Nipah Virus: The Bat Pathogen That Scares Virologists

In 1998, pig farmers in Malaysia began falling ill with a brain infection that doctors first struggled to name. By late April 1999, the CDC counted 257 cases of encephalitis (inflammation of the brain) and 100 deaths. The pathogen was a previously unknown virus, later named after the village where it was first isolated. Nipah virus has no licensed vaccine or treatment, and the World Health Organization estimates it kills 40 to 75 percent of those infected. This article explains how it works, where it comes from, and why virologists watch it closely.

Educational content, not medical advice. Talk to your doctor before changing any medication, treatment, or routine. Information here reflects current research but every person's situation differs.

How Nipah was discovered: the Malaysian pig outbreak

The first recognized Nipah outbreak ran from September 1998 to May 1999 in the Malaysian states of Perak, Negeri Sembilan, and Selangor. By April 27, 1999, the CDC reported that 257 cases and 100 deaths had been recorded, and laboratory results from 65 of those who died suggested recent Nipah infection. The CDC's current overview puts the final toll at more than 300 infected and more than 100 dead.

Pigs turned out to be the key. In one analysis of 65 confirmed cases in Negeri Sembilan, 56 people (86 percent) said they had touched or handled pigs before getting sick, and 36 of those 56 had contact with pigs that looked ill. The CDC's contemporary report found no sign of human-to-human spread: nurses, physicians, and pathologists who cared for patients did not develop encephalitis or show evidence of recent infection.

The response was blunt. Authorities culled roughly a million pigs, most of them in the three affected states, banned the movement of pigs within the country, and set up national surveillance to find and cull infected herds. The infection also reached Singapore, where 11 abattoir workers who handled pigs imported from Malaysia fell ill between March 13 and March 19, 1999. Singapore stopped importing Malaysian pigs and closed its abattoirs on March 19, and the CDC reported no new cases there afterward. WHO notes that no new outbreaks have been reported in Singapore or Malaysia since 1999.

What kind of virus is it?

Nipah virus belongs to the genus Henipavirus in the family Paramyxoviridae, which also contains the viruses that cause measles and mumps. It is an enveloped virus with a genome made of RNA. For a primer on the basic cycle that Nipah follows, see our guide to how viruses work from infection to replication.

Two proteins on the surface of the virus do most of the work. Think of them as a key and a grappling hook. The attachment protein, called G, is the key: it binds to receptors called ephrin-B2 and ephrin-B3 on host cells. The fusion protein, F, is the hook: once G has docked, F drives the viral envelope to merge with the cell membrane so the genome can enter. The antibody m102.4, described below, works by blocking the interaction between G and those two receptors.

This receptor choice helps explain two features of the disease. Ephrin-B2 and ephrin-B3 are found in many mammals, which fits the virus's wide host range. They also appear on neurons and on the cells lining blood vessels, which fits a disease that attacks both the brain and the vasculature. Two genetic lineages are usually discussed: one associated with Malaysia (often written NiV-M) and one with Bangladesh and India (NiV-B). The Hendra virus, discovered in Australia in the 1990s, is a close relative that affects horses and people.

The natural reservoir: flying foxes

Nipah circulates in fruit bats of the genus Pteropus, known as flying foxes, which belong to the family Pteropodidae. The CDC says these bats live across South and Southeast Asia, the South Pacific, and Australia. WHO adds that African Eidolon bats carry antibodies against Nipah and Hendra viruses, which suggests the viruses, or close relatives, may also occur in Africa.

A reservoir host is a species that carries a pathogen long term, usually without serious illness. The bats are not known to suffer obvious disease from it, and they can shed the virus in saliva and urine as they feed and roost. The virus is rarely a problem for the bat. It becomes a problem when the bat's feeding area overlaps with human food or farming.

Field data from Kerala, India, show how common the virus can be in a bat colony. During the 2018 outbreak, investigators sampled 52 Pteropus medius bats and found that 13 (25 percent) had throat or rectal swabs positive for Nipah virus. Genome sequences from the human cases were 99.7 to 100 percent similar to the virus in the bats, and about 96 percent similar to a Bangladesh strain. That placed the Kerala virus in the same broad lineage as the Bangladesh strains, though in its own cluster. The pattern is what a recent jump from bats would look like, not a chain stretching back to Bangladesh.

For the general mechanics of how animal viruses cross into people, see our explainer on how viruses jump from animals to humans.

Three routes from bat to human

Nipah has reached people by more than one road. This is part of what makes it hard to contain.

