Abstract
A survey for fish-borne zoonotic trematode (FZT) metacercariae in fish from the Northern Mountain Region of Vietnam was conducted from May to August 2014. The major objective was to evaluate the impact of water body types on the prevalence of FZT. A total of 846 fish (31 species) were collected from three water body types, including aquaculture ponds, rivers, and reservoirs. A pepsin digestion method was used for the recovery of metacercariae from fish. Overall, 29 of the 31 fish species were infected with FZTs. Prevalence was 56.8% in river fish, 46.7% in pond fish, and 29.3% in reservoir fish. The prevalence of Clonorchis sinensis metacercariae was 10.6% in reservoir fish, and 2.6% in river fish; fish from ponds were uninfected. The infection intensity of C. sinensis was 29.7 metacercariae/reservoir fish and 2.1 metacercariae/river fish. The prevalence of C. sinensis was highest (25.9%) in Toxabramis houdemeri. The intestinal flukes Haplorchis pumilio, Haplorchis taichui, Haplorchis yokogawai, Centrocestus formosanus and Procerovum varium were recovered from 29 fish species, with an overall prevalence of 46.8% and a mean infection intensity of 23.6 metacercariae. The results indicate that a C. sinensis sylvatic cycle involving wild fish species is important in the epidemiology of liver fluke and that consumption of raw or inadequately prepared wild-caught fish is a the major risk factor for human Clonorchiasis in this region. Both wild-caught and cultured fish present a high risk for infection with intestinal flukes. The habitat requirements of the snail host may be the important determining factor in favoring a sylvatic cycle of C. sinensis. This information is relevant to the development of effective prevention and control strategies, and emphasizes the priority that must be given to education of the local communities on the risks of consuming raw or improperly prepared fish dishes.
Introduction
F
Although intestinal flukes are very common in fish raised in aquaculture ponds, C. sinensis metacercariae have been infrequently recovered in cultured fish in Vietnam (Phan et al. 2010b, Clausen et al. 2012). O. viverrini metacercarie, however, have been reported in wild-caught fish in South and Central South Vietnam (Thu et al. 2007, Vo et al. 2014). This host pattern has also been reported elsewhere in East and Southeast Asia (Chai et al. 2007, Zhang et al. 2007, Chen et al. 2010, Li et al. 2013, Pitaksakulrat et al. 2013). The relative importance of different fish species and their specific habitats is important to the design of public health programs to prevent and control FZT, particularly liver flukes, and for the clarification of food safety risks associated with aquaculture, a major economic activity in these endemic countries. The recent discovery of high transmission of C. sinensis in fish in reservoirs in the North Vietnam Mountain Region (Thanh et al. 2016) offered an opportunity to investigate the relative risk of cultured and wild fish for infection with C. sinensis and intestinal FZTs in a rural community.
Materials and Methods
Study area
The Northern Mountain Region (NMR) of Vietnam occupies 103,000 km2, about one third of the country's area, and is inhabited by about 12 million people (15% of the national population) living in more than 2000 communes (Rambo 1997). It is home to more than 30 different ethnic groups (Vien 2003). The poverty rate of ethnic minorities in the NMR is 67.3% (Cuong 2012). Poor people living in or near wetland areas such as lakes, reservoirs, and rivers have limited opportunities for agricultural diversification and, therefore, are highly dependent on aquatic resources, especially fish (Anh et al. 2003). Five provinces in NMR were selected for the study, including Dien Bien, Lao Cai, Son La, Yen Bai, and Thai Nguyen.
Fish collection
The study was conducted from May to August 2014 in the five selected NMR provinces. Wild-caught fish were collected from local markets, or purchased directly from fishermen. Fish produced in aquaculture farms were purchased directly from the farmers. The total number of fish collected was 846, and comprised of 31 species. This included 450 fish from 7 rivers, 210 fish from 5 reservoirs, and 186 fish from 7 aquaculture ponds. Fish were transported in iceboxes to the laboratory based at the Research Institute for Aquaculture No. 1 in Bac Ninh province. These fish were identified to species based on the taxonomy guides in “Freshwater Fish in Vietnam” (Hao and Van 2001, Hao 2005a, b).
