Five years ago The Daily Mail newspaper ran an article titled “Could you catch TB from YOUR tabby?” and of course as a Microbiologist and an owner of 5 cats, I had to blog on that! However recently this subject has impacted my clinical work.
Last week I had a number of calls about patients who have had contact with “cats with TB”. The patients are very anxious and the GPs were sensibly seeking advice about what to do.
In fact none of the cats actually have tuberculosis (TB); M. tuberculosis doesn’t actually cause clinical disease in cats. The most common Mycobacterium spp. to infect cats is Mycobacterium microti (19%) with Mycobacterium bovis being the second most common (15%). Mycobacterial infection in cats is linked to being outdoors, hunting and fighting. It is therefore more common in adult cats, especially males.
M. microti and M. bovis are both part of a group of mycobacteria known as the MTB complex which also includes Mycobacterium tuberculosis. But humans very, very rarely get M. microti so why the fuss over tabby cats spreading TB? M. bovis is however a potential pathogen of humans and therefore there is a risk that infected cats can pass the infection on to their owners (or servants as the cats call them). NB Although cats don’t get M. tuberculosis… dogs can, so all of you dog owners out there beware!
M. bovis infection in cats is higher in areas of the country where there are also high rates of M. bovis in both livestock and wild animals, such as in the South West of the UK. So theoretically cats can spread the bacteria into the environment and therefore be a potential vector to spread infection between other animals, such as cattle, which can have a high economic impact. However in the UK, there are only 30-40 cases of M. bovis in humans a year, so it is a very rare infection. So why so many calls? Bear with me I’ll get to that.
M. bovis outbreak in cats in the UK
In 2018 M. bovis was diagnosed in 3 pedigree cats from 2 different households in 2 different geographical areas normally considered low risk for M. bovis. It was noted that all 3 cats were living exclusively indoors, which is unusual for cats with M. bovis infections.
Once published in the Veterinary Record and Veterinary Times the study acquired further notifications; 10 more cats from another 3 households all in separate geographical areas of the UK; Durham, Sussex, Norfolk, Devon and Essex. In total 6 sick and seven in-contact cats were identified all with evidence of M. bovis infection.
Overall there were 6 cats who were unwell with clinical infections and sadly 5 of these died. All 5 of the index cats presented with what appeared to be intestinal M. bovis infection. Strikingly, all had been fed the same commercially available raw food diet (the venison raw food diet from Natural Instinct) which has subsequently been voluntarily withdrawn by the manufacturer (batches withdrawn are those dated as best before between March 2019 and August 2019). One of the in-contact cats from these cases developed a respiratory illness with M. bovis whereas all of the other in-contacts had positive IGRA tests for M. bovis.
The epidemiological link to the raw food was compelling; despite an extensive investigation no other common link was found. Possible sources of exposure for these cats to M. bovis were excluded, including wildlife contact, access to raw milk, the presence of rodent populations inside the buildings in which the cats lived and exposure to known infectious humans. The fact that all of the index cases appear to have had a GI infection further supports it being something they have eaten. Apparently the outbreak is still ongoing and it is expected that more cases will be identified. At the time of writing there are over 90 feline cases.
Raw meat diets are becoming trendy with pet owners and big business for the manufacturers. It is estimated that it is worth $195 million in the USA alone and the number of manufacturers in the UK has increased from <5 a few years ago to now more than 80!
It is unclear if these types of raw meat diets are good or not for animals but some things are certain:
How does M. bovis infection present?
The most common presentation of M. bovis in cats is with localised nodular cutaneous lesions with ulcers and sinuses. These are usually found on what are referred to as the “fight and bite” sites such as the face, limbs and tail base. There is often associated lymphadenopathy (how the heck do you feel lymph glands in a cat?). It has been reported that about 1% of all cutaneous lesions in cats have Mycobacterium spp. in them although it is still unclear if these are the cause of infection or whether they are colonisers.
Pulmonary infection with M. bovis is rare in cats, and when it does occur it is usually due to haematogenous spread from a cutaneous focus. Pulmonary disease can present as either diffuse inflammation causing shortness of breath and a cough, or with enlarged pulmonary lymph nodes.
Gastrointestinal M. bovis is now very uncommon (in cats and humans) because the vast majority of milk is pasteurised to stop us humans getting gastrointestinal M. bovis. Most cat owners also know not to give a cat cow’s milk as it is very fattening. So an increase in gastrointestinal M. bovis is also very unusual.
