Viral Hemorrhagic Fever (Crimean-Congo, Ebola, Lassa, Marburg, Rift Valley, Yellow Fever)

Monograph
Tables
What's New
Reviews
History

VIRAL HEMORRHAGIC FEVER

Some 15 different RNA viruses from 4 families – the Arenaviridae, Bunyaviridae, Filoviridae and Flaviviridae – cause a similar illness in humans that is designated viral hemorrhagic fever (VHF). The syndrome is characterized by the sudden onset of fever, malaise, body aches and a variety of other nonspecific symptoms, which are followed over a period of days by the development of coagulation defects that can result in bleeding and an increase in vascular permeability that can lead to a fall in blood pressure, shock and death. The term VHF was first used in the early 1950s to designate an illness that occurred sporadically among soldiers fighting in the Korean War. That disease, now called hemorrhagic fever with renal syndrome, results from exposure to the excretions of hantavirus-infected rodents. Over subsequent decades, a number of additional viral infections, ranging from the ancient plague of yellow fever to the newly discovered New World disease of hantavirus pulmonary syndrome, have been found to fit the criteria for VHF.

This section focuses on the diseases caused by five different viruses from this group. The two filoviruses, Marburg and Ebola, produce the most severe form of VHF, with case fatality rates in large outbreaks in central Africa reaching 90%. Lassa virus, a member of the arenavirus family, is responsible for the third illness, Lassa fever, which is less severe than filoviral hemorrhagic fever, but much more common, as sporadic cases occur year-round in Sierra Leone and the surrounding region of West Africa. The last two diseases are caused by bunyaviruses: Crimean-Congo HF (CCHF), a tick-borne infection that occurs across a huge geographic area, from western China to the Balkans and southwards to the tip of Africa, and Rift Valley fever (RVF), which generally takes the form of sporadic human infections in east Africa, but can also occur in huge epidemics when heavy rainfall favors the proliferation of the mosquito vector.

Virology

All hemorrhagic fever viruses have single-stranded RNA genomes, which differ in polarity and in the number of segments (flaviviruses: positive-sense, 1 segment; filoviruses: negative-sense, 1 segment; arenaviruses: negative-sense, 2 segments; bunyaviruses: negative-sense, 3 segments). The agents also differ in virion morphology, with the arena-, bunya- and flaviviruses being spherical, while the filoviruses are long and filamentous. All hemorrhagic fever viruses have a lipid envelope derived from the host cell membrane in which their surface glycoprotein molecules are anchored.

Epidemiology

Nearly all types of VHF are zoonoses – that is, their causative agents are maintained through cycles of infection in wild animals, but are occasionally transferred to humans through direct physical contact with virus-containing blood or tissues, exposure to aerosolized excretions or the bite of a blood-feeding arthropod. (The two hemorrhagic fever viruses that can be maintained through person-to-person transmission are dengue, which lacks an animal reservoir and is transmitted among humans by mosquitoes, and yellow fever virus, which is maintained in wild primates, but is also carried from person to person by mosquitoes.) As is true of other zoonotic pathogens, the geographic distribution of each type of VHF is determined by the range of its reservoir host. Because zoonotic agents do not require human infection for their continued existence, they have not evolved efficient mechanisms of person-to-person transmission, and human illness therefore occurs either as single cases, with little or no secondary spread, or as hospital-associated outbreaks, when the lack of proper infection control measures permit health care workers and family caregivers to come into direct contact with the virus-containing body fluids of VHF patients.

Clinical Manifestations

VHF usually begins with the abrupt onset of fever and malaise, which are quickly followed by such nonspecific signs and symptoms as headache, muscle and joint pain, nausea, vomiting, abdominal pain, diarrhea – physical changes that reflect rising levels of proinflammatory cytokines and chemokines in the bloodstream. Host mediators also cause the vasodilatation and diffuse “vascular leak” that are the major physiologic abnormalities of VHF, which sometimes produce cutaneous flushing and soft-tissue edema, but are manifested most clearly through a fall in blood pressure. At the same time, systemic inflammation induces coagulation abnormalities, which may cause the development of a maculopapular rash, prolonged bleeding from needle puncture sites, conjunctival hemorrhages and easy bruising.

Despite the syndrome’s name, bleeding is rarely the cause of death in VHF; instead, the life-threatening lesion is a profound alteration of vascular function and loss of circulating plasma volume induced by high circulating levels of proinflammatory mediators. Major hemorrhage from the gastrointestinal and urinary tracts is generally seen only in moribund patients, as a terminal phenomenon. Severely ill patients develop intractable shock and may die during the second week of illness, while more fortunate individuals make a gradual recovery. Most hemorrhagic fever viruses do not invade the central nervous system, so changes in neurologic function usually result from a decrease in circulatory perfusion and resolve as the patient recovers.

