Viral Hemorrhagic Fevers

Definition

Viral hemorrhagic fever is an acute systemic illness classically involving fever, a constellation of initially nonspecific signs and symptoms, and a propensity for bleeding and shock.

The Pathogens

Viral hemorrhagic fever may be caused by more than 30 different viruses from four taxonomic families, Filoviridae, Arenaviridae, Bunyaviridae, and Flaviviridae ( Table 351-1 ), but not every virus in these families causes the syndrome. Recently discovered viruses such as severe fever with thrombocytopenia and heartland viruses (family Phenuiviridae family) have variably been considered as hemorrhagic fever viruses, thereby highlighting the extreme heterogeneity and perhaps increasing ambiguity of this categorization. All implicated viruses are single-stranded lipid-enveloped RNA viruses with small genomes (10 to 19 kilobases) that can be relatively easily inactivated in the environment.

TABLE 351-1

PRINCIPAL VIRUSES CAUSING HEMORRHAGIC FEVER

FAMILY AND VIRUS	DISEASE	GEOGRAPHIC DISTRIBUTION OF DISEASE	PRINCIPAL RESERVOIR/VECTOR	ANNUAL ESTIMATED CASES	CASE-TO-INFECTION RATIO	HUMAN-TO-HUMAN TRANSMISSIBILITY
FILOVIRIDAE
Ebola virus	Ebola virus disease	Central and West Africa	Fruit bat?	— ∗	1 : 1	High
Marburg virus	Marburg virus disease	Sub-Saharan Africa	Egyptian fruit bat (Rousettus aegyptiacus)	— ∗	1 : 1	High
ARENAVIRIDAE ^† ^, ^‡
Old World Group
Lassa	Lassa fever	West Africa	Rodent: natal mastomys or multimammate rat (Mastomys natalensis)	50,000-100,000	1 : 5-10	Moderate
Lujo ^§	Lujo HF	Zambia	Unknown, presumed rodent	Unknown	Unknown	Moderate to high
New World Group
Junín	Argentine HF	Argentine pampas	Rodent: corn mouse (Calomys musculinus)	≈100	1 : 1.5	Low
Machupo	Bolivian HF	Beni Department, Bolivia	Rodent: large vesper mouse (Calomys callosus)	≤50	1 : 1.5	Low
Guanarito	Venezuelan HF	Portuguesa State, Venezuela	Rodent: cane mouse (Zygodontomys brevicauda)	≤50	1 : 1.5	Low
Sabiá ^\|\|	Brazilian HF	São Paulo and Para States, Brazil	Unknown, presumed rodent	Unknown	1 : 1.5	Low?
Chapare ^¶	Chapare HF	Cochabamba and La Paz Departments, Bolivia	Unknown, presumed rodent	Unknown	Unknown	Unknown
BUNYAVIRIDAE ^†
Old World Complex
Hantaan, Seoul, Puumala, Dobrava-Belgrade, others	HF with renal syndrome	Hantaan: northeast Asia. Seoul: urban areas worldwide Puumala and Dobrava-Belgrade: Europe	Rodent Hantaan: striped field mouse (Apodemus agrarius) Seoul: brown or Norway rat (Rattus norvegicus) Puumala: bank vole (Clethrionomys glareolus) Dobrava-Belgrade: yellow-necked field mouse (Apodemus flavicollis)	50,000-150,000	Hantaan: 1 : 1.5 Others: 1 : 20	None
New World Complex
Sin Nombre, Andes, Laguna Negra, others	Hantavirus pulmonary syndrome	Americas	Rodent Sin Nombre: deer mouse (Peromyscus maniculatus) Andes: long-tailed colilargo (Oligoryzomys longicaudatus) Laguna Negra: little laucha or small vesper mouse (Calomys laucha)	50,000-150,000	Sin Nombre: 1 : 1 Others: up to 1 : 20	None, except for Andes virus
Rift Valley fever	Rift Valley fever	Sub-Saharan Africa, Madagascar, Saudi Arabia, Yemen	Domestic livestock/mosquitoes (sylvatic Aedes and others)	100-100,000 ∗ ^, ∗∗	1 : 100	None
Crimean-Congo HF	Crimean-Congo HF	Africa, Balkans, southern Russia, Middle East, India, Pakistan, Afghanistan, western China	Wild and domestic vertebrates/tick (primarily Hyalomma species)	≈500	1 : 1-2	High
FLAVIVIRIDAE
Yellow fever	Yellow fever	Sub-Saharan Africa, South America up to Panama	Monkey/mosquito ( Aedes aegypti , other Aedes and Haemagogus species)	5,000-200,000 ^††	1 : 2-20	None
Dengue	Dengue HF	Tropics and subtropics worldwide	Human/mosquito ( A . aegypti and albopictus )	100,000-200,000 ^††	1 : 10-100, depending on age, previous infection, genetic background, and infecting serotype	None
Omsk HF	Omsk HF	Western Siberia	Rodent/tick (primarily Dermacentor and Ixodes species)	100-200	Unknown	Not reported
Kyasanur Forest disease	Kyasanur Forest disease	Karnataka state, India; Yunnan Province, China; Saudi Arabia	Vertebrate (rodents, bats, birds, monkeys, others)/tick ( Haemaphysalis species and others)	≈500	Unknown	Not reported, but laboratory infections have occurred
Alkhumra HF ^‡‡	Proposed name: Alkhumra HF	Saudi Arabia, Egypt	Ticks?	≤50	Unknown	Not reported

