TRANSMISSION AND INFECTION DYNAMICS OF

THEILERIOSIS IN SOUTHERN ZAMBIA: THE EFFECT

OF ENVIRONMENTAL AND HOST FACTORS

Paul Fandamu

Dissertation submitted in fulfillment of the requirements for the degree of Doctor

(Ph.D.) in Veterinary Sciences

Promoters: Prof. Dr. ir. L. Duchateau Prof. Dr. ir. D. Berkvens

2005

Ghent University, Faculty of Veterinary Medicine Department of Physiology and Biometrics

Salisburylaan 133, B-9820, Merelbeke, Belgium

ii

This thesis is dedicated to my dear wife Jane and my two boys Mwape and

Masulani.

iii

ACKNOWLEDGEMENTS

In the first instance I would like to express my profound gratitude to my supervisors.

Prof. Luc Duchateau for his invaluable and constructive scientific guidance and tireless

support during the entire study period and during the preparation of all my published

papers and this thesis. I am grateful to Prof. Dirk Berkvens for his efforts in making

sure that my scholarship to undertake this work materialised and for his useful

guidance, encouragement, criticism, stimulating discussions and scientific advice before

and during the entire study period.

I am greatly indebted to all members of my study advisory committee (Dr. Tom Dolan,

Prof. Jozef Vercruysse, Prof. Luc Duchateau, Prof. Dirk Berkvens and Dr. Tanguy

Marcotty) for their support and constructive comments on the thesis and all the papers

that make part of this thesis.

My deepest appreciation goes to Dr. Niko Speybroeck for the time and patience he had

with me discussing the statistical analysis of the data I collected.

I am also grateful to all members of staff at the Institute of Tropical Medicine, Animal

Health Department, Antwerpen, who were always at hand to assist: Prof. Jef Brandt,

Prof. Redgi De Deken, Prof. Pierre Dorny, Dr. Maxime Madder, Dr. Peter Van den

Bossche, Dr. Eric Thys, Dr. Dirk Geysen, Danielle De Bois and many others.

The work presented in this thesis would not have been accomplished without the

support and assistance of the following people: Dr. Misheck Mulumba, Dr. Victor

Mbao, Dr. Jupiter Mtambo, Dr. Laurent Mostin, Dr. Patrick Mulenga, Dr. Christopher

Kubi, Dr. Michel Billiouw, Mr. Rik Elyn, Dr. George Chaka, Dr. Livwalii Mataa, all

iv

Veterinary Assistants, Laboratory Technicians and other members of staff of the former

ASVEZA project in both southern and eastern Zambia. The cooperation received from

all the traditional cattle farmers in southern Zambia whose animals were used in the

various studies, especially Mr. Bbuku senior of Nteme, Monze is greatly appreciated.

I am thankful to the Belgian Technical Cooperation (BTC) for providing the scholarship

for this work and I would more specifically like to thank Mr. Thierry Coppin for dealing

with all administrative matters of my scholarship.

I am grateful to Dr. Peter Sinyangwe, Director of Veterinary Services and Livestock

Development, Zambia, for the support and for allowing me to undertake this study

under the ASVEZA project and also for granting me the study leave to complete this

work. The Provincial Veterinary Officer for Southern Province, Zambia, Dr. Linous

Munsimbwe is also thanked for his support.

Finally, I can not forget to mention the love and support received from my wife Jane

and my two boys Mwape and Masulani for allowing me the time I spent on this work

and not always with them.

v

LIST OF ABBREVIATIONS

AIC Akaike’s Information Criterion

ASVEZA Assistance to the Veterinary Services of Zambia

CSO- Zambia Central Statistical Office, Zambia

CTLs Cytotoxic T-lymphocytes

ECF East Coast fever

EDTA Ethylenediamine tetra-acetic acid

ELISA Enzyme-linked immunosorbent assay

ENSO El Niño Southern Oscillation

FAO Food and Agriculture Organisation

FMD Foot and Mouth Disease

IFAT Indirect Fluorescent Antibody Test

ILRI International Livestock Research Institute

ITCZ Intertropical Convergence Zone

ITM Institute of Tropical Medicine

MCV Mean Corpuscular Volume

MEI Multiple El Niño Southern Oscillation Index

MHC Major histocompatibility complex

NALEIC National Livestock Epidemiology and Information Centre

NOAA National Oceanic and Atmospheric Administration

PC Principal Component

PCR Polymerase Chain Reaction

PCV Packed Cell Volume

RBC Red Blood Cell

STATACorp Stata Corporation statistical software

VA Veterinary Assistant

vi

TABLE OF CONTENTS

ACKNOWLEDGEMENTS......................................................................................... III

LIST OF ABBREVIATIONS........................................................................................V

TABLE OF CONTENTS............................................................................................. VI

CHAPTER 1 ....................................................................................................................1

LITERATURE REVIEW...............................................................................................1

1.1 Introduction ................................................................................................................................. 1

1.2 The parasite Theileria parva............................................................................................................... 2 1.2.1 Classification of the parasite .......................................................................................................... 2 1.2.2 Distribution of Theileria parva in Africa ....................................................................................... 3

1.2.2.1 Factors affecting the geographical distribution of Theileria parva ........................................ 5

1.3 The vector R. appendiculatus/ R. zambenziensis .............................................................................. 5 1.3.1 Distribution of R. appendiculatus/R. zambenziensis in Africa ....................................................... 6

1.3.1.1 Factors affecting the abundance and distribution of R. appendiculatus/R. zambenziensis ..... 6

1.4 Life cycle of Theileria parva................................................................................................................ 8 1.4.1 Cycle of T. parva in the bovine host .............................................................................................. 8 1.4.2 Life cycle of T. parva in the ixodid tick R. appendiculatus ........................................................... 9 1.4.3 Factors that affect T. parva infection in the tick........................................................................... 10

1.4.3.1 Factors related to the bovine host ......................................................................................... 10 1.4.3.2 Intrinsic factors related to the development of T. parva in the tick ...................................... 11 1.4.3.3 Eco-geographical origin of ticks........................................................................................... 12 1.4.3.4 Sex of ticks ........................................................................................................................... 12

1.4.4 Factors that affect T. parva infection in the host animal .............................................................. 13 1.4.4.1 Immunity to T. parva infections ........................................................................................... 14

1.5 Clinical signs and pathology.............................................................................................................. 15 1.5.1 Diagnosis of T. parva infections .................................................................................................. 17

1.6 Epidemiology of East Coast fever..................................................................................................... 18 1.6.1 Mechanisms in the epidemiology of ECF .................................................................................... 19

1.7 Control of ECF................................................................................................................................... 21

1.8 East Coast fever in Zambia/Southern Zambia ................................................................................ 23 1.8.1 Geographical information on Zambia .......................................................................................... 23 1.8.2 History of ECF in Zambia and Southern Zambia......................................................................... 26 1.8.3 East Coast fever control options in Southern Zambia .................................................................. 28

1.8.3.1 East Coast fever immunisations in Southern Zambia (the Infection and Treatment Method).......................................................................................................................................................... 29

1.9 El Niño Southern Oscillation and global climate variability.......................................................... 31

1.10 References......................................................................................................................................... 35

vii

CHAPTER 2 ..................................................................................................................52

OBJECTIVES OF THE THESIS ................................................................................52

CHAPTER 3 ..................................................................................................................53

THEILERIA PARVA SEROPREVALENCE IN TRADITIONALLY KEPT CATTLE IN SOUTHERN ZAMBIA AND EL NIÑO...............................................53

3.1 Introduction........................................................................................................................................ 54

3.2 Materials and Methods...................................................................................................................... 55 3.2.1 Areas of study and surveys........................................................................................................... 55 3.2.2 Serological tests............................................................................................................................ 56 3.2.3 Meteorological data...................................................................................................................... 56 3.2.4 Data analysis ................................................................................................................................ 57

3.3 Results ................................................................................................................................................. 58

3.4 Discussion............................................................................................................................................ 63

3.5 References........................................................................................................................................... 66

CHAPTER 4 ..................................................................................................................69

EAST COAST FEVER AND MULTIPLE EL NIÑO SOUTHERN OSCILLATION RANKS..............................................................................................69

4.1 Introduction........................................................................................................................................ 70

4.2 Materials and methods ...................................................................................................................... 71 4.2.1 Study area..................................................................................................................................... 71 4.2.2 Sampling ...................................................................................................................................... 71 4.2.3 Meteorological data...................................................................................................................... 72 4.2.4 Statistical methods........................................................................................................................ 72

4.3 Results ................................................................................................................................................. 74 4.3.1 Multiple El Niño Southern Oscillation indices............................................................................. 74 4.3.2 East Coast fever transmission and El Niño .................................................................................. 75

4.4 Discussion............................................................................................................................................ 76

4.5 References........................................................................................................................................... 79

CHAPTER 5 ..................................................................................................................82

RED BLOOD CELL VOLUME AS A PREDICTOR OF FATAL REACTIONS IN CATTLE INFECTED WITH THEILERIA PARVA KATETE................................82

5.1 Introduction........................................................................................................................................ 83

5.2 Materials and methods ...................................................................................................................... 84 5.2.1 Animals ........................................................................................................................................ 84 5.2.2 Haematological examinations ...................................................................................................... 85

viii

5.2.3 Statistical analysis ........................................................................................................................ 86

5.3 Results ................................................................................................................................................. 86 5.3.1 Packed Cell Volume (PCV) ......................................................................................................... 87 5.3.2 Mean Corpuscular Volume (MCV).............................................................................................. 87 5.3.3 Sizes of parasitised and unparasitised erythrocytes in T. parva infections................................... 90

5.4 Discussion............................................................................................................................................ 92

5.5 References........................................................................................................................................... 95

CHAPTER 6 ..................................................................................................................99

PERCEPTION OF CATTLE FARMERS ON EAST COAST FEVER IMMUNISATIONS IN SOUTHERN ZAMBIA.........................................................99

6.1 Introduction...................................................................................................................................... 100

6.2 Materials and methods .................................................................................................................... 101 6.2.1 Statistical analysis ...................................................................................................................... 102

6.3 Results ............................................................................................................................................... 102 6.3.2 Perception of ECF immunisations by cattle farmers .................................................................. 107 6.3.3 Official statistics on ECF in Southern Zambia ........................................................................... 109

6.4 Discussion and conclusions.............................................................................................................. 110

6.5 References......................................................................................................................................... 114

CHAPTER 7 ................................................................................................................116

GENERAL DISCUSSION..........................................................................................116

7.1 Introduction...................................................................................................................................... 116

7.2 Theileria parva sero-prevalence and transmission dynamics in traditionally kept cattle and the El Niño events......................................................................................................................................... 117

7.3 Predicting case fatalities in T. parva Katete experimentally infected animals ........................... 122

7.4 Cattle farmers’ perception of ECF immunisation in Southern Zambia ..................................... 124

7.5 Conclusions and future research .................................................................................................... 126

7.5 References......................................................................................................................................... 127

SUMMARY..................................................................................................................130

SAMENVATTING......................................................................................................134

CURRICULUM VITAE.............................................................................................138

Chapter 1

1

Chapter 1

Literature review

1.1 Introduction

Theileria parva, a tick-borne protozoan parasite, causes East Coast fever (ECF) in

cattle, a major livestock disease in Eastern, Central and Southern Africa. The disease is

transmitted transstadially by the three-host ticks Rhipicephalus appendiculatus and

Rhipicephalus zambeziensis. Larvae and nymphs of these ticks become infected when

feeding on infected bovid hosts and transmit the parasites in the next tick instar, i.e., as

nymph or adult (Ochanda et al., 1996).

East Coast fever was introduced into Southern Africa from Eastern Africa at the

beginning of the twentieth century. It was first reported in January 1902 in Zimbabwe

(then Southern Rhodesia) (Gray and Robertson, 1902). The disease was introduced

through a consignment of cattle brought from Tanzania through Mozambique (then

Portuguese East Africa) to restock the country after the rinderpest pandemic of 1896-

1897, when most animals were wiped out (Thomson, 1985). In 1903, Koch identified

the disease as a protozoan disease and proposed the name African Coast Fever that was

later changed to East Coast fever to emphasise the origin of the disease (Norval et al.,

1992). After a number of experiments, Lounsbury (1904) concluded that Rhipicephalus

appendiculatus ticks were the principal vectors of the disease. Theiler eventually named

the causal organism of the disease Theileria parva in 1904 (Norval et al., 1992).

The disease later spread throughout Southern Rhodesia and the neighbouring

Natal Province of South Africa, Swaziland and Southern Mozambique (Norval et al.,

1992). It was estimated that over 50% of cattle population in Zimbabwe died of the

disease. As a consequence, the disease disrupted ox-drawn transport upon which

2 Literature review

agriculture, commerce and mining activities depended at the time (Lawrence, 1991).

ECF was eradicated by 1960 in South Africa, Swaziland and Zimbabwe through

rigorous short-interval dipping (introduced in 1909), livestock movement controls,

quarantine and slaughter (Lawrence, 1985).

In this chapter we review the most important information concerning T. parva

and its vectors R. appendiculatus/R. zambenziensis and the different risk factors

involved in disease transmission. The emphasis is put on the environmental and host

factors that affect the transmission of T. parva, laying the core foundation of this thesis

work.

1.2 The parasite Theileria parva

Theileria parva is a protozoan parasite of domestic cattle and wild buffalo.

1.2.1 Classification of the parasite

The classification of Theileria has for many years been controversial (Irvin, 1987).

Parasites of the genus Theileria belong to the group classically known as the phylum

Protozoa, which has been recently, renamed the kingdom Protista by taxonomists

(Corliss, 1984). The most recent classification is by Levine (1988) in which the

following families in the order Piroplasmida were included: Anthemosomatidae,

Babesiidae, Theileriidae and Haemohormidiidae.

Chapter 1

3

Below is a classification of T. parva as revised by Levine et al. (1980):

Kingdom: Protista

Subkingdom: Protozoa

Phylum: Apicomplexa (apica complex visible, sexuality by syngamy)

Class: Sporozea (produce infective sporozoites)

Subclass: Piroplasmia (piroform, round, rod-shaped parasites)

Order: Piroplasmida (asexual and sexual reproduction; tick vector)

Family: Theileriidae (schizont stages in lymphocytes)

Genus: Theileria (piroplasms in erythrocytes lack pigment)

Species: Theileria parva

Uilenberg (1976) and Lawrence (1979) proposed a trinomial system of

classification of the three forms of T. parva; (1) T. parva parva for parasites causing

classical ECF, (2) T. parva lawrencei for parasites causing Corridor disease and (3)

T. parva bovis for those parasites causing January disease. However, this classification

was abandoned as the molecular genetic characterisation and cross immunity data did

not substantiate subspecies within the T. parva complex (Conrad et al., 1987, 1989;

Allsopp et al., 1989). It was therefore recommended that the T. parva parasites be

classified as either cattle- or buffalo-derived (Anonymous, 1989).

1.2.2 Distribution of Theileria parva in Africa

Theileria parva follows the distribution of the vector R. appendiculatus (Lawrence,

1991). Theileria parva is currently distributed within eleven countries in eastern, central

and Southern Africa where it is a major constraint to cattle production (Mukhebi et al.,

1992). The affected countries include Burundi, Democratic Republic of Congo, Kenya,

4 Literature review

Malawi, Mozambique, Rwanda, Sudan, Tanzania, Uganda, Zambia and Zimbabwe

(Figure 1.1).

Figure 1.1: Distribution range of Theileria parva (adapted from Norval et al., 1992;

Speybroeck et al., 1999 and Chaka, 2001) and the distribution of Rhipicephalus

appendiculatus (adapted from Berkvens et al., 1998; Speybroeck et al., 1999; Walker et

al., 2000 and Chaka, 2001).

Chapter 1

5

1.2.2.1 Factors affecting the geographical distribution of Theileria parva

There are a number of factors that are known to influence the occurrence of T. parva.

Among them are the following:

(a) The abundance and distribution of R. appendiculatus which depend on climate and

vegetation (Lessard et al., 1990).

(b) The presence of cattle, the main host of T. parva and also wild bovidae like the

African Buffalo (Syncerus caffer) (Lessard et al., 1990). Generally, in areas with low

cattle numbers, the prevalence of T. parva is minimal despite the presence of

R. appendiculatus.

(c) Resistance of the host cattle to ticks and tick-borne diseases. It is widely known that

local Bos indicus cattle (Zebu and Sanga) show much higher resistance to tick-borne

diseases than the Bos taurus breeds (De Castro, 1997). This resistance interferes with

the transmission of tick-borne diseases. Paling and Geysen (1981) also showed that the

Rwandan Ankole cattle (Sanga), which is similar to the Sanga of Southern Zambia had

a marked natural resistance to T. parva infection.

(d) Tick control policies that affect tick populations. In areas where high level of tick

control was practised and then relaxed or stopped for any reason, there has usually been

an increase in theileriosis cases and mortalities (Lessard et al., 1990).

1.3 The vector R. appendiculatus/ R. zambenziensis

Rhipicephalus appendiculatus and the closely related R. zambeziensis are three-host

ticks and are considered to be the principal vectors of T. parva.

6 Literature review

1.3.1 Distribution of R. appendiculatus/R. zambenziensis in Africa

Rhipicephalus appendiculatus occurs over large areas in Kenya, Uganda, Rwanda,

Burundi, Tanzania, Zambia, Malawi, Zimbabwe, Swaziland and South Africa (Norval et

al., 1992) (Figure 1.1). However, it has been observed that the distribution of

R. appendiculatus is by no means continuous, even in those countries in which it occurs

most commonly (Soneshine, 1993).

1.3.1.1 Factors affecting the abundance and distribution of R. appendiculatus/R.

zambenziensis

The distribution of R. appendiculatus/R. zambenziensis is influenced by several factors,

the most important of which are climate, vegetation and host availability. Throughout

the regions where R. appendiculatus occurs, climatic conditions vary and in response

the ticks have evolved different behavioural strategies. In tropical Eastern and Central

Africa, ticks experience no prolonged long dry seasons and all stages are present

throughout the year. No diapause occurs whatever the life cycle stage. In contrast, in

subtropical Central and Southern Africa, with well defined hot and cold seasons and a

single period of rainfall, there is a clearly defined pattern of seasonal occurrence which

is regulated by diapause in the unfed adult (Pegram and Banda, 1990; Berkvens et al.,

1995). Diapause is a neurohormonally mediated, state of low metabolic activity that is

associated with reduced morphogenesis, increased resistance to environmental extremes

and altered or reduced behavioural activity. Diapause not only affects the life cycle and

survival of R. appendiculatus but also influences the epidemiology of ECF. Where

diapause occurs, ticks cycle through one generation per year with almost no overlap in

feeding periods of nymphs and adults. Because transmission is dependent on the

infection in nymphs by recovered carrier hosts (larvae hardly pick up infection from

Chapter 1

7

carriers), the occurrence of the disease is largely restricted to the periods of adult

activity (Sonenshine and Mather, 1994). In areas where there is no diapause, all life

cycle stages occur on hosts simultaneously and the disease may be present throughout

the year with transmission from clinically ill animals as well as from carriers

(Sonenshine and Mather, 1994).

It is assumed that the seasonal pattern of R. appendiculatus is set by the adults,

which are only active under specific climatic conditions (related mainly to temperature

and rainfall/humidity) (Norval et al., 1992). To the contrary, Randolph (1993) reckons

that it was the timing of the questing activity of the desiccation-vulnerable larvae that

determined the pattern of the tick’s seasonal dynamics in Southern Africa. The

nighttime minimum temperatures determine condensation and saturation deficit and

thus the tick’s ability to replenish the moisture lost during the daytime. Larvae are most

sensitive to these factors. The timing of the activity of nymphs and adults is determined

by temperature-dependent developmental rates plus the delaying phenomenon of

diapause, in order to achieve maximum reproduction and survival ensuring the

occurrance of eggs and larvae during periods of optimal climatic conditions.

Vegetation is very important for R. appendiculatus as it provides favourable

microclimatic conditions and increases chances of ticks finding a host (Lessard et al.,

1990). In addition, vegetation increases chances of survival of the free-living stages of

the tick. Where vegetation cover is reduced by overgrazing and removal of trees,

R. appendiculatus tends to disappear. It has been observed that in many hotter and drier

areas of Central and Southern Africa especially the river valleys, R. zambeziensis tend

to replace R. appendiculatus (Walker et al., 1981). No diapause has been detected in

R. zambeziensis.

8 Literature review

The absence of R. appendiculatus from climatically suitable areas like those in

Western Africa has been attributed to the low densities of cattle and other herbivore

hosts (Norval et al., 1992).

1.4 Life cycle of Theileria parva

Theileria parva undergoes developmental cycles in both the bovine host and the tick

vector. The vector R. appendiculatus is a three-host cattle tick with each instar (larva,

nymph and adult) feeding on a different animal. Theileria parva transmission is

transstadial in that the larvae or nymphs that feed on infected bovine will transmit

T. parva infection in their subsequent instars as nymphs or adults. There is no

transovarial transmission (F.A.O., 1984; Brown et al., 1990; Ochanda et al., 1996).

1.4.1 Cycle of T. parva in the bovine host

During feeding, the infected tick will pass on sporozoites to a susceptible bovine

through its saliva. Following infection, the sporozoites invade the T and B-lymphocytes

of the host animal. The entry of the sporozoite into the lymphocytes involves a chain of

events that starts with the contact between sporozoite and lymphoid cell, in which the

cell membranes of the sporozoite and the host cell come into close proximity. A

zippering effect follows, whereby the sporozoite and the host cell membrane become

tightly attached culminating eventually, in the internalisation of the sporozoite (Shaw et

al., 1991). Once inside the lymphocyte, the sporozoite develops into a schizont.

Parasitised lymphocytes then transform to lymphoblasts, which have a capacity to

divide repeatedly. The proliferation of parasitised lymphoblasts is followed by

lymphodegeneration (F.A.O., 1984). The schizonts later on rupture to release

merozoites from the host cell. These merozoites then invade erythrocytes and

Chapter 1

9

subsequently develop to piroplasms. It is the presence of these piroplasms in the

infected animal that provide the source of infection for other ticks when they feed at the

time of parasitaemia (Figure 1.2).

1.4.2 Life cycle of T. parva in the ixodid tick R. appendiculatus

Theileria parva undergoes a sexual cycle in the tick to produce zygotes that invade gut

wall cells (Figure 1.2). Thereafter, kinetes are produced which are capable of entering

salivary gland cells. No subsequent development takes place until the tick starts to feed

in its next instar. Sporogony starts rapidly when the tick starts to feed and mature

sporozoites that are infective to cattle are produced within the salivary gland cells three

to five days later (F.A.O., 1984).

Figure 1.2: Life cycle of T. parva in cattle and the ixodid tick R. appendiculatus

10 Literature review

1.4.3 Factors that affect T. parva infection in the tick

Factors that affect infection levels of T. parva in the tick vector can be divided into

factors that are related to the bovine host, intrinsic factors related to the development of

T. parva in the tick, tick’s eco-geographical origin and sex of ticks.

1.4.3.1 Factors related to the bovine host

Young et al. (1996) identified six main factors that influenced T. parva experimental

infections in R. appendiculatus ticks: parasitaemia, leukopenia, packed cell volume

(PCV), sex of the animal, age of the animal and temperature.

Parasitaemia in cattle was found to be the most important factor that influenced

infection of T. parva in ticks (Purnell et al., 1974). This was also confirmed by Young

et al. (1996) who found that parasitaemia of over 5% had a marked effect on abundance

of infection. They also found that an increase of parasitaemia from 5% to 60% led to an

elevation in prevalence from 50% to 73% and a raised abundance in infected acini/tick

from 30 to 109 infected acini/tick.

Leukopenia, a common symptom associated with T. parva infection, can be used

as an index of disease severity (Dolan et al., 1984a). Young et al. (1996) found that the

leukocyte count was associated with the level of infection developing in the tick. A

reduction in PCV is an indicator of the development of anaemia and could also be used

to indicate the severity of the disease though this is normally not reported as a regular

symptom in T. parva infections (Wilde, 1967). Low PCV could be related to higher

infections in ticks (Young et al., 1996).

The other important factor is the sex of cattle. Young et al. (1996) found that

ticks fed on male animals had significantly higher infection rates than those fed on

female animals, though this was less important than other factors considered.

Chapter 1

11

Furthermore, cattle less than six months of age produced lower infections in ticks than

those above six months (Young et al., 1996).

It has been shown that ambient temperature influences T. parva infection in the

tick. Temperature below 180C and above 330C for longer periods inhibit the

transmission of T. parva by R. appendiculatus ticks. In addition, temperatures higher

than 400C or lower than 150C have been shown to prevent moulting of

R. appendiculatus ticks (Young and Leitch, 1981). In an experiment where ticks were

fed on infected cattle kept in cold and hot pens, ticks fed on infected cattle in a cold pen

produced higher infections than those fed on cattle in the hot pen (Young et al., 1996).

Low ambient temperature prolongs the feeding period of R. appendiculatus instars on

cattle, which in turn means higher quantities of blood ingested (Ochanda et al., 1988).

1.4.3.2 Intrinsic factors related to the development of T. parva in the tick

T. parva faces a number of hazards in the tick that affect its development. For instance

at the engorgement stage of tick feeding, many millions of piroplasms are ingested even

from cattle with low parasitaemia (Walker, 1990; Shaw and Young, 1994). However,

very few survive to become sporoblasts in the salivary glands because the ingested

piroplasms are destroyed by acid phosphate enzymes, which are secreted into the tick

gut during the preparatory stage of feeding (Walker, 1990).

Another hazard faced by Theileria in the gut is that of the phagocytic cells of the

tick gut. These cells ingest the red and white blood cells from cattle (Walker, 1990).

Even when the Theileria zygote survives within a phagocytic cell, it will remain in the

centre of the gut wall where these cells are active and will eventually face the barrier of

the complete gut wall when the zygote transforms to the kinete stage.

When a larva moults to a nymph or a nymph to an adult, the number of acini in

the glands increases five fold and one hundred fold respectively (Walker, 1990). During

12 Literature review

the process of replication and differentiation of the salivary gland cells, there may be

sufficient external markers of the identity of various acinar cells for the kinetes to select

the appropriate cell type out of the 18 that comprise the salivary glands (Walker et al.,

1985). Kinetes only survive in specific cell types even if they invade many other cells.

