D.T. de Waal1, R.C. Carter2, O. Matthee1 and R.J. Bagnall2
The early history of trypanosomosis in cattle in KwaZulu-Natal, South Africa, dates back to the latter part of the previous century, when it caused serious epidemics in northern KwaZulu. The advent of organo-chlorine insecticides provided suitable means to remove tsetse, and between 1945 and 1952 G. pallidipes was eradicated from northern KwaZulu-Natal by large-scale aerial spraying operations. Since 1954 only occasional and isolated cases of trypanosomosis have been reported. However, in 1990 bovine trypanosomosis re-emerged in this area. Using thick and thin blood smears, 10-15% of these animals in three districts, Ingwavuma, Ubombo and Hlabisa, were found to be infected with either T. congolense and/or T. vivax. By the end of 1993 more than 115,000 cattle were treated, costing nearly US$ 115,000, in treatment alone. Subsequently a more detailed survey was launched using the antigen detection ELISA to estimate the extent of the problem in the area.
Each of the four districts at risk (Nongoma, Ingwavuma, Ubombo and Hlabisa) was divided into sections (n=137) with one or more assigned to a diptank (n=106). Twenty cattle, selected systematically as they went through the diptank, were sampled (serum, haematocrit for packed cell volume (PCV) and buffy coat smears) per section. A total of 3166 samples were collected. Buffy coat smear examination revealed a Trypanosoma prevalence of between 2.4–9.8%. The Ag-ELISA results also showed a high prevalence of T. brucei. However, comparison of results of smear examination with Ag-ELISA showed overall rates of infection by the Ag-ELISA to be much higher, and these did not appear to reflect the clinical situation. It was therefore concluded that the cut-off optical density (OD) value of 0.05 for positive cases was not very sensitive and alternative criteria to determine positive cases must be found. Presently the Ag-ELISA cannot be used as a definitive tool to guide intervention studies in KwaZulu-Natal.
The early history of trypanosomosis in cattle in KwaZulu-Natal dates back to the previous century, when the then governor of Natal asked Sir David Bruce to investigate the notorious 'tsetse-fly disease' (=nakane, a Zulu word which has developed into the colloquial name nagana) in northern KwaZulu-Natal (then Zululand). He found both T. brucei and T. congolense in domestic animals in Zululand, the latter being responsible for the vast majority of cases of nagana in cattle while no less than 24% of the game animals examined harboured trypanosomes (Bruce 1895, 1897). Curson (1924, cited by Henning 1956) also found another pathogenic trypanosome, T. vivax, in domestic animals in Zululand. Bruce (1895) was the first to prove that nagana was spread by tsetse-fly Glossina spp and Jowett (1911) demonstrated that pathogenic trypanosomes can also be transmitted mechanically by other biting flies. Glossina spp was wide-spread in the western and eastern Transvaal (G. m. morsitans) and north-eastern KwaZulu-Natal (G. pallidipes, G. austeni and G. brevipalpis) in the early 1800's (Henning 1956). They disappeared from large parts of these areas with the rinderpest pandemic of 1896–1897.
Soon after the annexation of Natal by the British in 1887 game protection laws were enforced, as a result of which the indigenous antelopes multiplied rapidly. Parts previously devoid of game (due to excessive hunting) now swarmed with antelopes of all species. With this increase and extension of game, tsetse-fly and nagana spread. By 1894 the disease assumed such alarming proportions that its investigation was undertaken by Bruce (Bruce 1895). The rinderpest pandemic destroyed almost entire populations of wild animals. However, in parts of Zululand, including the present-day game reserves, pockets of wild animals and tsetse flies survived and by 1905 nagana was once again threatening livestock. Glossina pallidipes dispersed into farming areas causing serious epidemics (Henning 1956). The advent of organochlorine insecticides provided suitable means to remove tsetse flies, and between 1945 and 1952 G. pallidipes was eradicated from Zululand by large-scale aerial spraying operations (Henning 1956). From 1952–1987 only isolated cases of nagana were reported from Zululand. Most of the cases occurred around the St Lucia lake system where G. austeni and G. brevipalpis still occurred. At that time it was assumed that since both these Glossina spp favour dense, woody coverts, riverine thickets, or forests often associated with high rainfall or subsoil moisture, their distribution would be much more confined (than savanna tsetse flies) and will therefore never become a serious economic problem (Henning 1956). However, human and animal populations have steadily increased, resulting in closer contact between cattle and the shaded thicket and riverine habitats of the fly and in 1990 bovine trypanosomosis re-emerged in this area (Bosman 1990).
