Chapter 263 Other Viral Hemorrhagic Fevers
Viral hemorrhagic fevers are a loosely defined group of clinical syndromes in which hemorrhagic manifestations are either common or especially notable in severe illness. Both the etiologic agents and clinical features of the syndromes differ, but disseminated intravascular coagulopathy may be a common pathogenetic feature.
Six of the viral hemorrhagic fevers are caused by arthropod-borne viruses (arboviruses) (Table 263-1). Four are togaviruses of the family Flaviviridae: Kyasanur Forest disease, Omsk, dengue (Chapter 261), and yellow fever (Chapter 262) viruses. Three are of the family Bunyaviridae: Congo, Hantaan, and Rift Valley fever (RVF) viruses. Four are of the family Arenaviridae: Junin, Machupo, Guanarito, and Lassa viruses. Two are of the family Filoviridae: Ebola and Marburg viruses. The Filoviridae are enveloped, filamentous RNA viruses that are sometimes branched, unlike any other known virus.
Table 263-1 VIRAL HEMORRHAGE FEVERS (HFs)
| MODE OF TRANSMISSION | DISEASE | VIRUS |
|---|---|---|
| Tick-borne | Crimean-Congo HF* | Congo |
| Kyasanur Forest disease | Kyasanur Forest disease | |
| Omsk HF | Omsk | |
| Mosquito-borne† | Dengue HF | Dengue (four types) |
| Rift Valley fever | Rift Valley fever | |
| Yellow fever | Yellow fever | |
| Infected animals or materials to humans | Argentine HF | Junin |
| Bolivian HF | Machupo | |
| Lassa fever* | Lassa | |
| Marburg disease* | Marburg | |
| Ebola HF* | Ebola | |
| Hemorrhagic fever with renal syndrome | Hantaan |
* Patients may be contagious; nosocomial infections are common.
† Chikungunya virus is associated infrequently with petechiae and epistaxis. Severe hemorrhagic manifestations have been reported in some cases.
With some exceptions, the viruses causing viral hemorrhagic fevers are transmitted to humans via a nonhuman entity. The specific ecosystem required for viral survival determines the geographic distribution of disease. Although it is commonly thought that all viral hemorrhagic fevers are arthropod borne, 7 may be contracted from environmental contamination caused by animals or animal cells or from infected humans (see Table 263-1). Laboratory and hospital infections have occurred with many of these agents. Lassa fever and Argentine and Bolivian hemorrhagic fevers are reportedly milder in children than in adults.
Sporadic human infection with Crimean-Congo hemorrhage fever in Africa provided the original virus isolation. Natural foci are recognized in Bulgaria, western Crimea, and the Rostov-on-Don and Astrakhan regions; a somewhat similar disease occurs in Kazakhstan and Uzbekistan. Index cases were followed by nosocomial transmission in Pakistan and Afghanistan in 1976, in the Arabian Peninsula in 1983, and in South Africa in 1984. Outbreaks have been reported from Pakistan, Oman, and southern Russia. In the Russian Federation, the vectors are Hyalomma marginatum and Hyalomma anatolicum, which, along with hares and birds, may serve as viral reservoirs. Disease occurs from June to September, largely among farmers and dairy workers.
Human cases of Kyasanur Forest disease occur chiefly in adults in an area of Mysore State, India. The main vectors are 2 Ixodidae ticks, Haemaphysalis turturis and Haemaphysalis spinigera. Monkeys and forest rodents may be amplifying hosts. Laboratory infections are common.
Omsk hemorrhagic fever occurs throughout south-central Russia and northern Romania. Vectors may include Dermacentor pictus and Dermacentor marginatus, but direct transmission from moles and muskrats to humans seems well established. Human disease occurs in a spring-summer-autumn pattern, paralleling the activity of vectors. This infection occurs most frequently in persons with outdoor occupational exposure. Laboratory infections are common.
The virus causing RVF is responsible for epizootics involving sheep, cattle, buffalo, certain antelopes, and rodents in North, Central, East, and South Africa. The virus is transmitted to domestic animals by Culex theileri and several Aedes species. Mosquitoes may serve as reservoirs by transovarial transmission. An epizootic in Egypt in 1977-1978 was accompanied by thousands of human infections, principally among veterinarians, farmers, and farm laborers. Smaller outbreaks occurred in Senegal in 1987, Madagascar in 1990, and Saudi Arabia and Yemen in 2000-2001. Humans are most often infected during the slaughter or skinning of sick or dead animals. Laboratory infection is common.
