- Email: [email protected]
Distribution and prevalence of footrot in Bhutan
The Veterinary Journal The Veterinary Journal 171 (2006) 346–351 www.elsevier.com/locate/tvjl
Distribution and prevalence of footrot in Bhutan R.B. Gurung a, P. Tshering b, O.P. Dhungyel
c,* ,
J.R. Egerton
c
a
Regional Veterinary Laboratory, Bumthang, Bhutan Royal Veterinary Epidemiology Centre, Serbithang, Bhutan Faculty of Veterinary Science, University of Sydney, PM Bag 3, Camden, NSW 2570, Australia b
c
Accepted 9 November 2004
Abstract The first cases of footrot in Bhutan were reported in sheep in 1990 at the National Sheep Breeding Centre (NSBC), which supplies breeding animals to village sheep flocks throughout Bhutan. Despite the presence of footrot at the Centre the distribution of apparently disease-free sheep continued. Cases of footrot were reported in village flocks soon after the disease was diagnosed at NSBC. A national survey was designed to establish the distribution and prevalence of footrot in Bhutan. This detected footrot in 19/94 village sheep flocks surveyed. The 19 affected flocks were distributed among nine different administrative districts whereas the villages selected were in 13 of a total of 16 sheep growing districts. The highest within-flock prevalences were among the seven flocks sampled in Bumthang district (mean 20.4%). The prevalence of the disease within flocks was generally much lower in other affected districts and in three districts a single affected animal was identified in the sample of 14 sheep examined in each village. Nationally, footrot prevalence was estimated to be 3.1% (95% CI 2.16–4.04%). There was a positive association between the receipt of animals from NSBC and the presence of footrot. The prevalence of the disease was higher in flocks with a migratory system of management than in those using a sedentary system. The relative risk of there being footrot in a migratory flock was nine-times higher than in a non-migratory flock. Only one strain of Dichelobacter nodosus (serogroup B) was identified among the 234 isolates obtained from the 19 affected flocks. Sheep with footrot healed quickly when treated with a vaccine made from this strain. 2004 Elsevier Ltd. All rights reserved. Keywords: Footrot; Bhutan; Vaccination; Dichelobacter nodosus
1. Introduction In 1990, virulent footrot was diagnosed in sheep at the National Sheep Breeding Centre (NSBC), Bumthang, Bhutan, by Kelly (1990) and shown to be associated with isolates of Dichelobacter nodosus of serogroup B (Gurung, 2002). The Centre supplies breeding animals to sheep producers throughout Bhutan and animals continued to be supplied after footrot was diagnosed. Even apparently footrot-free sheep from affected flocks may relapse (Ghimire and Egerton, 1996) and *
Corresponding author. Tel.: +61 2 93511606; fax: +61 2 93511618. E-mail address: [email protected] (O.P. Dhungyel).
1090-0233/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.tvjl.2004.11.012
spread infection so for eight years potentially affected animals were supplied to all 16 of the districts in Bhutan that rear sheep. In Bhutan, sheep are managed either by a system which includes an annual summer migration with yaks to high altitude pastures returning to the villages in late autumn/winter, or a sedentary system in which animals are reared in and around the villages throughout the year. In an earlier, limited study conducted in villages surrounding NSBC, 12% of the animals were affected (Kelly, 1990). The present paper reports an investigation undertaken to establish the distribution and prevalence of footrot in Bhutan and to identify the strain(s) of D. nodosus involved.
R.B. Gurung et al. / The Veterinary Journal 171 (2006) 346–351
2. Materials and methods 2.1. National survey 2.1.1. Sampling procedure The survey was designed to determine the distribution of footrot throughout the sheep rearing districts of Bhutan and to estimate its prevalence with 95% confidence. In 1997 there were 29,108 sheep in Bhutan in 16 of its 20 administrative districts (Fig. 1). There were 251 defined villages in these 16 districts and thus the mean village sheep population at that time was 108. A two-stage sampling procedure was used. The sample size required for the present survey was based on an expected prevalence of 12% (Kelly, 1990) and was calculated using the software ‘‘Survey Toolbox for Livestock Diseases’’. Since there were no figures available for between and within village variance, values of 0.05 and 0.18 were used (assumed figures based on studies done in Southeast Asia; Cameron, 1999). Consequently it was calculated that we needed to sample 94/251 villages and to examine 14 animals per village flock in order to derive prevalence estimates with 95% confidence limits. Village flocks comprised the sheep owned by households within a village. Although owned by different households these sheep shared common pastures and were assumed to be exposed to the same diseases. Simple random sampling (SRS) was applied at the first stage giving every one of the 251 villages the same chance of being selected. Of the 94 villages selected 13 were, by chance, from the 16 sheep rearing districts. Two villages from one district had to be replaced: one for security reasons and the other as it had no sheep.
347
These villages were replaced with others from the same district. The identity of the villages to be sampled in each district was ascertained after the survey team reached each district headquarters and village names were substituted for the codes given in the design document. The second stage selection was done only when the survey team reached the selected villages. A list of households having sheep was prepared and households whose sheep were to be examined were then selected randomly from the list. The households were visited in the order of selection and all the sheep belonging to each were examined until the required number of 14 sheep for the village was reached. Because the number of sheep per household varied, the number of households sampled per village to obtain 14 sheep, also varied. 2.2. Inspection procedures Inspection of the migratory flocks was arranged during winter (November 2000–March 2001) when the animals had returned from the high altitude pastures. The non-migratory flocks were inspected in the period between August and October, 2001. The inspection team comprised a veterinarian (RBG), a laboratory technician, an extension worker from the district in which each village was located and a helper. The few animals belonging to households were confined, caught and examined. Individual feet of the animals were examined for signs of footrot and were scored 0, 1, 2, 3 or 4 depending on the severity of the lesion (Egerton and Roberts, 1971). Along with the clinical examination of animals a history of footrot in each of the villages was collected.
