate advantage of this T3qPCR over other diagnostic methods is its ability to detect mixed cases of Taenia spp. infection, a necessary tool that provides a more comprehensive understanding of the epidemiology of taeniasis and Taenia cysticercosis in both humans and animals in regions where all three species are sympatric [49]. For example, in communities such as those in the Central Highlands of Vietnam, where taeniasis is highly endemic, use of the T3qPCR allows the TP and risk factors for T. solium taeniasis carriers to be determined. Use of these diagnostic tools to inform taeniasis control programs will not only reduce the risk of T. solium NCC in humans, but will also assist in breaking the transmission cycle to backyard and free-roaming pigs, reducing economic losses to the small-holder farmers due to carcass condemnation [50–53]. Using the Bayesian approach described in this study, the multiplex qPCR had an estimated Se of 0.94 (95% CrI 0.88 to 0.98) compared to 0.83 (95% CrI 0.57 to 0.98) for a similar Taq- Man probe based multiplex qPCR described by Praet et al. (2013) [22] for the specific detection and differentiation of T. solium and T. saginata. The Sp of T3qPCR of 0.98 (95% CrI 0.95 to 1.00) was equivalent (0.99, 95% CrI 0.98 to 0.99) to the multiplex qPCR developed by Praet et al. (2013) [22].

Treatment and recovery followed by morphological and genetic identification of two adult tapeworms from a single individual supported the T3qPCR results to confirm the presence of T. asiatica in the Central Highlands of Vietnam for the first time, extending the known distribution of this tapeworm to southern Vietnam. This assay also confirmed

124

the co-existence of T. solium, T. saginata and T. asiatica in the Central Highlands region. The elimination of two adult tapeworms from a single individual also confirmed the ability of the T3qPCR to detect both species of worms as opposed to infection with hybrid tapeworm. In comparison to the KK thick smear, the T3qPCR offered a significantly greater Se (0.94, 95% CrI 0.88 to 0.99) for the detection of Taenia eggs in stool samples. There were 16 Taenia spp. T3qPCR-positive samples not detected by KK (Table 3.4). The innately lower Se of KK to detect eggs in faeces may explain this [54]. However, the unsuitability of microscopic-based techniques for the detection of Taenia spp. in faeces may also be compounded by the inability of Taenia spp. to actively shed/lay eggs through a uterine pore like Pseudophyllidians do so [55]. As opposed to microscopy-based techniques, the T3qPCR may be in addition to any released eggs, detecting sloughed tapeworm DNA present in the stool, however this needs to be confirmed. Most significantly, KK missed all four T. solium positive individuals confirmed by T3qPCR in this community. As a result, it is our opinion that microscopic based faecal examination (KK) is neither suitable nor recommended for screening for taeniasis. Therefore, qPCR or cAg-ELISA at a higher OD cut-off is recommended for screening T. solium carriers in community-based surveys in South East Asia where sympatric infection of all three human Taenia tapeworms is common [4, 26].

The estimated Se of the cAg-ELISA assay with an OD cut-off value at 0.55 using polyclonal antibodies from hyper-immunized rabbits against ES tapeworm antigens in this study was estimated at 0.82 (95% CrI 0.58 to 0.95), twofold higher than that of KK. There was a strong agreement with the T3qPCR, in which all three cAg-ELISA-positive samples were confirmed as T. solium infection by T3qPCR. The Se of the cAg-ELISA in this study was lower than that reported by Allan et al. (1996) [56] who determined test Se on purgation of a subset of coproantigen-positive cases. Using a Bayesian approach, Praet et al. (2013) [1] Showed that the diagnostic Se and Sp of cAg-ELISA was 0.85 (95% CrI 0.62 to 0.98) and 0.92 (95% CrI 0.85 to 0.96), respectively, similar to that estimated for this study. In total, thirteen of 21 cAg-ELISA OD 0.2-positive samples (Table 3.5) were not confirmed to be positive for Taenia by T3qPCR and microscopy. This suggests that the cAg-ELISA may cross-react with other parasites rather than Taenia spp [56]. cAg-ELISA shows excellent potential for a large scale

125

epidemiological community-based survey in regions where sympatric Taenia spp. exist as it is capable of specifically identifying T. solium carriers when the cut-off OD is set at ≥ 0.55. Therefore, choosing an appropriate OD cut-off value for cAg-ELISA seems to be an important diagnostic consideration when applying the assay for the diagnosis of T. solium taeniasis in different communities.

In conclusion, this study describes the development and test characteristics of a new T3qPCR for the detection and differentiation of all three human Taenia species in stool. The absence of a diagnostic gold standard necessitated the use of a Bayesian approach to estimate diagnostic test performance. In future, the usefulness of the T3qPCR for large-scaled epidemiological studies can be confirmed using a larger sample size of individuals, ideally in a community in which the prevalence of taeniasis is higher. The assay also has the potential use in anthelmintic efficacy trials targeting individual species of tapeworms, thereby directly allowing assessment of species-specific susceptibility to drugs as well as non-chemotherapeutic interventions.

126

3.6 References

[1] Ito A, Nakao M, Wandra T. Human taeniasis and cysticercosis in Asia. Lancet 2003;362:1918–20.

[2] Eom KS, Rim HJ. Morphologic descriptions of Taenia asiatica sp. n. Korean J Parasitol 1993;31:1–6.

[3] Eom KS, Jeon HK, Rim HJ. Geographical distribution of Taenia asiatica and related species. Korean J Parasitol 2009;47:S115–24.

[4] Ale A, Victor B, Praet N, Gabriël S, Speybroeck N, Dorny P, et al. Epidemiology and genetic diversity of Taenia asiatica: A systematic review. Parasites and Vectors 2014;7:45.

[5] Murrell KD, Dorny P, Flisser A, Geerts S, Kyvsgaard NC, McManus DP, et al. WHO/FAO/OIE Guidelines for the Surveillance, Prevention and Control of Taeniosis/Cysticercosis. Paris: OIE, WHO and FAO; 2005.

[6] Schantz PM, Cruz M, Sarti E, Pawlowski ZS. Potential eradicability of taeniasis and cysticercosis. Bull PAHO 1993;27:397–403.

[7] WHO. Atlas: Epilepsy Care in the World. Geneva: WHO press; 2005.

[8] Bowles J, McManus DP. Genetic characterization of the Asian Taenia, a newly described taeniid cestode of humans. Am J Trop Med Hyg 1994;50:33–44.

[9] Carabin H, Ndimubanzi PC, Budke CM, Nguyen H, Qian Y, Cowan LD, et al. Clinical manifestations associated with neurocysticercosis: a systematic review. PLoS Negl Trop Dis 2011;5:e1152.

[10] White AC. Neurocysticercosis: updates on epidemiology, pathogenesis, diagnosis, and management. Annu Rev Med 2000;51:187–206.

127

[11] Teerasukjinda O, Wongjittraporn S, Tongma C, Chung H. Asymptomatic giant intraventricular cysticercosis: A case report. Hawaii J Med Public Health 2016;75:187–9.

[12] Gonzales I, Rivera JT, Garcia HH, Cysticercosis Working Group in Peru. Pathogenesis of Taenia solium taeniasis and cysticercosis. Parasite Immunol 2016;38:136–46.

[13] Van De N, Le TH, Lien PTH, Eom KS. Current status of taeniasis and cysticercosis in Vietnam. Korean J Parasitol 2014;52:125–9.

[14] Mayta H, Talley A, Gilman RH, Jimenez J, Verastegui M, Ruiz M, et al. Differentiating Taenia solium and Taenia saginata infections by simple hematoxylin-eosin staining and PCR-restriction enzyme analysis. J Clin Microbiol 2000;38:133–7.

[15] Wilkins PP, Allan JC, Verastegui M, Acosta M, Eason AG, Hugo Garcia H, et al. Development of a serologic assay to detect Taenia solium taeniasis. Am J Trop Med Hyg 1999;60:199–204.

[16] Jeon HK, Eom KS. Immunoblot patterns of Taenia asiatica taeniasis. Korean J Parasitol 2009;47:73–7.

[17] Levine MZ, Lewis MM, Rodriquez S, Jimenez JA, Khan A, Lin S, et al. Development of an enzyme-linked immunoelectrotransfer blot (EITB) assay using two Baculovirus expressed recombinant antigens for diagnosis of Taenia solium taeniasis. J Parasitol 2007;93:409–17.

[18] Dorny P, Brandt J, Zoli A, Geerts S. Immunodiagnostic tools for human and porcine cysticercosis. Acta Trop 2003;87:79–86.

[19] Raoul F, Li T, Sako Y, Chen X, Long C, Yanagida T, et al. Advances in diagnosis and spatial analysis of cysticercosis and taeniasis. Parasitology 2013;140:1578–88.

128

[20] Mayta H, Gilman RH, Prendergast E, Castillo JP, Tinoco YO, Garcia HH, et al. Nested PCR for specific diagnosis of Taenia solium taeniasis. J Clin Microbiol 2008;46:286–9.

[21] Somers R, Dorny P, Geysen D, Nguyen LA, Thach DC, Vercruysse J, et al. Human tapeworms in north Vietnam. Trans R Soc Trop Med Hyg 2007;101:275–7.

[22] Praet N, Verweij JJ, Mwape KE, Phiri IK, Muma JB, Zulu G, et al. Bayesian modelling to estimate the test characteristics of coprology, coproantigen ELISA and a novel real-time PCR for the diagnosis of taeniasis. Trop Med Int Heal 2013;18:608–14.

[23] Guezala MC, Rodriguez S, Zamora H, Garcia HH, Gonzalez AE, Tembo A, et al. Development of a species-specific coproantigen ELISA for human Taenia solium taeniasis. Am J Trop Med Hyg 2009;81:433–7.

[24] Deplazes P, Eckert J, Pawlowski ZS, Machowska L, Gottstein B. An enzyme- linked immunosorbent assay for diagnostic detection of Taenia saginata coproantigens in humans. Trans R Soc Trop Med Hyg 1991;85:391–6.

[25] Allan JC, Craig PS, Garcia Noval J, Mencos F, Liu D, Wang Y, et al. Coproantigen detection for immunodiagnosis of echinococcosis and taeniasis in dogs and humans. Parasitol 1992;104:347–56.

[26] Aung AK, Spelman DW. Taenia solium taeniasis and cysticercosis in southeast Asia. Am J Trop Med Hyg 2016;94:947–54.

[27] Jeon HK, Chai JY, Kong Y, Waikagul J, Insisiengmay B, Rim HJ, et al. Differential diagnosis of Taenia asiatica using multiplex PCR. Exp Parasitol 2009;121:151–6.

[28] Yamasaki H, Allan JC, Sato MO, Nakao M, Sako Y, Nakaya K, et al. DNA differential diagnosis of taeniasis and cysticercosis by multiplex PCR. J Clin Microbiol 2004;42:548–53.

129

[29] Gordon CA, McManus DP, Acosta LP, Olveda RM, Williams GM, Ross AG, et al. Multiplex real-time PCR monitoring of intestinal helminths in humans reveals widespread polyparasitism in northern Samar, the Philippines. Int J Parasitol 2015;45:477–83.

[30] Nkouawa A, Sako Y, Nakao M, Nakaya K, Ito A. Loop-mediated isothermal amplification method for differentiation and rapid detection of Taenia species. J Clin Microbiol 2009;47:168–74.

[31] Lambert SB, Whiley DM, O’Neill NT, Andrews EC, Canavan FM, Bletchly C, et al. Comparing nose-throat swabs and nasopharyngeal aspirates collected from children with symptoms for respiratory virus identification using real-time polymerase chain reaction. Pediatrics 2008;122:e615–20.

[32] Ng-Nguyen D, Stevenson MA, Traub RJ. A systematic review of taeniasis, cysticercosis and trichinellosis in Vietnam. Parasit Vectors 2017;10:150.

[33] Allan JC, Avila G, Garcia Noval J, Flisser A, Craig PS. Immunodiagnosis of taeniasis by coproantigen detection. Parasitology 1990;101:473–7.

[34] Mwape KE, Phiri IK, Praet N, Muma JB, Zulu G, van den Bossche P, et al. Taenia solium infections in a rural area of Eastern Zambia-A community based study. PLoS Negl Trop Dis 2012;6:e1594.

[35] Thienpont D, Rochette F, Vanparijs OFJ. Diagnosing Helminthiasis by Coprological Examination. 2nd ed. Beerse: Janssen Research Foundation; 1986.

[36] Hoff R, Hoff R, Sleigh A, Mott K, Barreto M, de Paiva TM, et al. Comparison of filtration staining (bell) and thick smear (kato) for the detection and quantitation of schistosoma mansoni eggs in faeces. Trans R Soc Trop Med Hyg 1982;76:403–6.

[37] Teesdale CH, Amin MA. A simple thick-smear technique for the diagnosis of Schistosoma mansoni infection. Bull World Health Organ 1976;54:703–5.

130

[38] Katz N, Chaves A, Pellegrino J. A simple device for quantitative stool thick- smear technique in Schistosoma mansoni. Rev Do Inst Med Trop São Paulo 1972;14:397–400.

[39] Asaava LL, Kitala PM, Gathura PB, Nanyingi MO, Muchemi G, Schelling E. A survey of bovine cysticercosis/human taeniosis in Northern Turkana District, Kenya. Prev Vet Med 2009;89:197–204.

[40] Team R Core. R: A Language and Environment for Statistical Computing 2017.

[41] Van Chuong N. Status of helminthic infection in humans in 14 provinces spanning the Central and Central Highland regions of Vietnam. IMPE-QN 2005. http://www.impe-qn.org.vn/impe-qn/vn/portal/InfoDetail.jsp?area=58&cat=1068& ID=597 (accessed March 1, 2016).

[42] Branscum AJ, Gardner IA, Johnson WO. Estimation of diagnostic-test sensitivity and specificity through Bayesian modeling. Prev Vet Med 2005;68:145–63.

[43] Bouma A, Stegeman JA, Engel B, de Kluijver EP, Elbers ARW, de Jong MCM. Evaluation of diagnostic tests for the detection of classical swine fever in the field without a gold standard. J Vet Diagnostic Investig 2001;13:383–8.

[44] Gelman A. Inference and Monitoring Convergence. In: Gilks W, Richardson S, Spiegelhalter D, editors. Markov Chain Monte Carlo Pract., London: Chapman & Hall; 1996, p. 131–143.

[45] Yu B, Mykland P. Looking at Markov samplers through cusum path plots: a simple diagnostic idea. Stat Comput 1998;8:275-286–286.

[46] Casella G, Fienberg S, Olkin I. Statistical Design (Springer Texts in Statistics). vol. 102. New York: Springer; 2006.

[47] Raftery AE, Lewis SM. One long run with diagnostics: implementation strategies for Markov Chain Monte Carlo. Stat Sci 1992;7:493–7.

131

[48] Raftery EA, Lewis MS. How Many Iterations in the Gibbs Sampler? In: Benardo J, Berger J, Dawid A, Smith A, editors. Bayesian Stat. 4, New York: The Clarendon Press and Oxford University Press; 1992, p. 763–774.

[49] Ito A, Yanagida T, Nakao M. Recent advances and perspectives in molecular epidemiology of Taenia solium cysticercosis. Infect Genet Evol 2015;40:357–67.

[50] Bhattarai R, Carabin H, Proaño J V., Flores-Rivera J, Corona T, Flisser A, et al. Cost of neurocysticercosis patients treated in two referral hospitals in Mexico City, Mexico. Trop Med Int Heal 2015;20:1108–19.

[51] Rajkotia Y, Lescano AG, Gilman RH, Cornejo C, Garcia HH, Peru Cysticercosis Working Group. Economic burden of neurocysticercosis: results from Peru. Trans R Soc Trop Med Hyg 2007;101:840–6.

[52] Praet N, Speybroeck N, Manzanedo R, Berkvens D, Nforninwe DN, Zoli A, et al. The disease burden of Taenia solium cysticercosis in Cameroon. PLoS Negl Trop Dis 2009;3:e406.

[53] Choudhury AA, Conlan J V, Racloz VN, Reid SA, Blacksell SD, Fenwick SG, et al. The economic impact of pig-associated parasitic zoonosis in Northern Lao PDR. Ecohealth 2013;10:54–62.

[54] Llewellyn S, Inpankaew T, Nery SV, Gray DJ, Verweij JJ, Clements ACA, et al. Application of a multiplex quantitative PCR to assess prevalence and intensity of intestinal parasite infections in a controlled clinical trial. PLoS Negl Trop Dis 2016;10:e0004380.

[55] Price DL. Procedure Manual for the Diagnosis of Intestinal Parasites. 1st ed. Boca Raton: CRC Press; 1994.

[56] Allan JC, Velasquez-Tohom M, Torres-Alvarez R, Yurrita P, Garcia-Noval J. Field trial of the coproantigen-based diagnosis of Taenia solium taeniasis by enzyme-linked immunosorbent assay. Am J Trop Med Hyg 1996;54:352–6.

132

Table 3.1 Primers and probes for multiplex qPCR

Target Oligonucleotide GenBank Target gene Sequence (5’ - 3’) species accession # T. solium Forward primer LC004200 ITS-1 TGTTAGCAGCAGTTGTGATG Reverse primer CCAACTCCACCCAGATTGA Probe Cy5/CCAGCGCTG/TAO/CTGTTGACTGA/IAbRQSp T. asiatica Forward primer of AB107234 COX-1 GAGTACCAACAGGAATAAAGGT and T. asiatica AB271695 T. saginata and T. saginata Reverse primer of T. CAACACAATACCAGTCACACC asiatica and T. saginata Probe of T. asiatica FAM/AA+CTA+T+C+CA+CCA/IAbkFQ Probe of HEX/AA+CTA+T+T+CA+C+CA/IAbkFQ T. saginata EHV4 Forward primer M26171 Glycoprotein GATGACACTAGCGACTTCGA Reverse primer gB CAGGGCAGAAACCATAGACA Probe ROX/TTTCGCGTGCCTCCTCCAG/IAbRQSp Note: “+” symbols found in the probe sequences indicate the presence of a Locked Nucleic Acid Base; IAbRQSp: Iowa Black® RQ-Sp; IAbkFQ: Iowa Black® FQ

Table 3.2 gBlocks positive controls

Organisms GenBank accession Gene Star-end position 1 Amplicon length number T. solium and LC004200 ITS-1 92-183 235 bp T. asiatica AB107234 COX-1 938-1077 T. saginata AB271695 COX-1 938-1077 140 bp Note: 1 Original position in genBank

Table 3.3 Parameters of prior information

Parameter of interest Mode Lower limit 95% T3qPCR* Se 0.95 0.89 Sp 0.99 0.94 cAg-ELISA Se 0.85 0.62 Sp 0.92 0.90 KK Se 0.53 0.11 SP 1.00 1.00 *Modes and lower limits for test characteristic of T3qPCR calculated as the average of the estimates elicited from each of the six experts

133

Table 3.4 Results of three diagnostic tests for taeniasis

T3qPCR KK cAg-ELISA cAg-ELISA Individuals (OD>0.2) (OD>=0.55) T. solium T. saginata T. asiatica T. solium & T. saginata T. saginata & T. asiatica 0 0 0 0 0 0 0 0 306 0 1 0 0 0 1 1 0 1 0 1 0 0 0 1 0 0 5 0 0 0 0 1 1 0 0 1 0 0 0 1 0 0 1 0 1 0 0 0 0 1 0 1 0 1 0 1 0 0 0 0 1 0 2 1 0 0 0 0 0 1 1 3 0 1 0 0 0 0 0 0 6 0 0 0 0 1 0 0 0 3 0 0 0 0 0 0 1 0 13 Note: 0: Negative; 1: Positive.

Table 3.5 The agreement statistics of T3qPCR, cAg-ELISA and KK for the detection of Taenia spp. in stool

T3qPCR Kappa (95% CI) a Observed Positive Negative Total agreement (%) cAg-ELISA Positive 8 13 21 0.32 (0.21 to 0.43) 91.8 (OD >0.2) Negative 15 306 321 Total 23 319 342 cAg-ELISA Positive 3 0 3 0.86 (0.75 to 0.96) 99.7 (OD ≥0.55) Negative 1 338 339 Total 4 338 342 KK Positive 7 0 7 0.45 (0.36 to 0.53) 95.3 Negative 16 319 335 Total 23 319 342 a Kappa agreement level: K<0.2 Poor; 0.21-0.40 Fair; 0.41-0.60 Moderate; 0.61-0.80 Good; 0.81-1.00 Very good.

Table 3.6 The estimated characteristics of three diagnostic tests for the detection of Taenia spp. in stool

Test method Parameter of interest Median 95% credible interval (CrI) T3qPCR Se 0.94 0.88 to 0.98 Sp 0.98 0.95 to 1.00 cAg-ELISA Se 0.82 0.58 to 0.95 (OD ≥0.55) Sp 0.91 0.85 to 0.96 KK Se 0.52 0.07 to 0.94 Sp 0.94 0.89 to 0.98

134

Fig. 3.1 Map of study site. Map of Dak Lak province showing districts where sampling was carried out.

135

Fig. 3.2 Ct value comparison of multiplex to singleplex qPCR. Assay optimization to determine effects of multiplex PCR set up on Se and efficiency of PCRs compared to singleplex PCR using gBlock gene fragment (IDT Technologies) standard curve controls containing all PCR products.

136

Fig. 3.3 Infection status of human Taenia species. The venn diagram shows the proportional infection of Taenia spp. in Dak Lak province, Vietnam.

137

CHAPTER 4

The epidemiology of Taenia spp. infection and Taenia solium cysticerci exposure in humans in the Central Highlands of Vietnam

Presented as published:

138

4.1 Abstract

4.1.1 Background

Vietnam is endemic for taeniasis and T. solium cysticercosis. Despite this, information on the epidemiological characteristics of the diseases in the Central Highlands of Vietnam is poorly described. The aims of this study were to determine the epidemiological characteristics of taeniasis (Taenia spp.) and T. solium cysticerci exposure in humans in Dak Lak province in the Central Highlands, Vietnam.

4.1.2 Methods

This cross-sectional study was carried out in six villages in three districts of Dak Lak. A total of 190 households were visited. From each household, between one and five individuals were asked to donate a single faecal and blood sample and respond to a questionnaire. Serum samples were subjected to lentil lectin purified glycoprotein enzyme-linked immunoelectrotransfer blot assay to detect antibodies against T. solium cysticerci. Multiplex real-time PCR was used to detect Taenia spp. infection in faecal samples. A fixed-effects logistic regression model was developed to identify factors associated with the probability of Taenia spp. infection or T. solium cysticerci exposure risk. The contribution of each of identified factor was quantified using population attributable fractions.

4.1.3 Results

The prevalence of seroexposure to T. solium in Dak Lak was 5% (95% CI 3% to 8%). Consumption of raw vegetables, sourcing drinking water from lakes, streams or ponds and the practice of outdoor defaecation were identified as primary risk factors for the prevalence of T. solium cysticerci exposure, while consuming undercooked pork and beef, pork tongue and observing Taenia proglottids in stool were associated with Taenia spp. infection. Consumption of raw vegetables was attributed to 74% of T. solium cysticerci exposure-positive cases and consumption of undercooked beef was attributed to 77% of taeniasis cases in these communities.

139

4.1.4 Conclusions

The prevalence of T. solium seroexposure in Dak Lak is consistent with those reported in other regions of Vietnam. The identified risk factors associated with the prevalence of T. solium seroexposure and taeniasis infection in Dak Lak are modifiable and thus advocate for targeted community intervention programs to mitigating these risks.

4.2 Background

Taenia solium cysticercosis and human taeniasis are considered neglected tropical diseases by the World Health Organization [1]. When accidentally ingested either directly, or indirectly via contaminated food and water, eggs of T. solium hatch and lodge in the skeletal muscle, under the skin and/or within the central nervous system causing muscular-, subcutaneous- and/or neuro-cysticercosis. Neurocysticercosis is a common cause of epilepsy, syncope, paralysis and chronic headache [2, 3]. Globally, the disease has been estimated to cause approximately 2.8 million DALYs (disability- adjusted life-years) lost and approximately 300,000 individuals were estimated to be infected globally in 2010, resulting in over 28,100 deaths [4]. Consumption of raw or undercooked meat and/or visceral organs containing cysticerci of T. solium, Taenia saginata or Taenia asiatica cause taeniasis. Cysticercosis reduces the quality of life, is responsible for diagnostic and treatment costs in infected individuals and indirectly contributes to poverty due to losses in livestock production arising from organ condemnation and loss of market access [5].