Tick salivary glands regress to primordia after each instar has attained its

repletion weight and has dropped from the host. The glands redevelop during each

moult (Sauer et al., 1995), and entry of the kinetes into the acini appears possible after

moulting and the full development of the glands is completed (Fawcett et al., 1981). In

addition, the T. parva already present within the salivary glands of the nymph or adult

prior to feeding do not survive during the regression to primordia (Shaw and Young,

1994). Furthermore, if a nymph had become infected with T.parva when feeding as a

larva, the Theileria in its salivary glands will either be transmitted to the next feeding

host or die within the tick (Walker, 1990).

1.4.3.3 Eco-geographical origin of ticks

Norval et al. (1991) reported that ticks from Eastern Africa transmitted the disease

better than those from Southern Africa. Ochanda et al. (1998) also demonstrated that

higher infections of T. parva can develop in some ticks compared to other ticks

depending on the origin of the tick stocks.

1.4.3.4 Sex of ticks

Female ticks have higher infection rates and levels of T parva than the male ticks (Irvin

et al., 1981). It is reported that the major difference between the male and female

R. appendiculatus tick is the number of “e” cells in the type III acini, which are more

numerous in the females than in the males. T. parva parasites appear to develop mainly

in the “e” cells of the type III acini (Fawcett et al., 1982) although development may

Chapter 1

13

also occur in the type II acini especially when ticks feed on cattle with high levels of

parasitaemia (Walker, 1990).

1.4.4 Factors that affect T. parva infection in the host animal

One of the most important factors known to affect T. parva transmission in the host is

that of resistance (Norval et al., 1992). Individual animals acquire immunity during

their lifetime in response to infection, which may hinder infestation of ticks or

development of fatal infection. The ability to acquire this immunity is inherited, and this

ability varies between breeds and types of cattle. There is evidence that various cattle

breeds exhibit a different susceptibility to T. parva infection. Bos indicus breeds are

more resistant to T. parva than Bos taurus breeds (Norval et al., 1992). Ndungu et al.

(2005a) observed a difference in the way T. parva clinical disease developed between

B. indicus and B. Taurus cattle which they assigned to the differing degrees of

resistance. However, they failed to show a difference in susceptibility between the

breeds when using the length of the prepatent period as criterion. Ndungu et al. (2005b)

were also unable to demonstrate any difference in the infectivity or susceptibility of

B. indicus and B. taurus cattle to T. parva Muguga sporozoite stabilate. However, the

disease was more severe in Friesians and Borans than the endemic Zebus despite being

given the same dose. Ndungu and co-workers further showed that B. indicus cattle from

ECF-endemic areas were more resistant to T. parva infection than both B. indicus and

B. taurus cattle from ECF-free areas. Although the mechanism of resistance is not well

understood it is believed that it is likely to be due to a more effective cell mediated

response in Zebu cattle than other breeds (Baldwin et al., 1986). It should be pointed

out that resistance can be reduced in situations where there is malnutrition and stress of

the animal.

14 Literature review

Age related resistance to T. parva has never been demonstrated in cattle. Though

Koch et al. (1990) reported a lower incidence rate of field cases of January Disease in

calves younger than seven months, they failed to detect any age specific resistance to

T. parva bovis infection in calves in a trial.

1.4.4.1 Immunity to T. parva infections

Cattle that recover from ECF, either naturally or as a result of treatment with

tetracyclines or other drugs, are solidly immune to homologous challenge (Neitz, 1957;

Burridge et al., 1972). This type of immunity is mediated by cellular mechanisms

targeted at the parasitised lymphoblasts (Pearson et al., 1979; McKeever et al., 1999).

The major immune mechanism deployed by immune cattle against T. parva is the killer

function of parasite-specific, major histocompatibility complex (MHC) class- I

restricted CD8+ cytotoxic T-lymphocytes (CTLs) (Emery, 1981; Goddeeris et al., 1986;

McKeever et al., 1999). In immune cattle, Theileria-transformed cells express antigens

on their surfaces, which are recognised by effector or cytotoxic cells triggering the

killing of these infected cells (Pearson et al., 1979; Baldwin et al., 1987). Studies by

Eugui and Emery (1981) showed that cytotoxic lymphocytes were found in cattle

immune to T. parva and that the cytotoxicity is restricted to autologous targets. Other

studies by Emery (1981) showed that immunity to ECF can be transferred adoptively

between twins by using thoracic duct lymphocytes from the immunised partner. It

therefore seems that cellular immunity is one of the mechanisms involved in resistance

to T. parva infections (Pearson et al., 1982).

The role played by humoral antibodies in the immunity to ECF has been

previously studied. Wagner et al. (1974) and Muhammed et al. (1975) were unable to

confirm protection of cattle against T. parva infection by immune serum. However,

Musoke et al. (1982) were able to detect a sporozoite-neutralizing response after

Chapter 1

15

multiple challenges with T. parva. Protection in this case is said to be based on

neutralizing antibodies limiting the severity of infection by reducing the number of

sporozoites capable of invading lymphocytes. This principle has been exploited in the

development of a recombinant vaccine. Musoke et al. (1992) reported that

immunisation of cattle with Escherichia coli-derived recombinant p67, the major

surface antigen of T. parva sporozoites, gave approximately 70% protection against

severe disease after laboratory challenge. However, only a small proportion of

immunised cattle were able to show complete neutralisation of infection while the rest

showed schizont parasitosis of varying severity. In addition, Musoke et al. (2005)

reported that the p67 vaccine was able to reduce severe ECF in cattle by 47% and 52%

at the Kenyan coast and Central Kenya respectively. Kaba et al. (2005) also recorded a

protection level of 85% in cattle immunised with the p67 vaccine. McKeever et al.

(1999) however doubted whether the sporozoite-neutralizing response had a significant

role in the immunity against T. parva under natural circumstances. It has been suggested

that multiple challenges with T. parva in addition to the induction of antibodies could

result in boosting of cell-mediated immune responses against schizont infected cells

and/or induction of cell mediated responses of broader specificity (Irvin and Morrison,

1987).

1.5 Clinical signs and pathology

The clinical features of ECF and the severity and outcome of the disease have been

reported to be related to the parasite load (Radley et al., 1974; Irvin and Mwamachi,

1983; Dolan et al., 1984a) and can also be altered by host factors like breed, age and the

health status of an animal (Irvin and Mwamachi, 1983). Another factor that can modify

the course of the clinical disease is the strain of the T. parva parasite involved (Mbogo

16 Literature review

et al., 1996). Under experimental conditions, using either infected ticks or sporozoite

stabilate, the incubation period ranges from 8 to 12 days. The incubation period may be

much more variable in the field owing to differences in challenges experienced by the

cattle and may extend to beyond 3 weeks after attachment of infected ticks.

The initial clinical sign of ECF in cattle appears 7 to 15 days after attachment of

infected ticks. First a swelling of the draining lymph node is observed, usually the

parotid lymph node, for the ear is the preferred feeding site of the vector. This is

followed by a generalised lymphadenopathy in which superficial subcutaneous lymph

nodes such as the parotid, prescapular, and prefemoral lymph nodes, can easily be seen

and palpated (Irvin and Mwamachi, 1983). Fever ensues and continues throughout the

course of infection. This rise in temperature is rapid and is usually in excess of 39.5° C

(it may reach 42°C). As the disease progresses anorexia develops, rumination ceases

and the loss of condition follows. Other clinical signs may include lacrimation which

may be accompanied by photophobia, corneal opacity, nasal discharge, terminal

dyspnoea and diarrhoea. Pregnant animals may abort, especially in the pyrexic stage of

the disease. Before death the animal is usually recumbent, the temperature falls, and

there is a severe dyspnoea due to pulmonary oedema that is frequently seen as a frothy

nasal discharge. Death usually occurs within 30 days after infection of susceptible cattle

by infected ticks (Irvin and Morrison, 1987). Mortality in fully susceptible cattle can be

nearly 100 percent.

The pathogenic effect of T. parva is due to its ability to induce uncontrolled

proliferation of the infected lymphocytes (Hulliger et al., 1964; Malmquist et al., 1970).

Death of infected cattle is a result of this lymphoproliferation, membrane ‘leakage’ and

infiltration of lymphoid and non-lymphoid organs by parasitized cells which interfere

with the normal function of these organs, more in particular the lungs (Ndungu et al.,

Chapter 1

17

2005b). Generally, the muscles and fat appear normal when the carcass is opened.

However, serosal surfaces have petechial and ecchymotic haemorrhages, and serous

fluids may be present in body cavities. Haemorrhages and ulceration may be seen

throughout the gastrointestinal tract particularly in the abomasum and small intestine,

where necrosis of Peyer's patches can be observed. Lymphoid cellular infiltration

appears in the liver and kidney as white foci that have been referred to as pseudo-

infarcts. Fluid is usually present in the pericardium. The most striking post mortem

changes are normally seen in the respiratory organs. The lungs show interlobular

emphysema, severe pulmonary oedema and are hyperaemic, and the trachea and bronchi

are filled with fluid and froth.

1.5.1 Diagnosis of T. parva infections

The common field diagnosis for ECF is based on the clinical signs of the disease and

the demonstration of schizonts in the lymph nodes. In dead animals, impression smears

from cut lymph nodes or other lymphoid organs like the spleen are prepared as a

supplementary diagnostic test. A thin blood smear is also prepared to detect the

presence of piroplasms which appear 5-8 days after the detection of schizonts. These

smears are fixed in methanol and stained in 10% Giemsa stain for 20 to 30 minutes. The

detection of piroplasms only without schizonts, though indicating an infection with

T. parva is not considered to be positive in the diagnosis of clinical cases of T. parva as

this may only be showing the carrier status in a clinically normal animal or another

Theileria sp. (Norval et al., 1992).

Serological tests have been used in surveys to detect ECF recovered animals

(Irvin and Mwamachi, 1983). The most widely used serological test in Africa has been

the indirect fluorescent antibody test (IFAT) (Burridge and Kimber, 1972). Both the

18 Literature review

schizonts and piroplasms can be used as antigen in this test to detect antibodies though

the schizont antigen is preferred as it confers a longer duration of the serological

response. However, these antibodies are not protective but only indicative of previous

exposure to ECF (Irvin and Mwamachi, 1983). The major drawback of the IFAT test is

its lack of specificity as T. parva cross reacts with other Theileria parasites like

T. taurotragi and T. annulata (Burridge et al., 1974; Norval et al., 1992) and it is also

often difficult to assess the degree of fluorescence. Billiouw et al. (2005) reported that

the IFAT performs well when used in the epidemic period with high intensity of

T. parva transmission but afterwards it looses sensitivity though the specificity was not

affected. Other serological tests for diagnosis of ECF have been reviewed by Duffus and

Wagner (1980). Recently, a polymerase chain reaction (PCR) based test has been used

to detect T. parva infections (Ogden et al., 2003) and was described as the best test for

discriminating T. parva from acute infections caused by other Theileria species. The

PCR has also been used to detect the prevalence of T. parva in R. appendiculatus ticks

(Conrad et al., 1989; Young et al., 1996; Ogden et al., 2003). The other test used in the

detection of T. parva antibodies is the enzyme-linked immunosorbent assay (ELISA)

(Katende et al., 1998).

1.6 Epidemiology of East Coast fever

East Coast fever epidemiology is complex as it involves the understanding of a number

of factors related to the tick vector and the Theileria parasite and their interaction which

is mediated through alternative mammalian hosts (Young, 1981).

Chapter 1

19

1.6.1 Mechanisms in the epidemiology of ECF

Understanding the epidemiology of theileriosis requires a good knowledge of the

mechanisms involved in the maintenance of an endemic state (Young, 1981). Endemic

stability for T. parva is defined as a state where there is coexistence between host,

agent, vector and environment. In this situation, the large majority of the cattle

population is infected and immune by six months of age and little or no clinical disease

occurs (Moll et al., 1984). This state can be achieved when cattle possess a low innate

susceptibility to T. parva infection (Morzaria et al., 1988), R. appendiculatus infestation

is virtually continuous throughout the year (Yeoman, 1966) and a large majority of

calves are exposed to a low T. parva challenge. The epidemiological state of theileriosis

caused by the cattle-derived T. parva and transmitted by R. appendiculatus is

determined in a population of cattle, by using four indicators: herd antibody prevalence,

disease incidence, age group of cattle affected by the disease and case-fatality (Norval et

al., 1992). However, Billiouw (2005) suggested that the most important discriminating

criterion in describing the epidemiological states of ECF was the trend in the infection

prevalence. The other important criteria he put forward were the age at first contact and

the case-fatality ratio. He further proposed a new classification of the epidemiological

states with regard to ECF as: (a) epidemic when the age at first contact is

high/decreasing, case fatality is close to 100% and decreasing fast while the infection

prevalence is increasing, (b) first level endemic stability where age at first contact is less

than 2 years, case fatality is about 50% and infection prevalence is stable, (c) second

level endemic stability when age at first contact is less than 2 years, case fatality is

about 25% while infection prevalence is stable and (d) ultimate endemic stability where

age at first contact is less than six months, case fatality close to 0% and infection

prevalence stable.

20 Literature review

Young (1981) postulated that the endemic stability of theileriosis was a

combination of regulatory mechanisms involved in the development of protective

factors against Theileria infections in calves, the level of Theileria infections in the tick

vector population and the size of the vector population feeding on cattle.

The protective factors in the host calf vary considerably in different breeds of

cattle (Young, 1981; Norval et al., 1992). It has been shown that low numbers of Zebu

cattle die from ECF in endemically stable areas whereas a high numbers of Taurine

cattle would die if introduced in these areas (Moll et al., 1984, 1986).

The overall success of the transmission process of T. parva from the tick to the

host is determined by the size of the tick population and by the proportion of ticks that

is infected. In addition, the infection level in the tick population is influenced by the

presence of clinically diseased or carrier animals in the field (Norval et al., 1992). A

carrier animal is one that has survived the primary infection and afterwards maintains

the infectious parasite stages (piroplasms) in blood at levels high enough to infect ticks

but often too low to be detected by normal parasitological investigations (Medley et al.,

1992). Leitch and Young (1981) observed that in endemic areas 1-2% of the field ticks

are infected with T. parva and the majority of the infected ticks contain only one

infected salivary gland acinus. Therefore, the majority of calves in ECF endemic areas

would receive their challenge from one infected tick with only one infected acinus. This

will lead to low parasitaemia in calves which also produces low infections in ticks

feeding on them (Young, 1981). From Barnett’s results (1957) it would be expected that

100% of calves would survive this low challenge, as is effectively observed in the field

(Moll et al., 1986).

The vector population is affected by climate, host density and the development

of host resistance to tick infestation which reduces the feeding success, weight of replete

Chapter 1

21

ticks and the transmission of the Theileria parasite (Young, 1981). Non-seasonal

activity of tick vector instars occurs in more humid areas of the R. appendiculatus

distribution while seasonal activity occurs in association with bimodal rainfall patterns

especially in the highlands of Eastern Africa, and the unimodal rainfall pattern

throughout Southern Africa (Norval et al., 1992). This difference in tick behaviour leads

to different transmission dynamics of T. parva in Eastern Africa and Southern Africa. In

Eastern Africa, ticks are non-diapausing and the parasite can easily be maintained in the

tick vector rather than in the carrier animals. On the other hand, the R. appendiculatus in

Southern Africa undergo diapause in order to survive periods of unfavourable

conditions which do not permit normal activity of development during the long, dry

summer before the onset of the rains. The presence of diapausing populations of

R. appendiculatus plays an important role in T. parva dynamics in that the time between

nymphal and adult feeding is about six months (Rechav, 1981; Berkvens, 1990). This

will as a consequence reduce the transmission efficiency in that most parasites will die

off in the infected ticks. The decline in the T. parva infection levels in

R. appendiculatus nymphs over time has been demonstrated by Ochanda et al. (2003).

Management practices like zero grazing and seasonal calvings will tend to cause

variations in host availability and will as a result influence the abundance of ticks on

cattle.

1.7 Control of ECF

The early ECF control attempts were based on tick control (Dolan, 1999). Following the

discovery of the tick vector for ECF, regulations were developed for tick control using

arsenical acaricides, a concept introduced in Australia to control redwater transmitted by

Boophilus spp (Dolan, 1999). Fourteen ‘tick destroying’ agents were evaluated in Natal

22 Literature review

in 1909 to determine the frequency at which they could be used without risk to animals

(Watkins-Pitchford, 1909). Rigorous tick control measures were supported by

restriction on cattle importations and movement and by quarantine and slaughter

policies that eventually led to the elimination of ECF from Southern Africa by 1960

(Lawrence, 1992; Dolan, 1999). Acaricides are normally applied in dips, spray races, or

by hand spraying. More recently, "pour on" or "spot on" formulations have been

introduced. The application is usually on a weekly basis, but the rate has to be increased

when the challenge is high. The cost of this control measure is exorbitant, and the

farming economies in many countries in Africa are not able to afford it.

There are presently three effective drugs for the treatment of ECF: parvaquone

(Clexon and Parvaxone), buparvaquone (Butalex), and halofuginone lactate (Terit).

Each of these drugs has been introduced to the market within the last 20 years (Norval

et al., 1992). The availability of a therapeutic means of controlling ECF is a significant

development. However, there are two constraints to the widespread use of medication:

the drugs are too expensive for most African farmers, and rapid, accurate diagnosis and

immediate administration of the drug are required for effective therapy (Norval et al.,

1992).

Another method of ECF control is immunisation using the infection and

treatment method. Animals are inoculated with a dose of an infective sporozoite

stabilate prepared from ticks and treated simultaneously with tetracyclines

(Cunningham, 1977). Sporozoite stabilates are produced from adult ticks fed as nymphs

on infected cattle; the adult ticks are ground up in a medium after prefeeding on rabbits

for 4 days, and the sporozoite suspension is prepared by centrifugation and

cryopreserved as a stabilate.

Chapter 1

23

A more rational approach using integrated control has been suggested by Young

et al. (1988). These measures include effective fencing, pasture management, rotational

grazing to reduce the level of challenge, selection of tick resistant cattle and new

methods of immunization, with strategic acaricide application. This approach offers a

more satisfactory method of ECF control (Young et al., 1988).

1.8 East Coast fever in Zambia/Southern Zambia

1.8.1 Geographical information on Zambia

Zambia lies approximately between 8o30’– 18o South and 22o – 33o30’ East. It is a

landlocked country and shares boundaries with the Democratic Republic of Congo in

the north, Tanzania, Malawi and Mozambique in the east, Zimbabwe, Botswana and

Namibia in the south and Angola in the west. The country has an area of 752,614 square

kilometres and is divided into nine provinces namely Northern, Luapula, Copperbelt,

North-western, Eastern, Central, Lusaka, Western and Southern Provinces. It has a

human population of about 9.9 million and the annual human population growth rate

was estimated at 2.3% with a population density of 12.4 per square kilometre (CSO-

Zambia, 2000).

Zambia is composed of a series of plateaus ranging from 900-1500m. The

highest plateau which reaches a maximum altitude of 2000m is in the east and north-

east of the country. The lowest altitude occurs where the Zambezi River enters

Mozambique. The major depressions run from the Northeast to the Southwest and are

composed of the Luangwa Valley and the middle Zambezi Valley, which are southerly

extensions of the East African Rift Valley. The main swamp and flood plain regions are

the Bangweulu and Kafue flats (Figure 1.3).

24 Literature review

The climate in Zambia is tropical and is influenced by altitude. There are three

distinct climatic seasons: (a) hot and dry season from September to November, (b)

warm and wet season from December to April and (d) cool and dry season – from May

to August.

Annual rainfall averages 1400 millimetres in the northern part of the country and

decreases in the south to 600 millimetres. It is mainly the Intertropical Convergence

Zone (ITCZ) that influences the rainfall in Zambia. Mean relative humidity increases

from below 50% in the hot dry season to above 75% during the rainy season. Mean

minimum temperatures in June/July are 5 to 10oC in the Central, Southern and Western

Zambia and 10 to 13oC in the Eastern and North-eastern Zambia. Mean maximum

temperatures in October are highest in the valleys (Luangwa and Zambezi) where they

may exceed 35oC.

Chapter 1

25

Figure 1.3: Map of Zambia showing the major natural regions.

Zambia’s landmass is covered for 80% with Savannah woodlands made up of

the following types:

(a) Miombo woodland characterised by the presence of Brachystegia,

Julbernadia and Isoberlinia species and covering 60% of the woodlands.

(b) Munga woodland dominated by Acacia, Combretum and Terminalia species

that grow among tall grass in Mazabuka and Monze districts of southern

Zambia and Petauke district of eastern Zambia.

(c) Mushibe Woodland (Guibourtia coleosperma) on the western Kalahari sands

of Western Province.

26 Literature review

(d) Mopane woodland (Colophospermum mopane) found in the hot dry drier

valleys of the Luangwa and Zambezi rivers.

(e) Chipya woodland (Erthrophleum africanum, Parinari curatelliofolis,

Plerocarpus angolensis) around the northern lake areas.

(f) Lusese woodland (Burkea africana, Dialium engleanum, Baikiaea species,

and Clophospermum species) found on open grassy woodlands of western

Province and Namwala district of Southern Province.

Grasslands and swamps cover the remaining 20% of the land.

Zambia has a cattle population of about 2.6 million. It has 150 000 sheep, 1.27

million goats, 340 000 pigs and about 30 million chickens (F.A.O., 2004). The

traditionally indigenous cattle in Zambia can be classified in three main types.

(a) Angoni, a short – horned Zebu type found in the Eastern Province.

(b) Barotse, a long – horned Sanga type found in the Western Province.

(c) Tonga, a medium – horned Sanga type in the Southern and Central

Provinces.

The exotic breeds in the commercial sector include Friesians, Jersey, Guernsey, Sussex,

Brahman, and Hereford among others.

1.8.2 History of ECF in Zambia and Southern Zambia

The exact origin of ECF in Zambia is not well documented although it is assumed that it

has been introduced through neighbouring Tanzania (then German East Africa),

(Nambota et al., 1994).

The first case of ECF in Zambia was recorded in 1922 at a place called Fife in

Nakonde district of Northern Province (Chizyuka and Mangani, 1985; Musisi and

Peirce, 1981; Nambota et al., 1994). The disease later spread to Mbala, Kasama and

Chapter 1

27

Chinsali. After the Northern Province outbreaks, the disease was reported in Eastern

Province in Lundazi, Chipata, Chadiza and Katete districts in 1928 (Anonymous, 1974;

Chizyuka and Mangani, 1985). Between 1928 and 1945, there were no cases of

theileriosis reported in the country. The disease reappeared in 1946 when it was

diagnosed in Mbala district of the Northern Province and also in Chipata district of

Eastern Province. Since 1947, ECF has spread within the Northern and Eastern

Provinces where the disease has now become endemic (Chizyuka and Mangani, 1985;

Nambota et al., 1994).

A malignant form of theileriosis was detected in Southern Province in the

Hufwa area of Monze district during the 1977/78 rainy season (Anonymous, 1980;

Nambota et al., 1994). The appearance of the disease in the province has been very

difficult to explain due to lack of documented evidence. Nambota (1989) reported that

Southern Province was probably free of ECF prior to 1977. However, this is very

difficult to prove. Although the disease was first recorded in November 1977

(Anonymous, 1980), it is possible that the disease was already there. Grootenhuis

(1989) stated that most wild bovidae are carriers of Theileria parasites and that the

African buffalo has lived in harmony with T. parva and its vectors R. appendiculatus

even before cattle were introduced in sub-Saharan Africa. Indeed, the disease might not

have been well diagnosed prior to 1977 or the conditions might not have existed to

facilitate transmission of the disease. When it was reported in 1977, the disease was

referred to as Corridor Disease because the unusual appearance of the disease was

associated with extensive flooding of the Kafue flats that season pushing the National

Park buffalos to seek for grazing in closer proximity to traditional cattle and hence

aiding the transmission of the disease to cattle (Anonymous, 1980). However, the

Corridor disease is now thought to have been East Coast fever. There is currently no

28 Literature review

evidence to show that the buffalo is involved in the transmission of T. parva in the

province. Geysen et al. (1999) have shown that the disease in Southern Province is

similar to ECF caused by T. parva and that both Eastern and Southern Province T.

parva stocks have a common Eastern African origin.

The disease is now endemic in most plateau districts of the province and is

threatening cattle population in this region, which makes up approximately 45% of the

national herd (Nambota et al., 1994). The disease has been reported to have spread from

Southern Province to Lusaka, Central and Copperbelt Provinces (Nambota et al., 1997).

1.8.3 East Coast fever control options in Southern Zambia

The most widely used method of ECF control in Southern Zambia has been tick control.

However, the high costs of acaricides have forced some farmers to stop spraying or

dipping completely. This has been made worse by the new government policy of no

longer providing acaricides free of charge to farmers. Before, weekly application of

acaricides had been advocated in the rainy season which was then relaxed in the dry

season. However, this approach needs to be changed as theileriosis outbreaks in recent

years have been occurring almost throughout the year (Nambota et al., 1994) and in

some instances more cases have been observed in the dry season (Mulumba et al., 2000,

2001).

Antitheilerial drugs currently used to treat ECF in Southern Zambia include

Buparvaquone (“Butalex”, Coopers/Wellcome) 5mg/Kg i/m and Parvaquone

(“Parvaxone”, Bimeda) 20mg/Kg i/m. The problems with chemotherapy are the high

costs of antitheilerial drugs. Costs were estimated at between US$9.04 and US$27.31

Chapter 1

29

per animal (D’Haese et al., 1999). Furthermore, animals have to be diagnosed at an

early stage of the disease so that drugs can be given at the start of clinical signs.

The infection and treatment method of immunisation has been carried out in

Southern Zambia (Nambota, 1989) using the Muguga cocktail. A local T. parva

Chitongo strain vaccine has since been isolated and used in the province on a large scale

among traditional cattle farmers and has proved to be very effective.