During 1990 an increase in 'redwater' was reported by the local community in areas surrounding the game reserves. On further investigation nagana was diagnosed at a diptank east of the Hluhluwe Game Reserve. A survey was started using thick and thin blood smears from 20 emaciated cattle selected at diptanks in the area. This survey revealed that 5–15% of cattle in three districts (Ingwavuma, Ubombo and Hlabisa) were infected with either T. congolense and/or T. vivax (Figure 1). As soon as a positive diagnosis was made the whole diptank area was considered infected. The dip compound at the diptank was changed from amitraz (Triatix) to cyhalothrin (Grenade) and animals reported sick by the owner were treated, at first with diminazene (Berenil) and later with homidium bromide (Ethidium). Later blanket treatment with homidium bromide was applied. By the end of 1993 more than 115,000 cattle had been treated, costing nearly US$ 115,000 in treatment alone. The costs increase substantially when other costs like drug administration (equipment, travel, personnel, time, etc), change in dipping compound, etc, are added. Accurate data on production losses and actual mortality are not available but nagana was a major cause of death of animals which were already under severe stress caused by the drought conditions which prevailed in the area at that time. The treatments and change in dipping compound were very effective in rapidly reducing clinical cases. However, pyrethroid resistance developed in ticks and from April 1993 the diptanks reverted back to amitraz and every fifth animal was also treated with deltamethrin (Deca-spot) in order to maintain some level of fly control.
Figure 1. Prevalence of nagana in KwaZulu-Natal as determined by Ag-ELISA and BCT.
The nagana problem seems to be confined to some 16,000 km2 of northern KwaZulu-Natal comprising 426,000 people, 300,000 head of cattle and 130,000 small ruminants. Most of the area currently infested with nagana is used for traditional mixed farming and the presence of tsetse and nagana seriously handicaps development. The Food and Agriculture Organization of the United Nations (FAO) has also recently become involved by providing limited assistance under a technical cooperation programme for the long-term control/eradication of trypanosomosis in KwaZulu-Natal.
Clinically the disease is present in its chronic form suggesting low fly challenge. Even though trypanosome infections were detected, failure to adopt formal random techniques and the low sensitivity of the method employed prevented accurate estimation of the prevalence of trypanosomosis in the region. ILRAD was consulted in 1993 to assist in designing an unbiased cross-sectional study of cattle trypanosomosis in northern KwaZulu-Natal, using a more sensitive and specific diagnostic technique-the antigen detection ELISA (Ag-ELISA)-to estimate the extent of the problem in the area and to help with decisions on control strategies.
Recently specific monoclonal antibodies (MAb) were derived against non-variable trypanosomal antigens from in vitro propagated procyclic forms of T. congolense, T. vivax and T. b. brucei (Nantulya et al 1987). These antibodies were shown to detect species-specific circulating antigens in sera of infected cattle. The use of these specific MAbs in an antigen-trapping ELISA (Ag-ELISA) resulted in a sensitive and specific diagnostic test enabling many latent infections to be detected (Nantulya and Lindqvist 1989). Circulating antigens can be detected as early as 8–14 days post-infection and persist probably as long as the infection persist in the animal (Masake et al 1995), but are cleared from the circulation within two weeks, following drug cure (Nantulya and Lindqvist 1989). False negative results have been reported early on in infections and may occur due to low levels of circulating antigens (Nantulya and Lindqvist 1989; Masake et al 1995).
Table 1. Estimated prevalence of Trypanosoma spp infection in northern KwaZulu-Natal using Ag-ELISA and BCT.
| District | No. of sections |
No. of samples |
Ag-ELISA |
BCT Trypanosoma spp (%) |
||
T. brucei |
T. congolense |
T. vivax |
||||
| Ingwavuma | 38 | 886 | 0.389 |
0.449 |
0.336 |
3.4 |
| Ubombo | 43 | 1008 | 0.352 |
0.455 |
0.332 |
9.8 |
| Hlabisa | 50 | 1138 | 0.299 |
0.276 |
0.296 |
2.4 |
| Nongoma | 6 | 134 | 0.328 |
0.366 |
0.299 |
3.7 |
†95% confidence interval for the true prevalence.
