Canine Distemper Infections, with Special Reference to South Africa, with a Review of the Literature

INTRODUCTION Canine distemper (CD) is a severe, life-threatening disease with a worldwide distribution. It was first reported in Europe in 1761 8. During the first half of the 19th century it was the most lethal disease of dogs. It affects mainly domestic dogs but has become a serious problem in a wide range of hosts, threatening captive and free-ranging wildlife populations including several marine mammals such

clinicians need to make about the admission of suspect cases to their hospitals that could expose the hospital population to the CDV.As a rule, cases that present with acute catarrhal disease should not be admitted to an in-patient facility unless very strict isolation and barrier nursing can be applied.Specimens should be collected from these cases for diagnosis, and the dogs should then be managed as outpatients.A scenario that becomes very difficult to control is an isolation ward where puppies with Parvovirus infection are kept.The 2 diseases have clinical features in common and CD can be devastating in a ward of Parvovirus-infected puppies.Chronic CD that manifests with neurological signs or nasodigital hyperkeratosis is far less likely to be a source of infection.Because CDV remains a differential diagnosis in many cases showing CNS disease, it becomes impractical to separate these cases from the rest of the hospital population.The risk in such situations is so low that for practical purposes it can be regarded as insignificant.Dogs recovering from the infection are usually immune, are not persistently infected and do not shed the virus.Transplacental infection has been reported 93 , although the general consensus is that morbilliviruses are not transmitted vertically 86 .
The infection rate is significantly higher than the disease rate, with between 25 and 75 % of susceptible dogs clearing the infection without showing any clinical disease 62 .A constant supply of puppies or susceptible dogs is required to maintain the infection in a population.Several outbreaks of CDV infection have been described in both vaccinated and unvaccinated dog populations in developed and developing environments 26,51,82,130 .In a cross-sectional study of the canine population in a rural town in southern Africa, 5.5 % of the dogs examined were diagnosed with active CDV infection 130 .In a second study that compared the disease status of canine hospital patients from developed communities with those from developing communities over a 3-year period, a remarkable difference was found between the 2 groups.While 43.6 % of the dogs from the developing community had infectious disease, only 8.2 % of dogs from the developed community were similarly diagnosed.CD was diagnosed in 4.1 % of the infectious disease cases in the developing community dogs, but infection with CDV was rarely diagnosed in the dogs from the developed community 49 .This highlights the epidemiological requirement for CDV to persist in dogs, and the risk to certain southern African dog populations.The disease is constantly active in developing commu-nities, where outbreaks can be expected.Owners in these communities must be encouraged to be especially vigilant in keeping to annual vaccination practices.The estimated dog population in South Africa varies widely but conservative estimates place it at around 4 million (1999/2000).Approximately 1 million dogs will visit a veterinarian at least once a year (Pfizer and Intervet companies, South Africa, pers.comm., 1999).This implies that the greater proportion of the South African dog population is unvaccinated and probably also reproductively active.The conditions for persistent large-scale CDV disease are therefore ideal, as it has been estimated that at least 300 000 individuals are required to maintain a Morbillivirus in circulation 23 .This situation poses a threat to the domestic dog population, but the threat it poses to the wildlife resource of the country is equally serious.

PATHOPHYSIOLOGY
Dogs are exposed to the virus by aerosol droplets contacting the upper respiratory tract epithelium 11,62 .The virus multiplies there in tissue macrophages and spreads via the lymphatics to the tonsils and respiratory tract lymph nodes within 24 hours 16,92 .By d 2-4 other lymphoid tissues are infected, and by day 6 the spleen, gastrointestinal mucosa and hepatic Kupffer cells are infected.By this time there is a systemic reaction to the infection characterised by pyrexia and lymphopaenia.Further haematogenous spread of virus occurs to involve other epithelial tissues and the central nervous system (CNS).Virus has been demonstrated to travel cell-free in plasma as well as associated with leukocytes and thrombocytes 91,95 .By 2 weeks after infection, dogs with an adequate cellular and humoral response clear the infection, and no clinical signs of illness persist 75 .Intermediate immune responsiveness allows further spread of virus to epithelial tissues.Virus persists in some tissues (such as nervous tissue and epithelium, especially foot pads), resulting in delayed clinical signs in these tissues.A poor immune response at this point allows pan-systemic viral dissemination that causes the gastrointestinal and respiratory signs seen with acute infection 100,140 .Serum antibody response varies inversely with the severity of the disease 75 .The mortality in gnotobiotic dogs is similar to that caused by natural infections, which discounts the importance of secondary bacterial infections in disease severity despite their obvious involvement [96][97][98] .
