Dengue

Review

Dengue: an update
María G Guzmán and Gustavo Kourí

This review is an update of dengue and dengue

haemorrhagic fever (DHF) based on international 800 000
700 000 WPRO
and Cuban experience. We describe the virus SEARO
characteristics and risk factors for dengue and 600 000 Americas
DHF, and compare incidence and the case 500 000

Cases
fatality rates in endemic regions (southeast Asia, 400 000
western Pacific, and the Americas). The clinical
300 000
picture and the pathogenesis of the severe disease
are explained. We also discuss the viral, 200 000
individual, and environmental factors that 100 000
determine severe disease. Much more research is 0
necessary to clarify these mechanisms. Also 92 93 94 95 96 97 98 99 00
reviewed are methods for viral isolation and the 19 19 19 19 19 19 19 19 20
serological, immunohistochemical, and molecular Figure 1. World DF/ DHF reported cases 1992–2000 by WHO region.
methods applied in the diagnosis of the disease. SEARO=South East Asia, WPRO=Western Pacific.
We describe the status of vaccine development and
emphasise that the only alternative that we have Disease burden and global situation
today to control the disease is through control of Dengue fever (DF) and dengue haemorrhagic fever (DHF)
its vector Aedes aegypti. are increasingly important public health problems in the
Lancet Infectious Diseases 2001 2: 33–42 tropics and subtropics. Dengue has been recognised in over
100 countries and 2·5 billion people live in areas where
At the end of the last century, the world faced the dengue is endemic. Yearly, an estimated of 50–100 million
resurgence of many infectious diseases, dengue being one cases of DF and several hundred thousand cases of
of the most important in terms of morbidity and DHF occur, depending on epidemic activity. About
mortality. The dengue virus is transmitted to man by the 250 000–500 000 cases of DHF are officially notified annually;
bite of a domestic mosquito, Aedes aegypti being the however, the true incidence is not very well known.5–8 In
principal vector although some other species such as Aedes 1998, 1·2 million cases of dengue and DHF were reported to
albopictus are of importance. WHO, including 3442 deaths.7,8 Case fatality rates vary from
The disease has been described since 1779–1780; 0·5% to 3·5% in Asian countries.9 The disease is endemic in
however, there is evidences that a similar disease occurred the Americas, southeast Asia (SEAR), western Pacific (WPR),
earlier on several continents.1,2 Four viruses, dengue Africa, and the eastern Mediterranean, with the major disease
1 to 4, classified in an antigenic complex of the burden in the three first regions. Figure 1 shows the DF and
flavivirus genus, family flaviviridae, are the aetiological DHF reported cases by geographical regions from 1992 to
agents of this entity.3 These spherical agents of 2000 (data kindly provided by C Prasittisuk, WHO southeast
40–50 nm in diameter have a lipid envelope and a Asian regional office, K Palmer, WHO western Pacific
positive single stranded RNA. The viral genome of regional office, and J Arias, WHO American regional office).
approximately 11 kb in length encodes three structural During the past 8 years incidence of dengue has grown in the
proteins (capsid, C, membrane protein, M, envelope endemic areas, particularly the American region (AMR);
glycoprotein, E) and seven non-structural proteins however, in the past 3 years, the case fatality rate was higher
(NS1, NS2a, NS2b, NS3, NS4a, NS4b, and NS5). in southeast Asia and western Pacific regions (table 1). Much
The 5´and 3´non-coding regions are important for
Mar a G Guzm n is head of the Virology Department and director
regulating viral replication.4 The main biological the PAHO/WHO Collaborating Center for Viral Diseases, Pedro
properties of the viruses are located in the E protein, Kour Tropical Medicine Institute, Ciudad Habana, Cuba; and
including receptor binding, haemagglutination of Gustavo Kour is director of the Pedro Kour Tropical Medicine
erythrocytes, neutralising antibody induction, and Institute.
protective immune response.4 Correspondence: Professor Mar a G Guzm n, Virology Department
and PAHO/WHO Collaborating Center for Viral Diseases, Pedro
The spectrum of illness ranges from inapparent, mild Kour Tropical Medicine Institute, Autopista Novia del Mediod a,
disease to a severe and occasionally fatal haemorrhagic Km 6, PO Box Marianao 13, Ciudad Habana, Cuba.
clinical picture. Tel +53 7 220450; fax +53 7 246051; email [email protected]