Fig. 1. Administrative districts of Bhutan.
348
R.B. Gurung et al. / The Veterinary Journal 171 (2006) 346–351
2.3. Bacteriology Samples for bacteriological culture were taken from sheep with clinical signs of footrot in all districts where the disease was seen. Lesion material was streaked onto previously prepared plates of hoof agar (HA) and placed in anaerobic jars gassed with the GasPak system (BBL, Becton Dickinson). These jars were taken to the nearest regional diagnostic veterinary laboratory and incubated at 37 C for at least four days. Colonies resembling D. nodosus were selected, sub-cultured and subsequently prepared in pure culture for serogrouping. Methods for isolation and antigenic characterisation described earlier (Claxton, 1989; Claxton et al., 1983; Gurung, 2002) were applied to isolates taken from affected animals during the survey and from animals in the vaccination trial. 2.4. Vaccine preparation and vaccination Whole cell vaccine was prepared from isolates of serogroup-B D. nodosus collected from two villages in the Bumthang district. Cultures of these isolates were grown on 4% HA and subsequently on 2% HA. These cultures were harvested in formal phosphate buffered saline (FPBS). After pooling, the suspension was concentrated to contain sufficient cells so that there were approximately 5 · 109 D. nodosus cells/mL of vaccine. After tests for sterility, these cells were emulsified in an oil adjuvant. The ratio of cell suspension to oil was kept at 40:60, respectively. Animals in the vaccinated group received two doses of this preparation subcutaneously, four weeks apart. The first dose of vaccine was administered in the first week of August 2001. 2.5. Vaccination trial Animals from affected flocks from four villages in the Bumthang district, which had been identified during the survey, were used to evaluate response to vaccination. With one exception, animals remained in their village environments for the duration of the trial. Because of its remoteness, five cases from one village were purchased and transferred for the duration of the trial to the Regional Veterinary Centre, Bumthang. Altogether 41 sheep were enrolled and individually identified with ear tags. These sheep were allocated to two groups, with 31 allocated to the treated group and 10 to the non-vaccinated control group (when vaccination commenced there were 20 affected animals in the treated group and five in the control group). 2.6. Assessment All the feet of each sheep in the trial were examined and scored at days 0, 30, 60 and 120 after vaccination
to assess the clinical response to vaccination. Affected animals were defined as those with score 2 or greater in one or more feet. Blood samples were collected from all the animals in the trial at the time of each clinical assessment. Sera from these samples were used to measure agglutinating antibody responses to vaccination and/or infection. 2.7. Statistical analysis National prevalence was calculated using the prevalence analysis programme contained in Cameron (1999). Difference between prevalence in migratory and sedentary flocks and between flocks with and without a history of receiving breeding animals from NSBC were calculated using the v2 test (EpiInfo 2000, Version 1.1.2). Clinical lesion scores and antibody titres for each assessment from the vaccination trial were entered and stored in a spreadsheet (MS Excel 2000). Differences in proportions of affected animals and antibody titres between treatment and control groups were statistically analysed using the v2 test (EpiInfo 2000, Version 1.1.2) and paired t test (MS Excel 2000), respectively.
3. Results 3.1. National survey 3.1.1. Distribution of footrot The 94 village flocks selected into the survey from 16 sheep-growing districts were located in 13 of these districts (Fig. 1). There were four districts in which no affected flocks were detected. In the nine districts, where the disease was present the mean flock prevalence ranged from 11.1% (Sarpang) to 100% (Bumthang) (Table 1). Overall the survey indicated that 20.2% of the flocks in the 13 districts were affected. Footrot prevalence within affected flocks ranged from 0 to 20.4% but, with one exception, flocks in affected districts had a prevalence <6%. The exception was the Bumthang district where 20.4% of the 98 animals examined were affected. There were usually only one or two cases in affected flocks but three of the seven Bumthang flocks had four or more cases in the 14 sheep examined. Nationally it was estimated that 3.1% of sheep were affected (95%; CI 2.2–4.0). 3.2. Other associations with footrot occurrence 3.2.1. Migratory vs. sedentary flocks In the sample of flocks examined 47 were managed by the migratory system and 47 by the sedentary system (Table 2). There was a significantly higher proportion of affected flocks among the migratory ones (17/47) compared with the sedentary ones (2/47) P < 0.001.
R.B. Gurung et al. / The Veterinary Journal 171 (2006) 346–351
349
Table 1 Distribution of footrot in districts of Bhutan District
Flocks sampleda
Tsirang Dagana Sarpang Thimphu Gasa Wangdue Mongar Paro Trashigang Bumthang Trongsa Chukha Samtse
9 24 9 4 3 4 5 2 5 7 9 5 8
Totals
94
a b c
Affected flocks
Flock prevalence %
Footrot prevalence % in affected flocks
Husbandry
0 0 1 0 1 2 1 0 1 7 2 1 3
0 0 11.1 0 33.3 50 20 0 20 100 22.2 20 37.5
0 0 0.79 0 2.4 5.4 2.9 0 0 20.4 3.17 4.29 5.36
Sb S S Mc M M M M M M M S M
19
20.2
3.41
47 M; 47 S
Fourteen sheep in each village flock. S, sedentary husbandry. M, migratory husbandry.
