RESEARCH

Community surveys and risk factor analysis of human alveolar and cystic echinococcosis in Ningxia Hui autonomous region, China

Enquêtes en communauté et analyse des facteurs de risque pour les échinococcoses alvéolaire et cystique humaines dans la région autonome du Ningxia Hui, Chine

Encuestas comunitarias y análisis de los factores de riesgo de la equinococosis alveolar y quística humana en la región autónoma de Ningxia Hui, China

Yu Rong Yang^I; Tao Sun^I; Zhengzhi Li^I; Jianzhong Zhang^I; Jing Teng^I; Xongzhou Liu^I; Ruiqi Liu^I; Rui Zhao^I; Malcolm K Jones^II; Yunhai Wang^III; Hao Wen^III; Xiaohui Feng^III; Qin Zhao^III; Yumin Zhao^IV; Dazhong Shi^IV; Brigitte Bartholomot^V; Dominique A Vuitton^V; David Pleydell^V; Patrick Giraudoux^V; Akira Ito^VI; Mark F Danson^VII; Belchis Boufana^VII; Philip S Craig^VII; Gail M Williams^II; Donald P McManus^II,1

^INingxia Medical College, Yinchuan, Ningxia Hui Autonomous Region, China
^IIMolecular Parasitology Laboratory, Australian Centre for International and Tropical Health and Nutrition, The Queensland Institute of Medical Research, 300 Herston Road, Q 4006, Brisbane, Australia
^IIIHydatid Clinical Research Unit, Xingjiang Medical University, Xingjiang, China
^IVDepartment of Parasitology, Lanzhou Medical University, Gansu, China
^VWHO Collaborating Centre for Prevention and Treatment of Human Echinococcosis, University de Franche-Comte and University Hospital, Besancon, France
^VIDepartment of Parasitology, Asahikawa Medical College, Japan
^VIICestode Zoonoses Research Group, Bioscience Research Institute and School of Environment and Life Sciences, University of Salford, Salford, England

ABSTRACT

OBJECTIVE: To determine the true community prevalence of human cystic (CE) and alveolar (AE) echinococcosis (hydatid disease) in a highly endemic region in Ningxia Hui, China, by detecting asymptomatic cases.
METHODS: Using hospital records and "AE-risk" landscape patterns we selected study communities predicted to be at risk of human echinococcosis in Guyuan, Longde and Xiji counties. We conducted community surveys of 4773 individuals from 26 villages in 2002 and 2003 using questionnaire analysis, ultrasound examination and serology.
FINDINGS: Ultrasound and serology showed a range of prevalences for AE (0–8.1%; mean 2%) and CE (0–7.4%; mean 1.6%), with the highest prevalence in Xiji (2% for CE, 2.5% for AE). There were significant differences in the prevalence of CE, AE and total echinococcosis between the three counties and villages (with multiple degrees of freedom). While hospital records showed 96% of echinococcosis cases attributable to CE, our survey showed a higher prevalence of human AE (56%) compared to CE (44%). Questionnaire analysis revealed that key risk factors for infection were age and dog ownership for both CE and AE, and Hui ethnicity and being female for AE. Drinking well-water decreased the risk for both AE and CE.
CONCLUSIONS: Echinococcosis continues to be a severe public health problem in this part of China because of unhygienic practices/habits and poor knowledge among the communities regarding this disease.