In addition to resembling each other in their clinical signs and symptoms, the various types of VHF are characterized by similar changes in clinical laboratory tests, which relate to underlying physiologic changes. Thus, increased vascular permeability results in diminished plasma volume, which is reflected in turn in the complete blood count (rising hemoglobin concentration and hematocrit) and by a fall in albumin concentration, while the total protein generally remains normal. The drop in circulating blood volume causes impaired renal function, manifested by rising urea nitrogen and creatinine levels. Chemokines produced by infected macrophages evoke the release of leukocytes from the bone marrow, resulting in an increased percentage of immature granulocytes in blood smears, while the lymphocyte count may fall as a result of the loss of these cells through apoptosis over the course of infection.

The shift to a procoagulant state that accompanies systemic inflammation also leads to a consumption of coagulation factors, which will eventually be manifested in a prolongation of the prothrombin and partial thromboplastin times and a fall in the platelet count. Fibrin split products and D-dimers appear in the plasma as coagulopathy progresses to disseminated intravascular coagulation. In nearly all types of VHF, infection spreads to hepatic parenchymal cells, as reflected by the appearance of elevated levels of liver–associated enzymes, such as aspartate and alanine aminotransferase, in the plasma. In yellow fever and CCHF, severe hepatic involvement can lead to jaundice and deficits in coagulation factors.

All types of VHF are associated with a significant case-fatality rate, but they differ in the severity of the average case. For example, many cases of Crimean-Congo hemorrhagic fever, Lassa or yellow fever are mild and go unrecognized, but Ebola and Marburg hemorrhagic fever is almost always a severe disease, in which the case fatality rate ranges from 50-90%. Because the various types of VHF resemble each other in their clinical presentation and in the results of standard laboratory tests, specialized assays are required to make a specific diagnosis.

Laboratory Diagnosis

Rapid diagnostic tests for most types of VHF are based on the detection of viral antigens by enzyme-linked immunosorbent assay (ELISA) or of specific RNA sequences by reverse-transcription polymerase chain reaction (RT-PCR) in the blood or other body fluids. Because symptomatic patients are always viremic, they will test positive by such assays. However, it is not known whether these tests are sensitive enough to detect individuals who are still incubating the virus. In the United States, ELISA or RT-PCR testing for VHF can be performed by clinical laboratories that form part of the Laboratory Response Network. Obtaining a final confirmatory diagnosis relies on replication of the causative agent in cell culture in a BSL-4 containment laboratory and visualization of the characteristic viral particles by electron microscopy. Such tests are currently performed only by the Centers for Disease Control and Prevention (CDC) in Atlanta, GA and the US Army Medical Research Institute of Infectious Diseases (USAMRIID) in Frederick, MD.

Pathogenesis

The virus-infected macrophage plays the central role in the pathogenesis of VHF. All microbes that enter the body are confronted by these ubiquitous “sentinel” cells, but the hemorrhagic fever viruses are able to replicate prolifically within them, engendering large numbers of progeny virions and causing the release of proinflammatory cytokines, chemokines and vasoactive substances such as nitric oxide. Proteins encoded by hemorrhagic fever viruses are able to suppress human type I interferon responses, permitting their rapid dissemination throughout the body. As inflammatory mediators accumulate in the bloodstream, the diffuse dilatation of small vessels and increased permeability of their endothelial linings permit fluid and proteins to move out of the plasma into the interstitial space, causing a reduction in effective blood volume that produces circulatory insufficiency. and tissue factor, that in turn cause vasodilatation, increased vascular permeability and disseminated intravascular coagulation.

In addition to the damaging physiologic effects of a massive systemic inflammatory response, hemorrhagic fever viruses can cause direct tissue injury by infecting and destroying a variety of parenchymal cells. For most agents, hepatic parenchymal cells are the principal target, but the filoviruses are able to infect virtually all cell types, causing massive tissue damage.

The severity and high case fatality rate of VHF is also explained by a failure of adaptive immune responses, through the impaired function of virus-infected dendritic cells and the massive loss of lymphocytes through “bystander” apoptosis. In fatal cases, a terminal state of “immune paralysis” may develop resembling that seen in septic shock.