Each of the four districts at risk (Ingwavuma, Ubombo, Hlabisa and Nongoma) were divided into sections with one or more sections assigned to a diptank. There were a total of 137 sections and 106 diptanks (Table 1). A multistage sampling design was planned, with the primary unit being the section and the secondary unit, cattle. Twenty cattle were sampled per section and selected systematically as they went through the diptank.
Blood samples were collected in 10 ml vacutainer tubes, kept at ambient temperatures for a maximum of 24 hours (except in very hot [>30] areas). After clotting, the clot was removed, the sample centrifuged and the serum decanted into 2 ml plastic tubes and stored at -20°C before being tested with the Ag-ELISA or Latex agglutination test (only on some of the samples).
Blood was also collected in capillary tubes (containing heparin as anticoagulant) and centrifuged to determine the PCV and then used to prepare buffy coat smears and examined by dark ground/phase contrast microscopy (also referred to as the buffy coat technique (BCT) (Murray et al 1977). A questionnaire was completed for each section sampled.
The Ag-ELISA was performed on all serum samples using monoclonal antibodies capable of capturing species-specific non-variable antigens of T. congolense, T. vivax and T. brucei, as described by Nantulya and co-workers (Nantulya et al 1987; Nantulya and Lindqvist 1989).
A total of 3166 cattle, out of a population of 253,828 (= 1.2%), were sampled. The average PCV for the samples with a positive BCT result (n=139) was 32.1%, while the average PCV for the animals that tested negative (n=300) was 30.3%. It would therefore appear that the presence of Trypanosoma is unrelated to the PCV values in this study.
The results of the BCT and Ag-ELISA are summarised in Table 1. Figure 1 summarises the geographical prevalence of Trypanosoma spp as determined with the Ag-ELISA.
The Ag-ELISA revealed that T. brucei is also present in the animals and it would appear that T. congolense is more prevalent than T. brucei or T. vivax (although not statistically significant) at most diptanks.
The results of the Ag-ELISA and Latex agglutination tests are compared in Figure 2. It is clear that there are discrepancies between the two tests, the Latex test only identifying ca 10% (for T. congolense and T. vivax) of Ag-ELISA-positive animals (52% in the case of T. brucei), while only 4–8% (T. congolense and T. vivax, respectively) of Ag-ELISA-negative animals tested positive with the Latex test (46% in the case of T. brucei).
The diagnosis of trypanosomosis is notoriously difficult. Clinical signs are non-specific and parasitaemia is usually low and intermittent. Although the examination of blood smears alone is a relatively insensitive way to detect infection it is simple and of great practical significance as it can also be used to diagnose anaplasmosis, babesiosis and theileriosis. The most sensitive direct method of detecting T. congolense and T. vivax infections is by examination of wet preparations of the haematocrit buffy coat, under phase contrast illumination (Murray et al 1977).
Alternative laboratory approaches to diagnose trypanosomosis in animal populations are to demonstrate specific antibodies in the serum of infected animals for which several techniques have been described (reviewed by Nantulya 1990). Since antibodies persist for a long time, even after successful elimination of the infection in the animal (self cure or chemotherapy), these antibody detection techniques will only provide a presumptive diagnosis as it is not possible to distinguish between current and latent infections (Luckins et al 1979; Nantulya et al 1984, 1986). Another major problem of using serological tests to detect trypanosome antibodies is the lack of specificity of the antigens used and cross reactions occurring between Trypanosoma spp.