THE LANCET Infectious Diseases Vol 2 January 2002 33

For personal use. Only reproduce with permission from The Lancet Publishing Group.
 Review Dengue

population. 17 years after the last report, dengue 3 was

Table 1. Case fatality rate (per 1000 dengue cases) by geographical
reintroduced in Central America and in less than 7 years it
region, 1998–2000
had spread to Caribbean and South American countries
1998 1999 2000
producing DF and DHF epidemics.12
SEAR 14·14 9·64 3·37
In general those factors that enhance the contact
WPR 4·14 1·83 NR
between vector and host favour an increase in dengue
AMR NR 0·32 0·17
NR=not reported.
transmission.13 However, it is not possible to exclude other
factors that could influence emergence of this disease such as
climate change and virus evolution. WHO has reported that
more research on DF and DHF risk factors is needed to a temperature rise of 1–20C could result in an increase of the
explain this observation. risk population by several hundred million, with 20 000–
30 000 more fatal cases annually.14
Why the recent recrudescence of dengue and DHF? Dengue transmission is a complex phenomena where
Two main factors are directly responsible for the the factors mentioned above are all involved; however, living
emergence and re-emergence of DF and DHF: increases in conditions15 and specifically poverty, social inequalities, and
the density and geographic distribution of the vector, illiteracy constitute the general background.15,16
and in the rate and geographic range of virus transmission.
The first factor is widely influenced by major global Clinical picture
demographic changes, particularly unprecedented global Most dengue infections are symptomless or very mild
population growth and unplanned urbanisation resulting characterised by undifferentiated fever with or without rash
in substandard housing, and inadequate water supply and mainly in infants and young children. Older children and
waste management systems. The epidemiological situation adults may develop a mild febrile syndrome or typical
is worsening by deterioration of the health systems and of DF consisting of high fever, severe headache, myalgia,
mosquito-control programmes in most dengue-endemic arthralgia, retro-orbital pain, and maculopapular rash.
countries.5,10,11 The American region is perhaps the best Signs of skin bleeding such as positive tourniquet test,
example: an eradication campaign started at the end of the petechiae, or ecchymosis are observed in some patients. DF
1940s and most countries became free of the vector; cases with bleeding complications such as epistaxis,
however, during the 1960s and the 1970s A aegypti gingival bleeding, gastrointestinal bleeding, haematuria,
reinfested Central and South America.5 Today, only a few and hypermenorrhea can be observed during some
countries are free of the vector. Figure 2 shows the current epidemics. Thrombocytopenia has been also reported in
and potential distribution of A aegypti in the world some cases.17,18
according to WHO. DF is a very incapacitating disease; however, its
The increase in air travel allows the movement of the prognosis is favourable.
different serotypes, strains, and even genotypes of virus Plasma leakage is the major pathophysiological
from one region to another. Individuals in viraemic phase feature observed in DHF and differentiates this from
are able to introduce a new virus into a susceptible typical DF. High fever, bleedings, moderate to marked

Figure 2. Current and potential distribution of Aedes aegypti, WHO 1998.

34 THE LANCET Infectious Diseases Vol 2 January 2002

For personal use. Only reproduce with permission from The Lancet Publishing Group.
 Dengue
Review

thrombocytopenia (below 100 000/mL) and haemo- Table 2. Main signs and symptoms observed in DHF fatal cases,
concentration (haematocrit increase by 20%) characterise Cuba 1981 and 1997
severe disease. Hepatomegaly has been an important sign in
Clinical manifestation 1981 1997
different settings.17–20
adults adults
Both DF and DHF begin with a sudden rise in
Fever 88 100
temperature; after 3–4 days, signs of haemorrhage such as
Vomiting 81 100
petechiae, ecchymosis, epistaxis, or gingival or gastrointestinal
bleeding are observed in DHF cases. Plasma leakage such as Hepatomegaly 35 66·6