Antiviral Therapy

No antiviral drug is licensed for the treatment of VHF. The guanosine analog ribavirin (Virazole®), which is in clinical use for the treatment of respiratory syncytial virus infection in infants and in combination with interferon for the treatment of hepatitis C, inhibits the replication of a number of bunyaviruses and arenaviruses, and has been used to treat Lassa fever and CCHF. The drug acts through several pathways: blocking the host enzyme inosine monophosphate dehydrogenase, to decrease the size of intracellular guanosine di- and triphosphate pools; inhibiting 5' cap formation on viral messenger RNA; and interfering with RNA-dependent RNA polymerase activity. Because of these multiple mechanisms of action, drug-resistant viruses rarely emerge during the course of therapy. Ribavirin produces a reversible hemolytic anemia, and evidence of teratogenicity prevents its use in pregnant women. The drug’s inability to block the replication of some hemorrhagic fever agents, such as the filoviruses, remains unexplained.

Studies in nonhuman primates suggest that, in addition to the treatment of Lassa fever, ribavirin would also be of value for the therapy of New World arenavirus infections, and apparent benefit has been seen in limited evaluation against Argentine hemorrhagic fever. As is generally true of antiviral therapy, treatment is most effective when begun early in infection. Some apparent treatment failures, such as a recent open-label clinical trial of therapy of hantavirus pulmonary syndrome, actually reflect the current inability to diagnose disease early enough for antiviral therapy to be of benefit. Although active against yellow fever in a rodent model, ribavirin had no impact on infection in macaques, and also shows no activity against filoviruses.

Adjunctive Therapy

Human convalescent plasma has been widely used for the postexposure prophylaxis or treatment of some types of VHF, but only in the case of Argentine hemorrhagic fever has a controlled clinical trial provided evidence of efficacy. High-titer convalescent plasma was effective in treating laboratory primates for Lassa fever, but the results of limited trials in humans were not encouraging. Immune globulin has also been used as postexposure prophylaxis and treatment of CCHF, but no placebo-controlled trial has been reported.

INDIVIDUAL DISEASES

Ebola and Marburg Hemorrhagic Fever

Virology

The genus name of the filoviruses reflects their unique filamentous structure (the Latin word filum means thread). The two species, Marburg and Ebola, are identical in morphology. Virions are always 80 nm in diameter, but vary considerably in length. The genome consists of a single 19-kilobase strand of negative-sense RNA containing 7 genes. Transcription and genome replication are performed by a complex of the RNA-dependent RNA polymerase and three other virus-encoded proteins.

Marburg and Ebola virus replicate in identical fashion, except that the primary product of the glycoprotein gene of all Ebola species is a C-terminally truncated protein (“sGP”) that is secreted from infected cells. Production of full-length membrane-bound GP requires the post-transcriptional addition of a nucleotide to its messenger RNA during transcription. The release of sGP may contribute in some way to viral persistence in the filoviral maintenance host, but no role in the pathogenesis of Ebola hemorrhagic fever in humans has been identified.

Epidemiology

The filoviruses are the only agents of viral hemorrhagic fever for which no natural reservoir has been identified. As described below, virus has been recovered from a variety of wild primates, but because these animals die quickly from infection, they cannot serve as a natural reservoir for Ebola virus. A number of studies have provided evidence that fruit bats, widely distributed in central Africa, may be the reservoir host of both Ebola and Marburg virus. A number of captured bats have demonstrated anti-Ebola or -Marburg antibodies in serum and the presence of viral sequences in liver and spleen tissues by RT-PCR, but live virus has yet to be isolated from any of these animals.

Ebola or Marburg hemorrhagic fever of humans has occurred in 4 ways: exposure to an infected nonhuman primate, either in the wild or in captivity; presumed exposure to the unidentified reservoir host in Africa; contamination by virus-containing body fluids of patients; or as a result of a laboratory accident. Marburg virus was discovered when infected monkeys were inadvertently imported from Uganda to vaccine production facilities in Marburg, Germany and Belgrade, Yugoslavia in 1967, causing outbreaks of hemorrhagic fever that killed 7 of 31 infected persons.

Since the first outbreak, Marburg hemorrhagic fever has only occurred in Africa. Up until 2000, only 6 additional cases were identified, but in that year a large epidemic came to light in Watsa, in northeastern Democratic Republic of Congo (DRC, the former Zaire). More than 150 cases were eventually confirmed, with a fatality rate of greater than 80%. Genetic analysis of viral isolates showed that the epidemic had resulted from multiple independent introductions of virus into the local community of gold miners, and that little secondary spread of infection had taken place. In 2005, Marburg virus appeared unexpectedly in Angola, far from any previous site of human infection, causing some 250 cases of illness with a mortality that approached 90%. The spread of infection in a hospital pediatrics ward played a major role in amplifying the epidemic. In contrast to Watsa, only a single strain of virus was isolated.