Although it is reported that the Ag-ELISA is four to five times more sensitive than conventional techniques (thick and thin blood smear and BCT) (Masake and Nantulya 1991; Masake et al 1995) the estimated prevalence of trypanosomosis in KwaZulu-Natal appeared to be so much higher than expected that it appeared to in no way reflect the clinical position on the ground. Recent tsetse fly surveys in the area only collected G. austeni and G. brevipalpis and have also shown that the distribution of these fly species are much wider than generally accepted (i.e. distribution not confined only to dense, woody coverts, riverine thickets or forests) (E.M. Nevill, unpublished observations). The cut-off set for the Ag-ELISA (OD>0.05) seems to compromise the sensitivity and specificity (especially for T. brucei) of the test. Because population seroepidemiology may vary over time in different geographical regions, the cut-off should be adjusted based on local studies. There also seems to be a significant difference in the results of the Ag-ELISA and Latex agglutination tests. These apparent discrepancies result in severe problems in the interpretation and eventual control strategies. Many trypanosome-infected animals appear to be clinically normal and even if sub-clinical disease is present it may not be economical for treatment to be given.
Figure 2. Comparison of the Latex and ag-ELISA tests for Trypanosoma spp.
Other shortcomings of the Ag-ELISA test are:
On a national level diagnosis is essential for disease surveillance and monitoring and provides essential information for the formulation and execution of disease control programmes. Any surveillance programme should have a high probability of detecting a disease if it is present in a country or region. The techniques used in such a surveillance programme must be sensitive (detect true health), specific (minimum of false identification of a health status), be completed within reasonable time and be simple to perform. Accurate data is essential to place the tsetse fly and nagana problem in the broader context of socio-economic and agricultural development, natural resource management and land husbandry.
Bosman P.P. 1990. Trypanosomiasis (T. congolense) in South Africa. Office International de Epizooties: Disease Information 3:79–80.
Bruce D. 1895. In: Preliminary Report on the Tsetse Fly Disease or Nagana in Zululand. Ubombo, Zululand, December 1895. Durban: Bennett and David, South Africa.
Bruce D. 1897. In: Further Report on the Tesetse-fly Disease of Nagana in Zululand. Ubombo, Zululand, May 1896, London.
Henning M.W. 1956. In: Animal Diseases in South Africa. 3rd edition. Central News Agency Ltd., Pretoria, South Africa.
Jowett W. 1911. Further notes on cattle trypanosomiasis of Portuguese East Africa. Journal of Comparative Pathology and Therapeutics 24: 21–40.
Luckins A.G., Boid R., Rae P., Mahmoud M.M., Elmalik K.H. and Gary A.R. 1979. Serodiagnosis of infection with Trypanosoma evansi in camels in the Sudan. Tropical Animal Health and Production 11:1–12.
Masake R.A. and Nantulya V.M. 1991. Sensitivity of an antigen detection enzyme immunoassay for diagnosis of Trypanosoma congolense infections in goats and cattle. Journal of Parasitology 77:231–236.
Masake R.A., Moloo S.K., Nantulya V.M., Minja S.H., Makau J.M. and Njuguna J.T. 1995. Comparative sensitivity of antigen-detection enzyme immunosorbent assay and the microhaematocrit centrifugation technique in the diagnosis of Trypanosoma brucei infections in cattle. Veterinary Parasitology 56:37–46.
Murray M., Murray P.K. and McIntyre W.I.M. 1977. An improved parasitological technique for the diagnosis of African trypanosomiasis. Transactions of the Royal Society of Tropical Medicine and Hygiene 71:325–326.
Nantulya V.M. 1990. Trypanosomiasis in domestic animals: the problems of diagnosis. Revue Scientifique et Technique Office International des Epizooties 9:357–367.
Nantulya V.M. and Lindqvist K.J. 1989. Antigen-detection enzyme immunoassays for the diagnosis of Trypanosoma vivax, T. congolense and T. brucei infections in cattle. Tropical Medical Parasitology 40:267–272.
Nantulya V.M., Musoke A.J., Ruraringwa F.R. and Moloo S.K. 1984. Resistance of cattle to tsetse-transmitted challenge with Trypanosoma brucei or Trypanosoma congolense after spontaneous recovery from syringe-passaged infections. Infection and Immunity 43:735–738.
Nantulya V.M., Musoke A.J. and Moloo S.K. 1986. Apparent exhaustion of the variable antigen repertoires of Trypanosoma vivax in infected cattle. Infection and Immunity 54:444–447.
Nantulya V.M., Musoke A.J., Rurangirwa F.R., Saigar N. and Minja S.H. 1987. Monoclonal antibodies that distinguish Trypanosoma congolense, T. vivax and T. brucei. Parasite Immunology 9:421–431.
