Central point sampling from cattle in livestock markets in areas of human sleeping sickness

Abstract

We present the results of a study to determine the value of central point sampling in cattle markets as a means of estimating the trypanosomiasis (T. brucei s.l.) prevalence in the surrounding landscape in Uganda. We find that in the epidemic area Studied, central point sampling is a good predictor of prevalence in surrounding villages, but not in endemic areas. We also find that animals infected with trypanosomiasis are more likely to be brought for sale in livestock markets in endemic areas; we discuss these results in relation to the prevention of the spread of sleeping sickness. (c) 2005 Elsevier B.V. All rights reserved.

Full text

Page 1

Acta Tropica 97 (2006) 229–232
Central point sampling from cattle in livestock markets
in areas of human sleeping sickness
E.M. F` evrea,∗, A. Tilleya, K. Picozzia, J. Fyfea, I. Andersona,
J.W. Magonab, D.J. Shawa, M.C. Eislera, S.C. Welburna
aCentre for Tropical Veterinary Medicine, University of Edinburgh, Easter Bush, Roslin, Midlothian EH25 9RG, UK
bLivestock Research Institute, PO Box 96, Tororo, Uganda
Received 3 August 2005; received in revised form 16 November 2005; accepted 29 November 2005
Abstract
We present the results of a study to determine the value of central point sampling in cattle markets as a means of estimating
the trypanosomiasis (T. brucei s.l.) prevalence in the surrounding landscape in Uganda. We find that in the epidemic area studied,
central point sampling is a good predictor of prevalence in surrounding villages, but not in endemic areas. We also find that animals
infected with trypanosomiasis are more likely to be brought for sale in livestock markets in endemic areas; we discuss these results
in relation to the prevention of the spread of sleeping sickness.
© 2005 Elsevier B.V. All rights reserved.
Keywords: Sleeping sickness; Trypanosomiasis; Cattle; Market; Sampling
1. Introduction
Sleeping sickness, caused by the zoonotic try-
panosome Trypanosoma brucei rhodesiense, affects
large areas of eastern Africa. In southeast and eastern
Uganda, the disease occurs in discrete foci, although in
recent years there has been an expansion into previously
unaffected districts. In the virtual absence of wildlife in
theseareasofUganda,domesticlivestock—inparticular
cattle—havebecomeanimportantreservoirforthepara-
site, and a significant source of bloodmeals for the tsetse
host (Clausen et al., 1998; Okiria et al., 2002). This has
beenrecognizedbothinthepersistenceofsleepingsick-
ness during endemic periods (Waiswa et al., 2003), and
∗Corresponding author. Tel.: +44 131 650 8850;
fax: +44 131 651 3903.
E-mail address: Eric.Fevre@ed.ac.uk (E.M. F` evre).
intheepidemiczonesofsouth-eastern(Hideetal.,1996)
and more recently eastern Uganda (F` evre et al., 2001),
where18%ofcattlewerefoundtobeharbouringhuman-
infective parasites (Welburn et al., 2001). The current
epidemic in eastern Uganda is spreading northwards,
with cattle movements a likely driving factor (F` evre et
al., 2005).
In areas affected by T.b. rhodesiense sleeping sick-
ness, particularly during epidemics, a rapid response to
reducing the prevalence in the reservoir is essential to
prevent widespread transmission. However, it is often
onlythepassivesurveillanceofsleepingsicknessattreat-
ment centres that provides the first indication of disease
resurgence or spread, and reporting to health centres
is beset with problems (Odiit et al., 2004). Outbreaks
may occur in areas covering many hundreds of square
kilometres, so that assessment of the extent of infection
in the animal reservoir can be problematic. Strategies
designed to assess the degree of the problem without
0001-706X/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.actatropica.2005.11.005