Vietnam is considered endemic for taeniasis and T. solium cysticercosis [6, 7]. Previous research focusing on these diseases in Vietnam has mostly been carried out in the north of the country. A review of the literature shows that common risk factors for taeniasis and cysticercosis include consumption of raw or undercooked meat and vegetables, lack of functioning latrines, husbandry practices allowing pigs to free-roam and routine use of human waste for fertilizing and irrigating crops [8–12]. Information on the prevalence of and risk factors for taeniasis and T. solium cysticercosis in the central and southern areas of Vietnam is scarce and outdated [12]. In Dak Lak province, there is no information available on the infection status of T. saginata in livestock, however our research data on the seroprevalence of cysticercosis in pigs was 0.95% [13]. A previous 140

study carried out by our group using a newly developed multiplex real-time PCR found the prevalence of Taenia solium, T. saginata and T. asiatica taeniasis to be 1.2%, 5.8% and 1.5%, respectively in Dak Lak province [14]. In this study, our aims were to: i) estimate the seroprevalence of T. solium cysticerci exposure using the lentil-lectin purified glycoprotein enzyme-linked immunoelectrotransfer blot (LLGP-EITB) assay; ii) identify and quantify risk factors for human T. solium cysticerci exposure and taeniasis (Taenia spp. infection), that inform optimal risk mitigation measures in studied communities in Dak Lak province, Vietnam.

4.3 Methods

4.3.1 Study site and sampling

The study was deployed in the districts of Buon Don, Krong Nang and M’Drak in the province of Dak Lak in the Central Highlands of Vietnam between May and October 2015. M’Drak is located in the east of Dak Lak province with an average altitude of 400 m to 500 m and has a tropical monsoon climate typical of the Vietnam’s Central Coast. Krong Nang is situated in the north of the province at an altitude of 800 m. Buon Don, situated to the west of the province with an average elevation of 330 m and has a hot and dry climate. The standard of living in this area of Vietnam is generally poor. The literacy rate among local residents aged between 15 and 60 in rural areas was low [15]. Open defaecation using outdoor pit latrines is common practice and livestock access to these latrine areas is usually unrestricted. The practice of non-confinement of pigs and cattle is common with slaughter activities often carried out in backyards [12].

The sample size was calculated based on the previous study of Van De at al. [16]. The seroprevalence of T. solium cysticerci exposure was assumed to be 7% and there was 95% certainty that this estimate was within 5% of the true population value (i.e. the prevalence ranged from 2% to 12%). Ignoring the tendency for T. solium cysticerci exposure to cluster within households we estimated that a total of 100 samples were required. We then assumed the average number of individuals eligible for the study per household was three and the between household (cluster) variance in seroprevalence was around 1.14 times greater than the within household (cluster) variance, returning an intra-class correlation coefficient of 0.07 [17]. Our revised sample size, accounting for 141

the tendency of cluster within households, was 114 for each district and 342 for the whole province.

A sampling frame listing the name of all villages in Dak Lak province was obtained from the Sub-Department of Animal Health office within the Ministry of Agriculture. Villages eligible for sampling comprised those with more than 1000 pigs, as recorded by the Sub-Department of Animal Health. All eligible villages within each district were assigned a number and two numbers chosen at random to select villages from each district for inclusion in the study. A list of householder names within each selected village was obtained from each village head person, and household names were assigned a numeric code. The sheet of paper of numeric household codes for each village were cut into pieces and placed face-down on a table. The village head person was asked to select 50 households at random. Each village has a number of households between 200 and 300. All selected households were visited several days before the proposed sampling date to obtain consent from participants to take part in the study. From each household between one and five individuals were requested to take part in the study. Individuals eligible for inclusion in the study were healthy, not pregnant and over seven years of age. Individuals requested to take part in the study signed a consent form. Those that were under 18 years of age were required to provide written consent as well as written consent from either their parents or legal guardians. At the time of consent, each study participant was given a labelled stool container, with instructions that the container would be collected on the date of sampling, several days later. Five mL of venous blood was collected into plain clotting tubes from consenting study participants by local medical staff. The blood was kept at ambient temperature for clotting and then centrifuged at 3200 × g for 5 minutes to collect serum. The serum was stored at -20 °C until use.

4.3.2 Household questionnaire

Each study participant was requested to answer a prepared questionnaire. The questionnaire comprised of two sections. The first section collected information at the household level and was completed by the nominated head of the household. Questions included in this section solicited details on housing type, details of the type of latrines in 142

use (if any), the presence and type of backyard pigsties, the source(s) of drinking water and activities associated with cultivation, irrigation and fertilization of home or farm grown vegetables. The second part of the questionnaire solicited details at the individual study participant level including their age, gender, the highest level of education achieved, dietary practices, and hygiene. All questionnaires were conducted in local Vietnamese phraseology and the validity of the questionnaire was pre-tested on thirty individuals in another community in Dak Lak province before application to the field survey. In addition, interviewers were trained before administering the questionnaires.

4.3.3 Laboratory procedures

4.3.3.1 Detection of antibodies against T. solium cysticerci

Serum samples were subjected to LLGP-EITB assay to detect antibodies against T. solium cysticerci. This assay has a diagnostic sensitivity of 98% and a diagnostic specificity of 100%. The assay was performed as previously described by Tsang et al. (1989) [18] and Noh et al. (2014) [19].

4.3.3.2 Detection of Taenia spp. infection

A total of 342 faecal samples were subject to a multiplex real-time PCR (T3qPCR) to detect all three species of Taenia, notably, T. solium, Taenia saginata and Taenia asiatica. The assay has been reported to have a diagnostic sensitivity of 94% and a diagnostic specificity of 98%. The positive results of T3qPCR were confirmed using either a conventional multiplex PCR targeting T. solium, T. saginata and T. asiatica [20] or a singleplex PCR for Taenia sp. following DNA sequencing. The methods and results have been described and published elsewhere [14].

4.3.4 Statistical methods

Risk factors for T. solium cysticerci exposure and taeniasis (Taenia spp. infection) in humans in the communities of Dak Lak province were identified using logistic regression. The outcome of interest for this study was a binary response variable, Yi, taking the value 1 if an individual was T. solium cysticerci exposure-positive using the

143

LLGP-EITB or in the case of taeniasis, was positive for Taenia spp. infection by T3qPCR, and 0 otherwise.

This study did not use a simple random sampling design so to avoid erroneous point estimates of the strength of the association between hypothesized risk factors and the outcome of interest and their standard errors the data were analysed in a manner that respected the three-stage cluster design (villages within districts formed the primary sampling units, households within villages formed the secondary sampling units and individuals within households formed the tertiary sampling units). Sampling weights WI, provided an estimate of the inverse probability of an individual being selected for study (Lumley 2010) [21] and were quantified for each study participant as follows:

푉 퐻 푀 푊 = × × Equation 1 푣 ℎ 푚 In Equation 1, V and v represent the total number of villages in each district and the number of sampled villages in each district, respectively; H and h represent the total number of households in each sampled village and the number of sampled households in each village, respectively; and M and m represent the total number of individuals in each sampled household, and the number of sampled individuals in each household, respectively.

Unconditional associations between each of the hypothesized risk factors from the questionnaire and the outcome of interest were computed using the odds ratio. Hypothesized risk factors (explanatory variables) with unconditional associations that were significant at P < 0.20 level (2-sided) were selected for multivariable modeling.

A fixed-effects logistic regression model was developed where the probability of an individual being Taenia- or T. solium cysticerci exposure-positive was parameterized as a function of the explanatory variables with unconditional associations significant at P < 0.20, as described above. Explanatory variables that were not statistically significant were removed from the model one at a time, beginning with the least significant, until the estimated regression coefficients for the explanatory variables retained were significant at an alpha level of less than 0.05. Explanatory variables that were excluded

144

at the initial screening stage were tested for inclusion in the final model and were retained if their inclusion changed any of the estimated regression coefficients by more than 20%. Biologically plausible two-way interactions were tested and none were significant at an alpha level of 0.05.

To account for the hierarchical structure of the data, that is, villages within districts and households within villages and study participants within households, we extended the model to include village- and household-level random effect terms using the lme4 package [22] in R version 3.4.3 [23]. This extension to the model was informative because it provided the opportunity to distinguish the influence of unobserved explanatory variables operating at the village, household and individual level on Taenia- and T. solium cysticerci exposure-positive risk. In our multilevel model the variance of the random effect terms at the village and household level were negligible, and the point estimates of the regression coefficients and standard errors for the explanatory variables were all within 0.05 units of the regression coefficients and standard errors for the fixed-effects model. For parsimony, we report the results of the fixed-effects regression model adjusted for the three-stage sampling design using the survey package [24] in R.

Frequency histograms of the residuals from the fixed effects model and plots of the residuals versus predicted values were constructed to check that the assumptions of normality and homogeneity of variance had been met and to identify the presence of influential observations (data not presented).

The contribution of each of the explanatory variables in the final model on Taenia- and T. solium cysticerci exposure-positive risk was quantified using the population attributable fraction. The population attributable fraction for a given explanatory variable is the proportional reduction in outcome event incidence expected by either eliminating exposure to the variable or completely preventing the effects of exposure, assuming the explanatory variable is causative and assuming there has been no bias and sampling error in the study population [25]. Using the method of Toschke et al. (2007) [26], adjusted population attributable fractions were calculated for each of the explanatory variables that were significant in the mixed-effects logistic regression model.

145

4.4 Results

4.4.1 General description of study population and data structure

Table 4.1 presents the structure of the data. In total 342 individuals, from 190 households in six villages in three districts of Dak Lak province consented to take part in the study. A household had, on average, two individuals that consented to take part (minimum 1; maximum 5). In total, only 25% (95% confidence interval (CI) 19% to 32%, 47 of 190) of households possessed a latrine and 25% (95% CI 19% to 32%, 48 of 190) of households sourced their water for daily activities, including drinking, from lakes, streams or ponds. Untreated drinking water was consumed by 60% (95% CI 53% to 67%, 115 of 190) of households (Table 4.2). The literacy rate among participants was 56% (95% CI 51% to 62%, 193 of 342). Most participants (87%, 95% CI 80% to 88%) listed farming as their main occupational activity (Table 4.3). The percentage of study participants that did not wash their hands after defaecation or before eating was 35% (95% CI 31% to 41%, 121 of 339) and 44% (95% CI 39% to 50%, 150 of 339), respectively.

Total of 342 serum samples collected from study participants, three were excluded due to hemolysis resulting in 339 samples included for further analysis. Of 339 serum samples, antibodies against T. solium cysticerci were identified in 17 individuals (5%, 95% CI 3% to 8%). Among 342 faecal samples, 23 (6.7%, 95% CI 4.4% to 10%) were positive for Taenia spp. Single infection of T. solium and T. saginata were identified in three and 14 samples, respectively; mixed infections of T. solium and T. saginata occurred in one samples, and of T. saginata and T. asiatica in five samples [14].

4.4.2 Risk factors for T. solium cysticerci exposure

Of the data recorded using the questionnaire, we identified three factors associated with an individual’s likelihood of being exposure-positive to T. solium cysticerci based on the results of the LLGP-EITB assay. These were, the frequent consumption of raw vegetables, sourcing drinking water from lakes, streams or ponds and outdoor defaecation.

146

Estimated regression coefficients for consumption of raw vegetables, the source of drinking water and routine location of defaecation are provided in Table 4.4. After adjusting for other explanatory variables in the model, the odds of an individual who often consumed raw vegetables being T. solium cysticerci exposure-positive was 9.98 (95% CI 2.89 to 34.4) times that of an individual who consumed raw vegetables rarely. The odds ratio for those who routinely defaecated outdoors was similar, at 11.1 (95% CI 2.97 to 42.0). The adjusted population attributable fraction for the consumption of raw vegetables was 74% and thus, reducing the prevalence of raw vegetable consumption in the population is expected to reduce the exposure to T. solium cysticerci by 74%. Similarly, the population attributable fraction estimate for routine location of defaecation was 60%.

The odds of T. solium cysticerci exposure for those that routinely sourced their drinking water from streams, ponds or lakes was 3.94 (95% CI 1.29 to 11.9) times higher than the odds of T. solium cysticerci exposure for those that sourced their water from wells.

4.4.3 Risk factors for Taenia spp. infection

We identified five factors associated with an individual’s likelihood of infection with Taenia spp. gender, the frequent consumption of pork tongue, the frequent consumption of undercooked pork, frequent consumption of undercooked beef, and observation of Taenia proglottids shedding from either themselves, their relatives or neighbors (Table 4.5).

After adjusting for other explanatory variables in the model, the odds of Taenia spp. infection for an individual who frequently consumed pork tongue was 4.62 (95% CI 1.49 to 14.3) times the odds for an individual who consumed pork tongue only rarely. Males had higher odds of being Taenia spp. positive compared with females (OR 2.77, 95% CI 1.00 to 7.68).

Among the modifiable risk factors for Taenia spp. infection identified in this study, the adjusted population attributable fraction for frequent consumption of undercooked beef was 77%. The population attributable fractions for frequent consumption of

147

undercooked pork and pork tongue were 24% and 26%, respectively. Assuming undercooked beef consumption is a component cause of taeniasis risk, reducing the prevalence of this practice in the population is expected to reduce the overall prevalence of taeniasis by 77%.

4.5 Discussion

The seroprevalence of exposure to T. solium in rural communities in Dak Lak in the Central Highlands was 5% (95% CI 3% to 8%). This prevalence is relatively similar to other regions across Vietnam. Phan et al. (2001) [11] reported a 4% seroprevalence of cysticercosis across 22 provinces in the south of Vietnam. For studies in the north and northwest, where T. solium is reported to be endemic, the seroprevalence ranged from 2% to 6% [27–30]. In this study, various risk factors were evaluated to determine their association with seroexposure to T. solium using the LLGP-EITB assay. Of these, raw vegetable consumption, the source of drinking water and routine location of defaecation were significantly associated with an increase in T. solium cysticerci exposure (Table 4.4). It is known that risk factors for transmission and circulation of T. solium cysticercosis are numerous and may vary in different settings. The identified risk factors in Dak Lak were consistent with that of study carried out in the north and south of Vietnam [11, 31]. In contrast to studies in Vietnam’s north, the utilisation of night-soil was not associated with the increase in T. solium exposure observed in these study communities [12, 16]. Similar to the studies in western Sichuan (China) [32], Mbeya (Tanzania) [33] and northern India [34], routine open defaecation, utilisation of unsafe water sources and consumption of raw vegetables significantly increased the risk of cysticercosis in Dak Lak. However, Cherian et al. (2014) [35] reported that these factors were not contributive in individuals who had epilepsy in Kerala, India. Research on the epidemiological characteristics of human T. solium cysticercosis in Nigeria [36], Tanzania [37], and Guatemala [38] found that gender and age [39] were associated with cysticercosis seropositivity, in which the female gender and increased age were significantly linked to the prevalence of T. solium. In our study, age and gender did not contribute to seroexposure of T. solium cysticerci.

148

Each of the identified risk factors in this study are eminently modifiable and if effective interventions are deployed, we predict that there will be marked reduction (in the order of 60% to 74%) in the prevalence of exposure to T. solium cysticerci in this area of Vietnam. Dak Lak province has a population of approximately 1.8 million [40]; the survey adjusted apparent prevalence of exposure to T. solium cysticerci, as estimated in this study, at 5.01% equates to approximately 90,000 exposure-positive individuals. With a population attributable fraction for routine location of defaecation at 60%, and assuming an intervention program to educate inhabitants to practice defaecation indoors, we estimate that the number of exposure-positive individuals will be reduced to approximately 36,000 individuals if the program is 100% successful. Even with 50% efficacy of the intervention program, we estimate the number of exposure-positive individuals would be in the order of 63,000, a substantial reduction from the current prevalence estimate of 90,000. The expected savings to the public health sector (in terms of avoided diagnostics and treatments) is likely to be substantial. If both location of defaecation and raw vegetable consumption practices are modified concurrently, the expected population attributable fraction is 89%. Outdoor defaecation results in the contamination of the environment including water and vegetables; thus, to reduce or mitigate the burden of T. solium cysticerci exposure of inhabitants in the communities of Dak Lak province, this factor should be given the highest priority of implementation as an intervention program.

Taenia spp. infection risk in in this study was associated with gender, the consumption of pork tongue, consumption of undercooked pork and beef, and observation of Taenia proglottids shedding from either themselves, their relatives or neighbors (Table 4.5). Among these factors, the consumption of pork tongue, and consumption of undercooked pork and beef are modifiable. Consumption of undercooked beef had the highest population attributable fraction of 77% compared to other factors which ranged from 24% to 26%. Thus, an intervention program aimed at discouraging practices of consuming undercooked beef could potentially result in a 77% reduction in the number of cases of taeniasis in the region. The strong association of undercooked beef consumption with Taenia positivity could explain the dominant prevalence of T. saginata in this study population at 5.8% compared to 1.2% with T. solium and 1.5%

149

with T. asiatica [14], because beef containing T. saginata cysticerci is the cause of T. saginata taeniasis in humans.

Males had a higher likelihood of taeniasis compared to females. In rural communities of Dak Lak, males commonly consume undercooked beef, pork, or pork visceral organs while drinking alcohol as part of traditional cultural practices. Thus, educating the community on appropriate consumption of meat, targeting male members of the community is expected to have a substantial impact on the prevalence of taeniasis, although altering cultural dietary practices is recognized as a challenging approach to successful intervention. The most notable limitation of this study was the low prevalence of infection with individual species of Taenia, namely, T. solium, T. saginata and T. asiatica in study communities. This deterred our ability to build a model to predict risk factors associated with species-level Taenia infection in this study.

A three-stage cluster sampling design was used for this study with villages within districts as the primary sampling unit, households within villages as the secondary sampling unit and individuals within households as the tertiary sampling unit. Given the hierarchical structure of our data it was our a priori belief that observations made on individual study participants were not likely to be independent, justifying the use of regression modelling approaches that accounted for this non-independence. In our mixed-effects logistic regression model the variance of the random effect terms at both the village and individual household levels were negligible. Our inference is that in this population most of the unmeasured Taenia/cysticerci-exposure risks resided at the individual (as opposed to either the village or household) level. The practical implications of this finding are that in addition to addressing the modifiable risk factors identified in these analyses, health education programs need to be focused at the individual level.

Since many of the identified risk factors are traditional/behavioural practices in this area of Vietnam, the probability to induce permanent behavioural change from a single education campaign is likely to be small. A One Health approach utilising the expertise of anthropologists might be one way of effecting permanent behavioural change. Long term behavioural change in a population and on-going political support for health 150

interventions occur when aspects of health risk, the economic benefits of disease control and the economic costs associated with disease can be quantified [41]. Financial and human resources are critical factors influencing the success of disease intervention programs. If resources are not available to support an integrated, long-term control and eradication program, a selective approach provides a suitable alternative with the selection of those interventions that provide the most cost-effective return on investment being preferred [42, 43]. The resource for taeniasis, cysticercosis intervention programs in Dak Lak province is limited; therefore, a selective approach is necessary. As discussed above, health education programs involving both the human medical and veterinary health sectors are a feasible intervention option. Choosing a small number of risk factors that make the greatest contribution to the prevalence of taeniasis or cysticercosis are logical targets within an education campaign, for example, indoor latrine construction will ultimately prove the most cost-effective approach.

4.6 Conclusions

The prevalence of seroexposure to T. solium in the communities of Dak Lak is consistent with that of other regions in Vietnam. Identified risk factors are modifiable, and associated strongly with the prevalence of T. solium seroexposure and Taenia spp. infection in Dak Lak communities, thus advocating for targeted community intervention programs in mitigating these risks.

4.7 Declarations

4.7.1 Ethics approval and consent to participate

This study was reviewed and approved by The University of Melbourne Human Research Ethics Committee through (Ethics ID numer 1443512), and conducted under the supervision of the local Center for Public Health, Dak Lak, Vietnam. Informed and voluntary consent of each participant (or guardian) was obtained prior to their participation and inclusion in the research project through a Plain Language Statement and the participants consent established by use of a Consent Form.

151

4.7.2 Funding

This research was self-funded by RJT. This work was done with partial support for travel from the Faculty of Veterinary and Agricultural Sciences, University of Melbourne, Australia. DNN received PhD scholarship from Australia Awards Scholarships, Department of Foreign Affairs and Trade, Australia Government.

4.7.3 Authors' contributions

DNN designed study, analyzed data and wrote manuscript; MAS: assisted with analyses of data and edited paper; KB, TVP, VTN and TVV: assisted laboratory work and fieldwork; RJT provided material, supervised study and edited paper. All authors read and approved the final manuscript.

4.7.4 Acknowledgements

We are grateful to the Institute of Biotechnology and Environment Tay Nguyen University for providing resources and facilities for the fieldwork. We thank Drs. John Noh and Sukwan Handali from the Center for Diseases Control and Prevention, GA, USA for EITB analysis. The Authors are grateful to village medical staff at Buon Don, Krong Nang and M’Drak district for assisting with sample collection. The authors are most thankful to Assoc. Prof. Than Trong Quang, Ms. Nguyen Thi Ngoc Hien, Ms. Long Khanh Linh, and Ms. Nguyen Thi Lan Huong, who assisted with laboratory work and fieldwork.

152

4.8 References

[1] FAO/WHO. Multicriterial-based Ranking for Risk Management of Food-borne Parasites. Microbiological Risk Assessment Series No 23. vol. 23. Rome: WHO Press; 2014.

[2] Ndimubanzi PC, Carabin H, Budke CM, Nguyen H, Qian YJ, Rainwater E, et al. A systematic review of the frequency of neurocyticercosis with a focus on people with epilepsy. PLoS Negl Trop Dis 2010;4:e870.

[3] Garcia HH, Nash TE, Del Brutto OH. Clinical symptoms, diagnosis, and treatment of neurocysticercosis. Lancet Neurol 2014;13:1202–15.

[5] WHO. Taeniasis/cysticercosis. WHO 2018. http://www.who.int/news-room/fact- sheets/detail/taeniasis-cysticercosis (accessed February 22, 2018).

[6] Aung AK, Spelman DW. Taenia solium taeniasis and cysticercosis in southeast Asia. Am J Trop Med Hyg 2016;94:947–54.

[7] Wu HW, Ito A, Ai L, Zhou XN, Acosta LP, Lee Willingham A. Cysticercosis/taeniasis endemicity in southeast Asia: Current status and control measures. Acta Trop 2015;165:121–32.

[8] NIMPE. Taeniasis and Cysticercosis. Rev. Work. helminthiasis Control Act. period 2006-2011 Implement. Work. period 2012-2015, Ministry of Health of Vietnam; 2012.

[9] Vien HV, Hung NM, Thach DTC, Thuan LK, Dung DT, Hop NT, et al. Investigate Taenia infection in two communes of Tan Son district, Phu Tho province. J Malar Parasite Dis Control 2012;1:69–74.

153

[10] Van Tuan B, Chuong N Van. Situation of Taenia saginata infestation and effectiveness of some control measures in Dak Mon commune (Dak Glei) and Yaxier commune (Sa Thay), Kon Tum province. J Malar Parasite Dis Control 2014;1:41–8.

[11] Anh Tuan P, Dung TTK, Nhi VA. Sero-epidemiological investigation of cysticercosis in the southern provinces. J Malar Parasite Dis Control 2001;4:81–7.

[12] Ng-Nguyen D, Stevenson MA, Traub RJ. A systematic review of taeniasis, cysticercosis and trichinellosis in Vietnam. Parasit Vectors 2017;10:150.

[13] Ng-nguyen D, Noh J, Breen K, Stevenson MA, Handali S, Traub RJ. The epidemiology of porcine Taenia solium cysticercosis in communities of the Central Highlands in Vietnam. Parasit Vectors 2018;11:360.

[14] Ng-Nguyen D, Stevenson MA, Dorny P, Gabriël S, Vo T Van, Nguyen VT, et al. Comparison of a new multiplex real-time PCR with the Kato Katz thick smear and copro-antigen ELISA for the detection and differentiation of Taenia spp. in human stools. PLoS Negl Trop Dis 2017;11:e0005743.

[15] General Statistics Office of Vietnam. Percentage of literate population at 15 years of age and above by province 2016. https://www.gso.gov.vn/default_en.aspx?tabid=774 (accessed January 8, 2018).

[16] Van De N, Le TH, Lien PTH, Eom KS. Current status of taeniasis and cysticercosis in Vietnam. Korean J Parasitol 2014;52:125–9.

[17] Asaava LL, Kitala PM, Gathura PB, Nanyingi MO, Muchemi G, Schelling E. A survey of bovine cysticercosis/human taeniosis in Northern Turkana District, Kenya. Prev Vet Med 2009;89:197–204.

[18] Tsang VCW, Brand JA, Boyer AE. An enzyme-linked immunoelectrotransfer blot assay and glycoprotein antigens for diagnosing human cysticercosis (Taenia

154

solium). J Infect Dis 1989;159:50–9.