RÉSUMÉ

OBJECTIF: Déterminer la prévalence réelle dans les communautés des échinococcoses cystique (EC) et alvéolaire (EA) humaines (hydatidose) dans une région chinoise de forte endémicité (Ningxia) par détection des cas asymptomatiques.
MÉTHODES: A partir des registres hospitaliers et des types paysagers associés à un risque d'EA, on a sélectionné, dans les comtés de Guyuan, Longde et Xiji, une série de communautés à étudier, pour lesquelles on prévoyait un risque accru d'échinococcose humaine. Des enquêtes en communauté ont été effectuées en 2002 et 2003 sur 4773 individus issus de 26 villages, par application d'un questionnaire, exploration par ultrasons et analyse sérologique.
RÉSULTATS: Les explorations par ultrasons et les analyses sérologiques ont mis en évidence les valeurs de prévalence suivantes pour l'EA (0 - 8,1 %, moyenne : 2 %) et la CE (0 -7,4 %, moyenne : 1,6 %), les niveaux de prévalence les plus élevés étant atteints dans le comté du Xiji (2 % pour l'EC et 2,5 % pour l'EA). On a constaté des différences importantes dans la prévalence de l'EC, de l'EA et de l'ensemble des échinococcoses entre les trois comtés et les villages (plusieurs degrés de liberté étant relevés). Bien que les registres hospitaliers recensent une proportion des cas d'échinococcose attribuables à l'EC de 96 %, la présente enquête a fait ressortir une plus forte prévalence de l'EA (56 %) par rapport à l'EC (44 %). L'analyse par questionnaire a révélé que les principaux facteurs de risque d'infection étaient l'âge et la possession d'un chien pour l'EC et l'EA, et l'appartenance à l'ethnie Hui et au sexe féminin pour l'EA. La consommation d'eau provenant d'un puits semblait réduire le risque d'EA et d'EC.
CONCLUSIONS: L'échinococcose reste un grave problème de santé publique dans cette partie de la Chine en raison des pratiques et des habitudes peu hygiéniques et du manque de connaissances sur la maladie parmi les communautés.

RESUMEN

OBJETIVO: Determinar la verdadera prevalencia comunitaria de la equinococosis (hidatidosis) alveolar (EA) y quística (EQ) humana en una región de alta endemicidad de la enfermedad en Ningxia, China, detectando para ello los casos asintomáticos.
MÉTODOS: A partir de registros hospitalarios y patrones de paisaje de la EA, seleccionamos varias comunidades de estudio con riesgo probable de equinococosis humana en los cantones de Guyuan, Longde y Xiji. En 2002 y 2003 realizamos encuestas comunitarias entre 4773 personas de 26 aldeas, utilizando cuestionarios, ecografías y análisis serológicos.
RESULTADOS: La ecografía y la serología revelaron un intervalo de prevalencias de EA (0% - 8,1%; media: 2%) y EQ (0% - 7,4%; media: 1,6%), con la prevalencia más alta en Xiji (2% para la EQ, 2,5% para la EA). Había diferencias importantes de la prevalencia de EQ, EA y equinococosis total entre los tres cantones y aldeas (con varios grados de libertad). Aunque los registros hospitalarios mostraron un 96% de casos de equinococosis atribuibles a EQ, nuestra encuesta mostró una prevalencia de EA humana (56%) superior a la de EQ (44%). El análisis de los cuestionarios reveló que los factores de riesgo de infección más importantes eran la edad y el hecho de tener perro, tanto para la EQ como para la EA; y la pertenencia al grupo étnico Hui y el sexo femenino para la EA. Beber agua de pozo reducía el riesgo tanto de EA como de EQ.
CONCLUSIÓN: La equinococosis sigue siendo un grave problema de salud pública en esa parte de China, debido a unos hábitos y prácticas poco higiénicos y al escaso conocimiento de la enfermedad entre las comunidades.

Introduction

Echinococcosis is caused by adult or larval stages of cestodes belonging to the genus Echinococcus (Taeniidae). Larval infection (hydatid disease; hydatidosis) is characterized by long-term growth of metacestode (hydatid) cysts in the intermediate host.

Echinococcus granulosus and E. multilocularis - the two major species of medical and public health importtance - cause cystic echinococcosis (CE) and alveolar echinococcosis (AE), respectively. While both CE and AE are serious diseases, AE has a high fatality rate and poor prognosis if managed inappropriately.