Before introduction of vaccine, hundreds to thousands of cases of Argentine hemorrhage fever occurred annually from April through July in the maize-producing area northwest of Buenos Aires that reaches to the eastern margin of the Province of Cordoba. Junin virus has been isolated from the rodents Mus musculus, Akodon arenicola, and Calomys laucha laucha. It infects migrant laborers who harvest the maize and who inhabit rodent-contaminated shelters.
The recognized endemic area of Bolivian hemorrhagic fever consists of the sparsely populated province of Beni in Amazonian Bolivia. Sporadic cases occur in farm families who raise maize, rice, yucca, and beans. In the town of San Joaquin, a disturbance in the domestic rodent ecosystem may have led to an outbreak of household infection caused by Machupo virus transmitted by chronically infected Calomys callosus, ordinarily a field rodent. Mortality rates are high in young children.
In 1989, an outbreak of hemorrhagic illness occurred in the farming community of Guanarito, Venezuela, 200 miles south of Caracas. Subsequently, in 1990-1991, there were 104 cases reported with 26 deaths due to Guanarito virus. Cotton rats (Sigmodon alstoni) and cane rats (Zygodontomys brevicauda) have been implicated as likely reservoirs of Venezuelan hemorrhage fever.
Lassa virus has an unusual potential for human-to-human spread, which has resulted in many small epidemics in Nigeria, Sierra Leone, and Liberia. Medical workers in Africa and the USA have also contracted the disease. Patients with acute Lassa fever have been transported by international aircraft, necessitating extensive surveillance among passengers and crews. The virus is probably maintained in nature in a species of African peridomestic rodent, Mastomys natalensis. Rodent-to-rodent transmission and infection of humans probably operate via mechanisms established for other arenaviruses.
Until recently, the world experience of Marburg disease had been limited to 26 primary and 5 secondary cases in Germany and Yugoslavia in 1967, and to small outbreaks in Zimbabwe in 1975, Kenya in 1980 and 1988, and South Africa in 1983. But, in 1999 a large outbreak occurred in Congo Republic and a still larger outbreak in Uige Province, Angola, with 351 cases and 312 deaths in 2005. In laboratory and clinical settings, transmission occurs by direct contact with tissues of the African green monkey or with infected human blood or semen. A reservoir in bats has been demonstrated. It appears that the virus is transmitted by close contact between fructivorous bats and from bats by aerosol to humans.
Ebola virus was isolated in 1976 from a devastating epidemic involving small villages in northern Zaire and southern Sudan; smaller outbreaks have occurred subsequently. Outbreaks have initially been nosocomial. Attack rates have been highest in the birth- 1 yr old and 15-50 yr old age groups. The virus is closely related to Marburg virus. Ebola virus has been particularly active recently, with an outbreak in Kikwit, Zaire, in 1995, followed recently by scattered outbreaks in Uganda and Central and West Africa. The virus has been recovered from chimpanzees, and antibodies have been found in other subhuman primates, which apparently acquire infection from a zoonotic reservoir in bats. The mode of transmission to humans is unknown. Reston virus, related to Ebola, has been recovered from Philippine monkeys and pigs and has caused subclinical infections in workers in monkey colonies in the USA.
The endemic area of hemorrhage fever with renal syndrome (HFRS), also known as epidemic hemorrhagic fever and Korean hemorrhagic fever, includes Japan, Korea, far eastern Siberia, north and central China, European and Asian Russia, Scandinavia, Czechoslovakia, Romania, Bulgaria, Yugoslavia, and Greece. Although the incidence and severity of hemorrhagic manifestations and the mortality are lower in Europe than in northeastern Asia, the renal lesions are the same. Disease in Scandinavia, nephropathia epidemica, is caused by a different although antigenically related virus, Puumala virus, associated with the bank vole, Clethrionomys glareolus. Cases occur predominantly in the spring and summer. There appears to be no age factor in susceptibility, but because of occupational hazards, young adult men are most frequently attacked. Rodent plagues and evidence of rodent infestation have accompanied endemic and epidemic occurrences. Hantaan virus has been detected in lung tissue and excreta of Apodemus agrarius coreae. Antigenically related agents have been detected in laboratory rats and in urban rat populations around the world, including Prospect Hill virus in the wild rodent Microtus pennsylvanicus in North America and Sin Nombre virus in the deer mouse in the southern and southwestern USA; these viruses are causes of hantavirus pulmonary syndrome (Chapter 265). Rodent-to-rodent and rodent-to-human transmission presumably occurs via the respiratory route.