[19] Noh J, Rodriguez S, Lee YM, Handali S, Gonzalez AE, Gilman RH, et al. Recombinant protein- and synthetic peptide-based immunoblot test for diagnosis of neurocysticercosis. J Clin Microbiol 2014;52:1429–34.

[20] Jeon HK, Chai JY, Kong Y, Waikagul J, Insisiengmay B, Rim HJ, et al. Differential diagnosis of Taenia asiatica using multiplex PCR. Exp Parasitol 2009;121:151–6.

[21] Lumley T. Complex Surveys A Guide to Analysis Using R. New York, USA: John Wiley & Sons, Ltd.; 2010.

[22] Bates D, Machler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. J Stat Softw 2015;67:1–48.

[23] Team R Core. R: A Language and Environment for Statistical Computing 2017.

[24] Lumley T. Analysis of complex survey samples. J Stat Softw 2004;9:1–19.

[25] Whittemore AS. Statistical methods for estimating attributable risk from retrospective data. Stat Med 1982;1:229–43.

[26] Toschke AM, Rückinger S, Böhler E, Von Kries R. Adjusted population attributable fractions and preventable potential of risk factors for childhood obesity. Public Health Nutr 2007;10:902–6.

[27] Trung DD, Praet N, Cam TDT, Lam BVT, Manh HN, Gabriël S, et al. Assessing the burden of human cysticercosis in Vietnam. Trop Med Int Heal 2013;18:352–6.

[28] Somers R, Dorny P, Nguyen VK, Dang TCT, Goddeeris B, Craig PS, et al. Taenia solium taeniasis and cysticercosis in three communities in north Vietnam. Trop Med Int Heal 2006;11:65–72.

155

[29] Doanh NQ, Kim NT, De N V., Lung NL. Result of survey on taeniasis and cysticercosis humans and pigs in Bac Ninh and Bac Kan provinces. Vet Sci Tech 2002;9:46–9.

[30] Erhart A, Dorny P, Van De N, Vien H V, Thach DC, Toan ND, et al. Taenia solium cysticercosis in a village in northern Viet Nam: Sero-prevalence study using an ELISA for detecting circulating antigen. Trans R Soc Trop Med Hyg 2002;96:270–2.

[31] Ho ST. Study on genotype of pathogen, clinical, subclinical symptoms, treatment efficacy for taeniasis and cysticercosis patients in National Institute of Malariology, Parasitology and Entomology 2007-2010. Hanoi Medical University, 2013.

[32] Openshaw JJ, Medina A, Felt SA, Li T, Huan Z, Rozelle S, et al. Prevalence and risk factors for Taenia solium cysticercosis in school-aged children: A school based study in western Sichuan, People’s Republic of China. PLoS Negl Trop Dis 2018;12:e0006465.

[33] Kabululu ML, Ngowi HA, Kimera SI, Lekule FP, Kimbi EC, Johansen MV. Risk factors for prevalence of pig parasitoses in Mbeya Region, Tanzania. Vet Parasitol 2015;212.

[34] Prasad KN, Verma A, Srivastava S, Gupta RK, Pandey CM, Paliwal VK. An epidemiological study of asymptomatic neurocysticercosis in a pig farming community in northern India. Trans R Soc Trop Med Hyg 2011;105:531–6.

[35] Cherian A, Syam UK, Sreevidya D, Jayaraman T, Oommen A, Rajshekhar V, et al. Low seroprevalence of systemic cysticercosis among patients with epilepsy in Kerala-South India. J Infect Public Health 2014;7:271–6.

[36] Rebecca WP, Eugene II, Joshua K. Seroprevalence of antibodies (IgG) to Taenia solium among pig rearers and associated risk factors in Jos metropolis, Nigeria. J Infect Dev Ctries 2013;7:67–72.

156

[37] Mwanjali G, Kihamia C, Kakoko DVC, Lekule F, Ngowi H, Johansen MV, et al. Prevalence and risk factors associated with human Taenia solium infections in Mbozi district, Mbeya region, Tanzania. PLoS Negl Trop Dis 2013;7:e2102.

[38] Garcia-Noval J, Moreno E, de Mata F, Soto de Alfaro H, Fletes C, Craig PS, et al. An epidemiological study of epilepsy and epileptic seizures in two rural Guatemalan communities. Ann Trop Med Parasitol 2001;95:167–75.

[39] Edia-Asuke AU, Inabo HI, Mukaratirwa S, Umoh VJ, Whong CM, Asuke S, et al. Seroprevalence of human cysticercosis and its associated risk factors among humans in areas of Kaduna metropolis, Nigeria. J Infect Dev Ctries 2015;9:799–805.

[40] Dak Lak Provincial People’s Committee. An overview about Dak Lak province 2014. http://daklak.gov.vn/portal/page/portal/daklak/tong-quan- daklak?p_page_id=13371692&pers_id=&folder_id=15937106&item_id=159376 43&p_details=1.

[41] Gabriël S, Dorny P, Mwape KE, Trevisan C, Braae UC, Magnussen P, et al. Control of Taenia solium taeniasis/cysticercosis: The best way forward for sub- Saharan Africa? Acta Trop 2017;165:252–60.

[42] Winskill P, Harrison WE, French MD, Dixon MA, Abela-Ridder B, Basáñez M- G. Assessing the impact of intervention strategies against Taenia solium cysticercosis using the EPICYST transmission model. Parasit Vectors 2017;10:73.

[43] O’Neal SE, Moyano LM, Ayvar V, Rodriguez S, Gavidia C, Wilkins PP, et al. Ring-screening to control endemic transmission of Taenia solium. PLoS Negl Trop Dis 2014;8:e3125.

157

Table 4.1 Structure of the data from 342 study participants from six villages in M’Drak, Buon Don and Krong Nang districts

Number at the next highest level Level Number Mean Range Districts 3 - - Villages 6 2 2 Households 190 31 18 to 40 Humans a 342 2 1 to 5 a A total of 342 individuals participated in this study. The mean of individuals per household was 2 (range 1 to 5).

Table 4.2 General description of household data

Characteristic Frequency Percentage (95% CI) Number of households 190 District: M'Drak 58 30 (24 to 38) Buon Don 66 35 (28 to 42) Krong Nang 66 35 (28 to 42) Owning pigs: Yes 37 19 (14 to 26) No 153 81 (74 to 86) Presence of latrine: Yes 47 25 (19 to 32) No 143 75 (68 to 81) Source of drinking water: Pipe/well/rain water 142 74 (66 to 79) Lake/stream/pond 48 26 (19 to 32) Water treatment: Yes 75 39 (32 to 47) No 115 60 (53 to 67) Are vegetables fertilized? Yes 81 43 (35 to 50) No 109 57 (50 to 64)

158

Table 4.3 Demographic data for individual study participants

Characteristic Frequency Percentage (95% CI) Total number of participants 342 Age (years): < 25 63 18 (14 to 23) 25 – 60 223 65 (60 to 70) > 60 56 16 (13 to 21) Sex: Male 104 30 (26 to 36) Female 238 70 (64 to 74) Highest level of education: None 149 44 (38 to 49) Primary/secondary 184 54 (48 to 59) Tertiary 9 2.6 (1.3 to 5.1) Primary activity: Farming 298 87 (83 to 90) Livestock 14 4.1 (2.3 to 6.9) Civil service 5 1.5 (0.5 to 3.6) Other 25 7.3 (4.9 to 11) Ethnicity: Kinh 40 11 (8.6 to 15) Ede 302 88 (84 to 91)

159

Table 4.4 Risk factors associated with T. solium cysticerci exposure identified by a LLGP-EITB assay

Explanatory Number Number of Regression Adjusted OR P value variable positive individuals coefficient (SE) (95% CI) Intercept 17 339 -6.4694 (1.0615) <0.01 - Raw vegetable <0.01 consumption: Occasionally 4 226 Reference 1.00 Often 13 113 2.3006 (0.6318) 9.98 (2.89 to 34.4) a Area of defaecation b: <0.01

Close latrine 5 228 Reference 1.00 Outdoor 12 110 2.4125 (0.6763) 11.1 (2.97 to 42.0) Source of drinking <0.05 water: Other c 9 253 Reference 1.00 Stream/Lake/Pond 8 86 1.3702 (0.5684) 3.94 (1.29 to 11.9) a Interpretation: After adjusting for other explanatory variables in the model, the odds of an individual from any household (in any village in any district) who consumed raw vegetables often being T. solium cysticerci exposure was 9.98 (95% CI 2.89 to 34.4) times that of an individual from any household (in any village in any district) who consumed rare vegetables rarely. b A single individual with a missing value for area of defaecation excluded. c Drinking water sourced from wells, pipes, rainy and packed water

160

Table 4.5 Logistic regression model showing the effect of sex, dietary and observation of proglottids on the odds of a taeniasis case

Number Number of Regression Adjusted OR Explanatory variable P value positive individuals coefficient (SE) (95% CI) -6.6069 Intercept 23 342 < 0.01 - (1.0365) Sex: < 0.05 Female 8 238 Reference 1.00 2.77 (1.00 to Male 15 104 1.0183 (0.5209) 7.68) a Undercooked pork < 0.05 consumed: No 9 241 Reference 1.00 3.68 (1.20 to Yes 14 101 1.3023 (0.5695) 11.2) Pork tongue consumed: < 0.01

Rare 15 305 Reference 1.00 4.62 (1.49 to Often 8 37 1.5309 (0.5778) 14.3) Undercooked beef < 0.01 consumed: No 2 154 Reference 1.00 6.92 (1.42 to Yes 21 188 1.935 (0.8070) 33.7) Proglottids observed: < 0.01

No 12 302 Reference 1.00 13.36 (4.25 to Yes 11 40 2.592 (0.5842) 41.9) a Interpretation: After adjusting for other explanatory variables in the model, the odds of an individual who is male being taeniasis infection was 2.77 (95% CI 1.00 to 7.68) times that of an individual who is female.

161

CHAPTER 5

The epidemiology of porcine Taenia solium cysticercosis in communities of the Central Highlands in Vietnam

Presented as published:

162

5.1 Abstract

5.1.1 Back ground

Taenia solium cysticercosis, recognized as a neglected tropical disease by the WHO, is distributed mostly in developing countries of Latin America, sub-Saharan Africa and Asia. Pigs and humans act as intermediate hosts, acquiring T. solium cysticerci (larval stage) in their tissue, through the ingestion of T. solium eggs shed in the faeces of humans infected with adult tapeworms. The disease has a negative impact on rural economies due to losses in productivity arising from human disease, pork carcass condemnations and loss of market access. The aim of this study was to estimate the prevalence of T. solium cysticercosis in pigs in Dak Lak Province in the Central Highlands of Vietnam and to identify household level characteristics associated with T. solium porcine cysticercosis.

5.1.2 Methods

This was a cross-sectional study of household pigs in three districts of Dak Lak Province. A total of 408 households in six villages in three districts were visited between June and October 2015. A questionnaire was administered to the head of each household, and within each household, serum samples were collected from three pigs. Serum samples were analyzed using the recombinant T24H antigen in enzyme-linked immunoelectrotransfer blot assay and lentil lectin purified glycoprotein in EITB assay. A Bayesian, mixed-effects logistic regression model was developed to identify management factors associated with the probability of a household having at least one cysticercosis-positive pig.

5.1.3 Results

The seroprevalence of porcine T. solium cysticercosis in this study was low at 0.94 [95% confidence interval (CI) 0.51–1.68] cases per 100 pigs at risk, in agreement with other studies conducted throughout Vietnam. Scavenging of food and coprophagy were associated with T. solium cysticercosis (odds ratios 7.24 [95% CrI 1.73 to 114] and 13.0 [95% CrI 3.38 to 105], respectively).

163

5.1.4 Conclusions

This study proves that the seroprevalence of porcine cysticercosis in Dak Lak Province was as low as that of other studies conducted throughout Vietnam. Scavenging of food and coprophagy are modifiable factors, providing the opportunity to decrease the prevalence of porcine cysticercosis further in the province.

5.2 Background

Taenia solium cysticercosis is recognized as a neglected tropical disease by the WHO [1]. It is distributed mostly in developing countries of Latin America, sub-Saharan Africa and Asia [2]. Pigs and humans act as intermediate hosts, acquiring T. solium cysticerci (larval stage) in their tissue, through the ingestion of T. solium eggs shed in the faeces of humans infected with adult tapeworms. Consumption of raw and/or undercooked pork with active T. solium cysticerci may result in T. solium taeniasis in humans. The presence of porcine cysticercosis impacts negatively on an economy due to costs arising from carcass condemnation and negative impacts on market access and trade of pork.

Although Vietnam is located in a region endemic for T. solium [3], the data on porcine cysticercosis are limited. In addition, it is not clear whether porcine cysticercosis is endemic in the country. During 1994 to 2005, the highest reported prevalence of porcine cysticercosis in Vietnam among four carcass-based studies was less than 1% [4]. These estimates were, for the most part, based on studies conducted in commercial slaughterhouses, and therefore do not reflect the prevalence of cysticercosis in pig populations in rural communities that are not processed in commercial slaughterhouses [5]. Of the relatively small number of studies that have quantified the prevalence of porcine cysticercosis in the country, most were conducted prior to 2003 with a focus on the north of Vietnam [4]. To the best of our knowledge, the only study of porcine cysticercosis in the south of Vietnam was carried out in 1994 [6].

Dak Lak Province is located in the Central Highlands in the south of Vietnam. A recent study in the communities of the province demonstrated an apparent prevalence of 1.2% T. solium taeniasis [7]. In most communities in the province, pigs are free-roaming and 164

outdoor defaecation is common [8]. These characteristics are conducive for infection and transmission of T. solium cysticercosis to pigs. The aim of this study was to estimate the prevalence of cysticercosis in pigs in Dak Lak and to identify household level characteristics associated with T. solium porcine cysticercosis.

5.3 Materials and Methods

5.3.1 Study site and sampling

Fieldwork was conducted between June and October 2015 in Krong Nang, M’Drak and Buon Don districts in Dak Lak Province, Vietnam. These districts were chosen as the study sites based on their diverse geographical characteristics. The study sites have been described in detail elsewhere [7]. In brief, M’Drak is located in the east of Dak Lak Province with an average altitude of 400 m to 500 m and has a tropical monsoon climate typical of the Vietnam’s Central Coast. Krong Nang is situated in the north of the province at an altitude of 800 m. Buon Don, situated to the west of the province with an average elevation of 330 m and has a hot and dry climate. The standard of living in this area of Vietnam is generally poor. Open defaecation using outdoor pit latrines is common practice and livestock access to these latrine areas is usually unrestricted. The practice of non-confinement of pigs and cattle is common with slaughter activities often carried out in backyards [4].

A sampling frame listing the name of all villages in the three study districts was obtained from Sub-Department of Animal Health office. Villages eligible for sampling comprised those with more than 1000 pigs, as recorded by the Sub-Department of Animal Health within the Ministry of Agriculture. All eligible villages within each of the three study districts were assigned a number and two numbers were chosen at random to select villages from each district for inclusion in the study. Within each selected village, a list of householder names was obtained from the respective village head person, and each householder name was then coded with a number (the number of households per village ranged from 200 to 300). A sheet of paper was drawn up into squares and each square numbered from 1 to 300. The squares were cut into pieces and placed face-down on a table. The village head was then asked to select between 100 and 140 squares. The number on each selected square identified each household to be 165

sampled. All selected households were visited several days before sampling to obtain consent from participants. Within each district, blood samples were collected from pigs in each of the study households. At the time of each household visit pig owners were asked to complete a questionnaire on the number and type of pigs kept and details of demography, husbandry practices, and diet (Table 5.1). All questionnaires were conducted in local Vietnamese phraseology and their validity pre-tested on 30 pig owners in another community in Dak Lak Province before application to the field survey. In addition, interviewers were trained before administering the questionnaires. Pigs were selected at random by the member of research group for blood sampling. Pigs that were pregnant or ill, and pigs aged less than 2 months of age were excluded from sampling. At the time of each household visit 10 ml of blood was obtained from the cranial vena cava of each pig into plain blood collection tubes. The blood samples were allowed to clot at ambient temperature prior to centrifugation at 3200× g for 5 min to separate serum. Serum was dispensed into 1.5 ml aliquots and stored at -20 °C until use.

5.3.2 Sample size estimation

The aim of this study was to estimate the prevalence of porcine T. solium cysticercosis in Dak Lak Province. Based on a previous slaughterhouse based survey by Van De at al. [9], the prevalence of porcine cysticercosis was assumed to be 10%. Assuming 95% certainty that this estimate was within 5% of the actual population prevalence (i.e. cysticercosis prevalence ranged from 5% to 15%) and ignoring the possibility that cysticercosis positive pigs were clustered within households, we estimated that a total of 384 pigs were required to be sampled. We then assumed the average number of pigs eligible for random sampling per household was at least three and an intra-class correlation coefficient for T. solium cysticercosis of 0.07 [10] returning a design effect of 1.14. Our revised sample size, accounting for the likelihood that porcine cysticercosis clusters within households, was 384 × 1.14 = 438 for each of the three study districts.

5.3.3 Laboratory procedures

Pig serum samples were analyzed using the recombinant T24H antigen in enzyme- linked immunoelectrotransfer blot (rT24H-EITB) assay as described by Noh et al. [11]

166

and lentil lectin purified glycoproteins in EITB (LLGP-EITB) assay as previously described by Tsang et al. [12] and Gonzalez et al. [13]. Both the LLGP-EITB and rT24H-EITB assay are immunoblot methods. The LLGP-EITB detects antibodies to one or more of seven lentil lectin purified glycoproteins (LLGPs), namely GP50, GP42, GP24, GP21, GP18, GP14 and GP13 which are present in the soluble fraction of an extract of T. solium cysticerci [11]. Reaction to any of these 7 glycoprotein antigens is considered positive. The rT24H-EITB assay detects antibodies against rT24H antigen derived from 24- and 42-kDa glycoproteins of the LLGPs [14].

To ascertain the analytical specificity of the rT24H antigen, we subjected 29 cysticercosis-negative USA pig sera, 12 necropsy-positive T. solium-positive Peruvian pig sera and 4 T. hydatigena necropsy-positive Vietnamese pig sera to the rT24H-EITB. All USA pig sera and Vietnamese T. hydatigena pig sera were negative for the rT24H- EITB, and all Peruvian T. solium positive pig sera were positive on the rT24H-EITB. These preliminary results provided the basis of results show that under experimental conditions, the rT24H-EITB do not cross-react to pig sera with T. hydatigena.

Individual serum samples were screened in pools of four using the rT24H antigen in EITB assay format to detect the presence of antibodies against T. solium cysticerci. The rT24H antigen was utilized in the EITB assay as it offers the best overall diagnostic performance compared with other recombinant or synthetic antigens based on our preliminary data and previous human-based studies [15, 16]. Individual serum sample from each positive pools were then re-screened using rT24H-EITB and LLGP (native antigen of T. solium cysticerci) antigens. The LLGP-EITB has been used as the reference standard assay for serological diagnosis of T. solium cysticercosis in humans and pigs that has the specificity of 100% and sensitivity between 98–100% [12, 13]. Jayashi et al. [17] when validating the LLGP-EITB for naturally acquired porcine cysticercosis pointed out that the LLGP-EITB achieves optimal sensitivity of 78% (95% CI: 52–94%) and specificity of 76% (95% CI: 66–85%) when the assay reacts to ≥ 3 of 7 LLGP antigens. The diagnostic performance of LLGP-EITB for porcine T. solium cysticercosis was evaluated in Peruvian pigs, and the cross-reaction of the assay to T.

167

hydatigena was not known [13]. An individual serum sample was considered positive for T. solium antibodies if it was positive to both rT24H and native LLGP antigens.

5.3.4 Statistical analysis

Risk factors for porcine T. solium cysticercosis in the communities of Dak Lak Province were identified using logistic regression. In this study, the outcome of interest was a dichotomous variable where households where at least one pig was T. solium cysticercosis-positive were assigned a value of 1, and 0 otherwise. The association between each of a set of household-level candidate explanatory variables from the questionnaires and the outcome of interest were tested using unconditional odds ratios and the chi-square test. All explanatory variables associated with the outcome of being T. solium cysticercosis positive at an alpha level of < 0.20 using the chi-square test were selected for inclusion in the multivariable model.

A frequentist fixed-effects logistic regression model was developed in which the probability of a household having at least one cysticercosis-positive pig was parameterized as a function of the explanatory variables with significance of the chi- square test at P < 0.20, as described above. The significance of each explanatory variable in the model was tested using the chi-square test. Explanatory variables that were not statistically significant were removed from the model one at a time, beginning with the least significant, until the estimated regression coefficients for all the explanatory variables retained in the model were significant at an alpha level of < 0.05.

Frequency histograms of the residuals from the fixed-effects model and plots of the residuals versus predicted values were constructed to check that the assumptions of normality and homogeneity of variance had been met and to identify the presence of influential observations in the data

To account for the hierarchical structure of the data (households within villages) a village-level random effect term (푉), was included in the model as shown in Equation 1. District was not a significant predictor of household level T. solium cysticercosis

168

status at the alpha level of 0.05 and was therefore not considered further in the mixed- effects model.

푙표푔 = 훽 + 훽 푥 +⋯ + 훽푥 + 푉 + 휀 Equation 1

Due to the low prevalence of T. solium cysticercosis (12 of 1281 pigs were positive) regression coefficients for the mixed-effects logistic regression model were estimated using a Bayesian approach implemented in JAGS [18, 19]. Flat (uninformed) prior distributions were assumed for the intercept 훽 and each of the regression coefficients for the fixed effects 훽 … 훽. The village-level random effect term (푉) was parameterized as having a normal distribution with mean 0 and precision (inverse variance) 휏.

For each of the Bayesian regression analyses we ran the Markov chain Monte Carlo sampler for 40,000 iterations and discarded the first 1000 ‘burn-in’ samples. Convergence was visually assessed by plotting cumulative path plots for each of the monitored parameters [20, 21] and quantified using the Raftery & Lewis convergence diagnostic [22, 23]. Parallel chains were run using diverse initial values to ensure that convergence was achieved to the same distribution [24]. Posterior sample sizes were determined by running sufficient iterations to ensure that the Monte Carlo standard error of the mean was at least one order of magnitude smaller than the posterior standard deviation for each parameter of interest.

The results of the final mixed-effects logistic regression model are reported in terms of adjusted odds ratios for each explanatory variable. Assuming a causal relationship between a given explanatory variable and porcine cysticercosis, an adjusted odds ratio [and its 95% credible interval (CrI)] of > 1 indicates that, after adjusting for other variables in the model, the explanatory variable increased the risk of a pig being cysticercosis positive. An adjusted odds ratio (and its 95% CrI) of < 1 indicates that exposure to the explanatory variable was protective, and an OR of 1 indicates that the variable was not associated with porcine cysticercosis risk.

169

Statistical analyses were performed using the packages R2jags [25] and coda [26] implemented in R version 3.3.0 [27].

5.4 Results

5.4.1 General description of study population

A total of 1324 pig serum samples were collected in Krong Nang, M’Drak and Buon Don districts in Dak Lak Province. Of these, 1281 samples were eligible for further examination as 43 samples were excluded from analysis due to hemolysis or missing questionnaire information. All 1281 serum samples were screened in pools of 4 using the rT24H-EITB assay, and 10 pool samples were identified as positive. Twelve single samples among the 10 positive pool samples were positive for T. solium antibodies using the rT24H-EITB assay and all 12 single samples were positive using LLGP-EITB. The prevalence of T. solium cysticercosis in pigs in the study districts was 0.94 (95% CI: 0.51–1.68) per 100 pigs at risk. The 12 positive pigs belonged to 11 households in the three study districts. Of 203 households visited in M’Drak district, 9 (4.43%, 95% CI: 2.17–0.85%) possessed T. solium seropositive pigs. Among 70 visited households in Krong Nang district, two (2.8%, 95% CI: 0.49–10.8%) possessed T. solium cysticercosis positive pigs, and no seropositive pigs were identified in 135 households in Buon Don District.

The hierarchical structure of the data in this study is shown in Table 5.2. The 1281 pigs were from 408 households and, within each household, an average of three pigs were sampled (minimum 1; maximum 24).

Of the 408 households, 266 (65%, 95% CI: 60–70%) used a pit latrine; however, most of these latrines were of temporary construction, which animals were able to access. A total of 35% (95% CI: 30–40%) of householders responded that their family members practiced outdoor defection; children typically defaecated around the main household building while adults defaecated some distance from the main household building, within the confines of the household property. Seventy-six percent (95% CI: 72–80%) of households had a pigsty and 58% (95% CI: 53–63%) confined their pigs at all times. Free-roaming pigs habitually ranged around the village to seek food and to return to 170

their litters located under stilt housing in the afternoon. Approximately 7% (95% CI: 5– 10%) of households used water sourced from lakes, streams or ponds for their pigs. The remaining households used either rainwater or water from wells or pipes. Dogs were kept in 63% (95% CI: 59–68%) of the households that owned pigs (Table 5.3).