HF = hemorrhagic fever.

∗ Although some endemic transmission of the filoviruses (Ebola virus > Marburg virus) and Rift Valley fever virus occurs, these viruses are most often associated with outbreaks.

† The virus families Arenaviridae and Bunyaviridae are serologically, phylogenetically, and geographically divided into Old World (i.e., Africa and Asia) and New World (i.e., the Americas) complexes.

‡ In addition to the arenaviruses listed in the table, Flexal and Tacaribe viruses have caused human disease as a result of laboratory accidents. Another arenavirus, Whitewater Arroyo, has been noted in sick persons in California, but its role as a pathogen has not been clearly established.

§ Only five cases (four of them fatal) from one outbreak in 2008 have been noted. The index case came to South Africa from Zambia.

|| Only five cases (three fatal) have been noted, two of them from laboratory accidents.

¶ Discovered from a small outbreak in 2003 from which blood was obtained from one fatal case and Chapare virus isolated. Few other details have been reported. A cluster of five cases, three fatal, occurred on the outskirts of La Paz, Bolivia, in 2019.

∗∗ Although Rift Valley fever virus can be found throughout sub-Saharan Africa, large outbreaks usually occur in East Africa’s Rift Valley region.

†† Based on estimates from the World Health Organization. Significant underreporting occurs. Incidence may fluctuate widely in place and time.

‡‡ Alkhumra is considered by some to be a variant of Kyasanur Forest disease virus. Disagreement exists over the proper spelling of the virus, written as Alkhurma in some publications.

Epidemiology

Maintenance in Nature and Transmission to Humans

With the exception of dengue virus, for which humans can now be considered the reservoir, all hemorrhagic fever viruses are zoonotic, maintained in nature in mammalian reservoirs that tend to be specific for each virus (see Table 351-1 ). Although viral hemorrhagic fevers collectively can be found worldwide, the endemic area of any given hemorrhagic fever virus is usually smaller than the extent of its natural reservoir or arthropod vector. With the exception of dengue, yellow fever, and some hantaviruses, human infection is generally infrequent and humans are usually dead-end hosts.

Depending upon the virus, primary transmission to humans may be from contact with infected animal excreta or other body fluids or from the bite of an arthropod vector. Infection may occur through direct exposure of mucous membranes or broken skin to the infected blood or excreta of an animal reservoir, which usually occurs inadvertently or, in the case of the flaviviruses and most of the bunyaviruses, by the bite of an arthropod vector. Transmissibility between humans and pathogenicity vary with the specific virus and sometimes even among strains of the same virus. The infectious dose for most hemorrhagic fever viruses appears to be low, sometimes on the order of just a few virions. Airborne transmission is not a predominant mode of spread, if it occurs at all; however, studies in nonhuman primates show that transmission of many hemorrhagic fever viruses is possible through artificially created aerosols, thereby raising the possibility of their potential use as bioweapons ( Chapter 19 ).

Bat-Borne Viruses

The filoviruses (from the Latin filo , “thread,” referring to their filamentous shape), Marburg and Ebola, are perhaps the most feared of all hemorrhagic fever viruses. ^, Fruit bats appear to be the filovirus reservoir, although more conclusive evidence is needed for Ebola virus, with transmission to humans likely from exposure to infected bat excreta or saliva. Nonhuman primates, especially gorillas and chimpanzees, and other wild animals may become infected, presumably from similar bat exposure, and transmit filoviruses to humans through contact with blood and body fluids of these animals, usually in association with hunting. Nonhuman primates, which are also dead-end hosts who develop severe and usually fatal disease similar to that seen in humans, may be easier prey for hunters when sick. Because hemorrhagic fever viruses are rapidly inactivated by heating, infection probably occurs by exposure during butchering and preparation, rather than by consumption of cooked meat. The 2013 to 2016 outbreak of Ebola virus disease in West Africa dwarfed all prior Ebola outbreaks combined, with a reported 28,616 cases and 11,310 deaths. Sporadic outbreaks continue to occur in the Democratic Republic of the Congo, West Africa, and Uganda, either through primary infection from wild animals or renewed secondary infection from persistently infected humans (see later).

