Article Text

PDF

Paper
Tuberculosis in goats on a farm in Ireland: epidemiological investigation and control
  1. A. Shanahan, MVB1,
  2. M. Good, MVB, MRCVS2,
  3. A. Duignan, MVB, MRCVS2,
  4. T. Curtin, MVB, MRCVS3 and
  5. S. J. More, BVSc, PhD, FACVSc, MVB, DipECVPH, DipECBHM4
  1. Department of Agriculture, Fisheries and Food, District Veterinary Office, Farnham Street, Cavan, Ireland
  2. Department of Agriculture, Fisheries and Food, Agriculture House, Kildare Street, Dublin 2, Ireland
  3. Department of Agriculture, Fisheries and Food, District Veterinary Office, South Mall, Cork, Ireland
  4. Centre for Veterinary Epidemiology and Risk Analysis, School of Agriculture, Food Science and Veterinary Medicine, University College Dublin, Belfield, Dublin 4, Ireland
  1. Correspondence to Ms Good, e-mail: margaret.good{at}agriculture.gov.ie

This paper describes an outbreak of tuberculosis (TB) caused by Mycobacterium bovis in a dairy goat herd on a farm in Ireland, where 66.3 per cent of the herd tested positive to the single intradermal comparative tuberculin test (SICTT) at initial detection. An epidemiological investigation was conducted to determine the origin of the outbreak, considering issues such as animal movements and herd management practices. Infection was introduced with a consignment of goats, as determined by the variable number tandem repeat profile. Infection was eradicated using a test and cull programme involving the SICTT, the interferon-γ assay and a multiplex immunoassay (Enferplex TB).

Statistics from Altmetric.com

IN Ireland, the control of tuberculosis (TB), which is caused by Mycobacterium bovis, is well documented. An eradication programme for cattle was established in 1954 and the annual animal incidence now remains relatively stable, at approximately 0.4 per cent. Endemic infection in badgers (Meles meles), is a key impediment to eradication (More and Good 2006). M bovis infection occurs in many domestic, feral and exotic animals, including goats and sheep, and in human beings (O'Reilly and Daborn 1995). Cousins and others (1993) and Cousins (2001) cautioned that investigations should consider other in-contact domestic animals when TB has been confirmed in cattle. This has particular significance in cases where infected cattle and goats are kept on the same holding.

TB in goats can lead to generalised progressive disease with clinical signs such as respiratory disease, hoarse cough, weight loss and decreased milk production (Bernabé and others 1991). Infection in goats has been diagnosed in many countries, including South Africa (van den Heever 1984), unspecified African states (Ayele and others 2004), Russia (Kel'dybaev and Turkebaeva 1969), Australia (Cousins and others 1993), New Zealand (Coleman and Cooke 2001), France (Thorel 1980), Germany (Honeker 1951a, b), Spain (Alvarez and others 2008), the UK (Daniel and others 2009) and Brazil (Pignata and others 2009). TB in goats has previously been diagnosed at postmortem examination in Ireland (Department of Agriculture, Fisheries and Food [DAFF], unpublished observations).

EU Regulations (European Commission 2004a) require a TB control plan, approved by the national competent authority, for milking goats/sheep or other non-bovine species. Food business operators producing or collecting raw milk from species susceptible to TB (including goats) must ensure that it comes from herds regularly checked for TB under a control plan approved by the competent authority (ie, DAFF). This plan seeks first to establish the baseline status of any presence of infection with M bovis in the dairy goat/sheep herd by the use of the single intradermal comparative tuberculin test (SICTT) on all adults. Follow-up monitoring is conducted through further field testing (further biannual SICTT), complete postmortem examination of any goats presenting with clinical signs of TB and ongoing abattoir examination (with associated histopathology/microbiology of adult cull animals). In determining the frequency, section and percentage of the herd to be subjected to SICTT, other factors such as the husbandry (including outdoor access) and possible contact with other M bovis-susceptible species, including cattle and badgers, are also taken into consideration. All herds with cattle on the same holding are subjected to annual SICTTs. Since autumn 2008, all Irish goat herds supplying raw milk to an Irish food business operator must have a TB control plan.

