Abstract
The study was to understand Angiostrongylus cantonensis infection status in host animals and populations in Shenzhen. In 2016–2017, 10 different ecological environments were selected, and intermediate and definitive hosts collected at the sites were examined using the enzyme digestion and dissection method to determine their infection status. Meanwhile, serum was collected from outpatients and healthy people. The enzyme-linked immunosorbent assay test was performed to detect serum IgG-specific antibodies to A. cantonensis, and serological characteristics of the populations were analyzed. A total of 300 Achatina fulica samples had an A. cantonensis infection rate of 10.67% (32/300) and an average infection intensity of 68.7 per snail, whereas 302 Pomacea canaliculata samples had an infection rate of 6.29% (19/302) and an average infection intensity of 31.4 per snail. Although both infection rate and infection intensity were lower in P. canaliculata than in A. fulica, infection intensity was significantly different (p < 0.001). Among 238 definitive-host rodents, 22 were infected with A. cantonensis. The infection rate in Rattus norvegicus was 14.68% (16/109), significantly higher than that in Rattus flavipectus (p < 0.05). The seroprevalence of A. cantonensis in the 900 outpatients and 1500 healthy people was 7.11% (64/900) and 1.87% (28/1500), respectively. Thus, the infection rate was significantly higher in outpatients than in healthy people in Shenzhen (p < 0.001). This study revealed a wide distribution and the prevalence of A. cantonensis in host animals and populations in Shenzhen, therefore, it is necessary to strengthen the current monitoring of the disease to prevent a potential outbreak.
Introduction
A
Since the first major angiostrongyliasis outbreak in Wenzhou, mainland China, in 1997, more than 30 outbreaks have been recorded (Lv et al. 2008, Yang et al. 2013, Song et al. 2016). Angiostrongyliasis was listed as an emerging foodborne parasitic disease by the World Health Organization in 2002, followed by the Ministry of Health of China in 2003. Today, the parasite is still expanding its range and causes related disease in China (Lv et al. 2008, 2009, Lu et al. 2018). As a potentially dangerous foodborne parasitic disease, angiostrongyliasis should be paid great attention and surveilled.
In China, humans acquire angiostrongyliasis primarily via eating undercooked snails harboring third-stage larvae (Lv et al. 2010). A national survey indicates that the two invasive mollusks Pomacea canaliculata and Achatina fulica play a critical role in the prevalence of angiostrongyliasis (Lv et al. 2009). And the invasions of the snails have facilitated the endemicity of A. cantonensis through facilitating the establishment of the life cycle of the parasite (Lv et al. 2009). Rodents as definitive hosts can release first-stage larvae in the feces and cause the infections of snails (Valente et al. 2018). Thus, both the intermediate and definitive hosts should be included in the surveillance system for angiostrongyliasis.
Previous studies have shown that the two main intermediate hosts (P. canaliculata and A. fulica) were widely distributed in South China, including Shenzhen (Zhang et al. 2008, Deng et al. 2012, Yang et al. 2012). Also, both infections of A. cantonensis in intermediate snails and rodents have been recorded in the investigations in Shenzhen city (Zhang et al. 2008, Deng et al. 2012, Song et al. 2016). Furthermore, there were several cases of human eosinophilic meningitis caused by A. cantonensis happened in Shenzhen (Huang et al. 2010, Gao et al. 2011). Thus, Shenzhen have been considered the natural focus of A. cantonensis and should establish a surveillance system. Also, Shenzhen is a city of migrants with a large floating population, which may increase the risk of A. cantonensis import and export. In this survey, we investigated the prevalence of A. cantonensis in host animals and humans in Shenzhen to illustrate the comprehensive epidemiological profiles and assess the current status of human angiostrongyliasis.
Materials and Methods
Ethics statement
All animals were handled in strict accordance with good animal practice according to the Animal Ethics Procedures and Guidelines of the People's Republic of China. All participants were fully explained with the research contents and signed the informed consents. The study was reviewed and approved by the Animal Welfare and Ethics Committee of National Institute of Parasitic Diseases (NIPD), Chinese Center for Disease Control and Prevention (China CDC, IPD2016-12) and the Ethics Committee of Shenzhen Center of Disease Control and Prevention (SZCDC, 20160627).