Ebola virus first came to the attention of the world scientific community in 1976, when two large epidemics occurred almost simultaneously in Zaire and Sudan. The former was centered at a mission hospital, where virus was inadvertently transmitted by contaminated syringes to nearly 100 patients, all of whom developed fatal disease. Many of the doctors, nurses and family members who cared for the first wave of patients also became infected; the final death toll was 280 out of 318 cases. The source of the initial infection was never identified. Another large hospital-based outbreak caused by the Zaire species occurred in Kikwit, DRC in 1995, with a case fatality rate exceeding 80% among more than 300 patients. In that and other large filoviral hemorrhagic fever epidemics since that time, teams of international medical workers helped to bring the outbreak to an end through case identification, contact tracing and the establishment of an isolation ward. The 1976 Sudan outbreak also involved extensive secondary transmission of virus in the hospital setting. In contrast to epidemics caused by Ebola Zaire virus, the case fatality rate for that epidemic and for two more in 1979 and in 2000 were all approximately 50%. The 2000 outbreak, which occurred in Gulu, Uganda, was the largest yet recorded, with 425 patients.

Beginning in 1996, a new epidemiologic pattern was observed, in which humans became infected with Ebola Zaire virus through contact with sick or dead chimps or gorillas. In the first known outbreak of this type, 19 Gabonese villagers who butchered and ate a chimp found dead in the forest developed fulminant hemorrhagic fever, and most of them died. The virus is now spreading among populations of great apes in Gabon and neighboring Republic of Congo, causing both a massive die-off of these animals and a repeated epidemics among local residents. A third species of Ebola virus was identified in the Ivory Coast, where it infected a researcher who performed a necropsy on a dead chimpanzee. No other human cases have been recognized.

The fourth species of Ebola virus, the enigmatic Reston agent, made its world debut in 1989, when it caused an outbreak among captive primates recently imported from the Philipines to a monkey quarantine facility in suburban Virginia. The origin of the virus and its relationship to the 3 African Ebola species still have not been determined. None of the animal caretakers who were exposed to sick animals became ill, prompting some researchers to propose that Ebola Reston virus is avirulent for humans.

Clinical Manifestations

The fulminant illess caused by Ebola or Marburg virus displays all of the classic features of the hemorrhagic fever syndrome. Exposure to the agent is usually followed within 3-7 days by the abrupt onset of high fever, malaise and a variety of nonspecific signs and symptoms produced by high levels of circulating cytokines and chemokines. These early changes are soon followed by the development of a “vascular leak” (see below) that results in a fall in blood pressure, leading in almost all cases to intractable shock and death during the second week of illness. All patients show evidence of coagulation abnormalities, with hemorrhages in the conjunctiva and easy bruising. Bleeding from the gastrointestinal and genitourinary tracts is frequently seen during the terminal phase of illness, but is rarely the cause of death. Survival appears to require the rapid mobilization of a specific immune response: those patients who develop anti-Ebola IgM or IgG antibodies during the second week of illness are likely to recover, while the persistence of high levels of circulating virus in the absence of a detectable antibody response is predictive of death.

Laboratory Diagnosis

Because the clinical features of an isolated case of Ebola or Marburg hemorrhagic fever do not differ from those of other types of VHF or a number of other infectious diseases present in central Africa, specific diagnosis requires the detection of viral antigen by ELISA or viral RNA by RT-PCR. Saliva from sick persons is usually positive by both tests, but serum is a much more reliable sample early in the course of illness. Confirmatory testing of Ebola or Marburg hemorrhagic fever can be performed in Africa, at laboratories in Libreville, Gabon or in South Africa, or in the USA, at the CDC or USAMRIID. The same tests can also be performed in small mobile containment labs set up by international teams during outbreaks.

Pathogenesis

Infection of a laboratory primate with a few particles of Ebola or Marburg virus by injection, inhalation or placement in the mouth or eye initiates an inexorable “chain reaction,” in which replication in monocytes, macrophages and dendritic cells releases large numbers of new virions that spread rapidly to the same cell types in lymph nodes, liver and spleen and other tissues throughout the body. From those cells, infection then spreads further to infect hepatocytes and other parenchymal cells; only lymphocytes and neurons are known to remain free of infection. Infected cells undergo death through necrosis, causing massive tissue damage. At the same time, cytokines and chemokines released from infected macrophages cause vasodilatation and increased permeability, and cell-surface tissue factor triggers disseminated intravascular coagulation. The host immune response is markedly impaired by the destruction of virus-infected dendritic cells and a massive loss of lymphocytes through apoptosis.