China is endemic for both CE and AE, with a greater prevalence in the north and north-west.¹ Human cases of CE reported from 33 provinces/autonomous regions in China account for more than 98% of echinococcosis cases with Gansu, Ningxia Hui Autonomous Region (NHAR), Qinghai, Sichuan, and Xingjiang being co-endemic for CE and AE.^1,2 Red foxes and small mammals are the main definitive and intermediate hosts, respectively, for E. multilocularis in the NHAR.³E. granulosus is primarily transmitted between domestic dogs and sheep. CE is present throughout the NHAR, whereas AE occurs in three confluent mountainous counties (Guyuan, Haiyuan, Xiji) in southern NHAR.

We undertook a community survey in 2002 and 2003, using ultrasound and serology, to detect asymptomatic cases for assessing the true prevalence of human CE and AE among rural communities in Guyuan, Longde and Xiji. We also sought to identify risk factors for both these diseases.

Methods

Study area and population

Guyuan, Longde and Xiji counties are situated on the Liupan mountains (average altitude 2200 m above sea level). The socioeconomic structure, land-use and population density of these counties are representative of rural mountainous regions of NHAR. People in this region are poor; subsistence agriculture and livestock herding (sheep, goats, cattle) are the main income sources. During the course of the community surveys and searches of local government records, it became evident that the dog population had decreased dramatically by 1997 due to poisoning as the secondary effect of a poison-bait rodent control programme.

Half the population of Guyuan and Xiji is Han while the other half is Hui Chinese (a minority ethnic group, with their distinct religion (Islamic), lifestyle and customs). Though the majority (91%) living in Longde county are Han, their lifestyle is comparable to those from Guyuan and Xiji. Village populations ranged from 200 to 1900 people (average, 400). We also surveyed three local primary and middle schools in Xiji.

Community surveys

We used hospital records and "AE-risk" landscape profiles⁴ to select village communities in Guyuan, Longde and Xiji counties predicted at risk of human CE and/or AE. We received approval for the surveys from the Ethics Committee of Ningxia Medical College, and obtained written consent from all adult participants and parents of minors five years or older who agreed to participate. We conducted participant interviews using a questionnaire to collect demographic, epidemiological and risk factor data. We took a small blood sample from the ear lobe of each participant for specific antibody testing by enzyme linked immunosorbent assay (ELISA) using E. granulosus cyst fluid antigen B (AgB) and E. multilocularis crude protoscolex extract (EmP).^5,6 We performed an abdominal ultrasound (US) scan on each participant, using a portable ultrasonograph (Aloka, Japan, model 3.5 MG67N- 35F2.4) to differentiate advanced AE from CE in the liver.⁷ We used serum (prepared from 5 ml venous blood samples) from individuals with a surgical history of CE or AE, an abnormal US image and those with no cystic lesions detected by US during the surveys, as negative controls. We transported serum samples at 4ºC and later stored them at –20ºC before processing for ELISA.

We employed the WHO recommended US classification for CE⁸ and the PNM system, proposed by the European Network for Concerted Surveillance for classification of AE.⁹

Statistical analysis of data

We used Epi-Info and SPSS 11.5 to analyyse data from interviews/questionnaires, US scanning and serology. Differences among groups were compared using the c² test. Odds ratios and 95% confidence intervals for the multivariate analysis were calculated using multiple logistic regression models.

Results

Structure of the surveyed populations

The sex ratios (female:male) in the three counties were 1:1.09 in Guyuan and Longde (Table 1) and 1:1.11 in Xiji. The age range was 5–83 years (average, 32 years; females, 34 years, males, 30 years). Age structures were similar, except for peaks in the age group of 11–20 years in Guyuan and Xiji, and 31–40 years in Longde (Table 2). Surveyed subjects were of both Han and Hui nationality in Guyuan and Xiji counties, but in Longde, the residents were all Han (Table 1). The residents were mainly farmers (64%) or students (31%), with the remainder (5%) comprising businessmen, village leaders, civil servants, teachers, public health workers and military personnel.