Survey of intermediate and definitive hosts
Selection of survey locations
Shenzhen (113°43′–114°38′E, 2214°3′–22°52′N) is located on the eastern coast of Pearl River Estuary in Southern China, bordered by Daya Bay and Dapeng Bay in the East, Lintin Ocean in the west, and Hong Kong in the south, with an area of 199.7 km2 and an average annual temperature of 22.4°C. It has a subtropical monsoon climate with rich precipitation and vegetation. This survey adopted a stratified random sampling method and 10 sites were selected for sample collection, which covered 9 districts of all 10 districts in Shenzhen (Fig. 1). The collection sites consisted of city parks, certain communities with great vegetation, lakes, and gardens. The samples were collected every 3 months from April 2016 to October 2017. Each collection lasted for 2 h in the morning in the selected site. Fifty to 70 adult snails were collected from each site, for a total of 602 individuals. Ten to 30 rodents were collected from each site, for a total of 238 individuals.

Study area with the sampling sites in Shenzhen City.
Sample collection
Snails were collected by hand or fishnet from where the snails were likely to inhabit, including weeds, shrubs, ponds, ditches, rivers, vegetable fields, and country sides in the collection site, with the longitude and latitude of the collection location recorded by GPS and the species and breeding environments of the snails documented. The snails were identified by the technologists in the Laboratory of Parasitology in Shenzhen CDC according to the morphological characteristics (Liu et al. 1979, 1986).
Rodents were trapped opportunistically in the 10 sites. The trapping methods were referred to sampling procedure of vector infected by pathogens–Rodent (GB/T 28940–2012). The rat cages were placed where the rodents would normally appear at dusk and received in the next morning. The number of traps used was based on minimizing trapping effort and maximizing individuals caught. The captured rodents were then numbered and taken to the Disinfection and Vector Control Division of Shenzhen CDC for species identification following the handbook (Lu 1982).
Tests for A. cantonensis in snail and rodent hosts
Test for the infection status of freshwater snails such as P. canaliculata, A. fulica, Cipangopaludina cathayensis, and Alocinma longicornis. The collected freshwater snails were removed from their shells, crushed, and artificially digested using routine procedures (incubation in a solution containing 0.2% pepsin and 0.7% hydrochloric acid at 37°C for 2 h). The final resuspended sediment was poured into a flat Petri dish, and the nodules containing stage III larvae of A. cantonensis were then isolated and counted under a microscope to calculate the rate and intensity of infection (Lv et al. 2006, Liu et al. 2007). The identification of A. cantonensis larvae were according to distinct morphological criteria described previously (Ash 1970, Wei and Wu 2014).
Test for infection status of the definitive hosts: All captured rodents were anesthetized with ether, and their blood was drained from the femoral artery, with the serum reserved for detection of IgG antibody to A. cantonensis. The thoracic cavity was dissected, and the heart and lungs were removed and shredded in a glass Petri dish filled with physiological saline to visually inspect for adult A. cantonensis under Olympus (4 × –10 ×) dissecting microscopy, in the pulmonary artery and the right heart. Adult worms were identified according to the morphological characteristics (Lindo et al. 2002, Aghazadeh et al. 2015). The etiological examination for the presence of adult nematodes was taken as the gold standard. The serological test to detect IgG antibody to A. cantonensis was also performed using the enzyme-linked immunosorbent assay (ELISA) detection kit for rodents (Combined 20171003, China), following the manufacturer's protocols.
Population survey
Selection of survey subjects
Outpatient selection: Serum was collected from outpatients who were screened for foodborne parasitic diseases (Clonorchis sinensis, Paragonimus westermani, Cysticercus, Echinococcus granulosus, Toxoplasma gondii, Trchinella spiralis, and Ascaris lumbricoides) at the parasitological department of Shenzhen Center for Disease Control and Prevention from 2016 to 2017. The outpatients and their doctors were timely informed the detection results. And the patients diagnosed with A. cantonensis infections were further invited to complete an individual case questionnaire, including their basic information, eating habits, major symptoms, and signs.