Susceptibility to Antivral Drugs In Vitro and In Laboratory Animals

No licensed antiviral drug inhibits filovirus replication, either in vitro or in laboratory animals. For unknown reasons, ribavirin, which is quite active against many other negative-sense RNA viruses, has no effect on the these agents. Experimental therapies for Ebola and Marburg hemorrhagic fever can be divided into two types: those that directly block filoviral replication, and those that modify host responses to infection. The former are represented by antisense oligonucleotides and siRNA targeting sequences in viral mRNA, especially in the nucleoprotein and polymerase genes. Both approaches have been effective in models of Ebola hemorrhagic fever in mice, guinea pigs and nonhuman primates, providing treatment is begun either before or soon after virus challenge. If such treatment proves to lack toxicity, it could be very valuable for treating researchers, doctors or nurses who are accidentally exposed to a filovirus in the laboratory or while responding to a disease outbreak.

Several other types of therapy have also proven effective in nonhuman primates when begun soon after virus challenge. A recombinant nematode protein, rNAPc2, which inhibits the initiation of the extrinsic coagulation pathway by blocking the interaction between tissue factor and factor VIIa, and is undergoing clinical evaluation for a number of medical conditions, was surprisingly potent in the otherwise uniformly lethal model of Ebola Zaire virus infection in cynomolgus macaques. Single daily injections, begun the day of or the day after infection, completely prevented illness in 3 of 9 treated animals and prolonged survival in the others. As well as reducing signs of coagulopathy, treatment significantly reduced both inflammation and levels of circulating virus. Recombinant activated protein C, a licensed product that has shown benefit in the treatment of septic shock, also reduced morbidity and mortality in Ebola virus-infected macaques when provided as a continuous intravenous infusion. In addition to pointing the way to future therapies, these findings provide experimental confirmation that the severe morbidity and mortality of VHF result in large part from host responses to infection.

Antiviral Therapy

There is currently no approved therapy for filoviral hemorrhagic fever; treatment is supportive, based on maintenance of an adequate circulating blood volume until the immune system clears the infection. Experimental approaches that show promise for human use are described above.

Vaccines

A number of vaccines have protected mice, guinea pigs and laboratory primates against otherwise lethal challenge with Ebola or Marburg virus. These include a “prime-boost” approach, consisting of a series of injections of a DNA vaccine followed by a recombinant adenovirus, both encoding the viral surface glycoprotein (GP); a more rapid approach employing only the adenovirus vaccine; a chimeric vesicular stomatitis virus (VSV) in which the native surface GP has been replaced by that of Ebola or Marburg virus; and noninfectious Ebola or Marburg virus-like particles that present the GP and other antigens in an immunogenic fashion. These vaccines are typically administered in one or more doses beginning several weeks before virus challenge. The VSV vaccine has also protected laboratory primates when administered soon after inoculation of a lethal dose of Ebola or Marburg virus, apparently because it induces the rapid development of adaptive immune responses that otherwise fail to occur in filoviral hemorrhagic fever. The DNA and adenovirus vaccines are currently undergoing Phase I safety testing in humans.

Lassa Fever

Virology

The arenaviruses are spherical, lipid-enveloped viruses that frequently entrap host ribosomes during their formation, giving them a “sandy” appearance on electron microscopy (the Latin word arena means “sand”). The genome consists of two ambisense RNA molecules, in which proteins are encoded in both directions on the same strand. The large (L) segment encodes the RNA-dependent RNA polymerase and a Z protein that is thought to link the nucleocapsid to the lipid membrane. The small (S) strand encodes a nucleoprotein, that encapsidates the genomic RNA, and two surface glycoproteins, G1 and G2.

Epidemiology

The arenaviruses that cause hemorrhagic fever are divided into two groups: an “Old World” agent, Lassa virus, found in west Africa, and a number of “New World” viruses that cause severe disease in South America and the southwestern United States. Each agent is named after the location where it was first isolated. All of them are maintained in rodents, among which they spread through direct contact or excretions; none is transmitted by an arthropod vector. Humans become infected when they handle, butcher or eat the maintenance host or are exposed to its aerosolized urine or feces.

Lassa is by far the most common type of arenavirus hemorrhagic fever. Serosurveys have suggested that as many as 100,000 people become infected each year in west Africa, principally in the three coastal countries of Liberia, Guinea and Sierra Leone, where the causative agent is maintained in multimammate rats. The high incidence of human infection reflects frequent contact with these animals, which infest living areas and are a source of food.