1.8.3.1 East Coast fever immunisations in Southern Zambia (the Infection and

Treatment Method)

East Coast fever immunisation by the infection and treatment method involves two

simultaneous processes; one of actively infecting cattle with the parasite and the second

of treating the recipient cattle chemoprophylactically with tetracycline which acts

during the early stages of the disease, and results in mild or inapparent reactions

(Radley, 1981). The use of antibiotics emanated from the early work of Neitz (1953)

who found that when infected ticks were applied to cattle that were simultaneously

treated with Aureomycin the cattle underwent mild reactions and were immune on

rechallenge. However, this procedure was not practical as a method of immunisation in

the field (Dolan et al., 1984b). With the development of triturated infected tick

stabilates (Cunningham et al., 1973) and the use of long-acting tetracyclines (Radley et

al., 1975) this method has become more practical for large-scale field application.

The major complicating factor for successful immunisations against ECF using

this method has been that of strain selection. Different strains of T. parva exist in the

field (Radley et al., 1975; Dolan et al., 1980). In Eastern Africa an alternative was

found by way of combining different strains namely T. parva Muguga, T. parva

Kiambu 5 and T. lawrencei Serengeti (transformed), referred to as the Muguga cocktail

30 Literature review

(Radley et al., 1975; Radley, 1981). This trivalent vaccine has been successfully used to

immunise cattle in Malawi, Tanzania, Uganda and Zambia (Musisi et al., 1992).

However, many countries are reluctant to use foreign stocks, as there is a danger of

introducing these parasites. Geysen et al. (1999) have shown that one of the stocks

introduced with the trivalent vaccine and having a different genotype from that of local

parasites was widespread in Southern Zambia. As an alternative, local Theileria stocks

have been isolated and used to immunise cattle in areas like Eastern Province of

Zambia, the coast and central highlands of Kenya and in Zimbabwe (De Castro, 1997).

In Southern Zambia, the work of the Belgian government funded project,

Assistance to the Veterinary Services of Zambia (ASVEZA) was initiated in 1993 to

isolate a local strain that could be used in field immunisations against ECF. This work

started with the evaluation of local strains that had been isolated in the area by F. Musisi

in 1982 – 83 and 1985 (Geysen et al., 1999). These were T. parva Chitongo isolated in

Chitongo village of Namwala district in 1982 and T. parva Mandali isolated in Mandali

area of Monze district in Southern Province in 1985 (Nambota et al., 1997). After a

series of experiments, T. parva Chitongo gave promising results and was selected. This

T. parva stock was first kept as a cell culture at the International Livestock Research

Institute (ILRI), Kenya. The T. parva Chitongo stock has been used in several

experiments and in the field and has proved to confer protection in immunised cattle

against most of the local strains in the province with no documented breakthroughs. In

addition, two Friesian cattle immunised in 1997 with this stock and kept in a paddock

free of ticks in Mazabuka, Zambia, withstood a lethal heterologous challenge with

T. parva Katete, five years after initial immunisation (own unpublished observation).

Large-scale field ECF immunisations using the T. parva Chitongo stock started in 1999

in Mazabuka, Monze and Choma and later in Namwala and Kalomo in 2000. As at the

Chapter 1

31

end of 2002, a total of 37,012 calves had been immunised against ECF in Southern

Zambia (Anonymous, 2002). This programme has since continued.

The purpose of undertaking ECF immunisations is to maintain herd immunity

and achieve a stable state in cattle in areas where an unstable endemic situation would

exist without intervention or where the disease is threatening naive cattle. As

immunized animals remain carriers, immunisation may contribute to attaining and

improving endemic stability in an endemic area (Uilenberg, 1999) and thereby altering

the epidemiology of the disease. Such stability can only be assessed through studies

aimed at evaluating the evolution of the infection prevalence over time. However, such

evaluation has never been done in Zambia before. It is therefore important to determine

the effect of such control strategies on the transmission of the disease in the field.

On a regional basis, the lack or presence of seasonal climatic variability has been

shown to influence the transmission of T. parva because the tick’s survival and

abundance depends on a suitable climate (Sonenshine, 1993). However, no documented

evidence exists on how global climate variability affects the transmission of tick-borne

diseases such as ECF. In this thesis, we investigate the impact of global climate

variabilty on the transmission of T. parva in Zambia.

1.9 El Niño Southern Oscillation and global climate variability

The name El Niño was derived from the appearance of warm water off the coast of

Ecuador and Peru, noticeable around Christmas (El Niño meaning “little boy” refers to

the infant Jesus) (Kovats et al., 2003). The El Niño events were first recognised in the

late 19th century in Peru though they occurred already for millennia (Glantz, 1996).

From time to time the warming exceeds expected variation and persists for 12 to 18

months. This phenomenon occurs at intervals of 2 to 7 years and has been associated

with heavy rainfall and flooding on the west coast of Southern America (Glantz, 2001).

32 Literature review

Walker and Bliss (1932) identified substantial air pressure differences across the

Pacific and argued that this had a large impact on the world’s weather and was also the

main contributor to monsoon rainfall in India. The fluctuations in pressure difference

between Darwin (Australia) and Tahiti is what is termed the Southern Oscillation

(Kovats et al., 2003). It was not until the 1960s that El Niño and the Southern

Oscillation were linked and identified as oceanographic and atmospheric components of

the same phenomenon – i.e. the El Niño Southern Oscillation (ENSO) (Kovats, 2000).

The ENSO phenomenon is the largest natural interannual climate variability. It

has a major impact not only on the climate of the Pacific region where its centre of

action lies, but also on the entire climate system through atmospheric teleconnections

from regions of anomalous tropical heating (Glantz et al., 1991). The weather patterns

associated with ENSO include regional land and sea surface warming, changes in storm

tracks and changes in rainfall patterns, particularly in terms of heavy rain or prolonged

drought (Kovats, 2000). El Niño Southern Oscillation is monitored using the

Multivariate ENSO Index (MEI) based on the six main observed variables over the

tropical Pacific. These six variables are the sea-level pressure, zonal and meridional

components of the surface wind, sea surface temperatures, surface air temperature and

total cloudiness of the sky. The MEI is computed separately for each of the twelve

sliding bi-monthly seasons (Dec/Jan, Jan/Feb, …, Nov/Dec). After spatially filtering the

individual fields into clusters (Wolter, 1987), the MEI is calculated as the first unrotated

Principal Component (PC) of all six observed fields combined. This is done by

normalising the total variance of each field first, and then extracting the first PC on the

covariance matrix of the combined fields (Wolter and Timlin, 1993). All seasonal

values are standardised with respect to each season and to the 1950-1993 reference

period, so as to keep the MEI comparable. The bimonthly MEI values are then

Chapter 1

33

expressed in 1/1000 of standard deviations starting with Dec 1949/Jan 1950 through to

the last month. The standardised MEI values are normally expressed as negative or

positive. Negative values of the MEI represent the cold ENSO phase or La Niña, while

positive MEI value represent the warm ENSO phase (El Niño). The MEI are also

expressed in ranks, where there are 55 numbers in each column, the lowest number (1)

would denote the strongest La Niña case for that bimonthly season, while the highest

number (55) would indicate the strongest El Nino case. If terciles or thirds are used to

define La Niña, near normal and El Niño, MEI ranks 1-18 will denote strong to weak La

Niña, 19- 37 will be normal and 38-55 will denote weak to strong El Niño.

El Niño has been associated with natural disasters in many parts of the world

as it brings changes in risk of droughts, floods and tropical cyclones (Kovats et al.,

2003). During the 1997-1998 El Niño event for example, Southern Africa received

more than the average rainfall (World Meteorological Organisation, 2000), Kenya was

affected by excessive rainfall and flooding, and severe floods and mudslides occurred in

coastal regions of Ecuador and Peru. At the other extreme, Guyana, Indonesia and

Papua New Guinea were severely affected by drought (Kovats, 2000).

The transmission of many human vector-borne infectious diseases is affected

by climate (Kovats et al., 2003). McMichael and Kovats (2000) pointed out that field-

based epidemiological research of the climatic influence on disease requires sufficient

information enabling differentiation between the effects of co-existent climatic and non-

climatic factors. As such there are a number of factors that need to be taken into account

when assessing the scientific plausibility of a relationship between ENSO and health

outcomes. According to Kovats (2000) these factors include evidence that regional or

local climate effects are related to ENSO, that the health outcome is causally affected by

climate variables (rainfall, land surface temperature, sea surface temperature) and the

34 Literature review

evidence of an association between health impacts and climate variables or ENSO

parameter (MEI) with due consideration of confounders. Change in weather events can

have a significant effect on insect reproduction and mortality rates and can thus

influence the overall abundance of these disease vectors (Kovats, 2000). There are well-

reported epidemiological relationships between the presence of El Niño and increased

risk of diseases in human health like malaria and other mosquito-borne diseases in

specific geographic areas where climate anomalies are linked to ENSO cycle (Lindblade

et al., 1999; Kovats, 2000). El Niño therefore provides an opportunity to illustrate the

importance of the ecological basis for many diseases, and therefore, needs to be taken

into account by health professionals, policy makers and the general public (Kovats,

2000).

Chapter 1

35

1.10 References

Allsopp, B. A., Carrington, M., Baylis, H., Sohal, S., Dolan, T. T. and Iams, K. (1989).

Improved characterization of Theileria parva isolates using the polymerase

chain reaction and oligonucleotide probes. Molecular and Biochemical

Parasitology 35, 137-148.

Anonymous (1974). Annual Report of the Department of Veterinary and Tsetse

Services for the year 1972. Government Printer, Lusaka, Zambia.

Anonymous (1980). Annual Report for the Department of Veterinary and Tsetse

Control Services for the year 1977. Government Printer, Lusaka, Zambia.

Anonymous (1989). Nomenclature in Theileria. In: Theileriosis in Eastern, Central and

Southern Africa. Proceedings of a workshop on East Coast fever Immunisation

held in Lilongwe, Malawi, 20-22 September 1988. Dolan, T.T. (Editor), pp. 182-

186.

Anonymous (2002). Assistance to the Veterinary Services of Zambia (ASVEZA)

Annual internal report for the year 2002.

Baldwin, C. L., Malu, M. N., Kinuthia, S. W., Conrad, P. A. and Grootenhuis, J. G.

(1986). Comparative analysis of infection and transformation of lymphocytes

from African buffalo and Boran cattle with Theileria parva parva and Theileria

parva lawrencei. Infection and Immunity 53, 186-191.

Baldwin, C. L., Goddeeris, B. M. and Morrison, W. I. (1987). Bovine helper T-cell

clones specific for lymphocytes infected with Theileria parva (Muguga).

Parasite Immunology 9, 499-513.

Barnett, S. F. (1957). Theileriosis Control. Bulletin of Epizootic Diseases of Africa 5,

343-357.

36 Literature review

Berkvens, D. L. (1990). A study of the ecology of Rhipicephalus appendiculatus

complex with reference to the Eastern Province of Zambia. PhD thesis. Brunel

University, Uxbridge, UK.

Berkvens, D. L., Pegram, R. G. and Brandt J. R. (1995). A study of the diapausing

behaviour of Rhipicephalus appendiculatus and R. zambeziensis under quasi-

natural conditions in Zambia. Medical and Veterinary Entomology 9, 307-315.

Berkvens, D. L., Geysen, D. M., Chaka, G., Madder, M. and Brandt, J. R. (1998). A

survey of the ixodid ticks parasitising cattle in the Eastern Province of Zambia.

Medical and Veterinary Entomology 12, 234-240.

Billiouw, M. (2005). The Epidemiology of bovine theileriosis in the eastern province of

Zambia. PhD thesis. Ghent University, Ghent, Belgium.

Billiouw, M., Brandt, J., Vercruysse, J., Speybroeck, N., Marcotty, T., Mulumba, M.

and Berkvens, D. (2005). Evaluation of the Indirect Fluorescent Antibody Test

as diagnostic tool for East Coast fever in eastern Zambia. Veterinary

Parasitology 127, 189-198.

Brown, C. G. D., Hunter, A. G. and Luckins, A. G. (1990). Diseases caused by

protozoa. In: Sewell, M.M.H. and Brocklesby, D.W. (Editors). Handbook on

Animal Diseases in the Tropics. Bailliere Tindal, London, pp. 161-226.

Burridge, M. J., Morzaria, S. P., Cunningham, M. P. and Brown, C. G. D. (1972).

Duration of immunity to East Coast fever (Theileria parva infection of cattle).

Parasitology 64, 511-515.

Burridge, M. J. and Kimber, C. D. (1972). The Indirect Fluorescent Antibody Test for

experimental East Coast fever (Theileria parva infection of cattle): evaluation of

a cell culture schizont antigen. Research in Veterinary Science 13, 451-455.

Chapter 1

37

Burridge, M. J., Brown, C. G. D. and Kimber, C. D. (1974). Theileria annulata: Cross-

reactions between a cell culture schizont antigen and antigens of East African

species in the indirect fluorescent antibody test. Experimental Parasitology 35,

374-380.

Central Statistical Office (CSO, Zambia) (2000). Agriculture Analytical Report for the

2000 census of population and housing. pp 1-3.

Chaka, G. (2001) Eco-physiology of R. appendiculatus (Acari, Ixodidae) in Eastern

Zambia. Ph.D. thesis. Institute of Tropical Medicine, Antwerp, Belgium.

Chizyuka, H. G. B. and Mangani, M. P. C. (1985). Theileriosis in Zambia. In:

Immunisation against Theileriosis in Africa. Proceedings of a joint workshop

sponsored by the International Laboratory for Research in Animal Diseases

(ILRAD) and the Food and Agriculture Organisation of the United Nations

(F.A.O.), held in Nairobi, Kenya. 1-5 October 1984, Irvin, A.D. (Editor), pp. 41-

44.

Conrad, P. A., Iams, K., Brown, W. C., Sohanpal, B. and Ole-Moi Yoi, O. (1987). DNA

probes detect genomic diversity in Theileria parva stocks. Molecular and

Biochemical Parasitology 25, 213-230.

Conrad, R. E., Chen, P. and Dolan, T. T. (1989). Detection of Theileria parva in the

salivary glands of infected Rhipicephalus appendiculatus ticks with DNA

probes. In: Proceedings of Eighth Veterinary Hemoparasite Disease Conference,

held in St. Louis, USA, 10-12 April, pp. 10-12.

Corliss, J. O. (1984). The Kingdom of Prostista and its 45 phyla. Biosystems 17, 87-

126.

38 Literature review

Cunningham, M. P., Brown, C. G. D., Burridge, M. J. and Purnell, R. E. (1973).

Cryopreservation of infective particles of Theileria parva. International Journal

for Parasitology 3, 583-587.

Cunningham, M. P. (1977). Immunization of Cattle Against Theileria parva. In:

Immunity to Blood Parasites of Animals and Man. L.H. Miller, J. A. Pino, and

J.J. McKelvey, Jr. (Editors). New York: Plenum Press, pp. 189-207.

De Castro, J. J. (1997). Sustainable tick and tick-borne diseases control in livestock

improvement in developing countries. Veterinary Parasitology 71, 77-97.

D’Haese, L., Penne, K. and Elyn, R. (1999). Economics of theileriosis control in

Zambia. Tropical Medicine and International Health 4, A49-A57.

Dolan, T. T., Radley, D. E., Brown, C. G. D., Cunningham, M. P., Morzaria, S. P. and

Young, A. S. (1980). East Coast fever. IV. Further studies on the protection of

cattle immunised with a combination of Theilerial strains. Veterinary

Parasitology 6, 325-332.

Dolan, T. T., Young, A. S., Losos, G. J., McMillan, I., Minder, C. H. E. and Soulsby, K.

(1984a). Dose dependent responses of cattle to Theileria parva stabilate

infection. International Journal for Parasitology 14, 89-95.

Dolan, T. T., Linyonyi, A., Mbogo, S. K. and Young, A. S. (1984b). Comparison of

long-acting oxytetracycline and parvaquone in immunisation against East Coast

fever by infection and treatment. Research in Veterinary Science 37, 175-178.

Dolan, T. T. (1999). Dogmas and misunderstandings in East Coast fever. Tropical

Medicine and International Health 4, A3-A11.

Duffus, W. P. H. and Wagner, G. G. (1980). Comparison between certain serological

tests for diagnosis of East Coast fever. Veterinary Parasitology 6, 313-324.

Chapter 1

39

Emery, D. L. (1981). Adoptive transfer of immunity to infection with Theileria parva

(East Coast fever) between cattle twins. Research in Veterinary Science 30, 364-

367.

Eugui, E. M. and Emery, D. L. (1981). Genetically restricted cell-mediated cytoxicity in

cattle immune to Theileria parva. Nature 290, 251-254.

Fawcett, D. W., Doxsey, S. and Buscher, G. (1981). Salivary glands of the tick vector of

East Coast fever: I. Ultrastructure of type III acinus. Tissue Cell 13, 209-230.

Fawcett, D. W., Doxsey, S and Buscher, G. (1982). Salivary glands of the tick vector of

East Coast fever: IV. Cell type selectivity and host responsiveness to Theileria

parva. Tissue cell 14, 397-414.

Food and Agriculture Organisation of the United Nations (F.A.O.) (1984). Ticks and

Tick-borne Diseases Control. A Practical Field manual, Volume II, pp. 457-507.

Food and Agriculture Organisation of the United Nations (F.A.O.) (2004). F.A.O.

Production Year Book.

Geysen, D., Bishop, R., Skilton, R., Dolan, T. T. and Morzaria, S. (1999). Molecular

epidemiology of Theileria parva in the field. Tropical Medicine and

International Health 4, A21-A27.

Glantz, M. H., Katz, R. W. and Nicholls, N. (1991). Teleconnection linking worldwide

climate anomalies. Cambridge University Press, UK.

Glantz, M. H. (1996). Currents of change: El Niño’s impact on climate and society.

Cambridge: Cambridge University Press, UK.

Glantz, M. H. (2001). Once burned, twice shy? Lessons learned from the 1997/98 El

Niño. United Nations University Press, Tokyo, Japan.

40 Literature review

Goddeeris, B. M., Morrison, W. I., Teale, A. J., Bensaid, A. and Baldwin, C. L. (1986).

Bovine cytotoxic T-cell clones specific for cells infected with the protozoan

parasite Theileria parva: Parasite strain specific and class I major

histocompatibility complex restriction. Proceedings of the National Academy of

Science 83, 5238-5242.

Gray, C. E. and Robertson, W. (1902). Report on the Texas Fever or Redwater in

Rhodesia. Argus Printing & Publishing, Cape Town, p. 26.

Grootenhuis, J. G. (1989). The role of wildlife in the epidemiology of cattle theileriosis.

In: Theileriosis in Eastern, Central and Southern Africa. Proceedings of a

workshop on East Coast fever Immunisation held in Lilongwe, Malawi, 20 – 22

September 1988, Dolan, T.T. (Editor), pp. 129 – 136.

Hulliger, L., Wilde, J. K. H., Brown, C. G. D. and Turner, L. (1964). Mode of

multiplication of Theileria in cultures of bovine lymphocytic cells. Nature 203,

728-730.

Irvin, A. D., Boarer, C. D. H., Dobbelaere, D. A. E., Mahan, S. M., Masake, R. and

Ocama, J. G. R. (1981). Monitoring Theileria parva infection in adult

Rhipicephalus ticks. Parasitology 82, 137-147.

Irvin, A. D. and Mwamachi, D. M. (1983). Clinical and diagnostic features of East

Coast fever (Theileria parva infection of cattle). Veterinary Record 113, 192-

198.

Irvin, A. D. (1987). Characterization of species and strains of Theileria. Advances in

Parasitology 26, 145-197.

Chapter 1

41

Irvin, A. D. and Morrison, W. I. (1987). Immunopathology, immunology and

immunoprophylaxis of Theileria infections. In: Immune Responses in Parasitic

Infections: Immunology, Immunopathology and Immunoprophylaxis, Volume

III, Soulsby E.J.L. (Editor), CRC Press, Inc. Boca Raton, Florida, pp. 223-74.

Kaba, S. A., Musoke, J. M., Schaap, D., Schetters, T., Rowlands, J., Vermeulen, A. N.,

Nene, V., Vlak, J. M. and van Oers, M. M. (2005). Novel baculovirus-derived

p67 subunit vaccines efficacious against East Coast fever in cattle. Vaccine 23,

2791-2800.

Katende, J., Morzaria, S., Toye, P., Skilton, R., Nene, V., Nkonge, C. and Musoke, A.

(1998). An enzyme-linked immunosorbent assay for detection of Theileria

parva antibodies in cattle using a recombinant polymorphic immunodominant

molecule. Parasitology Research 84, 408-416.

Koch, H. T., Kambeva, L., Norval, R. A. I., Ocama, J. G. R., Masaka, S., Munatswa, F.

C., Honhold, N. and Irvin, A. D. (1990). Age resistance to Theileria parva bovis

infection in calves. Veterinary Parasitology 37, 197-206.

Kovats, R. S. (2000). El Niño and human health. Bulletin of the World Health

Organization 78(9).

Kovats, R. S., Bouma, M. J., Hajat, S., Worrall, E. and Haines, A. (2003). El Niño and

health. Lancet 362, 1481-1489.

Lawrence, J. A. (1979). The differential diagnosis of the bovine Theilerias of Southern

Africa. Journal of the South African Veterinary Association 50, 311-313.

Lawrence, J. A. (1985). Theileria parva in Southern Africa: Past and Present. In:

Immunisation against Theileriosis in Africa. Proceedings of a joint workshop

sponsored by ILRAD and F.A.O., held in Nairobi, Kenya, 1-5 October 1984,

Irvin, A.D. (Editor), pp. 123-124.

42 Literature review

Lawrence, J. A. (1991). Retrospective observations on the geographical relationship

between Rhipicephalus appendiculatus and East Coast fever in Southern Africa.

Veterinary Record 128, 180-183.

Lawrence, J. A. (1992). History of Bovine Theileriosis in Southern Africa. In: The

Epidemiology of Theileriosis in Africa, Norval, R.A. I., Perry, B.D and Young,

A. S. (Editors), Academic Press, London, pp. 1-39.

Leitch, B. L. and Young, A.S. (1981). Theileria infections in Rhipicephalus

appendiculatus ticks collected in the field. In: Advances in the Control of

Theileriosis. Proceedings of an International Conference held at ILRAD,

Nairobi, 9-13 February, 1981. Irvin, A. D., Cunningham, M. P. and Young, A.

S. (Editors), Martinus Nijhoff Publishers, The Hague, pp. 63-65.

Lessard, P., L’Eplattenier, R., Norval, R. A. I., Kundert, K., Dolan, T. T., Croze, H.,

Walker, J. B., Irvin, A. D. and Perry, B. D. (1990). Geographical information

systems for studying the epidemiology of cattle diseases caused by Theileria

parva. Veterinary Record 126, 255-262.

Levine, N. D., Corliss, J. O., Cox, F. E. G., Deroux, G., Grain, J., Honigberg, B. N.,

Leedale, G. F., Loeblich, A. R. II., Lom, J., Lynn, D. H., Nerinfield, F. G., Page,

F. C., Poljansky, G., Sprague, V., Vaura, J. and Wallace, F. G. (1980). A newly

revised classification of the protozoa. Journal of Protozoology 27, 37-38.

Levine, N. D. (1988). The Protozoan Phylum Apicomplexa. CRC Press, Boca Raton,

Volumes I and II.

Lindblade, K. A., Walker, E. D., Onapa, A. W., Katungu, J. and Wilson, M. L. (1999).

Highland malaria in Uganda: prospective analysis of an epidemic associated with

El Niño. Transactions of the Royal Society of Tropical Medicine and Hygiene 93,

480-487.

Chapter 1

43

Lounsbury, C. P. (1904). Transmission of African Coast fever. Agricultural Journal of

the Cape of Good Hope 24, 428 – 432.

Malmquist, W. A., Nyindo, M. B. A. and Brown, C. G. D. (1970). East Coast fever:

Cultivation in vitro of bovine spleen cell lines infected and transformed by

Theileria parva. Tropical Animal Health and Production 2, 139-145.

Mbogo, S. K., Kariuki, D. P., Ngumi, P. N. and Mchardy, N. (1996). A mild Theileria

parva parasite with potential for immunisation against East Coast fever.

Veterinary Parasitology 61, 41-47.

McKeever, D. J., Taracha, E. L. N., Morrison, W. I., Musoke, A. J. and Morzaria, S. P.

(1999). Protective mechanisms against Theileria parva: Evolution of vaccine

development strategies. Parasitology Today 15, 263-267.

McMichael, A. J. and Kovats, R. S. (2000). Strategies for assessing health impacts of

global environmental change. In: Implementing ecological integrity: restoring

regional and global environmental and human health, Crabbé, P. et al. (Editors).

Kluwer Academic Publishers, Dordrecht, pp. 217-231.

Medley, G. F., Perry, B. D. and Young, A. S. (1992). Preliminary analysis of the

transmission dynamics of Theileria parva in eastern Africa. Parasitology 106,

251-264.

Moll, G., Lohding, A. and Young, A. S. (1984). Epidemiology of theileriosis in the

Trans-Mara Division, Kenya. Husbandry and disease background and

preliminary observation on theileriosis in calves. Preventive Veterinary

Medicine 2, 801-831.

Moll, G., Lohding, A., Young, A. S. and Leitch, B. L. (1986). Epidemiology of

theileriosis in calves in an endemic area of Kenya. Veterinary Parasitology 19,

255-273.

44 Literature review

Morzaria, S. P., Irvin, A. D., Wathanga, J., D’Souza, D., Katende, J., young, A. S.,

Scott, J. and Gettinby, G. (1988). The effect of East Coast fever immunisation

and different acaricidal treatments on the productivity of beef cattle. Veterinary

Record 123, 313-320.

Muhammed, S. I., Lauerman, L. H. and Johnson, L. W. (1975). Effect of humoral

antibodies on the course of Theileria parva (East Coast fever) of cattle.

American Journal of Veterinary Research 36, 399-402.

Mukhebi, A. W., Perry, B. D. and Kruska, R. (1992). Estimated economics of

theileriosis control in Africa. Preventive Veterinary Medicine 12, 73-85.