Table 2 Proportions of flocks/sheep affected in migratory and sedentary systems of management Husbandry system
Flocks affected/not affected
Animals affected/not affected
Migratory Sedentary
17 2
30 45
37 4
625 650
Totals
19
94
41
1275
3.2.2. Flocks supplied with animals from NSBC There was a significant difference in the proportions of affected flocks with and without a history of receipt of animals from NSBC. Sixty three flocks had this history and 18 were affected. All 31 flocks with no history of receipt of animals were unaffected (P < 0.01) (Table 3). 3.2.3. Clinical assessment of footrot from national survey The 41 cases of footrot identified during the whole survey period were from 19 different flocks. Score 4 lesions (under-running of the hard horn of the hoof) were present in 22 (53.65%) of these, score 3 (under-running of soft horn) in 11 (26.80%) and score 2 (interdigital dermatitis) in eight (19.50%). In some cases sternal ulceration was also observed indicating a history of severe
Table 3 Presence of footrot and receipt of breeding animals from NSBC History
Affected
Not affected
Total
Received Not received
18 0
45 31
63 31
Totals
18
76
94
lameness and recumbency. Collectively the distribution of severity scores is indicative of virulent footrot. 3.3. Vaccination trial 3.3.1. Clinical results Signs of footrot disappeared quickly in vaccinated animals. There were only two affected animals in this group 30 days after vaccination began. These two animals had recovered by day 60 (Table 4). New cases developed in the control group between days 0 and 30. Cases in the control group started to resolve between days 30 and 60 and continued between days 60 and 120. Between the groups there were significant differences in proportions affected between the two groups 30 and 60 days after vaccination (P = 0.0001; P = 0.0019, respectively) but at days 0 and 120 the differences were not significant, statistically.
Table 4 Response to vaccination with D. nodosus serogroup B in 41 cases from village flocks Proportion of animals affected (GMATa)
Treatment b
Vaccination (31) Nil (10) a b
Day 0
Day 30
Day 60
Day 120
20/31 (13.18) 5/10 (13.18)
2/31 (4446) 7/10 (13.18)
0/31 (21,930) 4/10 (14.13)
0/31 (3373) 1/10 (15.85)
Geometric mean agglutinin titre. Vaccine was administered on days 0 and 30.
R.B. Gurung et al. / The Veterinary Journal 171 (2006) 346–351 160
20
Rainfall
Rainfall (cm)
140
Temperature
120
15
100 80
10
60 40
5
Temperature (C)
350
20 0
0 August
September
October
November
December
Months
Fig. 2. Weather data for August to December 2001 at Bumthang.
Local weather data for the duration of the trial (August–November) are shown in Fig. 2. The temperature and rainfall for the first six–eight weeks of this period were suitable for the transmission of footrot but later was probably too cold (Graham and Egerton, 1968). 3.3.2. Serology Agglutinin titres of sera collected from sheep in the vaccination trial are presented in Table 4. Geometric mean titres (GMT) in both the test and control groups were 13.18 on day 0. On day 30 the GMT in the vaccinates increased to 4446 and to 21,930 on 60 day. Even on day 120 the level remained at 3373 whereas in the control group it was negligible throughout the trial (Table 4). Vaccine provoked antibody production soon after vaccination and the protective level of antibody (Egerton and Burrell, 1970) persisted for at least four months. On day 0 there were no differences in antibody titres between the groups (P = 0.96) but were significantly higher in vaccinates (P < 0.05) on days 30, 60 and 120. 3.4. Bacteriology 3.4.1. Isolates and their identity Samples for bacteriology were collected from 49 animals, 41 of which yielded D. nodosus. From these positive cultures 234 isolates of D. nodosus were collected. Isolates were obtained from all the clinically affected village flocks. Provisional serogrouping of all these isolates by slide agglutination test (Claxton et al., 1983) revealed that 213 of them were serogroup B. Twenty one isolates cross-reacted with antiserum to other serogroups. All these isolates were subsequently tested in the microplate agglutination test (Egerton, 1973; Raadsma et al., 1995) and all were identified as serogroup B. Genotypically their OMP gene restriction fragment length polymorphisms (RLFPs) were indistinguishable from those of the isolates from NSBC (Gurung, 2002).