Route	Where	Mechanism
Through pigs	Malaysia and Singapore, 1998 to 1999	Pigs infected from bat-contaminated fruit and environment; people infected by handling pigs
Raw date palm sap	Bangladesh and parts of India	Bats feed on sap collected in open containers; people drink it raw
Bat-contaminated fruit or direct contact	Kerala and elsewhere	Exact source often unconfirmed; fruit, bat roosts, and wells have been investigated

In Bangladesh, WHO advises keeping bats away from date palm sap collection and boiling raw sap, because bats can contaminate sap as it is collected. Sap is not the only route; WHO also advises washing and peeling fruit and discarding any with bat bites.

In Kerala the source of infection is often not found. WHO said of the four cases reported in 2025 that the source was under investigation and that the cases did not appear to be linked, which suggests separate spillovers from the reservoir rather than a single chain.

Person-to-person spread: Faridpur and the hospital wards

Investigators found no clear person-to-person spread in the Malaysian outbreak, but Bangladesh did show it. In 2004, an outbreak in Faridpur District involved 36 patients (23 laboratory-confirmed and 13 probable), of whom 27 died, a fatality rate of 75 percent. The investigation published in Emerging Infectious Diseases in 2007 found that 33 of the 36 had had close contact with another ill person before they became sick.

One patient, a local religious leader, was at the center. Many family members and followers had close contact with him during his illness, and he was linked to 22 subsequent infections, the most of any case. Eight of ten household members became infected. Contact with him carried a strongly increased risk (odds ratio of 5.6 after adjustment), and handwashing after contact was associated with lower risk. Viral RNA was even found on a soiled hospital wall and bed frame used by a confirmed patient about five weeks earlier, and a patient who shared a bed with a confirmed case became infected.

Hospital settings recur in later outbreaks. In the 2023 Kerala outbreak, WHO reported that two of the other five cases were family members and the rest were hospital contacts, a pattern of nosocomial transmission, meaning spread within a hospital. In January 2026, two nurses in a private hospital in Barasat, West Bengal, tested positive; more than 190 contacts were tested and all were negative.

The takeaway is that Nipah does not spread through the air like influenza. It passes through close contact with infected bodily fluids, most often in households and clinics where people care for a seriously ill patient without protective equipment. That limits the speed of spread, which is the main reason outbreaks have stayed small.

The outbreak record

The table collects confirmed numbers from WHO and the studies cited above. Totals differ by source and cut-off date, and outbreak counts depend on how many mild or asymptomatic cases are found.

Outbreak	Cases	Deaths	Notes
Malaysia, 1998 to 1999	257 (as of April 27, 1999)	100	Pig-linked; no documented person-to-person spread
Singapore, 1999	11	Not stated in the CDC update	Abattoir workers handling imported pigs
Faridpur, Bangladesh, 2004	36	27	Person-to-person transmission documented
Kerala, India, 2018	19 (18 confirmed)	17 (about 89 percent)	First Kerala outbreak; WHO count as of July 2018 (some papers report 23 cases and 21 deaths by counting probable cases)
Kerala, 2019	1	0	Survived
Kerala, 2021	1	1
Kerala, 2023	6	2	Hospital contacts among the cases
Kerala, 2025	4 confirmed	2	First outbreak in Palakkad District
West Bengal, India, 2026	2 confirmed	Not stated	Two nurses; over 190 contacts tested negative

Over the five years before January 2026, India had about a dozen confirmed cases, all in Kerala.

Two cautions apply when reading these numbers. First, fatality rates look higher in small outbreaks partly because surveillance catches the sickest patients; mild or asymptomatic infections go unnoticed. Evidence that such infections exist comes from a study of contacts of 18 infected patients in India, in which 3 of 279 contacts were seropositive (they had antibodies) without symptoms. Second, fatality swings widely between small outbreaks, from 0 percent in Kerala in 2019 to 100 percent in 2021, so a single average hides large differences.

What the disease does

Incubation, the time between exposure and symptoms, is usually 3 to 14 days according to WHO, with rare reports of up to 45 days. The CDC gives 4 to 14 days. Early symptoms are fever, headache, cough, sore throat, and difficulty breathing. Some patients go on to confusion, drowsiness, and seizures, and some fall into a coma within 24 to 48 hours.

Nipah is notable for hitting the brain and, in some patients, the lungs at the same time. Severe disease involves encephalitis, and death is common in that group. Not everyone is severely affected, and some infections produce no symptoms.

Survival does not always mean full recovery. WHO states that long-term neurological conditions occur in about one in five survivors. A rarer and stranger pattern is relapsing or late-onset encephalitis, in which a person who recovered, or who never felt ill, develops encephalitis months or years later. A 2020 letter in Emerging Infectious Diseases by Malaysian neurologists Chong Tin Tan and Kum Thong Wong said the death rate was about 20 percent during the course of late-onset or relapsing encephalitis, and that up to 16 percent of asymptomatic, seropositive people had small lesions on brain MRI. Whether those people go on to develop disease is unknown.