Canine distemper virus is a significant cause of immune suppression in host animals 100,124 .Although the molecular basis for immune suppression is known for only a few viral diseases, canine distemper is characterised by viral replication or the presence of virus during the viraemic phase in T lymphocytes, B lymphocytes and macrophages.Necrosis and lymphoid depletion are the essential lesions in the lymphoid tissues 121 .Infection of cells involved in phagocytosis, antigen presentation and the non-specific effector aspects of cell-mediated immunity in dogs with distemper leads to an inability to generate an effective in vivo immune response 134 .Acutely-infected dogs fail to respond to intradermal skin tests with phytomitogens, and do not develop significant antibody responses.The latter suggests a direct viral effect on B cells or their precursor cells in the bone marrow 124 .
Spread to the CNS depends on the efficacy of the initial immune response and many dogs probably develop transient CNS infections without clinical signs.Virus probably gains access to the CNS via infected white cells in the CSF and thrombocytes 16,94 .Antigen is first detected in CNS capillary and venular endothelium and perivascular astrocytic foot processes 21,157 , but this occurs long before signs of CNS disease occur.It seems plausible that viral access is via CSF, as the first lesions seen are in the periventricular sub-pial tissue and in the choroid plexus.Areas of predilection for infection include the cerebral cortex, optic tracts, cerebral peduncles, cerebellum and spinal cord 21,131,135,153,157 .Many factors affect the type of CNS lesions observed, including the efficacy of the immune response and the neurotropic and immune-suppressive characteristics of the particular virus strain 144 .Chronic lesions are classically inflammatory and develop in dogs unsuccessful at clearing the virus from the CNS 27,120,157,165,166 .
The neurobiology of demyelination in CDV infection is of particular interest and has been the topic of several studies, as it is regarded as a model of human demyelinating disease 36,37,42,142 .The acute demyelinating lesions are non-inflammatory, develop during a period of severe immune suppression, and coincide with replication of virus in glial cells of the white matter 118,122,153,157 .An obvious explanation for this would be infection of the oligodendrocytes, which are the myelinproducing cells.Viral protein is, however, only very rarely demonstrated in these cells, and in fact it was initially believed that oligodendrocytes could not become infected with virus 158 .Astrocytes are the most commonly-infected cells.Although viral protein is not produced in oligoden-drocytes, viral mRNA corresponding to all viral genes has been demonstrated by in situ hybridisation in oligodendrocytes, and it has been shown that these cells will support transcription of all CDV genes 167,168,168 .Thus a restricted form of oligodendrocyte infection occurs and may well be the mechanism responsible for demyelination.Down-regulation of myelin gene transcription (the effective shutting-down of the oligodendroglial cell metabolic machinery) in the areas of demyelination in the CNS has been demonstrated 61 .
The mechanisms of chronic demyelination are varied, but appear to be related to the inflammation that characterises this phase of the disease 154,157 .As the immune competence of the host recovers, lymphocytes, plasma cells and macrophages accumulate in the perivascular regions, leading to tissue damage and even necrosis.This inflammation is characterised by intrathecal immunoglobulin production.Dogs that recover have decreasing levels of immunoglobulin production in the CSF whereas increasing levels are associated with progressive disease.Intrathecal production of antibodies to CDV is clearly an important mechanism of immunity, as clearance of the virus from the CNS is associated with recovery 41,80,139,147,159,162 .CSF anti-CDV titres may be higher than in the blood 155 .Intrathecally-produced antibodies to myelin are not a consistent feature of CNS CDV infection, and titres are also not related to disease severity 162 .A cellmediated response to myelin basic protein has been demonstrated in experimental disease but is poorly correlated to the occurrence of demyelination 145,155 .Thus, although autoimmunity has been suggested as a mechanism of progressive neural injury in CDV infection, it remains a topic of debate and is regarded by some to be an epiphenomenon not primarily involved in the demyelinating process 143,157 .On the other hand, major histocompatibility complex class II (MHC II) expression has been demonstrated in the CNS of CDV-infected dogs, especially in the microglia and astrocytes of the white matter 3,160 .The brain is usually regarded as an immunologically-privileged site that cannot express MHC II and thus cannot mount its own immune response.Aberrant antigen-presenting capacity by cells of the CNS has been reported in autoimmune-like diseases of the CNS such as multiple sclerosis, experimental allergic encephalitis and Theiler 's murine encephalomyelitis virus infection 103 .Expression of MHC II by microglia and astrocytes in CDV infections is probably a direct viral effect.The significance of this lies in the fact that microglia and astrocytes expressing MHC II would allow presentation of 'self antigens' that could lead to virus-independent demyelination.In chronic lesions, MHC II was found to be highly expressed despite the absence of viral antigen, suggesting that demyelination may be triggered and perpetuated by non-viral (self) antigens.
Antiviral immune responses have been blamed for 'bystander ' demyelination 28,65,66 .Macrophages constitute an important proportion of the cells that respond to infection.These cells have been shown to play an important role in producing reactive oxygen species that lead to selective destruction of oligodendrocytes as 'innocent bystander ' cells 65,67 .It has also been shown that CDV infection leads to enhanced macrophage activity and hence enhanced destructive activity 30 .