The Irish 2008 sheep and goat census (DAFF 2008) records 7279 goats in 383 herds, with the majority of the herds (86 per cent) holding 20 goats or fewer. Many owners use goat products for personal consumption only or are hobbyists. Thirty herds had more than 50 goats, including 28 that supplied milk (2008/09). In the 12 months from mid-2008, approximately 5800 goats were tested as required to establish TB status under the control plan. TB was diagnosed in three herds, including two (farms A and C) that are discussed in detail in this paper.

This paper presents an investigation to determine and describe the source of an outbreak of M bovis infection in a newly established milking goat herd in Ireland and the use of test and cull procedures in the eradication of the infection.

Materials and methods

Epidemiological investigation

An epidemiological investigation to determine the source and potential for within-herd transmission was conducted by evaluating both the goat herd (housing, feeding, husbandry, preventive health programme, goat movements) and the farm (the integrity of the farm boundary, other susceptible species and opportunities for cross-species contact). Relevant history, including details of herd mortality and morbidity, were established. A qualitative assessment was undertaken to determine the risk of introduction of M bovis infection and/or onward transmission, based on evaluation of inward movements to, and biosecurity practices on, the farm; TB prevalence in the locality; and the likelihood of within-herd infection spread. Further assessments and control measures to minimise the potential for within-herd transmission were implemented based on diagnostic test results as they became available.

Diagnostic methods

Detailed information about the SICTT for the diagnosis of TB in cattle is available elsewhere (Monaghan and others 1994, de la Rua-Domenech and others 2006). Briefly, the test is conducted by separately injecting 0.1 ml each of M bovis (bovine) and Mycobacterium avium (avian) purified protein derivative tuberculin (PPD) intradermally into defined sites on the neck. The test is read 72 hours later by comparing the relative increase in skin fold thickness (an in vivo cell-mediated response) at each injection site.

The SICTT was conducted on adult goats using standard cattle methods, with modifications. An area approximately 10 cm square was clipped on each side in the mid-cervical region using cordless clippers with a size 40 blade (Oster Power Pro; Sunbeam Products). Injection site skin thickness, rounded to the nearest millimetre, was measured using sliding callipers with broad jaws designed to distribute an even manually applied pressure, and contemporaneously recorded. Intradermal injection of bovine and avian PPD (ID-Lelystad), 0.1 ml of each, was delivered on opposite sides of the neck using McLintock syringes (Barr Knight McLintock) and confirmed by palpation of an intradermal ‘pea’. The test was read 72 hours later and any increase in skin thickness and/or oedema at the injection sites was recorded. Severe interpretation of the SICTT was conducted; therefore, animals with an increase of more than 2 mm at the bovine site and a bovine reaction greater than the avian reaction were deemed reactors. In other words, animals classed as inconclusive reactors under the standard interpretation as per EU Directive 64/432/EEC (European Commission 1964) were designated positive. SICTT-positive animals were slaughtered and subjected to postmortem examination. The SICTT was repeated at approximately 60-day intervals on all adults and kids over six months of age in the herd with additional blood testing. SICTT-positive goats were removed in the interval between the first and second SICTTs.

A clinical examination was conducted on each goat in conjunction with the SICTT, emphasising examination of superficial lymph nodes, respiratory rate, body condition score, temperature, auscultation of the lungs and milk screening. Milk specimens from dairy goats with evidence of clinical mastitis (one doe in this case) and pooled bulk tank specimens were submitted to the Central Veterinary Research Laboratory (CVRL) for M bovis culture. Following decontamination with 5 per cent oxalic acid, culture was carried out using one tube of Stonebrink's medium, one tube of Lowenstein-Jensen medium with added pyruvate (LJP medium) and on the MGIT 960 culture system (BD Biosciences; Becton Dickinson). The isolates from the individual goat milk sample were identified as Mycobacterium tuberculosis complex by colony and cording characteristics and by AccuProbe (Gen-Probe).