Among the 1281 pigs that were sampled, 41% (95% CI: 38–44%) were of local breed (Soc). A total of 27% (95% CI: 24–29%) of the sampled pigs were reported to regularly consume human faeces. A little over half of the pigs were routinely offered raw, unwashed vegetables (57%; 95% CI: 54–59%). Commercial and/or homemade bran was offered to 89% (95% CI: 86–91%) of pigs. The small proportion of pigs that were not supplied bran (11%, 95% CI: 9–12%) were scavenging for the most part (Table 5.4).

5.4.2 Risk factors for porcine T. solium cysticercosis

Of the data recorded using the questionnaire, we identified two factors associated with a pig’s likelihood of being T. solium cysticercosis positive: (i) frequent coprophagy of human faeces; and (ii) scavenging for food. Estimated regression coefficients for the mixed-effects logistic regression model provided in Table 5.5. After adjusting for the other explanatory variables in the model, the odds of a household where pigs routinely consumed human faeces being T. solium cysticercosis positive was 13.0 (95% CrI 3.38 to 105) times that of a household where pigs did not consume human faeces. The odds ratio for a household where pigs routinely scavenged for food was 7.24 [95% CrI 1.73 to 114].

5.5 Discussion

In low-income rural communities of Vietnam, a substantial proportion of the human population practice open defaecation and uncooked pork and beef consumption is relatively common. Allowing pigs to roam freely is a common husbandry practice in the region [4, 5, 9, 28, 29]. Despite all relevant risk factors for cysticercosis being present in each of the communities in this study, the prevalence of T. solium cysticercosis was low at 0.94 (95% CI: 0.51–1.68) cases per 100 pigs at risk. A similar, low prevalence of cysticercosis has been observed not only in Dak Lak Province but in other regions of Vietnam [4, 6, 30]. In 1994 Huan, in a cross-sectional study of pigs submitted for slaughter from 12 provinces in the south of Vietnam, reported a prevalence of 0.90 171

(95% CI: 0.45–1.76) cases per 100 pigs at risk [6]. In three studies carried out in 10 provinces in the north of Vietnam between 1999 and 2003, the prevalence of cysticercosis ranged from 0 to 0.06 cases per 100 pigs at risk [31–33]. In the Province of Bac Ninh in the north of Vietnam, a known foci of T. solium in humans, carcass examination of 26 village pigs identified no cases of T. solium cysticercosis. Instead, 10 pigs were positive for T. hydatigena cysticerci [34]. These findings are in agreement with those of Conlan et al. [35], who reported a low prevalence of T. solium cysticercosis of 0.8% in village pigs in Laos with a relatively high prevalence of T. hydatigena (22 cases per 100 pigs at risk) and a high prevalence of T. solium taeniasis. Conlan et al. [35] hypothesized that T. hydatigena is likely to cross-protect pigs from T. solium infection.

In Vietnam the prevalence of T. hydatigena cysticercosis in pigs has been reported to be high, ranging between 25–38% in the north and the prevalence has been strongly correlated with the presence of T. hydatigena infection in dogs [36]. In addition, most of the households that owned pigs in the three study districts also kept dogs and approximately one half of the households owning dogs reported that they had observed proglottids in their dog’s faeces (Table 5.3). All of the 10 free-roaming village pigs that were backyard slaughtered had T. hydatigena cysticerci present in the mesentery, stomach, spleen, and liver (personal observation from fieldwork). It is our inference that the relatively high prevalence of T. hydatigena infection in dogs is likely to be a major source of T. hydatigena cysticercosis in pigs. If this is true a cautioned approach to cysticercosis control in pigs would be advised if T. hydatigena were to be targeted through pig confinement and canine deworming programs. Eliminating or reducing pig exposure to T. hydatigena would likely result in an increase in the observed prevalence of T. solium in non-compliant free-roaming pigs, providing an increased public health risk to the community.

Among the investigated households that owned pigs, a little under one-quarter did not have a pigsty and most of the pigs that were kept were of the local breed (Table 5.4). Local breeds are preferred due to their ability to thrive under harsh raising conditions and poor feeding [12]. Allowing pigs to free-roam for food was common practice (Table

172

5.3). The odds of a household where pigs routinely scavenged for food being seropositive for T. solium cysticercosis was 7.24 (95% CrI 1.73 to 114) times greater than the odds of a household where pigs were fed commercial and/or homemade bran (Table 5.5). Similarly, the odds ratio for a household where pigs routinely consumed human faeces being seropositive was 13.0 (95% CrI 3.38 to 105) times greater than the odds of a household where this practice was not habitual. It is known that risk factors for transmission and circulation of porcine cysticercosis are numerous and may vary in different settings. We found that allowing pigs to free-roam and allowing pigs to routinely consume human faeces were associated with T. solium exposure, consistent with other studies [37–39]. Research on the epidemiological characteristics of porcine cysticercosis in Peru [40], Mozambique [41] and Mexico [42] found that older pigs were more likely to show evidence of exposure to T. solium compared to younger pigs, an association not identified in this study. Similarly, while studies from Zambia [43], Mexico [44], and Tanzania [45] showed that the prevalence of T. solium exposure was higher in male pigs compared with females, this association was not identified in this study. In this study, both of the risk factors identified are modifiable, which means that there exists an opportunity to decrease the prevalence of porcine cysticercosis even further. We propose that a combination of intervention measures including education and public awareness campaigns, strategies to reduce coprophagy among pigs and enhanced meat inspection (particularly backyard slaughtered stock) are likely to have the greatest impact on porcine cysticercosis risk with positive secondary effects on human health [46]. For this to be successful there is a need for commitment and support from local and/or central veterinary and medical health authorities.

5.6 Conclusions

The prevalence of porcine T. solium cysticercosis in this study was low at 0.94 (95% CI: 0.51–1.68) cases per 100 pigs at risk, in agreement with other studies conducted throughout Vietnam. Scavenging of food and coprophagy were associated with T. solium cysticercosis risk. Both of these characteristics are modifiable providing the opportunity to decrease the prevalence of porcine cysticercosis even further.

173

5.7 Ethics approval and consent to participate

This study was reviewed and approved by the Animal Ethics and scientific committee, Tay Nguyen University (reference number 50.KCNTY), and conducted under the supervision of the local Centre for Animal Health, Dak Lak, Vietnam. If a pig was found seropositive for T. solium, the local veterinary office was notified in order to follow up with carcass inspection post-slaughtered. The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

174

5.8 References

[1] FAO/WHO. Multicriterial-based Ranking for Risk Management of Food-borne Parasites. Microbiologycal Risk Assessment Series No 23. vol. 23. Rome: WHO press; 2014.

[2] Donadeu M, Lightowlers MW, Fahrion AS, Kessels J, Donadeu M, Lightowlers MW, et al. Taenia solium: WHO endemicity map update Taenia solium: carte d’endémicité actualisée de l’OMS pour 2016 2016;4:595–9.

[3] Aung AK, Spelman DW. Taenia solium taeniasis and cysticercosis in southeast Asia. Am J Trop Med Hyg 2016;94:947–54.

[4] Ng-Nguyen D, Stevenson MA, Traub RJ. A systematic review of taeniasis, cysticercosis and trichinellosis in Vietnam. Parasit Vectors 2017;10:150.

[5] Huong NT. Taeniasis and Cysticercosis in a Selected Group of Inhabitants from a Mountainous Province in North Vietnam. Prince Leopold Institute of Tropical Medicine, Antwerpen (Antwerp), Belgium, 2006.

[6] Huan L Van. Parasitic Helminths in Pigs in Several Southern Provinces and Preventative Measures. National Institution of Veterinary Research, 1994.

[7] Ng-Nguyen D, Stevenson MA, Dorny P, Gabriël S, Vo T Van, Nguyen VT, et al. Comparison of a new multiplex real-time PCR with the Kato Katz thick smear and copro-antigen ELISA for the detection and differentiation of Taenia spp. in human stools. PLoS Negl Trop Dis 2017;11:e0005743.

[8] Nguyen TH. Evaluation of market opportunities for producing local pigs in Dak Lak. Tay Nguyen J Sci 2009;5:21–6.

[9] Van De N, Le TH, Lien PTH, Eom KS. Current status of taeniasis and cysticercosis in Vietnam. Korean J Parasitol 2014;52:125–9.

175

[10] Asaava LL, Kitala PM, Gathura PB, Nanyingi MO, Muchemi G, Schelling E. A survey of bovine cysticercosis/human taeniosis in Northern Turkana District, Kenya. Prev Vet Med 2009;89:197–204.

[11] Noh J, Rodriguez S, Lee YM, Handali S, Gonzalez AE, Gilman RH, et al. Recombinant protein- and synthetic peptide-based immunoblot test for diagnosis of neurocysticercosis. J Clin Microbiol 2014;52:1429–34.

[12] Tsang VCW, Brand JA, Boyer AE. An enzyme-linked immunoelectrotransfer blot assay and glycoprotein antigens for diagnosing human cysticercosis (Taenia solium). Oxford Journals 1989;159:50–9.

[13] Gonzalez AE, Cama V, Gilman RH, Tsang VCW, Pilcher JB, Chavera A, et al. Prevalence and comparison of serologic assays, necropsy, and tongue examination for the diagnosis of porcine cysticercosis in Peru. Am J Trop Med Hyg 1990;43:194–9.

[14] Hancock K, Pattabhi S, Whitfield FW, Yushak ML, Lane WS, Garcia HH, et al. Characterization and cloning of T24, a Taenia solium antigen diagnostic for cysticercosis. Mol Biochem Parasitol 2006;147:109–17.

[15] Handali S, Klarman M, Gaspard AN, Noh J, Lee YM, Rodriguez S, et al. Multiantigen print immunoassay for comparison of diagnostic antigens for Taenia solium cysticercosis and taeniasis. Clin Vaccine Immunol 2010;17:68–72.

[16] Rodriguez S, Wilkins P, Dorny P. Immunological and molecular diagnosis of cysticercosis. Pathog Glob Health 2012;106:286–98.

[17] Jayashi CM, Gonzalez AE, Castillo Neyra R, Rodríguez S, García HH, Lightowlers MW. Validity of the enzyme-linked immunoelectrotransfer blot (EITB) for naturally acquired porcine cysticercosis. Vet Parasitol 2014;199:42–9.

[18] Kruschke JK. Review of doing Bayesian data analysis: A tutorial with R, JAGS, and Stan (second edition). Clin Neuropsychol 2017;31:1268–70.

176

[19] Plummer M. JAGS Version 3.4.0 user manual 2013:0–41.

[20] Yu B, Mykland P. Looking at Markov samplers through cusum path plots: a simple diagnostic idea. Stat Comput 1998;8:275-286–286.

[21] Robert CP, Casella G. Monte Carlo Statistical Methods. New York, USA: Springer New York; 2004.

[22] Raftery AE, Lewis SM. One long run with diagnostics: implementation strategies for Markov Chain Monte Carlo. Stat Sci 1992;7:493–7.

[23] Raftery EA, Lewis MS. How Many Iterations in the Gibbs Sampler? In: Benardo J, Berger J, Dawid A, Smith A, editors. Bayesian Stat. 4, New York: The Clarendon Press and Oxford University Press; 1992, p. 763–774.

[24] Gelman A. Inference and Monitoring Convergence. In: Gilks W, Richardson S, Spiegelhalter D, editors. Markov Chain Monte Carlo Pract., London: Chapman & Hall; 1996, p. 131–143.

[25] Su Y-S, Yajima M. Using R to Run “JAGS” 2015.

[26] Plummer M, Best N, Cowles K, Vines K, Sarkar D, Bates D, et al. Output Analysis and Diagnostics for MCMC 2016.

[27] Team R Core. R: A Language and Environment for Statistical Computing 2017.

[28] Anh Tuan P, Dung TTK, Nhi VA. Sero-epidemiological investigation of cysticercosis in the southern provinces. J Malar Parasite Dis Control 2001;4:81–7.

[29] Trung DD, Praet N, Cam TDT, Lam BVT, Manh HN, Gabriël S, et al. Assessing the burden of human cysticercosis in Vietnam. Trop Med Int Heal 2013;18:352–6.

177

[30] Willingham AL, Van De N, Doanh NQ, Cong D, Dung T V, Dorny P, et al. Current status of cysticercosis in Vietnam. Southeast Asian J Trop Med Public Health 2003;34:35–50.

[31] Khue P Van, Luc P. Veterinary Parasitology. Hanoi: Hanoi Agricultural Publishing House; 1996.

[32] Doanh NQ, Holland W, Vercruysse J, De N Van. Results of survey on cysticercosis pig in some nothern provinces in Vietnam. J Malar Parasite Dis Control 2002;6:76–82.

[33] De N Van, Chau L Van, Son DT, Chuyen LT, Hop NT, Vien HV, et al. Taenia solium survey in Hanoi. J Malar Parasite Dis Control 2004;6:93–9.

[34] Doanh NQ, Kim NT, De N V., Lung NL. Result of survey on taeniasis and cysticercosis humans and pigs in Bac Ninh and Bac Kan provinces. Vet Sci Tech 2002;9:46–9.

[35] Conlan J V., Vongxay K, Khamlome B, Dorny P, Sripa B, Elliot A, et al. A cross-sectional study of Taenia solium in a multiple taeniid-endemic region reveals competition may be protective. Am J Trop Med Hyg 2012;87:281–91.

[36] Lan NTK, Quyen NT, Hoat PC. The correlation between the prevalence of the tapeworm Taenia hydatigena in dogs and their larvae Cysticercus tenuicollis in cattle and pigs - The effect of tapeworm treatment in dogs. Vet Sci Tech 2011;18:60–5.

[37] Ngowi HA, Kassuku AA, Maeda GEM, Boa ME, Carabin H, Willingham AL. Risk factors for the prevalence of porcine cysticercosis in Mbulu District, Tanzania. Vet Parasitol 2004;120:275–83.

[38] Sikasunge CS, Phiri IK, Phiri AM, Dorny P, Siziya S, Willingham AL. Risk factors associated with porcine cysticercosis in selected districts of Eastern and Southern provinces of Zambia. Vet Parasitol 2007;143:59–66.

178

[39] Pouedet MSR, Zoli AP, Nguekam, Vondou L, Assana E, Speybroeck N, et al. Epidemiological survey of swine cysticercosis in two rural communities of West- Cameroon. Vet Parasitol 2002;106:45–54.

[40] Lescano AG, García HH, Gilman RH, Guezala MC, Tsang VCW, Gavidia CM, et al. Swine cysticercosis hotspots surrounding Taenia solium tapeworm carriers. Am J Trop Med Hyg 2007;76:376–83.

[41] Pondja A, Neves L, Mlangwa J, Afonso S, Fafetine J, Willingham AL 3rd, et al. Prevalence and risk factors of porcine cysticercosis in Angonia District, Mozambique. PLoS Negl Trop Dis 2010;4:e594.

[42] Sarti Gutierrez E, Schantz PM, Aguilera J, Lopez A. Epidemiologic observations on porcine cysticercosis in a rural community of Michoacan State, Mexico. Vet Parasitol 1992;41:195–201.

[43] Sikasunge CS. The Prevalence and Transmission Risk Factors of Porcine Cysticercosis in Eastern and Southern Provinces of Zambia. University of Zambia, 2005.

[44] García HH, Gilman RH, Gonzalez AE, Verastegui M, Rodriguez S, Gavidia C, et al. Hyperendemic human and porcine Taenia solium infection in Perú. Am J Trop Med Hyg 2003;68:268–75.

[45] Shonyela SM, Mkupasi EM, Sikalizyo SC, Kabemba EM, Ngowi HA, Phiri I. An epidemiological survey of porcine cysticercosis in Nyasa District, Ruvuma Region, Tanzania. Parasite Epidemiol Control 2017;2:35–41.

[46] Gabriël S, Dorny P, Mwape KE, Trevisan C, Braae UC, Magnussen P, et al. Control of Taenia solium taeniasis/cysticercosis: The best way forward for sub- Saharan Africa? Acta Trop 2017;165:252–60.

179

Table 5.1 Details of information about general and pig information

Requesting information Detail General information Address of house; number of people per household; presence and type and location of toilet; number of livestock per household; vegetables and crops management (source of water for irrigation, kind of manure for vegetables and crops); coordinate of household, pigsty, toilet and vegetable patch; elevation of household; number of dogs; location of defecation for dogs; efforts to disinfect dog’s faeces; observation of Taenia proglottids in dog’s faeces Pig information: Demography Sex; age; breed; number of pigs Husbandry management The presence and location of pigsty; whether pig confined or free- roaming; roaming range of pigs; whether pigs eat human faeces; whether pigs have access to human defecation area Diet management Drinking water sources for pigs, kind of food for pigs (commercial or handmade bran, scavenging), feeding raw vegetables to pigs or not, vegetables washed before feeding pigs

Table 5.2 Structure of the data from 1281 study pigs from six villages in M’Drak, Buon Don and Krong Nang districts

Number at the next highest level Level Number Mean Range Districts 3 - - Villages 6 2 2 Households 408 68 21 to 135 Pigs a 1281 3 1 to 24 a A total of 1281 pigs recruited in this study. The mean of pigs per household were three (range 1 to 24).

180

Table 5.3 General description of household data

Characteristic Frequency Percentage (95% CI) Number of households 408 - District: M'Drak 203 50 (45 to 55) Buon Don 135 33 (28 to 38) Krong Nang 70 17 (14 to 21) Presence of pit latrine: Yes 266 65 (60 to 70) No 142 35 (30 to 40) Presence of pigsty: Yes 311 76 (72 to 80) No 97 24 (20 to 28) Are pigs permitted to roam freely? Yes 172 42 (37 to 47) No 236 58 (53 to 63) Source of water for pigs: Pipe/well/rain water 379 93 (90 to 95) Lake/stream/pond 29 7.0 (4.9 to 10) Owning dogs: yes 259 63 (59 to 68) No 149 37 (32 to 41) Household observing proglottids in dog faeces: Yes 113 44 (37 to 50) No 146 56 (50 to 62) Faeces of dogs treated: Yes 3 1.1 (0.3 to 4.6) No 256 99 (96 to 99) Site of dog defecation: Inside compound 192 74 (68 to 79) Outside compound 67 26 (21 to 32)

181

Table 5.4 General description of pig data

Characteristic Frequency Percentage (95% CI) Number of pigs 1281 - Sex: Male 536 42 (39 to 45) Female 745 58 (55 to 61) Age (months) < 4 497 39 (36 to 42) 4 - 12 665 52 (49 to 55) > 12 119 9.3 (7.8 to 11) Breed: Imported breed 522 41 (38 to 44) Local breed (Soc) 759 59 (56 to 62) Allowed to roam freely: Yes 406 32 (29 to 34) No 875 68 (66 to 71) Human coprophagy: Yes 340 27 (24 to 29) No 941 73 (71 to 76) Main food: Commercial/handmade bran 1145 89 (86 to 91) Scavenging 136 11 (9.0 to 12) Provision of raw, unwashed vegetables a: Yes 723 57 (54 to 59) No 555 43 (41 to 46) a Three cases was excluded from analysis.

182

Table 5.5 Risk factors associated with T. solium cysticercosis positive in pigs

Number of pigs Regression MC Adjusted OR Explanatory variable Positive Total coefficient (SD) error (95% CrI) -5.9800 Intercept 12 1281 0.006 - (1.5480) Coprophagy of human faeces: No 3 941 Reference Yes 9 340 2.6600 (0.8470) 0.004 13.0 ( 3.38 to 105) Kind of food: Bran 6 1145 Reference Scavenging 6 136 2.1400 (1.0320) 0.003 7.24 (1.73 to 114) a Random effects b: Estimate SD Village 2.18 5.77 a Interpretation: The odds of T. solium cysticercosis positive for pigs that scavenged food was 7.24 (95% credible interval 1.73 to 114) times that of pigs that did not scavenge for food. b Variance and standard deviation of the variance of the village-level random effect

183

CHAPTER 6

Spatial distribution of Taenia solium exposure in humans and pigs in the Central Highlands of Vietnam

Presented as published:

Ng-Nguyen D, Traub RJ, Nguyen V-AT, Breen K, Stevenson MA. Spatial distribution of Taenia solium exposure in humans and pigs in the Central Highlands of Vietnam. PLoS Negl Trop Dis 2018;12:e0006810.

184

6.1 Abstract

6.1.1 Background

Taenia solium, a pork-borne parasitic zoonosis, is the cause of taeniasis and cysticercosis in humans. In Vietnam, poor sanitation, the practice of outdoor defecation and consumption of raw/undercooked pork have been associated with infection/exposure to T. solium in both humans and pigs. The broad-scale geographic distribution of the prevalence of T. solium varies throughout the country with infection restricted to isolated foci in the north and a more sporadic geographic distribution in the Central Highlands and the south. While cross-sectional studies have allowed the broad- scale geographic distribution of T. solium to be described, details of the geographic distribution of T. solium at finer spatial scales have not been described in detail. This study provides a descriptive spatial analysis of T. solium exposure in humans and pigs and T. solium taeniasis in humans within individual households in village communities of Dak Lak in the Central Highlands of Vietnam.

6.1.2 Methodology/Principal findings

We used Ripley’s K-function to describe spatial dependence in T. solium exposure positive and negative human and pig households and T. solium taeniasis exposure positive and negative households in villages within the districts of Buon Don, Krong Nang and M’Drak of Dak Lak province in the Central Highlands of Vietnam. The prevalence of exposure to T. solium in pigs in Dak Lak province was at 0.9 (95% CI 0.5 to 1.7) cases per 100 pigs at risk. The prevalence of exposure to the parasite in humans was somewhat higher at 5 (95% CI 3 to 8) cases per 100 individuals at risk. Spatial aggregations of T. solium exposure-positive pig and human households occurred in some, but not all of the villages in the three study districts. Human exposure-positive households were found to be aggregated within a distance of 200 to 300 m in villages in Krong Nang district compared with distances of up to 1500 m in villages in M’Drak district. Although this study demonstrated the aggregation of households in which either T. solium exposure- or taeniasis-positive individuals were present, we were unable to identify an association between the two due to the very low number of T. solium taeniasis-positive households. 185

6.1.3 Conclusions

Spatial aggregations of T. solium exposure-positive pig and human households occurred in some, but not all of the villages in the three study districts. We were unable to definitively identify reasons for these findings but speculate that they were due to a combination of demographic, anthropological and micro-environmental factors. To more definitively identify characteristics that increase cysticercosis risk we propose that cross-sectional studies similar in design to that described in this paper should be applied in other provinces of Vietnam.

6.1.4 Author summary

Taenia solium is a pork-bone zoonotic parasite. Humans acquire taeniasis from consumption of raw/undercooked pork contaminated with T. solium cysticerci. Pigs and humans acquire cysticercosis following consumption of food contaminated with eggs shed from the faeces of humans with T. solium taeniasis. In Vietnam, the geographic distribution of T. solium varies throughout the country with hotspots or foci of infection in communities in the North and a more sporadic distribution in the Central Highlands and the South. While information on the distribution at the regional and provincial level is available, there is no available information on the spatial distribution of T. solium at fine spatial scales and factors influencing its distribution at fine spatial scales have not been described in detail. In this cross-sectional study, we collected information on the geographic coordinates of study households and utilized spatial analytical techniques to quantify both the fine scale spatial pattern of exposure to T. solium as well as the tendency for T. solium exposure-positive households to be located close to other T. solium exposure-positive households (spatial autocorrelation) in three districts in Dak Lak province. We found that in some of the study villages T. solium exposure-positive households were more likely to be surrounded by other T. solium exposure-positive households. Human exposure-positive households were found to be aggregated within a distance of 200 to 300 m in villages in Krong Nang district; whilst spatial aggregation of pig exposure-positive households was found up to distances of 1500 m in villages in M’Drak district. Although households that had either T. solium exposure- or taeniasis- positive cases were aggregated, we were unable to quantify their spatial association due 186

to the extremely low number of T. solium taeniasis-positive households. This study shows that in the Central Highlands of Vietnam, T. solium exposure tend to cluster within foci. This information can be used to inform community intervention programs to lower its incidence in both humans and pigs.

6.2 Introduction

Taenia solium is a pork-borne zoonosis of major public health and economic importance. The parasite causes cysticercosis/neurocysticercosis in humans and pigs in many low-income communities in Latin America, Africa and Asia [1]. Poor sanitation, allowing pigs to free roam, the practice of outdoor defecation and consumption of raw/undercooked pork are risk factors for T. solium infection. Cysticercosis infection in humans and pigs occurs due to the accidental ingestion of T. solium eggs shed through the faeces of humans with T. solium taeniasis (tapeworm carriers). Taeniasis occurs when humans consume raw/undercooked pork with T. solium cysticerci.