It is important to realise that viral clearance results in CNS recovery and only persistent infection of the CNS can lead to a progressive smouldering inflammatory reaction.It is possible for virus to persist in white matter areas outside the inflammatory demyelinating lesions or even in the periphery of such lesions.Mechanisms allowing such persistence are poorly understood but are obviously important to the evolution of disease and are the topic of ongoing investigation 27 .
Cytokines play a significant role in any disease in which inflammation contributes to the disease process.To date there have been few studies evaluating the effect of cytokines in CDV infections 55, 68 .One of the hindrances to this type of work in dogs is the lack of species-specific reagents.One study evaluated mRNA expression of interleukin 1(IL-1) , IL-6, IL-8, IL-12, tumour necrosis factor " (TNF), interferon ( (IFN)and TGF$in whole blood using reverse transcriptase polymerase chain reaction (rt-PCR) 68 .No significant correlation between disease and cytokine mRNA expression was demonstrated.No cytokine transcripts could be found in the blood of any of the acutely viraemic dogs with a high viral load.This may have been due to immune suppression.Pro-and anti-inflammatory cytokines revealed no clear pattern of expression.It is highly probable that the form of disease, immune status of the individual and time of sampling all impact strongly on cytokine expression.Circulating levels of cytokine mRNA may also be poorly correlated with cytokine protein production at tissue level.Another study examined mRNA expression of various cytokines in the CSF of naturallyinfected dogs 55 .IL-10 was found to be the dominantly-expressed cytokine, although both pro-and anti-inflammatory cyto-kines were co-expressed in most cases, making it very difficult to link expression with pathogenic effect.A further problem is the heterogeneous nature of CNS lesions at any time.Both non-inflammatory and chronic inflammatory lesions co-exist in the same brain at any particular time.Furthermore, determining cytokine mRNA expression in CSF may not relate to what is actually taking place in the brain tissue.One case with advanced chronic lesions expressed only TNF and IL-12.Ideally, it would be possible to study the cytokine kinetics in brain tissue over time at the same time as being able to evaluate inflammatory cell phenotype.
T-cell subsets in the brains of natural CDV infections have been evaluated 163,164 .Acute demyelination (histologically noninflammatory) is accompanied by a diffuse influx of CD8+ cells.It is suggested that these cells may be responsible for early antibody-independent anti-viral cytotoxicity.The chronic inflammatory perivascular cuffs consist mainly of CD4+ lymphocytes, plasma cells and macrophages.This may represent a phase of antibody-dependent demyelination.These findings highlight the heterogeneous and possibly, in part, immunemediated plaque pathogenesis of CDV demyelination.

CLINICAL SIGNS
Signs vary depending on a number of factors such as the age of the host, host immune competence, virus strain and organ system affected 45,74,96 .It is estimated that between 50 and 70 % of CDV infections are sub-clinical 62 .Mild forms of the disease present with non-specific signs such as listlessness, loss of appetite and fever.More specific presentations include oculonasal discharge, coughing, dyspnoea, diarrhoea and vomiting.These signs are not specific for CD and in the experience of the authors Ehrlichia canis infection may be an important differential diagnosis for practitioners in areas where both diseases are present.Both diseases can present with similar secondary skin and chest infections and the typical haematological findings of leukopaenia, thrombocytopaenia and possibly a nonregenerative anaemia.Kennel cough may be another differential diagnosis.It has been the experience of the first author at the Onderstepoort Veterinary Academic Hospital (OVAH) that some puppies admitted to the isolation ward for suspected parvovirus-induced diarrhoea were subsequently found to suffer from CDV infection.This presents a particularly severe problem, as the introduction of CDV to a group of severely immunesuppressed Parvovirus-infected puppies 0038-2809 Jl S.Afr.vet.Ass.( 2001) 72( 3): 127-136 can be catastrophic.Problems with CDV infection following discharge from hospital of puppies that had parvovirus infection is likely in these circumstances.
The form of the disease that presents most commonly depends especially on the level of immune competence of the population and age of the affected dogs.Clinical signs therefore vary widely 154 .In unvaccinated dogs and unvaccinated puppies (of between 12 and 16 weeks of age when maternal antibody titres have waned) the severe generalised catarrhal form is most common.These dogs are febrile, with mucopurulent conjunctivitis, upper respiratory tract signs, depression, anorexia, vomiting and diarrhoea (which may be bloody).Severe dehydration and weight loss may be present.Adequate treatment in this group may reduce the mortality rate.
The most common form of the disease seen at the OVAH is the neurological form in older dogs.The disease occurs frequently enough to justify investigation for CDV infection in all dogs showing non-surgical spinal cord disease and in all dogs with brain signs.
In an outbreak described in an urban dog population of the Copenhagen area in Denmark, more than half of the cases presented with the catarrhal form.Many of these dogs also had CNS involvement and digital hyperkeratosis (so called 'hard-pad') 25 .In an outbreak described amongst unvaccinated sled dogs in northern Greenland (with a mortality of 1000 of a population of around 3000), most died within a week with classic catarrhal and nervous signs.Digital hyperkeratosis was not observed.Blindness was observed in a few survivors 26 .In a Finnish study of an outbreak in approximately 5000 vaccinated dogs with a mortality of 30 %, the majority showed classic catarrhal disease.A few cases of digital hyperkeratosis were noted.Nervous signs were noted but the proportion was not reported 51 .