DAFF veterinary staff conducted postmortem examinations at the abattoir where reactor animals were slaughtered. The examination included palpation and inspection of the lungs and mesenteric lymph nodes, with incision of the bronchial, mediastinal, hepatic and supramammary lymph nodes. Gross findings were recorded, and samples of visible lesions and other tissues including mammary glands and supramammary lymph nodes from each reactor animal were removed to the Regional Veterinary Laboratory (RVL) for more detailed examinations. Variable number tandem repeat (VNTR) strain typing was conducted by CVRL staff as described by Roring and others (2004).

The interferon-γ (IFN-γ) assay was conducted as described previously (Gormley and others 2006). The Enferplex ELISA was conducted as described by Whelan and others (2008). Neither test has yet been validated in this species.

A comparison of results from different diagnostic tests was measured using κ statistics and interpreted as follows: ≤0 Poor agreement, 0.01 to 0.2 Slight agreement, 0.21 to 0.4 Fair agreement, 0.41 to 0.6 Moderate agreement, 0.61 to 0.8 Substantial agreement, 0.81 to 1.0 Almost perfect agreement (Dohoo and others 2009).

Results

Farm A

Farm A is located in the south of Ireland, in an area of heavy, clay-like land with limited drainage, surrounded by heavy flat pasture land. It was acquired and specifically adapted in 2007 for the establishment of a dairy goat herd in place of the sheep enterprise that had previously existed on the farm. All goats for this new enterprise were sourced from farm B in October 2007. No other livestock were kept on the farm. In January 2008, significant health concerns including deaths occurred among the adult goats on farm A. Animals presented with signs of pneumonia, illthrift/emaciation, anorexia, refusal of concentrate feed and drop in milk yield. TB was diagnosed in April 2008 during postmortem examination of goats submitted dead or seriously ill to the local RVL. TB lesions were mainly localised in lung tissue and bronchomediastinal lymph nodes, but also in liver, spleen and internal lymph nodes (Sharpe and others 2010). Clinical cases of caprine M bovis infection and caseous lymphadenitis (CLA) were confirmed by bacteriological culture of M bovis and Corynebacterium pseudotuberculosis. Subsequently, 90 per cent of farm A adult goats were SICTT-positive, a milk sample from the bulk tank was confirmed M bovis-positive and the herd (299 adults and 151 kids born in 2008) was depopulated. As this herd was coinfected with both TB and CLA, the gross visible lesions were not necessarily tuberculous (Sharpe and others 2010). The cattle herds in the locality of farm A had a clear TB history and, subsequent to this outbreak, were checked by SICTT and found to be free of TB.

Farm B

Farm B is located in the Irish midlands, approximately 100 km south-east of Dublin. There were both sheep and goats on farm B during 2006 to 2009, but no cattle since 1995. Before 1995, there had been intermittent TB outbreaks in cattle on farm B and in neighbouring herds, with the most recent being in 2004. The dairy goat enterprise had commenced in 1999, with goats sourced from three separate herds. During 2007, goats from farm B had been moved to farms C (in February 2007) and A (in October 2007) (Fig 1). In October 2007, only 25 goats remained on farm B. These goats had been housed throughout 2007, and had not grazed during summer 2007 with the goats sold to farm A. A further 10 goats were later acquired. No unusual pattern of illness or mortality was reported in 2007, before sale of the goats. In May and July 2008, two full-herd SICTTs were conducted on the 35 farm B goats (all adults aged over six months), including 12 does in kid, following back-tracing from farm A, noting that the farm had both goats and also 20 ewes with their lambs. All goats on the holding were negative to both tests. Neighbouring cattle herds also tested negative during this period. Subsequently, however, on March 26, 2009, two of 30 pedigree Suffolk sheep gave a strong positive response to bovine tuberculin in a SICTT performed in the axilla region, and seven tested positive for C pseudotuberculosis (the causative organism of CLA).