Clinical Manifestations

Outbreaks of the disease have occurred in west African hospitals when the lack of proper infection control measures has resulted in widespread contact with body fluids of patients or the re-use of contaminated syringes. Lassa fever has occasionally been diagnosed in travelers who have entered the United States or Europe, but no secondary transmission has occurred.

Most cases of Lassa fever are asymptomatic or mild, but among patients sick enough to require hospitalization, the case fatality rate may be 15-20% or higher.

In contrast to other types of VHF, Lassa fever usually develops gradually, after a 1-3 week incubation period. The syndrome includes fever, sore throat, headache and myalgias, which may progress to severe illness characterized by neurologic dysfunction and shock. Although hemorrhage may occur in severe cases, it is not a prominent part of the syndrome. Patients occasionally show signs of meningitis, and virus has been isolated from the cerebrospinal fluid. For unknown reasons, permanent deafness is a major complication of the disease.

Laboratory Diagnosis

Virus may be isolated from the blood throughout the course of Lassa fever, and it is also present in urine, pharyngeal secretions and other body fluids. Specific testing is based either on ELISA for viral antigens or RT-PCR.

Pathogenesis

As in other types of VHF, viral replication in Lassa fever is centered in monocytes, macrophages and dendritic cells, with the development of high levels of circulating virus in the bloodstream and spread to the liver and other body tissues, including the central nervous system. The prognosis at the time of hospital admission is strongly determined by the level of virus in the blood; patients with the highest levels are at greatest risk of death. The severity of hepatic involvement, as indicated by the serum aspartate aminotransferase level, also correlates with a fatal outcome.

In contrast to Ebola and Marburg hemorrhagic fever, in which evidence of an adaptive immune response to the infecting pathogen is rarely detectable, a majority of Lassa fever patients have IgM antibodies to Lassa virus in their plasma at the time of hospital admission, and about half have IgG. Antibody and virus may co-circulate throughout the course of the illness, with no evident relationship to the eventual outcome. The induction of cell-mediated immunity appears to be a more important factor than the humoral response in controlling and eliminating viral replication.

Susceptibility to Antiviral Drugs in Vitro and in Laboratory Animals

Ribavirin is active against Lassa virus in vitro and also proved highly effective in preventing death in infected guinea pigs and rhesus macaques. Immune plasma was also effective in treating Lassa virus-infected laboratory primates, either alone or in combination with ribavirin, but did not provide any benefit when tested in humans in limited clinical trials, perhaps because the titers of neutralizing antibodies in the preparations used were too low to produce an effect.

Antiviral Therapy

Ribavirin has been accepted as an effective treatment of Lassa fever ever since a large study in Sierra Leone showed that patients treated with either the intravenous or the oral drug had a significantly better prognosis than an earlier series of patients with similar severity of illness who received no specific therapy. The fatality rate of cases with high viremia or significant liver involvement was reduced from roughly 50 to less than 20%. Treatment was most effective if begun within 6 days after the onset of illness.

Vaccines

A number of experimental Lassa fever vaccines have been effective in laboratory rodents and nonhuman primates, but none has yet advanced to human clinical trials.

Prevention

Control measures therefore focus on reducing exposure to the virus by eliminating the viral reservoir host, the multimammate mouse, from areas of human habitation.

Crimean-Congo Hemorrhagic Fever

Virology

Crimean-Congo hemorrhagic fever (CCHF) virus is a member of the genus Nairovirus in the family Bunyaviridae. The bunyavirus genome consists of 3 separate RNA molecules. The largest (L) segment encodes the RNA-dependent RNA polymerase, the middle (M) the two virion surface glycoproteins, and the small (S) a nucleoprotein that binds to the viral genome to form the nucleocapsid. The S segment also encodes a small nonstructural (NSs) protein that has been shown to act as an inhibitor of type I interferon production by blocking the transcription of new messenger RNA in the nucleus, and thus may aid the rapid dissemination of virus from its point of entry.

Epidemiology

As its name implies, CCHF occurs over a vast geographic area. The virus was first isolated in 1944, during an outbreak of severe disease in the Crimean peninsula of the Soviet Union, where wartime conditions increased human exposure to ticks or infected animals. In 1956, a virus was recovered from a severely ill patient in the Republic of Congo, and subsequent immunologic testing showed that the two pathogens were identical. CCHF virus has since been found in a huge arc extending from western China across southern Asia into Eastern Europe, and from Turkey southwards to the tip of Africa. In all of these areas, the virus is maintained through continuous infection of ticks, which carry it to a variety of wild and domestic animals, including rodents, hares, cattle, goats, sheep and birds, all of which may serve as amplifying hosts. These animals undergo a brief period of viremia that spreads the virus to more feeding ticks, but with the possible exception of ostriches, none of these reservoir species shows signs of illness. Migratory birds may carry ticks from one region to another, but apparently do not undergo infection.