Rodent-Borne Viruses

Arenaviruses (from the Latin arena , “sand”) are divided into Old World (or lymphocytic choriomeningitis/Lassa) and New World (or Tacaribe) complexes. Lassa and Lujo viruses are found in Africa, whereas Junín, Machupo, Guanarito, Sabiá, and Chapare viruses are found in South America. Although there may be subtle differences among the syndromes produced by the New World arenaviruses, they are usually grouped together simply as the South American hemorrhagic fevers.

The genus Hantavirus of the Bunyaviridae family is similarly divided into Old and New World complexes. The Old World hantaviruses, such as Hantaan, Seoul, and Puumala, among many others, classically cause hemorrhagic fevers with prominent renal involvement across Europe and Asia. New World hantaviruses, such as Sin Nombre and Andes, among many others, cause hantavirus pulmonary syndrome, sometimes also called hantavirus cardiopulmonary syndrome to emphasize the significant cardiogenic component of this disease.

Pathogenic arenaviruses and hantaviruses are maintained in nature through chronic subclinical infection in rodents of the Muridae family, with a strict pairing between the specific virus and rodent species. Transmission between rodents may be by vertical or horizontal transmission or both, depending on the specific virus. Transmission to humans occurs through exposure to rodent excreta, either from aerosols produced when rodents urinate or by direct inoculation to the mucous membranes, although the precise modes of transmission remain to be elucidated. Secondary aerosol generation is notoriously inefficient, so disturbing shed urine is a less likely mechanism of infection. In West Africa, Lassa virus is sometimes contracted when rodents are trapped and prepared for consumption or, more rarely, through a rodent bite. Experimental data suggest that humans may also be infected with arenaviruses by the oral route.

The rodents that transmit Lassa, Machupo, and many of the Old World hantaviruses commonly invade the peridomestic environment, thereby putting housewives, children, and others who spend time at home at risk. In contrast, the reservoirs for Junín and Guanarito viruses, as well as most of the New World hantaviruses, typically inhabit agricultural fields, wood lots, or other rural habitats, thereby putting outdoor workers, campers, and hikers at risk.

Mosquito-Borne Viruses

Rift Valley fever virus is maintained in domestic livestock, such as cattle, buffalo, sheep, goats, and camels, in which it often provokes spontaneous abortion. The virus may be transmitted to humans by direct exposure to these animals, especially during parturition, or by mosquitoes. Farmers, abattoir workers, and veterinarians are at particular risk.

Yellow fever virus is maintained in a cycle between monkeys and forest canopy mosquitoes. Sporadic cases occur when humans are bitten by these mosquitoes, known as the sylvatic or jungle cycle. Larger outbreaks occur when humans bring the virus back to more settled environments, where the urban mosquito Aedes aegypti can spread the virus directly between humans (the urban cycle). Aedes aegypti , which typically lay eggs in artificial containers around the home and bite during the day, become infective a few weeks after feeding on a viremic monkey or human. An estimated 90% of infections occur on the African continent, but large outbreaks have also been seen in South America in recent years, thereby prompting mass vaccination campaigns of millions of people and occasionally resulting in exported cases. Although Aedes aegypti , mosquitoes are commonly found across Asia, the Pacific, and Australia, endogenous transmission of yellow fever virus has never been recorded in these locations for unknown reasons.

Although nonhuman primates are also a reservoir for sylvatic strains of dengue, the virus is now adapted and maintained in humans, with a regular transmission cycle akin to that of urban yellow fever. Dengue virus is found in the tropics and subtropics nearly worldwide. The incidence of infection has grown dramatically in recent decades, especially in Asia and Latin America. Less than 10% of infected persons develop hemorrhagic fever, primarily children between the ages of 4 and 12 years.

Tick-Borne Viruses

The viruses that cause Crimean-Congo hemorrhagic fever, Omsk hemorrhagic fever, Kyasanur Forest disease, and Alkhumra hemorrhagic fever are maintained in small mammals, such as rodents, hares, and hedgehogs, among which the viruses are spread by ticks. Humans are infected either by tick bites or by exposure to contaminated blood or excreta of the reservoir animals. Ticks also spread Crimean-Congo hemorrhagic fever virus to large mammals, including cattle and other domestic livestock, whose transient and asymptomatic viremia puts farmers, abattoir workers, and veterinarians at risk.