Mulumba, M., Speybroeck, N., Billiouw, M., Berkvens, D. L., Geysen, D. M. and

Brandt, J. R. A. (2000). Transmission of theileriosis in the traditional farming

sector in the Southern Province of Zambia during 1995-1996. Tropical Animal

Health and Production 32, 303-314.

Mulumba, M., Speybroeck, N., Berkvens, D. L., Geysen, D. M. and Brandt, J. R. A.

(2001). Transmission of Theileria parva in the traditional farming sector in the

Southern Province of Zambia during 1997-1998. Tropical Animal Health and

Production 33, 117-125.

Musisi, F. L. and Peirce, M. A. (1981). Preliminary observations on theileriosis in

Zambia. In: Advances in the control of Theileriosis, Irvin, A.D., Cunningham,

M.P. and Young, A.S. (Editors), Martinus Nijhoff Publishers, The Hague, pp.

76-78.

Chapter 1

45

Musisi, F. L., Quiroga, J. C., Mutugi, J. J., Jacobsen, J., De Castro, J. J. and Di Guilio,

G. (1992). Immunisation against East Coast fever: Recent experiences with the

trivalent vaccine in East and Central Africa. In: Bovine Theileriosis, 3rd

Symposium on Tropical and Animal Health and Production. Paling, R. W.

(Editor), Office for International Co-operation, Utrecht, pp. 26-32

Musoke, A. J., Nantulya, V. M., Buscher, G., Masake, R. A. and Otim, B. (1982).

Bovine immune response to Theileria parva: neutralizing antibodies to

sporozoites. Immunology 45, 663-668.

Musoke, A., Morzaria, S., Nkonge, C., Jones, E. and Nene, V. (1992). A recombinant

sporozoite surface antigen of Theileria parva induces protein in cattle.

Proceedings of the National Academy of Sciences 89, 514-518.

Musoke, A., Rowlands, J., Nene, V., Nyanjui, J., Katende, J., Spooner, P., Mwaura, S.,

Odongo, D., Nkonge, C., Mbogo, S., Bishop, R. and Morzaria, S. (2005).

Subunit vaccine based on the p67 major surface protein of Theileria parva

sporozoites reduces severity of infection derived from field tick challenge.

Vaccine 23, 3084-3095.

Nambota, A. (1989). Immunisation against theileriosis in the Southern Province of

Zambia. In: Theileriosis in Eastern, central and Southern Africa. Proceedings of

a workshop on East Coast fever Immunisation, held in Lilongwe, Malawi. 20-22

September 1988, Dolan, T.T. (Editor), pp. 87-89.

Nambota, A., Samui, K., Sugimoto, C., Kakuta, T. and Onuma, M. (1994). Theileriosis

in Zambia: Etiology, Epidemiology and Control measures. Japanese Journal of

Veterinary Research 42, 1 - 18.

46 Literature review

Nambota, A. M., Lovelace, C. E. A., Chitambo, H., Kakuda, T., Sugimoto, C. and

Onuma, M. (1997). Characterisation of some Theileria parva stocks from

Zambia using Monoclonal Antibodies. Journal of Veterinary Medical Science

59, 1- 4.

Ndungu, S. G., Ngumi, P. N., Mbogo, S. K., Dolan, T. T., Mutugi, J. J. and Young, A.

S. (2005a). Some preliminary observations on the susceptibility and resistance

of different cattle breeds to Theileria parva infection. Onderstepoort Journal of

Veterinary Research 72, 7-11.

Ndungu, S. G., Brown, C. G. D. and Dolan, T. T. (2005b). In vivo comparison of

susceptibility between Bos indicus and Bos Taurus cattle types to Theileria

parva infection. Onderstepoort Journal of Veterinary Research 72, 13-22.

Neitz, W. O. (1953). Aureomycin in Theileria parva infection. Nature 171, 34-35.

Neitz, W. O. (1957). Theileriosis, Gonderioses and Cytauxzoonoses: a review.

Onderstepoort Journal of Veterinary Research 27, 275-430.

Norval, R. A. I., Lawrence, J. A., Young, A. S., Perry, B. D., Dolan, T. T. and Scott, J.

(1991). Theileria parva: Influence of vector parasite and host relationship on the

epidemiology of theileriosis in Southern Africa. Parasitology 102, 347-356.

Norval, R. A. I., Perry, B. D. and Young, A. S. (1992). The Epidemiology of

Theileriosis in Africa. Academic Press, London.

Ochanda, H., Young, A. S., Mutugi, J. J., Mumo, J. and Omwoyo, P. L. (1988). The

effect of temperature on the rates of transmission of Theileria parva infection to

cattle by its tick vector Rhipicephalus appendiculatus. Parasitology 97, 239-245.

Ochanda, H., Young, A. S., Wells, C., Medley, G. F. and Perry, B. D. (1996).

Comparison of the transmission of Theileria parva between different instars of

Rhipicephalus appendiculatus. Parasitology 113, 243-253.

Chapter 1

47

Ochanda, H., Young, A. S., Medley, G. F. and Perry, B. D. (1998). Vector competence

of 7 Rhipicephalid tick stocks in transmitting 2 Theileria parva parasite stocks

from Kenya and Zimbabwe. Parasitology 116, 539 – 545.

Ochanda, H., Young, A. S. and Medley, G. F. (2003). Survival of Theileria parva in its

nymphal tick vector, Rhipicephalus appendiculatus, under laboratory and quasi

natural conditions. Parasitology 126, 571-576.

Ogden, N. H., Gwakisa, P. Swai, E., French, N. P., Fitzpatrick, J., Kambarage, D. and

Bryant, M. (2003). Evaluation of PCR to detect Theileria parva in field-

collected tick and bovine samples in Tanzania. Veterinary Parasitology 112,

177-183.

Paling, R. W. and Geysen, D. (1981). Observations on Rwandan strains of Theileria

parva and the value of Theileria parva (Nyakizu) as a possible vaccine strain.

In: Advances in the Control of Theileriosis: Proceedings of an International

Conference Held at ILRAD, Nairobi, 9 – 13 February 1981, Irvin, A.D.,

Cunningham, M.P. and Young, A.S. (Editors), Martinus Nijhoff Publishers, The

Hague, pp. 238 – 241.

Pearson, T. W., Dolan, T. T., Stag, D. A. and Lundin, L. B. (1979) Cell-mediated

immunity to Theileria-transformed cell lines. Nature 281, 678-680.

Pearson, T. W., Hewett, R. S., Roelants, G. E., Stagg, D. A. and Dolan, T. T. (1982).

Studies on the induction and specificity of cytotoxicity to Theileria-transformed

cell lines. The Journal of Immunology 128, 2509-2513.

Pegram, R. G. and Banda, D. S. (1990). Ecology and phenology of cattle ticks in

Zambia: development and survival of free-living stages. Experimental and

Applied Acarology 8, 291-301.

48 Literature review

Purnell, R. E., Ledger, M. A., Omwoyo, P. L., Payne, R. C. and Peirce, M. A. (1974).

Theileria parva: variation in the infection rate of the vector tick, Rhipicephalus

appendiculatus. International Journal for Parasitology 4, 513-517.

Radley, D. E., Brown, C. G. D., Burridge, M. J., Cunningham, M. P., Peirce, M. A. and

Purnell, R. E. (1974). East Coast fever: Quantitative studies of Theileria parva

in cattle. Experimental Parasitology 36, 278-287.

Radley, D. E., Brown, C. G. D., Burridge, M. J., Cunningham, M. P., Kirimi, I. M.,

Purnell, R. E. and Young, A. S. (1975). East Coast fever. 1. Chemoprophylactic

immunisation of cattle against Theileria parva (Muguga) and five theilerial

strains. Veterinary Parasitology 1, 35-41.

Radley, D. E. (1981). Infections and treatment immunisation against theileriosis. In:

Advances in the Control of Theileriosis. Proceedings of an International

Conference held at ILRAD, Nairobi, 9-13 February, 1981, Irvin, A. D.,

Cunningham, M. P. and Young, A. S. (Editors), Martinus Nijhoff Publishers,

The Hague, pp. 227-236.

Randolph, S. E. (1993). Climate, satellite imagery and the seasonal abundance of the

tick Rhipicephalus appendiculatus in southern Africa: a new perspective.

Medical and Veterinary Entomology 7, 243-258.

Rechav, Y. (1981). Ecological factors affecting seasonal activity of the brown ear tick

Rhipicephalus appendiculatus. In: Tick Biology and Control. Proceedings of an

International Conference Held in Grahamstown, 27-29 January, 1981,

Whitehead, G. B. and Gibson, J. D. (Editors), Tick Research Unit Rhodes

University, Grahamstown, pp 187-192.

Sauer, J. R., McSwain, J. L., Bowman, A. S. and Essenberg, R. C. (1995). Tick salivary

gland physiology. Annual Review of Entomology 40, 245-267.

Chapter 1

49

Shaw, M. K., Tilney, L. G. and Musoke, A. J. (1991). The entry of Theileria parva

sporozoites into bovine lymphocytes: Evidence of MHC class I involvement.

Journal of Cell Biology 113, 87-101.

Shaw, M. K. and Young, A. S. (1994). The biology of Theileria species in Ixodid ticks

in relation to the parasite transmission. Advances in Disease Vector Research

10, 23-63.

Sonenshine, D. E. (1993). Biology of ticks. Volume II. Oxford University Press, New

York.

Sonenshine, D. E. and Mather, T. N. (1994). Ecological Dynamics of Tick-Borne

Zoonoses. Oxford University Press. New York.

Speybroeck, N., Belot, J., Madder, M., Mulumba, M, Brandt, J.R.A.., Geysen, D.M.,

Sinyangwe, P., Geerts, S. and Berkvens, D.L. (1999) A spatial representation of

the point prevalence of Theileria parva and the distribution of Rhipicephalus

appendiculatus during the period 1995-1998 in the Southern Province of

Zambia. In: Proceedings of the 7th annual meeting of the Flemish Society for

Veterinary Epidemiology and Economics, Laevens, H. (Editor). Antwerp,

Belgium. pp. 156-166.

Thomson, J. W. (1985). Theileriosis in Zimbabwe. In: Immunisation against

Theileriosis in Africa. Proceedings of a joint workshop sponsored by ILRAD

and F.A.O. held in Nairobi, Kenya. 1-5 October, 1984, Irvin, A.D. (Editor), pp.

48-57.

Uilenberg, G. (1976). Tick-borne livestock diseases and their vectors. 2. Epizootiology

of tick-borne diseases. World Animal Review 17, 8-15.

Uilenberg, G. (1999). Immunnization against diseases caused by Theileria parva: a

review. Tropical Medicine and International Health 4, A12-A20.

50 Literature review

Wagner, G. G., Duffus, W. P. H. and Burridge, M. J. (1974). The specific

immunoglobulin response in cattle immunised with isolated Theileria parva

antigens. Parasitology 69, 43-53.

Walker, G. T. and Bliss, E. W. (1932). World Weather. V. Mem Royal Meteological

Society 4, 53-84.

Walker, J. B., Norval, R. A. I. and Corwin, M. D. (1981). Rhipicephalus zambeziensis

sp. nov., a new tick from eastern and southern Africa, together with a

redescription of Rhipicephalus appendiculatus Neumann, 1901 (Acarina,

Ixodidae). Onderstepoort Journal of Veterinary Research 48, 87-104.

Walker, A. R., Fletcher, J. D. and Gill, H. S. (1985). Structural and histochemical

changes in the salivary glands of Rhipicephalus appendiculatus during feeding.

International Journal for Parasitology 15, 81-100.

Walker, A. R. (1990). Parasitic adaptations in the transmission of Theileria by ticks – a

review. Tropical Animal Health and Production 22, 23-33.

Walker, J. B., Keirans, J. E. and Horak, I. G. (2000). The genus Rhipicephalus (Acari:

Ixodidae). A guide to the Brown Ticks of the World. Cambridge University

Press.

Watkins-Pitchford, H. (1909). Dipping and tick destroying agents. Natal Agricultural

Journal 12, 436-459.

Wilde, J. K. H. (1967). East Coast fever. Advances in Veterinary Science 11, 207-259.

Wolter, K. (1987). The Southern Oscillation in surface circulation and climate over the

tropical Atlantic, Eastern Pacific, and Indian Oceans as captured by cluster

analysis. Journal of Climate and Applied Meteorology 26, 540-558

Chapter 1

51

Wolter, K. and Timlin, M. S. (1993). Monitoring ENSO in COADS with a seasonally

adjusted principal component index. In: Proceedings of the 17th Climate

Diagnostics Workshop, Norman, OK, NOAA/N MC/CAC, NSSL, Oklahoma

Clim. Survey, CIMMS and the School of Meteor., Univ. of Oklahoma, pp. 52-

57.

World Meteorological Organisation (WMO No. 905) with UNESCO, UNEP and ICSU

(2000). The 1997-1998 El Niño event: a scientific and technical retrospective.

Yeoman, G. H. (1966). Field vector studies of epizootic East Coast fever. I. A

quantitative relationship between R. appendiculatus and the epizootics of East

Coast fever. Bulletin of Epizootic Diseases of Africa 14, 113-140.

Young, A.S. (1981). The epidemiology of theileriosis in East Africa. In: Advances in

the Control of Theileriosis. Proceedings of an International Conference held at

ILRAD, Nairobi, 9-13 February, 1981, Irvin, A. D., Cunningham, M. P. and

Young, A. S. (Editors), Martinus Nijhoff Publishers, The Hague, pp. 38-55.

Young, A. S. and Leitch, B. L. (1981). Epidemiology of East Coast fever: Some effects

of temperature on the development of Theileria parva in the tick vector,

Rhipicephalus appendiculatus. Parasitology 83, 199-211.

Young, A. S., Groocock, C. M. and Kariuki, D. P. (1988). Integrated control of ticks

and tick-borne diseases of cattle in Africa. Parasitology 96, 403-441.

Young, A. S., Dolan, T. T., Morzaria, S. P., Mwakima, F. N., Norval, R. A. I., Scott, J.,

Sherriff, A. and Gettinby, G. (1996). Factors influencing infections in

Rhipicephalus appendiculatus ticks fed on cattle infected with Theileria parva.

Parasitology 113, 255-266.

52 Objectives of thesis

Chapter 2

Objectives of the thesis The general objectives of this thesis were to study the infection dynamics of ECF and

host factors in experimental T. parva infections in order to further optimize the ECF

immunisation programme in Southern Zambia.

In order to achieve these general objectives, the study comprises the following

specific objectives:

• Investigate the association between T. parva sero-prevalence and global climate

variability and/or the ECF immunisations based on a T. parva transmission

dynamics study in traditionally kept cattle in Southern Zambia.

• Investigate alternative indicators to better predict case fatality in T. parva

infected animals during clinical ECF.

• Investigate the perception of cattle farmers on the effect of the ECF

immunisations on calf mortality and identify the factors affecting the

acceptability of immunisation in Southern Zambia in order to make

immunisation campaigns more sustainable.

Chapter 3

53

Chapter 3

Theileria parva seroprevalence in traditionally kept cattle in Southern

Zambia and El Niño

Adapted from:

Fandamu, P., Duchateau, L., Speybroeck, N., Marcotty, T., Mbao, V., Mtambo, J., Mulumba, M. and Berkvens, D. (2005). Theileria parva

seroprevalence in traditionally kept cattle in Southern Zambia and El Niño. International Journal for Parasitology 35, 391-396.

54 T. parva sero-prevalence and El Niño

3.1 Introduction

East Coast fever (ECF), caused by the protozoan Theileria parva and transmitted by the

three-host ixodid tick Rhipicephalus appendiculatus, is economically the most

important tick-borne disease of cattle in eastern, central and southern Africa (Mukhebi

et al., 1992). The occurrence of T. parva in a particular area is influenced by the

abundance and distribution of the main vector, R. appendiculatus and the presence of

cattle, the main hosts. Knowledge of the distribution of the vector and disease is crucial

in assessing the extent and potential of the disease problem and in planning control

strategies (Lessard et al., 1990).

Serological surveys for tick-borne diseases like ECF have been used in many parts

of Africa to determine the extent of the disease and also disease pressure in a particular

area (Norval et al., 1985; Gitau et al., 1997; Maloo et al., 2001). In recent years, the

effects of global weather change on the epidemiology of vector borne diseases like

malaria have started receiving increasing attention (Bouma and Van der Kaay, 1996).

East Coast fever, like malaria, is likely to be sensitive to climatic changes especially in

areas of low endemicity. Indeed, in these areas the cattle population lacks protective

immunity and is prone to outbreaks when climatic conditions facilitate transmission.

Before this study, no T. parva sero-prevalence surveys covering all districts of

Southern Zambia had been reported. This paper presents results of large scale

serological surveys carried out twice per year from 1995 to 2002 in Southern Zambia.

The aim of the study was to study the evolution of T. parva sero-prevalence in

traditionally kept cattle in Southern Zambia and attempt to link the changes to climatic

factors and immunization. The study of the vector, R. appendiculatus phenology and

Chapter 3

55

distribution during the same period in the area has already been reported (Speybroeck et

al., 2002).

3.2 Materials and Methods

3.2.1 Areas of study and surveys

The Southern Province of Zambia lies between 25o10’ and 28o50’ East and

15o14’ and 18o South covering a total area of 83,000 km2. It is a plateau around 1200 m

altitude, bounded to the south and the east by the Zambezi River Valley (600 m

altitude), northward by the Kafue River Valley (900 m altitude) and westward by Kafue

National Park and Game Management Areas. The surveys were carried out in 52

veterinary camps (veterinary administrative divisions of a district) in Choma, Kalomo,

Livingstone, Mazabuka, Monze, Namwala, Gwembe, Siavonga and Sinazongwe

districts (Figure 3.1).

Surveys were organised in close collaboration with the Department of

Veterinary and Livestock Development and carried out by the Veterinary Assistants

(VA). Surveys were conducted in the months of March and September to incorporate

the respective effects of peaks in adult and nymphal R. appendiculatus challenges. The

aim of the surveys was to estimate the sero-prevalence as precisely as possible, taking

into account the laboratory capacity to perform the tests. Logistics prompted the choice

of the veterinary camp as the basic sampling unit. A maximum of 50 adult cattle per

veterinary camp (min = 23, mean = 48.97) were randomly sampled each time at the 52

veterinary camps twice per year (March and September) for eight years. A total of

27,526 cattle were sampled during the study period. The random sampling ensured that

the probability of re-sampling the same animals remained minimal, although the

occurrence of re-sampling cannot be excluded. Blood was collected from an auricular

56 T. parva sero-prevalence and El Niño

vein of each animal and immediately blotted on an area of more than 2 cm in diameter

on a Whatman filter paper (Whatman International, England, No. 4) and air-dried in the

shade for half an hour. The filter papers were then stored in small plastic bags filled

with Silicagel® and sent to the laboratory where they were kept at -20oC. The

maximum storage period for the plastic bags with filter papers at room temperature was

4 days.

3.2.2 Serological tests

The Indirect Fluorescent Antibody test (IFAt) (Burridge and Kimber, 1972) was

used on eluates from dried blood spots on filter papers to test for T. parva antibodies

using schizont antigen. Individual disks of dried blood of 5 mm were allowed to elute in

80 microlitres of PBS and shaken at room temperature for 1 hour in a microtitre plate.

Two dilutions (1/40 and 1/160, Mulumba et al., 2000) were used and the tests were

performed three times for each sample. Samples were considered positive when they

showed a positive reaction three times at 1/40 or once at 1/160.

3.2.3 Meteorological data

Multiple El Niño Southern Oscillation Index (MEI) values were retrieved from the

National Oceanic and Atmospheric Administration (NOAA) website,

http://www.cdc.noaa.gov/people/klaus.wolter/MEI/rank.html. El Niño Southern

Oscillation (ENSO) is used to describe both warm (El Niño) and cool (La Niña) ocean

atmospheric events in the tropical Pacific

(http://www.ggweather.com/enso/glossary.htm). The calculation of MEI is based on six

observed variables over the tropical Pacific

(http://www.cdc.noaa.gov/people/klaus.wolter/MEI/mei.html). The variables are the sea

surface temperature (S), surface air temperature (A), sea-level pressure (P), zonal (U)

Chapter 3

57

and meridional (V) components of the surface wind, and the total cloudness fraction of

the sky (C). The MEI is worked out for each of the sliding bi-monthly seasons (Dec/Jan,

Jan/Feb, …., Nov/Dec). The MEI is then calculated as the first unrotated Principal

Component (PC) of the six observed fields combined after spatially filtering the

individual fields into clusters (Wolter, 1987). The total variance of each of the fields is

first normalised and thereafter, the extraction of the first PC on the covariance matrix of

the combined fields is performed (Wolter and Timlin, 1993). All seasonal values are

standardized with respect to each season and to the 1950-93 reference period, so as to

keep the MEI comparable

(http://www.cdc.noaa.gov/people/klaus.wolter/MEI/mei.html). Multiple El Niño

Southern Oscillation Index values from 1 to 18 denote strong to weak La Niña

conditions, while 38-55 denote weak to strong El Niño conditions

(http://www.cdc.noaa.gov/people/klaus.wolter/MEI/mei.html).

3.2.4 Data analysis

The data were analysed with a logistic regression using Stata 8 (StataCorp., 2003).

The outcome variable was individual animal seropositivity, coded as 1 = positive and 0

= negative. Effects were expressed as odds ratios. Independent variables were district

(Choma, Gwembe, Kalomo, Livingstone, Mazabuka, Monze, Namwala, Siavonga and

Sinazongwe), area (plateau, valley and floodplain), sampling period (March and

September), immunised (0 = no vaccinations in district before sampling, 1 = vaccination

had started), and MEI Ranks. Model selection was based on likelihood ratio testing and

Akaike’s Information Criterion (AIC) (Lindsey, 1997). Because of colinearity between

certain variables (e.g. district and area), some analyses were performed using a subset of

the independent variables. Analyses of the immunisation effect were performed on a

subset of the districts (no immunisation in the valley districts). The influence of MEI

58 T. parva sero-prevalence and El Niño

was modelled in various ways with models being compared amongst them using AIC.

The MEI ranks were included in the final model as a five-month average centred six

months prior to the sampling period (i.e. the average of July, August, September,

October and November the previous year for the March sampling and January,

February, March, April and May for the September sampling). All analyses started from

a fully saturated model and likelihood ratios were used to simplify the models. Results

are reported as odds ratios with their 95% confidence intervals.

Figure 3.1: Map of Southern Zambia. Dots show the veterinary camps where sero-

prevalence surveys were carried out.

3.3 Results

Highest numbers of T. parva positive samples in cattle were recorded in the

plateau districts of Monze and Choma while the lowest were recorded in Siavonga and

Chapter 3

59

Sinazongwe (valley districts, Table 3.1). Taking into account variables ‘district’,

‘sampling period’ and ‘immunised’ and the subset of data pertaining to districts, where

immunisation had taken place, it was found that Monze and Choma had a significantly

higher sero-prevalence than Kalomo, Mazabuka and Namwala (all P < 0.001), that

higher T. parva sero-prevalence was recorded in the September sampling than in March

throughout the study period (P < 0.001) and that we could not show any effect of

immunisation on sero-prevalence levels (P = 0.577) (Table 3.2).

Table 3.1: Sample sizes and average Theileria parva sero-prevalence per district for

the period 1995-2002.

District Total sample

size

Average

sero-prevalence

Monze (Plateau) 4756 0.183

Choma (Plateau) 4958 0.181

Mazabuka (Plateau) 4259 0.146

Livingstone (Valley) 3720 0.074

Kalomo (Plateau) 3002 0.067

Gwembe (Valley) 503 0.058

Namwala (Floodplain) 2530 0.057

Sinazongwe (Valley) 1972 0.024

Siavonga (Valley) 1826 0.015

60 T. parva sero-prevalence and El Niño

Table 3.2: Odds ratios of Theileria parva sero-prevalence (1995 – 2002) in function

of district, month of sampling and immunisation status.

Taking into account MEI ranks (Table 3.3) yielded the results reproduced in

Table 3.4. This table uses all districts (even those where no animals have been

immunised to date). This analysis was compared to an analysis using only the districts

where animals had been immunised (results not shown) and it was found to yield

similar results for the main effects and interactions. The principal difference with Table

3.2 is that now (i.e. when correcting for the significant effect of MEI) immunisation

significantly increases sero-prevalence. The effect of increased MEI ranks is noted

during the March collections, not during the September samplings (odds ratio of

interaction significantly below unity).

Odds Ratio 95% confidence interval

District (compared to Choma)

Kalomo 0.33 0.28 - 0.38

Mazabuka 0.78 0.69 - 0.87

Monze 1.02 0.92 - 1.13

Namwala 0.27 0.23 - 0.33

Month of sampling (compared to March)

September 1.17 1.08 - 1.27

Immunised (compared to not immunised)

Immunised 1.02 0.94 - 1.12

Chapter 3

61

Table 3.3: Multiple El Niño Southern Oscillation Index values for the period 1994-

2002.

Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

1994 36 32 33 41 41 39 43 38 39 50 49 46

1995 47 45 45 32 36 37 31 27 20 16 16 18

1996 15 15.5 19 17 20 28 25 18 21 19 25 21

1997 22 20 22 42 49 54 54 54 54 54 53 53

1998 54 54 54 54 51 49 32 22 14 12 11.5 12

1999 10 9 11 11 12 16 14 11 10 11 13 7

2000 5 8 10 19 26 20 20 23 23 22 14 15

2001 19 15.5 17 23 32 30 30 33 24 21 22 28

2002 30 29 25 39 46 42 38 44 43 42 45.5 45

62 T. parva sero-prevalence and El Niño

Table 3.4: Odds ratios of Theileria parva sero-prevalence (1995 – 2002) associated

with district, month of sampling, immunisation status and MEI ranks.