Of the 4778 who participated in our 2002 and 2003 surveys, 4773 were residents of the surveyed areas and belonged to 26 communities within 16 townships in the three counties (Table 3). We surveyed approximately 25% of the total rural population from Guyuan, 80% from Longde and 12.5% from Xiji. Within Xiji, 2.3% of the population was urban and 73% rural. The average rural population coverage was 36%.

CE and AE prevalence determined by US

The overall echinococcosis prevalence determined by US was 3.6% (171/4773); 0–8.1% (mean 2%) for AE and 0–7.4% (mean 1.6%) for CE. The highest prevalence of echinococcosis was 4.5% in Xiji (2% for CE, 2.5% for AE), while Guyuan and Longde counties had prevalences of 0.9% (0.3% for CE, 0.6% for AE) and 0.6% (CE only), respectively (Table 1). There were significant differences in prevalence of CE, AE and total echinococcosis infections between Guyuan and Xiji (c² = 12.4, P <0.001; c² = 11.9, P <0.001; c² = 24.6, P <0.001), of AE between Xiji and Longde (c² = 2.9, P <0.05) and of total echinococcosis infections (c² = 4.3, P <0.05) (Table 1). There were no significant differences between Guyuan and Longde for CE and total echinococcosis infections, or between Xiji and Longde for CE infection, but a comparison of the three counties for CE, AE and total echinococcosis showed significant differences (c² = 14.2, P <0.001; c² = 16.6, P <0.001; c² = 30.3, P <0.001).

Some villages, especially in Xiji county, had a higher AE prevalence than CE. Some had either high or low prevalences of both diseases, and in some no cases were detected. We compared prevalences using a c² test with multiple degrees of freedom as many villages had low or zero prevalence for CE or AE. Our results showed a high degree of heterogeneity in prevalences among communities and villages within/ between counties (Table 3). We found highly significant (P <0.01) differences for CE, AE and total echinococcosis for village communities within Xiji county and between Xiji and Guyuan. However, there were no significant differences in echinococcosis prevalences within villages in Guyuan county.

We found no significant gender associations; AE cases occurred in 62.5% (60/96) females (c² = 7.8, P <0.01) and CE cases occurred in 58.6% (44/75) females (c² = 3.2, P = 0.07) (Table 1). There was a highly significant difference between the Hui and Han ethnic groups for AE (c² = 5.2, P <0.05) (Table 1).

The ages of AE and CE patients ranged from 19–73 years and from 18–79 years, respectively (Table 2). Age-prevalences increased from 0.12% in those aged 20 years or less to 6.8% for those above 61 years for AE and from 0.06% in those 20 years or less to 6.0% for those above 61 years for CE.

Soroprevalence among schoolchildren

Of the 121 females and 148 males (age 7–15 years) tested serologically using EmP and AgB antigens for antibody detection by ELISA (Table 4), we found significantly more EgB seropositive individuals (c² = 24.2, P <0.01) among Hao-Zi-Wan school students, all of whom were Han. Significantly more students from Nan-Wan school displayed sero-reactivity for EmP (c² = 8.3, P <0.01), all of whom were Hui. Seropositive students from Huo-Shi-Zhai local schools were predominantly of Hui ethnicity, in whom anti-EmP reactivity was significantly higher than for EgB (c² = 5.6, P <0.05). Seropositive rates between boys and girls in the different schools were not significantly different.

Questionnaire and relative risk analysis

We evaluated 4773 participants and estimated prevalence odds ratios, with P-values, for each potential risk factor for CE and AE obtained by univariate analyssis as a preliminary screen (Table 5; web version only, available from http://www. who.int/bulletin). The risk of having either AE or CE significantly increased among individuals older than 30 years, farmers and dog owners (especially those who had owned a dog for over 5 years). Increased risk for AE was associated with Hui ethnicity, female gender and drinking spring-water. There was decreased risk for AE infection among those who drank well-water and decreased risk for CE infection among those who drank unboiled water.