Dengue hemorrhagic fever (Chapter 261) and yellow fever (Chapter 262) cause similar syndromes in children in endemic areas.
The incubation period of 3-12 days is followed by a febrile period of 5-12 days and a prolonged convalescence. Illness begins suddenly with fever, severe headache, myalgia, abdominal pain, anorexia, nausea, and vomiting. After 1-2 days, fever may subside until the patient experiences an erythematous facial or truncal flush and injected conjunctivae. A 2nd febrile period of 2-6 days then develops, with a hemorrhagic enanthem on the soft palate and a fine petechial rash on the chest and abdomen. Less frequently, there are large areas of purpura and bleeding from the gums, nose, intestines, lungs, or uterus. Hematuria and proteinuria are relatively rare. During the hemorrhagic stage, there is usually tachycardia with diminished heart sounds and occasionally hypotension. The liver is usually enlarged, but there is no icterus. In protracted cases, central nervous system signs include delirium, somnolence, and progressive clouding of consciousness. Early in the disease, leukopenia with relative lymphocytosis, progressively worsening thrombocytopenia, and gradually increasing anemia occur. In convalescence there may be hearing and memory loss. The mortality rate is 2-50%.
After an incubation period of 3-8 days, both Kyasanur Forest disease and Omsk hemorrhagic fever begin with sudden onset of fever and headache. Kyasanur Forest disease is characterized by severe myalgia, prostration, and bronchiolar involvement; it often manifests without hemorrhage but occasionally with severe gastrointestinal bleeding. In Omsk hemorrhagic fever, there is moderate epistaxis, hematemesis, and a hemorrhagic enanthem but no profuse hemorrhage; bronchopneumonia is common. In both diseases, severe leukopenia and thrombocytopenia, vascular dilatation, increased vascular permeability, gastrointestinal hemorrhages, and subserosal and interstitial petechial hemorrhages occur. Kyasanur Forest disease may be complicated by acute degeneration of renal tubules and focal liver damage. In many patients, recurrent febrile illness may follow an afebrile period of 7-15 days. This 2nd phase takes the form of a meningoencephalitis.
Most Rift Valley fever infections have occurred in adults with signs and symptoms resembling those of dengue fever (Chapter 261). Onset is acute, with fever, headache, prostration, myalgia, anorexia, nausea, vomiting, conjunctivitis, and lymphadenopathy. The fever lasts 3-6 days and is often biphasic. Convalescence is often prolonged. In the 1977-1978 outbreak many patients died after showing signs that included purpura, epistaxis, hematemesis, and melena. RVF affects the uvea and posterior chorioretina; macular scarring, vascular occlusion, and optic atrophy occur, resulting in permanent visual loss in a high proportion of patients with mild to severe RVF. At autopsy in one report, extensive eosinophilic degeneration of the parenchymal cells of the liver were observed.
The incubation period in Argentine, Venezuelan, and Bolivian hemorrhagic fevers and Lassa fever is commonly 7-14 days; the acute illness lasts for 2-4 wk. Clinical illnesses range from undifferentiated fever to the characteristic severe illness. Lassa fever is most often clinically severe in white persons. Onset is usually gradual, with increasing fever, headache, diffuse myalgia, and anorexia (Table 263-2). During the 1st wk, signs frequently include a sore throat, dysphagia, cough, oropharyngeal ulcers, nausea, vomiting, diarrhea, and pains in the chest and abdomen. Pleuritic chest pain may persist for 2-3 wk. In Argentine and Bolivian hemorrhagic fevers, and less frequently in Lassa fever, a petechial enanthem appears on the soft palate 3-5 days after onset and at about the same time on the trunk. The tourniquet test result may be positive. The clinical course of Venezuelan hemorrhagic fever has not been well described.