Why this happens is not fully understood. One reading is that the virus stays hidden in the nervous system and flares again. Another is that an immune response to leftover virus does damage. Compare this with the best-known case of a virus hiding in nerve tissue, the rabies virus, which behaves very differently, or the family of herpesviruses that sit dormant for life. Neither is a clean model, and the question remains open.

Why Nipah is a WHO priority pathogen

WHO has named Nipah as a priority disease under its R&D Blueprint, which directs research money toward pathogens with epidemic potential and weak countermeasures. Four features explain the listing.

1. High lethality. A fatality rate of 40 to 75 percent, although partly shaped by who gets detected, is far above that of seasonal flu or the seasonal coronaviruses.
2. Person-to-person transmission. It is documented, though inefficient.
3. No licensed vaccine or drug. Care is supportive only.
4. A wide reservoir. The bats that carry it span several continents.

For comparison with other high-fatality viruses, see our profiles of Ebola virus and Marburg virus, and our overview of the deadliest viruses and what makes them dangerous.

What keeps Nipah from being a pandemic virus today is how it spreads. Documented chains of transmission have been short, and the virus has not shown efficient spread through the air. The concern is that the picture could change: a mutation that improves transmission, an outbreak in a crowded city with weak infection control, or a spillover that reaches many people at once. These are possibilities that surveillance exists to catch, not predictions. The same logic applies to other threats we cover, such as H5N1 pandemic risk, and to the way machine learning is being used for viral evolution prediction.

Treatment, antibodies, and vaccines

No specific treatment is approved. WHO says early supportive care, which means oxygen, organ support, rehydration, and monitoring, improves survival. Several candidate products exist.

Ribavirin. An older antiviral. In the Malaysian outbreak, an open-label study of 140 treated patients compared with 54 untreated controls associated treatment with a 36 percent reduction in mortality. Open-label studies without randomization are weak evidence, and the review notes its efficacy remains unclear.

m102.4. A human monoclonal antibody, meaning a lab-made antibody, that targets the G protein and blocks its binding to ephrin-B2 and ephrin-B3. In ferrets, a single dose given 10 hours after infection fully protected animals. In African green monkeys it protected when started up to 5 days after infection with the Malaysia-type virus, but only up to 3 days with the Bangladesh-type virus. In a phase 1 trial in 30 healthy adults the doses were safe and well tolerated. It has been given 14 times as compassionate therapy after high-risk exposures. Because those uses were not controlled trials, they cannot show that the drug prevented illness.

Remdesivir. In African green monkeys given a lethal dose of the Bangladesh-type virus, daily intravenous remdesivir starting 24 hours after infection for 12 days gave 100 percent survival. That is an animal result.

Vaccines. The leading vaccine candidates use different platforms.

Candidate	Developer	Status reported
ChAdOx1 NipahB	University of Oxford	In clinical trials; Oxford announced a phase II trial in Bangladesh in December 2025
HeV-sG subunit vaccine	Based on a Hendra virus glycoprotein	Reported in phase I trials as of a 2021 review

The Oxford vaccine uses a harmless adenovirus to carry the gene for a Nipah protein into cells, a design used in other vaccines. For the basics of how any vaccine trains the body, see our guide to how vaccines work, and for the mRNA approach, mRNA vaccines.

The central obstacle is how to prove a vaccine works. Outbreaks are small, scattered, and unpredictable, so the large efficacy trials used for common diseases are hard to run. Regulators and funders are therefore weighing alternatives, such as relying on antibody levels in healthy volunteers, or animal-based evidence when human trials are not feasible. A trial in Bangladesh, where outbreaks have been reported almost every year since 2001 according to WHO, is a step toward that evidence.

Prevention in affected regions

Prevention is aimed at the places where spillover and hospital spread occur. WHO's recommendations include keeping bats away from date palm sap collection, boiling fresh sap, washing and peeling fruit, discarding fruit with bat bites, and using protective gear around sick animals. In health facilities, WHO recommends single-patient rooms with contact, droplet, and airborne precautions as appropriate. The CDC adds regular handwashing and avoiding contact with the blood and bodily fluids of someone with Nipah. These are public health measures for outbreak settings; for personal health questions, see a clinician.

Kerala shows how much response speed matters. Its repeated outbreaks have been contained with contact tracing and isolation. After the 2025 cases, authorities had identified 723 contacts across five districts by July 17, per WHO. That scale of tracing for four cases is why a few infections do not become hundreds.

Common misconceptions

"Nipah spreads like COVID-19." It does not spread through the air in the way seasonal respiratory viruses do. Spread between people has been documented through close contact with an ill person, mostly in households and hospitals.