Fig 1

Timeline of key management and health-related events affecting goats on farms A, B and C during 2006 to 2009. CLA Caseous lymphadenitis, SICTT Single intradermal comparative tuberculin test, TB Tuberculosis

Farm C

Establishment, initial movements (February 2007 to March 2008)

Farm C was located in north-east Ireland. It was a mixed suckler beef-fattening and pig enterprise conducted over three non-contiguous fragments (approximately 140 ha).

The first goat consignment (140 young female goat kids of the Saanen, Toggenburg and British Alpine breeds from two weeks to two months of age) arrived in February 2007 (Fig 1). These goats had originated from farm B. The bulk of the remaining goats on farm B (the dams of the kids that had gone to farm C; adults plus followers including newly born kids) were sold to farm A in the south of Ireland in October 2007. No premovement testing for TB was carried out on any of these goats as this is not a statutory requirement.

On arrival at farm C, the kids were permanently housed in a dedicated shed with concrete flooring, straw bedding, solid block/galvanised walls on three sides and a wire mesh and silage feeding area on the fourth. The fourth wall opened into the winter housing accommodation for the suckler cow herd, so that the goats and suckler animals shared the same air space. This area was not badger-proof. The young kids were fed sheep milk substitute with a roughage diet of silage and sheep nuts introduced later. Occasionally, uneaten surplus silage from the goats was given to the suckler cattle herd. At no stage were the goats allowed to graze, as is common in the dairy goat industry. Anthelmintics and vaccinations were routinely administered as per schedules recommended for sheep. The goats remained in this shed for approximately 12 months.

In February 2008, two Saanen and two Toggenburg male breeding goats (bucks), approximately five months of age, were imported from Scotland's scrapie-monitored herds. They tested negative for bluetongue virus (both pre-export and postimport) and were kept isolated at the home farm. Relevant EU Animal Health Import Regulations (European Commission 2003, 2007) were complied with. No TB tests were done before or after import. In March 2008, these bucks were kept along with female goats, aiming for the first does to kid in late August, for a September milk production target. A dairy unit consisting of three sheds, a separate collection yard and milking parlour was constructed approximately 1.5 km from the home farm. Once goats moved into this unit, access by badgers to goats was no longer possible.

Risk assessment (July 2008)

In mid-2008, an investigation of farm C was requested, given the epidemiological linkages between farms A, B and C. Therefore, in July 2008, a risk assessment was conducted on farm C, to determine the urgency with which the infection status of the goat herd on farm C should be clarified. The goats were in varying stages of gestation, raising concerns that unnecessary intervention (including a full-herd SICTT) might induce stress abortions.

TB infection risk was assessed as low and the risk posed by TB infection, if present, was not deemed to be high, considering the following: goats were permanently housed; the goats had originated from farm B, where infection in goats had not been confirmed, when they were up to two months of age; goats were clinically healthy, in good body condition and with no mortalities noted; there was no evidence of contact with any infected species; there had been a recent negative full-herd SICTT on farm C cattle, including those that had shared air space with the goats and those that had eaten silage to which the goats had first access; there were no outward movements from this herd since its assembly; and no milk was being supplied.

The infection risk was considered unchanged at further farm visits up to October 2008.

Further movements (August 2008)

In August 2008, a further 33 goats were purchased from farm B, including 17 pregnant goats, two bucks and 14 non-pregnant females. Two clear SICTTs had been completed at a 60-day interval before arrival. These tests had been done by DAFF staff with the cooperation of the farmer as there was a suspicion of TB caused by M bovis in the goats on farm B based on follow-up epidemiological investigations from farm A. The 14 non-pregnant goats were housed separately at the dairy unit and the rest mixed with the pregnant does and bucks at the home farm. There was no outward movement of goats from farm C during this period and only one death was reported. No postmortem examination was done on this animal.