T. solium infection results in not only an economic burden in low-income communities due to loss of productivity in affected individuals and the cost of treatment, but also losses arising from the condemnation of pig carcasses destined for human consumption. It was estimated that approximately US $185 million and 2.1 million disability-adjusted life years (DALYs) were lost in north India due to human cysticercosis in 2011 [2]. It was estimated that the number of DALYs for human cysticercosis in Mozambique in 2007 [3] was 6.0 per thousand person-years, 0.2 in 2011 in Mexico [4] and 0.7 in 2012 in Tanzania [5]. The pork industry in four provinces of Laos estimated losses of between US $55,000 to 96,000 arising from 20% of carcasses identified as cysticerci- positive over a 21 month period [6]. In Latin America, the number of neurocysticercosis infections has been estimated to be between 11 and 29 million with 1.3 million individuals suffering from neurocysticercosis-related epilepsy [7]. Globally, cysticercosis was estimated to be the cause of over 28,000 deaths in 2010 [8].

Although risk factors for T. solium taeniasis and cysticercosis include outdoor defecation, the consumption of raw/undercooked pork and allowing pigs to free roam, transmission patterns and the prevalence of the parasite can vary considerably and may prove inconsistent within and/or between regions and communities [9] thus impeding 187

efforts to achieve parasite eradication. Because of the negative impact of T. solium on human health and the economy [10], controlling the disease is a priority. Understanding the spatial distribution of T. solium is one important step towards development of effective control strategies. Studies in communities of Latin America and Africa, where T. solium infection is hyperendemic, identified clustering of T. solium cysticercosis at both the household and community level [11–13] and a strong geographical association between T. solium carriers and cysticercosis in humans and pigs [14, 15]. Spatial analyses were used to define an appropriate radius for a control area in a community where T. solium was hyperendemic in Peru. Targeting interventions within this control area was effective in reducing the number of T. solium carriers and the sero-incidence of porcine cysticercosis [16].

T. solium is endemic in Vietnam. A systematic review of cross-sectional studies carried out in Vietnam between 1999 to 2011 showed that the prevalence ranged from 0 to 130 T. solium positive individuals per 1000 individuals at risk [17]. The distribution of T. solium in Vietnam is characterized by hotspots or foci of infection in communities in the northern provinces, including Phu Tho and Bac Ninh [18, 19]. In the Central Highlands and the south of the country the distribution of T. solium is sporadic [20]. While previous studies [15, 16, 17] have provided useful information at the regional and provincial level, we are aware of no investigations that have investigated the distribution of T. solium at finer spatial scales. With this background, the aim of this study was to describe the spatial distribution of households in which one or more individual humans or pigs were T. solium exposure positive and households in which one or more individual (humans) were T. solium taeniasis positive. Quantitative estimates of the prevalence and geographic distribution of T. solium exposure and T. solium taeniasis positive households will provide evidence to allow public health authorities to decide between treatment programs applied at the whole community level as opposed to treatment programs applied at either the individual household or small area level.

188

6.3 Materials and methods

6.3.1 Ethics statement

This study was reviewed and approved by the Behavioral and Social Sciences Human Ethics Sub-committee, the University of Melbourne (reference number 1443512) and the Animal Ethics and Scientific Committee, Tay Nguyen University (reference number 50.KCNTY). This study was conducted under the supervision of the local Center for Public Health and the local Center for Animal Health, Dak Lak, Vietnam. This research on pigs was based on the International Guiding Principle for Biomedical Research Involving Animals issued by the Council for the International Organization of Medical Sciences.

6.3.2 Field procedures

The cross-sectional study was carried out between May and October 2015 in Dak Lak province in the Central Highlands of Vietnam. Dak Lak is comprised of 15 districts with approximately 70% of the total population of 1.8 million people living in rural areas [21]. Within the province, three districts namely Buon Don, Krong Nang and M’Drak were chosen as the study sites based on their diverse geographic characteristics (Fig. 6.1). The characteristics of these districts have been described in detail elsewhere [22]. A sampling frame listing the name of all villages in Dak Lak province was obtained from Sub-Department of Animal Health office within the Ministry of Agriculture and Rural Development. Villages eligible for sampling comprised those with more than 1000 pigs, as recorded by the Sub-Department of Animal Health. All eligible villages within each district were assigned a number and two numbers chosen at random to select villages from each district for inclusion in the study. A list of householder names within each selected village was obtained from each village head person, and householder names were assigned a numeric code. A sheet of paper listing numeric household codes for each village were cut into pieces and placed face-down on a table. The village head person was asked to select 50 households at random for human sampling and between 100 and 140 households for pig sampling. All households selected for human sampling and pig sampling were visited several days before the proposed sampling date to obtain consent from householders to take part in the study. 189

Householders eligible for inclusion in the study were individuals who were healthy, not pregnant and over seven years of age. Householders requested to take part in the study signed a consent form. Those that were under 18 years of age were required to provide written consent as well as written consent from either their parents or legal guardians. At the time of consent each study participant was given a labeled stool container, with instructions that the container would be collected on the date of sampling, several days later. Sampling of households (Fig. 6.2) was carried out in two stages. Humans from each of the 50 consenting study households were sampled between May and October 2015; pigs from each of the 100 to 150 consenting study households were sampled between June and October 2015. At the time of each household visit, a questionnaire was administered to each of the study participants soliciting details regarding demography, sanitation and hygiene status, food culture and religion, practice of pig management and the longitude and latitude coordinates of the main doorway of entry of the dwelling used for sleeping.

On the date of sampling, consenting householders were visited by staff from the Sub- Department of Health and 5 mL of venous blood collected into plain clotting tubes from consenting study participants. Stool samples were fixed in 5% potassium dichromate (w/v) for molecular analysis.

Pigs that were pregnant, ill or aged less than two months of age were excluded from sampling. Approximately, 10 mL of blood was obtained from the cranial vena cava of each pig into plain blood collection tubes. Blood samples were allowed to clot at ambient temperature prior to centrifugation at 3200 × g for 5 minutes to collect serum. Serum was dispensed into 1.5 mL aliquots and stored at -20 °C until analysis.

The longitude and latitude of the main doorway of entry of sampled households was recorded using a handheld global positioning system (GPS) device (Garmin GPSMAP64, Taiwan).

190

6.3.3 Taenia solium carrier identification

Human stool samples were tested to determine T. solium tapeworm carriers using a real- time PCR (T3qPCR) described by Ng-Nguyen et al. (2017) [22]. The assay has been reported to have a diagnostic sensitivity of 94% and a diagnostic specificity of 98%.

6.3.4 Human T. solium exposure identification

Human serum samples were tested for the presence of antibody against T. solium cysticerci using a lentil-lectin purified glycoprotein-enzyme-linked immunoelectrotransfer blot (LLGP-EITB). This assay has a diagnostic sensitivity of 98% and a diagnostic specificity of 100%. The LLGP-EITB was performed using the methodology described by Tsang et al. (1989) [23].

6.3.5 Pig T. solium exposure identification

Pig serum samples were tested for the presence antibody against T. solium cysticerci using rT24H antigen in the enzyme-linked immunoelectrotransfer blot (EITB) format. The rT24H-EITB assay showed no cross-reaction to T. hydatigena and had a diagnostic sensitivity and specificity of 100% when tested on 29 cysticercosis- negative USA pig sera, 12 necropsy-positive T. solium-positive Peruvian pig sera and four T. hydatigena necropsy-positive Vietnamese pig sera [24]. The performance of the rT24H-EITB assay was carried out using the methodology described by Noh et al. 2014 [25]. Positive samples resulting from the rT24H-EITB assay were confirmed using the LLGP-EITB assay.

6.3.6 Spatial data analyses

Details collected during each of the household visits and laboratory test results were stored in a relational database (Microsoft Access 2007, Microsoft Corporation, Redmond, USA). Longitude and latitude coordinates of household locations (recorded in degrees, minutes and seconds) were converted to decimal degrees and re-projected to the Universal Transverse Mercator Zone 48N projection using the World Geodetic System 1984 datum.

191

Analyses were carried out to describe the spatial distribution characteristics of: (1) T. solium exposure-positive and T. solium exposure-negative households for humans; (2) T. solium exposure-positive and T. solium exposure-negative households for pigs; and (3) human or pig T. solium or T. solium taeniasis-positive and negative households.

Our classification of T. solium exposure-positive households for humans and pigs was based on the LLGP-EITB and rT24H-EITB assays, respectively. Classification of Taenia-positive households was based on parallel interpretation of the test results of the LLGP-EITB, rT24H-EITB and T3qPCR assays (positive for either T. solium exposure or T. solium taeniasis).

Ripley’s K-function [26] provides a summary measure of spatial dependence among point locations as a function of their Euclidean distance. The K-function is defined as the expected number of points that are located within a distance ℎ of an arbitrarily selected point location, divided by the overall density of points [27]. Where there is spatial dependence in a point pattern, point events are likely to be surrounded by other point events and, for small values of distance ℎ, 퐾(ℎ) will be relatively large. Conversely, if point events are regularly spaced, each point is likely to be surrounded by empty space and, for small values of distance ℎ, 퐾(ℎ) will be small. To facilitate inference, we developed separate K-function plots for T. solium exposure-positive and T. solium exposure-negative households. For each value of ℎ we then calculated the K-function difference as 퐷(ℎ) = 퐾(ℎ) − 퐾(ℎ). If exposure-positive households were spatially aggregated, over and above that of the exposure-negative households, then 퐷(ℎ) will appear graphically as peaks (or troughs) as a function of distance.

Three sets of Ripley's K-function analyses were carried out for: (1) human T. solium exposure-positive households and human T. solium exposure-negative households; (2) pig T. solium exposure-positive households and pig T. solium exposure-negative households; and (3) T. solium taeniasis exposure positive and T. solium taeniasis negative households.

Monte Carlo simulation was used to construct critical envelopes for each K-function difference plot. Here, we randomly assigned the observed number of positive 192

households across the population of study household locations and re-computed 퐷(ℎ) each time. The critical envelopes are based on 1000 Monte Carlo simulations of the data. Departures of the observed value of 퐷(ℎ) from the limits of the upper and lower critical envelopes provided an indication of spatial aggregation of exposure-positive households beyond that which would be expected by chance, and at what spatial scale.

6.4 Results

6.4.1 General data description

The total numbers of households visited for collecting samples from humans and pigs, respectively, were 190 and 408 in Buon Don, Krong Nang and M’Drak districts. Within the 190 households, a total of 342 individuals consented to participate in the study. The number of pigs sampled from the 408 households was 1281. Four of the 190 households (2.1%, 95% CI 0.6 to 5.6) housed T. solium tapeworm carriers and the percentages of households housing individuals and pigs that were T. solium exposure positive was 8.9% (17/190, 95% CI 5.5 to 14) and 2.7% (11/408, 95% CI 1.4 to 4.9), respectively. Amongst the 11 T. solium exposure-positive households for pigs, there was one household that had more than one pig antibody-positive to T. solium cysticerci; all other exposure-positive households had a single pig that was seropositive. All T. solium exposure-positive households for humans had a single individual that was antibody- positive. Of the 561 households that were visited for either human or pig sampling, 31 had either humans or pigs that were either human T. solium exposure-positive, pig T. solium exposure-positive or T. solium taeniasis positive (Table 6.1). There were 29 households with single infections; one household with exposure-positive pigs and one household had an individual infected with T. solium taeniasis and an individual that was T. solium exposure-positive. Human T. solium exposure- and taeniasis-positive households were present in all three districts. There was no pig T. solium exposure- positive households in Buon Don.

In three study districts, the prevalence of having a household latrine was relatively low ranging from 13% to 47%. This meant that outdoor defecation was a practice reported by between 15% and 74% of the study population. Our data showed that allowing pigs to

193

free roam was common practice in Buon Bon, Krong Nang, and M’Drak. The percentage of pigs that consumed human faeces was high in Buon Don and M’Drak (Table 6.2). Of 11 exposure-positive households for pigs, there were nine households in M’Drak. Pigs kept in seven of the nine households were allowed to roam freely within the village.

6.4.2 Distribution of pig T. solium exposure households

No exposed pigs were detected in Buon Don (Fig. 6.3c-d) and two exposed pigs were detected in Krong Nang (Fig. 6.4c-d). In M’Drak, the four exposed pigs that were detected were within a distance of 250 m of each other (Fig. 6.5c) and pairs of exposure-positive pig households were less than 100 m apart (Fig. 6.5d). The K- function difference plot for T. solium exposure in pigs in M’Drak shows 퐾(ℎ) in excess of 퐾(ℎ) up to a distance of 1500 m (Fig. 6.6f).

6.4.3 Distribution of human T. solium exposure households

There were small numbers of human T. solium exposure-positive households in close proximity in Krong Nang (Fig. 6.4b). The K-function difference plot for exposure to T. solium in humans in Krong Nang supported this observation, where 퐾(ℎ) was in excess of 퐾(ℎ) up to a distance of 200 to 300 m (Fig. 6.6c). The spatial distribution of human exposure-positive households in Buon Don (Fig. 6.3a-b) and M’Drak (Fig. 6.5a-b) were more regularly distributed; there was no evidence of significant differences between 퐾(ℎ) and 퐾(ℎ) up to a distance of 3000 m (Fig. 6.6a and 6.6e).

6.4.4 Distribution of Taenia-positive households

When we considered households that were either human T. solium exposure, taeniasis or pig T. solium exposure as a single group, the K-function difference plot showed all T. solium exposure-positive and taeniasis-positive households were aggregated up to a distance of 1000 m in M’Drak (Fig. 6.6g). Similar associations were evident in Buon Bon and Krong Nang but the observed K-function difference plot did not exceed the simulation envelope limits at any distance (Fig. 6.6b and 6.6d). On inspection, however,

194

we observed a group of taeniasis- and exposure-positive households in close proximity to each other in the village of Cu Mta in M’Drak district (Fig. 6.7).

6.5 Discussion

This study describes the fine scale spatial distribution of T. solium exposure in pigs and humans in Vietnam for the first time. The geographic distribution of T. solium exposure- and taeniasis-positive households varied markedly across the districts of Buon Don, Krong Nang and M’Drak of Dak Lak province. A prominent feature of this data is that the prevalence of T. solium exposure in both species and T. solium taeniasis was relatively low (in humans 9 exposure- and 2 taeniasis-positive households per 100 households at risk; in pigs 3 exposure-positive households per 100 households at risk) making it difficult to definitively identify characteristics of the spatial distribution of positive households that are likely to exist across all districts of Dak Lak, and indeed all districts of Vietnam.

Spatial aggregations of T. solium exposure-positive households for humans occurred in the village of Dlieya in Krong Nang district (Fig. 6.4b) but not in the villages of Ea Wer in Buon Don (Fig. 6.3b), and Krong Jing and Cu Mta in M’Drak (Fig. 6.5a-b)). Our K- function difference plots showed that T. solium exposure-positive households for humans showed the same pattern of spatial dependence as T. solium exposure-negative households in Buon Bon and M’Drak (Fig. 6.6a and 6.6e). In Krong Nang, compared with human exposure-negative households, human exposure-positive households were aggregated up to a distance of 200 to 300 m (Fig. 6.6c). We speculate that if the prevalence of exposure was higher in Buon Don and M’Drak and sufficient resources were available to allow larger sample sizes in each of the two districts to be collected, a similar pattern of spatial dependence would be evident.

Although spatial aggregation of T. solium exposure-positive households for humans in Krong Nang was beyond that expected by chance (and entirely due to a collection of five positive households in the village of Dlieya), its overall magnitude was relatively small (Fig. 6.6c). Dlieya is small village comprised of less than 200 households in a remote area of Krong Nang. In Dlieya the number of individuals per household was considerably larger than that of the other villages in the study (median 5; minimum 2 to 195

maximum 11) and a notable feature was that it was common for several generations of a family to live together in close proximity, and a highly prevalent custom was that food was shared with neighbours and relatives on a daily basis, often associated with community ceremonies (e.g. weddings and anniversary of deaths). In the district of Krong Nang, houses are typically surrounded by a large garden comprised of coffee or pepper trees. We hypothesize that in this district, where a high proportion of the study population were known to defaecate outdoors, individuals were more likely to defecate in their own garden (as opposed to communal areas) which means that it was more likely for T. solium eggs to be present in close spatial proximity to a given household where exposure-positive individuals were present. We speculate that the anthropological and fine-scale environmental characteristics of Dlieya were sufficient to allow spatial clustering of human exposure infection to be detected even in the presence of a modest sampling effort (푛 = 30 households).

Spatial aggregations of exposure to T. solium in pigs occurred but this was infrequent. In three study districts, there was a single aggregation of pig exposure-positive households in M’Drak (Fig. 6.5c). Our K-function difference plot for M’Drak (Fig. 6.6f) showed pig exposure-positive households were clustered within a distance of 1500 m. Presumably, this was due to the larger range over which free-roaming pigs forage. Copado et al., 2004 [26] reported that free-roaming pigs travel daily within a distance ranging from 1000 to 3000 m. In a 12 hour period, pigs traveled a distance of up to 4000 m and spent, on average, 47% of their time outside of their homestead [28]. Our findings are supported by those of Ngowi et al. (2010) [12] who conducted a cross- sectional study of cysticercosis in 784 pig-owning households in northern Tanzania. In the study of Ngowi et al. (2010) it was shown that porcine cysticercosis was clustered within the distance of 600 m and 10 km. Morales et al. (2008) [29] conducted a cross- sectional study of 562 pigs in the state of Morelos in Mexico in 2003. In this study the prevalence of porcine cysticercosis was relatively high (13%; 95% CI 11 to 17) and while free-roaming pigs had a greater risk of being cysticercosis-positive, no geographical clustering of positivity was found.

196

Spatial clustering of T. solium exposure in pigs in M’Drak could have been associated with the age of the resident pig population, the absence of pigsties and the regular habit of coprophagy amongst pigs. In total, there were 11 T. solium pig exposure-positive households in Dak Lak. Nine of these 11 positive households were in M’Drak, aggregated in groups of two to four households (Fig. 6.5c-d). Of the nine T. solium pig exposure-positive households in M’Drak, in seven households pigs were allowed to roam freely within the village, increasing the chance of exposure to T. solium eggs. The terrain in M’Drak is generally flat. The quality of the soil is poor supporting predominantly natural grasslands. For these reasons, the practice of allowing pigs to free roam is more common compared with the two other districts. M’Drak had a high proportion of pigs that were not confined (17% [211 of 1281], 95% CI 15 to 19). Of the total number of pigs sampled in this study, 511 (40%) were from M’Drak. Of the 511 M’Drak pigs that were sampled, it was reported that 193 (38%) consumed human faeces and 211 (41%) regularly scavenged for food (Table 6.2).

When we considered households that were human and/or pig T. solium exposure-positive or taeniasis positive as a single group, our K-function difference plot showed these positive households were aggregated up to a distance of 1000 m in M’Drak (Fig. 6.6g), but not in Buon Bon (Fig. 6.6b) and Krong Nang (Fig. 6.6d). O’Nea at al. (2012) and Pray et al. (2017) showed that human and/or porcine cysticercosis cases were strongly associated with the presence of tapeworm carriers [14, 15]. Individuals and pigs living in close proximity to tapeworm carriers are more likely to be infected with T. solium cysticercosis [11, 15, 30]. Given this unique data set, with contemporary sampling of humans and pigs, it was of interest to us to determine if there was a spatial dependence between T. solium exposure- and T. solium taeniasis-positive households. Although households that had either T. solium exposure- or taeniasis-positive cases were spatially aggregated, we were unable to identify an association between the two because of the extremely low number (푛 = 4) of taeniasis-positive households across the three study districts. On inspection we observed a group of taeniasis- and exposure-positive households in close proximity to each other in the village of Cu Mta in M’Drak district (Fig. 6.7). Madinga et al. (2017) [26] and Morales et al. (2008) [29] indicated that there was no spatial correlation of T. solium exposure in pigs and T. solium taeniasis.

197

Spatial aggregations of human and pig exposure-positive households occurred in some, but not all, of the villages in the three study districts. We can only speculate about the reasons for this pattern, as discussed above. Since cysticercosis occurs throughout Vietnam [17] it is likely that foci of infection are present in other areas. The relatively low prevalence of exposure to T. solium indicates that massive deworming programs in the communities of Dak Lak province are, for the most part, unnecessary. Instead, we recommend that if a human is identified as T. solium positive then either: (a) individuals resident in the immediate area should be tested to rule out the presence of an exposure or infection cluster; or (b) anthelmintic treatment is offered to individuals resident within a 2000 m radius of the identified case. With respect to the second approach, privacy issues would need to be handled appropriately, particularly in small communities.

A limitation of this study was that for logistic reasons sampling of humans and pigs were carried out independently resulting in a lack of overlap of the locations of households where humans and pigs were sampled (see, for example, Ea Nuol in the district of Buon Don, Fig. 6.3a and 6.3c). While this limited our ability to identify an association (if any) between human and pig T. solium exposure-positive households, assessment of the spatial dependence of exposure status by species (human, pigs) was possible. Although households that had either T. solium exposure- or taeniasis-positive cases were spatially aggregated, we were unable to quantify their spatial association due to the extremely low number of T. solium taeniasis-positive households.

198

6.6 References

[1] Donadeu M, Lightowlers MW, Fahrion AS, Kessels J, Donadeu M, Lightowlers MW, et al. Taenia solium: WHO endemicity map update Taenia solium: carte d’endémicité actualisée de l’OMS pour 2016. vol. 4. 2016.

[2] Singh BB, Khatkar MS, Gill JPS, Dhand NK. Estimation of the health and economic burden of neurocysticercosis in India. Acta Trop 2017;165:161–9.

[3] Trevisan C, Devleesschauwer B, Praet N, Pondja A, Assane YA, Dorny P, et al. Assessment of the societal cost of Taenia solium in Angonia district, Mozambique. BMC Infect Dis 2018;18:127.

[4] Bhattarai R, Budke CM, Carabin H, Proaño J V., Flores-Rivera J, Corona T, et al. Estimating the non-monetary burden of neurocysticercosis in mexico. PLoS Negl Trop Dis 2012;6:e1521.

[5] Trevisan C, Devleesschauwer B, Schmidt V, Winkler AS, Harrison W, Johansen MV. The societal cost of Taenia solium cysticercosis in Tanzania. Acta Trop 2017;165:141–54.

[6] Choudhury AA, Conlan J V, Racloz VN, Reid SA, Blacksell SD, Fenwick SG, et al. The economic impact of pig-associated parasitic zoonosis in Northern Lao PDR. Ecohealth 2013;10:54–62.

[7] Coyle CM, Mahanty S, Zunt JR, Wallin MT, Cantey PT, Jr CW, et al. Neurocysticercosis: Neglected but not forgotten. PLoS Negl Trop Dis 2012;6:e1500.

[8] Torgerson PR, Devleesschauwer B, Praet N, Speybroeck N, Willingham AL, Kasuga F, et al. World Health Organization estimates of the global and regional disease burden of 11 foodborne parasitic diseases, 2010: A data synthesis. PLoS Med 2015;12:e1001920.

[9] Komba EVG, Kimbi EC, Ngowi HA, Kimera SI, Mlangwa JE, Lekule FP, et al.

199

Prevalence of porcine cysticercosis and associated risk factors in smallholder pig production systems in Mbeya region, southern highlands of Tanzania. Vet Parasitol 2013;198:284–91.

[10] FAO/WHO. Multicriterial-based Ranking for Risk Management of Food-borne Parasites. Microbiological Risk Assessment Series No 23. vol. 23. Rome: WHO Press; 2014.

[11] Garcia HH, Gilman RH, Gonzalez AE, Verastegui M, Rodriguez S, Gavidia C, et al. Hyperendemic human and porcine Taenia solium infection in Peru. Am J Trop Med Hyg 2003;68:268–75.

[12] Ngowi H, Kassuku A, Carabin H, Mlangwa J, Mlozi M, Mbilinyi B, et al. Spatial clustering of porcine cysticercosis in Mbulu district, northern Tanzania. PLoS Negl Trop Dis 2010;4:e652.

[13] Widdowson M-A, Cook AJC, Williams JJ, Argaes F, Rodriguez I, Dominguez JL, et al. Investigation of risk factors for porcine Taenia solium cysticercosis: a multiple regression analysis of a cross-sectional study in the Yucatan Peninsula, Mexico. Trans R Soc Trop Med Hyg 2000;94:620–4.

[14] O’Neal SE, Moyano LM, Ayvar V, Gonzalvez G, Diaz A, Rodriguez S, et al. Geographic correlation between tapeworm carriers and heavily infected cysticercotic pigs. PLoS Negl Trop Dis 2012;6:e1953.