Isolation and identification of FZTs
Individual fish were weighed before examination for metacercariae by a pepsin digestion process. Fish weighing less than 100 g were processed whole. Individuals weighing more than 100 g were subdivided by taking samples from fish's head, including the gills, muscles of the main body, and the caudal and anal fins. The subsamples were pooled together (weighing ∼50 g) for digestion. Whole fish or pooled subsamples were then ground and then placed in 1% pepsin digestion medium to release metacercariae, following the methods described by Thien et al. (2007) and Chi et al. (2008). The digest sediment, after washing, was examined under a stereomicroscope and the metacercariae present were isolated and counted. Representative metacercariae were mounted on glass slides and observed under a compound microscope for taxonomic identification. To assist the morphological identification, some of the encysted metacercariae were excysted by pressing down on the coverslip to rupture the cyst and free the metacercariae, where they were more easily studied. Excystment was also achieved by subjecting the metacercarial cyst to trypsin digestion (0.4% sodium hydrogen carbonate, 1.0% trypsin, 0.89% NaCl) until they emerged from the cyst (Komiya 1965). The metacercariae was identified based on the morphological characters described in Yamaguti (1971), Velasquez (1973), Pearson and Ow-Yang (1982), Scholtz et al. (1991), and Sohn (2009). The differentiating features used for diagnosis were as follows. C. sinensis: The metacercarial cyst wall of C. sinensis is relatively thin and transparent, the ventral and oral suckers are prominent and are of equal size, a ventrogenital sac is absent, two testis are present, and the excysted metacercaria is elongate. Haplorchis spp.: the metacercaria possess a single testis, oral and ventral suckers are not equal in size, and the ventral sucker is a ventrogenital complex (sac), which is armed with hooklets (16–18 hooklets in H. taichui, 36–42 hooklets in H. pumilio, and 70 hooklets in Haplorchis yokogawai) and a gonotyl is absent. C. formosanus: similar to Haplorchis, but with two testes and an unarmed gonotyl. The oral sucker is encircled by 2 rows of 32 large spines, and the excretory vesicle is X or Y shaped. P. varium: the metacercaria of P. varium has a single testis, hooklets are absent on the ventral sucker, and it possesses an armed gonotyl.
To confirm morphological identifications a subset of encysted metacercariae were inoculated orally into either hamsters or mice following the protocol of Kay et al. (2009). Animals inoculated with presumptive liver fluke metacercariae were necropsied for worm recovery 30 days postinfection and those inoculated with intestinal fluke metacercariae after 5 to 7 days. The worms were fixed and stained following the procedures of Kay et al. (2009) and Vo et al. (2014). Identifications based on adult worm morphological characters were made using the diagnostic keys of Yamaguti (1971) and Pearson and Ow-Yang (1982).
Data analysis
Data were entered into Microsoft Excel datasheets (Microsoft Office, version 2010) and imported to STATA IC 12.1 (Corp LP) for analysis. Prevalence and intensity of FZT infections were calculated for each fish species and type of water body. The prevalence among infected fish species were compared by using logistic regression. Negative binomial regression in the generalized linear model, link (log), was used to compare differences in the intensity of FZTs among infected fish species and water body types. Significance was set at a p-value equal to or less than 0.05.
Results
The results of the investigation showed that FZTs are highly prevalent in fish collected from the three different water types in the NMR of Vietnam (Table 1); infections were detected in 29 of the 31 fish species sampled. The overall prevalence and mean intensity of FZT in fish from all water bodies were 47.4% and 25.5 metacercariae/infected fish, however, this varied significantly between the body water types (Table 1). FZT prevalence was highest in river fish and pond fish; pond fish, in turn, had a higher infection rate than reservoir fish. However, infection intensity in pond fish was significantly greater than in fish from either rivers or reservoirs.
Superscripts difference indicates significant (p ≤ 0.05) differences in prevalence and intensity—in a column, different letters indicate significant difference in the value; (mc) metacercariae/infected fish sample.
FZT, fish-borne zoonotic trematode.
Six species of FZT metacercariae were recovered from the fish, including C. sinensis, H. pumilio, H. taichui, H. yokogawai, C. formosanus, and P. varium (Table 2). The adult worms of C. sinensis recovered from experimentally infected hamsters exhibited dendritic testes, differentiating them from Opisthorchis spp. Adult worms of P. varium recovered from experimental mouse infections revealed the prominent genital expulsor (100–130 μm long). The identity of the other species of intestinal flukes (Haplorchis spp., and C. formosanus) that were recovered from experimentally infected mice was also confirmed from adult morphological features.