4. Discussion The two-stage survey described here was designed with an expectation that about 12% of animals would be found affected. This prevalence estimate was based on a prior survey in Bumthang district where the NSBC is located (Kelly, 1990). The national prevalence estimate derived from the survey was much lower – 3.1% CI 95%. The occurrence of footrot throughout the country was characterised by a wide variation in estimates of its prevalence from village to village. The impact of the actual prevalence being less than the assumed value of 12% is that sample sizes could have been too low to detect a very low prevalence of footrot. Half the flocks included in the survey were examined during the time of the year when footrot transmits readily and we could reasonably have expected a prevalence of >30% in any affected flock if the disease were present. In flocks where there was no footrot observed there was no other evidence of its presence-history of lameness, misshapen hooves or sternal ulceration. We suggest that the flock prevalence observed in this survey is also too low because of the relatively small sample taken. It is of interest though that the estimate of flock prevalence is similar to the mean of the flock prevalences reported by Locke and Coombs (Locke and Coombes, 1994) in a number of districts in southern New South Wales in Australia. Depending on environmental factors and the level of disease control used the prevalence of footrot within flocks is more likely to be 30% or more. The finding of a single mild case in samples of 14 from three village flocks suggests that there had been prior culling and or treatment of cases. These mild cases, which are not likely to be lame, are less likely to be recognised as footrot cases by farmers. We are confident that our description of them as footrot was correct, however, because D. nodosus was isolated from them. Flocks in the Bumthang and similar districts reflected more accurately the prevalence and severity
R.B. Gurung et al. / The Veterinary Journal 171 (2006) 346–351
of the form of footrot in Bhutan. Owners in this area are very conservative and as Buddhists do not dispose of diseased animals by slaughter for meat. They have little access to nor can they afford treatments especially during the spring/summer migration to the mountain pastures. The disease in their flocks was very similar to that described previously at NSBC (Gurung, 2002). There are other possible explanations for the higher prevalence in Bumthang. It is possible that, because of their proximity to NSBC, villagers in this district acquired comparatively more breeding stock from the Centre but no data exist to support this view. It is considered also that pastures around the Bumthang villages are more favourable for the persistence of the disease. The apparent difference in flock prevalence between those subject to migration and those kept permanently about the villages may be confounded by differences in the cultural attitudes of the people involved. In Nepal, however, in communities perhaps more culturally uniform, footrot was more prevalent in migrating sheep and goats than in sedentary animals (Ghimire and Egerton, 1996). We are more confident of a causal association between receiving animals from NSBC and the presence or absence of footrot in flocks. Similar proportions of those practising sedentary management to those who migrated, had a history of contact with NSBC (data not presented). Additional evidence for the role of the breeding centre in spreading footrot comes from the bacteriology conducted as a component of this survey and the nature of the disease observed in the villages. All isolates of D. nodosus obtained from affected animals and characterised were indistinguishable from those from infected sheep at NSBC. They were of the same serogroup B, had the same gelatin-gel activity (Palmer, 1993) and also had the same OMP gene RLFP patterns (Gurung, 2002). The rapid response of cases in village sheep to treatment with serogroup B vaccine is further evidence for the occurrence of a single strain of D. nodosus in Bhutan. The occurrence of single strain of D. nodosus within and between many flocks is unique. Normally many strains whether characterised by serogroup or gelatin gel activity are present (Claxton et al., 1983; Hindmarsh and Fraser, 1985; Locke and Coombes, 1994). We conclude that the footrot observed throughout Bhutan is the virulent form of the disease irrespective of the occurrence of only a few mild cases in some of the village flocks. In two situations, at the NSBC (Gurung, 2002) and in the flocks of Bumthang described here there were high proportions of severe cases (scores 3 and 4). The same strain of D. nodosus was present in all cases irrespective of their severity and cases responded to serogroup B vaccination. The disease in the village flocks is
351
most likely to have originated in animals purchased from the previously infected NSBC.
Acknowledgements This study was funded by the Australian Centre for International Agricultural Research (ACIAR) and the Royal Government of Bhutan. We thank Professor Richard Whittington and Dr. Jenny-Anne Toribio for reading the manuscript and making valuable suggestions. The technical assistance of Craig Kristo, University of Sydney, laboratory staff of RVL, Bumthang and REVC, Serbithang, Bhutan is sincerely acknowledged. We thank Tall Bennett group for supplying vaccine adjuvant.
References Cameron, A., 1999. Survey Toolbox for Livestock Diseases. A Practical Manual and Software Package for Active Surveillance in Developing Countries. ACIAR Publication. Claxton, P.D., 1989. Antigenic classification of Dichelobacter nodosus. In: Egerton, J.R., Young, W.K., Riffkin, G.C. (Eds.), Footrot and Foot Abscess of Ruminants. CRC Press Inc., Boca Raton, pp. 155– 166. Claxton, P.D., Ribeiro, L.A., Egerton, J.R., 1983. Classification of Bacteroides nodosus by agglutination tests. Australian Veterinary Journal 60, 331–334. Egerton, J.R., 1973. Surface and somatic antigens of Fusiformis nodosus. Journal of Comparative Pathology 83, 151–159. Egerton, J.R., Burrell, D.H., 1970. Prophylactic and therapeutic vaccination against ovine footrot. Australian Veterinary Journal 46, 517–522. Egerton, J.R., Roberts, D.S., 1971. Vaccination against ovine foot-rot. Journal of Comparative Pathology 81, 179–185. Ghimire, S.C., Egerton, J.R., 1996. Transmission of footrot in migratory sheep and goats of Nepal. Small Ruminant Research 22, 231–240. Graham, N.P.H., Egerton, J.R., 1968. Pathogenesis of ovine foot-rot: the role of some environmental factors. Australian Veterinary Journal 44, 235–240. Gurung, R.B., 2002. Study of footrot in Bhutan, MVSc. Thesis, The University of Sydney, Sydney. Hindmarsh, F., Fraser, J., 1985. Serogroups of Bacteroides nodosus isolated from ovine footrot in Britain. The Veterinary Record 116, 187–188. Kelly, A. 1990. Short term consultancy on flock health of Bhutan– Australia Sheep and Wool Development Project. Locke, R.H., Coombes, N.E., 1994. Prevalence of virulent footrot in sheep flocks in southern New South Wales. Australian Veterinary Journal 71, 348–349. Palmer, M.A., 1993. A gelatin test to detect activity and stability of proteases produced by Dichelobacter (bacteroides) nodosus. Veterinary Microbiology 36, 113–122. Raadsma, H.W., Attard, G.A., Nicholas, F.W., Egerton, J.R., 1995. Disease resistance in Merino sheep. IV. Genetic variation in immunological responsiveness to fimbrial Dichelobacter nodosus antigens, and its relationship with resistance to footrot. Journal of Animal Breeding and Genetics 112, 349–372.