Humans are exposed to CCHF virus in several ways. Most cases occur singly and unpredictably, when inhabitants of rural areas are bitten by infected ticks. Small outbreaks can also take place when slaughterhouse employees are exposed to virus-containing tissues or aerosolized body fluids of infected animals. However, there is no evidence that disease is transmitted by meat purchased in a market, suggesting that infectious virus persists only briefly after the death of the animal host. A third mode of transmission results from contact with the body fluids of patients, usually before the nature of illness has been recognized. Small outbreaks have occurred in hospital settings such as operating rooms when abdominal pain and gastrointestinal signs and symptoms have led to surgical intervention. Spread to health care workers can be prevented through standard barrier nursing measures.

Clinical Manifestations

The clinical course of CCHF resembles that of other types of severe hemorrhagic fever. After a 3-6 day incubation period, illness typically begins with the sudden onset of high fever, headache, severe muscle pain, nausea and vomiting. The presence of hyperemia of the face and chest, conjunctival congestion and a petechial rash on the trunk may help to distinguish CCHF from other febrile syndromes. Extensive bruises on the extremities and oropharyngeal and gastrointestinal bleeding often appear during the first week. The spread of infection to the liver can lead to palpable enlargement and pain in the right upper quadrant of the abdomen. As the illness progresses, a fall in blood pressure leads to shock, multi-organ failure and coma. The fatality rate has variably been reported to range from 5-50%, but recent reports from Turkey have shown that mild cases of CCHF are surprisingly common, and that the overall case fatality rate is in fact closer to 1%.

Laboratory Diagnosis

A specific diagnosis can be made through antigen-capture ELISA or RT-PCR testing of serum samples. Virus-specific IgM and IgG can generally be detected within 10 days after the onset of illness in patients who will survive their illness, but antibody responses may not be detectable in fatal cases. Laboratory tests during the course of CCHF show typical changes of VHF, with an increase in the total white blood cell count, especially immature neutrophils, but a fall in circulating lymphocytes; a marked drop in the platelet count; rising plasma levels of the liver enzymes AST and ALT; and the development of disseminated intravascular coagulation. In endemic areas, a presumptive diagnosis can be made based on an appropriate history and the presence of thrombocytopenia and elevated liver enzymes. Patients with the most severe liver involvement and coagulopathy are most likely to die from the disease.

Pathogenesis

Because CCHF virus does not cause disease in any laboratory animal other than suckling mice, pathogenesis studies have been restricted to the testing of blood samples obtained from a small number of patients and tissues from the few cases that have come to autopsy. In general, these data the course of CCHF appears to resemble that of other types of VHF, with monocytes and macrophages playing the central role in the initiation and spread of infection. Hepatic involvement may lead to extensive necrosis; whether this represents direct infection of hepatocytes by circulating virus or spread from adjacent Kupffer cells is not known. The spleen and other lymphoid tissues show extensive destruction of lymphocytes, which probably represents the same process of “bystander” apoptosis seen in filoviral hemorrhagic fever and other severe infections.

Susceptibility to Antivral Drugs In Vitro and In Laboratory Animals

The only licensed antiviral drug that inhibits CCHF virus replication in vitro is ribavirin, described above in the context of Lassa fever. Treatment was effective in preventing death in virus-infected suckling mice, but the absence of other animal models has prevented further in vivo testing.

Antiviral Therapy

Ribavirin therapy appears to be beneficial in reducing mortality and the severity and duration of illness, based on reports of its use in a few hospital-based outbreaks, but efficacy has not been demonstrated in controlled trials. In endemic areas, a reasonable approach to the patient with suspected CCHF, based on abnormalities in circulating liver enzymes and platelet count, would be to begin oral ribavirin, then move to intravenous therapy once the diagnosis has been confirmed by specific testing.

Adjunctive Therapy

Patients with CCHF benefit from general supportive therapy with careful volume replacement and the correction of coagulation abnormalities, including administration of platelets and coagulation factors.

Vaccines

Several countries endemic for CCHF have developed inactivated vaccines, such as formalin-treated suckling mouse brain preparations, but controlled trials to demonstrate efficacy have never been performed. No FDA-approved vaccine is available.

Prevention

Control measures are based primarily on reducing exposure to infected ticks and on efforts to protect health care workers through algorithms for the early recognition of CCHF patients.

Rift Valley Fever

Virology

Rift Valley fever (RVF) virus is a member of the genus Phlebovirus in the family Bunyaviridae. Viral genome organization and virion structure are similar to those of CCHF virus, as described above.