To allow for confounding effects we carried out a multivariate analysis using a logistic regression model (Table 6) to further investigate the risk factors, including those that had shown associations with disease in the univariate analysis. The possibility of AE or CE infection increased among individuals older than 30 years and dog owners, but not among farmers. Increased risk for AE was also associated with Hui ethnicity and being female, but not with drinking spring-water. Drinking well-water decreased the risk for AE and CE, while drinking unboiled water did not decrease the probability of being infected with CE.

Discussion

Despite the establishment of extensive and successful control programmes for CE and AE, E. granulosus and E. multilocularis continue to have a wide geographical distribution. This is a cause for concern as they are likely to persist or re-emerge in many endemic areas worldwide and also readily spread from endemic to non-endemic areas, causing severe disease and considerable economic loss.^1,10 Our study provides an update on a highly endemic focus for AE and CE in NHAR, China and exemplifies the serious public health problem that echinococcosis presents, particularly in poor rural communities.

A previous retrospective clinical study of echinococcosis incidences in southern NHAR identified highly endemic foci and a highly heterogeneous distribution of both CE and AE cases.² This survey of hospital records for the period 1994–2001 showed the incidence of combined cases of CE and AE to be 7 per 100 000 for southern NHAR, comppared with 1 per 100 000 for the northern zone, represented by the capital city, Yinchuan. Human AE cases were only recorded in a confluent area comprising the three southern counties of Guyuan, Haiyuan and Xiji despite CE cases occurring over the whole of NHAR.² While the hospital records showed an overall majority (96%) of cases resulting from CE, our community survey results revealed a higher prevalence of human AE (56%) compared to CE (44%), thereby demonstrating that hosppital records though relevant for CE are grossly inaccurate for AE case detection, particularly in this rural under-developed region. Although the overall human echinococcosis prevalence detected by US was 3.6%, the distribution was heterogeneous between counties, and also within Xiji. We found that while a retrospective survey of hospital records was beneficial for identifying regions of endemicity,² active community surveys revealed far more detailed information in rural areas, especially for AE.

The asymptomatic period for human AE and CE is 5–15 years, but CE is much more likely to be detected by the patients themselves and confirmed by local or regional health centres. Human echinococcosis is endemic in regions with poor economic conditions, accompanied by low education levels and poor medical facilities,¹¹ thus resulting in these predominantly rural people often failing to seek medical attention.¹² Therefore, active identification of human echinococcosis cases at the community level, especially in rural areas, is a necessary step for accurately evaluating the true extent of the disease in a given endemic focus.¹

The risk factors we identified for AE and CE - poor hygienic practices, femmale gender, low income, limited education and dog ownership - were similar to those reported in previous studies.^13–17 Although dogs from the neighbouring Gansu Province were known to be infected with E. multilocularis, this was not confirmed for dogs in Ningxia.^3,18 In Gansu Han communities, dog ownership was a risk factor for AE.¹⁹ In contrast to a previous study,20 in our survey echinococcosis was not significantly associated with uncooked food collected from the field or garden, livestock ownership, or home slaughtering. Our results revvealed some unexpected results, such as increased risk related to having some knowledge of human echinococcosis, probably due to misunderstanding of the true risk factor situation. This emphassizes the necessity for investigators to be aware of the possibility of bias in answers by interviewees or in the interpretation of risk factor data.

In contrast to another study,²¹ farming was not an important risk factor for echinococcosis in our analysis, probably because the majority of our participants came from a rural population, and were thus farmers.

As reported in another study²² our results showed that use of well-water decreased the risk of contracting either AE or CE. The wells in our study area were deep (because of the low water table), sturdy and less likely to be contaminated by faeces from wild animals or livestock, as protective lids covered the majority of wells.

