Statistics from Altmetric.com
Tuberculosis (TB), caused by Mycobacterium bovis, and caseous lymphadenitis (CLA), caused by Corynebacterium pseudotuberculosis, are typically chronic infectious pyogranulomatous diseases. Other similarities between the two diseases include gross postmortem findings, such as abscessation (Crawshaw and others 2008, Baird and Fontaine 2007) and lesion distribution. Both diseases may affect the lung, thoracic lymph nodes and viscera (Sanson 1988, Crawshaw and others 2008). These similarities may make it difficult to distinguish between the two diseases. Definitive diagnosis of TB and CLA is by culture (Baird 2007, Sharp 2007). However, in concurrent infections, culture of every lesion may not be possible. Dual infection with TB and CLA has not previously been reported. This short communication describes a concurrent outbreak of TB and CLA in milking goats, and also describes the value of Ziehl-Neelsen (ZN) and Gram staining of histological sections in distinguishing between both types of lesions where culture results were not available.
Clinical cases of TB and CLA were confirmed in a milking goat herd in April 2008 by bacteriological culture of lesions. The clinical signs observed included poor milk yield, poor body condition, chronic cough and enlarged superficial lymph nodes showing purulent discharge. The farmer, based in the south of Ireland, had bought in the already established milking herd of 290 adult goats the previous autumn from a midlands farm that was closing down its dairy goat enterprise. In May 2008, the adult goats in the herd were tested using the single intradermal comparative tuberculin test (SICTT), the routine test used in Ireland to identify animals infected with M bovis. The sensitivity and specificity of this test in goats are reported to be 83 per cent and 100 per cent, respectively (Gutiérrez and others 1998). The majority of goats showed a positive reaction to bovine purified protein derivative (PPD), and the whole herd was culled. Postmortem examinations were carried out on 15 of the bovine PPD-positive animals.
Lesion distribution was based on the results of bacteriological culture and, where these were unavailable, on ZN and Gram staining of histological sections (Table 1). The acid-fast bacilli in TB lesions were sparse, often seen as single bacilli or an occasional cluster in multinucleated giant cells (Fig 1), whereas Gram staining of CLA lesions readily revealed clumps of Gram-positive short rod-shaped bacteria (Fig 2). M bovis isolates from five goats were identified as belonging to spoligotype SB0146.
Of the 15 goats, eight had TB lesions only, three had CLA lesions only, three had both TB and CLA lesions, and no lesions were found in one goat. TB lesions mainly affected the lung and bronchomediastinal lymph nodes, and were also seen in other internal organs such as the liver, spleen and internal lymph nodes. In comparison, CLA lesions mainly affected the lymph nodes of the head and neck. The tissues affected by either disease included the tonsil and retropharyngeal lymph node.
This is the first report of concurrent infection by M bovis and C pseudotuberculosis in milking goats. Dual infection by M bovis and Mycobacterium paratuberculosis has been reported in goats (Bernabé and others 1991). TB lesions tended to affect mainly the lung and thoracic lymph nodes, while internal lymph nodes and viscera were affected to a lesser extent, similar to other reports of caprine TB (Bernabé and others 1991, Crawshaw and others 2008). In comparison, CLA mainly affected the lymph nodes of the head and neck, similar to other observations in caprine CLA (Batey and others 1986, Schreuder and others 1986, Brown and Olander 1987). Udder and associated lymph node lesions, described in both caprine TB (Crawshaw and others 2008) and caprine CLA (Baird 2007), were not seen in the animals studied, but may have been present in other animals in the outbreak.
A culture result was not available for all lesions. This was overcome by Gram and ZN staining of histological sections, with only two lesions yielding equivocal results. While Gram-positive bacterial clumps were easily seen in CLA lesions, prolonged staining with carbol-fuschin at 60°C (Gutiérrez Cancela and García Marin 1993) and careful microscopic examination were required to detect sparse ZN-positive acid-fast mycobacteria in TB lesions. This paucity of acidfast mycobacteria was similar to other cases of caprine TB (Sanson 1988, Gutiérrez Cancela and García Marin 1993) and may be due to the immune responses operating in the granulomatous inflammation of caprine mycobacteriosis (Ridley 1983), or may be strain-related (Gutiérrez Cancela and García Marin 1993).
Four of the 15 SCITT-positive goats had no TB lesions. It is possible that M bovis infection was at such an early stage in these animals that small lesions may have been missed or had not yet formed. Because C pseudotuberculosis has cell wall antigens that are similar to Mycobacterium species, goats and sheep with CLA can yield false-positive results on the tuberculin skin test (Brown and Olander 1987) or on Johne's disease serology (Manning and others 2007).
Spoligotype SB0146, a strain of M bovis mainly found in the Irish midlands, was identified in the isolates examined, indicating that the goats were infected by the same source and probably while they were on the midlands farm. The sources of TB and CLA in this outbreak were not conclusively identified. The midlands farmer had bought in goats from other Irish herds, a possible mode of entry of both pathogens (Baird 2003, Crawshaw and others 2008). There were no cattle on the farm and no TB reactors in cattle herds contiguous to the farm. However, the possibility of contact with TB-infected wildlife cannot be ruled out. The midlands farm also had a flock of 30 pedigree Suffolk sheep, two of which tested positive to bovine tuberculin in a SCITT and seven of which tested positive for antibodies to C pseudotuberculosis. These sheep could have been a source of CLA.
Practitioners should consider the possibility of concurrent CLA in bovine PPD-positive goats, especially if superficial lymph nodes are enlarged. The present study also showed that, although mycobacteria were sparse in contrast to easily identifiable corynebacteria, ZN and Gram staining of histological sections were valuable in distinguishing between the two diseases when culture results were unavailable.
The authors thank DAFF veterinary inspectors Ted Curtin, Cork District Veterinary Office, and Bob Treacy, Laois District Veterinary Office, for their work investigating both farms. They also thank Graham Baird, Scottish Agricultural College, Edinburgh, for carrying out the CLA serology, and Marion Barrett for ZN staining. They are grateful for the help of all DAFF staff, Backweston, including Maria Kirby, Claire Fahy, Joanne McLernon, Rosemarie Slowey, Áine O'Doherty, Martin Hill, Henrietta Cameron and Andrew Downes.
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.