pleural effusion, ascites, and hypoproteinaemia are common. Abdominal pain 58 83·3
Some patients deteriorate to circulatory failure, dengue shock Ascites 8 91·6
syndrome (DSS), presenting as rapid and weak pulse, narrow Pleural effusion 8 58·3
pulse pressure or hypotension, cold clammy skin, and altered Shock 100 100
mental status. Disease severity is classified as mild (grades I Haemorrhagic 65 100
and II) or severe (grades III and IV), the presence of shock manifestations
being the main difference.17,18 Petechiae 38 41·6
There is not a specific antiviral treatment but patients Haematemesis 35 58·3
usually recover after fluid and electrolyte supportive therapy, Melaena 4 25
particularly if early measures are applied. Early recognition of Vaginal bleeding 44 42·8
the warning signs of DHF (intense continuous abdominal Haemoconcentration 92 91·6
pain, persistent vomiting, and restlessness or lethargy) and Thrombocytopenia 71·8 83·3
early treatment are of utmost importance to reduce case Numbers given as %.
fatality rate.17,18,20
An iceberg characterises dengue virus infections. Most
cases are symptomless, followed, in increasing rarity, by One of the unusual manifestations of dengue infection is
undifferentiated fever, DF, and DHF. Studies of the 1997 the involvement of the central nervous system. Gubler et al27
DHF Cuban epidemic21 illustrate this fact (figure 3). concluded that neurologic disorders can occur in both DF
DHF has been primarily a disease of children.17,18 Reports and DHF. In DF, neurological symptoms range from
from 1975 to 1978 from areas where DHF is endemic revealed irritability and depression through mononeural palsies to
only eight of 629 patients in Indonesia and 18 of 694 in encephalitis with seizures and death. Encephalopathy in DHF
Thailand to be older than 15 years.19,22 However, the age could result from cerebral anoxia, oedema, intracanial
distribution of DHF cases has changed progressively, and is haemorrhage, and vessel occlusion. In general, encephalitic
different in the Americas to that observed in Asia. In the symptoms in DHF are attributable to liver failure and oedema
outbreaks in Cuba and Venezuela the disease occurred in all associated with leakage through the cerebral vasculature;
age groups, although about two-thirds of the fatalities were however, in DF, the pathogenesis of encephalopathy is less
among children.5,23,24 Similar observations have been made in clear. There is controversy over whether dengue viruses
Brazil and Puerto Rico.25,26 In general, an increase in the produce neurological disease as a non-specific complication
number of DHF adult cases has been observed during the or because of direct invasion of the brain in the manner of
1980s and the 1990s in countries such as Philippines and other flaviviruses such as Japanese encephalitis and St Louis
Malaysia. Table 2 illustrates some of the main signs and encephalitis. In a study of 378 Vietnamese patients with
symptoms observed in adult patients.24 suspected central nervous system infections, 4·2% were
infected with dengue viruses.28 Viruses were isolated or
detected by PCR in cerebrospinal fluid in some cases. With
the development of molecular diagnosis, dengue detection
from cerebrospinal fluid or brain has increased; however, the
12
question of contamination with blood has been raised. It is
fatalities
possible that haemorrhage or leakage through the blood-
brain barrier enables antibody and virus present in the blood
205 to move into the cerebrospinal fluid, or dengue virus could
DHF/DSS cases cross the blood-brain barrier and infect the cerebrospinal
fluid. Finally, dengue antigen has been detected by
immunoperoxidase stain in the brain of fatal cases.29 In
5208
DF/DHF cases
general, the pathological findings observed in the brain are
associated with cerebral oedema and haemorrhage without
evidences of encephalitis. More pathological studies are
17926 needed.
Dengue infections
Dengue diagnosis—still a need
Three factors have been fundamental in dengue diagnosis:
Figure 3. Reported and estimated DF/ DHF and dengue-2 infections development of ELISAs for dengue-specific IgM detection;
during the 1997 DHF Cuban epidemic.21 mosquito cell lines and monoclonal antibody development

THE LANCET Infectious Diseases Vol 2 January 2002 35

For personal use. Only reproduce with permission from The Lancet Publishing Group.
 Review Dengue

for viral isolation and identification; and, most recently, the recent development of protocols for dengue RNA
introduction of reverse transcriptase PCR for molecular quantification will allow improved study of the role of level
diagnosis and strain characterisation. These three methods of viraemia in severity of the disease.49
cover the serological, virological, and molecular diagnosis of Primary, secondary, and even tertiary dengue infections
dengue.30 can be observed taking into account the existence of four
Once an individual is infected, an incubation period of serotypes. During a primary infection, individuals develops
7–10 days occurs. 2 days before the onset of disease to 5–6 IgM after 5–6 days and IgG antibodies after 7–10 days. During
days after, virus could be recovered from blood. Serum a secondary infection high levels of IgG are detectable even
inoculation in mosquito cell lines such as A albopictus during the acute phase and they rise considerably over the