Healthy control selection: Registered and transient healthy residents of Shenzhen who had lived in Shenzhen >6 months, aged 1–69 years, had no history of parasitic diseases and did not show any symptoms of A. cantonensis infections, such as headache, vomiting, nausea, fever, and paresthesia, were considered as locals and selected as the target population. In the first sampling phase, cluster random sampling was used for randomly selecting two districts (namely, Futian District and Nanshan District). In the second phase, 10 random communities were chosen from the two selected districts. In the third phase, 1500 subjects (810 males and 690 females) were randomly chosen from the 10 selected communities for inclusion in the study. Around 2–4 mL of venous blood was collected from each subject and centrifuged at 1200 × g for 5 min. The serum was isolated and preserved at −20°C for further examination.
Population antibody test method
The specific IgG antibody to A. cantonensis was detected using a commercial ELISA kit (Combined 20171203, Shenzhen, China), following the manufacturer's protocols (Deng et al. 2007, Tong et al. 2010, Chen et al. 2016).
Statistical analysis
Data were entered into EpiData 3.1 and analyzed by IBM SPSS 21.0. Statistical analysis was performed using routine descriptive statistics and the chi-squared test (*p < 0.05, **p < 0.01, ***p < 0.001). With the etiological test as the gold standard for measuring the infection status of rodents, the sensitivity, specificity, and Youden index of the serological test for IgG antibodies to A. cantonensis were calculated.
Results
Infection status of snails
P. canaliculata and A. fulica were found in the 10 survey locations. Between them, 300 A. fulica samples had an A. cantonensis infection rate of 10.67% (32/300) and an average infection intensity of 68.7 per snail, and 302 P. canaliculata samples had an infection rate of 6.29% (19/302) and an average infection intensity of 31.4 per snail. Although both infection rate and infection intensity were lower in P. canaliculata than in A. fulica, only infection intensity was significantly different (p < 0.001); there was no significant difference in infection rate (χ 2 = 3.14, p > 0.05). No infected snails were found from the other two species (C. cathayensis and A. longicornis). There were no significant differences in the prevalence of A. cantonensis in snails among distinct sampling sites.
Infection status of rodents
A total of 238 definitive host rodents (109 Rattus norvegicus, 101 Rattus flavipectus, and 28 Suncus murinus) were captured. After dissection of the thoracic cavity and visual examination of the heart and lung, 22 rodents infected with A. cantonensis were found (total infection rate, 9.24%). The infection rate in R. norvegicus was 14.68% (16/109), which was significantly higher than in R. flavipectus (5.94%, 6/101) (p < 0.05). No infection was found in S. murinus samples. In the test for IgG antibody to A. cantonensis, 26 rodents had seroprevalence (10.92%). The sensitivity, specificity, and Youden index of the serological test were 91%, 97%, and 0.88, respectively. The different results of infection status between the etiological and serological tests are shown in Table 1.
Different Results of Angiostrongylus cantonensis Infection Status by Etiological and Serological Tests
“+” indicates positive results; “−” indicates negative results.
Population antibody test results
Total seroprevalence of A. cantonensis for different populations
The seroprevalence of A. cantonensis in the 900 outpatients and 1500 healthy people was 7.11% (64/900) and 1.87% (28/1500), respectively. Thus, the infection rate was significantly higher in outpatients than in healthy people in Shenzhen (p < 0.001). According to the questionnaire, among the positive outpatients, 59 of them (92.19%) had a history of eating raw food in the last half year, 49 of them (76.56%) had elevated eosinophil level, and 29 of them (45.31%) experienced neurological symptoms such as headache, nausea, vomiting, and so on.
The health education about A. cantonensis transmission and prevention was also given to the patients and their families.
Seroprevalence for different sexes
The infection rate was significantly higher in male outpatients (8.97%) than in female outpatients (5.09%) (p < 0.05). For healthy people, the infection rate was higher in males (2.35%) than in females (1.30%); however, the difference was not significant (p > 0.05). Results according to sex distribution are presented in Table 2.