Within the surveyed zone, vegetattion and wildlife, including foxes used to be more abundant in this mountainous area in the past and E. multilocularis infections in animals were well docummented in the 1980s.^3,23–25 During the past 50 years, the natural landscape (such as forest, pasture) in NHAR has been modified to an artificial landscape of farmland and urban areas thus increasing soil erosion and changing local climatic conditions with more frequent droughts, floods and pests, thereby affecting wildlife biodiversity.^19,26 The variable prevalence of human AE and CE among communities may be associated with different micro-environmental features of various localities, including a range of biotic and abiotic factors.^27–30

Our results showed a predominance of AE and CE infections in females in NHAR probably because women are more involved in farming and herding livestock, as well as more likely than men to come in contact with dogs. Increased susceptibility to Echinococcus infection, however, may also be affected by physiological and genetic factors.^31,32 The lack of gender differences in Echinococcus seropositivity among teenage schoolchildren suggests that there is a similar level of exposure to infection for boys and girls. The higher prevalence of infection in adult females may be the effect of physiology and farming or increased contact with dogs.¹⁹

In concordance with previous reports,^13,14,33 our survey revealed some association between infection prevalence and unhygienic behaviour or lifestyle. In one of the villages, where teenagers showed higher EmP than EgB sero-reactivity, tap water had replaced spring water in 1997 and the numbers of wild and owned dogs had also decreased considerably. We have evidence that foxes (susceptible rodent species for AE) occur in areas (identified by faeces) close to this community,³⁴ suggesting current continued transmission of AE. However, the combined effects of the change in water source and the reduction in definitive hosts (dogs) may explain the apparent interruption of CE transmission to humans in this locality. Although, current use of tap water seems safer than the previous use of unprotected spring or river water, the lifestyle of these rural people has not changed thus continuing AE transmission.

Conclusions

Unhygienic practices and habits emphasize the poor knowledge that NHAR communities have about echinococcosis, thus sustaining active transmission of E. granulosus and E. multilocularis. Since infection occurs though accidental ingestion of contaminated (with Echinococcus eggs released in dog or fox faeces) garden vegetables, water or soil,^35,36 residents in endemic areas should be educated about careful washing of hands and food, and the importance of drinking clean, uncontaminated water. Feeding of livestock offal to dogs should be avoided as it is a starting point for the E. granulosus life cycle.³⁵ We suggest that health education programmes integral to echinococcosis control be initiated early because most infections are acquired during childhood.^37,38 Control measures for CE and AE should focus on decreasing dog numbers and active anthelminthic dosing of owned dogs.^28,39

Acknowledgements

We thank Ningxia Medical College, University of Queensland and the NIH/ NSF "Transech" Echinococcosis China Working Group for providing technical assistance for the community surveys.

Funding: Ningxia Medical College, University of Queensland and the NIH/ NSF "Transech" Echinococcosis China Working Group provided financial support (Ecology of Infectious Diseases Program NIH/NSF#1565) for the community surveys.

Competing interests: none declared.

References

1. Craig PS, Rogan MT, Campos-Ponce M. Echinococcosis: disease, detection and transmission. Parasitol 2003;127 Suppl:S5-20.        

2. Yang YR, Sun T, Li Z, Li XP, Zhao R, Cheng L, et al. Echinococcosis, Ningxia, China. Emerg Infect Dis 2005;11:1314-6.        

3. Li WX, Zhang GC, Lin YG, Hong LX. The occurrence of Echinococcus multilocularis Leuckart, 1863, the natural animal hosts in China and its morphological study. Acta Zool Sinica 1985;31:365-71. In Chinese.        

4. Danson FM, Graham AJ, Pleydell DR, Campos-Ponce M, Giraudoux P, Craig PS. Multi-scale spatial analysis of human alveolar echinococcosis risk in China. Parasitol 2003;127 Suppl:S133-41.        

5. Craig PS, Deshan L, MacPherson CN, Shi DZ, Reynolds D, Barnish G, et al. A large focus of alveolar echinococcosis in central China. Lancet 1992; 340:826-31.        