Human-to-Human Transmission

Secondary human-to-human transmission occurs with many of the hemorrhagic fever viruses (see Table 351-1 ), but attack rates are generally low (less than 15% for Ebola Zaire virus) because transmission between humans requires direct contact with contaminated blood or body fluids. Nevertheless, as evidenced by the massive 2013 to 2016 outbreak of Ebola virus disease in West Africa, the risk of transmission can increase significantly when many sick patients overwhelm fragile health systems and infection prevention and control measures break down, especially when health care workers are insufficiently trained and equipped. Human-to-human transmission probably usually occurs through oral or mucous membrane exposure, most often in the context of providing care to a sick family member (community) or patient (nosocomial transmission). Funeral rituals that entail the touching of the corpse have been a major source of infection in some outbreaks. Infection through fomites cannot be excluded. In some cases, highly infectious “super spreaders” have been suspected, but there is no evidence for natural aerosol transmission between humans. Despite modern-day travel, imported cases of viral hemorrhagic fever remain infrequent and usually do not result in secondary transmission because of the more routinely maintained infection prevention and control practices in resource-rich countries.

With the exception of hantaviruses and some of the flaviviruses, infectivity parallels the clinical state; persons are most infectious late in the course of severe disease, when viral loads are high and patients shed virus into the environment through vomiting, diarrhea, and bleeding. The risk of transmission during the incubation period or from asymptomatic persons is negligible, although a case of Argentine hemorrhagic fever occurred after blood transfusion from a donor who was asymptomatic at the time the blood was taken.

Pathobiology

Microvascular instability and impaired hemostasis are the pathobiologic hallmarks of viral hemorrhagic fever. However, not all infections lead to these conditions, which are seen only in a subset of patients at the severe end of the disease spectrum.

With the exception of disease caused by the hantaviruses and some of the flaviviruses, the pathogenesis of viral hemorrhagic fever appears to be related to unchecked viremia, with most fatal cases failing to mount an effective immune response. In these diseases, infectious virus is cleared rapidly from the blood in survivors. In dengue, yellow fever, and hantavirus infections, by comparison, viral replication is usually controlled, and virus is cleared from the blood before the most severe phase of the disease. The unique process of antibody-mediated immune enhancement, in which secondary infection with a different dengue virus serotype is more severe than the primary one, may play a role in the pathogenesis of dengue hemorrhagic fever. Data from animal models suggest that cardiac inotropy may also be directly or indirectly inhibited in some viral hemorrhagic fevers, especially Lassa fever.

After inoculation, virus first replicates in dendritic cells and other local tissues, with subsequent migration to regional lymph nodes and then dissemination through the lymph and blood monocytes to a broad range of tissues and organs, including the liver, spleen, lymph nodes, adrenal glands, lungs, and endothelium. Migration of tissue macrophages results in secondary infection of permissive parenchymal cells. During the acute illness, virus can be found in a wide variety of body fluids, including blood, saliva, stool, and breast milk.

The interaction of virus with immune cells, especially macrophages and endothelial cells, results directly or indirectly (through soluble mediators) in cell activation and the unleashing of an inflammatory and vasoactive process consistent with the systemic inflammatory response syndrome. The synthesis of cell surface tissue factor triggers the extrinsic coagulation pathway. Impaired hemostasis may entail endothelial cell, platelet, or coagulation factor dysfunction. Disseminated intravascular coagulopathy (DIC; Chapter 161 ) is frequently noted, especially with Ebola, Marburg, and Crimean-Congo hemorrhagic fever viruses.

Inflammatory cell infiltrates, which are usually mild, consist of a mix of mononuclear cells and neutrophils. In some viral hemorrhagic fevers, such as Ebola virus disease, virus replication and dissemination are facilitated by virus-induced suppression of the host adaptive immune response. Failure of the immune response to adequately respond appears to be a major determinant of severity in Lassa fever.

Tissue damage may be mediated through direct necrosis of infected cells or indirectly through apoptosis of immune cells, as seen in other forms of septic shock. A dysregulated and uncontrolled host immune response leads to end-organ damage, with the most affected organs varying with the virus ( Table 351-2 ). For example, renal tubular necrosis and retroperitoneal edema are seen in hemorrhagic fever with renal syndrome, whereas interstitial pneumonitis and myocardial depression are the hallmarks of hantavirus pulmonary syndrome. The liver is particularly affected in yellow fever, with fatty degeneration, coagulative midzonal necrosis of hepatocytes, and the presence of Councilman bodies. The brain and meninges are particularly affected in Kyasanur Forest disease and Omsk hemorrhagic fever and often in the South American hemorrhagic fevers as well. Reticuloendothelial proliferation is seen in Kyasanur Forest disease, with marked erythrophagocytosis in the spleen.