[15] Pray IW, Ayvar V, Gamboa R, Muro C, Moyano LM, Benavides V, et al. Spatial relationship between Taenia solium tapeworm carriers and necropsy cyst burden in pigs. PLoS Negl Trop Dis 2017;11:e0005536.

[16] O’Neal SE, Moyano LM, Ayvar V, Rodriguez S, Gavidia C, Wilkins PP, et al. Ring-screening to control endemic transmission of Taenia solium. PLoS Negl Trop Dis 2014;8:e3125.

200

[17] Ng-Nguyen D, Stevenson MA, Traub RJ. A systematic review of taeniasis, cysticercosis and trichinellosis in Vietnam. Parasit Vectors 2017;10:150.

[18] Doanh NQ, Holland W, Tam PT, De N Van, Hoa LT, Hoan DH. The results of the studies on situation of humans and pigs contracting disease that cestode worm and its larvae cause for. Sci Technol J Agric Rural Dev 2006;4:56–8.

[19] Vien HV, Hung NM, Thach DTC, Thuan LK, Dung DT, Hop NT, et al. Investigate Taenia infection in two communes of Tan Son district, Phu Tho province. J Malar Parasite Dis Control 2012;1:69–74.

[20] Van Chuong N. Status of helminthic infection in humans in 14 provinces spanning the Central and Central Highland regions of Vietnam. IMPE-QN 2005. http://www.impe-qn.org.vn/impe- qn/vn/portal/InfoDetail.jsp?area=58&cat=1068&ID=597 (accessed March 1, 2016).

[21] General Statistics Office Of Vietnam. Area, population and population density by province. Gen Stat Off Vietnam 2017. http://www.gso.gov.vn/default_en.aspx?tabid=774 (accessed August 10, 2017).

[22] Ng-Nguyen D, Stevenson MA, Dorny P, Gabriël S, Vo T Van, Nguyen VT, et al. Comparison of a new multiplex real-time PCR with the Kato Katz thick smear and copro-antigen ELISA for the detection and differentiation of Taenia spp. in human stools. PLoS Negl Trop Dis 2017;11:e0005743.

[23] Tsang VCW, Brand JA, Boyer AE. An enzyme-linked immunoelectrotransfer blot assay and glycoprotein antigens for diagnosing human cysticercosis (Taenia solium). J Infect Dis 1989;159:50–9.

[24] Ng-nguyen D, Noh J, Breen K, Stevenson MA, Handali S, Traub RJ. The epidemiology of porcine Taenia solium cysticercosis in communities of the Central Highlands in Vietnam. Parasit Vectors 2018;11:360.

201

[25] Noh J, Rodriguez S, Lee YM, Handali S, Gonzalez AE, Gilman RH, et al. Recombinant protein- and synthetic peptide-based immunoblot test for diagnosis of neurocysticercosis. J Clin Microbiol 2014;52:1429–34.

[26] Madinga J, Kanobana K, Lukanu P, Abatih E, Baloji S, Linsuke S, et al. Geospatial and age-related patterns of Taenia solium taeniasis in the rural health zone of Kimpese, Democratic Republic of Congo. Acta Trop 2017;165:100–9.

[27] Pfeiffer D, Robinson T, Stevenson M, Stevens K, Rogers D, Clements A. Spatial Analysis in Epidemiology. London, UK.: Oxford University Press; 2008.

[28] Thomas LF, de Glanville WA, Cook EA, Fèvre EM. The spatial ecology of free- ranging domestic pigs (Sus scrofa) in western Kenya. BMC Vet Res 2013;9:46.

[29] Morales J, Martínez JJ, Rosetti M, Fleury A, Maza V, Hernandez M, et al. Spatial distribution of Taenia solium porcine cysticercosis within a rural area of Mexico. PLoS Negl Trop Dis 2008;2:e284.

[30] Lescano AG, Garcia HH, Gilman RH, Gavidia CM, Tsang VCW, Rodriguez S, et al. Taenia solium cysticercosis hotspots surrounding tapeworm carriers: Clustering on human seroprevalence but not on seizures. PLoS Negl Trop Dis 2009;3:e371.

202

Table 6.1 Spatial characteristics of T. solium exposure and taeniasis. Data structure of household and individual sampling

Infection status Human Pig Total

Total number of households visited 190 408 561 a T. solium exposure-positive households (pigs) - 11 11 T. solium exposure-positive households (humans) 17 - 17 Taeniasis-positive households 4 - 4 Taenia-positive households - - 31 b

Total number of tested study subjects 342 1281 1623 Total T. solium exposure-positive subjects 17 12 29 Total taeniasis-positive subjects 4 - 4 a Total of visited households for sampling humans, pigs, and both humans and pigs. b A total of 31 households had cases of either pig T. solium exposure-positive, human T. solium exposure- positive or T. solium taeniasis.

Table 6.2 Spatial characteristics of T. solium exposure and taeniasis. Pig management in the study districts of Dak Lak province

Location Total of pigs Free-roaming pig Human coprophagy Event % (95% CI) Event % (95% CI) Buon Don 323 164 51 (45 to 56) 133 41 (36 to 47) Krong Nang 447 31 6.9 (4.8 to 9.8) 14 3.1 (1.8 to 5.3) M’Drak 511 211 41 (37 to 46) 193 38 (36 to 42)

203

Fig. 6.1 Map of Dak Lak province showing the location of study sites. The black shaded areas show the study locations in Buon Don, Krong Nang and M’Drak districts. This figure was adapted from Ng- Nguyen et al. (2017) [22].

204

Fig. 6.2 A picture showing a typical household in rural areas in Dak Lak province. The space under the floor of the stilt house is used for the shelter of domestic animals.

205

Fig. 6.3 Location of households for human (a and b) and pig (c and d) T. solium exposure in Ea Noul (a and c) and Ea Wer (b and d) villages in Buon Don district. The solid red circles indicate the location of exposure-positive households; the open grey circles indicate the location of exposure-negative households.

206

Fig. 6.4 Location of households for human (a and b) and pig (c and d) T. solium exposure in Ea Ho (a and c) and Dlieya (b and d) villages in Krong Nang district. The solid red circles indicate the location of exposure-positive households; the open grey circles indicate the location of exposure-negative households.

207

Fig. 6.5 Location of households for human (a and b) and pig (c and d) T. solium exposure in Krong Jing (a and c) and Cu Mta (b and d) villages in M’Drak district. The solid red circles indicate the location of exposure-positive households; the open grey circles indicate the location of exposure-negative households.

208

Fig. 6.6 Estimated K-function differences for human and pig T. solium exposure- and Taenia- positive households in three studied districts. Horizontal Horizontal dotted lines at D(h) = 0 provides a reference for D(h) under complete spatial randomness. The solid line indicates the observed value of D(h). The light blue areas indicate the 95% tolerance limits for D(h) estimated from 1000 Monte Carlo simulations. Panel (a), (c) and (e) showing the estimation for human T. solium exposure respectively in Buon Don, Krong Nang and M’Drak; panel (b), (d) and (g) for Taenia-positive respectively in Buon Don, Krong Nang and M’Drak and panel (f) for T. solium exposure in pigs in M’Drak.

209

Fig. 6.7 Location of T. solium taeniasis- and exposure-positive households. A group of cysticercosis- positive households (red circles) occurred in close proximity of a taeniasis-positive household (green circles) in the village of Cu Mta of M’Drak district. The taeniasis-positive household (green circles) had a case of cysticercosis.

210

CHAPTER 7

The prevalence of Trichinella in pigs in the Central Highlands of Vietnam

211

7.1 Introduction

The nematode Trichinella is a globally distributed meat-borne parasitic zoonosis [1]. A number of domestic and wild animals including mammals, birds, reptiles and rodents are known definitive hosts of the parasite [1]. Infection in domestic and wild animals such as domestic and feral pigs, dogs, cats and raptorial birds occurs due to scavenging or being fed scraps of infected meat or via natural hunting behaviour. Humans may be infected when ingesting raw or undercooked infected meat [2]. No clinical signs have been observed in infected animals, whereas, humans with trichinellosis show nonspecific symptoms such as high fever, myalgia and periorbital oedema [3].

Trichinella has been reported to be present in Southeast Asia. The food-borne parasite has caused several outbreaks in the mountainous regions of Cambodia, southwest China, Lao PDR, Thailand, and Vietnam [4–7]. In Vietnam, a limited number of published studies on Trichinella in the northwest of the country have reported at least four human outbreaks involving 88 individuals in the last 15 years (reviewed by Nguyen et al., 2017 [8]). In affected provinces, including Son La and Dien Bien, 20% of domestic free-roaming pigs [9], 3.2% of wild boars (n = 62), 2.8% (n = 820) of rats and 4% (n = 125) of dogs [10] were found positive to Trichinella antibodies. Surprisingly, none of 261 confined wild boars resident on seven farms or 98 cats in these two provinces were seropositive [11]. In these areas, the outbreaks of trichinellosis were associated with the consumption of raw/undercooked pork and/or game meat [12]. In affected provinces of Dien Bien, Son La, Yen Bai, and Thanh Hoa, it is typical for inhabitants to consume raw/undercooked pork/wild pork. Non-confinement of pigs is common [13]. In other regions of Vietnam, including the Central Highlands, information on the prevalence of Trichinella is not available [8].

Dak Lak province in the Central Highlands of Vietnam has the same environmental conditions and pig management practices for the propagation of Trichinella in domestic and sylvatic hosts and dietary risk factors for zoonotic meat-borne disease. The province has approximately 810,000 adult pigs [14]. In rural, resource-poor communities approximately 16,000 pigs (approximately 2% of the adult pig population) are non- confined and permitted to scavenge for food [15]. Currently, there has been an increase

212

in the demand for free-roaming pork. The interprovincial market demand for free- roaming pork has increased by a factor of between 170% and 210% in recent years [15]. Free-roaming pigs are often sold directly from farmers to consumers. These pigs are slaughtered by consumers in their back yards and, as a result, are not inspected. In addition, the dietary practice of consumption of raw or undercooked meat and pork is common among rural Dak Lak inhabitants, which increases their risk of Trichinella infection. The aim of this study was to quantify the prevalence of Trichinella in pigs in Dak Lak province using the artificial digestion (AD) method.

7.2 Materials and methods

7.2.1 Study sites and sampling

This was a cross-sectional study carried out between May and October 2015 in Krong Nang, M’Drak and Buon Don districts in Dak Lak province. These districts were chosen as the study sites based on their diverse geographic characteristics. M’Drak is located in the east of Dak Lak province with an average altitude of 400 m to 500 m. The region has a tropical monsoonal climate typical of Vietnam’s Central Coast. Krong Nang is situated to the north at an altitude of 800 m. Buon Don, situated to the west of the province, has an average elevation of 330 m and has a hot and dry climate.

There are two pig production systems in these districts: semi-commercial farming and smallholder production systems. Semi-commercial farms are comprised of between 20 to 60 fattening pigs that are fed a mixture of 80% crop by-products and 20% commercial concentrates. Pigs from these farms are sold via pork traders with meat inspection carried out at district slaughterhouses in urban areas. In contrast to semi- commercial farms, smallholder production systems in the three studied districts comprise of one or two sows and less than 10 fattening pigs. In the study districts, local breeds are predominant in these smallholder farms. An additional characteristic of smallholder farms is that the pigs are permitted to freely roam around forests and/or villages to scavenge for food. Pigs on smallholder farms are not provided with a daily food ration and are instead fed crop by-products or human leftover food on an irregular basis. Pork is sold to consumers via village slaughterers, who function as pork traders within a village. In addition, pigs are often sold directly from farmers to 213

consumers. There is almost no activity of meat inspection at the village level since pork traders fear that their businesses will be shut down if infected carcasses are found by health officials.

For this study muscle samples were collected from pigs slaughtered at i) each of the district abattoirs (n = 3) and ii) from households where pigs from smallholder farms were slaughtered (n = 80). Samples were collected by trained local veterinarians (n = 10) or medical staff (n = 10). For each slaughtered pig approximately 50 to 100 grams of tissue from the diaphragm crus, the tongue, and masseter muscle [16] were collected. Data on age, sex, original location (domestic or wild pig), production system (commercial, smallholder backyard) and management practices (confinement or non-confinement) were obtained from pig traders (in the case of abattoir- slaughtered pigs) and original smallholder farmers or consumers (in the case of smallholder-slaughtered pigs).

7.2.2 Artificial digestion

For groups of five pigs, diaphragm, tongue, and masseter muscle samples were pooled based on the method described by Gamble et al. (2000) [17]. In brief, fat tissues were carefully removed from the muscle tissue with the amount of muscle dissection kept to a minimum. Twenty grams of pooled muscle was homogenised using a blender. In some situations where there was insufficient muscle tissue to yield a 20 g pooled sample, muscle tissue from healthy pork was used to make up the shortfall. Each 20 g pool was then digested in 2 litres of 1% pepsin-hydrochloric acid (Pepsin 1:10 000 NF Sigma Aldrich; Hydrochloric acid 37% Merck) and mixed for 40 minutes. The digest fluid was passed through a 180 µm sieve to retain any undigested material and then onto a 20 µm mesh that would then retain any Trichinella muscle larvae. The 20 µm mesh was rinsed forcefully using warm tap water until traces of the digested fluid were gone. The remaining sediment was rinsed into a gridded square petri dish with up to 20 mL water. The rinse was examined for Trichinella larvae using a dissecting microscope at 20 × magnification.

214

7.2.3 Statistical analysis

Because of the imperfect diagnostic test characteristics of AD (compared with PCR [18]) the estimated TP of Trichinella, accounting for pooling was calculated using the general approach described by Sergeant et al., 2008 [19] and Rogan & Gladen (1978) [20], modified for the low prevalence situation using Bayesian methods as described by Messam et al. (2008) [21].

In brief, a Bayesian approach accounts for uncertainty in the values of the diagnostic test characteristics of the pooled testing protocol. Where 푆푒 and 푆푝 represent the sensitivity and specificity of the individual tests (respectively), and 푆푒 and 푆푝 represent the sensitivity and specificity of the pooled testing protocols (respectively) and 푛 is the number of pools, the distribution of the number of test-positive pools is

푥 | 푇푃, 푆푒, 푆푝 ~ 푏푖푛표푚푖푎푙(푛, 퐴푃) where:

퐴푃 = 푆푒 × (1 − [1 − 푇푃] ) + (1 − 푆푝 × [1 − 푇푃] ) Equation 1 and

() 푆푒 = 1 − (1 − 푆푒) × 푆푝 Equation 2 and

푆푝 = 푆푝 . Equation 3 To estimate the 푇푃 of Trichinella at the individual animal level beta prior distributions for the 푆푒 and 푆푝 of AD were used. We assumed that the 푇푃 of Trichinella was greater than zero and that all prevalences were equally likely, which translates to a 푇푃 ~ 푏푒푡푎(1,1) distribution.

Markov chain Monte Carlo (MCMC) methods were used to derive posterior estimates of 푇푃 using WinBUGS [22, 23]. In WinBUGS, the MCMC sampler was run for 20,000 iterations and the first 1000 ‘burn in’ samples were discarded. The posterior distribution of 푇푃 was obtained by running sufficient iterations to ensure that the Monte Carlo standard error of the posterior means were at least one order of magnitude smaller than their posterior standard deviation [24]. Parallel chains were run using diverse initial

215

values to ensure that convergence was achieved to the same distribution. The point estimate and 95% CrI for 푇푃 is reported as the median and 0.025 and 0.975 quantiles of the posterior distribution of 푇푃. Because of the considerable uncertainty around the 푆푒 of AD we conducted a series of analyses where, 푇푃 was calculated assuming 푆푒 took values that ranged from 10 to 100% at 10% increments or were obtained from previous publications [25]. For all calculations the 푆푝 of AD was assumed to be 100%. Statistical analyses were carried out using the contributed package R2WinBUGS [26] in R [27].

7.3 Results

Of the 835 pigs sampled, 202 (24%), 334 (40%) and 298 (36%) were sourced from Krong Nang, Buon Don and M’Drak districts, respectively. Confined pigs comprised 87% (n = 730) and free-roaming pigs 13% (n = 105) of the study population. Of the confined pigs, 12% (n = 84) were raised in local households, 4% (n = 26) were wild- captured pigs and 85% (n = 620) were from commercial enterprises. All free-roaming pigs were raised in backyards of local small holder farmer households. AD did not identify the presence of larvae in any of the pigs sampled in this study. The estimated TP of Trichinella in pigs in these communities ranged from 0.38% (95% CrI 0.02 to 1.29) to 2.44% (95% CrI 0.12 to 9.23) (Table 7.1 and Fig. 7.1).

7.4 Discussion

Even though pig management conditions in Dak Lak provide conditions conducive to the propagation of Trichinella, we failed to recover any Trichinella larvae using AD. In the human trichinellosis outbreak described by Vu Thi et al. in 2013 and 2014 [10, 11], only free-roaming pigs and dogs were found seropositive to Trichinella as opposed to the confined seronegative pigs. The low proportion of free-roaming pigs tested in this study (13%, n = 105) may explain the absence of Trichinella positive pigs herein. Due to pork traders fearing that their businesses will be shut down if infected carcasses are found by health officials, slaughtering of free-roaming pigs in rural communities tends to be a discrete activity. This impacted our ability to obtain sufficient free-roaming pig sample numbers to reflect the proportion of free-roaming pigs consumed in this area of Vietnam.

216

Assuming the 푆푒 of AD ranged from 10% to 92% and the 푆푝 is 100% our estimate of the true individual animal prevalence of Trichinella in Dak Lak ranged from 2.4% (95% CrI 0.12 to 9.2) to 0.52% (95% CrI 0.02 to 1.9), respectively (Table 7.1 and Fig. 7.1). If the 푆푒 of AD was assumed 100%, the estimated TP of Trichinella was 0.38% (95%, CrI 0.02 to 1.3). In Argentina and Mexico where Trichinella are endemic, the prevalence of Trichinella in slaughtered pigs determined using AD is commonly less than 1% [28, 29]. This indicates that even in endemic regions the prevalence of Trichinella is quite low. Our estimates of the 푇푃 of Trichinella in pigs in Dak Lak province are therefore consistent with those from endemic countries such as Argentina and Mexico. Even though all samples in this study were AD test negative, we stress that absence of evidence is not evidence of absence [26] and our findings should be interpreted accordingly with routine existing advice regarding standards of pig housing and limiting access of pigs to raw meat maintained.

Although AD is considered the ‘gold standard’ for the detection of Trichinella, the limit of detection using this method is ≥ 3 larvae per gram (LPG) when digesting 1 g of pool muscle compared to PCR-based methods which can detect as low as 0.1 LPG [18, 30]. In a survey conducted by Vu Thi et al., (2010) [9] in Son La province, of 206 seropositive carcasses tested by PCR, Trichinella larvae were recovered from 11 samples and the burden of infection in Trichinella positive individuals was estimated to vary from 0.04 to 0.4 LPG. It is therefore possible that in this study larval burdens were lower than the limit of AD detection which resulted in missed positive cases [31].

7.5 Conclusions

Although Dak Lak province has environmental conditions and pig management practices conducive for the propagation of Trichinella in domestic and sylvatic hosts, we failed to detect Trichinella larvae in confined and free-roaming pig populations using AD. Further investigations targeting more representative sections of the pig population using cost competitive and sensitive serological and/or molecular diagnostic tools should provide a better estimate of the prevalence of Trichinella infection in Dak Lak in the future.

217

7.6 References

[1] Pozio E, Hoberg E, La Rosa G, Zarlenga DS. Molecular taxonomy, phylogeny and biogeography of nematodes belonging to the Trichinella genus. Infect Genet Evol 2009;9:606–16.

[2] FAO/WHO/OIE. Guidelines for the Surveillance, Management, Prevention and Control of Trichinellosis. Paris: FAO, WHO, OIE; 2007.

[3] Robertson LJ, Sprong H, Ortega YR, Van der Giessen JWB, Fayer R. Impacts of globalisation on foodborne parasites. Trends Parasitol 2014:37.

[4] Sayasone S, Odermatt P, Vongphrachanh P, Keoluangkot V, Dupouy-Camet J, Newton PN, et al. A trichinellosis outbreak in Borikhamxay Province, Lao PDR. Trans R Soc Trop Med Hyg 2006;100:1126–9.

[5] Barennes H, Sayasone S, Odermatt P, De Bruyne A, Hongsakhone S, Newton PN, et al. A major trichinellosis outbreak suggesting a high endemicity of Trichinella infection in northern Laos. Am J Trop Med Hyg 2008;78:40–4.

[6] Cui J, Wang ZQ, Xu BL. The epidemiology of human trichinellosis in China during 2004-2009. Acta Trop 2011;118:1–5.

[7] Van De N, Nga VT, Dorny P, Trung NV, Minh PN, Dung DT, et al. Trichinellosis in Vietnam 2015;92:1265–70.

[8] Ng-Nguyen D, Stevenson MA, Traub RJ. A systematic review of taeniasis, cysticercosis and trichinellosis in Vietnam. Parasit Vectors 2017;10:150.

[9] Vu Thi N, Dorny P, La Rosa G, To Long T, Nguyen Van C, Pozio E. High prevalence of anti-Trichinella IgG in domestic pigs of the Son La province, Vietnam. Vet Parasitol 2010;168:136–40.

218

[10] Thi N V, De N V, Praet N, Claes L, Gabriël S, Dorny P. Seroprevalence of trichinellosis in domestic animals in northwestern Vietnam. Vet Parasitol 2013;193:200–5.

[11] Vu Thi N, Nguyen VD, Praet N, Claes L, Gabriël S, Huyen NT, et al. Trichinella infection in wild boars and synanthropic rats in northwest Vietnam. Vet Parasitol 2014;200:207–11.

[12] Nguyen Van C, Vu Thi N, Nguyen Trong C. Result of survey and solution for trichinellosis in pigs in Bac Ha district, Son La province. Vet Sci Tech 2012;19:50–6.

[13] Huong NT. Taeniasis and Cysticercosis in a Selected Group of Inhabitants from a Mountainous Province in North Vietnam. Prince Leopold Institute of Tropical Medicine, Antwerpen (Antwerp), Belgium, 2006.

[14] General Statistics Office of Vietnam. Statistical Data. Gen Stat Off Vietnam 2017. http://www.gso.gov.vn/ (accessed December 22, 2015).

[15] Nguyen TH. Evaluation of market opportunities for producing local pigs in Dak Lak. Tay Nguyen J Sci 2009;5:21–6.

[16] Gottstein B, Pozio E, Nockler K. Epidemiology, diagnosis, treatment, and control of trichinellosis. Clin Microbiol Rev 2009;22:127–45.

[17] Gamble HR, Bessonov AS, Cuperlovic K, Gajadhar AA, van Knapen F, Noeckler K, et al. International Commission on Trichinellosis: Recommendations on methods for the control of Trichinella in domestic and wild animals intended for human consumption. Vet Parasitol 2000;93:393–408.

[18] Cuttell L, Corley SW, Gray CP, Vanderlinde PB, Jackson LA, Traub RJ. Real- time PCR as a surveillance tool for the detection of Trichinella infection in muscle samples from wildlife. Vet Parasitol 2012;188:285–93.

219

[19] Sergeant ESG, Nielsen SS, Toft N. Evaluation of test-strategies for estimating probability of low prevalence of paratuberculosis in Danish dairy herds. Prev Vet Med 2008;85:92–106.

[20] Rogan W, Gladen B. Estimating prevalence from results of a screening test. American Journal of Epidemiology. Am J Epidemiol 1978;107:71–6.

[21] Messam L, Branscum A, Collins M, Gardner I. Frequentist and Bayesian approaches to prevalence estimation using examples from Johne’s disease. Anim Heal Res Rev 2008;9:1–23.

[22] Branscum AJ, Gardner IA, Johnson WO. Bayesian modeling of animal- and herd-level prevalences. Prev Vet Med 2004;66:101–12.

[23] Lunn DJ, Thomas A, Best N, Spiegelhalter D. WinBUGS - A Bayesian modelling framework: Concepts, structure, and extensibility. Stat Comput 2000;10:325–37.

[24] Wakefield JC, Best NG, Waller L. Bayesian approaches to disease mapping. In: Elliott P, Wakefield J, Best N, Briggs D, editors. Spat. Epidemiol. Methods Appl., Oxford University Press, London; 2000, p. 118.

[25] Cuttell L. Wildlife Surveillance and Risk Assessment for Non-Encapsulated Trichinella Species in Mainland Australia. University of Queensland, 2013.

[26] Sturtz S, Ligges U, Gelman A. R2WinBUGS: A Package for Running WinBUGS from R. J Stat Softw 2005;12:1–16.

[27] Team R Core. R: A Language and Environment for Statistical Computing 2017.

[28] Ribicich M, Gamble HR, Rosa A, Bolpe J, Franco A. Trichinellosis in Argentina: an historical review. Vet Parasitol 2005;132:137–42.

