EPIDEMIOLOGY OF TSETSE-TRANSMITTED TRYPANOSOMSIS AND  THE PRESENCE OF DRUG RESISTANT TRYPANOSOMES IN THE ABAY (BLUE NILE) BASIN OF NORTHWEST ETHIOPIA 

EPIDEMIOLOGIE DE LA TRYPANOSOMOSE TRANSMISE PAS LES TSETSE ET PRESENCE DE TRYPANSOMOSES CHIMIORESISTANTS DANS LE BASSIN DE L’ABAY (NIL BLEU) AU NORD-OUEST DE L’ETHIOPIE

Shimelis Dagnachew1, Arun K. Sangwan2 & Getachew Abebe3

1Bahir Dar Veterinary Laboratory, Amhara Region Beureau of Agriculture, BAHIR DAR, Ethiopia
2Department of Veterinary Parasitology, College of Veterinary Sciences, CCS Haryana    Agricultural University, HISAR125004, India
3Faculty of Veterinary Medicine, Addis Ababa University, P. O. Box 34, DEBRE ZEIT, Ethiopia
gkibret@yahoo.com

Résumé

            Une étude a été menée de septembre 2003 à avril 2004 dans les weredas de Dembecha et de Jabitehenan, dans la région du bassin de l’Abay, au Nord-Ouest de l’Ethiopie, pour évaluer la résistance aux trypanocides sur le terrain. L’enquête entomologique a révélé que Glossina m. submorsitans était la seule espèce de tsétsé présente dans la zone, aux côtés d’autres tabanidés et muscidés piqueurs. Les densités apparentes des mouches étaient beaucoup plus élevées (p<0,05) vers la fin de la saison des pluies (1,08 mouche/ piège/jour ; 8,78 mouches/piège/jour et 91 mouches/piège/jour) que pendant la saison sèche (0,68 mouche/piège/jour ; 0,35 mouche/piège/jour et 7,33 mouches/piège/jour) respectivement pour Glossina m. submorsitans, les tabanidés et les muscidés. Dans les zones à faible altitude (moins de 1 600 m au-dessus du niveau de la mer), la densité apparente de Glossina m. submorsitans était beaucoup plus élevée (p<0,05) que dans les zones d’altitude moyenne (entre 1 600 et 2 000 m au-dessus du niveau de la mer) pendant les deux saisons. Au total, une centaine d’animaux, dont 50 de chacun des deux villages à haut risque identifiés dans la zone et ayant des similarités agroécologiques, ont été sélectionnés pour évaluer leur résistance au chlorure d’isométamidium (ISMM) et à l’acéturate de diminazène (DIM). Il n’y avait pas de différence notable entre les estimations de courbe de survie faites par Kaplan-Meier au sein du groupe-témoin et du groupe sous traitement dans les deux villages (p>0,05). Les résultats de l’efficacité du DIM ont révélé que 16 animaux sont devenus vulnérables à l’infection récurrente à T. congolense, mais il n’y avait pas une grande différence entre le taux d’incidence trypanosomienne et le taux de récurrence trypanosomienne.

Summary

Introduction

            Tsetse transmitted animal trypanosomosis is a serious constraint to livestock production and agricultural development in Ethiopia. A total of 14.8 million cattle, 6.12 million sheep and goats, 1 million camels and 1.23 million equines are at risk of contracting trypanosomosis (MOA, 1995). Due to the advance of tsetse flies into formerly free areas now 220,000 km2 areas is estimated to be affected by tsetse flies (NTTICC, 1996). These areas follow the Baro/Akobo, Omo/Ghibe and Abay/Didessa valleys of the large rivers in the country. These areas possess the most arable land with a high potential for agricultural development due to high annual rainfall (Jemal and Hugh-Johns, 1995). In the Abay basin areas of northwest Ethiopia, tsetse transmitted trypanosomosis is becoming a serious threat for livestock production and agricultural activity but the disease surveys are lacking in the area. For the effective way of controlling tsetse-transmitted trypanosomosis the knowledge of insect biology and ecology, and the status of the disease prevalence is of paramount importance (Leak, 1999). The aims of the present study were to determine the seasonal prevalence of trypanosomosis, apparent density, distribution and species of its vectors and assess the presence of drug resistance of drug resistant trypanosomes in naturally infected cattle in the field.

Materials and Methods

Study area

            The study was conducted in 8 Peasant Associations (PAs) of Dembecha and Jabitehenan Weredas of West Gojjam Zone (100 30’ N and 370 29’ E) in Amhara Regional State of Northwest Ethiopia. The climate alternates long summer rainfall (June-September) with winter dry season (December-March) with mean annual rainfall of 1200-1600 mm. The mean temperature is between 10-200C and the altitude ranges from 1400-2300 m. Temechan and Bir tributaries in West Gojjam Zone join before entering the main river Abay (Blue Nile) bordering the study PAs. Ponds and marshy areas are also found in the lowland areas of the study area. The five different vegetation types namely savanna grassland, forest, riverine and bush land along with the recently expanded cultivated lands are found. These different vegetation types are mainly found in areas below 1700 m whereas above this altitude the land is occupied by cultivated lands and small areas left for grazing purposes.

