Other formats

    Adobe Portable Document Format file (facsimile images)   TEI XML file   ePub eBook file  


    mail icontwitter iconBlogspot iconrss icon

Proceedings of the First Symposium on Marsupials in New Zealand

Tuberculosis in the Possum Trichosurus Vulpecula

page 163

Tuberculosis in the Possum Trichosurus Vulpecula


Since tuberculosis was first reported in possums in New Zealand in 1970 field surveys and experimental work have contributed significantly to our understanding of the epidemiology and pathology of the disease.

Infected possum communities have been discovered in 23 general localities throughout New Zealand. There is compelling evidence linking tuberculosis in possums to high and persistent reactor rates in cattle in the same environment.

The possum is very susceptible to the effect of infection with Mycobacterium bovis and will develop a progressive and fatal disease with wide dissemination of lesions. The lesions are characterised by a limited cellular response, extensive tissue necrosis and a high rate of multiplication of the bacilli.

It is highly probable that tuberculosis can maintain itself in a population of possums and serve as a continuing source of re-infection for cattle and as a risk to people handling possum carcases.


Tuberculosis is caused by acid-fast bacilli in the genus Mycobacterium, principally by M. tuberculosis, M. bovis and M. avium (the human, bovine and avian strains respectively).

Although recognition of tuberculosis in wild brushtail possums Trichosurus vulpecula is a relatively recent event, tuberculosis in captive animals has been recognised for sixty years. Moore (1903) noted "an enlargement of one of the stifle joints exactly in the nature of a strumous enlargement or abcess sometimes found in sheep. Post-mortem examination showed the lungs one mass of tubercles." These lesions were seen in 1895 in a young possum held in captivity. Subsequently tuberculosis was reported in a possum held in a zoological park (Scott 1928).

Tuberculosis in wild possums was first reported in 1970 (Ekdahl et al. 1970), and this report was an outcome of a growing awareness of an association between the persistence of tuberculosis in cattle in a tuberculosis control area and tuberculosis in possums in the same environment. The distinctive lesions of tuberculosis due to Mycobacterium bovis in the possum have been described page 164 (Ekdahl et al. 1970, Smith 1972) and Smith (1972) drew attention to the large numbers of organisms which may be present in these lesions.

There is now compelling circumstantial evidence linking tuberculosis in possums to a high and persistent reactor rate in cattle sharing the same environment and, although a number of feral or wild species may be infected with M. bovis, only the tuberculous possum appears to have attributes of a significant reservoir of infection for cattle (Animal Health Division 1976).

The susceptibility of the possum to M. tuberculosis and M. bovis infections has been established experimentally (Bolliger and Bolliger 1948). Intraperitoneal and intramuscular injections of tubercle bacilli resulted in death in 2 to 5 weeks and natural transmission of M. bovis from an experimentally infected possum to a cage mate was demonstrated.

Clinical Effects

No detailed clinical observations have been made on tuberculosis in the possum. Wild possums with advanced tuberculosis have occasionally been observed wandering about in the open during daylight hours. They appear dazed and do not respond to the approach of man.

Following experimental infection with M. bovis the onset of the illness was quite abrupt. The possums became quiet, less responsive to stimuli and less resistant to handling. They lost their appetite and body condition: the degree of loss being dependent on the duration of illness. Clinical signs were observed for approximately 14 days prior to death regardless of the length of the premonitory period (O'Hara et al. 1976).

Gross Lesions

The distribution of gross lesions as seen at post mortem of poisoned wild possums in three surveys is given in Table 1. These figures show that the lungs are the most commonly affected site (202/327 - 61.8% of cases). The superficial lymph nodes were affected on 203 occasions. As some possums would have lesions in more than one superficial lymph node the total number of possums with lesions in these sites would be somewhat less than 62.7%. Nevertheless, superficial lymph nodes would be the second most common site affected.

