Media centre

Marburg haemorrhagic fever

Fact sheet
November 2012


Key facts

  • The Marburg virus causes severe viral haemorrhagic fever in humans.
  • Case fatality rates in Marburg haemorrhagic fever outbreaks have ranged from 24% to 88%.
  • Rousettus aegypti, fruit bats of the Pteropodidae family, are considered to be natural hosts of Marburg virus. The Marburg virus is transmitted to people from fruit bats and spreads among humans through human-to-human transmission.
  • No specific antiviral treatment or vaccine is available.

Marburg virus is the causative agent of Marburg haemorrhagic fever (MHF), a disease with a case fatality ratio of up to 88%. Marburg haemorrhagic fever was initially detected in 1967 after simultaneous outbreaks in Marburg, from which the disease takes its name, and Frankfurt in Germany; and in Belgrade, Serbia.

Marburg and Ebola viruses are the two members of the Filoviridae family (filovirus). Though caused by different viruses, the two diseases are clinically similar. Both diseases are rare and have the capacity to cause dramatic outbreaks with high fatality rates.

Outbreaks

Two large outbreaks that occurred simultaneously in Marburg and Frankfurt in Germany, and in Belgrade, Serbia, in 1967, led to the initial recognition of the disease. The outbreak was associated with laboratory work using African green monkeys (Cercopithecus aethiops) imported from Uganda. Subsequently, outbreaks and sporadic cases have been reported in Angola, Democratic Republic of the Congo, Kenya, South Africa (in a person with recent travel history to Zimbabwe) and Uganda. In 2008, two independent cases were reported in travelers who visited a cave inhabited by Rousettus bat colonies in Uganda.

Transmission

Originally, human infection results from prolonged exposure to mines or caves inhabited by Rousettus bats colonies.

Transmission is mainly human-to-human, resulting from close contact with the blood, secretions, organs or other bodily fluids of infected persons. Burial ceremonies where mourners have direct contact with the body of the deceased can play a significant role in the transmission of Marburg. Transmission via infected semen can occur up to seven weeks after clinical recovery.

Transmission to health-care workers has been reported while treating Marburg patients, through close contact without the use of correct infection control precautions. Transmission via contaminated injection equipment or through needle-stick injuries is associated with more severe disease, rapid deterioration, and, possibly, a higher fatality rate.

Signs and symptoms

The incubation period (interval from infection to onset of symptoms) varies from 2 to 21 days.

Illness caused by Marburg virus begins abruptly, with high fever, severe headache and severe malaise. Muscle aches and pains are a common feature. Severe watery diarrhoea, abdominal pain and cramping, nausea and vomiting can begin on the third day. Diarrhoea can persist for a week. The appearance of patients at this phase has been described as showing “ghost-like” drawn features, deep-set eyes, expressionless faces, and extreme lethargy. In the 1967 European outbreak, non-itchy rash was a feature noted in most patients between 2 and 7 days after onset of symptoms.

Many patients develop severe haemorrhagic manifestations between 5 and 7 days, and fatal cases usually have some form of bleeding, often from multiple areas. Fresh blood in vomitus and faeces is often accompanied by bleeding from the nose, gums, and vagina. Spontaneous bleeding at venepuncture sites (where intravenous access is obtained to give fluids or obtain blood samples) can be particularly troublesome. During the severe phase of illness, patients have sustained high fever. Involvement of the central nervous system can result in confusion, irritability, and aggression. Orchitis has been reported occasionally in the late phase of disease (15 days).

In fatal cases, death occurs most often between 8 and 9 days after symptom onset, usually preceded by severe blood loss and shock.

Diagnosis

The differential diagnoses usually include malaria, typhoid fever, shigellosis, cholera, leptospirosis, plague, rickettsiosis, relapsing fever, meningitis, hepatitis and other viral haemorrhagic fevers.

Marburg virus infections can be diagnosed definitively only in laboratories, by a number of different tests:

  • enzyme-linked immunosorbent assay (ELISA);
  • antigen detection tests;
  • serum neutralization test;
  • reverse-transcriptase polymerase chain reaction (RT-PCR) assay; and
  • virus isolation by cell culture.

Tests on clinical samples present an extreme biohazard risk and are conducted only under maximum biological containment conditions.

Treatment and vaccine

Severe cases require intensive supportive care, as patients are frequently in need of intravenous fluids or oral rehydration with solutions containing electrolytes.

No specific treatment or vaccine is yet available for MHF. Several vaccine candidates are being tested but it could be several years before any are available. New drug therapies have shown promising results in laboratory studies and are currently being evaluated.