Entomological survey

            The apparent density of tsetse fly and other biting flies in relation to season, altitude, trap and vegetation types were studied at selected sites of the area. The apparent density was determined based on the mean catches in traps deployed and expressed as the number of fly catch/trap/day (Leak et al., 1987).  Entomological data were collected in the two seasons. The flies were caught with Monoconical, Biconical or NGU traps baited with acetone and three week old cow urine (Brightwell et al., 1987). A total of 142 traps, 70 in late rainy season and 72 in dry season were deployed just before sunrise and kept in position for 72 hours. The species of tsetse fly was identified based on morphological characteristic (Mulligan, 1970) and for other biting flies according to their morphological characteristics such as size, color, wing venation structure and proboscis at the genus level (Walle and Shearer, 1997). Sexing was done for tsetse fly just by observing the posterior end of the ventral aspect of abdomen by a hand lens as a result male flies easily identified by enlarged hypophgeum.

Parasitological survey

            The study animals constituted about 600 herds of 30,000 cattle in the study area. The sampling strategy was cluster-sampling method and herds were considered as clusters. The sample size was determined on the basis of the expected prevalence of 20% and absolute desired precision of 4% at confidence level of 95%. As a result a total of 384 animals were needed to be sampled (Martin et al., 1987). But in case of cluster sampling the subjects are not independent and hence larger sample is required. Therefore, as rule of thumb double the number of animals required for simple random sample are needed for cluster sampling (Martin et al., 1987). So the optimum sample size for this study was about 800 and so 814 samples from 19 herds and 834 samples from 18 herds were taken during the late rainy season and in the dry season respectively. Blood samples were collected from ear vein of each animal using two haematocrit capillary tubes. The blood samples were examined for the presence of trypanosomes by the dark ground buffy coat technique (Murray et al., 1977; Paris et al., 1982) and anaemia was estimated by packed cell volume (Woo, 1970).  Confirmation of trypanosome species was done using morphological characteristics (Mulligan, 1970).

Assessment of trypanocidal drug resistance (Isometamidium block treatment study)

            Based on the results of the trypanosome prevalence Melkachaba village from Gedebe PA of Dembecha Wereda and Ergib village from Weynema PA of Jabitehenan were selected for ISMM block treatment study, using purposive sampling method because it had a prevalence rate of trypanosome above 20% and the complain  is of drug failure mentioned by the interviewed people in these PAs were suggestive too. This study was conducted from December 10, 2003 to March 10, 2004. The grazing area and watering points as well as the husbandry systems of the study animals were similar. DIM and ISMM were the trypanocidal drugs used in the study area. A total of 100 animals 50 from each village were selected with a simple random method. All these animals were treated with diminazene aceturate (DiminazeneTM, Lot No.9687, exp. 07-2006, Farvet Laboratories B.V., Bladel-Holland) at a dose rate of 7mg/kg body weight to eliminate the existing trypanosome infection. Cattle of each village were then randomized into 25 in ISMM treatment group and 25 in control (sentinel) group following the methods proposed by Eisler et al., 2000. Animals in each group were ear tagged using yellow plastic tags, which allow easy identification of animals during each visit for parasitological examination. After two weeks of the blanket treatment (day 0), one group was treated with ISMM (TrypamidiumTM, Lot/Batch: W391971 Aut. Av/Exp.:06/2008, MERIAL -17, rue Bourgelat 69002 LYON - France) at a dose rate of 1mg/kg body weight and the other group was left as untreated control group. Body weight of each animal was estimated using heart girth measuring tape (Arora et al., 1981). Blood examination of animals for trypanosome infection was conducted every two weeks starting from day of blanket treatment up to 12 weeks following ISMM block treatment i.e. day-14, 0, 14, 28, 42, 56, 70 and 84. In each group, cattle found infected with trypanosomes were treated with diminazene aceturate at a dose rate of 7mg/kg body weight.

Data analysis

            Stata version 7.0 Software (Stata, 2000) was employed for the analysis and interpretation of data collected. The apparent fly catches in relation to variables measured (season, altitude level, vegetation and trap types) were analyzed using Kruskal Wallis test. The prevalence of trypanosomosis with variables (altitude levels, sex and age) was compared by c2-test. The mean PCV of parasitaemic and aparasitaemic animals were compared with Student’s t-test. Survival analysis of time was done to ascertain the occurrence of ISMM resistance. Survival time is the time to first detection of trypanosome up to 8 weeks post treatment with ISMM at 1mg/kg body weight. The efficacy of diminazene aceturate treatment was assessed on the basis of whether or not parasitaemia followed within two weeks after each treatment of any cattle on day 0 to 84 (Rowlands et al., 2001). Interpretation of survival data (Eisler et al., 2000): If fewer than 25% of control cattle become infected within 8 weeks of exposure, then the challenge was insufficient to warrant ISMM prophylaxis, which would be undesirable on grounds of cost, possible side effects and unnecessary drug pressure tending to develop drug resistance. If more than 25% ISMM treated cattle become infected within 8 weeks of exposure, this was strongly suspicious of the occurrence of drug resistant trypanosome. Where there was evidence of drug resistance on the grounds of the number of ISMM treated cattle becoming infected within 8 weeks of exposure it may nevertheless be worth continuing prophylaxis in situations where the ratio of the mean hazard rate for the control and prophylaxis herds over 1-8 weeks greater than 2. The proportion of cattle becoming infected 8 weeks after ISMM treatment was calculated as number of cattle infected (failure) during 8 weeks after the start of ISMM block treatment study divided by the total number of cattle presented at day 14 when the 1st case diagnosed in ISMM block treatment study. Since all survival times were not exactly known Kaplan-Meier survival curves plotted for the control and treatment groups of animals found in two villages, to estimate probability of surviving upto 8 weeks post treatment study. Log-rank and Wilcoxon (Breslow) tests were used for the statistical evaluation of the equality of the survivor functions of the control and ISMM treatment groups of cattle. To analyze data on the efficacy of diminazene aceturate, trypanosome incidence rate and trypanosome infection recurrence (Rowlands et al., 2001) at each village were compared using Fisher’s exact test. Trypanosome infection recurrence rate was defined as the proportion of cattle that were found infected with same species of trypanosome among the total number of animals that were treated with diminazene aceturate at dose rate of 7mg/kg body weight before two weeks. Significant level was determined at p<0.05 for all statistical results.