page 165
Table 1. Distribution of gross lesions in poisoned wild possums.
No % No % No % No %
Lung 21 80.1 81 55.1 100 65 202 61.8
Liver 12 46.2 28 19.0 20 13 60 18.3
Spleen 2 7.7 14 9.5 9 5.8 25 7.6
Kidney 2 7.7 18 12.2 6 3.9 26 7.9
Adrenal 2 7.7 3 2.0 - - 5 1.5
Axillary lymphnodes 2 7.7 46 31.3* 81 52.6* 129 39.4
Inguinal " " 4 15.4 17 11.6+ 28 18.2+ 49 14.9
Mediastinal " " 3 11.5 8 5.4§ 30 19.5§ 41 12.5
Mesenteric " " 7 26.9 27 18.4 31 20.1 65 19.8
Gastric " " 4 15.4 2 1.4 1 0.6 7 2.1
Hepatic " " 3 11.5 14 9.5 14 9.1 31 9.4
Iliac/sublumbar lymphnodes 1 3.9 5 3.4 5 3.2 11 3.3
Peripharyngeal/cervical lymphnodes - - 3 2.0 22 14.3 25 7.6
Total No. of possums 26 147 154 327
APossum tuberculosis survey - Waikato region 1974 (Lake 1974)
BPossums from Hohonu Mountain 1974 (Animal Health Division 1974)
CPossums from Buller and Inangahua counties 1974 (Cook 1974)

Lesions have also been recorded in splenic and renal lymph nodes, prostate and mammary glands, stomach, intestine and heart (Animal Health Division 1974, Cook 1974, Lake 1974). Bone infections also occur occasionally.

The distribution of lesions in experimentally infected possums is shown in Tables 2 and 3. Following subcutaneous inoculation on the medial aspect of the left thigh the nature and distribution of lesions at post-mortem resembles those seen in naturally infected wild possums. Intranasal instillation produced a less widespread distribution of lesions with involvement of somatic lymph nodes confined to the head and neck only (O'Hara et al. 1976).

page 166
Table 2. Distribution of gross lesions in subcutaneously infected possums (O'Hara et al. 1976).

Table 2. Distribution of gross lesions in subcutaneously infected possums (O'Hara et al. 1976).

Table 3. Distribution of gross lesions in intranasally infected possums (O'Hara et al. 1976).

Table 3. Distribution of gross lesions in intranasally infected possums (O'Hara et al. 1976).

Somatic lymph node lesions are typically a soft fluctuating abscess with a diameter up to 4 cm containing a semi-liquid lime green pus. On occasions these abscesses open and form sinus tracts, discharging their contents. White to yellow nodules up to 2 cm in diameter are the most commonly seen lesions in visceral organs and lymph nodes. They may be multilobular and have a more page 167 caseous centre than a somatic lesion. In some possums miliary white lesions will be seen in the lungs, liver, spleen or kidneys. The lungs of other possums may have generalised grey-white consolidation of part of a lobe or lobes.


Histologically the lesions can vary from typical tubercle formation to poorly organised lesions composed mainly of necrotic tissue.

Lake (1974) describes the typical areas of nodular consolidation as being composed of granulomatous tissue, with no distinctive fibrous capsule and with varying amounts of amorphous eosinophilic debris centrally. Reactive infiltrating cells included lymphocytes, plasma cells, an occasional eosinophil, macrophages, giant cells and many neutrophils. Neutrophils were in greatest concentration around the area of caseous material. No mineralisation was observed. Acid-fast bacteria were often present in large numbers; both free and phagocytosed in macrophages. Smith (1972) described the reaction as being mainly suppurative and estimated that approximately 5 × 109 organisms were present in 1 gm of tissue.

Typical tubercles with epithelioid cells, Langhans giant cells and capsule formation have been seen on rare occasions.

Preliminary observations from experimental work (O'Hara et al. 1976) have suggested that the earliest lesions appear to be small foci of necrosis of the invaded tissue infiltrated by neutrophils. These foci are invaded by macrophages which accumulate in considerable numbers but appear to be randomly arranged and do not organize as epithelioid cells as seen in other species. At this stage small to moderate numbers of bacilli are present. Coagulative necrosis occurs at the centre of the macrophage nodules, the necrotic tissue becomes caseous and bacilli are usually more numerous. While caseation does occur, there is usually a substantial amount of coagulative necrosis indicating that necrosis is proceeding at a rapid rate. The granuloma does not acquire a significant capsule and the formation of satellite granulomas is common. The large lesions found in the lungs and lymph nodes consist of large necrotic centres surrounded by a narrow rim of macrophages and neutrophils. Liquefaction is common in these large lesions and at this stage the number of bacilli present is reduced.

page 168

The limited cellular response, lack of organisation of the response, extensive tissue necrosis and high rate of multiplication of bacilli are the histological hallmarks of a susceptible host.