How is M. bovis diagnosed?
It is just as tricky to diagnose M. bovis infection in cats as it is in humans. The most common way traditionally in both species is with culture. This is widely available for humans but not so much for cats. Mycobacterial culture has to be done in a Category 3 facility; a contained laboratory with biological safety cabinets (often known as a “Cat 3” lab although there aren’t 3 cats present… honest). These types of labs are present in all standard hospital laboratories but Veterinary Surgeons would have to use human reference laboratories in the UK to process their samples and this is very expensive. Positive cultures can be identified as M. bovis by molecular tests but this is not suitable to do on the tissue directly as the tests are too insensitive; the same applies to human tissue.
In fact none of the cats actually have tuberculosis (TB); M. tuberculosis doesn’t actually cause clinical disease in cats. The most common Mycobacterium spp. to infect cats is Mycobacterium microti (19%) with Mycobacterium bovis being the second most common (15%). Mycobacterial infection in cats is linked to being outdoors, hunting and fighting. It is therefore more common in adult cats, especially males.
M. microti and M. bovis are both part of a group of mycobacteria known as the MTB complex which also includes Mycobacterium tuberculosis. But humans very, very rarely get M. microti so why the fuss over tabby cats spreading TB? M. bovis is however a potential pathogen of humans and therefore there is a risk that infected cats can pass the infection on to their owners (or servants as the cats call them). NB Although cats don’t get M. tuberculosis… dogs can, so all of you dog owners out there beware!
M. bovis infection in cats is higher in areas of the country where there are also high rates of M. bovis in both livestock and wild animals, such as in the South West of the UK. So theoretically cats can spread the bacteria into the environment and therefore be a potential vector to spread infection between other animals, such as cattle, which can have a high economic impact. However in the UK, there are only 30-40 cases of M. bovis in humans a year, so it is a very rare infection. So why so many calls? Bear with me I’ll get to that.
M. bovis outbreak in cats in the UK
In 2018 M. bovis was diagnosed in 3 pedigree cats from 2 different households in 2 different geographical areas normally considered low risk for M. bovis. It was noted that all 3 cats were living exclusively indoors, which is unusual for cats with M. bovis infections.
Once published in the Veterinary Record and Veterinary Times the study acquired further notifications; 10 more cats from another 3 households all in separate geographical areas of the UK; Durham, Sussex, Norfolk, Devon and Essex. In total 6 sick and seven in-contact cats were identified all with evidence of M. bovis infection.
Overall there were 6 cats who were unwell with clinical infections and sadly 5 of these died. All 5 of the index cats presented with what appeared to be intestinal M. bovis infection. Strikingly, all had been fed the same commercially available raw food diet (the venison raw food diet from Natural Instinct) which has subsequently been voluntarily withdrawn by the manufacturer (batches withdrawn are those dated as best before between March 2019 and August 2019). One of the in-contact cats from these cases developed a respiratory illness with M. bovis whereas all of the other in-contacts had positive IGRA tests for M. bovis.
The epidemiological link to the raw food was compelling; despite an extensive investigation no other common link was found. Possible sources of exposure for these cats to M. bovis were excluded, including wildlife contact, access to raw milk, the presence of rodent populations inside the buildings in which the cats lived and exposure to known infectious humans. The fact that all of the index cases appear to have had a GI infection further supports it being something they have eaten. Apparently the outbreak is still ongoing and it is expected that more cases will be identified. At the time of writing there are over 90 feline cases.
Raw meat diets are becoming trendy with pet owners and big business for the manufacturers. It is estimated that it is worth $195 million in the USA alone and the number of manufacturers in the UK has increased from <5 a few years ago to now more than 80!
It is unclear if these types of raw meat diets are good or not for animals but some things are certain:
- Cats are obligate carnivores and should not be given vegetarian diets… it will do them harm.
- Home prepared raw diets should be avoided as it is extremely difficult to make these nutritionally balanced.
- There is a significant increased risk of infectious diseases in animals fed raw meat including salmonellosis, E. coli O157, listeriosis, campylobacter, toxoplasmosis and other parasites, which all pose a potential zoonotic risk being spread from the animal to the owners, shedding microorganisms into the home environment and through handling of such products.