Table 263-2 CLINICAL STAGES OF LASSA FEVER
| STAGE | SYMPTOMS |
|---|---|
| 1 (days 1-3) | General weakness and malaise. High fever, >39°C, constant with peaks of 40-41°C |
| 2 (days 4-7) | Sore throat (with white exudative patches) very common; headache; back, chest, side, or abdominal pain; conjunctivitis; nausea and vomiting; diarrhea; productive cough; proteinuria; low blood pressure (systolic <100 mm Hg); anemia |
| 3 (after 7 days) | Facial edema; convulsions; mucosal bleeding (mouth, nose, eyes); internal bleeding; confusion or disorientation |
| 4 (after 14 days) | Coma and death |
From Richmond JK, Baglole DJ: Lassa fever: epidemiology, clinical features, and social consequences, Br Med J 327: 1271–1275, 2003.
In 35-50% of all patients, these diseases may become severe, with persistent high temperature, increasing toxicity, swelling of the face or neck, microscopic hematuria, and frank hemorrhages from the stomach, intestines, nose, gums, and uterus. A syndrome of hypovolemic shock is accompanied by pleural effusion and renal failure. Respiratory distress resulting from airway obstruction, pleural effusion, or congestive heart failure may occur. A total of 10-20% of patients experience late neurologic involvement, characterized by intention tremor of the tongue and associated speech abnormalities. In severe cases, there may be intention tremors of the extremities, seizures, and delirium. The cerebrospinal fluid is normal. In Lassa fever, nerve deafness occurs in early convalescence in 25% of cases. Prolonged convalescence is accompanied by alopecia and, in Argentine and Bolivian hemorrhagic fevers, by signs of autonomic nervous system lability, such as postural hypotension, spontaneous flushing or blanching of the skin, and intermittent diaphoresis.
Laboratory studies reveal marked leukopenia, mild to moderate thrombocytopenia, proteinuria, and, in Argentine hemorrhagic fever, moderate abnormalities in blood clotting, decreased fibrinogen, increased fibrinogen split products, and elevated serum transaminases. There is focal, often extensive eosinophilic necrosis of liver parenchyma, focal interstitial pneumonitis, focal necrosis of the distal and collecting tubules, and partial replacement of splenic follicles by amorphous eosinophilic material. Usually bleeding occurs by diapedesis with little inflammatory reaction. The mortality rate is 10-40%.
After an incubation period of 4-7 days, illness begins abruptly with severe frontal headache, malaise, drowsiness, lumbar myalgia, vomiting, nausea, and diarrhea. A maculopapular eruption begins 5-7 days later on the trunk and upper arms. It becomes generalized and often hemorrhagic and exfoliates during convalescence. The exanthem is accompanied by a dark red enanthem on the hard palate, conjunctivitis, and scrotal or labial edema. Gastrointestinal hemorrhage occurs as the severity of illness increases. Late in the illness, the patient may become tearfully depressed with marked hyperalgesia to tactile stimuli. In fatal cases, patients become hypotensive, restless, and confused and lapse into coma. Convalescent patients may experience alopecia and may have paresthesias of the back and trunk. There is a marked leukopenia with necrosis of granulocytes. Disseminated intravascular coagulopathy and thrombocytopenia are universal and correlate with severity of disease; there are moderate abnormalities in concentrations of clotting proteins and elevations of serum transaminases and amylase. The mortality rate of Marburg disease is 25-85%, and that of Ebola hemorrhagic fever 50-90%. High viral loads in acute-phase blood samples convey a poor prognosis.
In most cases, HFRS is characterized by fever, petechiae, mild hemorrhagic phenomena, and mild proteinuria, followed by relatively uneventful recovery. In 20% of recognized cases, the disease may progress through 4 distinct phases. The febrile phase is ushered in with fever, malaise, and facial and truncal flushing. It lasts 3-8 days and ends with thrombocytopenia, petechiae, and proteinuria. The hypotensive phase, of 1-3 days, follows defervescence. Loss of fluid from the intravascular compartment may result in marked hemoconcentration. Proteinuria and ecchymoses increase. The oliguric phase, usually 3-5 days in duration, is characterized by a low output of protein-rich urine, increasing nitrogen retention, nausea, vomiting, and dehydration. Confusion, extreme restlessness, and hypertension are common. The diuretic phase, which may last for days or weeks, usually initiates clinical improvement. The kidneys show little concentrating ability, and rapid loss of fluid may result in severe dehydration and shock. Potassium and sodium depletion may be severe. Fatal cases manifest as abundant protein-rich retroperitoneal edema and marked hemorrhagic necrosis of the renal medulla. The mortality rate is 5-10%.