TABLE 351-2

PATHOBIOLOGIC AND CLINICAL ASPECTS OF VIRAL HEMORRHAGIC FEVERS

DISEASE	INCUBATION PERIOD (DAYS)	ONSET	BLEEDING	RASH	JAUNDICE	HEART	LUNG	KIDNEY	CENTRAL NERVOUS SYSTEM	EYE	CASE-FATALITY RATIO	CLINICAL MANAGEMENT
FILOVIRIDAE
Ebola virus disease	3-21	Variable	++	+	+	+	+	+	+	++ ∗	40-85% ^†	Monoclonal antibodies
Marburg virus disease	3-21	Abrupt	++	+	+	+	+	+	+	++	22-85% ^‡	Supportive
ARENAVIRIDAE
Lassa fever	5-16	Gradual	+	+ ^§	0	++	+	++	+	0	20%	Ribavirin
Lujo HF	9-13	Abrupt	++	+	0	?	+	+	+	0	80%	Ribavirin
South American HF ^\|\|	4-14	Gradual	+++	+	0	++	+	0	+++	0	15-40%	Ribavirin, convalescent plasma
BUNYAVIRIDAE
HF with renal syndrome	9-35	Abrupt	+++	0	0	++	+	+++	+	0	<1-50%, depending on specific virus	Ribavirin
Hantavirus pulmonary syndrome	7-35	Gradual	0 (except for Andes virus infection)	0	0	+++	+++	+	+	0	<1-50%, depending on specific virus	Supportive, consider high volume hemofiltration and ECMO
Rift Valley fever ^¶	2-5	Abrupt	++	+	++	+?	0	+	++	++	Up to 50% in severe forms	Ribavirin?
Crimean-Congo HF	1-12 ∗∗	Abrupt	+++	0	++	+?	+	0	+	0	15-30%	Ribavirin?
FLAVIVIRIDAE
Yellow fever	3-6	Abrupt	+++	0	+++	++	+	++	++	0	20-50%	Supportive
Dengue HF	3-15	Abrupt	++	+++	+	++	+	0	+	0	Untreated: 10-15% Treated: ≤1%	Supportive
Omsk HF	3-8	Abrupt	++	0	0	+	++	0	+++	+	1-3%	Supportive
Kyasanur forest disease	3-8	Abrupt	++	0	0	+	++	0	+++	+	3-5%	Supportive
Alkhumra HF ^††	3-8	Abrupt	++	+	+	+	+	0	++	+	20-25%	Supportive

ECMO = extracorporeal membrane oxygenation; HF = hemorrhagic fever; 0 = sign not typically noted/organ not typically affected; + = sign occasionally noted/organ occasionally affected; ++ = sign commonly noted/organ commonly affected; +++ = sign characteristic/organ involvement severe.

∗ Potentially sight-threatening uveitis, early cataract formation, and other ocular sequelae are noted in up to one third of survivors.

† Four species of Ebola virus are known to cause disease in humans, with varying associated case-fatality ratios during outbreaks: Zaire, 85%; Sudan, 55%; Bundibugyo, 40%; and Tai Forest (formerly Côte d’Ivoire), 0% (only one recognized case, who survived).

‡ The case-fatality ratio was 22% in the first recognized outbreak of Marburg virus disease in Germany and Yugoslavia in 1967 but has been consistently above 80% in outbreaks in central Africa, where the virus is endemic. Possible reasons for this discrepancy include differences in quality of care, strain pathogenicity, route and dose of infection, underlying prevalence of immunodeficiency and comorbid illnesses, and genetic susceptibility.

§ A morbilliform or maculopapular rash almost always occurs in persons with lighter skin, who are usually expatriates, but for unclear reasons is rarely present in darker-skinned Africans from the endemic area.

|| Data are insufficient to distinguish between the syndromes produced by the various arenaviruses found in the Americas. They are thus frequently grouped as the South American hemorrhagic fevers.

¶ Hemorrhagic fever, encephalitis, and retinitis may be seen in Rift Valley fever independently of each other.

∗∗ The incubation period of Crimean-Congo HF varies with the mode of transmission: typically 1 to 3 days after tick bite and 5 to 6 days after contact with infected animal blood or tissues.

†† Based on preliminary observations. Fewer than 100 cases have been reported.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here