Sex Distribution of A. cantonensis Infection in Different Groups
χ 2 = 5.124, p < 0.05.
χ 2 = 2.206, p > 0.05.
χ 2 = 6.549, p < 0.05.
Infection status of different age groups
The subjects were divided into three age groups: 0–20, 21–50, and ≥51 years. The chi-squared test showed that the total infection rate and the healthy-population infection rate in the age group 21–50 years were significantly higher than in the other two age groups. The age distribution results are shown in Table 3.
Age Distribution of A. cantonensis Infection in Different Groups
χ 2 = 15.805, p < 0.001.
χ 2 = 6.254, p < 0.05.
χ 2 = 18.809, p < 0.001.
Discussion
Shenzhen is a city located on the southern coast of Guangdong Province. Because of its warm, humid climate and rich plantation, Shenzhen is a suitable breeding area for rodents and mollusks such as snails, which are the definitive and intermediate hosts of A. cantonensis (Thiengo et al. 2013). Hence, a “rat–snail–rat” life cycle of this nematode is easily formed, which constitutes the natural epidemic source of the disease (Zhang et al. 2009, Moreira et al. 2013, Eamsobhana 2014). In recent years, a few clinical cases of A. cantonensis infection in Shenzhen have been reported. Previous investigations on the infection rate in the intermediate and definitive hosts of A. cantonensis indicated that natural epidemic sources possibly exist in Shenzhen (Zhang et al. 2008, Huang et al. 2010). These findings support the need for further study on the A. cantonensis infection status in host animals and populations in Shenzhen to help with prevention and risk assessment of the disease.
A. cantonensis has no specific parasite host; consequently, its intermediate mollusk hosts may comprise a variety of species and may be widely distributed. Thirteen species of terrestrial and freshwater mollusks that can be infected naturally have been reported domestically, and according to their natural infection rate and the way in which they are consumed by humans, the snail species that are of epidemiological significance include A. fulica, P. canaliculata, Bellamya aeruginosa, and C. cathayensis (Hollingsworth et al. 2013, Martinalonso et al. 2015). However, variations exist in the reported host species and infection rate of A. cantonensis in different areas of China. In this present study in Shenzhen, among the four investigated freshwater snail species, A. cantonensis infection was found only in the terrestrial A. fulica and amphibian P. canaliculata. The infection rate in A. fulica was 10.67%, which is lower than the average in Hainan (22.66%) (Hu et al. 2011) and Shenzhen back in 2008 (15.56%) (Zhang et al. 2008). Meanwhile, the infection rate in P. canaliculata was 6.29%, which is lower than the average in Hainan Province (12.36%) (Hu et al. 2011) and Shenzhen back in 2008 (10.3%) (Zhang et al. 2008). These variations in infection rate are likely due to differences in the selection of survey areas, breeding environments, and seasons for sampling. This study also found that under the same conditions, the infection rate in A. fulica was higher than in P. canaliculata, which is consistent with the survey results from Guangdong Province (Chen et al. 2011), Guangzhou (Yang et al. 2012), and mainland China (Lv et al. 2009). This is possibly because the chance of infection for P. canaliculata is relatively low; as an amphibian snail, it is infected only when consuming excrement of infected rodents that is discharged into the water. On the contrary, as A. fulica lives in the same environment with rodents, its chance of infection is higher.