Diagnosis of these viral hemorrhagic fevers depends on a high index of suspicion in endemic areas. In nonendemic areas, histories of recent travel, recent laboratory exposure, or exposure to an earlier case should evoke suspicion of a viral hemorrhagic fever.
In all viral hemorrhagic fevers, the viral agent circulates in the blood at least transiently during the early febrile stage. Togaviruses and bunyaviruses can be recovered from acute-phase serum samples by inoculation into tissue culture or living mosquitoes. Argentine, Bolivian, and Venezuelan hemorrhagic fever viruses can be isolated from acute-phase blood or throat washings by intracerebral inoculation into guinea pigs, infant hamsters, or infant mice. Lassa virus may be isolated from acute-phase blood or throat washings by inoculation into tissue cultures. For Marburg disease and Ebola hemorrhagic fever, acute-phase throat washings, blood, and urine may be inoculated into tissue culture, guinea pigs, or monkeys. The viruses are readily identified on electron microscopy, with a filamentous structure differentiating them from all other known agents. Specific complement-fixing and immunofluorescent antibodies appear during convalescence. The virus of HFRS is recovered from acute-phase serum or urine by inoculation into tissue culture. A variety of antibody tests using viral subunits is becoming available. Serologic diagnosis depends on demonstration of seroconversion or a fourfold or greater increase in immunoglobulin G antibody titer in acute and convalescent serum specimens collected 3-4 wk apart. Viral RNA may also be detected in blood or tissues with use of reverse transcriptase polymerase chain reaction analysis.
Handling blood and other biologic specimens is hazardous and must be performed by specially trained personnel. Blood and autopsy specimens should be placed in tightly sealed metal containers, wrapped in absorbent material inside a sealed plastic bag, and shipped on dry ice to laboratories with biocontainment safety level 4 facilities. Even routine hematologic and biochemical tests should be done with extreme caution.
Mild cases of hemorrhagic fever may be confused with almost any self-limited systemic bacterial or viral infection. More severe cases may suggest typhoid fever; epidemic, murine, or scrub typhus; leptospirosis; or a rickettsial spotted fever, for which effective chemotherapeutic agents are available. Many of these disorders may be acquired in geographic or ecologic locations endemic for a viral hemorrhagic fever.
Ribavirin administered intravenously is effective in reducing mortality in Lassa fever and HFRS. Further information and advice about management, control measures, diagnosis, and collection of biohazardous specimens can be obtained from Centers for Disease Control and Prevention, National Center for Infectious Diseases, Special Pathogens Branch, Atlanta, Georgia 30333 (404-639-1115).
The therapeutic principle involved in all of these diseases, especially HFRS, is the reversal of dehydration, hemoconcentration, renal failure, and protein, electrolyte, or blood losses. The contribution of disseminated intravascular coagulopathy to the hemorrhagic manifestations is unknown, and the management of hemorrhage should be individualized. Transfusions of fresh blood and platelets are frequently given. Good results have been reported in a few patients after the administration of clotting factor concentrates. The efficacy of corticosteroids, ε-aminocaproic acid, pressor amines, and α-adrenergic blocking agents has not been established. Sedatives should be selected with regard to the possibility of kidney or liver damage. The successful management of HFRS may require renal dialysis.
Although whole blood transfusions from Ebola virus–immune donors are thought to be therapeutic, studies in a monkey model were unable to confirm this outcome.
A live-attenuated vaccine (Candid-I) for Argentine hemorrhagic fever (Junin virus) is highly efficacious. A form of inactivated mouse brain vaccine is reported to be effective in preventing Omsk hemorrhagic fever. Inactivated RVF vaccines are widely used to protect domestic animals and laboratory workers. HFRS inactivated vaccine is licensed in Korea, and killed and live-attenuated vaccines are widely used in China. A vaccinia-vector glycoprotein vaccine provides protection against Lassa fever in monkeys. A single dose of a recombinant vesicular stomatitis virus vaccine containing surface glycoproteins from Ebola and Marburg viruses is effective in preventing virus hemorrhagic fevers due to several strains of filovirus in a monkey model.