Natural hosts of Marburg virus

In Africa, the Old World fruit bats of the family Pteropodidae, particularly species belonging to the genus Rousettus aegyptiacus are considered natural hosts for Marburg virus. There is no apparent disease in the fruit bats. As a result, the geographic distribution of Marburg virus may overlap with the range of Rousettus bats.

Marburg virus in animals

African green monkeys (Cercopithecus aethiops) imported from Uganda were the source of infection for humans during the first Marburg outbreak.

Experimental inoculations in pigs with different Ebola viruses have been reported and show that pigs are susceptible to filovirus infection and shed the virus. Therefore pigs should be considered as a potential amplifier host during MHF outbreaks. Although no other domestic animals have yet been confirmed as having an association with filovirus outbreaks, as a precautionary measure they should be considered as potential amplifier hosts until proven otherwise.

Prevention

Precautionary measures for pig farms in endemic zones

Precautionary measures are needed in pig farms in Africa to avoid pigs becoming infected through contact with fruit bats. Such infection could potentially amplify the virus and cause or contribute to MHF outbreaks.

Reducing the risk of infection in people

In the absence of effective treatment and human vaccine, raising awareness of the risk factors for Marburg infection and the protective measures individuals can take to reduce human exposure to the virus, are the only ways to reduce human infections and deaths.

During MHF outbreaks in Africa, public health educational messages for risk reduction should focus on:

  • Reducing the risk of bat-to-human transmission arising from prolonged exposure to mines or caves inhabited by fruit bats colonies. During work or research activities or tourist visits in mines or caves inhabited by fruit bat colonies, people should wear gloves and other appropriate protective clothing (including masks).
  • Reducing the risk of human-to-human transmission in the community arising from direct or close contact with infected patients, particularly with their body fluids. Close physical contact with Marburg patients should be avoided. Gloves and appropriate personal protective equipment should be worn when taking care of ill patients at home. Regular hand washing should be performed after visiting sick relatives in hospital, as well as after taking care of ill patients at home.
  • Communities affected by Marburg should make efforts to ensure that the population is well informed, both about the nature of the disease itself and about necessary outbreak containment measures, including burial of the dead. People who have died from Marburg should be promptly and safely buried.

Controlling infection in health-care settings

Human-to-human transmission of Marburg virus is primarily associated with direct contact with blood and body fluids, and Marburg virus transmission associated with provision of health care has been reported when appropriate infection control measures have not been observed.

Health-care workers caring for patients with suspected or confirmed Marburg virus should apply infection control precautions to avoid any exposure to blood and body fluids and to direct unprotected contact with possibly contaminated environment. Therefore, provision of health care for suspected or confirmed Marburg patients requires specific control measures and reinforcement of standard precautions, particularly hand hygiene, use of personal protective equipment (PPE), safe injection practices, and safe burial practices.

Laboratory workers are also at risk. Samples taken from suspected human and animal Marburg cases for diagnosis should be handled by trained staff and processed in suitably equipped laboratories.

WHO response

WHO has been involved in all past Marburg outbreaks by providing expertise and documentation to support disease investigation and control.

Recommendations for infection control while providing care to patients with suspected or confirmed Marburg haemorrhagic fever is provided in the: Interim infection control recommendations for care of patients with suspected or confirmed filovirus (Ebola, Marburg) Haemorrhagic Fever, March 2008.

WHO has created an aide–memoire for standard precautions in health care. Standard precautions are meant to reduce the risk of transmission of bloodborne and other pathogens. If universally applied, the precautions would help prevent most transmission through exposure to blood and body fluids. Standard precautions are recommended in the care and treatment of all patients regardless of their perceived or confirmed infectious status.

They include the basic level of infection control and include hand hygiene, use of personal protective equipment to avoid direct contact with blood and body fluids, prevention of injuries from needle sticks and from other sharp instruments, and a set of environmental controls.

Table: Chronology of major Marburg Haemorrhagic Fever outbreaks.

Year Country Virus subtype Cases Deaths Case fatality rate
2008 Netherland Marburg 1 1 100%
(ex-Uganda)
2008 USA Marburg 1 0 0%
(ex-Uganda)
2007 Uganda Marburg 4 2 50%
2005 Angola Marburg 374 329 88%
1998 to 2000 DRC Marburg 154 128 83%
1987 Kenya Marburg 1 1 100%
1980 Kenya Marburg 2 1 50%
1975 South Africa Marburg 3 1 33%
1967 Yugoslavia Marburg 2 0 0%
1967 Germany Marburg 29 7 24%
Share

For more information contact:

WHO Media centre
Telephone: +41 22 791 2222
E-mail: mediainquiries@who.int