Page 2

230
E.M. F` evre et al. / Acta Tropica 97 (2006) 229–232
requiring sampling from many disparate geographical
locationswouldthereforegreatlyimproveefficiencyand
reduce costs. In contrast to other infectious diseases of
livestocksuchasfoot-and-mouthdisease,animalmove-
ment restrictions are not put in place to prevent sleeping
sickness and livestock are brought to market regularly
from across risk areas.
Using PCR based methods for identification of ani-
mals infected with T. brucei s.l., we investigated the
value of central point sampling strategies; we tested
the hypothesis that livestock markets could be used as
central point sampling locations representative of the
surrounding area. This study was carried out prior to a
large-scale sleeping sickness control programme which
targeted livestock in this region.
2. Methods
2.1. Study design
Surveys were conducted in three districts in Uganda:
Soroti district, where sleeping sickness is currently epi-
demic, and Tororo and Kamuli districts, both endemic
areas.InUganda,eachdistrictoperatesoneortwomajor
livestock markets in which animals from across the dis-
trict are traded, and a number of smaller ones dealing
mainly with inter-village trade. We sampled each of the
major markets in the districts studied. Cattle were also
sampled in four to six typical and arbitrarily chosen vil-
lagesfromthesurroundingareaineachdistrict.Thiswas
a cross-sectional study, conducted in the first quarter of
2003; in both the markets and villages, we sampled all
the cattle present on the sampling day, with a minimum
of 50 animals in each location, sufficient to detect an
effect in a village/market with an estimated maximum
cattle population of 100 animals and an expected preva-
lence of up to 4% for T. brucei s.l. (Waiswa et al., 2003;
Magona et al., 2005), with a 95% confidence level.
2.2. Sample analysis
Bovine ear vein blood was collected into heparinised
capillary tubes and applied directly onto Whatman FTA
cards (Whatman Inc., NJ, USA). These were analysed
for the presence of T. brucei s.l. using species specific
primers, as described by Picozzi et al. (2002). Differ-
encesinthenumberofT.bruceis.l.positiveandnegative
animalsamongmarketsandvillagesintheepidemicand
endemic districts were assessed using generalised linear
models with binomial errors in S+ (Insightful, Seattle,
USA).
3. Results
Fig. 1 shows the PCR-based prevalence for T. bru-
cei s.l. in cattle from markets and villages within the
three districts studied. The epidemic area of Soroti
had a significantly higher prevalence of T. brucei
s.l. than the endemic areas of Kamuli and Tororo
Fig.1. BoxplotsoftheprevalenceofT.bruceis.l.inmarketsandvillagesinKamuli(twomarkets;fourvillages),Tororo(twomarkets;fourvillages)
andSorotidistricts(twomarkets;fivevillages)ofUganda.Sorotiisdenotedasanepidemiczone,KamuliandTororoasendemic.Whiskersrepresent
range of data, while bars and boxes the 25–75% percentile.

Page 3

E.M. F` evre et al. / Acta Tropica 97 (2006) 229–232
231
(χ2
action between epidemic/endemic status and sampling
location (χ2
1=10.78; p=0.001). There was no signif-
icant difference between T. brucei s.l. prevalence in
markets and villages within Soroti district (χ2
p=0.967); this study was undertaken prior to the imple-
mentation of control activities in that region. Within this
epidemic area, prevalence determined for markets pro-
vided reasonable estimates of prevalence in surrounding
villages. In contrast, in the endemic areas, markets did
not accurately reflect village T. brucei s.l. prevalence;
marketshadasignificantlyhigherproportionofinfected
animals (χ2
1=29.20; p<0.001).
1=18.37; p<0.001); however, there was an inter-
1=0.001;
4. Discussion
Thedatapresentedsuggestthatpriortotheimplemen-
tation of sleeping sickness control in the epidemic area
studied (Soroti district), T. brucei s.l. prevalences in cat-
tle at markets could be used to estimate the prevalence
of the parasite in the surrounding area; Soroti was the
onlyT.b.rhodesiensesleepingsicknessepidemicareain
Uganda at the time of this study. In the endemic zones
studied, market prevalences were not representative of
those in the surrounding area; however, for assessment
of the extent of T. brucei s.l. prevalence in livestock dur-
ingsleepingsicknessepidemics,marketsmaybeusedas
a convenient central point. This has benefits in terms of
reducingcostsandimprovingtheefficiencyofscreening
programmes,andmightbeappliedinemergingepidemic
areas to which sleeping sickness has recently spread
(F` evre et al., 2005). Sampling of cattle would not, how-
ever, replace the need to screen humans in an effort to
identifyandtreatpatientsinfectedwithT.b.rhodesiense.
The treatment of cattle at markets in epidemic areas
would also prevent the spread of T.b. rhodesiense in the
animal reservoir, though this should be accompanied by
monitoring for the possible development of drug resis-
tance (Barrett, 2001). Our results also show that while
themarketasacentralpointapproachisinappropriatein
endemic areas, there is a need for researching and test-
ing cost-effective strategies for efficient sampling where
cattle and human trypanosomiasis are well established.
Improving our understanding of the levels of infection
in reservoir populations where sleeping sickness per-
sists should have important impacts on the efficiency
of reservoir-targeted interventions.
At present, microscopy is the best diagnostic tool
available for the identification of trypanosomes in the
field, and the detection of the SRA gene—the marker
for T.b. rhodesiense (Welburn et al., 2001; Gibson et al.,
2002)isnotfield-adapted.WhileT.bruceis.l.prevalence
continues to be used as a proxy for T.b. rhodesiense,
Coleman and Welburn (2004) have shown that the ratio
of T. b. brucei: T.b. rhodesiense in a given population is
fairly constant, at 3:1. The level of T. brucei s.l. preva-
lence in a reservoir population in a sleeping sickness
affected area is thus indicative of the prevalence of T.b.
rhodesiense in that same reservoir population.
T. brucei s.l. does not result in severe clinical disease
in local (primarily zebu) breeds of cattle. However, it is
likely, that animals carrying T. brucei s.l. infections are
also infected with other, more pathogenic species of try-
panosome such as T. vivax and T. congolense (Magona
et al., 2003), and the overt clinical signs associated with
these infections may encourage farmers to attempt to
sell animals, accounting for infection rates at markets in
endemic areas higher than those in the surrounding vil-
lages. In addition, if animals infected with pathogenic
trypanosomes are brought to market on the basis of poor
performance (e.g. loss of body condition), such animals
may not be rapidly sold to traders; farmers and traders
have a tendency of returning to the market with the
unsold animals on subsequent market days, leading to
a build up of infected animals in the market—which
may greatly contribute to the high prevalence recorded.
The preferential sale of potentially infected animals in
areas serving as a source of livestock for country-wide
restockingprogrammeshasimplicationsfordiseasecon-
trol policy (F` evre et al., 2001), and emphasises the need
for a better understanding of the livestock trade and for
controlmeasuresdesignedspecificallyforpreventingthe
long-range spread of zoonotic diseases.
Acknowledgements
This work was funded by the Animal Health Pro-
gramme of the UK Department for International Devel-
opment(DFID)andtheCunninghamTrust,althoughthe
views expressed are those of the authors and not neces-
sarilythoseofthefundingbodies.WethanktheDirector
of the Livestock Research Institute, Uganda, for hosting
thefieldcomponentsoftheresearch,andtheDistrictVet-
erinary Officers in Kamuli, Tororo and Soroti for their
support. DJS is funded by the Wellcome Trust.
References
Barrett,M.P.,2001.VeterinarylinktodrugresistanceinhumanAfrican
trypanosomiasis? Lancet 358, 603–604.
Clausen,P.H.,Adeyemi,I.,Bauer,B.,Breloeer,M.,Salchow,F.,Staak,
C., 1998. Host preferences of tsetse (Diptera: Glossinidae) based
on bloodmeal identifications. Med. Vet. Entomol. 12, 169–180.