(C6-36) and Aedes pseudoscutellaris (AP61) are the most next 2 weeks. IgM levels are lower and in some cases absent
common method for virus isolation.30–32 However, during secondary infection. IgM antibodies suggest a recent
inoculation of samples directly into mosquitoes, specifically infection although they are still present after 2–3 months.
adult or larval inoculation of Toxorhynchites spp mosquitoes High titres of IgG are a criterion of secondary infection.17,18,50
(Tx amboinensis, Tx splendens) is the best isolation system in There are different methods to detect both IgM and IgG
terms of sensitivity.33,34 Unfortunately this last method is not immunoglobulins; however, ELISA is the most widely used
available in most endemic countries. In routine diagnosis, the in routine practice.50–52 The sensitivity of IgM ELISA ranges
C6-36 cell line has become most widely used. 7–10 days post- from 90 to 97% compared with the gold standard
inoculation in cell lines or 14 days post-inoculation in haemagglutination-inhibition test. Some false positive
mosquitoes, virus identification is done by immuno- reactions can be observed in less than 2% of cases and also a
fluorescence assay with serotype-specific monoclonal low or negative IgM reaction in secondary infections.50,53
antibodies.30,35,36 Viral isolation rates are up to 36% with IgM false positive reactions are illustrated with the data from
C6-36 cell lines or up to 80% by direct mosquito Cuban dengue surveillance in the period 1998–1999 where
inoculation.32,36 A rapid centrifugation assay for dengue virus no dengue circulation was detected (table 3).
improved the isolation rate, even in tissue samples.37 Commercial kits are available for serological diagnosis,
Several PCR protocols for dengue detection have been but they still need careful evaluation.53–55
described that vary in the RNA extraction methods, The capture ELISA for IgM detection is the most useful
genomic location of primers, specificity, sensitivity, and the serologic procedure currently available and it is widely
methods to detect PCR products and to determine the recommended for serological surveillance allowing health
serotype.38–41 Reverse transcriptase PCR has provided one of authorities to be alerted before there is an increase in
the most important steps in the molecular diagnosis of number of cases and severity of illness.17,56 One example of
dengue virus. A rapid assay was developed by Lanciotti this occurred in Havana, a city of two millions inhabitants,
et al,38 which allows the detection of virus in viraemic sera in September 2000. The national dengue surveillance system
with consensus primers located in the C and prM genes. A detected a positive IgM sample from a dengue suspected
second PCR with specific primers allows serotype case. Five more related cases were detected in the following
identification (DNA products of different sizes according to days. Rapid surveillance and established control actions
the dengue serotype are obtained). allowed this outbreak to be controlled in less than 3 months.
PCR has been applied to dengue diagnosis (with sera, About 135 DF confirmed cases were reported, dengue 3 and
tissue from fatal cases, mosquito pool, infected cell cultures, dengue 4 viruses being the aetiological agents. No DHF cases
and mosquito larvae), molecular surveillance, and genetic were observed. This epidemic was located in three health
strain characterisation. areas of Havana City. No cases were reported in the rest of
PCR combined with the nucleotide sequencing and the country and no DHF cases were observed (MG Guzman,
restriction enzyme analysis has become a powerful tool for unpublished observation).
dengue strain characterisation.42–47 Nucleotide sequencing PAHO/WHO guidelines for prevention and control of
studies have allowed classification of dengue viruses into dengue in the Americas commend the usefulness of
different genotypes according to their nucleotide sequence. epidemiological surveillance with laboratory support
Rico-Hesse42 studying the E/NS1 gene junction (including dengue IgM detection and viral isolation) for
demonstrated the presence of five genotypes for both rapid detection of dengue epidemics.17
dengue 1 and dengue 2 viruses. Others have obtained similar Antigen detection in tissues and serum samples
results by studying a fragment encoding aminoacids 29–94 complete the dengue diagnosis. Immunohistochemistry
in the E protein of 28 dengue 2 and aminoacids 28–87 in 35 with specific antibodies has allowed detection of dengue
dengue 1 isolates.43,44 Dengue 3 has been classified into four antigen in liver, spleen, lung, and lymph nodes from fatal
subtypes by sequencing the M and E structural genes of 23
geographically and temporally distinct strains, and dengue 4 Table 3. IgM false positive reaction, dengue surveillance, Cuba,
has been classified into two genotypes by studying the 1998–1999*.
complete E gene of 19 strains.44,45 A direct correlation Year False positive reactions/total (%)
between viraemia, disease severity, and explosiveness of 1998 26/4794 (0·54)
epidemic transmission has been reported previously. Illness 1999 9/10012 (0·08)
severity seems to correlate with the level of circulating virus. Total 35/14806 (0·23)
Viraemia levels should also correlate with mosquito
*Samples tested by IgM antibody capture ELISA
infection rates and thus epidemic transmission rates.48 The