Odds Ratio 95% confidence interval

District (compared with Choma)

Gwembe 0.31 0.21 - 0.46

Kalomo 0.32 0.28 - 0.38

Livingstone 0.41 0.35 - 0.48

Mazabuka 0.76 0.68 - 0.85

Monze 0.97 0.87 - 1.07

Namwala 0.28 0.23 - 0.33

Siavonga 0.08 0.05 - 0.11

Sinazongwe 0.13 0.10 - 0.17

Immunised (compared to not immunised)

Immunised 1.43 1.28 - 1.60

Month of sampling (compared to March)

September 2.58 2.16 - 3.09

MEI rank (unit increase) + interaction month-of-sampling X MEI

MEI 1.03 1.03 - 1.04

Sep X MEI 0.97 0.97 - 0.98

The average T. parva sero-prevalence using all 52 veterinary camps and average

MEI ranks for the study period are shown in Figure 3.2. This figure clearly shows the

strong association between the two.

Chapter 3

63

10

15

20

25

30

35

40

45

1995 1996 1997 1998 1999 2000 2001 2002

Year

0.05

0.1

0.15

0.2

Figure 3.2: Relationship between annual average MEI values (full line) and Theileria

parva sero-prevalence (dashed lines), illustrating an approximate six-month delay

between a MEI peak and a sero-prevalence peak.

3.4 Discussion

The serological surveys showed T. parva parasites to be widespread in most

districts of Southern Zambia, with the exception of the valley districts of Sinazongwe

and Siavonga. The highest values of T. parva prevalence were observed in the plateau

districts of Monze, Choma and Mazabuka. The T. parva prevalence is related to the

reported abundance of R. appendiculatus ticks on animals in these districts (Speybroeck

et al., 2002). In the valley districts, sero-prevalence was generally lower than those

64 T. parva sero-prevalence and El Niño

reported in plateau districts. It appears that ECF fails to establish itself in these areas

because of the low tick numbers for long periods (Speybroeck et al., 2002). This is

likely due to extreme climatic conditions in these areas where temperatures exceed 40oC

in the hot dry season. Additionally, Young and Leitch (1981) indicated that if

temperatures below 180C and above 330C existed for long periods, they were likely to

inhibit the transmission of T. parva because these extreme temperatures were

detrimental to the development of T. parva in ticks.

A higher sero-prevalence was recorded in the September surveys throughout the

study period. The nymphs of R. appendiculatus in Southern Zambia are more abundant

on cattle in the period May to August (Pegram and Banda, 1990). The higher sero-

prevalence in September must therefore be a result of the R. appendiculatus nymphal

challenge, which, as also noted by Ochanda et al. (1996), results in more sub-clinical

cases than after adult tick challenge.

The most important finding of this study is the strong positive association between

an El Niño event and increased T. parva sero-prevalence with a 6-month lag. The

biological explanation is probably that the increased rainfall and humidity in Southern

Africa, that accompanies an El Niño event favours T. parva survival. It is interesting to

note that the increase in sero-prevalence occurs during the March collections (end of the

rains): Mulumba et al. (2000) already showed in their longitudinal studies that better

rains increased ECF incidence, especially during the rains.

The importance of this positive association between an El Niño event and

increased T. parva sero-prevalence is two-fold. It offers a unique opportunity to predict

up to 6 months in advance an increase in ECF incidence (measured here as increase in

sero-prevalence), giving ample time to prepare control measures (e.g. increased dipping

in epidemic areas, more intensive immunisation campaigns in endemic areas). A readily

Chapter 3

65

available index replaces longitudinal monitoring of sentinel herds, requiring information

not only on climate, tick burdens and phenology, but also on the abundance of the

various tick species involved (Mulumba et al., 2000, 2001). On the down side, the

dominating influence of climate on the sero-prevalence precludes the use of this

parameter to evaluate the effect of an immunisation campaign (one of the original aims

of this study), at least in Southern Zambia. It remains to be demonstrated whether this

finding can be extended to other regions of Southern, Central and Eastern Africa.

However, it must be clear that, provided MEI ranks can be used, the advantage far

outweighs this disadvantage. We recommend that similar analyses be undertaken for

other areas and possibly other vector-borne diseases of veterinary importance

66 T. parva sero-prevalence and El Niño

3.5 References

Bouma, M. J. and Van der Kaay, H. J. (1996). The El Niño Southern Oscillation and the

historic malaria epidemics on the Indian subcontinent and Sri Lanka: an early

warning system for future epidemics? Tropical Medicine and International Health

1, 86-96.

Burridge, M. J. and Kimber, C. D. (1972). The Indirect Fluorescent Antibody Test for

experimental East Coast fever (Theileria parva infection of cattle): evaluation of a

cell culture schizont antigen. Research in Veterinary Science 13, 451-455.

Gitau, G. K., Perry, B. D., Katende, J. M., McDermott, J. J., Morzaria, S. P. and Young,

A. S. (1997). The prevalence of serum antibodies to tick-borne infections in cattle

in small-holder dairy farms in Murang’a District, Kenya: a cross sectional study.

Preventive Veterinary Medicine 30, 95-107.

Lessard, P., L’eplattenier, Norval, R. A. I., Kundert, K., Dolan, T. T., Croze, H.,

Walker, J. B., Irvin, A. D. and Perry, B. D. (1990). Geographical information

systems for studying the epidemiology of cattle diseases caused by Theileria

parva. Veterinary Record 126, 255-262.

Lindsey, J. K. (1997). Applying Generalized Linear Models. Springer, New York.

Maloo, S. H., Thorpe, W., Kioo, G., Ngumi, P., Rowlands, G. J. and Perry, B. D.

(2001). Seroprevalences of vector-transmitted infections of small-holder dairy

cattle in coastal Kenya. Preventive Veterinary Medicine 52, 1-16.

Mukhebi, A. W., Perry, B. D. and Kruska, R. (1992). Estimated economics of

theileriosis control in Africa. Preventive Veterinary Medicine 12, 73-85.

Chapter 3

67

Mulumba, M., Speybroeck, N., Billiouw, M., Berkvens, D.L., Geysen, D. M. and

Brandt, J. R. A., (2000). Transmission of theileriosis in the traditional farming

sector in the Southern Province of Zambia during 1995-1996. Tropical Animal

Health and Production 32, 303-314.

Mulumba, M., Speybroeck, N., Berkvens, D.L., Geysen, D. M. and Brandt, J. R. A.

(2001). Transmission of Theileria parva in the traditional farming sector in the

Southern Province of Zambia during 1997-1998. Tropical Animal Health and

Production 33, 117-125.

Norval, R. A. I., Fivaz, B. H., Lawrence, J. A. and Brown, A. F. (1985). Epidemiology

of tick-borne diseases of cattle in Zimbabwe. III. Theileria parva group. Tropical

Animal Health and Production 17, 19-28.

Ochanda, H., Young, A. S., Wells, C., Medley, G. F. and Perry, B. D. (1996).

Comparison of the transmission of Theileria parva between different instars of

Rhipicephalus appendiculatus. Parasitology 113, 243-253.

Pegram, R. G. and Banda, D. S. (1990). Ecology and phenology of cattle ticks in

Zambia: Development and survival of free-living stages. Experimental and

Applied Acarology 8, 291-301.

Speybroeck, N., Madder, M., Van den Bossche, P., Mtambo, J., Berkvens, N., Chaka,

G., Mulumba, M., Brandt, J., Tirry, L. and Berkvens, D. (2002). Distribution and

phenology of ixodid ticks in southern Zambia. Medical and Veterinary

Entomology 16, 430-441.

StataCorp. (2003). Stata/SE 8.0 Statistical software for windows, Stata Corporation,

College Station, Texas, USA.

68 T. parva sero-prevalence and El Niño

Wolter, K. (1987). The Southern Oscillation in surface circulation and climate over the

tropical Atlantic, Eastern Pacific, and Indian Oceans as captured by cluster

analysis. Journal of Climate and Applied Meteorology 26, 540-558.

Wolter, K. and Timlin, M. S. (1993). Monitoring ENSO in COADS with a seasonally

adjusted principal component index. In: Proceedings of the 17th Climate

Diagnostics Workshop, Norman, OK, NOAA/N MC/CAC, NSSL, Oklahoma

Clim. Survey, CIMMS and the School of Meteor., Univ. of Oklahoma, pp. 52-

57.

Young, A. S. and Leitch, B. L. (1981). Epidemiology of East Coast fever: Some effects

of temperature on the development of Theileria parva in the tick vector,

Rhipicephalus appendiculatus. Parasitology 83, 199-211.

Chapter 4

69

Chapter 4

East Coast fever and Multiple El Niño Southern Oscillation Ranks

Adapted from:

Fandamu, P., Duchateau, L., Speybroeck, N., Mulumba, M. and Berkvens, D. (2005). East Coast fever and Multiple El Niño Southern Oscillation

Ranks. Veterinary Parasitology (in press).

70 ECF contacts and El Niño

4.1 Introduction

The protozoan Theileria parva causes East Coast fever (ECF) in African cattle.

The disease is transmitted transstadially by the ticks Rhipicephalus appendiculatus and

the closely related Rhipicephalus zambeziensis: larvae and nymphs can become infected

with T. parva while feeding on an infected host and transmit it in the next stage. ECF

remains one of the most important cattle diseases in terms of economic losses (168

million US$ in 1989, MacInerney et al., 1992) and restriction of livestock development

in Eastern, Central and Southern Africa (Norval et al., 1992).

Previous studies (Mulumba et al., 2000, 2001) explained transmission patterns of

ECF in Southern Zambia, but the disease is very complex and incidence remains

difficult to predict. The above studies demonstrated that the majority of contacts in

Southern Zambia were mainly due to nymphal transmission during the early part of the

dry season in years with below-average rainfall, while during wet years adult ticks were

equally responsible for transmission. Mulumba and co-workers also observed that in dry

years a R. appendiculatus stock, indistinguishable from R. zambeziensis, was found to

infest cattle. However, this tick stock was no longer found when normal rains returned.

They concluded that this change of tick stock played a very important role in the

transmission of ECF in the area. Predicting disease pressure and infection peaks thus

relies on having precise knowledge of a series of parameters, including climate and tick

phenology and taxonomy. Thus it is still difficult to optimise control programmes.

The current work was aimed at investigating the possibility of a link between El

Niño and disease pressure, thereby providing a simple way to forecast outbreaks.

El Niño, the warming phase of the Southern Oscillation, has been associated with

malaria epidemics in Eastern Africa (Patz et al., 2002; Lindblade et al., 1999), in

Colombia (Bouma et al., 1997) and in India and Sri Lanka (Bouma and Van der Kaay,

Chapter 4

71

1996). To the best of our knowledge, no association between this climate index and

transmission of a parasite of veterinary importance has been documented. An

understanding of the role that global weather changes play in ECF transmission is useful

for forecasting and early warning. In this study, we investigate the association between

T. parva contacts in cattle and El Niño using the Multiple El Niño Southern Oscillation

Index (MEI) Ranks.

4.2 Materials and methods

4.2.1 Study area

The study was carried out in the Southern Province of Zambia, which has a total surface

area of 85,000 km2 and a cattle population of circa 575,000 (Anonymous, 2003). Two

sentinel herds were studied: one at Nteme (Monze) on the plateau (16o10’S, 27o29’E,

1050 m) and the other at Halubilo in the Gwembe valley (16o26’S, 27o42’E, 780 m) as

shown in Figure 4.1. The sentinel herds were selected on the basis of preliminary

studies done by the Assistance to the Veterinary Services of Zambia (ASVEZA) project

in 1993/1994 (Mulumba et al., 2000). Briefly, the plateau herd was typical of the

endemic disease situation, while the valley herd represented the epidemic situation.

Management of the sentinel herds was traditional with no tick control and animals were

let to graze on communal pastures. The nine-year study started in January 1994 and

ended in December 2002.

4.2.2 Sampling

All calves born in the two sentinel herds were followed up from birth until their first

contact with T. parva. Age at and calendar time of first contact were recorded. Age at

first contact is used as a measure to assess whether an area was endemic or epidemic

72 ECF contacts and El Niño

(Billiouw et al., 2002). Animals were examined clinically weekly. Blood slides were

made weekly and checked for parasitaemia. Biopsies were made and examined for

schizonts in case of swollen lymph nodes. Blood was collected weekly to test for

T. parva antibodies using the Indirect Fluorescent Antibody test (IFAt) (Burridge and

Kimber, 1972). The cut-off titre for seropositivity was taken at 1/160. Seroconversion

was defined as three weeks consecutive sero-positive samples and the date of contact

with T. parva was set one month before the date of first sero-conversion (Billiouw et

al., 1999).

4.2.3 Meteorological data

Rainfall data were collected from meteorological recording stations in the study

areas and also the Meteorological Department in Lusaka, Zambia. Multiple El Niño

Southern Oscillation Index (MEI) Ranks were retrieved from the National Oceanic and

Atmospheric Administration (NOAA) website (http://www.cdc.noaa.gov , 2004).

El Niño/Southern Oscillation (ENSO) is used to describe both warm (El Niño) and cool

(La Niña) ocean atmospheric events in the tropical Pacific (http://www.ggweather.com ,

2000). The MEI is a composite index using sea surface temperatures, surface air

temperatures, sea-level pressure, zonal (i.e. east-west) surface wind, meridional (i.e.

north-south) surface wind and total amount of cloudiness. Multiple El Niño Southern

Oscillation Index ranks from 1-18 denote strong to weak La Niña conditions, while the

range 38-55 denotes weak to strong El Niño conditions.

4.2.4 Statistical methods

The data were analysed with a Cox proportional hazard model in Stata SE/8.0

(StataCorp. 2003), with the time to first T. parva contact as response and the year of

birth and area (epidemic versus endemic area) as covariates. The Cox proportional

Chapter 4

73

hazard model is a semi-parametric model in that it makes no assumptions about the

specific hazard function. It only assumes proportionality of the hazards. The time to first

contact is inversely related to the instantaneous probability or hazard of first contact (the

probability of having a first T. parva contact at this instance given that the animal has

survived without T. parva contact up to that time. Traditionally, results are presented in

terms of hazard ratios. In this study, we use the ratio of the hazard for animals born in a

particular year to the hazard for animals born in 1996 (animals born in 1996 had the

lowest hazard of first T. parva contact, i.e. had on average a higher age at first contact).

Animals born in 2002 and calves showing seropositivity when less than one month old

were excluded from the analysis to minimise the possibility of detecting passively

derived maternal antibodies (Moll and Lohding, 1984; Billiouw et al., 2002).

ZIMBABWE

BOTSWANA

Kariba L

ake

Kafue N

ational

Parc

Choma

Kalomo

Sinazo

ngwe

Siavonga

Monze

Mazabuka1050 m

1200 m

1050 m900 m

600 m

Gwembe

#

Livingstone

Namwala#

#

18°3

0' 18°30'

18°0

0' 18°00'

17°3

0' 17°30'

17°0

0' 17°00'

16°3

0' 16°30'

16°0

0' 16°00'

15°3

0' 15°30'

15°0

0' 15°00'

14°3

0' 14°30'

14°0

0' 14°00'

13°3

0' 13°30'

23°00'

23°00'

23°30'

23°30'

24°00'

24°00'

24°30'

24°30'

25°00'

25°00'

25°30'

25°30'

26°00'

26°00'

26°30'

26°30'

27°00'

27°00'

27°30'

27°30'

28°00'

28°00'

28°30'

28°30'

29°00'

29°00'

29°30'

29°30'

30°00'

30°00'

30°30'

30°30'

N

40 0 40 80 Kilometers

Figure 4.1: Map of Southern Zambia. The two big dots show sentinel herd localities

on the plateau in Monze and in the Gwembe valley.

74 ECF contacts and El Niño

4.3 Results

4.3.1 Multiple El Niño Southern Oscillation indices

The MEI indices indicating the presence or absence of El Niño are shown in

Figure 4.2. Of interest for the present study is the distinct El Niño episode in 1997-

1998, flanked on both sides by La Niña conditions. Multiple El Niño Southern

Oscillation Index ranks rise again from 1999 onwards to new El Niño levels by 2002.

The onset of the El Niño event was accompanied by heavier rainfall (Table 4.1).

Table 4.1: Rainfall figures for Monze and Gwembe districts where the sentinel herds

are located.

Year Monze (Nteme) Gwembe (Halubilo)

1994 460.0 mm 401.9 mm

1995 490.0 mm 338.9 mm

1996 710.3 mm 754.0 mm

1997 1031.8 mm 992.2 mm

1998 515.6 mm 434.0 mm

1999 833.2 mm 653.3 mm

2000 850.7 mm 880.9 mm

2001 762.4 mm 770.9 mm

2002 603.4 mm 458.4 mm

Chapter 4

75

4.3.2 East Coast fever transmission and El Niño

An animal’s year of birth influenced its hazard of coming into contact with

T. parva at the two locations. Figure 4.2 shows the hazard ratios estimated with a Cox

regression model for the different years of birth compared to the year of birth 1996. In

the same figure MEI ranks are presented. The analysis showed that hazard ratios were

higher on the plateau than in the valley in most of the years. However, hazard ratios

went up sharply when El Niño was present (year 1997) (p < 0.001). Hazard ratios were

higher in the epidemic area (Halubilo, valley) than in the endemic area of Nteme

(plateau) when El Niño was present (p < 0.001). Hazard ratios in Nteme were also seen

to increase in the year when El Niño was present though not as high as at Halubilo. The

hazard ratios at the endemic area remained high in the following year even when El

Niño was waning. It must be noted that most of the animals born in 1997 (height of the

El Niño event) become infected later in the year or in the following years. There is thus

a delay between the El Niño event and increased probability to become infected by

T. parva. It is also for this reason that no effect is discernible for the second (much

smaller) El Niño event in 2002-2003 (this would be so for the 2002 cohort).

76 ECF contacts and El Niño

0

10

20

30

40

50

60

Ja

n-9

3

Ja

n-9

4

Ja

n-9

5

Ja

n-9

6

Ja

n-9

7

Ja

n-9

8

Ja

n-9

9

Ja

n-0

0

Ja

n-0

1

Ja

n-0

2

Period

ME

I-v

alu

es

0

5

10

15

20

25

30

35

Ha

za

rd

ra

tio

Figure 4.2: Monthly MEI values (continuous line) and hazard ratios for Nteme

(endemic area, full line) and Halubilo (epidemic area, dotted line) with 1996 as

reference category.

4.4 Discussion

This study confirms the results found by Fandamu et al. (2005), where

population-level antibody levels against T. parva were used, rather than the more

accurate age at first contact in different calf cohorts. It is well known that El Niño, the

warming phase of the Southern Oscillation produces heavier than normal rainfall in

some parts of the world (Lindblade et al., 1999). The results from this study are in

agreement with the findings of Kovats et al. (2003) who reported excessive rainfall in

some areas of the Southern Africa region during the 1997-1998 El Niño event. The

heavier than normal rainfall is indeed very likely related to the increased disease

transmission due to higher tick numbers, especially so in the drier valley region of

Chapter 4

77

Southern Zambia. The hazard ratios in the epidemic valley area (Halubilo) were

significantly higher than the endemic area of Nteme (plateau) when El Niño was

present. This may be explained by the low challenge in 1994, 1995 and 1996 that

resulted in a large population of susceptible animals. High tick numbers and a large

population of susceptible animals result in a high number of infections, creating a pool

of infective animals, which in turn infect the current population of larvae and nymphs,

thus ensuring that newborn animals become infected at an early age. In other words, in

arid/dry areas like those of Southern Zambia, R. appendiculatus may not survive easily,

reducing the chances for the ticks to become infective. This way, ECF is not common in

the area, and therefore the population lacks protective immunity and is more vulnerable

when conditions for transmission improve. The effect of the higher rainfall is also short-

lived in these arid areas, as hazards immediately returned to their normal levels. In the

endemic area of Nteme, high hazard ratios were recorded even a year after El Niño had

ended. This is possibly a result of the El Niño effects persisting for longer periods in

areas with more rainfall on average, e.g. because higher tick population densities do not

collapse so quickly as in more arid areas.

Our results indicate a strong association between the presence of El Niño and the

increased transmission of ECF in cattle especially in the drier valley areas of Southern

Zambia. Though this is a veterinary study, it is essentially in agreement with Nicholls

(1993) who suggested that the incidence of most vector-borne diseases in humans was

associated with El Niño Southern Oscillation. It must be pointed out that the higher

probabilities of contact follow the high MEI ranks: high MEI ranks are recorded when

the animals are born, i.e. this study provides evidence that MEI ranks have predictive

power as far as ECF outbreaks in low rainfall areas (e.g. valley regions) are concerned.

78 ECF contacts and El Niño

Farmers could therefore be forewarned of major outbreaks in order to prevent or at least

reduce losses and the use of this simple index will facilitate better disease control.

Chapter 4

79

4.5 References

Anonymous. (2003). Southern Province, Zambia, Department of Veterinary and

Livestock Development Annual Report for 2002.

Billiouw, M., Mataa, L., Marcotty, T., Chaka, G., Brandt, J. and Berkvens, D. (1999).

The current epidemiological status of bovine theileriosis in eastern Zambia.

Tropical Medicine and International Health 4, A28-A33.

Billiouw, M., Vercruysse, J., Marcotty, T., Speybroeck, N., Chaka, G. and Berkvens, D.

(2002). Theileria parva epidemics: a case study in eastern Zambia. Veterinary

Parasitology 107, 51-63.

Bouma, M. J. and Van der Kaay, H. J. (1996). El Niño Southern Oscillation and the

historic malaria epidemics on the Indian subcontinent and Sri Lanka: an early

warning system for future epidemics? Tropical Medicine and International

Health 1, 86-96.

Bouma, M. J., Poveda, G., Rojas, W., Chavasse, D., Quiñones, M., Cox, J. and Patz, J.

(1997). Predicting high-risk years for malaria in Colombia using parameters of

El Niño Southern Oscillation. Tropical Medicine and International Health 2,

1122-1127.

Burridge, M. J. and Kimber, C. D. (1972). The Indirect Fluorescent Antibody Test for

experimental East Coast fever (Theileria parva infection of cattle): evaluation of a

cell culture schizont antigen. Research in Veterinary Science 13, 451-455.

El Niño-Southern Oscillation Glossary. (2000);

http://www.ggweather.com/enso/glossary.htm

80 ECF contacts and El Niño

Fandamu, P., Duchateau, L., Speybroeck, N., Marcotty, T., Mbao, V., Mtambo, J.,

Mulumba, M. and Berkvens, D. (2005). Theileria parva seroprevalence in

traditionally kept cattle in Southern Zambia and El Niño. International Journal for

Parasitology 35, 391-396.

Kovats, R. S., Bouma, M. J., Hajat, S., Worrall, E. and Haines, A. (2003). El Niño and

health. Lancet 362, 1481-1489.

Lindblade, K. A., Walker, E. D., Onapa, A. W., Katungu, J. and Wilson, M. L. (1999).

Highland malaria in Uganda: prospective analysis of an epidemic associated with

El Niño. Transactions of the Royal Society of Tropical Medicine and Hygiene 93,

480-487.

MacInerney, J. P., Howe, K. S. and Schepers, J. A. (1992). A framework for the

economic analysis of disease in farm livestock. Preventive Veterinary Medicine

13, 137 154.

Moll, G. and Lohding, A. (1984). Epidemiology of theileriosis in the Trans-Mara

division, Kenya: husbandry and disease background and preliminary

investigations on theileriosis in calves. Preventive Veterinary Medicine 2, 801-

831.

Mulumba, M., Speybroeck, N., Billiouw, M., Berkvens, D. L., Geysen, D.M. and

Brandt, J. R. A. (2000) Transmission of theileriosis in the traditional farming

sector in the Southern Province of Zambia during 1995-1996. Tropical Animal

Health and Production 32, 303-314.

Mulumba, M., Speybroeck, N., Berkvens, D. L., Geysen, D. M. and Brandt, J. R. A.

(2001). Transmission of Theileria parva in the traditional farming sector in the

Southern Province of Zambia during 1997-1998. Tropical Animal Health and

Production 33, 117-125.

Chapter 4

81

National Oceanic and Atmospheric Administration (NOAA) website, last update 5 May

2004. http://www.cdc.noaa.gov/people/klaus.wolter/MEI/rank.html

Nicholls, N. (1993). El Niño-Southern Oscillation and vector-borne diseases. Lancet

342, 1284-1285.

Norval, R. A. I., Perry, B. D., Young, A. S. (1992). The Epidemiology of Theileriosis in

Africa. Academic Press, London.

Patz, J. A., Hulme, M., Rosenzweig, C., Mitchel, T. D., Goldberg, R. A., Githeko, A.

K., Lele, S., McMichael, A. J. and Le Sueur, D. (2002). Regional warming and

malaria resurgence. Nature 420, 627-628.

StataCorp. (2003). Stata/SE 8.0 Statistical software for windows, Stata Corporation,

College Station, Texas, USA.

82 predicting fatal T. parva infections

Chapter 5

Red blood cell volume as a predictor of fatal reactions in cattle infected

with Theileria parva Katete

Adapted from:

Fandamu, P., Marcotty, T., Brandt, J. R. A., Duchateau, L., Speybroeck, N., Dolan, T. T., and Berkvens, D. (2005). Red blood cell volume as a

predictor of fatal reactions in cattle infected with Theileria parva Katete. Onderstepoort Journal of Veterinary Research (in press).

Chapter 5

83

5.1 Introduction

Theileria parva is a tick-borne protozoan parasite of cattle that causes an acute, often

fatal lymphoproliferative disease known as East Coast fever (ECF). Theileria parva

multiplies in bovine lymphocytes by schizogony resulting in the production of

merozoites that invade red blood cells (RBCs) and develop into piroplasms (Conrad et

al., 1986). The major symptoms in T. parva infections are enlargement of lymph nodes,

especially those adjacent to sites of parasite inoculation, pyrexia, lympho-destruction

resulting into leukopenia, interstitial pneumonia, pulmonary oedema and death (Irvin

and Mwamachi, 1983; Jura and Losos, 1980; Irvin and Morrison, 1987). Other signs

include blood-tinged diarrhoea, lacrimation and corneal opacity, splenic enlargement

and lymphomata in the kidneys. Animals usually die within 30 days after infection

(Irvin and Morrison, 1987). Some animals die early within 24 days as a result of

respiratory distress while others die later following secondary infections and wasting

(Irvin and Morrison, 1987; Irvin and Mwamachi, 1983). Anaemia has not been

recognised as a major feature in classical T. parva infections (Maxie et al., 1982) but it

has been described in cattle infected with certain stocks of T. parva (Norval et al., 1992;

Mbassa et al., 1994). The pathogenesis of the anaemia in T. parva infections is not well

understood (Mbassa et al., 1994). In infections with the related T. annulata and

T. mutans parasites, anaemia is common and is attributed to the replication of the

parasite in the erythrocytes (Young et al., 1978; Norval et al., 1992). The severity of

T. parva infections has been linked to the quantity of the infective dose (Radley et al.,

1974; Dolan et al., 1984), to the T. parva stock (Mbogo et al., 1996) and to the cattle

breed (Paling and Geysen, 1981) but there are no reports of the role of anaemia in the

pathogenicity or virulence of T. parva.

