Seroepidemiological study of Taenia saginata cysticercosis in Swaziland

Research in Veterinal?; Science 1993, 55, 287-291

Seroepidemiological study of Taenia saginata cysticercosis in Swaziland G. HUGHES, M. HOQUE, M. S. TEWES, S. H. WRIGHT, L. J. S. HARRISON, Centre for Tropical

Veterinary Medicine, Easter Bush, RosIin, Midlothian EH25 9RG

A cross-sectional study of Taenia saginata eystieercosis in Swaziland using a serodiagnostic ELISA for parasite antigen is described. The seroprevalence and the levels of parasite antigen were compared in the sera of cattle from different geographical localities, and from areas of high or low population density. Cattle from the Lowveldt region, which has a hot and dry climate relative to the other areas investigated, exhibited significantly higher serum antigen levels. Seroprevalence was also higher in the Lowveldt but this difference was not found to be significant. Within the Lowveldt, antigen levels were found to be slightly elevated in cattle from more highly populated areas. It is suggested that either human behaviour and/or practices in animal husbandry, or increased susceptibility of cattle to reinfection at certain times of the year, may enhance transmission in the Lowveldt since climatic conditions in this region are not conducive to transmission. BOVINE cysticercosis is caused by the metacestode stage of the zoonotic cestode, Taenia saginata. Although natural infections rarely produce clinical symptoms, the economic significance of this helminth to the livestock industry may be considerable, especially in less developed countries (WHO 1983). Downgrading and condemnation of carcases may cause substantial losses to the livestock industry and reductions in beef exports. Grindle (1978) estimated that in Botswana and Kenya, annual lost revenues may total £0.5 million and £1.0 million, respectively. Studies which identify endemic areas will allow important foci of disease transmission to be targeted in control programmes, and will also provide a better understanding of local transmission charac-

teristics. Prevalence of T saginata cysticercosis is believed to be high in Africa (Pawlowski and Schultz 1972) although detailed information is available from only a few selected countries (Harrison and Sewell 1991). The present study was designed to investigate the prevalence and distribution of bovine cysticercosis in Swaziland, and constitutes an extension of studies first undertaken by Tewes (1987). The infection was identified using a serodiagnostic enzyme-linked immunosorbent assay (ELISA) for the detection of T saginata antigen in serum (Harrison et al 1989). Despite its relatively small size (17,364 km2), Swaziland comprises several distinct geographical regions, the mountainous Highveldt, the hilly Middleveldt and Lubombo Plateau, and the savannah-like Lowveldt. Climate also varies with region, being hotter and drier at lower altitudes. Such environmental diversity provided an excellent opportunity to investigate some of the parameters which may influence the transmission of this cestode. Materials and methods

Project design Cattle were selected for the study at their local dipping centres which are distributed throughout Swaziland. Twenty cattle from each of 30 diptanks were studied (600 animals in all). Each diptank was classified according to the geographical area or veldt in which it was located, as well as to the density of the local population (Table 1). Population density and geographical location were estimated by superimposing appropriate maps over a diptank distribution map. The distribution of diptanks involved in this investigation is illustrated in Fig 1. 287

288

G. Hughes, M. Hoque, M. S. Tewes, S.H. Wright, L. J. S. Harrison

TABLE 1: Classification of Swaziland diptanks used in the present study according to geographical area and local population density Population density (people km-2) < 24 > 24

Lowveidt 9 2

Number of diptanks Middleveldt Highveldt 5 6

3 5

Diagnostic procedures Collection of cattle sera. Serum samples were obtained from the cattle at their local diptanks in Swaziland. A positive control serum was obtained from a calf experimentally infected with 5000 to 10,000 viable T saginata eggs. A sample was taken when the animal was slaughtered 12 weeks later. The numbers of live and dead metacestodes were determined following dissection of the carcase. Sera were collected from 20 naive cattle and used as negative controls, although serum from only one animal was used routinely. All sera were stored at-20°C until use.

/,

-. %

Detection of cysticercosis, T saginata antigen in the sera of the study cattle was detected using a specific ELISA developed and described by Harrison et al (1989). A mouse monoclonal antibody (mAb), designated HP10, was used. This mAb is reactive with an epitope present on a glycoprotein found both on the surface and in the secretions/excretions of T saginata cysticerci. The homologous biotin-conjugated mAb was used as the indicator system and sera being assayed were always undiluted. The optical density (OD) of the reaction product was measured on a Titertek Multiscan ELISAplate reader (Flow Laboratories) at 450 nm. Positive and negative control sera were run on each plate to check and correct for day to day and plate to plate variations. This assay has been shown to be highly specific for T saginata and T solium. Sera from cattle experimentally and naturally infected with a variety of helminths, including Echinococcus granulosus and Fasciola gigantica, and also with protozoons, did not produce false positive results (Harrison et al 1989). All animals with infection intensities of 200 cysts or more can be diagnosed (Harrison et al 1989), although at least 43 per cent of cattle with less then 30 cysts can be detected following the same procedure (Onyango-Abuje et al 1993).