Typically, neurological signs follow the catarrhal signs within about 2-3 weeks 10,11,62 .As demonstrated by previous studies and observations cited above, many dogs will present for the first time with neurological signs preceded by no or only very mild signs.In one report, only one third of cases with neurological disease had a previous history of systemic signs, and another third had no previous history of disease at all 148 .Once CNS disease manifests, it is usually progressive, without a relapsing course, although a chronic relapsing course has been described 73 .Neurological disease often appears localised according to clinical examination, despite the diffuse viral and lesion localisation 148 .Signs are very varied but commonly include seizures 105 , blindness, central vestibular disease 52 , cerebellar 87,118,142,154 , brain stem and spinal cord disease 135 .Myoclonus occurs in about one third of cases and is most commonly associated with CDV, but should not be regarded as pathognomonic, as it is seen in other causes of menigoencephalomyelopathies 77,78,146 .It is seen most commonly in the muscles of mastication and the limbs, but may be seen in any muscle group 77,78 .Naso-digital keratosis is reported to occur in about 8 % of cases with neurological disease 148 .Bi-or unilateral optic neuritis is often the cause of blindness.Swelling of the optic disc may be seen in the acute disease.Ill-defined grey to pink densities may be seen in the tapetal and non-tapetal regions.Chronic disease may cause well-delineated hyperrefletic retinal areas 24,57,88,107 .The immune compromising nature of the disease 92,100,140 results in the frequent occurrence of secondary bacterial infection of the skin and respiratory tract which are features of the acute phase of the disease.The activation of other unusual and CNS infections owing to the immune suppression caused by CDV has been reported, and these infections include combinations of CDV and Toxoplasma gondii 5 0 , Neospora caninum 148 , Filaroides hirthi 32 , cryptosporidiosis 149 , Pneumocystis carinii 141 and canine infectious hepatitis virus 89 .
CDV infection has been associated with inflammatory joint diseases and hypertrophic osteodystrophy (HOD) in dogs 20,116 .In a study of experimental infection of gnotobiotic dogs CDV infection of the proximal metaphyses of the humerus was prominent.Antigen was abundant in the marrow, osteoclasts and osteoblasts, where it had a direct cytolytic effect, even causing cell necrosis.There was a strong association between CDV antigen presence and osseous changes.CDV antigen has been reported in the bone of humans with Paget's disease and an association between dog ownership and this disease in man has been postulated 58 .There is, however, very little evidence for this, and other studies have failed to support this hypothesis 54,85,114,115,123,129 .A similar link between CDV infection and multiple sclerosis in man has been suggested, but there is no proof of this association, and CDV should be regarded as an animal disease with no public health risk 36,37,39,60,76,101,133 .In an earlier study linking CDV infection and HOD, dogs injected with serum from HOD patients developed CDV signs and the infection was confirmed at necropsy 69 .CDV has also been demonstrated by rt-PCR, in situ hybridisation and southern blotting in the bone tissue of dogs with HOD 112,113 .It is possible that all dogs with CDV infection develop sub-clinical bone disease, especially considering that dogs with HOD develop constitutional signs of disease such as fever and malaise.Infection during puppyhood may cause enamel hypoplasia in the permanent teeth seen in adulthood 22,45 .
A rare progressive condition that has come to be called 'old dog encephalitis' is purported to occur in dogs that have survived CDV infection.It is characterised by widespread perivascular lymphoplasmacytic infiltration in areas of demyelination and neuronal degeneration.In some cases it has also been described as a sclerosing panencephalitis with infiltration and replacement of nervous tissue by a dense astrocytic network.Clinical signs are typically restricted to lesions typical for the cerebrum and exclude involvement of the brain stem, spinal cord and cerebellum 104 .
A link between juvenile pyoderma and CDV has been suggested.A pustular skin disease is not uncommon in CDV infection and has been described as an 'allergic response to the virus' 6 .In a recent report, 3 of 4 dogs recently vaccinated with a modified live CDV vaccine developed juvenile cellulitis and radiographic evidence of HOD.The skin disease was successfully treated in all 4 dogs with prednisolone.Distemper-like skin lesions have also been described following inoculation with a modified live virus vaccine 34 .The concurrence of HOD and juvenile cellulitis has been reported previously 35,106,117 .

EXPERIENCE WITH THE DISEASE IN SOUTH AFRICA
A retrospective study of cerebrospinal fluid samples submitted to the Department of Veterinary Tropical Diseases, Faculty of Veterinary Science, University of Pretoria, Onderstepoort, from dogs with clinical signs compatible with CDV infection was conducted during 1998 and 1999.