Results

Entomological survey

            A total of 13,927 flies were caught during the late rainy season and 1,731 during the dry season. Tsetse fly accounted for 1.12% and 7.79%, tabanids 8.45% and 4.27% while muscids 90.42% and 87.92% during the late rainy and the dry seasons respectively. The only tsetse fly species detected was G. m. submorsitans. The tabanid flies included Tabanus, Haematopota and Chrysops while the Muscids were mainly Stomoxys. The mean catches of flies by three different types of traps (Monoconical, Biconical and NGU) during the first study season (late rainy season) are shown in table I. There was a significant difference between trap types for the mean catches of flies (Kruskal-Wallis test, c2 = 8.38 with 2 d.f. and p=0.0151). Since monoconical trap performed the best in the study area, it was used for determination of apparent fly density and statistical description.

Table I.  The mean fly catches in three trap types during the late rainy season.

There was significant difference between trap types for the mean catches of flies (Kruskal-Wallis test, c2 = 8.38 with 2 d.f. and p=0.0151).

Figure 1. Tsetse apparent density in late rainy season and in dry season in the Abay basin areas of northwest Ethiopia.

             There was significant difference between seasons in the apparent density of tsetse (Kruskal-Wallis test, χ2 = 4.319 with 1 d.f. p = 0.0377).

            The apparent fly density (fly/trap/day) was 1.08, 8.78 and 91 in the late rainy season and 0.68, 0.35 and 7.33 in the dry season for tsetse, tabanids and muscids respectively. There was a significant difference between seasons in the apparent density of tsetse (Kruskal-Wallis test, χ2 = 4.319 with 1 d.f. p = 0.0377) and other biting flies also (Kruskal-Wallis test, χ2 =13.268 with 1 d.f. p =0.0003). Altitude had a significant effect on the apparent density of tsetse in both the seasons (Kruskal-Wallis test, χ2 =3.65 with 1 d.f. p = 0.0002). Tsetse fly sexing in the study period indicated that the predominant sex was females with 88.78% counts during the late rainy season and 53.33% in the dry season. Female flies were caught at altitude levels up to 1780m while male flies were caught up to 1650m.

Parasitological and haematological findings

            The overall prevalence of trypanosomosis during both the seasons and relative prevalence of different trypanosome species are shown in table II. T. congolense was the most prevalent species followed by T. vivax and T. brucei. Mixed infections of T. congolense and T. vivax were also recorded. The prevalence was higher in the late rainy season than the dry season with a significant difference. The risk of infections in the dry season was 0.7 times (Odds ratio) lower than late rainy season (95% CI=0.53-0.92).

            The prevalence of trypanosome infection in late rainy season varied significantly between the lowland 19.87% (95% CI=0.16-0.23) and midland areas 13.39% (95% CI=0.10-0.17) (p=0.015, c2=5.9205 at 1d.f) and animals in the midland area were at lower risk from lowland area by 0.7 times odds ratio from lowland areas (95% CI=0.42-0.91). The predominant trypanosome infection was T. congolense at both altitude levels with 73.9% (95% CI=0.63-0.82) and 53.2 % (95% CI=0.38-0.67) in the lowland and midland areas respectively. The prevalence of trypanosomosis in the dry season was significantly varied in the lowland 17.62% (95% CI=0.14-0.21) and midland areas 6.54% (95%CI=0.04-0.09) (p = 0.000, c2 = 23.55 at 1 d.f) as animals in the midland area were at lower risk than animals in lowland areas with 0.3 odds ratio (95% CI= 0.15-0.47). The predominant trypanosome infection in lowland areas was T. congolense with 94.8% (95% CI=0.87-0.98) while in the midland areas it was T. vivax 53.8 % (95% CI=0.33-0.73).

Table II: Prevalence of trypanosome infection in two seasons in the Abay basin area of northwest Ethiopia

T.c=Trypanosoma congolense, T.v=T. vivax, T.b= T. brucei, mixed =T. congolense and T. vivax.

There was statistical significant difference between seasons (p=0.007, c2 = 7.3 at 1 d.f).