Data analysis

/,

Y?

i

./

0 I 0

!

50 m i l e s I 80 k i l o m e t r e s

FIG 1:The distribution of diptanks in Swaziland which were involved in the present study. Blood was sampled from 20 cattle at each tank. 1 = Highveldt; 2 = Middleveldt; 3 = Lowveldt

To calculate a cut off point between positive and negative ELISA readings, the standard deviation of the 20 negative control sera readings was multiplied by 2.58 and added to the mean. This gave a cut off point of 0-110. Since the negative control sera values were normally distributed, the probability of a negative ELISAresult exceeding this cut off point is 0.005. The ratios of positive and negative results for cattle from the different geographical areas, and at different local population densities, were compared using Z2 analyses. A two-way analysis of variance (ANOVA) was used to compare ELISAreadings of cattle from different veldts and local population densities. Before this analysis, ELISA data were log transformed (logl0 [OD + 1]) to minimise overdispersion effects and the homogeneity of the variances was verified. Significance was located using a Tukey test.

Taenia saginata cysticercosis in Swaziland TABLE 2: Proportion of cattle exhibiting a positive Tsaginata antigen ELISA response in each geographical area of Swaziland

Positive (%) n

Lowveldt

Middleveldt

Highveldt

32-7 220

25.9 220

23.1 160

TABLE 3: Proportion of cattle exhibiting a positive Tsaginata antigen ELISA response at low and high local population densities in Swaziland Population density (people km-2) 24 or less (low) more than 24 (high) Positive (%) n

29-4 320

25.7 280

289

ent veldts and population densities are illustrated with standard error bars in Fig 2. Significantly greater amounts of antigen, as indicated by the ODs, occurred in cattle from the Lowveldt than from the Highveldt area (two-way ANOVA followed by Tukey test; F2, 594 = 4.86; P<0.01). ELISA values were not significantly influenced by local population density, although they were found to be higher in cattle from highly populated areas of the Lowveldt (two-way ANOVA; FI, 594 = 0.09; P>0.05). Also, no significant interaction was detected between veldt and population density (ANOVA; F2, 594 = 1.62; P>0.05).

Discussion Results

Cysticercosis prevalence Overall seroprevalence of T saginata cysticercosis in Swaziland, as diagnosed by the detection of circulating T saginata products using the specific EHSA, was estimated at 28 per cent (n = 600). A greater proportion of cattle showed evidence of infection in the Lowveldt (Table 2), but no significant difference between the numbers of infected and uninfected cattle in the different geographical areas was discovered (X2 = 5.19, df = 2; P>0.05). There was little difference between the prevalence of T saginam in cattle originating from areas of high and low population densities (Table 3; %2 = 1.26 [with Yates' correction ]; df = 1; P>0.05).

Level of circulating T saginata antigen Mean ELISA readings for cattle from the differ0.40

0.30

0.20

0,10

0.00

Lowveldt

Middleveldt

Highveldt

FIG 2: Mean ELISA readings (-+SE) of cattle from different geographical areas and local population densities in Swaziland. Open bars represent readings from cattle from areas of low population density (24 people km-2 or less); hatched bars represent readings from cattle from areas of high population density (more than 24 people km-2)

T saginata cysticercosis is endemic in certain areas of Africa (Pawlowski and Schultz 1972) and results from this study suggest that the parasite may be widespread in Swaziland, with an estimated 28 per cent of cattle seropositive for T saginata antigen. This is in accordance with the meat inspection findings of Mitchell (1973, 1977) although this work suggested that the prevalence of the disease was slightly lower. Only 14 per cent of cattle at the Matsapa abattoir in Swaziland were found to harbour T saginata cysticerci (Mitchell 1977). The discrepancy between these figures is likely to be at least partly attributable to the different methods of diagnosis. Meat inspection may grossly underestimate the prevalence of T saginata infection. Only 38 per cent of a sample of infected Kenyan calves were detected at meat inspection, apparently because cysticerci are not always found at the predilection sites (Walther and Koske 1980). As one would predict, detection was particularly unsuccessful at lower levels of infection. Only 31 per cent of cattle with 20 cysticerci or less were identified by standard meat inspection procedures (Walther and Koske 1980). In contrast, the ELISA specific for T saginata antigen described by Harrison et al (1989) has been shown to identify 47 per cent of infected cattle overall and 43 per cent of 30 animals with 30 cysticerci or less (Onyango-Abuje et al 1993). Furthermore, no clear association between the animals diagnosed 'positive' by the two methods can be determined (Onyango-Abuje et al 1993), probably because the ELISA detects circulating T saginata antigen from live infections only, whereas at meat inspection, carcases with dead and calcified cysticerci may be