One hundred and thirty-three CSF samples were submitted during this 2-year period.Thirty-four (25 %) were found to be positive for CDV-specific IgG.Of these 34 cases, 23 showed nervous signs only (including tremors, vestibular syndrome, hind-limb ataxia, progressive ataxia, seizures, myoclonus, depression, periodic neck pain, generalised body pain and dementia).A combination of neurological signs with gastrointestinal signs was seen in 4 cases.No dogs were observed to have a combination of neurological and respiratory signs.The typical non-neurological signs seen were purulent oculonasal discharge (4 dogs) and respiratory signs (2 dogs) that included pneumonia-induced coughing and increased respiratory rate.The gastrointestinal signs observed were anorexia, diarrhoea and vomiting (4 dogs).No cases of digital hyperkeratosis (so called 'hardpad disease') were noted.Nineteen of the 34 CSF CDV-specific IgG-positive dogs were less than a year old, and 12 of these presented with only neurological signs.The remaining 15 dogs were older than a year, 2 being older than 7 years.Fourteen of these 15 dogs older than a year showed only neurological signs.No seasonal incidence was noted over the 2 years.The results of this study highlighted several important issues about the disease as it is seen at the OVAH.These include: At a satellite clinic of the Medical University of Southern Africa that served a more resource-deprived community than the OVAH, catarrhal signs alone and in combination with neurological signs were more common than neurological signs alone 33,49,130 .
In a retrospective evaluation of 154 dog brains presented to the OVI for rabies diagnosis between July 1992 and July 1993, a final diagnosis of CDV was made in 26 cases.Diagnosis was based on immunoperoxidase staining for inclusion bodies in at least 1 of 6 standard sections.Other common diagnoses included cerebral babesiosis, E. canis infection and granulomatous meningoencephalitis (J J van der Lugt, pers.comm., 1999).

DIAGNOSIS
Diagnosing canine distemper in the live animal with confidence remains difficult, yet very important because of the implications of the disease.Identifying typical clinical signs still remains the most important initial diagnostic step.Because the disease often has a poor prognosis and has a high infectious potential, ante mortem confirmation of the diagnosis is preferred.The many varied clinical presentations of the disease (especially in its neurological form) may also dictate that clinical signs are by no means typical, further emphasising the need for a definitive diagnostic test.Numerous different methods that have been used for diagnosis testify to the difficulties experienced in this regard.
The most common haematological finding is lymphopaenia.In a local study 86 % of 30 dogs with CDV infection were lymphopaenic 33 .Thrombocytopaenia may also be present 15 .A left shift neutrophilia may be helpful in confirming the presence of significant secondary bacterial infection.In the South African setting, because of the frequency of other concurrent infectious diseases, haematological findings may be varied if mixed infections are present.Other immune-suppressive infections (such as ehrlichiosis or parvovirus infections) can induce very similar haematological abnormalities.This is especially relevant because many of the dogs suspected of having distemper infection live in developing communities where infectious diseases are very common and vaccination is seldom practised.Viral inclusions in the white or red cells are rarely observed.Serum biochemistry is usually unremarkable.Radiology may demonstrate an interstitial (viral) pneumonia in early cases and an alveolar pattern with bacterial pneumonia later in the course of the disease.
CSF and serological abnormalities and the demonstration of viral inclusions in epithelial cell cytology (by fluorescent antibody techniques) are usually the most practical and useful means of confirming a diagnosis.
The diagnosis of inflammatory diseases of the canine CNS is usually difficult, with at least one third of cases remaining undiagnosed in the live animal despite thorough investigation 146 .However, results of analysis of the CSF of CDV-infected dogs are frequently useful in diagnosis.Raised protein content and lymphocytic pleocytosis are common 1,41,64,81,139,146,156 .In the study by Carter, 89 % of the 30 CDV positive dogs examined had CSF protein concentrations in excess of 0.5 g/ (normal is <0.25 g/ ), and lymphocytes were also the most frequently-encountered cells 33 .Finding anti-CDV IgG antibodies in the CSF is definitive in CDV encephalitis, as the antibody is locally produced 41,64,80,146,147,156 .This is not the case in vaccinated dogs or dogs without CNS CDV infection.Vaccinated or exposed and cured dogs should all have a blood IgG titre that, if present in CSF, would indicate blood contamination during CSF collection.This problem can be overcome by determining concomitant Parvovirus titres in both serum and CSF.If there has been blood contamination of the CSF, a Parvovirus titre will be present in the CSF if it is present in the blood.The only way that this could have occurred is by contamination, as anti-Parvovirus antibodies are not synthesised intrathecally.An IgG index may also be calculated comparing CSF CDV IgG titres to CSF albumin concentrations 147 .