36 THE LANCET Infectious Diseases Vol 2 January 2002

For personal use. Only reproduce with permission from The Lancet Publishing Group.
 Dengue
Review

cases.57 There are scarce reports of

•Age •Number of
antigen detection in serum, probably •Sex susceptible
because of the low sensitivity of the •Race •Vector high
employed system in the context of a •Nutritional Individual Epidemiological density
high number of secondary infections Status •Wide viral
risk risk
and the presence of immune •Secondary circulation
factors factors
complexes of virus and antibodies.58,59 infection •Hyperende-
Although dengue diagnosis has •Host micity
improved, we still need better tools for response
an early, rapid, specific, sensitive, and
non-expensive diagnosis.

What we know about DHF and

what we don’t Viral
For years, DHF pathogenesis has been factors
a controversial matter. Some workers
argued that secondary infection was
the main factor in the severity of this
disease, whereas others pointed to
•Strain virulence
viral virulence.60–65 Today, the majority
•Serotype
view is that secondary infection is the
main risk factor for DHF; however, Figure 4. Risk factors for DHF/DSS: an integral hypothesis.
other factor such as viral virulence and
host characteristics are also of utmost importance. other hand, when no neutralising antibodies are present,
DHF occurs as a consequence of a very complex heterotypic antibodies form complexes with dengue
mechanism where virus, host, and host immune response viruses, which infect mononuclear phagocytes with
interact to give the severe disease in 2–4% of individuals enhanced efficiency and as a consequence a higher number
with secondary infection.21 An integral hypothesis for the of cells are infected. This phenomenon has been called
development of DHF epidemics was published in 198765 antibody dependent enhancement (ADE).66,67 In an elegant
taking into account the international and Cuban experiences study, Kliks and colleagues68 demonstrated that pre-
on DHF (figure 4). The intersection of three groups of infection sera of children experiencing non-severe dengue
factors (host, viral, and epidemiological factors) determine 2 secondary infection had cross-reactive dengue-
the occurrence of a DHF epidemic. A high vector density, a neutralising antibodies, whereas sera from children with
high virus circulation, and a susceptible population (at risk severe dengue 2 secondary infection did not. Furthermore,
of secondary infection) are necessary to get a high number the sera of children with severe infection exhibited
of DHF cases.65 In general the epidemiological and viral significant enhancement activity, supporting the fact that
factors are the determinants for an epidemic of disease. heterologous antibodies are an important controlling
Individual risk factors such as sex, race, and chronic diseases determinant of infection enhancement. It is understood
are predisposing factors that make the disease more frequent today that dengue virions and IgG antibody to dengue virus
in a certain race or age group. However, the pre-existence of form virus-antibody complexes, whose binding to Fc␥R via
antibodies is the main individual risk factor, but not the only the Fc portion of IgG results in augmentation of dengue
one, for the occurrence of severe disease. Individual risk virus infection.66,69
factors are the ones that determine the appearance of the Epidemiological and serological studies done both in
disease in a particular person in a given population. The Thailand and Cuba are good examples of the importance of
presence or absence of these individual risk factors, in the secondary infections as risk factor of DHF. Since the first
matrix of epidemiological and viral factors, determines observations done by Halstead et al in 1970, DHF has been
whether or not all persons with a secondary type of infection present in situations where more than one serotype
present the clinical picture of DHF. circulates.17,21,62,69,70
Antibody levels may have a dominant role in driving Children are in higher risk of DHF than adults, as have
dengue infection to more or fewer infected cells. Epitopes been reported in Thailand and some others southeast Asian
present on E protein are able to induce neutralising countries.19,71 These studies showed that age-specific DHF
antibodies, both homologous and heterologous. People incidence was bimodal, with severe cases peaking at 7
infected with one serotype maintain a life-long protective months of age and again at 3–5 years of age (figure 5).61,72
immunity to infection by the homologous virus, although DHF or DSS occurred in infants who acquired maternal
protective immunity to infection with heterologous dengue antibody and subsequently experienced a dengue
serotypes is short-lived. It has been proposed that infection. In general, children less that 1 year of age were
neutralising antibodies down regulate the severity of the hospitalised almost exclusively during primary dengue
disease.66 During a secondary infection with a different infections. These infants were born to dengue immune
serotype, the presence of low amount of heterotypic mothers.73 On the other hand, children 3–5 years old have
neutralising antibodies could prevent severe disease; on the DHF during a secondary infection.

THE LANCET Infectious Diseases Vol 2 January 2002 37

For personal use. Only reproduce with permission from The Lancet Publishing Group.
 Review Dengue

enhances uptake of dengue particles by target cells through

18 increasing Fc cell receptors.81 Others cytokines such as IL-6,
16 IL-8, and IL-10 are also increased. A protein of 22–25 kDa has