6. Craig PS, Rogan MT, Allan JC. Hydatidosis and cysticercosis-larval cestodes. In: Gillespie SH, Hawkey PM, editors. Medical parasitology: A practical approach. London: IRL Press; 1995. p. 209-37.        

7. Bartholomot G, Vuitton DA, Harraga S, Shi da DZ, Giraudoux P, Barnish G, et al. Combined ultrasound and serologic screening for hepatic alveolar echinococcosis in central China. Am J Trop Med Hyg 2002;66:23-9.        

8. Macpherson CN, Bartholomot B, Frider B. Application of ultrasound in diagnosis, treatment, epidemiology, public health and control of Echinococcus granulosus and E. multilocularis. Parasitol 2003;127 Suppl:S21-35.        

9. Pawlowski ZS, Echert J, Vuitton DA, Ammann RW, Kem P, Craig PS, et al. Echinococcosis in humans: clinical aspects, diagnosis and treatment. In: Eckert J, Gemmell MA, Meslin X, Pawlowski ZS, editors. WHO/OIE Manual on Echinococcosis in Humans and Animals: a Public Health Problem of Global Concern. World Organisation for Animal Health: Paris, France; 2001. p 20-66.        

10. McManus DP, Zhang W, Li J, Bartley PB. Echinococcosis. Lancet 2003; 362:1295-304.        

11. Vuitton DA, Zhou H, Bresson-Hadni S, Wang Q, Piarroux M, Raoul F, et al. Epidemiology of alveolar echinococcosis with particular reference to China and Europe. Parasitol 2003;127 Suppl:S87-107.         

12. Craig PS, Rogan MT, Allan JC. Detection, screening and community epidemiology of taeniid cestode zoonoses: cystic echinococcosis, alveolar echinococcosis and neurocysticercosis. Adv Parasitol 1996;38:169-250.         

13. Carmona C, Perdomo R, Carbo A, Alarez C, Monti J, Grauert R, et al. Risk factors associated with human cystic echinococcosis in Florida, Uruguay: results of a mass screening study using ultrasound and serology. Am J Trop Med Hyg 1998;58:599-605.         

14. Dowling PM, Torgerson PR. A cross-sectional survey to analyse the risk factors associated with human cystic echinococcosis in an endemic area of mid-Wales. Ann Trop Med Parasitol 2000;94:241-5.        

15. Dowling PM, Abo-Shehada MN, Torgerson PR. Risk factors associated with human cystic echinococcosis in Jordan: results of a case–control study. Ann Trop Med Parasitol 2000;94:69-75.         

16. Gottstein B, Saucy F, Deplazes P, Reichen J, Demierre G, Busato A, et al. Is high prevalence of Echinococcus multilocularis in wild and domestic animals associated with disease incidence in humans? Emerg Infect Dis 2001; 7:408-12.         

17. Schantz PM, Wang H, Qiu J, Liu FJ, Saito E, Emshoff A, et al. Echinococcosis on the Tibetan Plateau: prevalence and risk factors for cystic and alveolar echinococcosis in Tibetan populations in Qinghai Province, China. Parasitol 2003;127 Suppl:S109-20.         

18. Shi D. The first report of Echinococcus multilocularis in domestic dogs in Zhang County of Gansu Province. Chin Med J 1995;108:615-7.         

19. Craig PS, Giraudoux P, Shi D, Bartholomot B, Barnish G, Delattre P, et al. An epidemiological and ecological study of human alveolar echinococcosis transmission in south Gansu, China. Acta Trop 2000;77:167-77.         

20. Nasrieh MA, Abdel-Hafez SK, Kamhawi SA, Craig PS, Schantz PM. Cystic echinococcosis in Jordan: socioeconomic evaluation and risk factors. Parasitol Res 2003;90:456-66.         

21. Kern P, Ammon A, Kron M, Sinn G, Sander S, Petersen LR, et al. Risk factors for alveolar echinococcosis in humans. Emerg Infect Dis 2004;10:2088-93.         