Immunocytological examination of smears from the conjunctiva, tonsillar, genital or respiratory epithelium may demonstrate virus if smears are made early in the course of disease (within 2 weeks post-infection) 1,19 .Finding antigen by this method in buffy coat smears would also confirm a diagnosis, but only during the early brief viraemic phase of the disease (usually within 10 days post-infection).As antibody titres rise, antigen is either bound and masked from immuno-staining, or disappears from the epithelial surfaces altogether.A recent report has demonstrated the usefulness of CDV-specific immunoperoxidase histochemical test in skin biopsies (foot pad, nasal mucosa and haired skin form the neck) in diagnosis.Biopsies from the haired skin from the neck were as likely to be positive as epithelium from the other sites in this study.This method proved useful in acute and subacute disease 71 .Although the experience base with this method is small, its efficiency, ease and cost-effectiveness should encourage wider use.Immunochemical staining for antigen has also been successful in cells from CSF 4 and urine 29,136 .
Detection of serum IgG and IgM titres has been widely used.Because of the practice of vaccination and the frequent exposure of dogs to self-limiting infection, detecting a positive IgG titre in serum is of very little value in confirming a diagnosis.A positive IgM titre, on the other hand, should be interpreted as recent exposure (during the last 3 weeks) to vaccine virus or natural infection 62 .
With the advent of molecular biological technology, rt-PCR has been used to detect viral nucleic acid in blood, serum and CSF 56,68,137 .This highly sensitive and specific mode of diagnosis can be expected to be widely used in future.
Virus isolation is useful, but special arrangements need to be made with the diagnostic laboratory concerned.This technique is usually reserved for research environments and has proven useful to differentiate natural infections from vaccine-induced disease in exotic species.Vaccine strains can replicate in macrophage, lymphocyte and epithelial cell cultures, whereas field strains require several blind passages to replicate in epithelial cell lines 72 In a South African study Carter 33  This study concluded that the following tests, in order of preference, were helpful in confirming the ante mortem diagnosis of CDV: 1. IFA for IgG in the CSF in cases of CDV encephalopathy.2. IFA for IgM in the serum.

Direct fluorescent antibody testing
for CDV antigen in conjunctival scrapings.

Direct fluorescent antibody testing for
CDV antigens in cells in the CSF of cases with CDV encephalopathy.These findings have proven useful and practical in the resource-deprived communities of South Africa.A similar survey of diagnostic tests in dogs from developed areas has not been performed, and would probably yield slightly different results because the acute catarrhal form of the disease is rare in this population of dogs.Most CDV disease in developed communities is of the chronic neurological type.In this setting, CSF IgG titres are most useful, but fluorescent antibody testing of cellular material for analysis is less useful.Reverse transcriptase PCR of CSF fluid may prove to be useful in the future.

CANINE DISTEMPER VIRUS INFECTION IN WILDLIFE
Although CD is an important disease in dogs, there have been some alarming distemper outbreaks in non-domestic species, which further highlights the importance of control of the dog reservoir.Eight of the 11 families of the order Carnivora are susceptible to CDV, with a mortality rate that varies widely between species 5,13,72 .Of the species in which CDV infection has been reported, the following occur in South Africa: lion, leopard, wild dog, hyaena, foxes and otters 70,102,132 .Of these species, exposure to CDV has been demonstrated serologically in lions, leopards and wild dogs.However, no catastrophic mortality due to CDV infection has been reported in southern Africa 102,152 .Distemper outbreaks have been reported in captive lions, tigers, leopards and jaguars in North America 14 .Most of these animals die of CNS disease following respiratory and gut involvement.
Outbreaks have been reported in marine mammals, including seals (so called phocine distemper) 125 and striped dolphins.Thousands died in the Mediterranean between 1990 and 1992 43,46,84,125,150,161 .It is most likely that these outbreaks originated in terrestrial carnivores 53 .
A massive CD outbreak in the lion population of Serengeti (Tanzania) killed about 30 % of the lion population of that area in 1994.The same area experienced CDV-related deaths in bat-eared foxes, African wild dogs, spotted hyaena, common jackal and silver-backed jackal.The source of these infections is purported to be the domestic dog population of about 30 000 that surrounds the park 132 .Several years later 55 % of the lions of Maasai-Mara were found to be serologically positive for CDV, indicating exposure 90 .Preventing the spread of this disease to wildlife would require vaccinating all dogs around the conserved areas or vaccinating the wildlife, which is not practical in free-ranging wild animals.Developing a vaccine that is safe in captive wildlife remains a challenge, as there have been several reports of disease caused by the use of modified live CDV vaccines 40,47,79,110,119 .It would be ill-advised to use a live vaccine in captive wild animals.In this regard, inactivated vaccines (which are no longer commercially available) are best used.Recombinant or subunit vaccines would be desirable for captive wildlife 83,126,127,138 .

