Rabies

What it is

Rabies is an acute encephalitis caused by rabies virus, of the genus Lyssavirus (family Rhabdoviridae); other related lyssaviruses circulate in bats. The virus is shed in saliva and transmitted mainly by bites; less often, by contamination of wounds or mucous membranes with saliva.

At the bite site the virus multiplies in muscle and enters peripheral nerve endings. From there it travels up by retrograde axonal transport to the spinal cord and brain (centripetal spread), barely triggering an immune response. After replicating in the brain, it travels down the nerves to other tissues (centrifugal spread), especially the salivary glands. The incubation period is very variable, from weeks to months, depending on the bite site, the viral dose and the strain. The animal can shed virus in saliva a few days before showing signs.

It is an almost always fatal zoonosis: the dog is the source of the vast majority of human cases worldwide, mainly in Asia and Africa. WHO, WOAH, FAO and their partners hold the "Zero by 30" goal: no human deaths from dog-mediated rabies by 2030. It is notifiable in many countries and its control follows official programmes.

Who it affects

All mammals are susceptible. Where canine rabies is endemic, the main reservoir is the dog, especially the stray or free-roaming owned population that is not vaccinated. In other regions it is maintained by wildlife (foxes, raccoons, skunks, jackals and bats, including vampire bats in Latin America), which transmit it to dogs, cats, livestock and people. Unvaccinated dogs that roam freely are at greatest risk.

Anatomy involved

• Brain: the target organ. The virus causes a non-suppurative polioencephalomyelitis, with microscopic lesions that are often mild for the severity of the disease. Negri bodies form in the cytoplasm of neurons; these eosinophilic inclusions are typical of the hippocampus (Ammon's horn) in carnivores and of the cerebellar Purkinje cells in ruminants, and their absence does not rule out the disease. Involvement of the brainstem explains paralysis of swallowing and of the jaw, and involvement of the limbic system the behavioural changes. For diagnosis the brainstem is sampled together with the cerebellum and hippocampus.

Clinical signs

In dogs the course usually lasts less than 10 days from the first signs to death, and three overlapping phases are described:

• Prodromal phase (a few days): behavioural change (a friendly dog becomes withdrawn or a shy one seeks contact), mild fever, licking or scratching at the bite site.
• Furious form: restlessness, irritability, unprovoked aggression, biting at objects, disorientation, a hoarse bark and seizures.
• Paralytic or dumb form: jaw paralysis (dropped jaw), dysphagia, hypersalivation and ascending paralysis ending in respiratory paralysis and death. Drooling saliva and dysphagia make the owner think the dog has something stuck in its throat, and examining the mouth by hand is a classic route of exposure.

Diagnosis

• There is no reliable ante-mortem test in animals; signs only allow the disease to be suspected.
• Diagnosis is post mortem, on the brain: the direct fluorescent antibody test (FAT) on brain tissue impressions is the reference test recommended by WOAH. The direct rapid immunohistochemical test (dRIT) and RT-PCR are also used. Histopathology looking for Negri bodies is insensitive and is no longer used alone.
• The head or brain must be sent chilled, not frozen or fixed in formalin, to the official laboratory, handled by trained and vaccinated staff.
• Healthy dog or cat that has bitten a person: it is kept under observation for 10 days. If it is still healthy at the end of that period, it was not shedding virus in saliva at the time of the bite. If it becomes ill, it is euthanised and the brain is tested.

Treatment and prevention

There is no treatment in animals: an animal with signs of rabies is euthanised and its brain tested. Management of exposed animals is set by each country's regulations: in general, vaccinated animals receive an immediate booster and are kept under observation, and unvaccinated animals undergo prolonged quarantine or euthanasia.

• Dog vaccination: inactivated rabies vaccine is effective and essential in endemic areas and where required by law. Mass dog vaccination that achieves high coverage and maintains it with regular campaigns is the most effective way to reduce human cases; it eliminated canine rabies from Western Europe and North America and has drastically reduced it in Latin America.
• Dog population management, responsible ownership and integrated bite surveillance between veterinary and public health services (One Health).
• In people: immediate, thorough washing of the wound with soap and water and post-exposure prophylaxis (vaccine and, depending on the exposure, rabies immunoglobulin), which is highly effective if given in time. Veterinarians and laboratory staff at risk receive pre-exposure prophylaxis.

In the cat

Cats are infected mainly by bites from wildlife or rabid dogs, and in some countries they are the domestic species with the most reported cases, so their vaccination is just as important. In cats the furious form predominates: sudden behavioural changes, intense aggression with unprovoked attacks, abnormal vocalisation, mydriasis and muscle fasciculations, followed by incoordination, paralysis and death. A rabid cat is very dangerous because it bites and scratches quickly. The brain lesions and post-mortem diagnosis are the same as in the dog, and a cat that has bitten a person is also observed for 10 days.