14 been associated with the pathogenesis of DHF. This cytotoxic
12 factor is able to induce increased capillary permeability in
mice, is capable of reproducing in mice all the pathological
Cases

10
lesions that are seen in human beings, and has been detected
8
in sera of DHF patients.82
6 Dengue virus through an indirect more than a direct
4 mechanism could mediate endothelial cell activation. It has
2 been demonstrated that dengue infected peripheral blood
monocytes in ADE conditions generate soluble mediators that
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 activate endothelial cells through the enhanced expression of
adhesion molecules such as VCAM-1 and ICAM-1.83 At the
Age
same time, high levels of TNF␣ (induced by infected
Figure 5. Age distribution in children with DHF/DSS, Cuba 1981.23 monocytes and by T-cell activation or both) could be
responsible in part for transient vascular damage. The role of
It has been suggested that baseline microvascular TNF␣ in the pathogenesis of the disease is critical, and it
permeability in children is considerably greater than that of probably initiates several processes relating to plasma leakage
adults and this could explain, at least partly, why DHF is and haemorrhage.83 Avirutnan and colleagues84 have shown
more frequently observed in children.74 that infection of human endothelial cells with dengue virus
There seems to be no time limit to sensitisation after a induces the secretion of RANTES and IL-8, and the formation
primary dengue infection. The 1997 Cuban epidemic clearly of nonlytic complement complexes.
demonstrated that dengue 2 DHF still occurs 16–20 years after King et al demonstrated the potential role of mast
the primary dengue 1 infection.21,24 How the homotypic and cells/basophils in the pathogenesis of the disease.85 They
heterotypic antibodies raised after the primary dengue reported that mast cell/basophil KU812 cells are permissive
infection change in time in terms of neutralising titre, to dengue virus infection with the production of viral
immune enhancement ability, and avidity are important particles and vasoactive cytokine production. Furthermore,
matters that deserve careful study. they demonstrated that virus-antibody complexes are much
Besides secondary infection, chronic diseases such as more potent that virus alone in inducing cells activation.
bronchial asthma and diabetes have been suggested as risk Finally, a high release of IL-6 and IL-1␤ was also observed.
factors for DHF. Finally, whites have higher risk of developing Both cytokines could activate endothelial cells modulating
DHF than blacks. Dengue 2 virus is known to replicate to the expression of the adhesion molecules as well as altering
higher concentration in the peripheral blood cells of whites endothelial cell morphology.85
compared with those of blacks.24 Complement activation as a result of immune
Neutralising antibodies are key factors in the complexes (virus-antibody) or immune activation and
aetiopathogenesis of the disease; however, the cellular cytokine production could be also involved in the
immune response is also of importance. It has been mechanism of plasma leakage.
demonstrated that memory dengue T lymphocyte response Haemorrhage appears to be due to several factors such
after a primary infection includes both serotype-specific and as vasculopathy, thrombocytopenia, platelet dysfunction,
serotype-cross-reactive T lymphocytes.75 NS3 protein seems to and prothrombin-complex deficiency.86 However, the
be the major target for CD4+ and CD8+ T cells, although mechanisms that trigger events leading to haemorrhage have
some T cell epitopes have been recognised in other proteins not been well studied.
such as envelope and capsid.76,77 The magnitude of Development of antibodies potentially cross-reactive to
proliferation to heterologous dengue serotypes is variable plasminogen (due to a similarity in 20 aminoacid sequence of
depending on different factors such as the serotype causing dengue E glycoprotein and a family of clotting factors) could
the primary infection and the ethnicity of the individual.78 have a role in causing haemorrhage in DHF.86 The increased
These findings support the possibility that during a secondary destruction or decreased production of platelets could result
infection T cells become activated due to interactions with in thrombocytopenia.86 Virus-antibody complexes have been
infected monocytes. Recent observations suggest a massive detected on the platelet surface of DHF patients suggesting a
T-cell activation during DHF, which could explain partly if role for immune-mediated destruction of platelets.87 The
not totally the mechanism of plasma leakage through cytokine release of high levels of platelet-activating factor by
production and infected cell lysis by CD4+ and CD8+ monocytes with heterologous secondary infection may
dengue-specific T lymphocyte. Cytokines could be released explain the haemorrhage, given that platelet-activating factor
either directly from monocytes/macrophages as a result of may induce platelet consumption and augment adhesiveness
infection or after interactions between infected and immune of vascular endothelial cells resulting in thrombocytopenia.88
cells, or both.78,79 The presence of IgM antibodies in the sera of DHF cases that
Cytokines that may induce plasma leakage such as cross-reacted with platelets has been demonstrated.89 These
interferon ␥, interleukin (IL) 2, and tumour necrosis factor autoantibodies were able to cause platelet lysis and could be
(TNF) ␣ are increased in DHF cases.79,80 Also, interferon ␥ involved in the pathogenesis of the disease.