A. cantonensis adults parasitize the circulatory system of rodents, with the majority in the right atrium and pulmonary arteries. In this study, the serological test used to examine the IgG antibody to A. cantonensis was assessed by comparing the results to those of the diagnostic test, considered the gold standard, which was conducted by dissection of the rodent's cardiopulmonary tissue to find A. cantonensis adults. On comparison, the results of the serological and etiological tests agreed well, providing future reference for the investigation of A. cantonensis infection in rodents. Nevertheless, compared with the gold standard, the serological test produced a few false positive and negative results. Because the serodiagnosis is an indirect method to detect the presence of A. cantonensis (Chen 1986), the false positive results mostly were caused by cross reactions of crude antigen extracts to antibodies to other rodent parasites (Eamsobhana and Yong 2009). The presence of detectable antibodies to A. cantonensis in rat serum coincided with appearance of larvae in feces, which meant that false negative results would have occurred in specimens from early infective stage (Chen 1986, Wilkins et al. 2013). However, due to the many advantages of ELISA, such as its reliable, convenient technical features and potential for automation, it is still an efficient screening test for A. cantonensis in hosts. The results also showed a total infection rate of 9.24% in rodents, which is lower than that found in Brazil, Canary Islands, Queensland, and the Philippines (Martin-Alonso et al. 2011, Simões et al. 2014, Aghazadeh et al. 2015, Tujan et al. 2016), but higher than that found in Phitsanulok province, Thailand (Vitta et al. 2011). Among four common rodent species (R. norvegicus, R. flavipectus, S. murinus, and Mus musculus) captured in this survey, infection was found only in R. norvegicus, consistent with the survey findings from other areas in Guangdong Province (Qiu-An et al. 2017). The infection rate in R. flavipectus, S. murinus, and M. musculus was low possibly because these species mainly consume plant foods; however, as the sample size of this survey was small, further investigation is required to confirm the results.
Because of the clinically confirmed A. cantonensis infection cases in Shenzhen, this epidemiological survey was conducted in both outpatients of the parasitological department and healthy people. The results showed that the seroprevalence of A. cantonensis in the outpatients was notably higher than in the healthy population. These outpatients were voluntary visitors of the parasitological department, and they often had a history of eating raw food and presented with abnormal clinical manifestations or symptoms of elevated eosinophils. Thus, their 7.11% seroprevalence rate represents, to a certain extent, the infection rate in the high-risk population in Shenzhen. The healthy volunteers were randomly selected, so their 1.84% seroprevalence rate represents the average infection rate in the local residents. This rate indicates the presence of hidden infectants of A. cantonensis in the healthy population in Shenzhen, which can be related to the population's frequent contact with the host animals. The survey results also presented different epidemiological characteristics between the outpatients and healthy population. Sex analysis revealed that although differences between the outpatients and healthy population were not significant, the infection rate in males was significantly higher than in females. A possible explanation is that males have more social opportunities and are more encouraged to eat raw food. In the age analysis, the infection rate was slightly higher in people aged 21–50 years than in people aged <20 or >50 years, possibly because people aged 21–50 years are mostly young and spend more time socializing and dining outside their homes, leading to a higher chance of infection. Furthermore, these results suggest that infection with foodborne diseases such as A. cantonensis is closely related to poor hygienic practices.
The analyses and results of this survey can contribute to the fundamental understanding of A. cantonensis infection status in host animals and populations in Shenzhen. The survey indicates that there remains a potential risk of an A. cantonensis outbreak in Shenzhen; thus, monitoring, prevention, and control measures are necessary. Other factors that are closely related to the population density and infection of snails and rodents, including geographical and environmental factors (such as temperature, rainfall, plants, and soil) and socioeconomic and health conditions, are recommended in future studies. In addition to be infected by eating undercooked snails, people can also get angiostrongyliasis by consuming paratenic hosts such as freshwater shrimp, crabs, and frogs, by inadvertent ingestion via contact with the debris produced in preparing the snails, and even by consumption of vegetables or other produce contaminated by infected mucus (Cowie 2013). So, this study can also be improved by increasing the number of samples and species of the host animals such as taking the paratenic hosts into the surveillance system. Also, more-detailed investigations and analyses should be performed on the antibody positivity of the population to fully understand all possible infection pathways in local residents, which helps for the completion of the epidemiological profile of A. cantonensis in Shenzhen and the construction of an integrated strategy to control it.
Footnotes
Acknowledgments
This research was funded by the National Key Research and Development Program of China (Nos. 2016YFC1202000, 2016YFC1202001); the authors thank the following people: Dr. Liu Yi and Dr. Wang Tieqiang, who work in Guangming CDC, for helping catching snails and rodents; Dr. Zheng Wei, who works in Baoan CDC, for helping catching snails and rodents.
Author Disclosure Statement
No competing financial interests exist.