"The fatality rate is exactly 75 percent." The 40 to 75 percent range is an estimate that varies with the outbreak, the strain, the quality of care, and how many mild cases are detected. Kerala's fatality rates have ranged from 0 percent in 2019 to 100 percent in 2021, in outbreaks of a single person.

"There is a vaccine in the clinic." None is approved. Candidates are in clinical trials.

"Pigs are the main source." Pigs mattered in the 1998 to 1999 outbreak. In Bangladesh and India, date palm sap and other bat-contaminated food are the more common routes.

"Once you survive, you are in the clear." Many survivors recover well, but WHO reports long-term neurological conditions in about one in five, and relapsing encephalitis has been described months to years later.

Frequently asked questions

Q: How does Nipah virus spread from person to person? A: Through close contact with the bodily fluids of an infected person, such as saliva, respiratory secretions, urine, and blood. It is not known to spread through the air over long distances. Family members and health workers caring for patients without protection have been most affected, as in the 2004 Faridpur outbreak, where 33 of 36 cases had close contact with another ill person.

Q: Why are fatality rates so different between outbreaks? A: Several factors likely contribute: how many mild cases are detected, the strain, access to intensive care, and chance in small outbreaks. Malaysia's outbreak had roughly a 39 percent fatality rate (100 of 257 cases as of April 1999), the 2004 Faridpur outbreak in Bangladesh 75 percent (27 of 36), and Kerala's 2018 outbreak about 89 percent (17 of 19, per WHO). Direct comparison is risky because detection methods differ.

Q: Can you get Nipah from fruit bats have touched? A: Yes, it is considered possible. WHO advises washing and peeling fruit and discarding any fruit with bat bites. Raw date palm sap is the best-documented food route.

Q: Is there an approved treatment? A: No. Supportive care is the standard. Ribavirin, remdesivir, and the antibody m102.4 have been studied, but ribavirin results are inconclusive, remdesivir data are from animals, and m102.4 has been given under compassionate use but is not approved.

Q: Why does Nipah matter if outbreaks are so small? A: Small outbreaks can still be deadly, strain health systems, and signal the chance of a larger one. WHO lists it as a priority disease because it combines high lethality, person-to-person spread, no licensed countermeasures, and a wide animal reservoir. The record to date is of repeated, contained events.

Q: What are the long-term effects? A: WHO says about one in five survivors develops long-term neurological conditions. Relapsing or late-onset encephalitis occurs in a minority, and Malaysian neurologists put the death rate during such episodes at about 20 percent. Estimates come mostly from the Malaysian outbreak and may not apply elsewhere.

Q: How close are we to a vaccine? A: Oxford's ChAdOx1 NipahB is among the leading candidates, with a phase II trial in Bangladesh announced in December 2025. WHO says no vaccine is approved, and no Nipah vaccine has completed the trials needed for licensing.

The takeaway

Nipah virus is a case study in how a dangerous pathogen can stay small. It is lethal, it comes from animals that are widespread, and it has found several ways into the human food chain and the hospital ward. What has limited it so far is mostly how it spreads: through close contact, in chains that have burned out within a few generations. That is not a guarantee. The lesson of Malaysia, Bangladesh, and Kerala is that the response, from culling in 1999 to contact tracing of 723 people in 2025, has done the work that vaccines and drugs have not yet been able to do. The countermeasure pipeline, led by the Oxford vaccine trial in Bangladesh, is the part of the story most likely to change in the next few years.

Sources

• World Health Organization. Nipah virus: fact sheet, updated January 29, 2026.
• Centers for Disease Control and Prevention. About Nipah virus, updated February 13, 2026.
• CDC. Update: Outbreak of Nipah virus, Malaysia and Singapore, 1999. MMWR.
• Gurley ES, et al. Person-to-person transmission of Nipah virus in a Bangladeshi community. Emerging Infectious Diseases, 2007.
• Yadav PD, et al. Nipah virus outbreak in Kerala State, India, 2018. Emerging Infectious Diseases, 2019.
• WHO. Disease Outbreak News: Nipah virus, India, August 7, 2018.
• Tan CT, Wong KT. Infections among contacts of patients with Nipah virus, India. Emerging Infectious Diseases, 2020.
• WHO. Disease Outbreak News: Nipah virus, India, August 6, 2025.
• WHO. Disease Outbreak News: Nipah virus, India, January 30, 2026.
• Review of Nipah therapeutics, including m102.4, ribavirin, and remdesivir, on PubMed Central.
• University of Oxford. World's first Phase II Nipah virus vaccine trial launch, December 15, 2025.

---

Read next:

• BK Virus, JC Virus & Polyomaviruses Explained
• Chikungunya Virus Explained
• Dengue Virus Explained — Why It's Spreading and How It Works