Distribution and prevalence of footrot in Bhutan R.B. Gurung a, P. Tshering b, O.P. Dhungyel
c,* ,
J.R. Egerton
c
a
Regional Veterinary Laboratory, Bumthang, Bhutan Royal Veterinary Epidemiology Centre, Serbithang, Bhutan Faculty of Veterinary Science, University of Sydney, PM Bag 3, Camden, NSW 2570, Australia b
c
Accepted 9 November 2004
Abstract The first cases of footrot in Bhutan were reported in sheep in 1990 at the National Sheep Breeding Centre (NSBC), which supplies breeding animals to village sheep flocks throughout Bhutan. Despite the presence of footrot at the Centre the distribution of apparently disease-free sheep continued. Cases of footrot were reported in village flocks soon after the disease was diagnosed at NSBC. A national survey was designed to establish the distribution and prevalence of footrot in Bhutan. This detected footrot in 19/94 village sheep flocks surveyed. The 19 affected flocks were distributed among nine different administrative districts whereas the villages selected were in 13 of a total of 16 sheep growing districts. The highest within-flock prevalences were among the seven flocks sampled in Bumthang district (mean 20.4%). The prevalence of the disease within flocks was generally much lower in other affected districts and in three districts a single affected animal was identified in the sample of 14 sheep examined in each village. Nationally, footrot prevalence was estimated to be 3.1% (95% CI 2.16–4.04%). There was a positive association between the receipt of animals from NSBC and the presence of footrot. The prevalence of the disease was higher in flocks with a migratory system of management than in those using a sedentary system. The relative risk of there being footrot in a migratory flock was nine-times higher than in a non-migratory flock. Only one strain of Dichelobacter nodosus (serogroup B) was identified among the 234 isolates obtained from the 19 affected flocks. Sheep with footrot healed quickly when treated with a vaccine made from this strain. 2004 Elsevier Ltd. All rights reserved. Keywords: Footrot; Bhutan; Vaccination; Dichelobacter nodosus
1. Introduction In 1990, virulent footrot was diagnosed in sheep at the National Sheep Breeding Centre (NSBC), Bumthang, Bhutan, by Kelly (1990) and shown to be associated with isolates of Dichelobacter nodosus of serogroup B (Gurung, 2002). The Centre supplies breeding animals to sheep producers throughout Bhutan and animals continued to be supplied after footrot was diagnosed. Even apparently footrot-free sheep from affected flocks may relapse (Ghimire and Egerton, 1996) and *
Corresponding author. Tel.: +61 2 93511606; fax: +61 2 93511618. E-mail address: [email protected] (O.P. Dhungyel).
1090-0233/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.tvjl.2004.11.012
spread infection so for eight years potentially affected animals were supplied to all 16 of the districts in Bhutan that rear sheep. In Bhutan, sheep are managed either by a system which includes an annual summer migration with yaks to high altitude pastures returning to the villages in late autumn/winter, or a sedentary system in which animals are reared in and around the villages throughout the year. In an earlier, limited study conducted in villages surrounding NSBC, 12% of the animals were affected (Kelly, 1990). The present paper reports an investigation undertaken to establish the distribution and prevalence of footrot in Bhutan and to identify the strain(s) of D. nodosus involved.
R.B. Gurung et al. / The Veterinary Journal 171 (2006) 346–351
2. Materials and methods 2.1. National survey 2.1.1. Sampling procedure The survey was designed to determine the distribution of footrot throughout the sheep rearing districts of Bhutan and to estimate its prevalence with 95% confidence. In 1997 there were 29,108 sheep in Bhutan in 16 of its 20 administrative districts (Fig. 1). There were 251 defined villages in these 16 districts and thus the mean village sheep population at that time was 108. A two-stage sampling procedure was used. The sample size required for the present survey was based on an expected prevalence of 12% (Kelly, 1990) and was calculated using the software ‘‘Survey Toolbox for Livestock Diseases’’. Since there were no figures available for between and within village variance, values of 0.05 and 0.18 were used (assumed figures based on studies done in Southeast Asia; Cameron, 1999). Consequently it was calculated that we needed to sample 94/251 villages and to examine 14 animals per village flock in order to derive prevalence estimates with 95% confidence limits. Village flocks comprised the sheep owned by households within a village. Although owned by different households these sheep shared common pastures and were assumed to be exposed to the same diseases. Simple random sampling (SRS) was applied at the first stage giving every one of the 251 villages the same chance of being selected. Of the 94 villages selected 13 were, by chance, from the 16 sheep rearing districts. Two villages from one district had to be replaced: one for security reasons and the other as it had no sheep.
347
These villages were replaced with others from the same district. The identity of the villages to be sampled in each district was ascertained after the survey team reached each district headquarters and village names were substituted for the codes given in the design document. The second stage selection was done only when the survey team reached the selected villages. A list of households having sheep was prepared and households whose sheep were to be examined were then selected randomly from the list. The households were visited in the order of selection and all the sheep belonging to each were examined until the required number of 14 sheep for the village was reached. Because the number of sheep per household varied, the number of households sampled per village to obtain 14 sheep, also varied. 2.2. Inspection procedures Inspection of the migratory flocks was arranged during winter (November 2000–March 2001) when the animals had returned from the high altitude pastures. The non-migratory flocks were inspected in the period between August and October, 2001. The inspection team comprised a veterinarian (RBG), a laboratory technician, an extension worker from the district in which each village was located and a helper. The few animals belonging to households were confined, caught and examined. Individual feet of the animals were examined for signs of footrot and were scored 0, 1, 2, 3 or 4 depending on the severity of the lesion (Egerton and Roberts, 1971). Along with the clinical examination of animals a history of footrot in each of the villages was collected.