Superscripts difference indicate significant (p ≤ 0.05) differences in prevalence and intensity—in a column, different letters indicate significant difference in the value.
Prevalence and intensity of individual FZT species varied in fish collected from the different types of water bodies (Table 2). Although pond fish were not infected with C. sinensis, it was commonly recovered from river and reservoir fish. There was no significant difference in C. sinensis prevalence between river and reservoir fish, however, the intensity of infection in reservoir fish was significantly higher than in river fish. The intestinal FZTs were highly prevalent in all three water body types. H. pumilio was the dominant FZT species, its prevalence and intensity in fish was consistently high and not statistically different among fish from different water body types. The prevalence of H. taichui in reservoir fish was lower than in fish from ponds and rivers, but the difference in intensity was not significant. The prevalence of metacercariae of H. yokogawai was relatively low overall and did not vary significantly between water body types. The prevalence and intensity of C. formosanus was greatest in the river fish. Metacercariae of P. varium were only recovered from fish collected from ponds and reservoirs, but the prevalence and intensity of metacercariae were not statistically different between these two water types.
Comparison of FZT species prevalence and intensity from different fish species revealed significant host interspecific variation. All of the eight fish species collected from ponds were infected with metacercariae of at least one of five intestinal flukes (Table 3). H. pumilio metacercariae were found in seven fish species, and its prevalence and intensity varied significantly between fish species (p ≤ 0.001); highest prevalence was observed in Prochilodus lineatus, Anabas testudineus, Hypophthalmichthys molitrix, Cyprinus carpio and Ctenopharyngodon idellus. The intensity of infection was greatest in C. idellus and C. carpio. Metacercariae of H. taichui was recovered from six species of pond fish and the prevalence was significantly different among the fish species (p = 0.025), while the intensity was not statistically different (p = 0.769). C. formosanus was recovered from four fish species (A. testudineus, C. idellus, C. carpio, and H. molitrix), but there was no significant difference in prevalence and intensity among fish species (p ≥ 0.085). H. yokogawai and P. varium were found only in A. testudineus, which had a low infection (both prevalence and intensity) of H. yokogawai, but a high prevalence and infection intensity of P. varium metacercariae.
Six of seven fish species collected from reservoirs were infected with FZT metacercariae (Table 4). Three fish species, Cultrichthys erythropterus, Hemiculter leucisculus, and Toxabramis houdemeri, were positive for C. sinensis metacercariae. Prevalence and intensity were significantly different among the fish species (p ≥ 0.007). Both prevalence and intensity of C. sinensis was significantly higher in T. houdemeri compared with C. erythropterus and H. leucisculus. All six fish species harbored H. pumilio, and the prevalence was highest in T. houdemeri, followed by C. carpio and Clarias fuscus. The intensity of infection was not significantly different between these species (p = 0.343). Low prevalence and intensity of H. yokogawai were observed in T. houdemeri. C. formosanus was recovered from only one fish species, C. erythropterus. None of the reservoir fish was infected with P. varium metacercariae.
Twenty-four fish species (of 26) collected from the rivers were infected with one or more FZT species (Table 5). C. sinensis was present in H. leucisculus and C. erythropterus both at low prevalence and intensity. H. pumilio was recovered from 23 fish species; the prevalence and intensity varied significantly between the fish species (p ≤ 0.001). Fourteen fish species were infected with C. formosanus, however prevalence and intensity of infection did not differ significantly among the fish species. Thirteen river fish species were infected by H. taichui; although prevalence did not differ significantly between species of fish (p = 0.571), differences in intensity of infection were significant (p = 0.030). P. varium was recovered from four fish species and its intensity varied significantly between fish species (p ≤ 0.0001); differences in prevalence were not significant. Only one fish species, A. testudineus, was infected with H. yokogawai.
Discussion
The results of this study strongly implicate wild fish as the major source of C. sinensis infection for people in North Vietnam who consume raw or inadequately prepared fish dishes. This greater risk from wild-caught fish in North Vietnam is similar to that reported from neighboring South China, where wild fish are also a higher risk than cultured fish when comparisons have been made (Yu et al. 2003, Lin et al. 2005, Zhang et al. 2007, Sohn et al. 2009); in Korea wild fish are also the major intermediate host (Chai et al. 2005, Kim et al. 2008, Cho et al. 2010). The risk for infection with liver fluke from consumption of wild fish is increased by local people's preference for them for raw fish dishes (Table 6).