PREVENTION
Vaccination remains the mainstay of preventing CDV infection in dogs, and therefore a large amount of research and a large industry has developed around this practice.Maternally-derived antibody provides temporary protection to neonates for a period that varies from days to months.This phenomenon has been responsible for the use of a modified live measles virus vaccine.The 2 viruses are antigenically closely related, and the measles virus is not neutralised by the maternal antibody in 6-week-old puppies 12 .The second and all subsequent vaccines are modified live virus vaccines, as by this time maternal antibody titre has waned to the point of being inconsequential.This long-accepted practice of 'heterotypic' vaccine use has been questioned, as ongoing usage of measles vaccine means that progeny of bitches vaccinated with measles vaccine may possess maternal antibodies against measles virus that would interfere with early measles vaccine in puppies born to them.It has been shown that a high-titre CDV vaccine is even more effective in protecting 6-week-old puppies against experimental distemper challenge than the traditional human measles vaccine 34 .It is generally accepted that vaccination should take place every 2-4 weeks from the age of 6 weeks until 12 weeks, and then yearly thereafter with a modified live virus vaccine 7,31,44 .Frequent vaccination has been practised in an attempt to reduce losses early in life.It has been shown that 'maximal' vaccination (every week from the age of 2 weeks until 10 weeks and then at 12, 16, 18 and 24 weeks) gave no better protection than a 'minimal' regime (5, 9, 13, 17 and 22 weeks) in Greyhound puppies.There was also no adverse effect on cellular immunity of the maximal protocol 111 .Inactivated vaccines do not provide protection against experimental challenge infection 9 .A canary pox recombinant vaccine has recently been developed that provided good protection 127 and is now commercially available.
The need for yearly booster vaccination is often questioned, and in fact in developed countries the recommendation is now that booster vaccinations are only required every 3 years following the initial puppy series and a booster at one year 44 .Boosters for geriatric animals are discouraged 44 .Sufficient levels of neutralising antibodies remain in serum for longer than 6 years after vaccination 31 .It has, however, been claimed that, on the basis of serum antibody titres, annual vaccination should be maintained 109 .On the other hand, it has been suggested that humoral immune mechanisms alone do not account for immunity and that protection lasts much longer than indicated by antibody titres 63 .In a country such as South Africa, where most dogs are unvaccinated and reproductively active, and where a large wildlife resource is potentially at risk, it would be unwise to recommend anything other than yearly booster vaccination.
Not all vaccines are equally effective, as has been demonstrated in numerous comparative studies on the efficacy of various products.Outbreaks of disease have been described in well-vaccinated populations, as indicated above.Despite these reports, the chances of outbreaks are greatly reduced in well-vaccinated populations 128 .It should, however, always be remembered that no vaccine is 100 % effective in 100 % of animals.When individuals present with signs of CDV infection, the disease cannot be ruled out on the basis of an adequate vaccination history.
There are numerous reports of adverse effects of CDV vaccination.Introduction of a foreign protein will always be associated with a certain amount of risk ranging from mild urticarial skin reactions to life-threatening anaphylaxis.It has been recommended that caution should be exercised when using a modified live vaccine in dogs infected with Parvovirus, as this may predispose dogs to neurological disease 99 .Cases of post-vaccinal encephalitis were recorded in dogs in various parts of Britain after the administration of a particular batch of vaccine 38,108 .It has also been shown that there is a temporal relationship between the use of canine vaccines (not specifically CDV) and immune-mediated haemolytic anaemia 48 .Full-blown CDV infection with mortality has been reported following the use of modified live virus vaccines in various wildlife species, as described above.

MANAGEMENT
There is no effective specific antiviral treatment.All treatment focuses on patient support and nursing care.Before treating an infected dog, the infectious nature of the disease and prognosis should be discussed with the owner.The possibility of neurological signs developing should be discussed with owners.This remains an important reason for euthanasia but before this is done, it is worth considering treatment for a period of time in a select group of patients, as some dogs may recover.Patients for which euthanasia should be discussed include: 1) those dogs with neurological disease that compromises quality of life (such as intractable seizure activity, severe myoclonus or incapacitating spinal cord disease), 2) severe and worsening catarrhal forms of the disease, or 3) dogs that come from multiple-dog households where other dogs may be at risk of infection.
Supportive treatment for dogs likely to be shedding virus should be adminis-tered on an outpatient basis as far as possible unless the dogs can be barrier nursed in isolation.This is frequently a problem in the private practice situation in South Africa where most isolation facilities are still used for Parvovirus-infected puppies.CDV suspect dogs should not be treated in proximity to these patients.Antibiotic cover (broad spectrum and effective in the respiratory tract), fluid and electrolyte balance (including anti-emetics if required), nutritional support and seizure control form the basis of a management plan.Good nursing care such as keeping the animal warm and dry and cleaning oculonasal discharges should be applied.Various homeopathic treatment claims have been made.These are scientifically unsubstantiated.