84 predicting fatal T. parva infections

The objective of this study was to confirm the decline of packed cell volume

(PCV) in animals infected by T. parva Katete and to associate the magnitude of this

decline with the severity of the clinical reaction. In addition, we investigated the

association between the size of red blood cells and severity and lethality of ECF

infection. Finally, we compared the diameter of infected and non-infected cells to

evaluate whether there is a preference of T. parva merozoites for larger or smaller cells.

5.2 Materials and methods

5.2.1 Animals

A retrospective analysis was conducted using haematological data from 18 T. parva

susceptible Belgian Black and White dairy cattle aged approximately ten months that

had been used in T. parva experiments between 1996 and 2000. The animals (16 heifers

and 2 steers) were housed and fed on hay and concentrates and water ad libitum. These

experiments were performed under the control of the Ethical Committee for

Experimental Animals of the Institute of Tropical Medicine of Antwerp (Belgian

registration number LA 1100120) which guarantees that animals did not suffer

unnecessarily.

Animals were inoculated with T. parva Katete stabilates derived from the same

reference stock V1 (1983). This stock was isolated from cattle in Katete district of

Eastern Province of Zambia (Berkvens et al., 1989). The stabilates were administered

randomly and independently of MCV and PCV values. The stabilates were injected

subcutaneously in front of the right parotid lymph node. Blood samples were collected

from the external jugular vein in ethylene diaminotetraacetic acid (EDTA) tubes for

PCV and MCV measurements. All samples were processed on the day of collection.

Chapter 5

85

Blood and lymph smears were prepared daily from day 7. Theileria parva infections

were confirmed by the presence of schizonts in the lymph node biopsies 7-20 days after

inoculation. Classification of reactions was according to Morzaria et al. (1987). Non-

lethal reactions were recorded when infected animals (confirmed by the presence of

schizonts) survived for more than 24 days without chemotherapy.

5.2.2 Haematological examinations

The PCV and the mean corpuscular volume (MCV) were measured directly using a

Technicon H1E haematology analyzer (Bayer, Tarrytown, NY) together with other

haematological parameters. The corpuscular size is determined by direct measurement

and not derived from PCV and red blood cell counts (Feldman et al., 2000). Additional

examinations were carried out on blood smears to determine the mean diameters of

infected and non-infected RBCs. Red blood cells were measured prior to detection of

piroplasms in blood (days 0 – 7), at 1% parasitaemia and when maximum parasitaemia

was recorded. Measurements were done using a micrometer (Mahr 46EH Digital

micrometer head, Germany) fitted on a microscope at x1000 magnification.

Examinations were performed around the tail of the smears where the cells were evenly

spread. Twenty circular RBCs were measured from 3-4 fields for slides obtained before

parasitaemia. Following the detection of piroplasms, the sizes of all infected cells in a

particular field were measured, provided that they contained a single piroplasm and

were circular. Thereafter, four unparasitised RBCs with normal shape and nearest to an

infected cell were measured and an average size calculated.

86 predicting fatal T. parva infections

5.2.3 Statistical analysis

Data analysis was done in Stata SE/8.0 (StataCorp., 2003). The PCV and MCV profiles

were drawn using cubic splines (five cross-median knots). The PCV, MCV,

parasitaemia and RBC diameters were analysed using mixed models with animals as

random effect. To ensure normality of the response variables, PCV values were arcsine

transformed (Osborne 2002) while RBC diameters and parasitaemia values were

transformed using logarithms. Lethality group (lethal or non-lethal), period of infection

(day 0-7 or day 14-21) and the period x lethality group interaction were used as

categorical fixed effects for PCV and MCV regressions. Separate analyses were done on

RBC diameters before and during parasitaemia using the lethality group, the presence of

piroplasms and the interaction between these two as categorical fixed effects.

5.3 Results

The animals that suffered lethal reactions died on average 20 days after infection (18 –

23 days) following a period of high fever and parasitosis. The reactions of all animals

undergoing non-lethal infections were also characterised by fever and parasitosis.

However, parasitaemia was significantly higher in lethal than non-lethal reactions (p =

0.004) in the period 14-21 days. The mean parasitaemia in the lethal group was 3.88

(1.54 – 9.77) while that of the non-lethal group was 0.51 (0.18 – 1.41).

Chapter 5

87

5.3.1 Packed Cell Volume (PCV)

Packed cell volume profiles for animals with lethal and non-lethal reactions are shown

in Figure 5.1. In the early stage of infection (days 0 – 7) there was no significant

difference in PCV values between animals with lethal and those with non-lethal

reactions (p = 0.915) (Figure 5.3a). In both groups, the PCV was significantly lower in

the later stage compared to the early stage (p < 0.001). However, the PCV in the later

stage of infection (days 14 – 21) decreased significantly more in lethal reactions (p <

0.001).

Figure 5.1: Packed cell volume (PCV) as a function of days post infection in animals

with non-lethal and lethal reactions. Open circles are individual observations and the

solid line corresponds to a cubic spline function fitted through the observations.

5.3.2 Mean Corpuscular Volume (MCV)

The MCV profiles following T. parva infection are shown in Figure 5.2. When the

MCV values were compared in the early stage of infection, there was a statistically

significant difference between animals that developed lethal reactions and those that did

88 predicting fatal T. parva infections

not (p < 0.001). Animals that had lethal infections had lower initial MCV values

(smaller cells) than those that suffered non-lethal infections. The box-and-whisker plot

in Figure 5.3b further shows that animals that suffered lethal reactions had a lower

median MCV value than those that had non-lethal reactions. Late in the infection, MCV

values remained lower in animals with lethal reactions (p < 0.001). When plotting the

initial frequency of RBC sizes in individual animals, the distribution of RBC sizes was

wider in animals with non-lethal reactions (Figure 5.4). While the lower limit of sizes

was the same in both groups, the upper limit was lower in lethal T. parva infection.

Figure 5.2: Mean corpuscular volume (MCV) as a function of days post infection in

animals with non-lethal and lethal reactions. Open circles are individual observations

and the solid line corresponds to a cubic spline function fitted through the observations.

non-lethal reaction lethal reaction

Chapter 5

89

Figure 5.3: The box-and-whisker plots of (a) the average packed cell volume (PCV)

and (b) mean corpuscular volume (MCV) values for cattle infected with T. parva Katete

in the first seven days post infection. Dots represent outlier values. Boxes represent the

25th, 50th (median) and 75th percentiles (including outliers). Whiskers include all

observations except outliers.

90 predicting fatal T. parva infections

5.3.3 Sizes of parasitised and unparasitised erythrocytes in T. parva infections

Animals that suffered lethal reactions had smaller RBC diameters before parasitaemia

was detected (p < 0.001) (Table 5.1). In addition, for both lethal and non-lethal

reactions, the diameter of parasitized RBCs was significantly smaller than non-

parasitised cells at low and maximum parasitaemia (p < 0.001). The average size of

uninfected cells was significantly larger during the parasitaemia of lethal reactions than

before.

Table 5.1: Mean diameter (95% confidence interval) of red blood cells (µm).

Lethal Non-lethal

Before

parasitaemia

4.87

(4.81 - 4.93)

5.19

(5.12 - 5.26)

Infected RBC 4.46

(4.41 - 4.51)

4.60

(4.55 - 4.65) Low

parasitaemia Non-infected

RBC

5.05

(5.01 - 5.09)

5.21

(5.17 - 5.25)

Infected RBC 4.45

(4.40 - 4.51)

4.59

(4.54 - 4.65)

During

parasitaemia

Maximum

parasitaemia Non-infected

RBC

5.04

(5.00 - 5.08)

5.20

(5.16 - 5.24)

Chapter 5

91

LETHAL

NON-LETHAL

92 predicting fatal T. parva infections

Figure 5.4: Red blood cell volume (in fl) histograms from animals in the study taken

in the first stage of infection (0 - 7days) using a haematology analyzer. Animals 1 – 10

had lethal reactions; 11 – 18 had non-lethal reactions.

5.4 Discussion

This study confirmed that anaemia occurs in animals infected with the T. parva Katete

stock. In the early stage of infection, the PCV profiles were similar in both outcome

groups but the decrease in PCV was steeper in animals undergoing lethal reactions as

the infection progressed.

Before day 24 in both lethal and non-lethal infections, the MCV did not respond,

presenting as a normocytic (non-regenerative) anaemia. This contrasts with our

unpublished experimental observations with the T. parva Katete infections in Belgian

Black and White cattle where during recovery following chemotherapy the MCV values

increased, showing a regenerative response. It is possible that chemotherapy destroys

most of the schizonts, halts lympho-destruction and allows the homeostatic mechanism

to switch focus from a lymphoid to an erythroid response.

The other major finding was that animals that suffered lethal reactions had lower

initial MCV values. Before infection, these animals were not anaemic as their PCV

values were within normal range. No effect of age on MCV could be observed in this

study as the number of observations was limited and the range of ages narrow. Birgel

Junior et al. (2001) reported that PCV and RBC counts were significantly higher in

younger Jersey cattle while MCV did not seem to be significantly affected by age. In

contrast Penny et al. (1966) working with different breeds (Friesian, Hereford,

Shorthorn, Charolais, Guernsey and Ayrshire) with ages ranging from less that 2 years

to over 10 years reported that PCV increased in bulls of ages three to eight years

Chapter 5

93

compared to other age groups while there was no significant difference in RBC and

MCV values with age.

One possible explanation for the deaths in animals with low MCVs could be that

there is an increased affinity of T. parva for small RBCs which contributes to increased

destruction or clearance of infected cells and subsequently to anaemia. This hypothesis

is supported by the fact that uninfected red blood cells were larger during parasitaemia

than before in lethal reactions in our study. Indeed, if the smallest cells of a pool

become infected, the average size of the remaining cells will increase as observed.

Waugh et al. (1992) reported that aging RBCs lose both surface area and volume

and are smaller than younger cells. Older (smaller) cells might be more easily infected

by merozoites than younger (larger) cells. Shaw and Tilney (1995) suggested that the

deformability of the erythrocyte membrane may be a crucial factor in determining host

cell susceptibility to invading T. parva merozoites. It may be that older cells lose some

of the characteristics that prohibit the initial interaction with merozoites, making it

easier for the parasite to enter the cell. Therefore, as parasitized cells are lysed or

damaged by the parasite, or parasitized and non parasitized cells are coated with parasite

antigen and removed by the immune system, animals with a pool of old cells may be

more likely to develop anaemia than animals with a population of younger cells. It has

been reported that following T. parva infections, there are considerable losses in

parasitized RBCs and anaemic changes in animals with severe parasitaemia (Barnett

1957). Additionally, Shiono et al. (2004) reported an increasing number of IgG bound

red blood cells that were prone to removal by phagocytosis in animals infected with

Theileria sergenti when a rise in parasitaemia was recorded. Otsuka et al. (2001) also

reported immunological phagocytosis of IgG bound red blood cells in dogs infected

94 predicting fatal T. parva infections

with Babesia gibsoni resulting into severe anaemia. Therefore, animals with a higher

proportion of smaller cells may experience a higher fatality rate as parasite invasion of

these fragile cells would lead to more severe anaemia. The anaemia may exacerbate the

respiratory distress caused by pneumonia and oedema induced by the infiltration of the

lungs with infected and dividing lymphoblasts (Irvin and Morrison, 1987) that is

considered to be the cause of death in most T. parva infections (Irvin and Mwamachi,

1983).

The invasion of older RBCs observed in this study is in contrast to the findings

of Mbassa et al. (1994) who reported that T. parva merozoites infected immature stages

of erythrocytes. A possible explanation for this difference could be that Mbassa and

colleagues were studying natural T. parva challenge in Tanzania and were thus looking

at a different parasite and cattle with a different genetic make up.

These findings offer the possibility of predicting the possible outcome of

T. parva infection based on an assessment of the initial MCV values and could

contribute to improved diagnosis and control. It must be noted that MCV is only one

additional parameter affecting the outcome of T. parva which is equally influenced by

the parasite strain, the dose and cattle breed. Our observations pertain only to T. parva

Katete in Belgian Black and White dairy animals and need to be extended to other

breeds and other T. parva stocks. It might also have relevance for other blood parasites

of animal health importance such as Babesia spp.

Chapter 5

95

5.5 References

Barnett, S. F. (1957). Theileriasis Control. Bulletin of Epizootic Diseases of Africa 5,

343-357.

Berkvens, D. L., Geysen, D. M. and Lynen, G. M. (1989). East Coast fever

Immunisation in the Eastern Province of Zambia. In: Theileriosis in Eastern,

Central and Southern Africa. Proceedings of a workshop on East Coast fever

Immunisation held in Lilongwe, Malawi, 20-22 September 1988, Dolan, T. T.

(Editor), pp. 83-86.

Birgel Junior, E. H., D’angelino, J. L., Benesi, F. J. and Birgel, E. H. (2001). Reference

values of the erythrogram of Jersey breed, raised in São Paulo state. Arquivo

Brasileiro de Medicina Veterinária e Zootecnia 15(2).

Conrad, P. A., Denham, D. and Brown, C. G. D. (1986). Intraerythrocytic multiplication

of Theileria parva in vitro: an ultrastructural study. International Journal for

Parasitology 16, 223-229.

Dolan, T. T., Young, A. S., Losos, G. J., McMillan, I., Minder, C. H. E. and Soulsby, K.

(1984). Dose dependent responses of cattle to Theileria parva stabilate.

International Journal for Parasitology 14, 89-95.

Feldman, B. F., Zinkl, J. G. and Jain, N. C. (2000). Schalm’s Veterinary Haematology.

Fifth Edition. Lippincott Williams and Wilkins, Baltimore USA.

Irvin, A. D. and Mwamachi, D. M. (1983). Clinical and diagnostic features of East

Coast fever (Theileria parva infection of cattle). Veterinary Record 113, 192-

198.

96 predicting fatal T. parva infections

Irvin, A. D. and Morrison, W. I. (1987). Immunopathology, immunology and

immunoprophylaxis of Theileria infections. In: Immune Responses in Parasitic

Infections: Immunology, Immunopathology and Immunoprophylaxis, Volume

III, Soulsby E.J.L. (Editor), CRC Press, Inc. Boca Raton, Florida, pp. 223-74

Jura, W. G. Z. and Losos, G. J. (1980). A comparative study of the diseases in cattle

caused by Theileria Lawrencei and Theileria parva. 1. Clinical signs and

parasitological observations. Veterinary Parasitology 7, 275-286.

Maxie, M. G., Dolan, T. T., Jura, W. G. Z., Tabel, H. and Flowers, M. J. A. (1982). A

comparative study of the diseases in cattle caused by Theileria parva or

T. Lawrencei: II. Hematology, clinical chemistry, coagulation studies and

complement. Veterinary Parasitology 10: 1-19.

Mbassa, G. K., Balemba, O., Maselle, R. M. and Mwaga, N. V. (1994). Severe anaemia

due to haematopoietic precursor cell destruction in field cases of East Coast

fever in Tanzania. Veterinary Parasitology 52, 243-256.

Mbogo, S, K., Kariuki, D. P., Nguni, P. N. and McHardy, N. (1996). A mild Theileria

parva parasite with potential for immunisation against East Coast fever.

Veterinary Parasitology 61, 41-47.

Morzaria, S. P., Irvin, A. D., Voigt, W. P. and Taracha, E. L. N. (1987). Effect of timing

and intensity of challenge following immunisation against East Coast fever.

Veterinary Parasitology 26, 29-41

Norval, R. A. I., Perry, B. D. and Young, A. S. (1992). The Epidemiology of

Theileriosis in Africa. Academic Press, London.

Osborne, W. J. (2002). Notes on the use of data transformations. Practical Assessment,

Research and Evaluation 8, 6.

Chapter 5

97

Otsuka, Y., Yamasaki, M., Yamato, O. and Maede, Y. (2002). The effect of

macrophages on the erythrocyte oxidative damage and the pathogenesis of

anaemia in Babesia gibsoni-infected dogs with low parasitaemia. Journal of

Veterinary Medical Science 64, 221-226.

Paling, R. W. and Geysen, D. (1981). Observations on Rwandan strains of Theileria

parva and the value of Theileria parva (Nyakizu) as a possible vaccine strain.

In: Irvin, A.D., Cunningham, M.P. and Young, A.S. (Eds.), Advances in the

Control of Theileriosis: Proceedings of an International Conference Held at

ILRAD, Nairobi, 9 – 13 February 1981. Martinus Nijhoff Publishers, The

Hague, pp. 238 – 241.

Penny, R. H. C., Scofield, A. M. and Cembrowicz, H. (1966). Haematological values

for the clinically normal bull. British Veterinary Journal 122, 239-247.

Radley, D. E., Brown, C. G. D., Burridge, M. J., Cunningham, M. P., Peirce, M. A. and

Purnell, R. E. (1974). East Coast fever; quantitative studies of Theileria parva in

cattle. Experimental Parasitology 36, 278-287.

Shaw, M. K. and Tilney, L. G. (1995). The entry of Theileria parva merozoites into

bovine erythrocytes occurs by a process similar to sporozoite invasion of

lymphocytes. Parasitology 111, 455-461.

Shiono, H., Yagi, Y., Kumar, A., Yamanaka, M. and Chikayama, Y. (2004).

Accelerated binding of autoantibody to red blood cells with increasing anaemia

in cattle experimentally infected with Theileria sergenti. Journal of Veterinary

Medicine B51, 39-42.

StataCorp. (2003). Stata/SE 8.0. for Windows Statistical Software, Stata Corporation,

College Station, Texas.

98 predicting fatal T. parva infections

Waugh, R. E., Narla, M., Jackson, C. W., Mueller, T. J., Suzuki, T. and Dale, G. L.

(1992). Rheologic properties of senescent erythrocytes: loss of surface area and

volume with red blood cell age. Blood 79, 1351-1358.

Young, A. S., Purnell, R. E., Payne, R. C., Brown, C. G. D. and Kanhai, G. K. (1978).

Studies on the transmission and course of infection of a Kenyan strain of

Theileria mutans. Parasitology 67, 99-115.

Chapter 6

99

Chapter 6

Perception of cattle farmers on East Coast fever immunisations in

Southern Zambia

Adapted from:

Fandamu, P., Thys, E., Duchateau, L. and Berkvens, D. (2005). Perception of cattle farmers on East Coast fever immunisations in Southern Zambia.

Tropical Animal Health and Production (in press).

100 Perception on ECF immunisations

6.1 Introduction

Southern Zambia covers a total surface area of 85,283 square kilometres and is the most

important cattle rearing area of the country. The province accounts for the largest

portion (24.4%) of the total cattle-raising households in Zambia (CSO, Zambia, 2000),

holding 742,524 head of cattle (28.3% of the country’s total cattle population)

(NALEIC, Zambia, 2002). However, cattle productivity in the area is threatened by

important cattle diseases such as East Coast fever (ECF), Foot-and-mouth disease

(FMD) and Anthrax. East Coast fever, a tick borne disease, is economically the most

important in Southern Province. It is caused by the protozoan Theileria parva, and

transmitted by the brown ear ticks Rhipicephalus appendiculatus and Rhipicephalus

zambenziensis. The disease poses a threat to cattle upon which most local people

depend for draught power and income. High cattle mortalities due to this disease have

continued to be reported especially in the poorly managed traditional herds.

Previously, the Veterinary Department in Zambia attempted to control this

disease by enforcing a compulsory dipping programme (Berkvens, 1991; Mulumba et

al., 2000). However, a number of problems were encountered such as high costs to

sustain the programme, infrastructural problems and farmers not willing to continuously

bring their animals for dipping (Berkvens, 1991). Therefore a more cost effective and

appropriate method of control was required.

In 1993, the Assistance to the Veterinary Services of Zambia Project (ASVEZA)

funded by the Belgian government was initiated to develop a better control strategy for

ECF in Southern Zambia. This involved identifying a local T. parva strain that could be

used for field immunisations. Theileria parva Chitongo was identified and it was proven

that immunisation with this strain protected calves against most T. parva strains in the

province. Since 1999 immunisation of cattle against ECF using the treatment and

Chapter 6

101

infection method was implemented as one of the most important control methods in

Southern Zambia. However, no study has ever been conducted to assess the perception

of farmers on these ECF immunisations in the province. Knowledge of how cattle

farmers perceive a given intervention is critical to the adoption, hence success of the

implementation of such a control strategy. Tatchell (1981) pointed out that when people

understand a control option and its benefits, they are likely to respond positively.

The present study was therefore designed to record the perception of cattle

farmers on the impact of ECF immunisations in ECF endemic areas of Southern Zambia

and to determine the major factors influencing that perception.

6.2 Materials and methods

A survey based on a face-to-face interview was designed to collect data on the

perception of traditional cattle farmers in the five immunising districts (Choma,

Kalomo, Mazabuka, Monze and Namwala) of Southern Zambia. The study was

conducted in June and July 2003. This coincided with the immunisation campaigns in

the area.

Stratified random sampling was used. Out of the 20-30 immunisation points in

each of the immunising districts 10 were selected. At least 3 farmers with immunised

animals were randomly selected at each immunisation point, giving 30-40

farmers/district. A total of 179 farmers were interviewed.

Veterinary Assistants (VAs) from the ASVEZA Project speaking the local

Tonga language were trained to interview the farmers based on a questionnaire. This

questionnaire included information on personal details, number of animals kept,

knowledge of ECF and ECF immunisations, reasons for using ECF immunisations,

preffered control strategy, affordability of ECF immunisations and the perception of the

102 Perception on ECF immunisations

ECF immunisation method. The questions were in English but later translated into the

local Tonga language and all responses were recorded on the questionnaire in English.

In addition, information on the incidences of ECF from 1998 to 2002 and other

relevant data was collected from the various official veterinary reports in these five

districts. This was done to compare farmers’ reports on the disease incidence to those of

official government reports.

6.2.1 Statistical analysis

Data analysis was carried out in STATA SE/8.0 (StataCorp., 2003). A logistic

regression model was fitted to compare the districts with respect to whether farmers had

reported ECF cases in their herds before. To investigate whether the farmer’s perception

of the effectiveness of ECF immunisation influenced the number of immunised animals,

a negative binomial regression with number of immunised animals as response variable

and farmer’s perception (very effective versus effective) as independent variable was

fitted as there were no respondents who stated that immunisations were ineffective.

Finally, the relationship between the number of cattle owned by the farmer and the

number of immunised calves was also investigated by a negative binomial regression.

6.3 Results

A total of 179 farmers from five districts of Southern Zambia owning a total of 10,582

cattle were interviewed.

Chapter 6

103

6.3.1 Sample characteristics and sanitary situation after the start of the ECF

immunisations

Table 4.1: Districts, total number of cattle farmers interviewed and cattle owned.

District District area

(sq km)

Total cattle kept Calves Cattle farmers interviewed

(n)

Choma 6,960.26 1,487 (14.1%) 476 38 (21.2%)

Monze 4,648.12 2,629 (24.8%) 786 46 (25.7%)

Mazabuka 6,313.98 1,339 (12.7%) 402 34 (19.0%)

Namwala 21,084.05 3,987 (37.7%) 876 31 (17.3%)

Kalomo 31,524.73 1,140 (10.8%) 319 30 (16.8%)

Total 70,531.14 10,582 (100.0%) 2,859 179 (100.0%)

Approximately 38% of the cattle in the questionnaire came from Namwala district.

Table 6.2: Ranking of cattle farmers by herd size and by district.

Herd size

% of farmers Choma Kalomo Monze Mazabuka Namwala n

1-10 12 5(13%) 5(17%) 3(7%) 7(21%) 2(6%) 22 11-20 21 7(18%) 7(23%) 10(22%) 9(26%) 4(13%) 37 21-30 19 8(21%) 5(17%) 10(22%) 6(18%) 5(16%) 34 31-40 9 4(11%) 3(10%) 0(0%) 4(12%) 5(16%) 16 41-50 6 2(5%) 2(7%) 4(9%) 1(3%) 1(3%) 10 51-60 4 4(11%) 0(0%) 1(2%) 0(0%) 3(10%) 8 61-70 6 2(5%) 3(10%) 4(9%) 2(6%) 0(0%) 11 71-80 4 3(8%) 0(0%) 2(4%) 1(3%) 1(3%) 7 81-90 4 0(0%) 1(3%) 3(7%) 1(3%) 3(10%) 8 91-100 4 1(3%) 3(10%) 2(4%) 1(3%) 1(3%) 8 101-150 5 2(5%) 1(3%) 5(11%) 0(0%) 1(3%) 9 >150 5 0(0%) 0(0%) 2(4%) 2(6%) 5(16%) 9 Total 100 38 30 46 34 31 179

104 Perception on ECF immunisations

The largest proportion of cattle farmers had a herd of 11-20 cattle (21%). Mazabuka

was represented best in this group (26%).

Table 6.3: Farmers reporting ECF cases in their herds before

District No Yes Total (n)

Choma 0 (0.0%) 38 (100%) 38

Monze 1 (2.2%) 45 (97.8%) 46

Mazabuka 4 (11.8%) 30 (88.2%) 34

Namwala 3 (9.7%) 28 (90.3%) 31

Kalomo 1 (6.5%) 29 (93.5%) 30

Total 9 (5.03%) 170 (94.97%) 179

Of the farmers interviewed only 5% reported not to have experienced ECF in their herds

before. There was no significant difference between districts in reporting ECF in their

herds before the questionnaire (p = 0.180).