290

G. Hughes, M. Hoque, M. S. Tewes, S.H. Wright, L. J. S. HatTison

classified 'positive'. In the context of the present study, however, it is also noteworthy that cysticercosis prevalence tends to decline with increasing host age (Grindle 1978). Cattle in Swaziland are normally slaughtered at about four years old (B. McCartan, personal communication), whereas most of those involved in this study ranged from one to three years old. Of particular interest was the distribution of T saginata cysticercosis within the country. Although overall seroprevalences were similar, significantly greater antigen levels were detected in cattle from the hotter and drier Lowveldt region of Swaziland. Since there is a positive relationship between the quantity of antigen detected in serum and live cysticercus burden (Tewes 1987), it is possible that high intensity infections are concentrated in the Lowveldt. This would perhaps not have been anticipated since T saginata ova are not believed to survive well under hot and dry conditions (Wachira et al 1991). Indeed, a humid climate was thought to be partly responsible for the relatively high prevalence of T saginata cysticercosis in cattle from the Lachinsk region of Azerbaijan (Nadzhafov 1987). It may be significant that in a seroepidemiological study of T solium cysticercosis in South African schoolchildren, mean antibody titres were found to be significantly higher in children from a 'dry' area subject to extreme temperatures, than in those from a wetter area (Shasha and Pammenter 1991). It is, nevertheless, difficult to explain these findings. There may be more prolonged contact with humans in the Lowveldt and hence increased exposure to infection from the definitive host. The level of exposure to infection may also be dictated by human behaviour and animal husbandry practices. In the Manyatta encampments in Kenya, for example, herdsmen protect their cattle from predation by enclosing them in an area surrounded by acacia scrub fences, a practice which is conducive to intensive T saginata transmission (Nelson 1972). An alternative explanation is that, in the Lowveldt, if conditions detrimental to egg survival persist for prolonged periods each year, cattle may not be continuously exposed to T saginata eggs and their immunity to reinfection may wane. Such a scenario has been referred to as the 'endemic steady state' by Gemmel (1986) and would allow reinfection and thus, presumably, result in elevated serum antigen levels, at certain times of the

year. In contrast, cattle from areas where climatic conditions are more favourable for egg survival are more likely to be continuously exposed to viable eggs. This has been termed the 'hyperendemic steady state' and could result in an acquired immunity to reinfection and lower cyst burdens, and hence lower antigen levels overall (Gemmel 1986). In the Lowveldt itself, the antigen levels of cattle from highly populated areas were higher than from more rural areas, although not significantly so. If the level of antigen detected is positively associated with cyst burden, these results suggest a greater intensity of infection in more populated areas. This phenomenon has been reported previously (Kozakiewicz 1986, Collins and Pope 1990) and may, again, be attributed to a close association with T saginata-infected humans, either directly or indirectly. Cattle which are allowed to graze in the vicinity of municipal sewage treatment works are, for example, thought to be at considerable risk of exposure to T saginata eggs, perhaps from streams contaminated with effluent, or from vectors such as birds (Collins and Pope 1990, IlsCe et al 1990). Within the constraints of this study, it is impossible to determine the actual routes of T saginata transmission between infected humans and cattle in Swaziland. Nevertheless, the discovery of greater quantities of T saginata antigen in cattle from highly populated areas of the Lowveldt implies that infection intensities are higher there. The presence of high intensity infection is, itself, suggestive of enhanced cysticercosis transmission. Consequently, the Lowveldt region of Swaziland may be an important focus for T saginata cysticercosis transmission, particularly in areas which are more densely populated. The mechanisms through which this transmission operates need to be elucidated.

Acknowledgements The authors wish to thank Dr A. Hunter and members of the University of Edinburgh veterinary expedition to Swaziland (1985/1986) who collected the serum samples as part of a serological survey of tick-borne diseases.