[29] Ortega-Pierres MG, Arriaga C, Yepez-Mulia L. Epidemiology of trichinellosis in Mexico, Central and South America. Vet Parasitol 2000;93:201–25.

220

[30] Gamble HR. Sensitivity of artificial digestion and enzyme immunoassay methods of inspection for trichinae in pigs. J Food Prot 1998;61:339–43.

[31] Forbes LB, Rajic A, Gajadhar AA. Proficiency samples for quality assurance in Trichinella digestion tests. J Food Prot 1998;61:1396–9.

221

Table 7.1 Estimated true prevalence of Trichinella with assuming test Sp of 100%

Test sensitivity (%) Larvae burden (LPG) True prevalence (%) 95% CrI 100 NA 0.38 0.02 to 1.29 a 92 0.25 0.52 0.02 to 1.86 a 88 0.50 0.54 0.02 to 1.95 a 72 0.15 0.68 0.03 to 2.42 60 NA 0.84 0.04 to 2.74 50 NA 0.84 0.05 to 3.03 40 NA 1.06 0.06 to 3.96 30 NA 1.27 0.07 to 4.58 20 NA 1.71 0.06 to 6.65 10 NA 2.44 0.12 to 9.23

NA: Not applicable; a Test sensitivity was obtained from Cuttell (2013) [25]

Fig. 7.1 Estimated true prevalence of Trichinella with assuming test Sp of 100%. The dark grey area indicates 95% credible interval of estimated true prevalence

222

CHAPTER 8

General discussion

223

8.1 Aims of the study revisited

Vietnam is classified as a low-to-middle income country by the World Bank [1, 2]. The country is comprised of eight regions: the Southeast, the Red River Delta, the Mekong River Delta, the Northeast, Northwest, North Central Coast, South Central Coast and Central Highlands. The Central Highlands is located in the southern central part of the country and comprises five provinces, Kon Tum, Gia Lai, Dak Lak, Lam Dong and Dak Nong. The socio-economic standing of this region is low compared with the rest of the country and resources for addressing issues related to low rates of education and high rates of poverty is limited. Low numbers of physicians and veterinarians per capita and relatively limited public health facilities contribute to low standards of accessibility to medical care. Health and environmental education campaigns are infrequent which contributes to the poor standards of sanitation and hygiene observed, which include routine practices of outdoor defaecation and unsafe eating habits. Vietnam is among the top five countries that raise pigs in the world [3]. However, the majority of pig production systems consist of smallholder (household) farms, common in rural regions [4]. The pig husbandry system in rural regions of the Central Highlands is characterised by the use of agricultural by-products for feeding pigs, that are allowed to free-roam [4, 5].

Taenia and Trichinella are neglected zoonotic parasites that are endemic in most underdeveloped countries and regions where extensive pig raising is practiced and where the consumption of undercooked meat is commonplace [6]. Current estimates of the prevalence of Taenia and Trichinella at the community level in Vietnam is lacking. The little research that has been conducted, has focussed on the north of the country and, given the substantial changes arising from economic development that has occurred in recent years, it is likely that the results of these few studies are now outdated and not indicative of current conditions.

In the Central Highlands, despite the presence of a number of factors conducive to either infection or exposure to Taenia and Trichinella, there is little or no information on the prevalence of and risk factors for taeniasis, T. solium cysticercosis, and Trichinella in humans and pigs. A lack of robust, quantitative data deprives government veterinary and public health officers of information that can be used to devise appropriate measures to

224

manage these parasites in pig and human populations. For these reasons, the primary aims of this thesis were to determine the risk factors associated with the seroprevalence of human and pig T. solium exposure, and true prevalence of Taenia tapeworm species infection in humans and Trichinella in pigs. In addition, this PhD study aimed to identify the spatial distribution of T. solium taeniasis and cysticerci exposed pigs and humans and ascertain the spatial relationship between them.

8.2 Limitation of this thesis

The research that comprises this dissertation was self-funded, limiting the ability to carry out investigations comprised of large numbers of participants over wide geographic areas. Hence, the conclusion on the TP of taeniasis in humans and Trichinella exposure in pigs, as well as the seroprevalence of T. solium cysticerci exposure in humans and pigs may not be as strong as if a larger sample size were screened. Subsequently, risk factors associated with taeniasis and T. solium seroexposure in humans and pigs would also be strengthened if a larger sample size were utilised.

Given the budget limitation, we were not able to utilise ‘gold standard’ methods to confirm the present T. solium cysticerci infections in humans (chapter 4) and pigs (chapter 5) using ultrasound or CT / MRI scans. The reported results in this thesis were thus limited to T. solium seroexposure which cannot definitively confirm the presence of a current or true infection in humans and pigs but rather, their ability to mount an antibody response to the antigens of T. solium. This research was thus unable to provide the most accurate and clinically relevant picture of T. solium cysticerci infection in the region. The specificity and accuracy of the intervention methods proposed to the communities (Chapter 8) may thus be reduced as a consequence. In addition, the more sensitive methods (using ELISA, WB or PCR) to screen Trichinella in pig populations (Chapter 7) were not utilised due to the budgetary constraints; the TP of Trichinella in pigs may be underestimated.

This research was conducted in the rural communities of Dak Lak province where facilities are generally lacking along with limited numbers of trained personnel available to provide assistance with field research. 225

Budgetary limitations also meant that we could not screen for porcine cysticercosis using additional LLGP antigens in the EITB assay format (Chapter 5), which has been shown to be the most reliable serological ante-mortem test for porcine T. solium seroexposure [7]. Utilisation of rT24H antigens in the EITB assay however provided a reasonable compromise. Ideally, post-mortem examinations of pigs that were found to be seropositive for T. solium exposure using the rT24H-EITB assay would have provided confirmation of serological results, however this was not possible given constraints of having to pay the householder for replacement pigs. Similar to the aforementioned situation in humans, the outcome we measured was therefore T. solium exposure, as opposed to T. solium infection. The proposal intervention methods for eliminating or reducing the infection in humans and pigs may not fix the current situation in the communities as the true and/or present infections were not confirmed.

Moreover, given the relatively limited sample size and low numbers of individuals positive for taeniasis, we were unable to estimate risk factors for species-level taeniasis in Dak Lak province (Chapter 4). Similarly, the relatively low numbers of pigs positive for T. solium exposure hindered our ability to build a strong model for the association of risk factors for T. solium exposure in pigs, as well as detered efforts to identify the spatial relationship between T. solium taeniasis and cysticercosis in humans and pigs (Chapter 6).

In the Central Highlands of Vietnam slaughter of free-roaming pigs is typically carried out discretely since pork traders fear that their businesses will be shut down if cysticerci-infected carcasses are found by health officials. Furthermore, free-roaming pigs are sold directly from farmers to consumers and pigs that are purchased tend to be slaughtered by consumers, typically in their back yard. Both of these factors negatively impacted on our ability to collect muscle samples from slaughtered community pigs for the Trichinella study. Since the composition of the pigs for Trichinella study did not reflect the general population of pigs, it is possible that the prevalence reported in this study is not an accurate estimate of the TP of Trichinella infection in Dak Lak. A relatively small sample size and the fact that all of the muscle samples that were tested were found to be negative for Trichinella meant that we were unable to identify the

226

species/genotypes and risk factors of Trichinella infection in pigs in the communities of Dak Lak province.

8.3 Application of new and multiple diagnostic tests

8.3.1 Diagnosis of taeniasis

The question of how many Taenia tapeworm species exist and the prevalence of taeniasis infection in humans in Dak Lak province were questions that were addressed in this dissertation. In North Vietnam, three species of Taenia tapeworms were confirmed for the first time in 2001 [8]. Prior to this study, experts presumed that T. asiatica was not present in the South of Vietnam and T. saginata was dominant over T. solium. This assumption was based on the finding that 100% of 15 Taenia spp. proglottids collected in the South Vietnam (in the provinces of Lam Dong, Kon Tum, Dak Lak, Ha Tinh, Phu Yen, Ninh Thuan, Can Tho and Ho Chi Minh) were identified as T. saginata [9].

The poor sensitivity of light microscopy to identify eggs of Taenia in faeces in previously conducted helminth surveys in Vietnam is likely to have resulted in underestimation of the TP of taeniasis infection, and it is possible that there may have been errors in identifying Taenia species on the basis of light microscopic examination of proglottids alone (Chapter 2). Our study in Dak Lak province reinforced this critique of the literature: of 342 individuals, the AP of taeniasis estimated using the KK technique was 2% (95% CI 0.9 to 4.4). The AP increased to 6.7% (95% CI 4.4 to 10) when the same samples were tested using a newly developed multiplex real-time PCR (T3qPCR). Furthermore, when the imperfect Se and Sp of the multiplex real-time PCR were taken into account the TP of taeniasis was estimated to be 5.1% (95% CI 2.7 to 8.6).

The advantage of the T3qPCR compared with the cAg-ELISA and KK is its ability to discriminate Taenia species and with substantially greater diagnostic Se and Sp compared to the KK technique (Chapter 3). The utilisation of T3qPCR allowed the prevalence of this tapeworm in the Central Highlands of Vietnam to be determined and negated the assumption that T. asiatica is restricted to the north of the country. In 227

addition, this assay assisted in proving the sympatric existence of all three human tapeworms in the region for the first time and confirmed the dominance of T. saginata over T. solium in Dak Lak province with TP estimates of 5.8% (95% CI 3.7 to 9.0) and 1.2% (95% CI 0.4 to 3.2), respectively. The T3qPCR provides a suitable tool for large-scale community surveys due to its ability to screen samples in a high throughput fashion.

8.3.2 Diagnosis of Taenia solium cysticercosis in humans

Options for choosing a reliable serodiagnostic test for T. solium cysticercosis in humans and pigs are limited. The LLGP-EITB assay is considered the reference serological test using native LLGPs antigen derived from T. solium cysticerci. The assay requires expensive equipment and technical expertise, thus it is difficult to apply in developing countries (usually Taenia endemic countries) where actually need for eliminating the parasite. Besides this, an ELISA using B158C11A10 and B60H8A4 monoclonal antibodies has been used widely in the field and has recently been commercialised (the apDia Cysticercosis Antigen ELISA, apDia-ELISA), despite little information on diagnostic test characteristics being available [10]. In this study, utilization of antibody test (LLGP-EITB assay) means that the current T. solium cysticercosis infection is not known as the assay can pick up the case of passively obtained antibodies or transition antibodies. In Dak Lak as well as Vietnam where T. solium is endemic, a reliable serological assay to confirm the true status of cysticercosis infection is necessary to obtain the true burden of cysticercosis in the region to propose a suitable strategy for intervention programs.

8.3.3 Diagnosis of Taenia solium cysticercosis in pigs

In Dak Lak, the dog tapeworm, T. hydatigena, is endemic [11, 12] along with the sympatric presence of T. solium and T. asiatica which use pigs as their intermediate hosts. For these reasons it is a challenge to investigate the prevalence of porcine T. solium cysticercosis in the province using ante-mortem serological methods. Post- mortem examination is a reliable means for diagnosing T. solium cysticercosis in pigs, but this was impractical because most of the pigs reared in rural communities of Dak

228

Lak are slaughtered on an individual and irregular basis within backyards. Palpation of the tongue to detect cysts may have been possible, however this is a relatively insensitive procedure. Moreover, this technique is not only time consuming and but also presents a health and safety risks for those conducting the examination [7, 13, 14].

In a large-scale survey, serology is the preferred diagnostic method. However, antibody and antigen-ELISAs reportedly cross-react with T. hydatigena [15–20] and are therefore not suitable for use as a diagnostic test for Taenia solium cysticercosis in pigs in Dak Lak. In this study, serum samples from pigs were subjected to the rT24H-EITB assay to detect antibodies against T. solium cysticerci. This antigen worked well in different platforms of EITB, MAPIA, MICT and Quick ELISA [21–30] and initially shows highly specific to T. solium.

8.4 Epidemiological characteristics of Taenia in Vietnam

The systematic review of taeniasis and T. solium cysticercosis (Chapter 2) helped to provide an overview of the main epidemiological features of pork-borne zoonotic diseases in Vietnam. Taeniasis and cysticercosis in humans were the outcomes of interest with TP estimates ranging from zero to 13% across the various study locations. The TP of human cysticercosis in some regions was relatively high, arising from sporadic foci of T. solium. In contrast to human cysticercosis, the TP of porcine cysticercosis was relatively low in most study locations (less than 1%). The use of tests with a relatively low diagnostic Se (e.g. post-mortem carcass examination) and slaughterhouse-based surveys were presumed to be the reason for the low prevalence of porcine cysticercosis in Vietnam (Chapter 2). Consumption of undercooked and/or raw meat were identified as the main risk factors for taeniasis whilst consumption of raw vegetables and the use of night-soil for fertilisation of local produce were risk factors for T. solium exposure.

In Chapter 2, we hypothesised that the low AP of T. solium cysticercosis in pigs was due to use of a diagnostic test with relatively low Se (carcass examination) and/or the internal validity of the study’s findings was poor because of selection bias: the population of pigs slaughtered at abattoirs were not typical of the general population of pigs in each of the study regions. In Chapter 5, where the seroprevalence of T. solium 229

exposure in pigs was estimated using a community-based survey (where the population of pigs that took part in the study were typical of the general population of pigs in Dak Lak), we showed that even with a good diagnostic test (the rT24H-EITB assay) the prevalence of T. solium exposure was low, supporting the hypothesis of protective cross-reactivity between T. hydatigena and T. solium.

The sympatric existence of all three human tapeworm species along with the presence of T. hydatigena increases the complexity of diagnosing T. solium cysticercosis in pigs in Dak Lak province. The overall AP of taeniasis in the province was 6.7% (95% CI 4.4 to 10) in which T. saginata taeniasis was the most dominant infection (5.8%, 95% CI 3.7 to 9.0) followed by T. asiatica and T. solium taeniasis with 1.5% (95% CI 0.5 to 3.6) and 1.2% (95% CI 0.4 to 3.2), respectively (Chapter 3). Factors associated with Taenia spp. taeniasis infection were frequent consumption of undercooked beef and undercooked pork, and frequent consumption of pork tongue. The identified risk factors supported the presence of the three human tapeworms in the region as consumption of undercooked beef may cause T. saginata taeniasis and consumption of undercooked pork or pork tongue may cause T. solium and/or T. asiatica taeniasis. Because the prevalence of undercooked beef consumption was high and because the association between this consumption and taeniasis was relatively strong, elimination of exposure to this risk factor, e.g. via using education campaigns, was estimated to decrease the prevalence of taeniasis by 77%. Elimination of exposure to consumption of undercooked pork and pork tongue was estimated to decrease the prevalence of taeniasis by 24% and 26%, respectively (Chapter 4). Parasitic diseases may have occult and insidious impacts on human health and thus local economies in comparison to bacterial and virological diseases which normally occur acutely with obvious impacts. Due to this, parasitic diseases often do not receive much attention and governmental interest, especially in low-income countries such as Vietnam. Our findings therefore help to raise governmental awareness in Dak Lak as well as Vietnam about taeniasis and T. solium cysticercosis. This study provides a scientific basis and evidence to allow the economic costs and the benefits of intervention to be calculated for T. solium cysticercosis and taeniasis. This would be an important next step to increase the likelihood of advocacy and action from local governments to

230

implement community control measures for these parasites. Finally, the quantified risk factors for taeniasis and T. solium exposure assist in the proposal of efficient strategies to reduce or eliminate the burden of Taenia in Dak Lak.

Pigs are expected to have a higher probability of T. solium cysticerci infection compared with humans because pigs have more frequent exposure to each of the known risk factors for cysticercosis (i.e. free-roaming, scavenging and coprophagy of human stools). While humans and pigs share the same environmental conditions in Dak Lak province, the prevalence of T. solium exposure was substantially higher in humans (5%) than it was in pigs (0.94%) (Chapter 2). This result is in concordance with findings reported by Conlan et al. (2012) [31], who showed that village pigs in Laos possessed a very low prevalence of T. solium cysticercosis (0.8%) compared with that of T. hydatigena (22%), even though the prevalence of T. solium in humans in the same area was relatively high (5.7%). The low prevalence of T. solium porcine cysticercosis was hypothesised by Conlan and co- workers to be due to protective cross-reactivity between T. hydatigena and T. solium.

The literacy rate among the 342 participants that took part in the study described in Chapter 4 was 56% and approximately 91% of participants worked in animal husbandry and agricultural sectors at the smallholder level. If the study participants were typical of the wider population of Dak Lak it is likely that the general level of awareness of hygiene and sanitation and its relationship with health is low. This statement is supported by the relatively high percentage of participants who did not their wash hands after defaecation and before eating (35% and 44%, respectively) (Chapter 4).

Fig. 8.1 provides a schematic diagram of transmission pathways for taeniasis and T. solium seroexposure in the communities of Dak Lak province. Environmental factors such as water sources, the presence of vegetables or soil may be contaminated with Taenia eggs shed from tapeworm carriers (either T. solium, T. saginata or T. asiatica) due to outdoor defaecation was commonly observed within the study communities. Inhabitants may be infected with T. solium cysticerci when drinking untreated water sourced from streams or wells and/or the consumption of raw, unwashed vegetables contaminated T. solium eggs. These factors contributed significantly to T. solium cysticerci exposure in humans with the adjusted population attributable fraction for 231

consumption of raw vegetables and drinking water sourced from lakes, streams or ponds at 74% and 56%, respectively.

The practice of defaecating outdoors in combination with the mobility of humans and animals is conducive to the spread of Taenia eggs into the environment. Understanding the spatial distribution of T. solium taeniasis and cysticercosis and the spatial relationship between taeniasis and cysticercosis provides important background information when developing strategies to either control and /or eliminate T. solium infection. In Chapter 6 we identified a statistically significant positive spatial autocorrelation among human T. solium exposure-positive households in the district of Krong Nang. We speculate that anthropological and fine-scale environmental characteristics within the villages of Krong Nang were sufficient to give rise to spatial clustering of human T. solium exposure to be detected even in the presence of a modest sampling effort (n = 30 households).

8.5 What are the distinctive characteristics of Dak Lak?

The presence of T. asiatica together with T. saginata and T. solium in Dak Lak makes the province typical of most regions in Southeast Asia, harbouring the sympatric existence of all three human tapeworms. While in Vietnam’s North [32, 33] and Samosir Island (Indonesia) [34] T. asiatica was observed to be the dominant Taenia species, T. saginata is dominant in Dak Lak province, similar to Thailand [35] and Laos [31]. The presence of all three Taenia species with the co-existence of dog tapeworm, T. hydatigena, adds to the complexity for understanding the epidemiological features of these Taenia spp. parasites in Dak Lak province. In contrast to the Northern Vietnam where T. solium is characterized by hotspots or foci of infection, the prevalence of T. solium in this study region was low and homogenously dispersed, indicating that T. solium may not pose as serious a public health problem as originally hypothesised.

It is known that different settings may possess different risk factors for the circulation of the parasite. For example, in the mountainous regions of Vietnam’s North, permitting pigs to roam, the lack of latrines and utilization of human faeces (night-soil) for fertilization were identified to be associated with T. solium cysticercosis in pigs (Willingham III et al., 2003). In contrast, the use of night-soil was not a risk factor for 232

porcine cysticercosis in Dak Lak (Chapter 5). The cultivation practices between the two regions are quite different and although night-soil is utilised as a fertilizer for vegetables in the North, this is rarely practiced in the Central Highlands.

Similarly, routine coprophagy of human faeces and scavenging for food were significant risk factors for cysticercosis exposure in pigs (Chapter 5). Although examined, factors such as sex, breed and age of pigs which were associated with porcine cysticercosis in Peru, Mexico, Mozambique, Zambia and Tanzania [40-45], were not found associated with seroexposure. This may due to the difference in biological (the presence of T. hydatigena), social, cultural, or environmental characteristics of the region.

8.6 Trichinella study - The first attempt

This study provided an effort to quantify the prevalence of Trichinella infection in pigs in the Central Highlands of Vietnam for the first time. Trichinella larvae were not found in any of the 835 pork muscle samples screened using the AD method; it is possible that the prevalence reported in this study is not an accurate estimate of the TP of Trichinella infection in Dak Lak. In endemic regions, the prevalence of Trichinella infection in pigs is often less than 1% [36, 37],.. The low sample size may be one of the reasons for the negative result. Furthermore free-roaming pigs are commonly at higher risk of Trichinella infection compared with confined pigs. In the northwest of Vietnam no positive cases of Trichinella were detected in confined pigs [38]. The low proportion of free-roaming pigs (13%, n = 150) tested as part of this study may explain the absence of the parasite in this cohort. In addition, the AD method is known to possess poor diagnostic sensitivity compared with serological and molecular methods and may have resulted in false negative detections.

While the AD method is useful for confirming the result of immunological and molecular diagnostic tests it is known to have relatively poor Se when the burden of Trichinella is low [39]. Assuming the Se of the AD test ranged from 10% to 90% and the Sp was 100%, the TP of Trichinella in Dak Lak was estimated to be anywhere between 0.09% (95% CI 0 to 0.44) and 0.8% (95% CI 0.03 to 4.34), respectively (Chapter 7). In fact, our TP estimate assumes 92% diagnostic Se (0.52%, 95% CrI 0.02 to 1.9) and this results in a prevalence of trichinellosis observed in regions such as Mexico 233

and Argentina where the disease is endemic, with a mere 1% of pigs at slaughter testing positive to AD [36, 37]. From this TP estimation, it is possible that Trichinella is indeed present in Dak Lak as well as the Central Highlands of Vietnam, however could not be definitively ascertained owing to relatively low sample size. Future work subjecting current sera and muscle samples to immuno- or molecular- diagnostic assays will allow an estimate of prevalence to be calculated with improved confidence. .

Because of what appears to be a low prevalence of Trichinella and the relatively poor Se of AD to detect Trichinella in the tested samples, this dissertation was unable to identify risk factors for porcine Trichinella in Dak Lak. Although none of the pigs sampled in this study were Trichinella larvae positive, we cannot assume that the risk of trichinellosis in humans is negligible in this area of Vietnam. Further investigations targeting representative sections of the pig population using cost competitive and sensitive serological and/or molecular diagnostic tools should provide a better estimate of the prevalence of porcine Trichinella infection and risks for human trichinellosis in Dak Lak in the future.

8.7 Recommendations for controlling of the pork-borne zoonoses

The current study demonstrated the high risk of Taenia infection and possibility of the circulation of Trichinella in the communities of Dak Lak province. Both Taenia and Trichinella have the same pig hosts; as the result, the measures for control these pork- borne zoonotic diseases are relatively similar.

There is a broad group of control methods for taeniasis, cysticercosis and Trichinella infection such as community education on health, hygiene and sanitary, meat inspection, pig management, and treatment of taeniasis carriers as well as vaccination of pig populations. Research has pointed out that the implementation of a single method will not result in a satisfactory reduction of disease morbidity or transmission. A combination of approaches focussing on both human and animal hosts should be implemented to control and eliminate the zoonotic parasites. In the communities of Dak Lak province, a One Health approach involving community education involving pig owners, medical- and veterinary- sectors would be asuitable approach and comply with the findings of this study. 234

8.7.1 Community education

Community education includes providing villagers with information about the primary pork-borne zoonotic parasites of importance and their relationship with human and animal health. In turn, this knowledge will guide the development of appropriate sanitation, hygiene and animal management strategies to avoid infection and/or stop the circulation of these parasites within communities. For example, the health education program resulted in a reduction of 43% in porcine cysticercosis incidence in 21 intervention villages in Tanzania (Ngowi et al., 2009) [3]. Health education although important [40, 41], should form a component of a larger or more integrated intervention program in order to obtain a high degree of translational efficacy.

The findings of this study showed that the prevalence of Taenia spp. taeniasis in the communities of Dak Lak province was strongly associated with males, observation of proglottids and the consumption of raw or undercooked meat (pork tongue, beef and pork). T. solium seroexposure in humans associated with the consumption of raw vegetables, sourcing water from streams, lakes or ponds and outdoor defaecation. In pigs, allowing pigs to roam freely, the practice of outdoor defaecation amongst humans and coprophagy of human faeces were the risk factors associated with T. solium seroexposure. The messages included in public health education programs should therefore address each of these factors.