Further investigations, initial control measures (October 2008)

In October 2008, farm C started to supply milk. Under the TB control plan established by the competent authority, and as a food business operator, a full-herd SICTT on all 190 adult goats was conducted. This revealed an apparent herd prevalence of 66.3 per cent (123 positives, three inconclusives) (Table 1).

Table 1

Single intradermal comparative tuberculin test (SICTT), Enferplex tuberculosis (TB), interferon-γ (IFN-γ) assay and gross postmortem results from a goat farm in Ireland (farm C) in October and November 2008, in three risk groups of animals

There were a further three severe inconclusive reactors, and one animal with a subcutis reaction. Reactors commonly had oedematous raised plaques, and marked bovine reactions (skin increases of 20 to 60 mm, reactions occupying the full clipped area, areas of necrosis and exudation) were observed in some animals. On the reading day, many reactors were reluctant to put their heads into the milking parlour head rack due to pain at the injection sites on contact with the rack. Clinical examinations of reactors detected enlarged submandibular and prescapular lymph nodes, pyrexia and milk drop in lactating does. In most reactors, the avian reactions were slight (skin increases of 2 to 6 mm). Animals that tested negative had no measurable avian skin increase.

Using the SICTT results, control measures were implemented to minimise the risk of further within-herd transmission, including early isolation of reactors, attention to effective biosecurity and the formation and management of specific ‘risk’ groups, as follows: Reactor group, 126 SICTT-positive/inconclusive animals, including 100 milking does, 23 dry/pregnant does and three bucks; At-risk group, 50 SICTT-negative goats (47 milking does and three bucks) with substantial contact with reactor animals; and Low-risk group, 14 non-pregnant does moved, following a negative SICTT, from farm B in August 2008. These had been housed in isolation, had not been in contact with rest of herd and were each SICTT-negative (Table 1).

One goat (goat 12) in the reactor group had been recently treated for clinical mastitis. Milk samples were collected from both this animal and from the bulk tank. The CVRL reported, four weeks later, that M bovis had been cultured from the sample collected from goat 12 but not the bulk tank.

Further investigations (November 2008 to April 2009)

In November 2008, all 126 goats in the reactor group were culled, although not for human consumption. Blood testing was conducted on the adult herd (no contemporaneous SICTT was conducted at this time). In the October 2008 SICTT reactor group, 66 goats (selected based on convenience) were tested using the IFN-γ assay, and 62 of these were positive. In the at-risk and low-risk groups, nine of 50 (18 per cent) and none of 14 were IFN-γ-positive, respectively. Furthermore, 124 of the 126 (98.4 per cent) goats in the October 2008 SICTT reactor group were Enferplex TB-positive, as were nine of 50 (18 per cent) of the at-risk group, but none of the 14 goats in the low-risk group (Table 1). Goats with a positive blood test were also culled later in November. In the course of the examination at the abattoir where 63 reactors were sent for immediate slaughter, many goats had gross visible lesions. However, it was possible that this herd also was coinfected with both TB and CLA, and examinations of tissues from these goats are continuing.

Five goats were examined in detail by the CVRL to confirm the presence of M bovis. The VNTR profile identified in each case was 11, 7, 4, 3, 2 (loci sequence Qub 11a, ETRA, MIRU 26, VNTR4052, VNTR 1895). The same strain had been identified in M bovis isolates from farm A goats and from badgers from the midlands, in the locality of farm B.

In December 2008, the IFN-γ assay was conducted on all adults and kids that remained on farm C (no contemporaneous SICTT was performed at this time). In total, two of 98 kids were IFN-γ-positive, as was one animal in the at-risk group. These were culled from the herd and all had positive lesions at postmortem examination, consistent with caprine infection with M bovis. All 14 animals in the low-risk group were IFN-γ negative.