38 THE LANCET Infectious Diseases Vol 2 January 2002

For personal use. Only reproduce with permission from The Lancet Publishing Group.
 Dengue
Review

In spite of this knowledge, it is still uncertain what kind of with heterotypic antibodies. Recently, intraserotype re-
host and virus-specified factors determine why certain combination was demonstrated.98,99 The implications of these
individuals have only mild DF while others develop DHF. genetic changes for epidemics, severity of disease outcome,
Nucleotide and deduced aminoacid sequences of the different and virus emergence in new areas remain to be determined
serotypes have been studied to define possible molecular In addition to the effect of strain and serotype of virus in
markers of attenuation and virulence.90–92 Sequence determining the magnitude of an epidemic and severity of the
comparison of virulent and attenuated strains has disease, the serotype that produces the secondary infection
demonstrated that mutations could be important for and particularly the serotype sequence of infection are
attenuation.90 important. Although the four dengue viruses are able to
Two important genotypes for dengue 2 virus have been produce DHF cases, dengue 2 and dengue 3 are the most
identified, one, from southeast Asian origin, related to most frequently associated with the severe disease. Dengue 1
of the DHF epidemics in southeast Asia and the Americas; infection followed by dengue 2 has been associated with DHF
and the other, the American genotype, only related to DF epidemics, although in hyperendemic areas it is not easy to
epidemics in the American region.91,93,94 The 1981 Cuban define the virus producing the primary infection.100
epidemic has been considered one of the most severe DHF Finally, escape mutants could be another factor that adds
epidemics to date in the Americas.23 The genomic study of to the complexity of the DHF phenomenon.101
Cuban dengue 2 strains and the RNA from a liver sample of a Still a matter for discussion is the target cell for dengue
fatal case demonstrated their similarity with southeast Asian virus replication. Cells of the monocyte/macrophage lineage
strains, at least at the studied fragment.93,95 A similar have long been viewed as the primary target cells for dengue
epidemiological situation to that in Cuba was reported in viruses. Dengue antigen has been detected in Kuffer cells,
Iquitos, Peru. A dengue 1 epidemic occurred in 1990 followed alveolar macrophages, mononuclear phagocytes in the skin,
5 years later by a dengue 2 epidemic.94 No DHF cases were and circulating monocytes.102 The viruses have been isolated
observed. The dengue 1 viruses that affected Cuba in 1977 or detected in organs such as lymphoid organs, liver, spleen,
and Iquitos in 1990 were similar; by contrast, the 1981 Cuban kidney, and brain.24,102 Immune effector cells such as
dengue 2 strain belonged to an Asian genotype and the monocytes and T lymphocytes, and non-immune cells such
Peruvian dengue 2 virus belonged to the American as hepatocytes, endothelial cells, and brain cells have been
genotype.93,94 These results suggest that the Asian genotype reported as potential hosts. Langerhans and dendritic cells
possess a virulence determinant that is absent from viruses could be targets for dengue viruses and could play an
originating in the Americas. important part in the pathogenesis of dengue infection
At least two branches of the Asian genotype are circulating through an increase in virus load and cell activation.102,103
in the Americas, one related to old dengue 2 strains that have Dendritic cells may be required for establishing a primary
been isolated in Cuba in 1981, Venezuela in 1994, and Mexico immune response, producing virus particles, and TNF␣ and
in 1995, and the other isolated in Jamaica in 1981, Cuba in interferon ␣ production. They could be the early, primary
1997, and many other Latin-American countries and related target of dengue virus in natural infection and the vigour of
to the most recent Thailand dengue 2 strains. Both branches cell-mediated immunity could be modulated by the relative
are linked to DHF cases.91,93,95 Recently, some aminoacid presence or absence of interferon ␥ in the microenvironment
changes on M and E proteins of dengue 2 strains have been surrounding the virus-infected dendritic cells.104
related to DHF epidemics.91 These aminoacid changes are
present in both 1981 and 1997 dengue 2 Cuban strains (MG Dengue control—a challenge?
Guzman, unpublished observation). Today we are closer to getting a dengue vaccine, although
In a recent study, dengue 2 Thailand strains were problems remain to be solved. For decades, scientists have
classified in three subtypes, according the non-synonymous considered that a dengue vaccine should provide protective
aminoacid replacements and the authors proposed that immunity to the four serotypes to avoid the ADE
clinical severity depend both on the molecular structure of the phenomenon.105 However, we are facing a new challenge.
viruses and the serological response of patients.92 Subtype I The report of DHF 20 years after the primary infection and
included one strain isolated from a DSS case with a secondary the higher severity observed when secondary infection
serological response, subtype II included strains from DHF occurs after long interval compared with shorter interval
with secondary serological response and DF cases with a give a new dimension to this disease and reinforce the need
primary serological response, and subtype III included only to get a long-lasting immunity to the four viruses.70
DF cases with both primary or secondary serological The absence of an animal model, poor understanding of
response.
The relation of specific genotypes with severe disease has Strategies for a dengue vaccine
also being observed for dengue 3 virus.12,96 Conventional (inactivated and attenuated vaccines)
The significant genetic variation between genotypes could Non recombinant (structural and non-structural purified
be responsible of differences in virus interactions with proteins, synthetic peptides)
macrophages and suggest that certain strains are more Recombinant subunit (Escherichia coli, baculovirus, yeast)
virulent than others are. Morens and Halstead97 reported that Recombinant vector
virulence differences between dengue 2 strains could be Infectious cDNA clone
associated with subtle antigenic differences that affect DNA vaccines
the degree to which strains form immune complexes