            The PCV (mean ± SD %) values of parasitaemic and aparasitaemic animals during the late rainy season were 20.7±3.5 and 26.6± 4.3 (p<0.001, 95%CI=25.3-25.9) while during the dry season 21.4± 3.6 and 26.6±4.3 (p<0.001, 95%CI=25.4-25.9) respectively. The range of PCV values in parasitaemic animals was from 11-35% and in aparasitaemic animals from 14-43% in the late rainy season while in the dry season from 14 - 32% in parasitaemic animals and in aparasitaemic animals from 16 - 44%. The PCV (mean ± SD %) values of animals in the lowland area was 24.8±4.8 (95% CI=24.4-25.2) and in the midland area 26.6±4.4 (95% CI=26.1-27.1) in the late rainy season while in the dry season the mean PCV (%) values was 24.5±4.1 (95% CI=24.1-24.9) in lowland areas and 26.8± 4.1(95% CI=26.5-27.3) in the midland areas. The analysis of variance between mean PCV values in the lowland and midland areas were significantly vary in the two seasons (p=0.000, F=30.8 with 3 d.f.). The mean PCV (%) values of parasitaemic animals in late rainy season in the lowland areas was 20.3± 3.5 (95% CI=19.5-21.0) and in the midland areas 21.6±3.5 (95% CI=20.5-22.6) while for aparasitaemic animals 25.9±4.4 (95%CI=24.2-26.2) and 27.4±4.0 (95% CI=25.4-28.1) respectively. In the dry season the mean PCV (%) values of parasitaemic animals were 21.4± 4.2 (95% CI=20.5-22.4) in the lowland areas and 21.2±3.6 (95% CI=19.7-22.7) in the midland while for aparasitaemic animals 26.3±3.9 (95% CI=25.9-26.5) and 27.2±3.8 (95% CI=26.8-27.7) respectively.

Assessment of trypanocidal drug resistance (Isometamidium block treatment study)

            On day -14 the prevalence of trypanosome infection was 30% (95% CI=0.17-0.44) in Ergib out of which 24% were T. congolense infections while in Melkachaba it was 34% (95% CI=0.21- 0.48)  out of which 28% of infections were T. congolense. At day 0 of the ISMM block treatment study 4% infection in treatment groups and 12% infection in control group (average 7%) in Ergib while 12% infection in treatment groups and 8% infection in control group of Melkachaba village average (10%) were diagnosed. Fourteen days after ISMM block treatment, one animal with T. congolense infection in the treatment group and 2 animals with T. congolense and T. vivax infection in the control group were found in Ergib village where as in Melkachaba 5 animals were parasitaemic with T. congolense in the treatment group and 3 animals with T. congolense infection in the control group. The mean prevalence rate of trypanosome infection during 8 weeks period at Ergib was 7.52% in the treatment group and 13.6% in the control group while in Melkachaba village 22.4% and 18.12% respectively. The prevalence of trypanosome infection in the ISMM block treatment study for two villages are shown in figures 2 and 3.

Figure 2: Prevalence of trypanosomosis in the control and ISMM treated groups of cattle in Ergib village in Abay Basin areas of northwest Ethiopia

Figure 3: Prevalence of trypanosomosis in the control and ISMM treated groups of cattle in Melkachaba village in Abay Basin areas of northwest Ethiopia

Survival analysis

            The 25% survival time for the treatment and control groups of cattle in Ergib was 42 days while in Melkachaba it was 28 days and there was overlapping in days in both groups at each village, however, the confidence interval varies between groups in each village as shown in table III. The mean hazard rates of the treatment and control groups of cattle in Ergib were 0.0056 and 0 .0097 while in Melkachaba 0.0178 and 0.0122 (table IV) respectively

Table III. Mean hazard rate and 25% survival time  of control and treatment groups upto day 56 of the ISMM block treatment study in Ergib and Melkachaba villages in the Abay basin areas of northwest Ethiopia.

a=Total no. of cattle presented at day 14 when the first case was diagnosed

b=Number cattle infected (failure) divided by total time at risk from day 0 to 56 of ISMM block treatment study

c=Time by which 25% of the animal become parasitaemic as a result of trypanosome infection.

Table IV. Hazard ratio and proportion of cattle becoming  parasitaemic in control and treatment groups  upto day 56 of ISMM block treatment study in Ergib and Melkachaba villages in the Abay basin areas of northwest Ethiopia.

a=Total no. of cattle presented at day 14 when the first case was diagnosed.

 b=The ratio of mean hazard rate of control to treatment groups of cattle in the ISMM block treatment study since day 0 upto day 56

c=The proportion of cattle becoming parasitaemic during the 8 weeks period of ISMM block treatment study since day 0.

            The hazard ratio of control to treatment groups of cattle for Ergib was 1.73 and for Melkachaba 0.68 and more than 25% of the animals at each village of the treatment group were found infected with trypanosome by day 56 (8th week) of the ISMM block treatment study. The probability to survive 8 weeks after the start of ISMM block treatment study varied between village and between cattle. However there was no statistical significant difference between Kaplan-Meier survival estimates of the control and treatment groups at each village (P>0.05) as shown in figures 4 and 5..

            In both villages the two tests performed on the Kaplan-Meier survivor function curves of the control and treatment groups of cattle showed no significant difference on the probability of surviving from infection by trypanosome pathogen consequently the prophylactic activity of ISMM was not as effective as recommended by drug manufacturers and the level of resistance was high along the previous indices

            Log-rank test (p =0.2503) and Wilcoxon (Breslow) test (p = 0.2815) for equality of survivor functions revealed no significant difference between control and treatment groups

Figure 4. Kaplan-Meier survival estimates and statistical test for the equality of the survivor function of the control and treatment groups of cattle in Melkachaba village in the Abay basin areas of northwest Ethiopia.

Group 0=Control group in Ergib village in the ISMM block treatment study from day 0 to day 84, Group1=Treatment group in Ergib village in the ISMM block treatment study from day 0 to day 84.