22. Torgerson PR, Karaeva RR, Corkeri N, Abdyjaparov TA, Kuttubaev OT, Shaikenov BS. Human cystic echinococcosis in Kyrgystan: an epidemiological study. Acta Trop 2003;85:51-61.         

23. Hong LX, Lin GY. Studies on the development and histopathology of alveolar cestode of Echinococcus multilocularis in human and animal hosts. Endemic Dis Bull 1987;2:51-3. In Chinese.         

24. Li W. Investigation of hosts of Echinococcus multilocularis in Ningxia. Endemic Dis Bull 1986;1:131-5. In Chinese.         

25. Wang XT. Hydatidosis/Echinococcus. In: Wang XT, editor. Scientific investigation in Liupan Mountains, Ningxia. Yinchuan: Ningxia People’s Press; 1989. p. 248-51. In Chinese.         

26. Giraudoux P, Craig PS, Delattre P, Bao G, Bartholomot B, Harraga S, et al. Interactions between landscape changes and host communities can regulate Echinococcus multilocularis transmission. Parasitol 2003;127 Suppl:S121-31.         

27. Anuradha, Pal R, Katiyar JC. Sex-influenced population kinetics of Leishmania donovani in hamsters. Indian J Exp Biol 1990;28:876-9.         

28. Gemmell MA. Australasian contributions to an understanding of the epidemiology and control of hydatid disease caused by Echinococcus granulosus - past, present and future. Int J Parasitol 1990;20:431-56.         

29. Nelson GS. Hydatid disease: research and control in Turkana, Kenya. 1. Epidemiological observations. Trans R Soc Trop Med Hyg 1986;80:177-82.        

30. Veit P, Bilger B, Schad V, Schafer J, Frank W, Lucius R. Influence of environmental factors on the infectivity of Echinococcus multilocularis eggs. Parasitol 1995; 110:79-86.         

31. Vuitton DA. The ambiguous role of immunity in echinococcosis: protection of the host or of the parasite? Acta Trop 2003;85:119-32.         

32. Vuitton DA, Zhang SL, Yang Y, Godot V, Beurton I, Mantion G, et al. Survival strategy of Echinococcus multilocularis in the human host. Parasitol Int 2006. In press.         

33. Abu-Hasan N, Daragmeh M, Adwan K, Al-Qaoud K, Abdel-Hafez SK. Human cystic echinococcosis in the West Bank of Palestine: surgical incidence and seroepidemiological study. Parasitol Res 2002;88:107-12.         

34. Giraudoux P, Pleydell D, Raoul F, Quere J, Wang Q, Yang Y, et al. Transmission ecology of Echinococcus multilocularis: What are the ranges of parasite stability among various host communities in China? Parasitol Int 2006. In press.         

35. Craig PS, Macpherson CN, Watson-Jones DL, Nelson GS. Immunodetection of Echinococcus eggs from naturally infected dogs and from environmental contamination sites in settlements in Turkana, Kenya. Trans R Soc Trop Med Hyg 1988;82:268-74.         

36. World Health Organization. Guidelines for treatment of cystic and alveolar echinococcosis in humans. WHO Informal Working Group on Echinococcosis. Bull World Health Organ 1996;74:231-42.         

37. Gorbach SL, Barza M. Where have all the case reports gone? Clin Infect Dis 2001;32:1.         

38. Kammerer WS, Schantz PM. Echinococcal disease. Infect Dis Clin North Am 1993;7:605-18.         

39. Ito A, Romig T, Takahashi K. Perspective on control options for Echinococcus multilocularis with particular reference to Japan. Parasitol 2003;127 Suppl: S159-72.        

(Submitted: 23 August 2005 - Final revised version received: 14 February 2006 - Accepted: 17 February 2006)

1 Correspondence to Dr McManus (email: donM@qimr.edu.au).