Mawer E B 1998 Detection of canine distemper virus in 100 % of Paget's disease samples by in situ-reverse transcriptase-polymerase chain reaction.Bone 23: 171-175 115.Mee A P, Sharpe P T 1993 Dogs, distemper and Paget's disease.Bioessays 15: 783-789 116.Mee A P, Webber D M, May C, Bennett D, Sharpe P T, Anderson D C 1992 Detection of canine distemper virus in bone cells in the metaphyses of distemper-infected dogs.Journal of Bone and Mineral Research 7: 829-834 117.Meier H, Clark S T, Schnelle G B, Will D H 1957 Hypertrophic osteodystrophy associated with disturbance if vitamin C synthesis in dogs.Journal of the American Veterinary Medical Association 130: 483 118.Mitchell W J, Summers B A, Appel M J 1991 Viral expression in experimental canine distemper demyelinating encephalitis.Journal of Comparative Pathology 104: 77-87 119.Montali R J, Bartz C R, Teare J A, Allen J T, Appel M J, Bush M 1983 Clinical trials with canine distemper vaccines in exotic carnivores.Journal of the American Veterinary Medical Association 183: 1163-1167 120.Muller C F, Fatzer R S, Beck K, Vandevelde M, Zurbriggen A 1995 Studies on canine distemper virus persistence in the central nervous system.Acta Neuropathologica (Berlin) 89: 438-445 121.Muneer M A, Farah I O, Newman J A, Goyal S M 1988 Immunosuppression in animals.British Veterinary Journal 144: 288-301 122.Mutinelli F, Vandevelde M, Griot C, Richard A 1989 Astrocytic infection in canine distemper virus-induced demyelination.Acta Neuropathologica (Berlin) 77: 333-335 123.O'Driscoll J B, Anderson D C 1985 Past pets and Paget's disease.Lancet 2: 919-921 124.Olsen R G, Krakowka S 1984 Immune dysfunctions associated with viral infections.Compendium of Continuing Education for the Practicing Veterinarian 6: 422-430 125.Osterhaus A 1989 A morbillivirus causing mass mortality in seals.Vaccine 7: 483-484 126.Paoletti E 1996 Applications of pox virus vectors to vaccination: an update.Proceedings of the National Academy of Sciences USA 93: 11349-11353 127.Pardo M C, Bauman J E, Mackowiak M 1997 Protection of dogs against canine distemper by vaccination with a canarypox virus recombinant expressing canine distemper virus fusion and hemagglutinin glycoproteins.American Journal of Veterinary Research 58: 833-836 128.Patronek G J, Glickman L T, Johnson R, 0038-2809 Jl S.Afr.vet.Ass.(2001) 72(3): 127-136 Emerick T J 1995 Canine distemper infection in pet dogs: II.A case-control study of risk factors during a suspected outbreak in Indiana.Journal of the American Animal Hospital Association 31: 230-235 129.Ralston S H, Digiovine F S, Gallacher S J, Boyle I T, Duff G W 1991 Failure to detect paramyxovirus sequences in Paget's disease of bone using the polymerase chain reaction.Journal of Bone and Mineral Research 6: 1243-1248 130.Rautenbach G H, Boomker J, de Villiers I 1991 A descriptive study of the canine population in a rural town in southern Africa.Journal of the South African Veterinary Association 62: 158-162 131.Raw M E, Pearson G R, Brown P J, Baumgartner W 1992 Canine distemper infection associated with acute nervous signs in dogs.Veterinary Record 130: 291-293 132.Roelke-Parker, M E, Munson L, Packer C, Kock R, Cleaveland S, Carpenter M, O'Brien S J, Pospischil A, Hofmann-Lehmann R, Lutz H 1996 A canine distemper virus epidemic in Serengeti lions (Panthera leo).Nature 379: 441-445 133.Rohowsky-Kochan C, Dowling P C, Cook S 1995 Canine distemper virus-specific antibodies in multiple sclerosis.Neurology 45: 1554-1560 134.Rouse B T Horohov D W 1986 Immunosuppression in viral infections.Reviews of Infectious Diseases 8: 850-873 135.Schoning P, Layton C E 1992 Canine distemper with spinal cord lesions.Zentralblatt für Veterinärmedizin Reich B 39: 571-574 136.Shen D T, Gorham J R, Pedersen V 1981 Viruria in dogs infected with canine distemper.Veterinary Medicine (Small Animal Clinics) 76: 1175-1177 137.Shin Y, Mori T, Okita M, Gemma T, Kai C, Mikami T 1995 Detection of canine distemper virus nucleocapsid protein gene in canine peripheral blood mononuclear cells by RT-PCR.Journal of Veterinary Medical Science 57: 439-445 138.Sixt N, Cardoso A, Vallier A, Fayolle J, Buckland R, Wild T F 1998 Canine distemper virus DNA vaccination induces humoral and cellular immunity and protects against a lethal intracerebral challenge.Journal of Virology 72: 8472-8476 139.Sorjonen D C, Cox N R, Swango L J 1989 Electrophoretic determination of albumin and gamma globulin concentrations in the cerebrospinal fluid of dogs with encephalomyelitis attributable to canine distemper virus infection: 13 cases (1980-1987).Journal of the American Veterinary Medical Association 195: 977-980 140.Stevens D R, Osburn B I 1976 Immune deficiency in a dog with distemper.Journal of the American Veterinary Medical Association 168: 493-498 141.Sukura A, Laakkonen J, Rudback E 1997 Occurrence of Pneumocystis carinii in canine distemper.Acta Veterinaria Scandinavica 38: 201-205