In cattle

In Latin America bovine rabies is transmitted mainly by the common vampire bat, Desmodus rotundus, which feeds on cattle blood and bites mainly on the neck, withers, ears, tail base and limbs. It causes major economic losses in tropical and subtropical areas. In other regions cattle are infected by bites from dogs or wild carnivores.

In cattle paralytic rabies predominates. Signs include behavioural changes, frequent hoarse bellowing, hypersalivation and difficulty swallowing, tenesmus, and ataxia and paresis of the hind limbs that progress to recumbency and death within a few days. Some show a furious form, with aggression and charging. Because hypersalivation and dysphagia resemble oesophageal obstruction, examining the mouth or pharynx without protection is a source of exposure for farmers and veterinarians.

Diagnosis is made on the brain with the same tests as in dogs; in cattle Negri bodies are seen mainly in the Purkinje cells of the cerebellum. Control relies on official programmes: vaccination of cattle in risk areas, surveillance and notification of outbreaks, and control of vampire bat populations by the official services.

In the horse

Horses are infected by bites from wild carnivores or rabid dogs and, in Latin America, also from vampire bats. The species is rarely affected in countries free of terrestrial rabies, but in endemic areas cases occur regularly, and handling the mouth and the neurological examination expose owners and veterinarians.

Signs are highly variable and non-specific, so rabies must be considered in any horse with acute, progressive neurological disease. It can begin as lameness, colic, ataxia and paresis of the hindquarters or difficulty swallowing, and progress to hyperaesthesia, loss of tail and anal tone, fever and recumbency, with death in less than a week. Some horses show a furious form, with aggression, biting and self-mutilation. Rabies is a mandatory differential diagnosis for West Nile virus encephalitis and equine herpesvirus 1 myeloencephalopathy: until it has been ruled out, the horse should be handled with gloves and protection.

Diagnosis is made on the brain with the same tests as in other species. Vaccination of horses with inactivated vaccine is recommended in endemic areas. Horse-specific evidence is weak: the clinical description comes mainly from case series, such as Green et al. (1992), with 21 horses that died a mean of 4.47 days after the first signs, and there are no field trials of vaccine protection in this species.

Key studies

• Hampson et al. (2015), model of the global impact of dog-mediated rabies, built from literature searches and country questionnaires. It estimated about 59,000 human deaths per year (95% CI 25,000–159,000), more than 3.7 million disability-adjusted life years and 8.6 billion US dollars in annual losses. Dog vaccination accounted for only 1.5% of the costs and livestock losses for 6%. It concludes that dog vaccination is the most effective way to reduce the burden and that investment in it has been inadequate. Uncertainty is wide and several authors worked for vaccine manufacturers.
• Hampson et al. (2009), transmission study based on detailed tracing of rabid animals in Tanzania. The basic reproductive number of canine rabies was about 1.2 in the study area and below 2 throughout its historical range, which makes control by vaccination feasible. Rapid turnover of the dog population was the main obstacle, so regular campaigns are needed to maintain immunity between them. It is based on a single study area.
• Rattanavipapong et al. (2019), systematic review of 19 transmission modelling studies, carried out for the WHO vaccine advisory group. The basic reproductive number was below 2 in all studies but one; all concluded that dog vaccination reduces rabies in dogs or people, and most considered 70% annual coverage adequate. Coverage, dog density and birth rate were the decisive factors. These are model results, not trials.
• Tepsumethanon et al. (2004), cohort of 1,820 dogs and 332 cats that were ill or had bitten, observed for at least 10 days in Thailand. All 957 dogs and 94 cats with confirmed rabies died within 10 days of admission. It supports the 10-day observation of biting animals. It is a single centre and time was counted from admission, not from the bite.
• Soler-Tovar and Escobar (2025), analysis of Pan American Health Organization data on vampire-bat-transmitted bovine rabies from 1970 to 2023. There was an average of 450 outbreaks per year in countries where D. rotundus lives, with at least 6 animals dying per outbreak; Brazil, Colombia, Peru and Mexico were the most affected. Outbreaks were more frequent between 2000 and 2020 than in earlier decades, although smaller. These are passive surveillance data, subject to under-reporting.
• Oliveira et al. (2022), analysis of surveillance data on rabies in horses in Brazil from 2010 to 2019. 1,290 cases were reported, mainly in São Paulo (21.7%) and Mato Grosso (13.3%); Espírito Santo had the highest incidence, at 139.7 cases per 100,000 horses, and three high-risk clusters were found in the southeast and centre-west. There were cases in every region of the country and a rising trend in the northeast. These are passive surveillance data, subject to under-reporting.