Fig. 1. Administrative districts of Bhutan.
348
R.B. Gurung et al. / The Veterinary Journal 171 (2006) 346–351
2.3. Bacteriology Samples for bacteriological culture were taken from sheep with clinical signs of footrot in all districts where the disease was seen. Lesion material was streaked onto previously prepared plates of hoof agar (HA) and placed in anaerobic jars gassed with the GasPak system (BBL, Becton Dickinson). These jars were taken to the nearest regional diagnostic veterinary laboratory and incubated at 37 C for at least four days. Colonies resembling D. nodosus were selected, sub-cultured and subsequently prepared in pure culture for serogrouping. Methods for isolation and antigenic characterisation described earlier (Claxton, 1989; Claxton et al., 1983; Gurung, 2002) were applied to isolates taken from affected animals during the survey and from animals in the vaccination trial. 2.4. Vaccine preparation and vaccination Whole cell vaccine was prepared from isolates of serogroup-B D. nodosus collected from two villages in the Bumthang district. Cultures of these isolates were grown on 4% HA and subsequently on 2% HA. These cultures were harvested in formal phosphate buffered saline (FPBS). After pooling, the suspension was concentrated to contain sufficient cells so that there were approximately 5 · 109 D. nodosus cells/mL of vaccine. After tests for sterility, these cells were emulsified in an oil adjuvant. The ratio of cell suspension to oil was kept at 40:60, respectively. Animals in the vaccinated group received two doses of this preparation subcutaneously, four weeks apart. The first dose of vaccine was administered in the first week of August 2001. 2.5. Vaccination trial Animals from affected flocks from four villages in the Bumthang district, which had been identified during the survey, were used to evaluate response to vaccination. With one exception, animals remained in their village environments for the duration of the trial. Because of its remoteness, five cases from one village were purchased and transferred for the duration of the trial to the Regional Veterinary Centre, Bumthang. Altogether 41 sheep were enrolled and individually identified with ear tags. These sheep were allocated to two groups, with 31 allocated to the treated group and 10 to the non-vaccinated control group (when vaccination commenced there were 20 affected animals in the treated group and five in the control group). 2.6. Assessment All the feet of each sheep in the trial were examined and scored at days 0, 30, 60 and 120 after vaccination
to assess the clinical response to vaccination. Affected animals were defined as those with score 2 or greater in one or more feet. Blood samples were collected from all the animals in the trial at the time of each clinical assessment. Sera from these samples were used to measure agglutinating antibody responses to vaccination and/or infection. 2.7. Statistical analysis National prevalence was calculated using the prevalence analysis programme contained in Cameron (1999). Difference between prevalence in migratory and sedentary flocks and between flocks with and without a history of receiving breeding animals from NSBC were calculated using the v2 test (EpiInfo 2000, Version 1.1.2). Clinical lesion scores and antibody titres for each assessment from the vaccination trial were entered and stored in a spreadsheet (MS Excel 2000). Differences in proportions of affected animals and antibody titres between treatment and control groups were statistically analysed using the v2 test (EpiInfo 2000, Version 1.1.2) and paired t test (MS Excel 2000), respectively.
3. Results 3.1. National survey 3.1.1. Distribution of footrot The 94 village flocks selected into the survey from 16 sheep-growing districts were located in 13 of these districts (Fig. 1). There were four districts in which no affected flocks were detected. In the nine districts, where the disease was present the mean flock prevalence ranged from 11.1% (Sarpang) to 100% (Bumthang) (Table 1). Overall the survey indicated that 20.2% of the flocks in the 13 districts were affected. Footrot prevalence within affected flocks ranged from 0 to 20.4% but, with one exception, flocks in affected districts had a prevalence <6%. The exception was the Bumthang district where 20.4% of the 98 animals examined were affected. There were usually only one or two cases in affected flocks but three of the seven Bumthang flocks had four or more cases in the 14 sheep examined. Nationally it was estimated that 3.1% of sheep were affected (95%; CI 2.2–4.0). 3.2. Other associations with footrot occurrence 3.2.1. Migratory vs. sedentary flocks In the sample of flocks examined 47 were managed by the migratory system and 47 by the sedentary system (Table 2). There was a significantly higher proportion of affected flocks among the migratory ones (17/47) compared with the sedentary ones (2/47) P < 0.001.
R.B. Gurung et al. / The Veterinary Journal 171 (2006) 346–351
349
Table 1 Distribution of footrot in districts of Bhutan District
Flocks sampleda
Tsirang Dagana Sarpang Thimphu Gasa Wangdue Mongar Paro Trashigang Bumthang Trongsa Chukha Samtse
9 24 9 4 3 4 5 2 5 7 9 5 8
Totals
94
a b c
Affected flocks
Flock prevalence %
Footrot prevalence % in affected flocks
Husbandry
0 0 1 0 1 2 1 0 1 7 2 1 3
0 0 11.1 0 33.3 50 20 0 20 100 22.2 20 37.5
0 0 0.79 0 2.4 5.4 2.9 0 0 20.4 3.17 4.29 5.36
Sb S S Mc M M M M M M M S M
19
20.2
3.41
47 M; 47 S
Fourteen sheep in each village flock. S, sedentary husbandry. M, migratory husbandry.