In February 2009, a SICTT was conducted on all adults and the more mature kids, and blood testing of the entire herd of 149 goats was performed, using the IFN-γ assay and Enferplex TB. All animals were SICTT-negative, although one goat showed a slight reaction to avian PPD. All 149 goats tested negative to the IFN-γ assay and Enferplex TB. Further testing of all animals (SICTT, IFN-γ assay and Enferplex TB) was conducted in April 2009 and all animals were negative to all tests.

Discussion

In this outbreak, it is most likely that infected goats (from farm B in the Irish midlands) were the source of infection. The VNTR profile of the strain of M bovis that is mainly found in the Irish midlands was isolated in adult goats on farms A and C. It is probable that infection was introduced to the goat herd before February 2007 and became established on farm B (the birth herd of these goats) either through direct exposure with infected wildlife or indirectly through contact with contaminated feed. Epidemiological investigations have previously implicated contaminated feed and/or water troughs, and it is known that badgers, in which TB is endemic in Ireland, will visit feed stores and water troughs used by other species. The farmer of farm B had reported signs of badger activity during 2007 in a maize crop adjacent to the land where his goats had grazed before being sold to farm A. However, this would appear not to be when the herd first became infected, since it later transpired that at least one kid must have been infected before it moved to farm C in February 2007. M bovis was also confirmed, in 2009, in one of the two SICTT-positive sheep from farm B and the one sheep positive for C pseudotuberculosis. C pseudotuberculosis has cell wall antigens that are similar to those of the Mycobacterium species; therefore, goats and sheep with CLA can yield false-positive results on the tuberculin skin test (Brown and Olander 1987). Although cattle and goats shared the same air space in the shed on farm C, and the cattle had eaten surplus goat silage, it is unlikely that these cattle were the source of infection. The farm C cattle had tested negative using the SICTT both before and following the outbreak. The goats that remained on farm B in late 2007/early 2008 were almost certainly uninfected, as these and other introduced goats subsequently had two consecutive negative SICTTs. Based on the epidemiological pattern of presentation and the VNTR results, it is unlikely that the UK imports (to farm C) were a source of infection. Only one strain of M bovis infection was found in this outbreak, consistent with a single source of infection.

TB was first confirmed in goats on farm A in April 2008. Many of these goats had visible lesions at slaughter following full-herd depopulation, although some were likely as a result of coinfection with CLA (Sharpe and others 2010). Herd prevalence was extremely high. On farm A, M bovis was cultured in milk from the bulk tank, and at least one of the goats examined initially at the commencement of the outbreak (but before TB confirmation) had evidence of mastitis, which was later confirmed as due to M bovis. Pooled colostrum on farm A and possibly on farm B, and close contact on both farm A and C, may each have contributed to within-herd transmission. The isolation of the 14 young dry stock does in a separate air space on farm C, and their subsequent negative skin test, would suggest that transmission was facilitated by direct contact with infected animals, rather than transmission by indirect contact via shared personnel. Cattle shared the same air space as the goats during the housing period and were occasionally fed surplus uneaten silage from the goats. However, all these cattle subsequently tested negative.

Lesions consistent with caprine infection by M bovis were observed at the abattoir and the pathology is expected to reflect the findings in farm A, where TB lesions in the subset of reactors where detailed examination was performed mainly affected the lung and bronchomediastinal lymph nodes (Sharpe and others 2010). As mentioned previously, the observation of lesions in the abattoir must be interpreted with care, given the possibility of coinfection with CLA. Nonetheless, respiratory excretions were likely to be infectious given the extensive within-herd spread and the detection of a reactor to all three tests after only six weeks of contact. The separation of the herd into groups based on the results of the diagnostic tests and controlling the degree of contact helped to minimise further goat-to-goat transmission.