Figure 5. Kaplan-Meier survival estimates and statistical test for the equality of the survivor function of the control and treatment groups of cattle in Melkachaba village in the Abay basin areas o of the control and treatment groups of cattle in Melkachaba village in the Abay basin areas of of the control and treatment groups of cattle in Melkachaba village in the Abay basin areas of  northwest Ethiopia f

Control group in Melkachaba village in the ISMM block treatment study from day 0 to day 84, Treatment group in Melkachaba village in the ISMM block treatment study from day 0 to day 84.

            Log-rank test (p=0.2310) and Wilcoxon (Breslow) test (p=0.1159) for equality of survivor functions showed no significant difference between control and treatment groups.

Efficacy of diminazene aceturate treatment

            In Ergib village a total of 39 cattle (24 from control and 15 from treatment group) were found at least once parasitaemic in 84 days follow up. Out of these 26 were during the 8 weeks time and 22(87.5%) were due to T. congolense infection. Seven animals from 39 became parasitaemic at least twice during the 84 days of the study, 4 from control group and 3 from treatment group.  Five animals (2 from treatment group and 3 from control group) were recurrent infections during the 84 days while 3 of 5 recurrent infections were within 8 weeks time. All the recurrent parasitaemia were due to T. congolense infections. In Melkachaba village a total of 70 cattle (32 from control group and 38 from treatment group) were found to be at least once parasitaemic in 84 days follow up.  Out of these 52 were during the 8 weeks time and 47(90.62%) were due to T. congolense infection. A total of 21 animals (11 from treatment group and 10 from control group) of 70 became parasitaemic at least twice. On the other hand 11 (5 from treatment group and 6 from control group) of the 21 multiple parasitaemic cases were recurrent infections. Hundred percent of the recurrent infections were due to T. congolense. Even though the required doses of diminazene aceturate were given  some animals remained positive for trypanosome infections with T. congolense up to day 56 and this was detected in both villages. The trypanosome incidence rate and trypanosome infection recurrence indicated no significant difference during the 14 weeks observation time with Fisher’s exact test (p> 0.05). Higher recurrence infection of 68.75% was detected in Melkachaba than in Ergib village.

Discussion

            Glossina m. submorsitans was the only species of tsetse fly found in the areas with a higher apparent density of 1.08 fly/trap/day in the late rainy season and a lower density of 0.68 fly/trap/day in the dry season. This might be due to an absolute increase in the number of tsetse flies due to favorable environment such as enough moisture, vegetation growth and suitable habitat or spread of flies from the rivers and thickets where they usually inhabit during the dry season, to more open areas during the rains (Brightwell et al., 1987). Leak et al. (1987) cited the latter as possible reason for the high densities of G. pallidipes obtained during dry season when the traps were deployed in the Ghibe river valley. The increases in tsetse apparent density during the wet season has been reported in Ethiopia (Msangi, 1999) in Somalia (Mohammed-Ahimed et al., 1989) and Cote d’Ivore, Togo, Gabon and Zaire (Leak et al. 1987).

            The result of tsetse fly survey agreed well with the general knowledge on the ecology of tsetse species found in southwest Ethiopia for the morsitans group. Typical habitat pattern were found in the study area for the savannah species G. m. submorsitans that prefers for savanna grass, riverine, and forest ecology. The geographical distribution of G. m. submorsitans is concentrated in the lowland area as climatic conditions are more favorable. Most of the tsetse flies in the present study were caught in the lowland areas and the apparent density decreased as altitude increased. This trend supports earlier works by Langridge (1976), Tikubet and Gemechu (1984) and Leak and Mulatu (1993). Slingenbergh (1992) discussed the invasion of G. m. submorsitans in to the upper Didessa valley, which suggested that the invasion of tsetse began during the 1970s and was responsible for an evacuation of the human population from the Didessa valley at that time. The Didessa and Angar rivers are both tributaries of the Abay (Blue Nile) river. Ford et al. (1986) reported that 5902 km2 of the river basin of the Angar, Didessa and Wama valleys were infested by G. m. submorsitans and G. tachinoides. G. m. submorsitans has a more wider distribution habitat than G. tachinoides and G. pallidipes and is also an efficient vector of pathogenic trypanosome to domestic livestock. The advance of G. m. submorsitans in the Abay basin areas of northwest Ethiopia as seen in the present study could have great importance regarding the epidemiology of bovine trypanosomosis and human settlement. NGU traps are efficient for savanna species (Leak et al., 1987) but in this case monoconical traps were the best of three trap types used during tsetse fly sampling. Glossina m. submorsitans was detected in western Ethiopia Gullele/Tolly (Leak and Mulatu, 1993) and they indicated that biconical traps were not efficient. The apparent densities of the tabanid and muscid flies were 6 fly/trap/day and 91 fly/trap/day respectively in the late rainy season and 0.43 fly/trap/day and 7 fly/trap/day respectively in the dry season. Despite a high density of these biting flies, mechanically transmitted trypanosomes are not predominant in the area. The role of biting flies in the mechanical transmission of the T. vivax is not well studied while T. evansi is absent from the area.