References COLLINS, G. H. & POPE, S. E. (1990) Cysticercosis bovis in cattle in New South Wales. Australian Veterinary Journal 67, 228-229

T a e n i a s a g i n a t a cysticercosis in S w a z i l a n d GEMMEL, M. A. (1986) A critical approach to the concepts of control and eradication of echinococcosis/hydatidosis and taeniasis/cysticercosis. In Proceedings of the Sixth International Congress of Parasitology. Ed M. J. Howell. Canberra, Australian Academy of Science. pp 465-472 GRINDLE, R. J. (1978) Economic losses resulting from bovine cysticercosis with special reference to Botswana and Kenya. Tropical Animal Health and Production I0, 127-140 HARRISON, L. J. S. & SEWELL, M. M. H. (1991) The zoonotic Taeniae of Africa. In Parasitic Helminths and Zoonoses in Africa. Eds C. N. L. MacPherson and P. S. Craig. London, Unwin Hymand. pp 54-82 HARRISON, L. J. S., JOSHUA, G. W. P., WRIGHT, S. H. & PARKHOUSE, R. M. E. (1989) Specific detection of circulating surface/secreted glycoproteins of viable cysticerci in Taenia saginata cysticemosis. Parasite Immunology 11, 351-370 ILSgiE, B., KYVSGAARD, N. C., NANSEN, P. & HENRIKSEN, S. A. (1990) Bovine cysticercosis in Denmark. A study of possible causes of infection in farms with heavily infected animals. Acta Veterinaria Scandinavica 31, 159-168 KOZAKIEWICZ, B. (1986) Ways of distribution of bovine cysticemosis in rural environment. In Proceedings of the Second International Symposium on Taeniasis/Cysticercosis and Echinococcus]Hydatidosis, 2-7 December 1985, Ceske Budejovice, Czechoslovakia. Eds J. Prokopic and J. Sterba. Parzitologicy Ustav CSAV, Ceske Budejovica, Czechoslovakia. pp 327-335 MITCHELL, J. R. (i973) Some aspects of epidemiology and the prevalence of Cysticercus boris in Africa. Bulletin of Epizootic Diseases of Africa 21, 133 MITCHELL, J. R. (1977) An abattoir survey of helminths in cattle in Swaziland. Journal of ttle South African Veterinary Association 48, 53~54 NADZHAFOV, I. G. (1987) Intensity of taeniasis transmission in foci in different climatic/geographic zones of Azerbaijan: a basis for rational

291

control measures. Meditskinskaya Pamzitologiya

i Parazitarnye

Bolezni 2, 38-41 NELSON, G. S. (1972) Human behaviour in the transmission of parasitic diseases. In Behavioural Aspects of Parasite Transmission. Eds E. U. Canniing and C. A. Wright. London, Academic Press. pp 109-122 ONYANGO-ABUJE, J. A., OPICHA, M., NGINYI, J. M., RUGUTT, M. K., GACHIHI, M.K., WRIGHT, S, H. & HARRISON, L. J. S. (1993) Recent advances in diagnosis of bovine cysticercosis: detection of circulating antigen in live cattle. East AJ?ican Agricultural and Forestry Journal (In press) PAWLOWSKI, Z. & SCHULTZ, M. G. (1972) Taeniasis and cysticercosis (Toenia saginata). In Advances in Parasitology. Volume 10. Ed B. Dawes. London, Academic Press. pp 219-239 SHASHA, W. & PAMMENTER, M. D. (1991) Sero-epidemiological studies of cysticercosis in school children from two rural areas of Transkei, South Africa. Annals of Tropical Medicine and Parasitology 85, 349-355 TEWES, M. S. (1987) Specific detection of Taenia saginata infection in naturally infected cattle. MSc thesis, University of Edinburgh WACHIRA, T. M., MACPHERSON, C. N. L. & GATHUMA, J. M. (1991) Release and survival of Echinococcus eggs in different environments in Turkana, and their possible impact on the incidence of hydatidosis in man and livestock. Journal of Helminthology 65, 55-61 WALTHER, M. & KOSKE, J. K. (1980) Taenia saginata cysticercosis: A comparison of routine meat inspection and carcase dissection results in calves. Veterinary Record 106~ 401-402 WORLD HEALTH ORGANIZATION (1983) Guidelines on Surveillance, Prevention and Control of Taeniasis/Cysticercosis. Eds M. A. Gemmel, Z. Matays, Z. Pawlowski and E. J. L. Soulsby. VPH 83/49, Geneva. p 207

Received January 13, 1993 Accepted April 19, 1993