Prevention of mosquito-borne and tick-borne infections includes use of repellents, wearing of tight-fitting clothing that fully covers the extremities, and careful examination of the skin after exposure with removal of any vectors found. Diseases transmitted from a rodent-infected environment can be prevented through methods of rodent control; elimination of refuse and breeding sites is particularly successful in urban and suburban areas.
Crimean-Congo hemorrhagic fever, Lassa fever, Marburg disease, and Ebola hemorrhagic fever may be transmitted in hospital settings. Patients should be isolated until they are virus-free or for 3 wk after illness. Patients’ urine, sputum, blood, clothing, and bedding should be disinfected. Disposable syringes and needles should be used. Prompt and strict enforcement of barrier nursing may be lifesaving. The mortality rate among medical workers contracting these diseases is 50%. A few entirely asymptomatic Ebola infections result in strong antibody production.
Al-Hazmi A, Al-Rajhi AA, Abboud EG, et al. Ocular complications of Rift Valley fever outbreak in Saudi Arabia. Ophthalmology. 2005;112:313-318.
Centers for Disease Control and Prevention. Update: management of patients with suspected viral hemorrhagic fever. MMWR Morb Mortal Wkly Rep. 1995;44:475-479.
Centers for Disease Control and Prevention. Update: outbreak of Rift Valley fever—Saudi Arabia, August-November 2000. MMWR Morb Mortal Wkly Rep. 2000;49:982-985.
Geisbert TW, Hensley LE, Jahrling PB, et al. Treatment of Ebola virus infection with a recombinant inhibitor of factor VII a tissue factor: a study in rhesus monkeys. Lancet. 2003;362:1953-1958.
Geisbert TW, Geisbert JB, Leung A, et al. Single injection vaccine protects nonhuman primates against Marburg virus and three species of Ebola virus. J Virol. 2009;83:7296-7304.
Huggins JW, Hsiang CM, Cosgriff TM, et al. Prospective, double-blind, concurrent, placebo-controlled trial of intravenous ribavirin therapy of hemorrhagic fever with renal syndrome. J Infect Dis. 1991;164:1119-1127.
Leroy EM, Baize S, Volchkov VE, et al. Human asymptomatic Ebola infection and strong inflammatory response. Lancet. 2000;355:2210-2215.
Leroy EM, Telfer P, Kumulungui B, et al. A serological survey of Ebola virus infection in central African nonhuman primates. J Infect Dis. 2004;190:1895-1899.
Leroy EM, Kumulungui B, Pourrut X, et al. Fruit bats as reservoirs of Ebola virus. Nature. 2005;438:575-576.
McCormick JB, King IJ, Webb PA, et al. Lassa fever: effective therapy with ribavirin. N Engl J Med. 1986;314:20-26.
McCormick JB, King IJ, Webb PA, et al. A case-control study of the clinical diagnosis and course of Lassa fever. J Infect Dis. 1987;155:445-455.
2009 Outbreak news: Ebola Reston in pigs and humans, Philippines. Weekly Epid Record. 2009;7:49-50.
Peters CJ. Marburg and Ebola—arming ourselves against the deadly filoviruses. N Engl J Med. 2005;352:2571-2574.
Pourrut X, Delicat A, Rollin PE, et al. Spatial and temporal patterns of Zaire Ebolavirus antibody prevalence in the possible reservoir bat species. J Infect Dis. 2007;196:S176-S183.
Richmond JK, Baglole DJ. Lassa fever: epidemiology, clinical features, and social consequences. Br Med J. 2003;327:1271-1275.
Smego RAJr, Sarwari AR, Siddiqui AR. Crimean-Congo hemorrhagic fever: prevention and control limitations in a resource-poor country. Clin Infect Dis. 2004;38:1731-1735.
Towner JS, Pourrut X, Albarino CG, et al. Marburg virus infection detected in a common African bat. PLoS ONE. 2007;22:e764.
Towner JS, Rollin PD, Bausch DG, et al. Rapid diagnosis of Ebola hemorrhagic fever by reverse transcription-PCR in an outbreak setting and assessment of patient viral load as a predictor of outcome. J Virol. 2004;78:4330-4341.