            The higher prevalence of bovine trypanosomosis was found in the low altitude areas along the river valleys of Bir, Temechan and Abay compared to the mid altitude areas. The seasonal occurrence of the disease was also consistent with the general knowledge of the vectors of trypanosomosis and hence it was higher during the late rainy season. The most prevalent trypanosome species in tsetse-infested areas of Ethiopia are T. congolense and T. vivax. Rowlands et al. (1993) reported a prevalence rate of 37% for T. congolense in southwest Ethiopia. Abebe and Jobre (1996) reported an infection rate of 58.5% for T. congolense, 31.2% for T. vivax and 3.5% for T. brucei in southwest Ethiopia. Different workers Afewerk et al, 2000; Muturi, et al., 2000; Tewelde et al., 2004) reported a prevalence of 17.2%, 21% and 17.5% in Metekel district, Southern Rift Valley and Upper Didessa Valley of tsetse infested regions respectively and the dominant species was T. congolense. The prevalence of bovine trypanosomosis in North Omo Zone in the dry and wet season was 14.2% and 21.5% (Muturi, et al., 2000) respectively. The dominant trypanosome species was T. congolense (66.1%) followed by T. vivax (20.8%). The predominance of T. congolense infection in cattle may be due to the high number of serodemes of T. congolense as compared to T. vivax and the development of better immune response to T. vivax infected animals (Leak 1999; McLennan 1980). There was significant difference (p<0.05) between the seasons as higher prevalence of trypanosomosis was found in the late rainy season (17.07%) than dry season (12.35%). The risks of trypanosomosis in cattle were 0.5 times lower in the dry season than in the late rainy season. The concurrent tsetse survey at the same time in the same altitude areas revealed that higher apparent density was seen in the late rainy season than the dry season. Similar results reported by Muturi et al., (1999) in North Omo Zone where higher prevalence of trypanosomosis was found in the wet season than the dry season.

            As anaemia is the classical symptom of the disease pathogenicity (Murray and Dexter, 1988) the low PCV in parasitaemic animals could have contributed in reducing the mean PCV for cattle in the lowland area and midland areas. Other diseases considered to be affecting the PCV values in animals in the study area are helmenthiasis, tick-borne diseases and nutritional imbalances. On the other hand most of the parasitaemic animals in the lowland areas were in good body condition despite having low PCV. This could be attributed to the fact that animals in the low altitude were at high plane of nutrition due to availability of sufficient pasture compared to animals in mid and high altitudes. In the two villages T. congolense was the dominant parasite and similar results were reported in tsetse infested areas particularly in savannah species of tsetse (Afewerk et al., 2000; Tewelde et al., 2004).

            The survival time analysis showed that 25% of the control group of cattle became infected with trypanosome parasite by day 42 in Ergib while equal proportion of cattle in the control group of cattle in Melkachaba were infected with trypanosome by day 28 of the study period. There was overlapping of the control and treatment group of cattle in both villages for the 25% survival time but the confidence interval varied between groups. The results indicated that the tsetse challenge was sufficiently high at all villages and a prophylactic regimen is more valuable if trypanosomosis is to be controlled efficiently but the prophylactic effect of ISMM was found to be low. However, protective levels of the treatment groups of cattle were the same in both villages and the prophylactic way of treatment was not effective and other approaches should be found. Based on Eisler et al. (2000), challenge was insufficient to warrant ISMM prophylaxis when fewer than 25% of the sentinel (control) cattle become infected within 8 weeks of exposure. This was justified on the grounds of cost, possible side effects and unnecessary drug pressure tending to develop drug resistance. In both villages more than 25% of the ISMM treated groups of cattle 27% and 61% in Ergib and Melkachaba were parasitaemic in 8 weeks following ISMM block treatment and this suggests is resistance of T. congolense against ISMM. Resistance against this drug was strongly suspected when more than 25% of ISMM treated cattle became parasitaemic within 8 weeks of exposure (Eisler et al., 2000). The ratio of the mean hazard rate for both villages was lower than 2 and hence it might be of no value to continue using ISMM at these villages. The similar results were reported by Tewelde et al. (2004) in one village of southwest Ethiopia from three villages.  In areas with evidence of drug resistance on the grounds of the ISMM treated cattle becoming infected within 8 weeks of exposure, it may nevertheless be worth continuing prophylaxis in situations where the ratio of the mean hazard rate for the sentinel and prophylaxis herds over 8 weeks is greater than 2 (Eisler et al., 2000). The Kaplan-Meier survival curves also in agreement with the results reported in the present findings. In general the probability to survive was some what higher in the treated group in Ergib than in Melkachaba but in both villages ISMM treatment had insignificant effect on survival time. ISMM treatment had minimal effect on survival times in high drug resistance situation in coastal areas of Kenya (Eisler et al., 2000) and southwest Ethiopia (Tewelde et al., 2004). The period of ISMM prophylaxis in infected cattle in the field was less than 1 month (Afewerk et al., 2000) in northwest Ethiopia. Rowlands et al. (1993) from southwest Ethiopia under laboratory condition indicated 1mg/kg body weight ISMM was less than 28 days in cattle challenged with clone of T. congolense that in therapeutic trials had been shown to be highly resistant to ISMM. The present result is also in agreement with previous works in southwest Ethiopia (Scott and Pegram, 1974; Codjia et al., 1993; Mulugeta et al., 1997; Assefa and Abebe, 2001; Chaka and Abebe, 2003; Tewelde et al., 2004), northwest Ethiopia (Afewerk et al., 2000) and in Southern Rift valley of Ethiopia (Muturi, 1999; Ademe and Abebe, 2001).