Key objectives during this outbreak included rapid detection and removal of any potentially infected animals, minimisation of any zoonotic risks and limiting of any further spread while maintaining a viable enterprise, if possible.

The results of the initial risk assessment on farm C proved somewhat misleading, given the later understanding of infection on this farm. Based on the evidence available at the time, the potential for infection in this herd was assessed as low. The birth herd of these animals had completed two negative skin tests, and clinical examinations did not reveal any significant abnormalities, apart from occasional coughing on exertion. The animals had good body condition scores, all were clinically healthy and, once lactation had commenced, the animals' milk yield was good. Furthermore, at the time it was felt that infection could have been either sourced from within, and subsequently confined to, farm A in the south, or to have commenced on farm B after the kids had left for farm C. Based on the evidence available, it was perceived that there was a need to reassess the health status of goats on farm C only once the animals had started to supply milk, as per the TB control plan requirement introduced in autumn 2008.

During this outbreak, the SICTT proved a very effective means to identify potentially infected animals. However, the SICTT is difficult to apply on goats due to the very fine nature of their skin (often less than 2 mm thick). It is necessary to use an oblique angle and a McClintock syringe with a very short (5/16 inch) needle. The SICTT was limited to adult goats and some of the more mature kids, as it was considered too problematic to conduct on the very thin-skinned younger kids. The results were interpreted using the severe (bovine) interpretation, an approach not previously documented in goats in Ireland. Based on analysis of skin results and subsequent gross lesion detection at the abattoir, it seems that any increase in skin thickness at the bovine site could be considered a positive result, particularly in herds already confirmed as being infected or with close epidemiological links to confirmed infection. Indeed, in housed goats with no CLA or other concurrent mycobacterial infection, a comparative (avian) test may not be necessary. However, further analysis will be required to substantiate this observation. The skin response of goats to the SICTT was also very marked. Most of the bovine site reactions showed classic signs of heat, pain, swelling and associated lymph node enlargement. Three goats that were SICTT-inconclusive (under the severe interpretation, ie, bovine response 0 to 2 mm less than the avian response) had visible lesions at the abattoir suggestive of TB.

There was substantial agreement between the SICTT and IFN-γ test during this outbreak (κ=0.800) (Table 1). At the first test, in October 2008, 62 of 66 SICTT-positive animals on farm C were also IFN-γ-positive. When the IFN-γ test was conducted on the full herd in December 2008, three positive animals were identified. At this stage of the investigation, the IFN-γ test proved very useful in identifying potentially infected kids, noting that these animals were only approximately four weeks old at that time and that no SICTT was performed during December. In cattle, the IFN-γ test is not recommended in such young animals (Olsen and others 2005). However, the same issues of interpretation in goats in this study or in a control set of goats including kids specifically assayed during this period did not appear (E. Gormley, personal communication). There was almost perfect agreement between the Enferplex TB and SICTT (κ=0.867) (Table 1). In November 2008, 124 of 126 SICTT-positive animals from the October SICTT were also Enferplex TB-positive.

The SICTT results were the first results available and thus were used as the basis for subsequent risk categorisation and separation of the goats into the three risk categories. Given subsequent events, this strategy appears to have been useful in limiting further within-herd transmission by isolating all skin reactors in a separate air space. It also facilitated the introduction of rational biosecurity measures including disinfectant footbaths and controlled personnel movement (never moving from infected to negative animals in any single day). Following each round of testing, positive/reactor goats were rapidly removed. Then, cleaning and disinfection of all housing, bedding, the milking parlour and the bulk milk tank was conducted. All bulk tank milk was disposed of, all manure and bedding were removed, and housing was power-washed with a 4 per cent solution of washing soda. Disinfection was then undertaken using a DAFF-approved disinfectant (Antec Hyperox; Antec International) (DAFF 1975). From this point, all milk supplied to the processor was facilitated under permit and only from goats that had tested negative to the screening tests. All positive goats were dried off as quickly as possible, noting that these animals were heading towards peak lactation. There was a noticeable decline in the body condition of some of the reactor animals, precipitated by the stress induced by this forced drying off.