To reduce/eliminate Taenia spp. taeniasis in the communities of Dak Lak province, the massage of education is to encourage consumption of cooked pork and beef amongst inhabitants, especially targeting males who commonly consume these delicacies when consuming alcohol. The message of using a safe drinking water supply, boiling water, washing or cooking of vegetables prior to consumption and using a closed latrine system should be conveyed to the communities to reduce the risk of infection with T. solium cysticerci among resident. The local residents should be educated to confine pigs and the message of the benefit of confining pigs should be delivered to pig owners. The optimal choices to convey education messages to villages are through community loudspeakers, through village head persons, village communal meetings, at schools and/or social organisations such as women’s or youth groups. Among the three study 235

districts in Dak Lak province, porcine T. solium seroexposure in M’Drak district was spatially aggregated. This implies that one or more of the risk factors listed above were particularly prevalent in M’Drak and that education programs should be targeted accordingly. Since many of the aforementioned risk factors are traditional practices in this area of Vietnam, the probability to induce permanent behavioural change from a single education campaign is likely to be small. As proposed above, a One Health approach utilising the expertise of anthropologists might be one way of effecting permanent behavioural change.

8.7.2 Veterinary measures

It is essential that veterinarians retain their role as professional advisors for disease prevention and control in animals. Taenia and Trichinella are zoonotic parasites, and intervention measures developed by veterinary health workers are a key step towards breaking parasite lifecycles and reducing the risk of transmission of infection to humans. There are a lot of veterinary measures to reduce/eliminate the prevalence of T. solium cysticercosis in pigs such as meat inspection, improving pig husbandry, restricting animal movement, treatment of pigs with anthelmintics and vaccination [42]. The findings of this study indicates that improvements in pig husbandry is necessary and will have the great impact on parasite control in Dak Lak province as allowing pigs to roam freely and allowing pigs to ingest human faeces were the risk factors for T. solium exposure.

Pigs are produced in a variety of production systems in Dak Lak province ranging from free-roaming village pigs living on waste, backyard pigs, pigs on small family farms, and large scale integrated farms with high levels of biosecurity [43]. In rural communities of Dak Lak as well as the rest of Vietnam, pig rearing is a temporary activity often associated with low-income families and regarded as a backup financial source with a very low investment outlay. Farmers see no advantage in confining pigs when faced with the financial burden of constructing a pigsty, particularly in times of food shortage. In Dak Lak province where rice production is not a major activity, householders are unable to utilise rice by-products as a source of food for pigs. The immediate economic benefit of allowing pigs to free roam is obvious to most farmers in 236

this area of Vietnam [44]. Gilman et al., (1999) [44] demonstrated a reduction in the incidence of T. solium cysticercosis when pigs were tied or corralled and fed with rice by-products. Similarly, Braae et al. [45] proved that confined pigs had comparable prevalence of T. solium cysticercosis to free-ranging pigs.

To eliminate the practice of allowing pigs to free roam in Dak Lak two key points should be solved. Firstly, materials for pigsty construction and foodstuff resources for pigs should be made more affordable for poor families. Confinement of pigs within a corral using local materials such as bamboo, coffee or rubber trunk to construct a fence could be an option. Secondly, encouragement of local residents to cultivate maize, cassava or sweet potato and even golden snails as a food resources for pigs should be considered. Semi-industrial pig husbandry systems that utilise local feed stuffs and affordable and simple pig housing should be showcased to poor communities to show their benefit to both pig production and health. The utilisation of anthelmintics such as oxfendazole for treatment of porcine cysticercosis is an intervention strategy. Finally, if none of the aforementioned solutions are feasible, it may be necessary to encourage poor families to change their husbandry system from pigs to other, low overhead livestock species such as poultry or rabbits.

8.7.3 Medical measures

Medical measures to eradicate T. solium cysticercosis and Taenia spp. taeniasis include detection and treatment of tapeworm carriers and ensuring appropriate sanitation and hygiene practices.

Detection of tapeworm carriers and chemotherapy with cestodicides in taeniasis patients destroys the source of infection for humans and pigs by eliminating the parasite. Mass medication with cestodicides [46] would provide one option for reducing the prevalence of infection in humans; however, this method seems not feasible for Dak Lak as infection is low and sporadic. Since there is an association betweenTaenia spp. infection and males and with individuals consuming raw/undercooked beef/pork in the communities of Dak Lak province, a risk based approach to target cestodicidal treatment to this group of people is advised. As praziquantel is not ovicidal, it is

237

imperative these individuals be advised to defecate in a closed latrine and if this is not possible, dispose of their stool via incineration / burning to ensure that eliminated tapeworms do not in turn result in further contamination of the environment.

Taeniasis can be detected through self-diagnosis, microscopic examination of stools, copro-PCRs, copro-ELISAs or copro-WB. The association of taeniasis infection with the observation of Taenia proglottids in this study suggests that self-report from infected patients is the feasible approach. To carry out self-diagnosis, individuals need to be equipped with appropriate knowledge about clinical symptoms and in this regard local health workers should play an active education role within their communities.

In Chapter 6, we showed that the relatively low prevalence of exposure to T. solium and although cases of human and porcine T. solium seroexposure in certain locations occurred in close proximity of each other, the regular distribution was found in the most locations of study. These indicate that massive deworming programs in the communities of Dak Lak province are, for the most part, unnecessary. Instead, we recommend that if a case of T. solium seropositive identified then either: (a) individuals resident in the immediate area should be tested to rule out the presence of an exposure or infection cluster; or (b) anthelmintic treatment is offered to individuals resident within a 1500 m radius of the identified case.

Outdoor defecation was a significant factor for T. solium seropositive humans (Chapter 4); this activity may result in free-roaming pigs ingesting human feces (Chapter 5). Public health workers should play an active role in guiding inhabitants on how to build functional latrines. Building a latrine should be cheap and easy to construct in order for it to be translatable to inhabitants in low-income regions. In Dak Lak province as well as remote regions, building a pit latrine seems an economic solution. In case pit latrines are not affordable, guiding the practice of defecating into a deep hole to avoid widespread spread of pathogens into environment or preventing pigs from coprophagy will assist.

238

8.7.4 Other measures to ensure the success of intervention programs

The requirements and preconditions that need to be considered when formulating control and eradication programs for any infectious disease fall into two main general categories: (1) technical-epidemiological, and (2) socio-economic, operational and administrative [47]. The technical-epidemiological preconditions are:

a) There should be a main tool (control measure that is effective in breaking disease transmission). This tool should be simple to apply and be relatively inexpensive. b) The disease should have epidemiological features that facilitate effective case detection in the advanced stages of the eradication program.

The socio-economic, operational and administrative preconditions are:

a) The disease must be of recognised socio-economic importance (at either or both the national and international level). b) There should be a specific reason for elimination (as opposed to control). c) There should be adequate administrative, operational and financial resources available for the entire eradication program. d) There should be an absence of cultural or social conditions that render the methods used to eliminate disease unacceptable to stakeholders.

While much of this dissertation has considered the epidemiological features of taeniasis and cysticercosis, aspects of the socio-economic, operational and administrative preconditions will now be considered.

Long term behavioural change in a population and on-going political support for health interventions occur when aspects of health risk, the economic benefits of disease control and the economic costs associated with disease can be quantified [42]. Financial and human resources are critical factors influencing the success of disease intervention programs. If resources are not available to support an integrated, long-term control and eradication program, a selective approach provides a suitable alternative with the selection of those interventions that provide the most cost-effective return on investment being preferred [46, 48]. The resource for taeniasis, cysticercosis and Trichinella 239

intervention programs in Dak Lak province is limited; therefore, a selective approach is necessary. As discussed above, health education programs involving both the human medical and veterinary health sectors are a feasible intervention option. Choosing a small number of risk factors that make the greatest contribution to the prevalence of taeniasis or cysticercosis are logical targets within an education campaign. The population attributable fraction estimates calculated in Chapter 4 are particularly useful in this regard. As mentioned previously, field data and simulations carried out using disease transmission models [49] show that implementation of a single intervention strategy does not, in general, lead to a satisfactory reduction in disease morbidity and/or transmission. This being the case, and given the scarcity of resources in Dak Lak and recognising that effective disease control is dependent on biological, socioeconomic, behavioural and environmental factors [50], a One Health approach in its very simplest form would involve asking human medical and veterinary health practitioners to deliver the same public health messages when interacting with members of the community. For example, a medical health professional visiting a household that observes pigs free roaming should take time to explain to the householder about the disease risks associated with this practice. Conversely, a veterinarian visiting a household that becomes aware that outdoor defaecation is taking place should explain the risks arising from this practice.

Finally, it is important to recognise that interventions implemented for other disease conditions, for example the WASH (Water, Sanitation and Hygiene) program implemented in Dak Lak to reduce the prevalence of intestinal parasitism, can also have a positive impact on control of taeniasis and cysticercosis.

8.8 Future directions

The presence of all four species of taeniid cestode parasites, T. solium, T. saginata, T. asiatica and T. hydatigena that share the same intermediate or definitive host generates a complex situation when trying to understand aspects of disease epidemiology including prevalence, incidence and factors influencing disease transmission. In Dak Lak, these four taeniid species are sympatric so a comprehensive program of research

240

targeting multiple species (including humans) is necessary to provide a clear picture of infection status and the level of interaction between these parasites.

Most of the serological tests that have been applied to porcine cysticercosis were developed initially for diagnosis of human cysticercosis. Pigs are likely to be exposed to many species of taeniid cestode parasites due to their foraging habits. For this reason, the potential for nonspecific positive serological reactions in porcine T. solium serology is advantageous. It has been shown that many serologically positive pigs are found to have no cysts when necropsied [51–56]. In addition, no assessment has been made of the potential for cross-reactivity when testing for porcine cysticercosis due to exposure to T. saginata, T. asiatica and T. hydatigena [10]. In future, studies that compare serological test results with post-mortem findings are required to quantify Taenia species-specific diagnostic specificity with greater certainty.

A low prevalence of porcine T. solium cysticercosis combined with a relatively high background prevalence of T. hydatigena has been observed not only in Dak Lak province but also in other areas of Vietnam. This situation has also been observed in Laos by Conlan and co-workers (2012) [31]. Conlan et al. (2012) [31] assumed that exposure to T. hydatigena was cross-protective for T. solium in pigs. An animal-based study in which pigs are experimentally infected and tested is necessary to confirm this hypothesis.

The T3qPCR was a useful tool for identification and differentiation of all three Taenia spp. The usefulness of the T3qPCR for large-scale epidemiological studies can be confirmed using a larger sample size of individuals, ideally in a community in which the prevalence of taeniasis is relatively high.

Findings from this study showed the dominance of T. saginata taeniasis in the region. Even though this tapeworm is not a cause of serious health problems in humans, it is associated with direct and indirect economic losses due to reduction of productivity in patients with T. saginata taeniasis and beef carcass condemnation. Intervention programs are necessary to reduce to the prevalence of this parasite in Dak Lak.

241

Further evaluations of the prevalence of Trichinella in domestic and feral pigs using more sensitive serological and molecular diagnostic tools (e.g. Ab-ELISA/WB and real-time PCR directly from muscle digest) using a larger sample size and sampling appropriately from each of the major pig production classes in Dak Lak is necessary to provide an estimate of the prevalence and species of Trichinella infecting pigs with greater accuracy and precision.

242

8.9 References

[1] United Nations. Human Development Index and its components. United Nations Dev Program 2014. http://hdr.undp.org/en/composite/HDI (accessed December 23, 2015).

[2] World Bank. World Development Indicators. World Bank 2014. http://databank.worldbank.org/data/reports.aspx?source=world-development- indicators (accessed December 23, 2015).

[3] Dang-Nguyen TQ, Tich NK, Nguyen BX, Ozawa M, Kikuchi K, Manabe N, et al. Introduction of various Vietnamese indigenous pig breeds and their conservation by using assisted reproductive techniques. J Reprod Dev 2010;56:31–5.

[4] Lemke U, Mergenthaler M, Rö ßler R, Huyen LTT, Herold P, Kaufmann B, et al. Pig production in Vietnam – a review. CAB Rev Perspect Agric Vet Sci Nutr Nat Resour 2008;3.

[5] Nguyen TH. Evaluation of market opportunities for producing local pigs in Dak Lak. Tay Nguyen J Sci 2009;5:21–6.

[6] Dorny P, Praet N, Deckers N, Gabriel S. Emerging food-borne parasites. Vet Parasitol 2009;163:196–206.

[7] Gonzalez AE, Cama V, Gilman RH, Tsang VCW, Pilcher JB, Chavera A, et al. Prevalence and comparison of serologic assays, necropsy, and tongue examination for the diagnosis of porcine cysticercosis in Peru. Am J Trop Med Hyg 1990;43:194–9.

[8] De N V., Hoa LT, Doanh NQ, Ngoc NB, Cong LD. Report on a new species of Taenia (Taenia asiatica) in Hanoi, Vietnam. J Malar Parasite Dis Control 2001;3:80–5.

243

[9] De N V., Hoa LT. Molecular Identification of Human Taenia saginata and Taenia asiatica in Vietnam. Vietnam J Med 2006;1:42–50.

[10] Lightowlers MW, Garcia HH, Gauci CG, Donadeu M, Abela-Ridder B. Monitoring the outcomes of interventions against Taenia solium: Options and suggestions. Parasite Immunol 2016;38:158–69.

[11] Lan NTK, Quyen NT, Hoat PC. The correlation between the prevalence of the tapeworm Taenia hydatigena in dogs and their larvae Cysticercus tenuicollis in cattle and pigs - The effect of tapeworm treatment in dogs. Vet Sci Tech 2011;18:60–5.

[12] Nguyen MTT, Gabriël S, Abatih EN, Dorny P. A systematic review on the global occurrence of Taenia hydatigena in pigs and cattle. Vet Parasitol 2016;226:97–103.

[13] Boa ME, Kassuku AA, Willingham IAL, Keyyu JD, Phiri IK, Nansen P. Distribution and density of cysticerci of Taenia solium by muscle groups and organs in naturally infected local finished pigs in Tanzania. Vet Parasitol 2002;106:155–64.

[14] Dorny P, Phiri IK, Vercruysse J, Gabriel S, Willingham IAL, Brandt J, et al. A Bayesian approach for estimating values for prevalence and diagnostic test characteristics of porcine cysticercosis. Int J Parasitol 2004;34:569–76.

[15] Dorny P, Brandt J, Zoli A, Geerts S. Immunodiagnostic tools for human and porcine cysticercosis. Acta Trop 2003;87:79–86.

[16] Cheng RWK, Ko RC. Cross-reactions between crude antigens of larval Taenia solium (Cysticercus cellulosae) and other helminths of pigs. Vet Parasitol 1991;39:161–70.

[17] Ko RC, Ng TF. Evaluation of excretory/secretory products of larval Taenia solium as diagnostic antigens for porcine and human cysticercosis. J Helminthol

244

1998;72:147–54.

[18] Kumar D, Gaur SNS. Serodiagnosis of porcine cysticercosis by enzyme-linked- immunosorbent-assay (ELISA) using fractionated antigens. Vet Parasitol 1987;24:195–202.

[19] Pathak KML, Allan JC, Ersfeld K, Craig PS. A western blot and ELISA assay for the diagnosis of Taenia solium infection in pigs. Vet Parasitol 1994;53:209–17.

[20] Pinto PSA, Vaz AJ, Germano PML, Nakamura PM. Performance of the ELISA test for swine cysticercosis using antigens of Taenia solium and Taenia crassiceps cysticerci. Vet Parasitol 2000;88:127–30.

[21] Handali S, Gonzalez AE, Hancock K, Garcia HH, Roberts JM, Gilma RH, et al. Porcine antibody responses to Taenia solium antigens rGP50 and sTS18var1. Am J Trop Med Hyg 2004;71:322–6.

[22] Handali S, Pattabhi S, Lee Y-M, Silva-Ibanez M, Kovalenko VA, Levin AE, et al. Development and evaluation of porcine cysticercosis QuickELISA in Triturus EIA Analyzer. J Immunoass Immunochem 2009;31:60–70.

[23] Bueno EC, Scheel CM, Vaz AJ, Machado LR, Livramento JA, Takayanagui OM, et al. Application of synthetic 8-kD and recombinant GP50 antigens in the diagnosis of neurocysticercosis by enzyme-linked immunosorbent assay. Am J Trop Med Hyg 2005;72:278–83.

[24] Hancock K, Khan A, Williams FB, Yushak ML, Pattabhi S, Noh J, et al. Characterization of the 8-kilodalton antigens of Taenia solium metacestodes and evaluation of their use in an enzyme-linked immunosorbent assay for serodiagnosis. J Clin Microbiol 2003;41:2577–86.

[25] Hancock K, Pattabhi S, Greene RM, Yushak ML, Williams F, Khan A, et al. Characterization and cloning of GP50, a Taenia solium antigen diagnostic for cysticercosis. Mol Biochem Parasitol 2004;133:115–24.

245

[26] Hancock K, Pattabhi S, Whitfield FW, Yushak ML, Lane WS, Garcia HH, et al. Characterization and cloning of T24, a Taenia solium antigen diagnostic for cysticercosis. Mol Biochem Parasitol 2006;147:109–17.

[27] Handali S, Klarman M, Gaspard AN, Dong XF, LaBorde R, Noh J, et al. Development and evaluation of a magnetic immunochromatographic test to detect Taenia solium, which causes taeniasis and neurocysticercosis in humans. Clin Vaccine Immunol 2010;17:631–7.

[28] Handali S, Klarman M, Gaspard AN, Noh J, Lee YM, Rodriguez S, et al. Multiantigen print immunoassay for comparison of diagnostic antigens for Taenia solium cysticercosis and taeniasis. Clin Vaccine Immunol 2010;17:68–72.

[29] Lee YM, Handali S, Hancock K, Pattabhi S, Kovalenko VA, Levin A, et al. Serologic diagnosis of human Taenia solium cysticercosis by using recombinant and synthetic antigens in QuickELISATM. Am J Trop Med Hyg 2011;84:587–93.

[30] Gilman RH, Tsang VCW, Cysticercosis THE, Group W. Serodiagnosis Of neurocysticercosis using synthetic 8-KD proteins: Comparison of assay formats 2005;73:771–6.

[31] Conlan J V., Vongxay K, Khamlome B, Dorny P, Sripa B, Elliot A, et al. A cross-sectional study of Taenia solium in a multiple taeniid-endemic region reveals competition may be protective. Am J Trop Med Hyg 2012;87:281–91.

[32] Van De N, Le TH, Lien PTH, Eom KS. Current status of taeniasis and cysticercosis in Vietnam. Korean J Parasitol 2014;52:125–9.

[33] Somers R, Dorny P, Geysen D, Nguyen LA, Thach DC, Vercruysse J, et al. Human tapeworms in north Vietnam. Trans R Soc Trop Med Hyg 2007;101:275–7.

[34] Wandra T, Depary AA, Sutisna P, Margono SS, Suroso T, Okamoto M, et al. Taeniasis and cysticercosis in Bali and North Sumatra, Indonesia. Parasitol Int

246

2006;55 Suppl:S155-60.

[35] Anantaphruti MT. Current status of taeniasis in Thailand. Korean J Parasitol 2013;51:37–42.

[36] Ribicich M, Gamble HR, Rosa A, Bolpe J, Franco A. Trichinellosis in Argentina: an historical review. Vet Parasitol 2005;132:137–42.

[37] Ortega-Pierres MG, Arriaga C, Yepez-Mulia L. Epidemiology of trichinellosis in Mexico, Central and South America. Vet Parasitol 2000;93:201–25.

[38] Vu Thi N, Nguyen VD, Praet N, Claes L, Gabriël S, Huyen NT, et al. Trichinella infection in wild boars and synanthropic rats in northwest Vietnam. Vet Parasitol 2014;200:207–11.

[39] Gamble H, Pozio E, Bruschi F, Nöckler K, Kapel C, Gajadhar A. International Commission on Trichinellosis: Recommendations on the use of serological tests for the detection of of Trichinella infection in animals and man. Parasite 2004;11:39.

[40] Mwidunda SA, Carabin H, Matuja WBM, Winkler AS, Ngowi HA. A school based cluster randomised health education intervention trial for improving knowledge and attitudes related to Taenia solium cysticercosis and taeniasis in Mbulu district, northern Tanzania. PLoS One 2015;10:e0118541.

[41] Sarti E, Flisser A, Schantz PM, Gleizer M, Loya M, Plancarte A, et al. Development and evaluation of a health education intervention against Taenia solium in a rural community in Mexico. Am J Trop Med Hyg 1997;56:127–32.

[42] Gabriël S, Dorny P, Mwape KE, Trevisan C, Braae UC, Magnussen P, et al. Control of Taenia solium taeniasis/cysticercosis: The best way forward for sub- Saharan Africa? Acta Trop 2017;165:252–60.

[43] Lekule FP, Kyvsgaard NC. Improving pig husbandry in tropical resource-poor communities and its potential to reduce risk of porcine cysticercosis. Acta Trop 247

2003;87:111–117.

[44] Gilman RH, Garcia HH, Gonzalez AE, Dunleavy M, Verastegui M, Evans C, et al. Shortcuts to development: methods to control the transmission of cysticercosis in developing countries. In: Garcia HH, Martinez SM, editors. Taenia solium Taeniasis/Cysticercosis, Lima, Peru: Editorial Universo; 1999, p. 313–326.

[45] Braae UC, Magnussen P, Lekule F, Harrison W, Johansen MV. Temporal fluctuations in the sero-prevalence of Taenia solium cysticercosis in pigs in Mbeya Region, Tanzania. Parasit Vectors 2014;7:574.

[46] O’Neal SE, Moyano LM, Ayvar V, Rodriguez S, Gavidia C, Wilkins PP, et al. Ring-screening to control endemic transmission of Taenia solium. PLoS Negl Trop Dis 2014;8:e3125.

[47] NoRabinowitz P. Human-Animal Medicine Title: How Do Laboratories Fit Into the Picture? Am. Assoc. Vet. Lab. Diagnosticians 53rd Annu. Conf., Minneapolis: American Association of Veterinary Laboratory Diagnosticians; 2010.

[48] Winskill P, Harrison WE, French MD, Dixon MA, Abela-Ridder B, Basáñez M- G. Assessing the impact of intervention strategies against Taenia solium cysticercosis using the EPICYST transmission model. Parasit Vectors 2017;10:73.

[49] Kyvsgaard NC, Johansen MV, Carabin H. Simulating transmission and control of Taenia solium infections using a Reed-Frost stochastic model. Int J Parasitol 2007;37:547–58.

[50] Wardrop NA, Thomas LF, Atkinson PM, de Glanville WA, Cook EAJ, Wamae CN, et al. The influence of socio-economic, behavioural and environmental factors on Taenia spp. transmission in Western Kenya: Evidence from a cross- sectional survey in humans and pigs. PLoS Negl Trop Dis 2015;9:e0004223.

248

[51] Muro C, Gomez-Puerta LA, Flecker RH, Gamboa R, Barreto PV, Dorny P, et al. Porcine cysticercosis: possible cross-reactivity of Taenia hydatigena to GP50 antigen in the enzyme-linked immunoelectrotransfer blot assay. Am J Trop Med Hyg 2017.

[52] Sciutto E, Martinez JJ, Villalobos NM, Hernandez M, Jose M V, Beltran C, et al. Limitations of current diagnostic procedures for the diagnosis of Taenia solium cysticercosis in rural pigs. Vet Parasitol 1998;79:299–313.

[53] Sciutto E, Hernandez M, Garcia G, Aluja AS de, Villalobos ANM, Rodarte LF, et al. Diagnosis of porcine cysticercosis: a comparative study of serological tests for detection of circulating antibody and viable parasites. Vet Parasitol 1998.

[54] Jayashi CM, Gonzalez AE, Castillo Neyra R, Rodríguez S, García HH, Lightowlers MW. Validity of the enzyme-linked immunoelectrotransfer blot (EITB) for naturally acquired porcine cysticercosis. Vet Parasitol 2014;199:42–9.

[55] Gavidia CM, Verastegui MR, Garcia HH, Lopez-Urbina T, Tsang VCW, Pan W, et al. Relationship between serum antibodies and Taenia solium larvae burden in pigs raised in field conditions. PLoS Negl Trop Dis 2013;7:e2192.

[56] Devleesschauwer B, Aryal A, Tharmalingam J, Joshi DD, Rijal S, Speybroeck N, et al. Complexities in using sentinel pigs to study Taenia solium transmission dynamics under field conditions. Vet Parasitol 2013;193:172–8.

249

Fig. 8.1 The pattern for transmission and circulation of taeniasis and T. solium exposure in humans and pigs in communities of Dak Lak province

250

Minerva Access is the Institutional Repository of The University of Melbourne

Author/s: Nguyen, Dinh

Title: The epidemiology of pork-borne parasitic zoonoses in rural communities in the central highlands of Vietnam

Date: 2018

Persistent Link: http://hdl.handle.net/11343/219896

File Description: Complete thesis

Terms and Conditions: Terms and Conditions: Copyright in works deposited in Minerva Access is retained by the copyright owner. The work may not be altered without permission from the copyright owner. Readers may only download, print and save electronic copies of whole works for their own personal non-commercial use. Any use that exceeds these limits requires permission from the copyright owner. Attribution is essential when quoting or paraphrasing from these works.