Mycobacterium bovis is the most common isolate from tuberculous possums. Ninety out of one hundred and nine isolates from possums were typed as M. bovis. M. vaccae was isolated on seventeen occasions and M. avium and M. fortuitum on one occasion each (Ekdahl 1976).

M. vaccae is a rapidly growing organism and is considered to be saprophytic. It has been isolated most commonly from possums which had lesions histologically attributed to other causes although in three out of 20 cases a dual infection of M. Vaccae and M. bovis was found. Experimental inoculation of possums with a suspension of M. vaccae failed to produce lesions (O'Hara et al. 1976). Thus, it appears as if M. vaccae is an opportunist which can colonise lesions in possums.


The susceptibility of the possum to the effects of M. bovis infection has prompted speculation on, and some investigation into the cell mediated immunity of the species.

Artificially infected possums showed no significant hypersensitivity to intradermal tuberculin whereas BCG vaccinated possums did show a slight degree of reaction. The vaccinated possums did show an increased resistance to subsequent infection manifest by longer survival times, fewer and less widely distributed lesions and the development of fewer but larger pulmonary lesions. The latter lesions were large but discrete areas of lung replaced by caseous tissue (O'Hara et al. 1976). These findings suggest that the possum may be a useful model in studying the relationship between hypersensitivity and immunity as the role of hypersensitivity in the immune response of animals to mycobacteria is not completely clear (Salvin and Neta 1975, Youmans 1975).

It has been found that possum splenic leucocytes are less readily induced to transform by low levels of mitogen than are rabbit cells (Moriarty 1973). Lymphocyte transformation is one in vitro method of assessing cell mediated immunity.

page 169

Thus, there is some evidence that cell mediated immunity as a means of controlling tuberculosis is not as well developed in the possum as in other species.


The prevalence of tuberculosis in infected possum populations as determined by post-mortem examination of poisoned animals has varied with 20% being the highest recorded (Davidson 1976) - see Table 4.

Affected possums have most frequently been found in scrub or bush within two kilometres of cattle pastures (Davidson 1976).

Table 4. Prevalence of tuberculosis in populations as determined by post-mortem examination of poisoned, trapped or shot possums.
Mokihinui river 1970 not recorded 12% (1)
Buller County 1970-71 202/4193 4.8% (2)
Inangahua County 1970-71 89/1715 5.2% (2)
Hohonu mountain 1973 115/1486 7.7% (3)
Waikato 1973 2/1629 0.1% (4)
Taumarunui & Otorohanga Counties 1974 26/1602 1.6% (4)


To date tuberculosis has not been recorded in wild possums in Australia so we presume that infection occurred once they became established in New Zealand. Thus there can be little doubt that the possums were originally infected from cattle. When this occurred it is impossible to say. It is likely that cattle-to-possum transmission occurred independently in widely separated areas in New Zealand (Davidson 1976).

page 170

Possum-to-possum transmission of tuberculosis has been found to occur readily under experimental conditions. It has occurred between possums in direct contact (Bolliger and Bolliger 1948, O'Hara et al. 1976), close contact in adjacent cages and by aerosol over distances up to 180 cm (O'Hara et al. 1976).

It is possible to infect possums experimentally by intraperitoneal and intramuscular injections or by feeding infected material (Bolliger and Bolliger 1948), also by subcutaneous inoculation and intranasal instillation (O'Hara et al. 1976). All these factors suggest that transmission of tuberculosis could readily occur in possum populations especially when possums are in close association such as in the denning site. It is highly probable that the disease can maintain itself in a wild population and, on occasions, result in spread to neighbouring populations.