            The assessment of the efficacy of diminazene aceturate showed no statistical significant difference. However, the repeated recurrence infection of trypanosome after treatment with the recommended dosage in both villages was observed particularly in Melkachaba where 68.8% of recurrent infections recorded. Therefore, the recurrent infections might not be only due to new infections but also to resistant strains. Hence instead of fully depending on treatment by diminazene aceturate the principle of sanative pairs of treatment should be practiced in both villages. The comparisons of incidence rate and trypanosome infection recurrence rate in southwest Ethiopia showed drug resistant trypanosomes against diminazene aceturate (Rowlands et al, 2001) using a long-term study. Whiteside (1960) indicated that under-dosing, irregular use of prophylactics and discontinuation while at risk time and high incidence of trypanosomosis are the root cause of development of drug resistance. Clausen et al. (1992) and Geerts and Holmes (1998) stressed that the prolonged and frequent use of trypanocides in high challenge areas resulted in high selection pressure for resistance as well. The epidemiology of drug resistance population of trypanosome is dynamic when the incidence progressively spread within the population. For instance in Ghibe 7% recurrent infection in 1986 increased to 14% in 1989 (Rowlands et al., 1993). Transmission of resistant trypanosome by tsetse do not change the strains resistant after passage.  The trait remains stable for long time and is spread by cattle movement and or spread of tsetse population (Moloo and Kutuza, 1990). The method used here generates useful information on the efficacy of ISMM and diminazene aceturate to trypanosome populations present in the area where laboratory limitations are common. Hence 14-36 weeks prophylactic efficacy of ISMM was not recommended by this result. However, the method usually required long follow up and the interference of farmers on the study animals should be considered.

Acknowledgments

            We thank the Faculty of Veterinary Medicine of Addis Ababa University, Amhara Region Bureau of Agriculture, Amhara Region Agricultural Research Institute, Bahir Dar Veterinary Laboratory, Dembecha and Jabitehenan Weredas Office of agriculture and Debre genet Orthodox Church Monastery for their financial, logistic and other supports. The contribution of Dr. Nega Tewelde is also appreciated.

References

Abebe, G., Jobre, Y. (1996). Trypanosomosis: A threat to cattle production in Ethiopia. Rev. Med. Vet. 147, 897-902.

Ademe, M., Abebe, G. (2001). Field study on drug resistance trypanosomes in cattle (Bos indicus) in Kindo Koysha wereda, Southern Ethiopia. Bull. Anim. Hlth. Prod. 48, 131-138.

Afewerk, Y., Clausen, P. H., Abebe, G., Tilahun, G., Mehlitz, D. (2000). Multiple-drug resistant Trypanosoma congolense population in village cattle of Metekel district, northwest Ethiopia. Acta Trop.76, 231-238.

Arora, V. K., Sharma, R. C., Singh, B. P., Tamor, N. S., (1981). A note of body measurement in Hariana cow. Vet. Res. J. 4, 181-182.

Assefa, E., Abebe, G. (2001). drug resistant Trypanosoma congolense in naturally infected donkeys in North Omo zone, southern Ethiopia. Vet. Parasitol. 99, 261-271.

Brightwell, R., Dransfield, R.D., Korku, C.A., Golder, T.K., Tarimo, S.A. (1987). A new trap for Glossina pallidipes. Tropical Pest Management  33, 151-159.

Chaka, H. and Abebe, G. (2003). Drug resistant trypanosomes: a threat to cattle production in the southwest of Ethiopia. Revue Élev. Med. vét. Pays  trop. 56 (1-2): 33-36.

Clausen, P.-H., Sidibe, I., Kabore, I. & Bauer, B. (1992). Development of Multiple Drug Resistance of Trypanosoma Congolense in Zebu Cattle Under High Natural Tsetse Fly Challenge in the Pastoral Zones of Samorogouan, Burkina Faso. Acta Trop. 51, 229-236.

Codjia, V., Mulatu, W., Majiwa, P. A. O., d’Ieteren, G. D. M.& Peregrine, A. S. (1993). epidemiology of bovine trypanosomosis in the Ghibe valley, southwest Ethiopia. Occurrence of population of Trypanosoma congolense resistant to diminazene, isometamidium and homidium. Acta Trop. 53, 151-163.

Eisler, M.C., McDermott J., Mdachi R., Brandt J., Murilla G.A, Sinyangwe L., Mubanga, J., Machila N., Mbody Weightambo, H., Coleman, P.G., Clausen P.-H., Bauer, B., Sidibe, I., Geerts, S. & Peregrine, A.S. (2000). Rapid method for the assessment of trypanocidal drug resistance in the field. In: Proceeding of the 9th Symposium of the International Society for Veterinary Epidemiology and Economics (ISVEE9), Breckenridge, Colorado, USA, 6-11 August. 

Ford, J., Makin, M. J., Grimbler & R. J.  (1976). Trypanosomosis Control Program for Ethiopia. Ministry of Overseas Development of Great Britain,. p.1-30.

Geerts, S. and Holmes, P.-H. (1998). Drug management and parasite resistance in bovine trypanosomosis in frica. PAAT Technical and Scientific Series, no.1, vii+31p.FAO, Rome (Italy):                        

ISBN 92-5-104185-7.

Jemal, A., and Hugh-Johns, M. E. (1995). Association of tsetse control with health and productivity in the Didessa valley, western Ethiopia. Prev. Med. 22, 29-40.

Langridge, W.P. (1976). Tsetse and Trypanosomosis Survey of Ethiopia. Ministry of Overseas Department UK.  Pp.1-40.

Leak, S. G. A and Mulatu, W. (1993). Advance of Glossina morsitans submorsitans and G. pallidipes along the Ghibe river system in southwest Ethiopia. Acta Trop.,  55, 91-95.

Leak, S.G.A. (1999). Tsetse Biology and Ecology: Their Role in the Epidemiology and Control of Trypanosomosis. CABI publishing in association with the ILRI,. p152-210.