Table 6.4: Year in which cattle farmers started immunising their animals by district

District Year Choma Monze Mazabuka Namwala Kalomo Total

1999 14 8 11 4 7 44(25%)

2000 14 19 7 11 9 60(33%)

2001 3 12 9 7 6 37(21%)

2002 7 7 7 9 8 38(21%)

Total 38 46 34 31 30 179(100%)

Most of the interviewed cattle farmers started immunising their cattle in 2000 (33%),

one year after immunisation campaigns were launched.

Chapter 6

105

Table 6.5: Calves immunised in 2002 by interviewed farmers, total cattle and number

of VAs and their transport situation per district.

District Calves

immunised

Cattle

owned

Total no.

of VAs

VAs with

motorbikes

Choma 293 1,487 19 (20.9%) 8

Monze 590 2,629 23 (25.3%) 4

Mazabuka 281 1,339 16 (17.6%) 6

Namwala 641 3,987 17 (18.7%) 6

Kalomo 237 1,140 16 (17.6%) 9

Total 2042 10,582 91 (100%) 33 (36.3%)

The highest percentage of calf immunisation was observed in Namwala (31%) followed

by Monze (29%) and the lowest percentage was noted for Kalomo (12%). There was a

strong relationship between the number of calves immunised and the total number of

cattle owned by farmers (p < 0.001). No association was found between the number of

calves immunised and number of VAs or the number of VAs with motorbikes in all the

districts.

106 Perception on ECF immunisations

Table 6.6: Death of calves reported by farmers after the 2002 immunisations

District Deaths reported

after immunisations

Deaths reported to local VA

Choma 23 12

Monze 29 23

Mazabuka 11 8

Namwala 21 13

Kalomo 2 1

Total 86 57 (66.3%)

Out of the 2042 calves immunised against ECF by the interviewed farmers in 2002,

4.2% died from various causes. The results indicate that most of these deaths were

reported in Monze (33.7%), with Kalomo reporting the lowest (2.3%). Only 66.3% of

these deaths were reported to the local Veterinary Assistant.

Table 6.5: Causes of death in calves reported by farmers to have died after ECF

immunisations

Cause of death number of deaths Percentage Period after immunisation

ECF (confirmed) 5 6 2 less than 35days, 3(1-12months) ECF (suspected) 4 5 1 - 12 months Anaplasmosis 30 35 1 - 12 months Heartwater (suspected) 4 5 over 12 months Black quarter 12 14 1 - 12 months Babesiosis 2 2 1 - 12 months Bloat 3 3 1 - 12 months Injuries 1 1 less than 35 days Diarrhoea infections 2 2 1 - 12 months

Unknown 23 27 5(<10days), 6(1-12months), 11(10-

35days), 1(> 1year)

Total 86 100

Chapter 6

107

Most of the 86 deaths reported by farmers seemed to have been due to anaplasmosis

(35%) while confirmed ECF deaths accounted for 6%. Most of the unknown deaths

were not reported to VAs. In cases reported to the VAs, blood and lymph smears were

taken and post-mortem conducted where possible. The information on all deaths and

VA’s action was solely based on farmers’ reports.

6.3.2 Perception of ECF immunisations by cattle farmers

Table 6.6: Responses given by farmers in each district on how they perceived the

immunisation effect.

Response

District Very effective Effective Don't know Total

Choma 33 5 0 38 Monze 41 5 0 46 Mazabuka 26 6 2 34 Namwala 23 7 1 31 Kalomo 30 0 0 30

Total 153 (85%) 23 (13%) 3 (2%) 179

Results indicate that 85% of farmers interviewed described ECF immunisations as being

very effective while 2% had no idea of the effect of immunisations. There were no

farmers stating that ECF immunisations were not ineffective.

Significantly more calves were immunised by farmers who judged immunisation to be

very effective compared to farmers who judged it to be merely effective (p < 0.001).

108 Perception on ECF immunisations

Table 6.7: Number and percentage of farmers owning cattle according to herd sizes

in relation to ECF control interventions preferred.

Control intervention preferred by farmers Total Herd size 1 2 3 4 5 6 n 1-10 0 9(41%) 1(5%) 0 11(50%) 1(5%) 22 11-20 0 20 (54%) 1(3%) 0 16(43%) 0 37 21-30 1(3%) 21(62%) 0 0 12(35%) 0 34 31-40 0 9(56%) 1(6%) 0 6(38%) 0 16 41-50 0 5(50%) 0 0 5(50%) 0 10 51-100 0 21(50%) 0 1(2%) 19(45%) 1(2%) 42 101-150 0 5(55%) 0 0 4(44%) 0 9 >150 0 2(22%) 1(11%) 0 6(67%) 0 9

Total 1(0.6%) 92(51.4%) 4(2.2%) 1(0.6%) 79(44.1%) 2(1.1%) 179 Control interventions: 1 = treatment, 2 = immunisation, 3 = tick control, 4 = treatment & immunisation, 5 = immunisation & tick control, 6 = immunisation, tick control & movement controls.

The majority of farmers (51.4%) interviewed preferred immunisation as a control

strategy for ECF, with those owning 21-30 animals recording the highest of this group.

Only 2.2% of the respondents preferred tick control. The least preferred strategies were

treatment (0.6%) and treatment & immunisation (0.6%).

Table 6.8: Reasons given by farmers on the willingness to continue immunising their

calves (n = 179):

Reason Number of cattle farmers (n)

Percentage

Results from first calves immunised were good

22 12.3%

Convinced that immunised calves are protected

135 75.4%

Seen from fellow farmers immunising regularly that their animals have stopped dying.

22 12.3%

Total 179 100.0%

Chapter 6

109

Table 6.9: Immunisation benefits mentioned by cattle farmers

Benefits Number of cattle farmers (n)

Percentage

Immunised calves don’t die from ECF compared to those that are not immunised.

146 81.6%

Immunised calves are easily treated and irregular dipping is possible.

22 12.3%

Immunisation is less expensive than other control strategies.

11 6.1%

Total 179 100.0%

Out of the 10 cattle farmers who had failed to immunise their animals in the previous

year (2002), 3 said that the immunising team never reached their area, 2 said that they

did not have calves to immunise and 5 said that they had no money to pay for the

immunisations.

6.3.3 Official statistics on ECF in Southern Zambia

Table 6.10: Reported ECF cases and deaths, 1998 – 2002. (Source: Southern

Province Veterinary Office Annual Report for the year 2002).

District ECF

cases

1998

ECF

deaths

1998

ECF

cases

1999

ECF

deaths

1999

ECF

cases

2000

ECF

deaths

2000

ECF

cases

2001

ECF

deaths

2001

ECF

cases

2002

ECF

deaths

2002

Choma 1425 NI 2046 1637 497 NI 825 269 828 249

Monze 2107 NI 1121 325 1019 263 816 325 1005 244

Mazabuka 1837 1252 1346 918 969 537 640 378 232 110

Namwala 390 166 2125 1516 549 268 193 106 558 331

Kalomo 2899 1652 1220 42 1201 508 410 142 340 156

Total 8658 3070 7858 4438 4235 1576 2884 1220 2963 1090

NI = No information

110 Perception on ECF immunisations

The table shows that reported ECF cases dropped from 8658 in 1998 (before

immunisations started) to 2963 in 2002 (after four years of ECF immunisations) –

giving a drop in reported cases of 66%.

Table 6.11: Number of calves immunised against ECF in five districts of Southern

Zambia from 1999 to 2002. (Source: ASVEZA-south Mazabuka, Annual internal report

for the year 2002).

District Immunisation period Total

May-

99 Sep-99

May-00

Sep-00 May-01

Oct-01 May-02

Oct-02

Monze 349 389 1003 654 2102 2336 2765 2935 12,533

Choma 389 237 121 702 1159 1627 1820 2995 9,050

Mazabuka 142 122 106 680 642 1096 1190 1299 5,277

Kalomo 0 205 454 580 532 918 1405 1300 5,394

Namwala 0 0 0 988 640 632 912 1586 4,758

Sub-Total 880 953 1,684 3,604 5,075 6,609 8,092 10,115

Total /Year 1,833 5,288 11,684 18,207

Grand total 37,012

The table indicates that the number of calves immunised yearly in the five districts

increased almost 10 times from 1999 to 2002.

6.4 Discussion and conclusions

All the farmers in the study were traditional cattle farmers with those owning 11 – 20

cattle comprising the largest group (21%). In this study 95% of farmers interviewed

Chapter 6

111

indicated that they had experienced ECF in their herds before, confirming that this

disease was endemic in these districts.

It was interesting to note that in the period that immunisation had been

implemented most cattle farmers (33%) only presented their animals for immunisations

in the second year. These results show that some farmers will only adopt a new

intervention after seeing its benefits from fellow farmers first. In addition, the number

of calves immunised was only associated with the number of cattle a farmer had and the

farmer’s knowledge of the benefits from the immunisations and not to the number of

Veterinary Assistants or their transport situation in a given district. Farmers with more

cattle are likely to have more draught power hence increased area under cultivation and

increased income from crops that enables them to pay for the immunisations. Probably

the number of VAs with motorbikes (36.3%) for all the districts are too few compared

to the areas of operation covered to make any meaningful impact on their extension

abilities. Woods et al. (2003) pointed out that provision of motorcycles to extension

officers increases their interactions with farmers and enhances their importance as

sources of information on management practices. This then contributes to improved

adoption of new and recommended interventions by farmers. Elyn (2002) reported the

lack of transport and/or inadequate transport both motorised (motorbikes) and non-

motorised (bicycles) as one of the problems veterinary assistants in Zambia were facing.

The study shows that a number of deaths do occur after animals have been

immunised against ECF. However, some of these deaths are not reported to the local

VA. Anaplasmosis, a tick-borne disease, seemed to be the main cause of death in cattle

after ECF immunisations based on farmers’ reports. This may mainly be due to the fact

that most farmers tend to relax on their tick control programmes once animals have been

immunised against ECF. The findings of this study are not in agreement with the

112 Perception on ECF immunisations

findings of Berkvens (1991) in Eastern Province where anaplasmosis was not a problem

after ECF immunisations. The possible explanation for the situation in Eastern Province

could be that of the absence of Boophilus decoloratus the main vector of this disease

which is said to have been replaced by B. microplus (Berkvens et al., 1998). However,

in Southern Zambia, B. decoloratus was found throughout the area and B. microplus

was not recorded (Speybroeck et al., 2002). In addition, Jongejan et al. (1988) reported

that Bos indicus cattle have a high resistance to B. microplus ticks thus reducing the

chances of transmitting the parasites. This is probably a contributing factor in that most

of the cattle in Eastern Province are indigenous Angoni (short-horned Zebu) cattle

(Berkvens et al., 1998) that are resistant to anaplasmosis whereas in Southern Zambia

most of the animals kept in the study area are crosses with Bos taurus breeds (own

observations). Farmers thus need to be advised to strategically continue

dipping/spraying their animals (at reduced frequencies) after ECF immunisations to

reduce tick burdens and control other tick borne diseases.

Most respondents in the study (85%) perceived ECF immunisation as being very

effective and only 2% had no knowledge on the effectiveness of immunisations. This

has led to most of them (51.4%) preferring immunisation to other ECF control

strategies. This is confirmed by the ever increasing numbers of calves immunised each

year in all the districts as seen from the official reports. In addition, the reduction in the

number of officially reported ECF cases and deaths may be partly due to increased

adoption of the ECF immunisations in the districts concerned. A combination of

immunisation and tick control came second (44.1%). Economically, immunisation

combined with seasonal tick control is the most profitable way of managing a herd

under a traditional production system (Minjauw et al., 1999). Therefore, farmers need to

be advised to go for this option.

Chapter 6

113

A number of farmers identified lack of money and inability of immunising

teams to reach their areas (most remote areas) as reasons for not having immunised their

animals. Woods et al. (2003) reported that the distance between a farmer’s household

and a veterinary facility had a negative influence on the way veterinary services were

used. The decision-making process by farmers on using recommended animal health

practices is also determined by their available income from the sale of crops and other

activities and their priority ranking of other necessities (Elyn, 2002). Financial

assistance to farmers and improving the mobility of veterinary assistants by providing

them with reliable motorised transport may increase the adoption of this important ECF

control strategy.

It is evident from this study that knowledge of the perception of livestock

owners on a new disease control intervention (e.g. ECF immunisation) is critical to the

success of implementing such a programme. The perception of small scale farmers

towards a particular technology affects their decision to adopt it (Adesina and Baidu-

Forson, 1995). These findings may be used in planning and implementing new animal

disease control programmes in the country.

114 Perception on ECF immunisations

6.5 References

Adesina, A. A. and Baidu-Forson, J. (1995). Farmers’ perception and adoption of new

technology: evidence from analysis in Burkina Faso and Guinea, West Africa.

Agriculture Economics 13, 1-9.

Berkvens, D. L. (1991). Re-assessment of tick control after immunization against East

Coast fever in the Eastern Province of Zambia. Annales de la Société Belge de

Médecine Tropicale 71 Supplement 1: 87-94.

Berkvens, D. L., Geysen, D. M., Chaka, G., Madder, M. and Brandt, J. R. A. (1998). A

survey of the ixodid ticks parasitising cattle in the Eastern Province of Zambia.

Medical and Veterinary Entomology 12, 234-240.

Central Statistical Office (CSO, Zambia) (2000). Agriculture Analytical Report for the

2000 census of population and housing. pp 32-45.

Elyn, R. (2002). Evaluation of delivery of animal health services in the Eastern and

Southern Provinces of Zambia, ASVEZA Report pp 37-50.

Jongejan, F., Perry, B. D., Moorhouse, P. D. S., Musisi, F. L., Pegram, R. G. and

Snacken, M. (1988). Epidemiology of bovine babesiosis and anaplasmosis in

Zambia. Tropical Animal Health and Production 20, 234-242.

Minjauw, B., Rushton, J., James, A. D. and Upton, M. (1999). Financial analysis of East

Coast fever control strategies in traditionally managed Sanga cattle in Central

Province of Zambia. Preventive Veterinary Medicine 38, 35-45.

Mulumba, M., Speybroeck, N., Billiouw, M., Berkvens, D. L., Geysen, D. M. and

Brandt, J. R. A. (2000). Transmission of theileriosis in the traditional farming

sector in the Southern Province of Zambia during 1995-1996. Tropical Animal

Health and Production 32, 303-314.

Chapter 6

115

National Livestock Economic and Information Centre (NALEIC) (2002). National

livestock census figures. Lusaka, Zambia.

STATACORP. (2003). Stata/SE 8.0. for Windows Statistical Software, Stata

Corporation, College Station, Texas.

Speybroeck, N., Madder, M., Van den Bossche, P., Mtambo, J., Berkvens, N., Chaka,

G., Mulumba, M., Brandt, J., Tirry, L. and Berkvens, D. (2002). Distribution and

phenology of ixodid ticks in Southern Zambia. Medical and Veterinary

Entomology 16, 430-441.

Tatchell, R. J. (1981). Current methods of tick control with special reference to

Theileriosis. In: Advances in the control of Theileriosis, Irvin, A. D.,

Cunningham, M. P. and Young, A. S. (Editors), Martinus Nijhoff Publishers,

The Hague, pp. 148-159

Woods P. S. A., Wynne H. J., Ploeger H. W. and Leonard, D. K. (2003). Path analysis

of subsistence farmer’s use of veterinary services in Zimbabwe. Preventive

Veterinary Medicine 61, 339-358.

116 General discussion

Chapter 7

General discussion

7.1 Introduction

East Coast fever (ECF) caused by the apicomplexan protozoa T. parva is economically

the most important tick-borne disease of cattle in Zambia and Southern Zambia in

particular. The disease exerts a lot of constraints in terms of limiting the productivity of

the existing smallholder cattle herds and restricting the adoption of specialised dairy

cattle by smallholder farmers. Other economic losses are the mortality, treatment and

other control costs. The transmission and infection dynamics of this disease in Southern

Zambia is not yet fully understood and it remains difficult to predict outbreaks as they

are the result of a complex interplay between environment, tick population and host

population (Mulumba et al., 2000, 2001). In order to design and implement good and

viable ECF control strategies, there is need to identify all the risk factors and quantify

their effect in disease transmission and infection dynamics in this area. In addition, the

infection and treatment method has been introduced in Southern Zambia as a method of

ECF control and its influence on disease transmission in the field and its impact in the

immunised animals needs to be evaluated. The objectives of this thesis work were all

focused on having a better understanding of the transmission and infection dynamics of

T. parva so that outbreaks and fatalities could be predicted.

This discussion is divided into three sections. In the first section, we highlight

the important findings of the T. parva cross-sectional serological surveys conducted in

all districts of Southern Zambia and also discuss the conclusions drawn from a

longitudinal study carried out for a period of eight years in two sentinel herds with

regard to T. parva contacts in traditionally kept cattle. This section discusses the

Chapter 7

117

evolution of the disease in the area and we link the disease evolution with the

immunisations that took place and the variability in climatic factors. We also discuss the

link between T. parva outbreaks and El Niño Southern Oscillation Ranks and the

implications it has on disease control.

The infection dynamics of T. parva stocks in the field in Zambia has not been

fully understood and the clinical picture in animals affected has not been documented. It

is well acknowledged that different stocks of T. parva from different locations exhibit

varying degrees of severity in infected animals. It is therefore important that the clinical

picture of the disease is well understood. We explore the possibility of using some

haematological parameters to predict case fatalities in T. parva experimentally infected

cattle during the clinical phase of the disease. The potential of using these findings in

immunisation and pickup trials is discussed in the second section. The perception of

traditional cattle farmers on ECF immunisations in Southern Zambia with regard to the

effectiveness and adoption of this control strategy is discussed in the third section.

7.2 Theileria parva sero-prevalence and transmission dynamics in traditionally

kept cattle and the El Niño events

Sero-prevalence surveys for T. parva have been used before to assess the extent of

exposure of cattle in the field (Gitau et al., 1997). Risk factors can also be evaluated

with the help of sero-epidemiological surveys. Gitau et al. (1997) reported that T. parva

antibody prevalences were strongly associated with the agro-ecological zones, breed of

cattle and grazing system. Studies to link T. parva sero-prevalences with potential risk

factors have never been conducted before in the Southern part of Zambia.

In the current study, the evolution of T. parva seroprevalence over a eight-

year period in all nine districts of Southern Zambia was evaluated as a function of

118 General discussion

geographical zone, time of sampling, climate variability and ECF immunisation. The

results from this study (Chapter 3) are that T. parva is more prevalent on the plateau

than in the valley or flat plains. This is in agreement with the observations of

Speybroeck et al. (2002) that the tick vector, R. appendiculatus, responsible for

transmitting T. parva, is widespread in the plateau districts compared to the valley or

flat plain districts. The strong positive association between sero-prevalence and ENSO,

the periodic climatic phenomenon, was remarkable. More T. parva sero-positive

samples were observed during the El Niño years (1997/98) compared to other years in

the study period. Furthermore, the study has demonstrated that the T. parva infection

prevalence increases dramatically six months after an El Niño event. This finding which

has never been recorded before in Southern Zambia and Zambia as a whole is highly

relevant as it provides an opportunity to veterinary officials to apply preventive

measures. For instance, farmers could be encouraged to invest more in intensive ECF

immunisations in endemic areas and tick control could be employed in epidemic areas.

This study further revealed increases in sero-prevalence in the March samplings

(rainy season) during the El Niño years in contrast to other years when the sero-

prevalence was always higher during the September samplings (dry season). This can

only be explained by the fact that the increased rainfall and humidity that come with an

El Niño event in Southern Africa provide favourable conditions for R. appendiculatus

and T. parva survival, facilitating the disease transmission.

No significant effect of immunisation on sero-prevalence could be detected. The

immunisation effect was completely masked by El Niño: including El Niño in the model

led to a significant immunisation effect.

Even with the current level of immunisation in the field, increased sero-

prevalence could not be detected widely. One possible explanation could be that the

Chapter 7

119

IFAT test (as used in this study) is insensitive in certain situations and as Billiouw et al.

(2005) pointed out it lacks sensitivity to pick up T. parva antibodies in non-epidemic

situations.

We have shown that climate variability in this study measured by MEI ranks

has a strong positive correlation with increased T. parva seroprevalence in cattle in

Southern Zambia. The MEI values which are easily available from the NOAA website

should therefore now be used when planning ECF control programmes in this part of

Zambia. The T. parva incidence forecasts based on MEI values will allow early

intervention, which will as well mitigate effects of epidemics and improve the cost-

effectiveness of control activities.

The transmission dynamics of T. parva in the field has been monitored using

longitudinal studies in which individual animals are followed from birth until they come

in contact with the disease (Moll et al., 1986; Mulumba et al., 2000, 2001). This way

information is obtained not only on the disease incidence, trends and time effects but

also on case fatalities and the prevalence of maternal antibodies in new-borne calves

(Billiouw et al., 2002). In Chapter 4 the transmission dynamics of ECF in Southern

Zambia were investigated based on such a longitudinal study and the El Niño Southern

Oscillation (ENSO) is evaluated as a predictor for ECF outbreaks.

El Niño Southern Oscillation (ENSO) has been used in diseases of man

especially malaria to predict epidemics (Bouma and Van der Kaay, 1996). Theileria

parva shares a number of features to Plasmodium, the parasite causing malaria such as

the life cycles and the transmission patterns in that both these diseases are transmitted

by vectors, R. appendiculatus ticks and Anopheles mosquitoes respectively. Although

much work has been done on malaria and its association with El Niño events, no work

has ever been reported to relate this climatic variability with T. parva outbreaks or any

120 General discussion

other disease of veterinary importance. It is a known fact that the transmission of most

vector-borne diseases tends to occur within seasonal patterns, in which the role of

rainfall and temperature is very important. Kovats (2000) stated that rainfall and

temperature were the principal drivers of the biological processes by which ENSO

affected health. Vectors carrying most pathogens are exposed to ambient weather and

their densities are influenced by the variability in weather which may as well affect the

way the disease is transmitted. In this study, we endeavour to explain the transmission

dynamics of T. parva in the Southern part of Zambia using ENSO cycles, looking for

similarities in the reported malaria transmission in man.

Results from the field studies described in Chapter 4 indicate that the hazard of

an animal coming into contact with T. parva went up sharply in the 1997/98 season

when El Niño was present compared to years without El Niño. The hazard rate rose in

both the epidemic and endemic areas during El Niño. The rate was higher in the

epidemic area during El Niño, whereas the hazard rate was always higher in the

endemic areas in other periods. The increased rainfall received during the El Niño years

of 1997/98 seemed to have favoured the increase in ticks on animals in our study and

aided in T. parva transmission as the micro-climate improved for these vectors. The

effect on the disease transmission was noticed more in otherwise arid area (Gwembe

valley – epidemic area). As Norval et al. (1992) stated, the survival and development of

free-living stages of R. appendiculatus are mainly influenced by the microclimatic

conditions prevailing in the habitats in which the ticks occur. Furthermore, Branagan

(1973) stated that the survival of R. appendiculatus was dependent on humidity and that

aridity was the limiting factor for the development and eclosion of the desiccation-

vulnerable eggs, excluding the establishment of permanent populations in arid areas.

Therefore, an improvement in rainfall and humidity in arid areas like the Gwembe

Chapter 7

121

valley in this study will tend to increase the survival of R. appendiculatus ticks

significantly. It is known that the abundance of the feeding stages of ticks on cattle

controls the level of T. parva infection within the tick population and hence the success

of T. parva transmission from the tick to the host animal (Norval et al., 1992). Since

animals in the arid areas always receive a low challenge to ticks and T. parva

transmission rates are not sufficient to provide protective immunity, a large susceptible

population existed when conditions for tick abundance and T. parva transmission

improved. It appears from our study that T. parva transmission in Southern Zambia is

restricted by the climate that is too dry, but that small changes in environmental

conditions can easily trigger an epidemic in such regions. There is evidence in human

health that increased rainfall in arid regions and droughts in humid climates can cause

malaria epidemics (Kovats et al., 2003). Allan et al. (1998) reported that the 1997-98 El

Niño was associated with heavy rainfall and flooding in North-eastern Kenya, a region

normally too dry for malaria transmission. However, from January to May 1998 a major

falciparum malaria epidemic occurred in a population that had no immunity. This is

probably also the case with T. parva transmission in Southern Zambia and possibly

other areas.

Kovats et al. (2003) suggested that evidence for an association between disease

risk and ENSO was more robust when based on long time-series studies. Our evidence

is indeed based on such a study. We followed up animals for a long period of time for

disease transmission and managed to link it to climate variability. The strong positive

association between T. parva outbreaks and ENSO in this study will not only contribute

to the understanding of ECF epidemiology in Southern Zambia but will also be used to

predict epidemics in both the epidemic and endemic areas allowing sufficient time to

mobilize resources.

122 General discussion

7.3 Predicting case fatalities in T. parva Katete experimentally infected animals

As stated in Chapter 5, the severity of T. parva infection has been described to be

determined by the quantity of the infective dose (Radley et al., 1974; Dolan et al., 1984)

and the strain of of the parasite (Mbogo et al., 1996). Up to now, anemia or the drop in

PCV has never been considered to be a major clinical sign in T. parva infections and

has never been associated with the severity of a clinical reaction. This may be due to the

fact that most of the early work on T. parva was carried out in Eastern Africa using the

local stocks (especially those constituting the Muguga cocktail). In the Eastern African

stocks anemia does not seem to be a major problem. For the Zambian T. parva stock,

however, the PCV dropped in all the animals with even more significant reductions

recorded in those undergoing lethal reactions, confirming that anaemia occurs in

infected animals. The mechanisms leading to anemia due to this T. parva stock is

largely unkown. One of the hypothesis is that parasite antigen-coated infected or

noninfected RBCs are removed by the immune system of the host animal, especially

during periods of increased parasitaemia. Parasitaemia was found to be higher in lethal

reactions than non-lethal reactions in the period 14 – 21 days. Studies of Shiono et al.

(2004) working with T. sergenti and Otsuka et al. (2001) working with B. gibsoni have

shown that when these infections occur the number of IgG-bound RBCs in circulation

that are prone to removal by phagocytes increases, especially with high parasitaemia.

This removal of IgG-bound cells then leads to anemia in the affected animals.