There are two ways in which man can disseminate tuberculosis from one population of possums to another; by the transportation of infected carcases to a part distant from the area of collection, and by introducing tuberculous cattle into a previously uninfected possum habitat (Davidson 1976).

The potential risk of infected possums contaminating pastures which could then be grazed by cattle has been realised for some time (Ekdahl et al. 1970). The importance of possums as foci of infection for cattle is now more clearly understood with infected possum communities discovered in some 23 general localities in New Zealand (Davidson 1976), and compelling circumstantial evidence linking tuberculosis in possums to high and persistent reactor rates in cattle sharing the same environment (Animal Health Division 1976). The transmission between possum and cattle is presumably through contamination of cattle pasture and possibly also by investigation of possum carcases by cattle (Davidson 1976).

Possum trappers and people involved in autopsies on possums in survey work are obviously exposed to a significant health hazard (Ekdahl et al. 1970, Davidson 1976).

To date only one case of human tuberculosis has been ascribed to handling diseased possums, this was a case of a wound infection in a trapper (Davidson 1976). Protective clothing is made available to all people involved in survey work. Possum trappers can minimize the risk by limiting their handling of obviously diseased carcases and by maintaining good hygiene during and after page 171 handling carcases. It is important that trappers are taught to recognise lesions which may be tuberculous in possums.


The problem of tuberculosis in wild possums in New Zealand bears many similarities with tuberculosis in the badger Meles meles in South-west England. In 1971 M. bovis was isolated from a badger carcase found on a farm in Gloucestershire where bovine Tb existed in cattle (M.A.F.F. 1976). Since then studies on tuberculosis in wild badgers have been made (Muirhead et al. 1974, Gallagher et al. 1976), and experimental work on transmission of infection, viability of the bacillus and badger ecology and behaviour have been initiated (M.A.F.F. 1976).

Behavioural changes have been reported in some badgers with advanced lesions. These badgers may leave their sets in search of more easily available food supplies or they may be rejected by the remaining set occupants (Muirhead et al. 1974). In possums I think the former reason most likely as, at least in experimental studies, they do not appear to reject sick animals and have continued to sleep with the carcase of a recently dead animal.

Inhalation is thought to be the main route of infection in badgers because of the high prevalence of lung or pulmonary lymphnode lesions, the dusty conditions that can prevail in sets, the finding of dust aggregations in lungs and the knowledge that dust particles can render the lung more susceptible to tuberculosis (Gallagher et al. 1976). The possum similarly has a high prevalence of lung infections but it is unlikely that dust would be a factor in the pathogenesis of the disease. The possibility of aerosol infection directly from infected possums occupying the same denning site is obviously high.

This intimate association in the denning site may also account for the high prevalence of superficial lymph node lesions in the possum: there could be direct contamination of wounds with infection localising in the regional lymph nodes. Bite and scratch wounds inflicted during the mating season may also be a factor in this method of spread of the disease.

The kidney is considered to be a predilection site for lesions following haematogenous spread in the badger (Gallagher et al. 1976). Such a site has not been defined in the possum. The superficial lymph nodes have been suggested (Animal Health Division 1974). The lungs should also be considered because of page 172 the high prevalence of lung lesions in possums and the finding that lungs were always affected following subcutaneous inoculation. Unfortunately no correlations have been made between the different sites affected in possums so we do not know the usual distribution of lesions within an individual animal. This information could give some information on the method of transmission of tuberculosis in the possum.

There is similarity between the gross lesions in the two species though it appears as if the badger develops miliary lesions in the lung more often. Histologically there appears to be no great differences between the two species.

As in the possum the tuberculin test was found not to react in experimentally infected badgers (M.A.F.F. 1976). The mild cellular reaction in field cases of badger tuberculosis has suggested that the badger has a low hypersensitivity rating, that the development of lesions is likely to be slow and the clinical course more prolonged than might be expected in the bovine. This hypothesis is supported by the finding that only two-thirds of the infected badgers examined had distinct visible lesions (Gallagher et al. 1976). The possum could also be considered to have a low hypersensitivity rating but experimental work has shown that the clinical course of the disease is rapid following inoculation with large numbers of organisms. It will require more work to see if the above hypothesis holds true when low numbers of organisms are inoculated.