Table 2 Proportions of flocks/sheep affected in migratory and sedentary systems of management Husbandry system
Flocks affected/not affected
Animals affected/not affected
Migratory Sedentary
17 2
30 45
37 4
625 650
Totals
19
94
41
1275
3.2.2. Flocks supplied with animals from NSBC There was a significant difference in the proportions of affected flocks with and without a history of receipt of animals from NSBC. Sixty three flocks had this history and 18 were affected. All 31 flocks with no history of receipt of animals were unaffected (P < 0.01) (Table 3). 3.2.3. Clinical assessment of footrot from national survey The 41 cases of footrot identified during the whole survey period were from 19 different flocks. Score 4 lesions (under-running of the hard horn of the hoof) were present in 22 (53.65%) of these, score 3 (under-running of soft horn) in 11 (26.80%) and score 2 (interdigital dermatitis) in eight (19.50%). In some cases sternal ulceration was also observed indicating a history of severe
Table 3 Presence of footrot and receipt of breeding animals from NSBC History
Affected
Not affected
Total
Received Not received
18 0
45 31
63 31
Totals
18
76
94
lameness and recumbency. Collectively the distribution of severity scores is indicative of virulent footrot. 3.3. Vaccination trial 3.3.1. Clinical results Signs of footrot disappeared quickly in vaccinated animals. There were only two affected animals in this group 30 days after vaccination began. These two animals had recovered by day 60 (Table 4). New cases developed in the control group between days 0 and 30. Cases in the control group started to resolve between days 30 and 60 and continued between days 60 and 120. Between the groups there were significant differences in proportions affected between the two groups 30 and 60 days after vaccination (P = 0.0001; P = 0.0019, respectively) but at days 0 and 120 the differences were not significant, statistically.
Table 4 Response to vaccination with D. nodosus serogroup B in 41 cases from village flocks Proportion of animals affected (GMATa)
Treatment b
Vaccination (31) Nil (10) a b
Day 0
Day 30
Day 60
Day 120
20/31 (13.18) 5/10 (13.18)
2/31 (4446) 7/10 (13.18)
0/31 (21,930) 4/10 (14.13)
0/31 (3373) 1/10 (15.85)
Geometric mean agglutinin titre. Vaccine was administered on days 0 and 30.
R.B. Gurung et al. / The Veterinary Journal 171 (2006) 346–351 160
20
Rainfall
Rainfall (cm)
140
Temperature
120
15
100 80
10
60 40
5
Temperature (C)
350
20 0
0 August
September
October
November
December
Months
Fig. 2. Weather data for August to December 2001 at Bumthang.
Local weather data for the duration of the trial (August–November) are shown in Fig. 2. The temperature and rainfall for the first six–eight weeks of this period were suitable for the transmission of footrot but later was probably too cold (Graham and Egerton, 1968). 3.3.2. Serology Agglutinin titres of sera collected from sheep in the vaccination trial are presented in Table 4. Geometric mean titres (GMT) in both the test and control groups were 13.18 on day 0. On day 30 the GMT in the vaccinates increased to 4446 and to 21,930 on 60 day. Even on day 120 the level remained at 3373 whereas in the control group it was negligible throughout the trial (Table 4). Vaccine provoked antibody production soon after vaccination and the protective level of antibody (Egerton and Burrell, 1970) persisted for at least four months. On day 0 there were no differences in antibody titres between the groups (P = 0.96) but were significantly higher in vaccinates (P < 0.05) on days 30, 60 and 120. 3.4. Bacteriology 3.4.1. Isolates and their identity Samples for bacteriology were collected from 49 animals, 41 of which yielded D. nodosus. From these positive cultures 234 isolates of D. nodosus were collected. Isolates were obtained from all the clinically affected village flocks. Provisional serogrouping of all these isolates by slide agglutination test (Claxton et al., 1983) revealed that 213 of them were serogroup B. Twenty one isolates cross-reacted with antiserum to other serogroups. All these isolates were subsequently tested in the microplate agglutination test (Egerton, 1973; Raadsma et al., 1995) and all were identified as serogroup B. Genotypically their OMP gene restriction fragment length polymorphisms (RLFPs) were indistinguishable from those of the isolates from NSBC (Gurung, 2002).