The risk of infection by heterophyid intestinal FZT from both cultured and wild fish was, not unexpectedly, high in this study. As has been reported previously in Vietnam and elsewhere, Haplorchis spp. are very common in fish and among people in communities where raw fish are widely eaten (Dung et al. 2007, Thien et al. 2007, Chi et al. 2008, Sohn et al. 2009, Cho et al. 2010, Phan et al. 2010a, b, Chai et al. 2013). The frequent occurrence of intestinal FZT in fish from all types of water bodies is probably a reflection of the low fish and definitive host specificity of these heterophyids (which are found in both mammals and fish-eating birds), and the wide distribution of the major snail host, Melanoides tuberculata. This snail species is very common in both farm ponds and in rice fields and canals (Dung et al. 2010, Madsen et al. 2015).
Although intestinal flukes are generally mild with regard to clinical effects, in heavy infections intestinal pain, diarrhea, lethargy, anorexia, and weight loss may result (Chai et al. 2007). Heavy infections with H. taichui have been associated with intestinal inflammation and irritable bowel syndrome (Sukontason et al. 2000, Watthanakulpanich et al. 2010). From a public health standpoint, the mitigation of risk from FZT, regardless of whether liver or intestinal fluke, should be an important objective.
The absence of C. sinensis in cultured fish is very likely due to biological/ecological factors affecting the distribution of the snail intermediate host. The liver flukes, C. sinensis, O. viverrini, and O. felineus utilize species of snails belonging to the genera Alocinma, Bithynia, Thiara and Parafossarulus (Bithynidae family). In Southeast Asia, these include several species of Bithynia (Petney et al. 2012, Hung et al. 2013), which have a preference for water that is slow moving (e.g., irrigation ditches, streams, marshes, rice fields, and lakes) rather than for still water farm ponds (Brockelman et al. 1986, Dung et al. 2010, Petney et al. 2012, 2013, Madsen and Hung 2014, Wang et al. 2014, Hung et al. 2015, Madsen et al. 2015). The biotic and abiotic features of these nonpond water bodies may satisfy the ecological requirements of Bithynia better than the conditions found in fish farm ponds, which are normally stagnant and highly polluted. Recent research on the physical qualities of optimal waters supporting Bithynia demonstrates a significant influence of factors such as water depth and temperature, the level of dissolved oxygen, turbidity, salinity, and especially the pH of the water (Brockelman et al. 1986, Lohachit 2004, Ngern-Klun et al. 2006).
The relative importance of various species of reservoir hosts for C. sinensis requires greater assessment. Domestic cats and dogs, because they are generally allowed to roam free between the domestic and sylvatic habitats, are undoubtedly important in this regard. However, little is known on the relative importance of wild or feral mammals as reservoir hosts. A better understanding of the relative biomass and the ecology of C. sinensis within sylvatic and synanthropic biotopes is needed because these issues are of consequence for public health programs. Prevention of C. sinensis infection in humans, because of the high risk from wild-caught fish, must be based, at least for the foreseeable future, on education of consumers on the risks from consuming raw or inadequately prepared fish and consequences of infection. Well-designed, carefully implemented and monitored education programs can be beneficial. For example, education on the risks of infection in Thailand have led to a decrease in the frequency of consumption of raw, partially cooked, or fermented fish (Jongsuksuntigul and Imsomboon 2003).
The prevention of transmission of FZT from cultured fish, however, can also be implemented by interventions in controlled aquaculture systems. Effective interventions, such as those developed in extensive field trials (Khamboonraung et al. 1997, Clausen et al. 2015), have demonstrated it is possible to reduce fish FZT infections in ponds in a prevention program based on both farm household education and practical improvements in farm management practices and pond infrastructure. A recent example of this type of an intervention program is detailed in Clausen et al. (20015), which describes a 3-year trial in fish farms in Vietnam.
Footnotes
Acknowledgments
This research is funded by Vietnam National Foundation for Science and Technology Development (NAFOSTED) under grant number 106-YS.05-2013.26. The authors wish to express their special thanks to Mrs. Tran Thi Hang, Nguyen Thi Nguyen, and Pham Thi Thanh for their significant assistance in the examination of metacercariae. Special thanks also extended to Mr. Nguyen Van Hao for valuable help in identifying fish species.
Author Disclosure Statement
No competing financial interests exist.