Page 4

232
E.M. F` evre et al. / Acta Tropica 97 (2006) 229–232
Coleman, P.G., Welburn, S.C., 2004. Are fitness costs associated with
resistance to human serum in Trypanosoma brucei rhodesiense?
Trends Parasitol. 20, 311–315.
F` evre, E.M., Coleman, P.G., Odiit, M., Magona, J.W., Welburn, S.C.,
Woolhouse, M.E.J., 2001. The origins of a new Trypanosoma bru-
cei rhodesiense sleeping sickness outbreak in eastern Uganda.
Lancet 358, 625–628.
F` evre,E.M.,Picozzi,K.,Fyfe,J.,Waiswa,C.,Odiit,M.,Coleman,P.G.,
Welburn, S.C., 2005. A burgeoning epidemic of sleeping sickness
in Uganda. Lancet 366, 745–747.
Gibson,W.,Backhouse,T.,Griffiths,A.,2002.Thehumanserumresis-
tanceassociatedgeneisubiquitousandconserevedinTrypanosoma
bruceirhodesiensethroughoutEastAfrica.Infect.Genet.Evol.25,
1–8.
Hide, G., Tait, A., Maudlin, I., Welburn, S.C., 1996. The origins,
dynamics and generation of Trypanosoma brucei rhodesiense epi-
demics in East Africa. Parasitol. Today 12, 50–55.
Magona, J.W., Mayende, J.S.P., Olaho-Mukani, W., Coleman, P.G.,
Jonsson, N.N., Welburn, S.C., Eisler, M.C., 2003. A comparative
study on the clinical, parasitological and molecular diagnosis of
bovine trypanosomosis in Uganda. Onderstepoort J. Vet. Res. 70,
213–218.
Magona, J.W., Walubengo, J., Odiit, M., Okedi, L.A., Abila, P.,
Katabazi, B.K., Gidudu, A.M., Olaho-Mukani, W., 2005. Implica-
tionsofthere-invasionofSoutheastUgandabyGlossinapallidipes
ontheepidemiologyofbovinetrypanosomosis.Vet.Parasitol.128,
1–9.
Odiit, M., Shaw, A., Welburn, S.C., F` evre, E.M., Coleman, P.G.,
McDermott, J.J., 2004. Assessing the patterns of health-seeking
behaviourandawarenessamongsleeping-sicknesspatientsineast-
ern Uganda. Ann. Trop. Med. Parasitol. 98, 339–348.
Okiria, R., Okuna, N.M., Magona, J.W., Mayende, J.S.P., 2002. Sus-
tainability of tsetse control by subsequent treatment of 10% of
a previously treated Ugandan cattle population with 1%, w/v
deltamethrin. Trop. Anim. Health Prod. 34, 105–114.
Picozzi, K., Tilley, A., F` evre, E.M., Coleman, P.G., Magona, J.W.,
Odiit, M., Eisler, M.C., Welburn, S.C., 2002. The diagnosis of try-
panosomeinfections:applicationsofnoveltechnologyforreducing
disease risk. Afr. J. Biotechnol. 1, 39–45.
Waiswa, C., Olaho-Mukani, W., Katunguka-Rwakishaya, E., 2003.
Domestic animals as reservoirs for sleeping sickness in three
endemic foci in south–eastern Uganda. Ann. Trop. Med. Parasitol.
97, 149–155.
Welburn,S.C.,Picozzi,K.,F` evre,E.M.,Coleman,P.G.,Odiit,M.,Car-
rington, M., Maudlin, I., 2001. Identification of human-infective
trypanosomes in animal reservoir of sleeping sickness in Uganda
by means of serum-resistance-associated (SRA) gene. Lancet 358,
2017–2019.