- Normal cleaning practices in domestic kitchens are unlikely to kill M. bovis which can survive in standing water for 60 days and requires pasteurisation at >70oC for 15 seconds. Therefore M. bovis may persist on kitchen surfaces even after the owners feel that the surfaces have been ‘cleaned’.
How does M. bovis infection present?
The most common presentation of M. bovis in cats is with localised nodular cutaneous lesions with ulcers and sinuses. These are usually found on what are referred to as the “fight and bite” sites such as the face, limbs and tail base. There is often associated lymphadenopathy (how the heck do you feel lymph glands in a cat?). It has been reported that about 1% of all cutaneous lesions in cats have Mycobacterium spp. in them although it is still unclear if these are the cause of infection or whether they are colonisers.
Pulmonary infection with M. bovis is rare in cats, and when it does occur it is usually due to haematogenous spread from a cutaneous focus. Pulmonary disease can present as either diffuse inflammation causing shortness of breath and a cough, or with enlarged pulmonary lymph nodes.
Gastrointestinal M. bovis is now very uncommon (in cats and humans) because the vast majority of milk is pasteurised to stop us humans getting gastrointestinal M. bovis. Most cat owners also know not to give a cat cow’s milk as it is very fattening. So an increase in gastrointestinal M. bovis is also very unusual.
How is M. bovis diagnosed?
It is just as tricky to diagnose M. bovis infection in cats as it is in humans. The most common way traditionally in both species is with culture. This is widely available for humans but not so much for cats. Mycobacterial culture has to be done in a Category 3 facility; a contained laboratory with biological safety cabinets (often known as a “Cat 3” lab although there aren’t 3 cats present… honest). These types of labs are present in all standard hospital laboratories but Veterinary Surgeons would have to use human reference laboratories in the UK to process their samples and this is very expensive. Positive cultures can be identified as M. bovis by molecular tests but this is not suitable to do on the tissue directly as the tests are too insensitive; the same applies to human tissue.
The other old fashioned way to look for possible TB in humans was to do a Mantoux test. This used a small amount of tuberculin (a liquid containing TB proteins) injected intradermally and the skin reaction examined to see if the body reacted to the proteins. This was tricky and time consuming to do in humans and tuberculin is now in very short supply. Apparently it can’t be used in cats because they just don’t react to tuberculin and so this isn’t an option.
The final way to diagnose M. bovis is to do a blood test on the cat for an interferon-gamma release assay (IGRA). We use IGRAs in humans a lot; examples include Quantiferon and T-spot. IGRAs test for the presence of immune cells in peripheral blood that react to proteins produced by the bacteria from the MTB complex, including M. bovis and M. microti. Further tests can be performed on the blood sample to determine if the IGRA reaction is due to M. bovis rather than M. microti.
How is M. bovis in cats treated?
The normal antibiotics used to treat M. bovis in cats are Rifampicin, Azithromycin and the fluoroquinolone Pradofloxacin. This is different to those used in humans; M. bovis is inherently resistant to Pyrazinamide, so human treatment is with Isoniazid, Rifampicin and Ethambutol for 2 months followed by Isoniazid and Rifampicin for a further 7 months (9 months in total). Presumably neither Isoniazid nor Ethambutol can be used in cats?
Five of the 6 clinically affected cats underwent initial treatment but 4 had to be euthanized on clinical grounds; 1 of the index cats was so unwell that treatment wasn’t thought to be appropriate and they too were euthanized. Of the 7 cats who were asymptomatic but had positive IGRA tests 2 were euthanized because their owner was immunosuppressed and they were felt to be a risk to their owner (presumably it was thought they shouldn’t be rehomed with positive IGRA for M. bovis). The remaining 5 cats had positive IGRAs but did not require treatment; they remain well and are being monitored for weight loss or other clinical signs.
In humans we would treat M. bovis with 3 drugs known to be active (Isoniazid, Rifampicin, Ethambutol) whereas vets are using Rifampicin, a fluoroquinolone called Pradofloxacin and a macrolide, Azithromycin. I find the “triple” aspect of the treatment regimen for M. bovis in cats curious. I understand using a fluoroquinolone as these are active against MTB complex bacteria including M. bovis but in humans we do not use macrolides as studies have shown them to have no activity and therefore make no difference in treatment.