There is no internationally approved test and interpretation methodology for TB in goats. In addition, there is no TB eradication scheme provided for in legislation, and no compensation applicable for goat herds in Ireland. Consequently, all decisions were taken in consultation with the farmers involved, with due regard to test results available, the epidemiological findings and the possibility of preserving the herd as a viable productive unit. In the early stages of bovine TB eradication, isolation of infected groups of animals was commonly practised, until culling was feasible (Ritchie 1959). The same veterinary principles were conducted here.

The affected goat herds were restricted under S.I.101/2008 (Irish Statute Book 2008); movements of goats or milk in or out of the herd were allowed only under permit from the local Divisional Veterinary Officer. The trading status of the associated cattle herd on farm C was immediately suspended and a full-herd SICTT was conducted on all farm cattle to assess whether there had been any cross-species transmission. All the cattle tested negative at the end of November 2008, despite some having had close contact with the goats and some sharing of fodder during the housing period of the previous winter. The goat herd on farm C was SICTT-negative in autumn 2009 and also in 2010. Farm A commenced repopulation in 2009 from herds that had recent TB tests with negative results.

A number of control measures were recommended to minimise any possible zoonotic threat. On confirmation of the SICTT results, all personnel were referred to their doctor for a Mantoux test and to determine their BCG status (Health Protection Surveillance Centre 2010). No personnel had any positive reaction to the Mantoux test and chest radiographs showed normal results. Staff were instructed not to consume any unpasteurised milk, and increased biosecurity measures were recommended and implemented in the workplace.

This investigation highlights the necessity for increased public awareness of the susceptibility of goats to TB, especially among those who could be classed as hobbyist goat keepers (many of whom consume raw milk or milk products on-farm and are not legally obliged to have a TB control plan in place). Positive milk cultures were obtained from an individual animal (but not the bulk tank) on farm C and from the bulk tank on farm A. There are public health risks associated with the consumption of unpasteurised milk and/or milk products in an environment where TB is prevalent (eg, Doran and others 2009). Goat milk from both of these farms was subjected to pasteurisation and so, in these cases, was not a risk to consumers. The implementation of a control plan to screen for TB in dairy goat/sheep herds, as outlined under EC Regulation 853/2004 (European Commission 2004a), is vitally important to ensure food safety and public health and to guard against the threat of M bovis-infected milk.

Registration and identification of all goats in herds is a legal requirement under EU legislation (European Commission 2004b) and previous establishment of the TB status of any goats intended for sale/purchase would be a necessary control measure in order to track and trace movements in the event of disease breakdowns and to prevent disease introduction into TB-free herds.

Acknowledgments

The authors thank Joanne McLernon of CVRL Backweston for conducting the VNTR typing, Tracy Clegg from CVERA for conducting the statistical analyses, and Ronan Sugrue and other staff from DAFF's Veterinary Public Health Division for coordinating the slaughter at the abattoir. The authors thank CVRL and RVL DAFF staff for their assistance in the abattoir and for confirmation of TB on farms A and C, DVO staff for tracing the movement of animals to and from the various herds and for assisting in the abattoir and on the infected farms, and also Bob Treacy, Laois District Veterinary Office, for conducting testing and sampling on farm B. In particular, the authors thank Alan Johnson, Limerick RVL, for the initial detection of TB in goats from farm A, and Anne Sharpe, CVRL Backweston, for confirmation of CLA on farms A and B. The Enfer Scientific staff who optimised the Enferplex system for goats and the UCD staff who likewise optimised the IFN-γ assay for use in goats are thanked. Finally, the authors thank the farmers whose herds were the subject of this investigation.

References

View Abstract

Footnotes

  • Provenance not commissioned; externally peer reviewed

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.