The large numbers of possums examined in surveys have precluded the practicality of examining tissues from all animals microbiologically as was done in the British survey. But, in light of their finding one-third of infected animals with no distinct visible lesions, I feel that it is now necessary for us to do such a detailed survey. This does not mean that a full microbiological examination is necessary in all surveys as infected animals without gross lesions can hardly be considered a significant source of infection for cattle. This information is necessary to better understand the pathogenesis and epidemiology of the disease.

Parallels can also be seen in the epidemiology of the disease in the two situations. Both species were most likely originally infected from tuberculous cattle and now are both maintaining the disease in their own populations and serving as a reservoir of infection for cattle. The mechanics of transmission between the wild species and cattle has been suggested to be pasture in both instances.

page 173

These comparisons have been made to give a better understanding of the problem of tuberculosis in wild animals. By pooling information from both sources one is then in a better position to formulate a policy aimed at solving such a problem.


Animal Health Division 1974. Tuberculosis in possums. Hohonu Mountain M.A.F./N.Z.F.S. Project 117. Animal Health Division Technical Report, 1974.

Animal Health Division 1976. Tuberculosis in opossums. Animal Health Division, M.A.F., Technical Reports, 1976.

Cook, B.R. 1974. Tuberculosis in possums. Buller and Inangahua Counties. Animal Health Division Special Report, 1974.

Bolliger, A. & Bolliger, W. 1948. Experimental transmission of tuberculosis to Trichosurus vulpecula . The Australian Journal of Science 10: 182–183.

Davidson, R.M. 1976. The opossum as a source of tuberculosis to man and animals. Tuberculosis seminar, Hamilton, N.Z., August 1976.

Ekdahl, M.O. 1976. Classification of mycobacterial organisms isolated in New Zealand. Tuberculosis seminar, Hamilton, N.Z., August 1976.

Ekdahl, M.O. , Smith, B.L. & Money, D.F.L. 1970. Tuberculosis in some wild and feral animals in New Zealand. N.Z. Veterinary Journal 18: 44–45.

Gallagher, J. , Muirhead, R.H. & Burn, K.J. 1976. Tuberculosis in wild badgers (Meles meles) in Gloucestershire: pathology. Veterinary Record 98: 9–14.

Lake, D.E. 1974. Opossum tuberculosis survey - Waikato region 1974 .

M.A.F.F. 1976. Bovine tuberculosis in badgers. Ministry of Agriculture, Fisheries and Food Report, November 1976.

Moore, J. 1903. Tuberculosis in an Australian opossum. Veterinary Journal 8: 283.

Moriarty, K.M. 1973. A possible deficiency of cell-mediated immunity in the opossum, Trichosurus vulpecula, in relation to tuberculosis. N.Z. Veterinary Journal 21: 167–169.

Muirhead, R.H. , Gallagher, J. & Burn, K.J. 1974. Tuberculosis in wild badgers in Gloucestershire: epidemiology. Veterinary Record 95: 552–555.

O'Hara, P.J. , Julian, A.F. & Ekdahl, M.O. 1976. Tuberculosis in the opossum (Trichosurus vulpecula): an experimental study. Tuberculosis seminar, Hamilton, N.Z., August 1976.

Salvin, S.B. & Neta, R. 1975. A possible relationship between delayed hypersensitivity and cell-mediated immunity. American Review of Respiratory Diseases 111: 373–377.

page 174

Scott, H.H. 1928. Tuberculosis in marsupials. Proceedings of the Zoological Society of London 1928 : 249–256.

Smith, B.L. 1972. Tuberculosis in the opossum. N.Z. Veterinary Journal 20: 199.

Youmans, G.P. 1975. Relation between delayed hypersensitivity and immunity in tuberculosis. American Review of Respiratory Diseases 111: 109–118.

1 Sources cited: (1) Davidson 1976 (2) Cook 1974 (3) Animal Health Division 1974 (4) Lake 1974.

* Axillary and post scapular lymph nodes

+ External iliac lymph nodes

§ Mediastinal and bronchial lymph nodes