The finding of a strong positive association between animals with initial lower

MCV values (small RBCs) and increased severity and lethality to infection strengthens

this hypothesis. In addition, we have shown that small RBCs were more likely to be

Chapter 7

123

infected with T. parva than large ones. There are a number of factors contributing to

lower MCV values or smaller RBCs in cattle. They include chronic iron deficiency,

chronic inflammatory disease (Feldman et al., 2000) and aging of RBCs (Waugh et al.,

1992). Since all animals were closely monitored before and after the infections and did

not shown any sign of either iron deficiency or inflammatory disease, these could be

ruled out. We may hypothesise therefore that the cells were small because they were

old. The mechanisms to explain why some animals are not able to maintain a good

balance of old and young cells in circulation is beyond the scope of this work. As stated

in Chapter 5 older cells may easily be penetrated by T. parva merozoites leading to high

parasitaemia as observed in this study. This will then lead to the parasitized cells being

lysed or damaged by the parasite and unparasitised or parasitized cells being coated

with parasite antigen and eventually removed by the immune system. This subsequently

will cause anaemia in the affected animals. We may assume that this anaemia plays a

role in aggravating the respiratory distress in an animal leading to death. The above

findings suggest that cattle with a higher pool of smaller RBCs in circulation may be

more likely to succumb to T. parva infections. As such we could use MCV values

before starting T. parva experimental infections (especially immunisation pick up trials)

to predict the outcome of clinical reactions. This will help to make early decisions on

either treating or euthanizing an animal during experiments to prevent suffering. In

addition the findings add to the knowledge of the reactions enduced by T. parva Katete.

The use of MCV values to evaluate the course of a T. parva clinical reaction has never

been used before and we believe that there is a lot of potential in using it. However,

there is need to pursue further this finding with other T. parva stocks, in different cattle

breeds and other important blood parasites in animal health.

124 General discussion

7.4 Cattle farmers’ perception of ECF immunisation in Southern Zambia

In any disease control intervention strategy the perception of farmers with regard to

acceptability and sustainability of such programmes needs to be assessed. Studies of this

nature are not only critical to the success of the programme but also enable animal

health control planners to determine the impact of the control strategy on the disease

epidemiology. This is so because the rate at which farmers adopt a control strategy will

have quite some influence on the disease transmission in the field.

Results from this study have shown that most farmers (85%) perceved ECF

immunisations to be very effective and therefore more and more farmers have started

immunising their cattle, mainly after seing the benefits from fellow farmers immunising

their cattle. As stated in Chapter 6 the number of calves immunised was strongly

associated with the farmers’ knowledge of the benefits. This is very encouraging

especially where implementation of this control strategy is concerned as it is clearly

indicating that farmers have started appreciating the impact of ECF immunisations in

their herds. The other important finding in this study was the increasing number of

deaths caused by another tick-borne disease, namely anaplasmosis, in the immunised

calves. This was attributed mainly to the relaxation in tick control after immunisations.

This was found to be unusual especially when compared to the findings of Berkvens

(1991) in Eastern Province where anaplasmosis was not a problem in calves after ECF

immunisations. As stated in Chapter 6 the main difference is that the major vector,

B. decoloratus responsible for transmitting anaplasmosis is absent in Eastern Province

while it is abundant in Southern Zambia. The other tick species, B. microplus, that is

found in Eastern Province, is not an effective vector of anaplasmosis in this area either,

as most of the cattle here are the B. indicus type which are resistant to B. microplus ticks

(Jongejan et al., 1988). This is not the case in Southern Zambia where most of the cattle

Chapter 7

125

kept are B. taurus crosses. This finding, which has never been documented before in

Southern Zambia, is very important as farmers will now be advised to apply strategic

tick control interventions to protect their cattle against anaplasmosis after ECF

immunisations.

A number of problems were identified with regard to the adoption of

immunisation as the ECF control strategy. These included the lack of money on the

farmers’ side and the inability of government extension officers to reach the more

remote areas. For the extension staff, there was mainly the lack of adequate transport,

both motorised (motorcycles) and non-motorised (bicycles). From this study we

recommend that farmers should be assisted with small loans or advised on how they can

raise funds epecially from the sale of livestock or through the sale of crops or savings in

order to pay for the immunisations. Extension officers need to be provided with reliable

motorised transport by the government if this important ECF control strategy has to

yield meaningful results on a large scale in the area. The sustainability of this exercise

in the initial stages largely depends on the government veterinary service in making

efforts to sensitise farmers to adopt this control strategy. In the long run sustainability

will depend on more farmers able and willing to pay for this service so that more

stabilates could be made available or procured by the government. With its popularity

among cattle farmers, this method has been accepted and seems to be sustainable. If this

trend continues we could expect some effects on the transmission dynamics of the

disease in the near future.

126 General discussion

7.5 Conclusions and future research

We have shown in the studies carried out in this thesis that the transmission of T. parva

in the southern part of Zambia is mainly influenced by the variability in weather. The

T. parva sero-prevalence went up in all areas during an El ñino event masking the

effects of ECF immunisation despite the fact that animals were immunised to a large

extent. In addition, the presence of El ñino increased the incidence of T. parva contacts

causing epidemics especially in the usually dry areas of Southern Zambia where the

disease incidence is normally low. We therefore conclude that weather variability

expressed as the Multivariate ENSO Index (MEI) is the single most important risk

factor influencing ECF transmission dynamics in Southern Zambia. We recommend that

MEI indices be used in predicting ECF outbreaks, ECF infection prevalence and that

this concept is encorporated when planning ECF control programmes. The use of the

host MCV and PCV values in predicting the outcome of T. parva clinical infections is a

significant finding which will undoubtedly contribute to the early diagnosis and control

of this important disease and is also relevant in experimental infections.

Future research should be directed to the extension of the use of ENSO variables in

managing other tick borne diseases of cattle like anaplasmosis and babesiosis. It may

also have relevance to human medicine to see if malaria epidemics follow the same

pattern in Southern Zambia and Zambia as a whole. Research should also be focused

on studying the role of MCV in the pathogenesis of other T. parva stock infections in

different breeds of cattle and also the mechanisms involved in the development of

anaemia in animal infected with this T. parva Katete from Zambia.

Chapter 7

127

7.5 References

Allan, R., Nam, S. and Doull, L. (1998). MERLIN and malaria epidemic in north-east

Kenya. Lancet 351, 1966-1967.

Berkvens, D. L. (1991). Re-assessment of tick control after immunization against East

Coast fever in the Eastern Province of Zambia. Annales de la Société Belge de

Médecine Tropicale 71 Supplement 1: 87-94.

Billiouw, M., Vercruysse, J., Marcotty, T., Speybroeck, N., Chaka, G. and Berkvens, D.

(2002). Theileria parva epidemics: a case study in eastern Zambia. Veterinary

Parasitology 107, 51-63.

Billiouw, M., Brandt, J., Vercruysse, J., Speybroeck, N., Marcotty, T., Mulumba, M.

and Berkvens, D. (2005). Evaluation of the Indirect Fluorescent Antibody Test

as diagnostic tool for East Coast fever in eastern Zambia. Veterinary

Parasitology 127, 189-198.

Bouma, M.J. and Van der Kaay, H.J. (1996). El Niño Southern Oscillation and the

historic malaria epidemics on the Indian subcontinent and Sri Lanka: an early

warning system for future epidemics? Tropical Medicine and International

Health 1, 86-96.

Branagan, D. (1973). Observations on the development and survival of the ixodid tick

Rhipicephalus appendiculatus Neumann, 1901 under quasi-natural conditions in

Kenya. Tropical Animal Health and Production 5, 153-165.

Dolan, T. T., Young, A. S., Losos, G. J., McMillan, I., Minder, C. H. E. and Soulsby, K.

(1984). Dose dependent responses of cattle to Theileria parva stabilate.

International Journal for Parasitology 14, 89-95.

Feldman, B. F., Zinkl, J. G. and Jain, N. C. (2000). Schalm’s Veterinary Haematology.

Fifth Edition. Lippincott Williams and Wilkins, Baltimore USA.

128 General discussion

Gitau, G. K., Perry, B. D., Katende, J. M., McDermott, J. J., Morzaria, S. P. and Young,

A. S. (1997). The prevalence of serum antibodies to tick-borne infections in cattle

in small-holder dairy farms in Murang’a District, Kenya: a cross sectional study.

Preventive Veterinary Medicine 30, 95-107.

Jongejan, F., Perry, B. D., Moorhouse, P. D. S., Musisi, F. L., Pegram, R. G. and

Snacken, M. (1988). Epidemiology of bovine babesiosis and anaplasmosis in

Zambia. Tropical Animal Health and Production 20, 234-242.

Kovats, R. S. (2000). El Niño and human health. Bulletin of the World Health

Organization 78(9).

Kovats, R. S., Bouma, M. J., Hajat, S., Worrall, E. and Haines, A. (2003). El Niño and

health. Lancet 362, 1481-1489.

Mbogo, S, K., Kariuki, D. P., Nguni, P. N. and McHardy, N. (1996). A mild Theileria

parva parasite with potential for immunisation against East Coast fever.

Veterinary Parasitology 61, 41-47.

Moll, G., Lohding, A., Young, A. S. and Leitch, B. L. (1986). Epidemiology of

theileriosis in calves in an endemic area of Kenya. Veterinary Parasitology 19,

255-273.

Mulumba, M., Speybroeck, N., Billiouw, M., Berkvens, D. L., Geysen, D. M. and

Brandt, J. R. A. (2000). Transmission of theileriosis in the traditional farming

sector in the Southern Province of Zambia during 1995-1996. Tropical Animal

Health and Production 32, 303-314.

Mulumba, M., Speybroeck, N., Berkvens, D. L., Geysen, D. M. and Brandt, J. R. A.

(2001). Transmission of Theileria parva in the traditional farming sector in the

Southern Province of Zambia during 1997-1998. Tropical Animal Health and

Production 33, 117-125.

Chapter 7

129

Norval, R.A.I., Perry, B.D., Young, A.S. (1992). The Epidemiology of Theileriosis in

Africa. Academic Press, London.

Otsuka, Y., Yamasaki, M., Yamato, O. and Maede, Y. (2002). The effect of

macrophages on the erythrocyte oxidative damage and the pathogenesis of

anaemia in Babesia gibsoni-infected dogs with low parasitaemia. Journal of

Veterinary Medical Science 64, 221-226.

Shiono, H., Yagi, Y., Kumar, A., Yamanaka, M. and Chikayama, Y. (2004).

Accelerated binding of autoantibody to red blood cells with increasing anaemia

in cattle experimentally infected with Theileria sergenti. Journal of Veterinary

Medicine B51, 39-42.

Speybroeck, N., Madder, M., Van den Bossche, P., Mtambo, J., Berkvens, N., Chaka,

G., Mulumba, M., Brandt, J., Tirry, L. and Berkvens, D. (2002). Distribution and

phenology of ixodid ticks in southern Zambia. Medical and Veterinary

Entomology 16, 430-441.

Radley, D. E., Brown, C. G. D., Burridge, M. J., Cunningham, M. P., Peirce, M. A. and

Purnell, R. E. (1974). East Coast fever; quantitative studies of Theileria parva in

cattle. Experimental Parasitology 36, 278-287.

Waugh, R. E., Narla, M., Jackson, C. W., Mueller, T. J., Suzuki, T. and Dale, G. L.

(1992). Rheologic properties of senescent erythrocytes: loss of surface area and

volume with red blood cell age. Blood 79, 1351-1358.

130 Summary

Summary East Coast fever is a protozoan disease caused by the parasite T. parva and transmitted

by the three host ticks R. appendiculatus and the closely related R. zambenziensis. In

Zambia and Southern Zambia in particular, this disease poses a major constraint on the

improvement of cattle production especially in the traditionally kept cattle herds. This

study focuses mainly on the effects of environmental and host factors on the

transmission and infection dynamics of T. parva in the southern part of Zambia.

This thesis consists of the following parts: (1) the literature review and the

objectives of the study, (2) investigation on the transmission and infection dynamics of

T. parva in Southern Zambia and on the farmers’ perception of ECF immunisations and

(3) the general discussion where results are put in a wider perspective.

In Chapter 1, the important aspects on the history of East Coast fever, the

parasite causing the disease and the vectors including topics on the epidemiology,

transmission, immunology and control of the disease are reviewed. Emphasis in the

literature review is on the transmission and infection dynamics of the disease and risk

factors involved. The first chapter finishes with reviewing El niño Southern Oscillation

(ENSO) and its influence on the transmission of vector-borne infectious diseases.

Chapter 2 spells out the objectives of the thesis. The general objectives of this

thesis were to study the infection dynamics of ECF and host factors in experimental

T. parva infections in order to further optimize the ECF immunisation programme in

Southern Zambia.

Summary

131

In Chapter 3 results from sero-epidemiological surveys conducted in all the

nine districts of southern Zambia and involving 27,526 cattle over a period of eight

years are presented. It was demonstrated that T. parva, the parasite causing East Coast

fever (ECF) was found throughout Southern Zambia. However, higher values of

T. parva sero-prevalence were observed in the plateau districts of Monze, Choma and

Mazabuka than in the valley districts of Siavonga and Sinazongwe. Our results revealed

a strong association between high T. parva sero-prevalence and the presence of the

periodic climatic phenomenon known as the El Niño Southern Oscillation. More

T. parva sero-positive samples were recorded during El Niño years (1997/98) (P <

0.001) than other years in the study period. This association has led to the conclusion

that Multiple El Niño Southern Oscillation Indices could be used to predict years with

high or low ECF infection prevalence thereby contributing to the improved and

preventive control of ECF in the area.

Chapter 4 presents findings of a long-term study involving traditionally kept

cattle in Southern Zambia in two sentinel herds, one on the plateau (endemic area) and

the other in the valley region (epidemic area), with regard to T. parva transmission. The

study focussed on determining whether global weather changes had any influence on

disease transmission. The results from this study showed a strong association between

increased T. parva contacts in cattle and the presence of an El Niño event. This further

supports the finding of the first study that clearly linked the effects of climatic

variability to disease incidences. We propose that in Southern Zambia, the simple and

readily available Multiple El Niño Southern Oscillation Index (MEI) Ranks be used in

planning ECF control programmes and early warning.

132 Summary

The third study (Chapter 5) investigated haematological changes taking place in

cattle experimentally infected with the apicomplexan protozoa T. parva Katete. A

comparison of mean corpuscular volume (MCV) and packed cell volume (PCV) was

made between cattle undergoing lethal and non-lethal reactions. This work confirmed

that anaemia occurs in animals infected with T. parva Katete. The fall in PCV was

steeper in lethal reactions compared to non-lethal reactions. Our results show that

animals with initially lower MCV values are more prone to fatal reaction, despite

having normal PCV profiles before the experimental infection. The study also showed

that small red blood cells are more likely to be infected with T. parva. These findings

suggest that animals with a higher proportion of small red blood cells in circulation will

be more likely to succumb to T. parva infections.

The perception of cattle farmers on East Coast fever immnunisations in southern

Zambia has never been evaluated before. Therefore, a fourth study (Chapter 6) was

performed using a structured questionnaire to assess their perception on East Coast

fever (ECF) immunisation and to derive its impact on calf mortality in their herds. In

total 179 farmers from five districts in southern Zambia were interviewed. The majority

of farmers (85%) perceived ECF immunisations as being very effective and about half

of them (51.4%) preferred immunisations to other ECF control strategies. The study

shows that the number of calves immunised was strongly associated with the farmer’s

perception on the benefits of immunisations. There was no association between the

number of calves immunised and the number of veterinary assistants in a given district

or their transport situation. Overall mortality in ECF immunised calves from various

causes stood at 4.2%. Based on farmers’ reports, the majority of these deaths seemed to

have been caused by anaplasmosis, a tick borne disease which may be as a result of

Summary

133

relaxation on tick control after ECF immunisations. Strategic tick control after

immunisations to prevent other tick-borne diseases, especially anaplasmosis, is

recommended. The reasons identified by farmers for not immunising their animals

included failure by immunising teams to reach certain areas, not having calves of

immunising age and lack of money. These findings provide valuable information on

how livestock farmers perceive and adopt new animal disease control programmes in a

certain context. We recommend that information obtained in this study be used in the

planning and implementation of ECF immunisation programme in Southern Zambia.

In the general discussion (Chapter 7), the results of the different chapters are

discussed, highlighting the most important findings and implications on disease control.

We emphasise the use of MEI values in predicting high T. parva incidences and

outbreaks in Southern Zambia and the use of MCV and PCV values in predicting

fatalities in clinically infected animals.

134 Summary

Samenvatting Theileriose, ook wel “East Coast fever (ECF)” genoemd, is een protozoaïre ziekte die

veroorzaakt wordt door de parasiet Theileria parva en overgebracht wordt door de drie-

gastheren teken Ripicephalus appendiculatus en de daaraan verwante Ripicephalus

zambenziensis. Deze ziekte houdt een verbetering van de veeteeltproductie tegen in

Zambia and Zuidelijk Zambia, vooral bij de traditionele gehouden veestapels. Deze

studie legt de klemtoon op het effect van omgevings- en gastheer factoren op de

transmissie en infectiedynamiek van T. parva in the zuidelijk deel van Zambia.

Deze doctoraatsthesis bestaat uit de volgende delen: (1) een literatuuronderzoek

en de objectieven van de studie, (2) onderzoek naar transmissie en infectiedynamiek van

T. parva in Zuidelijk Zambia en naar de perceptie van de veehouders omtrent de ECF

immunisatie en (3) een algemene discussie die de resultaten in een algemener kader

plaatst.

In Hoofstuk 1 wordt een overzicht gegeven van de geschiedenis van theileriose

en wordt de parasiet die de ziekte veroorzaakt en de vector besproken met betrekking tot

de epidemiologie, transmissie, immunologie en controle van de ziekte. In de

literatuurstudie worden transmissie, infectiedynamiek en bijhorende risicofactoren in

detail besproken. Dit eerste hoofdstuk besluit met een overzicht van “El niño Southern

Oscillation (ENSO) “ en zijn invloed op de transmissie van infectieuze ziekten die

afhankelijk zijn van een vector om overgedragen te worden.

In Hoofdstuk 2 worden de objectieven van de thesis opgelijst. De algemene

doelstelling van deze thesis bestaat erin de infectiedynamiek van ECF en

Samenvatting

135

gastheerfactoren in experimentele T. parva infecties te bestuderen om de ECF

immunisatie campagnes in Zuidelijk Zambia verder te optimaliseren.

In Hoofstuk 3 worden de resultaten besproken van de sero-epidemiologische

surveys die in alle 9 districten van Zuidelijk Zambia werden uitgevoerd over een

periode van 8 jaren, en waarbij en in totaal 27,526 runderen werden getest. Uit deze

survey bleek dat T. parva, de parasiet die ECF veroorzaakt, algemeen voorkomt in

Zuidelijk Zambia. In de plateau districten Monze, Choma and Mazabuka was de

T. parva sero-prevalentie hoger dan in de vallei districten Siavonga en Sinazongwe. Er

was ook een sterke associatie tussen hoge T. parva sero-prevalentie en het voorkomen

van het periodisch weerkerend klimatologisch fenomeen ENSO. Meer T. parva sero-

positieve stalen werden opgemeten tijdens El Niño jaren (1997/98) (P < 0.001) dan in

de andere jaren van de studieperiode. Op basis van deze associatie concludeerden we

dat “Multiple El Niño Southern Oscillation Indices (MEI)” zou kunnen gebruikt worden

om te voorspellen of in een bepaald jaar de ECF infectie prevalentie hoog zal zijn, wat

dan toelaat om betere en preventieve maatregelen te nemen om ECF te controleren.

Hoofdstuk 4 is gebaseerd op de resultaten van een longitudinale studie waarbij

twee kuddes, die op een traditionele manier worden gehouden, intensief en over lange

tijd opgevolgd worden met betrekking tot T. parva transmissie. De eerste kudde bevindt

zich op het plateau in een ECF endemisch gebied, de andere kudde in de vallei in een

ECF epidemisch gebied. In de studie wordt de relatie tussen globale

weersveranderingen en ziekte-overdracht onderzocht. Uit de resultaten bleek dat er een

sterk verband bestaat tussen verhoogd T. parva contact en de aanwezigheid van een

El Niño gebeurtenis. Deze studie is bijgevolg een bevestiging van de vorige studie die

een zelfde verband aantoonde tussen klimatologische variabiliteit en incidentie van

136 Summary

ECF. Vandaar stellen we ook voor dat in Zuidelijk Zambia gebruik zou gemaakt

worden van de eenvoudig te bekomen en te gebruiken MEI Ranks bij het plannen van

ECF control campagnes en om veetelers te waarschuwen.

In de derde studie (Hoofdstuk 5) wordt onderzocht welke hematologische

veranderingen plaatsgrijpen in runderen die experimenteel geïnfecteerd zijn met het

apicomplex protozoa T. parva Katete. Het gemiddelde corpusculaire volume (MCV)

and het “packed cell volume” (PCV) werd vergeleken tussen runderen met letale en

niet-letale reacties. Anemie trad op bij alle dieren die geïnfecteerd werden met T. parva

Katete, maar PCV daalde sterker in dieren met een letale reactie. Bovendien werd

aangetoond dat dieren met initieel lage MCV waarden meer kand hebben op een letale

reactie, ondanks het feit dat alle dieren normale PCV profielen hadden voor the

experimentele infectie. De studie toonde verder ook aan dat kleine rode bloed cellen

meer kans hebben om met T. parva geïnfecteerd te worden. Bijgvolg hebben dieren met

een hogere proportie kleine rode bloed cellen in circulatie een hogere kans om te

bezwijken aan T. parva infecties.

The perceptie van veehouders met betrekking tot ECF immunisatie in Zuidelijk

Zambia is in het verleden nog nooit geëvalueerd. Daarom werd een vierde studie

(Hoofdstuk 6) opgezet en werd aan de hand van een questionaire de perceptie van

veehouders met betrekking tot ECF immunisatie en de impact van de immunisatie op

kalfmortaliteit in kaart gebracht. In totaal werden 179 veehouders uit vijf districten in

Zuidelijk Zambia geïnterviewd. Het overgrote deel van de veehouders (85%)

beschouwden ECF immunisatie als zeer effectief en ongeveer de helft ervan (51.4%)

gaf de voorkeur aan immunisatie eerder dan andere ECF controle strategiën. De studie

Samenvatting

137

toonde aan dat het aantal geïmmuniseerde kalveren sterk geassocieerd was met de

perceptie van de veehouder omtrent de voordelen van immunisatie. Er bestond voorts

geen associatie tussen het aantal geïmmuniseerde kalveren en het aantal veterinaire

assistenten in een bepaald district of hun transportmogelijkheden. De totale mortaliteit

in geïmmuniseerde kalveren was gelijk aan 4.2%. Volgens informatie van de

veehouders werd het grootste aandeel veroorzaakt door anaplasmosis, tevens een ziekte

die overgebracht wordt door teken. Dit zou bijgevolg kunnen veroorzaakt zijn doordat

tekencontrole na ECF immunisatie minder sterk wordt afgebouwd. Strategische controle

van teken na ECF immunisatie blijft dus aangeraden om andere ziekten die door teken

worden overgebracht, en dan vooral anaplasmosis, te voorkomen. Veehouders

rapporteerden als redenen waarom ze niet immuniseerden dat bepaalde gebieden niet

bereikt werden door de teams die immuniseren, dat ze geen kalveren hadden die konden

geïmmuniseerd worden en/of er een geldgebrek was. Dit zijn belangrijke conclusies

aangaande de perceptie van veehouders met betrekking tot nieuwe controleprogrammas

voor een bepaalde dierziekte. Deze informatie zou tevens moeten gebruikt worden bij

het plannen en implementeren van nieuwe ECF immunisatie campagnes Zuidelijk

Zambia.

In de algemene discussie (Hoofdstuk 7) worden de resultaten van de verschillende

hoofdstukken besproken en de belangrijkste conclusies en implicaties voor controle van

ECF benadrukt. We onderstrepen het belang van het gebruik van MEI waarden in het

voorspellen van hoge T. parva incidenties en ECF uitbraken in Zuidelijk Zambia en

daarnaast het gebruik van MCV en PCV waarden in het voorspellen van letale reacties

in geïnfecteerde dieren.

138 Curriculum vitae

Curriculum vitae

Born in Samfya, Zambia on 4th October 1964.

Married to Jane, and has two sons, Mwape and Masulani.

Education

1985 Completed Secondary School Education at Kenneth Kaunda

Secondary School, Chinsali, Zambia with a School Certificate

1987 - 1992 Studied for a Bachelor of Veterinary Medicine Degree (BVM)

at the University of Zambia

1998 - 1999 Masters Degree in Tropical Veterinary Medicine (MSc- Trop.

Vet. Med) at the University of Edinburgh, Centre for Tropical

Veterinary Medicine, Scotland, UK.

Additional Education

May – July 2002 Attended course in Epidemiology, Data Handling and

Changing Roles of State Veterinarians at the Centre for Ticks

and Tick-borne Diseases (CTTBD) in Lilongwe, Malawi.

Professional Experience

1993 - 1997 District Veterinary Officer for Choma, southern Zambia in the

Department of Veterinary Services, Ministry of Agriculture

1998 - Now Senior Veterinary Officer - Livestock Disease Control,

Southern Province, Ministry of Agriculture, Zambia.

February 2000 –

December 2002

Seconded to the Belgian funded Assistance to the Veterinary

Services in Zambia (ASVEZA) project that dealt with the

production of a local East Coast fever (ECF) immunizing

stabilate to use in the field in southern Zambia. Work

Curriculum vitae

139

responsible for included ECF epidemiological studies,

laboratory work, tick studies, ECF stabilate production,

titrations and post immunisation follow-ups to test the efficacy

of the stabilates

January – June 2000 Teaching Certificate students studying Animal Production and

Health at the Zambia Institute of Animal Health in Mazabuka,

Zambia.

Professional membership

Registered with the Zambian Board of Veterinary Surgeons,

Executive Member of the Veterinary Association of Zambia, southern Zambia

Representative 2000 – 2003.