Further studies in Cystic Fibrosis patients suggest that Azithromycin prophylaxis may actually predispose to mycobacterial infection by adversely affecting how macrophages deal with the bacterium. Therefore I’m doubtful that the macrolide, Azithromycin, would provide any benefit to these infected cats… they are essentially on “double” not “triple” therapy. The reason for triple therapy is to prevent resistance occurring during treatment. The fact that 5 of the 6 symptomatic cats didn’t really respond to treatment makes me even more certain therapy was not adequate to treat M. bovis. Unfortunately, not being a vet, I cannot access the paper which provides evidence to support this treatment in cats.
So why have I had so many calls from GPs?
I think that when these issues come to the attention of the public through the media patients seek reassurance from their GP especially if they have had a flu like illness recently. M. bovis infection is a risk factor for cat owners but unsurprisingly GPs are not familiar with this zoonotic infection. My advice to GPs is if an owner becomes unwell with fever, weight loss, night sweats or a cough, for example, then TB should always be part of the differential diagnosis. However, it would be worthwhile just asking if the patient is a cat owner and whether the cat eats a raw meat diet. But do keep in mind that infection with M. bovis in humans is very rare and it is more likely the patient has another reason for their symptoms.
The final way to diagnose M. bovis is to do a blood test on the cat for an interferon-gamma release assay (IGRA). We use IGRAs in humans a lot; examples include Quantiferon and T-spot. IGRAs test for the presence of immune cells in peripheral blood that react to proteins produced by the bacteria from the MTB complex, including M. bovis and M. microti. Further tests can be performed on the blood sample to determine if the IGRA reaction is due to M. bovis rather than M. microti.
How is M. bovis in cats treated?
The normal antibiotics used to treat M. bovis in cats are Rifampicin, Azithromycin and the fluoroquinolone Pradofloxacin. This is different to those used in humans; M. bovis is inherently resistant to Pyrazinamide, so human treatment is with Isoniazid, Rifampicin and Ethambutol for 2 months followed by Isoniazid and Rifampicin for a further 7 months (9 months in total). Presumably neither Isoniazid nor Ethambutol can be used in cats?
Five of the 6 clinically affected cats underwent initial treatment but 4 had to be euthanized on clinical grounds; 1 of the index cats was so unwell that treatment wasn’t thought to be appropriate and they too were euthanized. Of the 7 cats who were asymptomatic but had positive IGRA tests 2 were euthanized because their owner was immunosuppressed and they were felt to be a risk to their owner (presumably it was thought they shouldn’t be rehomed with positive IGRA for M. bovis). The remaining 5 cats had positive IGRAs but did not require treatment; they remain well and are being monitored for weight loss or other clinical signs.
In humans we would treat M. bovis with 3 drugs known to be active (Isoniazid, Rifampicin, Ethambutol) whereas vets are using Rifampicin, a fluoroquinolone called Pradofloxacin and a macrolide, Azithromycin. I find the “triple” aspect of the treatment regimen for M. bovis in cats curious. I understand using a fluoroquinolone as these are active against MTB complex bacteria including M. bovis but in humans we do not use macrolides as studies have shown them to have no activity and therefore make no difference in treatment.
Further studies in Cystic Fibrosis patients suggest that Azithromycin prophylaxis may actually predispose to mycobacterial infection by adversely affecting how macrophages deal with the bacterium. Therefore I’m doubtful that the macrolide, Azithromycin, would provide any benefit to these infected cats… they are essentially on “double” not “triple” therapy. The reason for triple therapy is to prevent resistance occurring during treatment. The fact that 5 of the 6 symptomatic cats didn’t really respond to treatment makes me even more certain therapy was not adequate to treat M. bovis. Unfortunately, not being a vet, I cannot access the paper which provides evidence to support this treatment in cats.
So why have I had so many calls from GPs?
I think that when these issues come to the attention of the public through the media patients seek reassurance from their GP especially if they have had a flu like illness recently. M. bovis infection is a risk factor for cat owners but unsurprisingly GPs are not familiar with this zoonotic infection. My advice to GPs is if an owner becomes unwell with fever, weight loss, night sweats or a cough, for example, then TB should always be part of the differential diagnosis. However, it would be worthwhile just asking if the patient is a cat owner and whether the cat eats a raw meat diet. But do keep in mind that infection with M. bovis in humans is very rare and it is more likely the patient has another reason for their symptoms.
RSS Feed