4. Discussion The two-stage survey described here was designed with an expectation that about 12% of animals would be found affected. This prevalence estimate was based on a prior survey in Bumthang district where the NSBC is located (Kelly, 1990). The national prevalence estimate derived from the survey was much lower – 3.1% CI 95%. The occurrence of footrot throughout the country was characterised by a wide variation in estimates of its prevalence from village to village. The impact of the actual prevalence being less than the assumed value of 12% is that sample sizes could have been too low to detect a very low prevalence of footrot. Half the flocks included in the survey were examined during the time of the year when footrot transmits readily and we could reasonably have expected a prevalence of >30% in any affected flock if the disease were present. In flocks where there was no footrot observed there was no other evidence of its presence-history of lameness, misshapen hooves or sternal ulceration. We suggest that the flock prevalence observed in this survey is also too low because of the relatively small sample taken. It is of interest though that the estimate of flock prevalence is similar to the mean of the flock prevalences reported by Locke and Coombs (Locke and Coombes, 1994) in a number of districts in southern New South Wales in Australia. Depending on environmental factors and the level of disease control used the prevalence of footrot within flocks is more likely to be 30% or more. The finding of a single mild case in samples of 14 from three village flocks suggests that there had been prior culling and or treatment of cases. These mild cases, which are not likely to be lame, are less likely to be recognised as footrot cases by farmers. We are confident that our description of them as footrot was correct, however, because D. nodosus was isolated from them. Flocks in the Bumthang and similar districts reflected more accurately the prevalence and severity
R.B. Gurung et al. / The Veterinary Journal 171 (2006) 346–351
of the form of footrot in Bhutan. Owners in this area are very conservative and as Buddhists do not dispose of diseased animals by slaughter for meat. They have little access to nor can they afford treatments especially during the spring/summer migration to the mountain pastures. The disease in their flocks was very similar to that described previously at NSBC (Gurung, 2002). There are other possible explanations for the higher prevalence in Bumthang. It is possible that, because of their proximity to NSBC, villagers in this district acquired comparatively more breeding stock from the Centre but no data exist to support this view. It is considered also that pastures around the Bumthang villages are more favourable for the persistence of the disease. The apparent difference in flock prevalence between those subject to migration and those kept permanently about the villages may be confounded by differences in the cultural attitudes of the people involved. In Nepal, however, in communities perhaps more culturally uniform, footrot was more prevalent in migrating sheep and goats than in sedentary animals (Ghimire and Egerton, 1996). We are more confident of a causal association between receiving animals from NSBC and the presence or absence of footrot in flocks. Similar proportions of those practising sedentary management to those who migrated, had a history of contact with NSBC (data not presented). Additional evidence for the role of the breeding centre in spreading footrot comes from the bacteriology conducted as a component of this survey and the nature of the disease observed in the villages. All isolates of D. nodosus obtained from affected animals and characterised were indistinguishable from those from infected sheep at NSBC. They were of the same serogroup B, had the same gelatin-gel activity (Palmer, 1993) and also had the same OMP gene RLFP patterns (Gurung, 2002). The rapid response of cases in village sheep to treatment with serogroup B vaccine is further evidence for the occurrence of a single strain of D. nodosus in Bhutan. The occurrence of single strain of D. nodosus within and between many flocks is unique. Normally many strains whether characterised by serogroup or gelatin gel activity are present (Claxton et al., 1983; Hindmarsh and Fraser, 1985; Locke and Coombes, 1994). We conclude that the footrot observed throughout Bhutan is the virulent form of the disease irrespective of the occurrence of only a few mild cases in some of the village flocks. In two situations, at the NSBC (Gurung, 2002) and in the flocks of Bumthang described here there were high proportions of severe cases (scores 3 and 4). The same strain of D. nodosus was present in all cases irrespective of their severity and cases responded to serogroup B vaccination. The disease in the village flocks is
351
most likely to have originated in animals purchased from the previously infected NSBC.
Acknowledgements This study was funded by the Australian Centre for International Agricultural Research (ACIAR) and the Royal Government of Bhutan. We thank Professor Richard Whittington and Dr. Jenny-Anne Toribio for reading the manuscript and making valuable suggestions. The technical assistance of Craig Kristo, University of Sydney, laboratory staff of RVL, Bumthang and REVC, Serbithang, Bhutan is sincerely acknowledged. We thank Tall Bennett group for supplying vaccine adjuvant.
References Cameron, A., 1999. Survey Toolbox for Livestock Diseases. A Practical Manual and Software Package for Active Surveillance in Developing Countries. ACIAR Publication. Claxton, P.D., 1989. Antigenic classification of Dichelobacter nodosus. In: Egerton, J.R., Young, W.K., Riffkin, G.C. (Eds.), Footrot and Foot Abscess of Ruminants. CRC Press Inc., Boca Raton, pp. 155– 166. Claxton, P.D., Ribeiro, L.A., Egerton, J.R., 1983. Classification of Bacteroides nodosus by agglutination tests. Australian Veterinary Journal 60, 331–334. Egerton, J.R., 1973. Surface and somatic antigens of Fusiformis nodosus. Journal of Comparative Pathology 83, 151–159. Egerton, J.R., Burrell, D.H., 1970. Prophylactic and therapeutic vaccination against ovine footrot. Australian Veterinary Journal 46, 517–522. Egerton, J.R., Roberts, D.S., 1971. Vaccination against ovine foot-rot. Journal of Comparative Pathology 81, 179–185. Ghimire, S.C., Egerton, J.R., 1996. Transmission of footrot in migratory sheep and goats of Nepal. Small Ruminant Research 22, 231–240. Graham, N.P.H., Egerton, J.R., 1968. Pathogenesis of ovine foot-rot: the role of some environmental factors. Australian Veterinary Journal 44, 235–240. Gurung, R.B., 2002. Study of footrot in Bhutan, MVSc. Thesis, The University of Sydney, Sydney. Hindmarsh, F., Fraser, J., 1985. Serogroups of Bacteroides nodosus isolated from ovine footrot in Britain. The Veterinary Record 116, 187–188. Kelly, A. 1990. Short term consultancy on flock health of Bhutan– Australia Sheep and Wool Development Project. Locke, R.H., Coombes, N.E., 1994. Prevalence of virulent footrot in sheep flocks in southern New South Wales. Australian Veterinary Journal 71, 348–349. Palmer, M.A., 1993. A gelatin test to detect activity and stability of proteases produced by Dichelobacter (bacteroides) nodosus. Veterinary Microbiology 36, 113–122. Raadsma, H.W., Attard, G.A., Nicholas, F.W., Egerton, J.R., 1995. Disease resistance in Merino sheep. IV. Genetic variation in immunological responsiveness to fimbrial Dichelobacter nodosus antigens, and its relationship with resistance to footrot. Journal of Animal Breeding and Genetics 112, 349–372.