Leak, S. G. A  and Mulatu W. (1993). Advance of Glossina morsitans submorsitans and G. pallidipes along the Ghibe river system in southwest Ethiopia. Acta Trop., 55, 91-95.

Leak, S.K.A., Woume, K.A., Colardelle, C., Duffera, W., Feron, A. Mulingo, M., Tikubet, G. Toure, M & Yangari, G.  (1987). Determination of tsetse challenge and its relationship with trypanosomosis prevalence. In: livestock production in tsetse infested areas of Africa, ATLN, 43-52, Nairobi, Kenya.

Maclennan, K.J.R. (1980). tsetse transmitted trypanosomosis in relation to the rural economy. Wld. Anim. Rev. 36, 2-22.

Martin, S. W. Meek, A. H. and Willeberg P. (1987). Veterinary Epidemiology. Principles and Methods. Iowa State University Press/Ames.

Ministry of Agriculture (1995). Ethiopian Ruminant Livestock Development Strategy. Ministry of Agriculture (MoA), Addis Ababa, Ethiopia,. p.112-113.

Mohammed-Ahemed, M. M., Ahmed, A.I. and Ishag, A. (1989). Trypanosome infection rate of Glossina morsitans submorsitans in Bahr el Arab, South Darfur Province, Sudan. Trop. Anim. Hlth. Prod. 21, 239-244.

Mollo, S.K. and Kutuzu, S.B. (1990). Expression of resistance to isometamidium and diminazene in Trypanosoma congolense in boran cattle infected by Glossina morsitans centralis. Acta Trop. 47, 79-89.

Msangi S. (1999). distribution, density and infection rates of tsetse in selected sites of southern rift valley of Ethiopia. MSc Thesis, FVM, AAU Ethiopia.

Mulligan, H. W. (Ed.) (1970). The African Trypanosomosis. Ministry of Overseas Development and George Allen and Unwin, London.

Mulugeta, W., Wilkes, J., Mulatu, W., Majiwa, P. A. O., Masake, R. and Peregrine, A. S. (1997). long-term occurrence of Trypanosoma congolense resistant to diminazene, isometamidium and homidium in cattle at Ghibe, Ethiopia. Acta Trop. 64, 205-217.

Murray M., and Dexter M. (1988). Anaemia in bovine African trypanosomiasis. Acta Trop. 45: 389-432.

Murray M., Murray P.K. and Mc Intyre W. I. M. (1977).  An improved parasitological technique for the diagnosis of African trypanosomosis. Trans. R. Soc.Trop. Med. Hyg. 71, 325-326.

Muturi, K. S., Msangi, S., Munstermann, S., Clausen, P.-H., Abebe, G., Getachew, T., Bergenie, B. and Assefa, M. (1999). Trypanosomosis risk assessment in selected sites of the southern rift valley of Ethiopia, International Scientific Council for Trypanosomiasis Research and Control (ISCTRC) 25th Meeting, Mombasa, Kenya, Publication No.120.

NTTICC, (1996). Annual Report. Ministry of Agriculture, National Tsetse and Trypanosomosis Investigation and Control Center (NTTICC). Bedelle, Illubabor, Ethiopia. pp.29.

Paris, J., Murray, M. and McOimba, F. (1982). a comparative evaluation of the parasitological techniques currently available for the diagnosis of African trypanosomosis in cattle. Acta Trop. 39, 307-316.

Rowlands, G. J., Leak, S. G. A., Peregrine, A. S., Nagda, S. M., Mulatu, W. and d’Ieteren, G. D. M. (2001). The incidence of new and the prevalence of recurrent trypanosome infection in cattle in southwest Ethiopia exposed to a high challenge with drug-resistant parasite. Acta Trop. 79, 149-163.

Rowlands, G. J., Mulatu, W., Authie, E., Leak, S. G. A. and Peregrine, A. S. (1993). Epidemiology of bovine trypanosomosis in the Ghibe valley, southwest Ethiopia. Acta Trop. 53, 135-150.

Slingenbergh, J. (1992). Tsetse Control and Agricultural Development in Ethiopia. Wld.  Anim.  Rev., (FAO) 70-71, 30-36. 

Stata Corporation (2000). Intercooled Stata Version 7.0 For Windows 95/98/NT. University Drive East College   Station, Texas, USA.

Tewelde, T., Abebe, G., Eisler, M. C., McDermott, J., Greiner, M., Afewerk, Y., Kyule, M., Munstermann, S., Zessin, K.-H. and Clausen, P.-H. (2004). Application of field methods to assess isometamidium resistance of trypanosomes in cattle in western Ethiopia. Acta Trop. 90, 163-170.

Tikubet, G. and Gemechu T. (1984). Altitudinal distribution of tsetse flies in the Fincha valley (western part of Ethiopia). Insect. Sci. Application.   5, 389-395.

Waller, R. and Shearer, D. (1997). Veterinary Entomology. Arthropod Ectoparasites of Veterinary Importance. Champman and Hall, London, p. 141-193. 

Whiteside, E. F. (1960). recent work in Kenya on the control of drug-resistant cattle trypanosome. In: International Scientific Council for Trypanosomiasis Research and Control.8th meeting. Jos Nigeria Publication 62. Commission for Technical Co-operation, South of Sahara, London.

Woo, P. T. K. (1970). Haematocrit centrifugation technique for the diagnosis of African trypanosomosis. Acta Trop. 27, 384-386.