Epidemiology

RVF virus causes severe disease in livestock and humans across the entire extent of sub-Saharan Africa, extending north along the Nile Valley into Egypt. The agent is maintained through cyclic infection of Culex mosquitoes and rodents, but human exposure can occur through multiple pathways. Aedes mosquitoes and biting flies can transmit RVF virus to livestock animals and to humans; high levels of viremia then support further mosquito transmission. Infection of sheep and goats induces abortion, exposing farmers to high concentrations of virus in blood and placental tissues, and persons who butcher animals may also acquire the disease. Although most human infections occur sporadically, increases in vector populations can produce large outbreaks. An epidemic in 1977 involved more than 200,000 human infections in Egypt, with a case fatality rate of 0.3%. Current methods of satellite imaging and rainfall prediction are facilitating advanced planning for disease outbreaks. In 2000, RVF virus was carried across the Red Sea to Yemen through the local sheep trade, raising concerns that the disease may become established in the Middle East.

Clinical Manifestations

RVF is highly varied in its clinical manifestations. The majority of infected humans either remain asymptomatic or develop a mild, nonspecific flu-like illness with fever, headache, myalgias, and gastrointestinal symptoms, which usually resolves within 1-2 weeks. About 5% of patients progress within days of the onset of illness to a hemorrhagic fever syndrome, characterized by a petechial rash, mucosal bleeding and signs of severe liver involvement. Severely ill patients show elevated circulating liver enzymes and jaundice, and may die in hepatic failure.

A small percentage of RVF patients develop other forms of illness that usually develop several weeks after the initial symptoms, suggesting that they may at least in part be a consequence of the immune response to infection. Some patients develop signs of central nervous system disease, beginning with headache and progressing to severe or fatal encephalitis. Others develop ocular involvement, manifested as retinal inflammation and hemorrhage, often bilateral, which results in partial or complete loss of vision in up to half of cases.

Laboratory Diagnosis

When available, ELISA or RT-PCR testing can detect antigens or RNA of RVF virus in the bloodstream of acutely ill patients. A diagnosis can also be made through the detection of RVF virus-specific IgM during the course of illness.

Pathogenesis

The detailed pathogenesis of RVF is not known. As in other types of VHF, initial infection by mosquito bite or inhalation of virus-containing material probably involves monocytes, macrophages and dendritic cells, from which the virus disseminates to other lymphatic tissues and the liver. Severe illness is characterized by extensive hepatic infection and necrosis. The delayed ophthalmic and neurologic syndromes may represent host immune responses to viral invasion of those tissues.

Susceptibility to Antivral Drugs In Vitro and In Laboratory Animals

Ribavirin has been active in suppressing the replication of RVF virus in cell culture and in protecting laboratory animals against lethal virus challenge. Interferon and specific immune serum were protective in nonhuman primates, but these approaches have not been extended to human therapy.

Antiviral Therapy

There is no specific therapy for Rift Valley fever. Even though ribavirin was highly protective in laboratory animals, treatment has given equivocal results in humans, and is not currently recommended.

Vaccines

A formalin-inactivated, cell culture-derived vaccine is in widespread use as a veterinary vaccine in Africa, and is available in the United States as an Investigational New Drug.

Prevention

Control measures during outbreaks focus on vector control and vaccination of livestock to restrict the spread of infection.

READING LIST

1. Bausch D, Sprecher AG, Jeffs B, Boumandouki P. Treatment of Marburg and Ebola hemorrhagic fevers: a strategy for testing new drugs and vaccines under outbreak conditions. Antiviral Research 2008.

2. Bray M. Highly pathogenic RNA viral infections: challenges for antiviral research. Antiviral Research 2008.

3. Bray M. Pathogenesis of viral hemorrhagic fever. Current Opinion in Immunology 2005;17: 399-403. [PubMed]

4. Bray M, Murphy FA. Filovirus research: knowledge expands to meet a growing threat. Journal of Infectious Diseases 2007;196 Suppl 2:S438-43. [PubMed]

5. Ergonul O. Treatment of Crimean-Congo hemorrhagic fever. Antiviral Research 2008.

6. Geisbert T, Jahrling P. Exotic emerging viral diseases: progress and challenges. Nature Medicine 2004;10:S110-21.[PubMed]

7. Gowen B, Holbrook M. Animal models of highly pathogenic RNA viral infections: hemorrhagic fever viruses. Antiviral Research 2008.

8. Khan SH, Goba A, Chu M, et al. New opportunities for research in the pathogenesis and treatment of Lassa fever. Antiviral Research 2008.