THE LANCET Infectious Diseases Vol 2 January 2002 39

For personal use. Only reproduce with permission from The Lancet Publishing Group.
 Review Dengue

9 Halstead SB. Is there an inapparent dengue explosion? Lancet 1999;

Search strategy and selection criteria 353: 1100–01
10 Guzman MG, Kouri G, Bravo J. La emergencia de la fiebre
The references covered in this review come mostly from a hemorragica del dengue en las Americas: re-emergencia del dengue.
search of journals and books published in English and Rev Cub Med Trop 1999; 51: 5–13
Spanish in the past decades. The review includes the most 11 Guzman MG, Kouri G, Pelegrino JL. Enfermedades virales
emergentes. Rev Cub Med Trop 2001; 53: 5–15
advance knowledge on dengue today. 12 Figueroa R, Ramos C. Dengue virus (serotype 3) circulation in
endemic countries and its reappearance in America. Arch Med Res
2000; 31: 429–30
13 Monath TP. Epidemiology of yellow fever: current status and
the pathogenesis of the disease, and poor financial support speculations on future trends. In Saluzzo JF, Dodet B, eds. Factors
are other main problems. The panel shows some of the in the emergence of arboviruses diseases. Paris: Elsevier, 1997:
strategies used to produce a dengue vaccine.106–110 The first 143–56
14 WHO. The world wealth report. Life in the 21st century. A vision
human studies have been conducted for the attenuated for all. Geneva: WHO, 1998.
tetravalent vaccine and have demonstrated that it is safe and 15 Reiter P. Global warming and mosquito-borne disease in USA.
immunogenic in both children and adults. Results obtained Lancet 1996; 348: 622
with chimeric viruses in mice and non-human primates 16 Farmer P. Social inequalities and emerging infectious diseases.
Emerg Infect Dis 1996; 2: 259–69
suggest their possible usefulness in people. Independently of 17 Pan American Health Organization. Dengue and dengue
the applied strategy, careful human studies must be hemorrhagic fever in the Americas: guidelines for prevention
conducted to demonstrate the usefulness of the vaccine and control. Scientific publication no. 548. Washington: PAHO,
1994
candidates and its innocuousness. 18 WHO. Dengue haemorrhagic fever: diagnosis, treatment,
Until a dengue vaccine is available, vector control is the prevention and control. Geneva: WHO, 1997.
only way to diminish dengue transmission. But how can the 19 Nimmannitya S. Clinical spectrum and management of dengue
haemorrhagic fever. Southeast Asian J Trop Med Pub Health 1987;
mosquito be controlled when control programmes 20: 325–30
deteriorate, the economical situation in most of endemic 20 Martinez E. Dengue hemorragico en criancas. La Habana: Editorial
countries is unfavourable, and the population does not Jose Marti, 1992: 1–180
21 Guzman, MG, Kouri G, Valdes L, et al. Epidemiological studies on
recognise dengue as a priority? dengue in Santiago de Cuba, 1997. Am J Epidemiol 2000; 152:
Control of dengue transmission is harder today than 793–99.
ever before, but some principles remain fundamental if 22 Sumarmo, Wulur H, Jahja E, Gubler DJ, Suharyono W,
Sorensen K. Clinical observations on virologically confirmed fatal
control is to be achieved: political will (financial support, dengue infections in Jakarta, Indonesia. Bull World Health Organ
human resources), improvement of public health 1983; 61: 693–701
infrastructure and vector control programmes, intersectorial 23 Kouri GP, Guzman MG, Bravo JR, Triana C. Dengue hemorrhagic
fever/dengue shock syndrome: lessons from the Cuban epidemic,
coordination (partnerships among donors, the public 1981. Bull World Health Organ 1989; 67: 375–80.
sector, civil society, non-governmental organisations, and 24 Guzman MG, Alvarez M, Rodriguez R, et al. Fatal dengue
the private and commercial sectors), active community haemorrhagic fever in Cuba, 1997. Int J Infect Dis 1999; 3: 130–35.
participation, and reinforcement of health legislation. 25 Zagne SMO, Alves VGF, Nogueira RMR, Miagostovich MP,
Lampe E, Tavares W. Dengue hemorrhagic fever in the state of Rio
Ministries of health must direct control, and must establish de Janeiro, Brazil: a study of 56 confirmed cases. Trans R Soc Trop
epidemiological and entomological surveillance and Med Hyg 1994; 88: 677–79.
education campaigns for the community. It is fundamental 26 Rigau-Perez, the Puerto Rico Association of Epidemiologists.
Clinical manifestations of dengue hemorrhagic fever in Puerto Rico,
that the community recognises its responsibility in dengue 1990–1991. Rev Panam Salud Publica 1997; 1: 381–88.
control. 27 Gubler DJ, Kuno G, Waterman SH. Neurologic disorders associated
with dengue infection. Proceedings of the International Conference
on Dengue/Dengue Hemorrhagic Fever; Kuala Lumpur, Malaysia;
References 1983: 290–306
1 Rush AB. An account of the bilious remitting fever, as it appeared 28 Solomon T, Dung NM, Vaughn DW, et al. Neurological
in Philadelphia in the summer and autumn of the year 1780. manifestations of dengue infection. Lancet 2000; 355: 1053–59.
Medical inquiries and observations. Philadelphia: Prichard and 29 Nogueira RMR, Miagostovich MP, Cunha RV, et al. Fatal primary
Hall, 1789: 104–117 dengue infections in Brazil. Trans R Soc Trop Med Hyg 1999; 93: 418
2 Hirsch A. Dengue, a comparatively new disease: its symptoms.
Handbook of geographical and historical pathology. London: 30 Guzmán MG, Kourí G. Advances in dengue diagnosis.
Syndenham Society, 1883; 1: 55–81. Clin Diagnostic Immunol 1996; 3: 621–27.
3 Calisher CH, Karabatsos N, Dalrymple JD, et al. Antigenic 31 Race MW, Williams MC, Agostini CFM. Dengue in the Caribbean:
relationships between flaviviruses as determined by cross- virus isolation in a mosquito (Aedes pseudoscutellaris) cell line.
neutralization tests with polyclonal antisera. J Gen Virol 1989; 70: Trans R Soc Trop Med Hyg 1979; 73: 18–22
37–43 32 Kuno G, Gubler DJ, Velez M, Oliver A. Comparative sensitivity of
4 Chang GJ. Molecular biology of dengue viruses. In: Gubler DJ, three mosquito cell lines of dengue viruses. Bull World Health Organ
Kuno G, eds. Dengue and dengue haemorrhagic fever. New York: 1985; 63: 279–86.
CAB International, 1997: 175–98 33 Gubler D, Rosen L. A simple technique for demonstrating
5 Pinheiro FP, Corber SJ. Global situation of dengue and dengue transmission of dengue virus by mosquitoes without the use of
haemorrhagic fever and its emergence in the Americas. World vertebrate hosts. Am J Trop Med Hyg 1976; 25: 146–50.
Health Statist Quart 1997; 50: 161–68 34 Rosen L, Shroyer DA. Comparative susceptibility of five species of
6 Rigau-Pérez JG, Clark GG, Gubler DJ, Reiter P, Sanders EJ, Toxorhynchities mosquitoes to parenteral infection with dengue and
Vorndam AV. Dengue and dengue haemorrhagic fever. Lancet other flaviviruses. Am J Trop Med Hyg 1985; 34: 805–09.
1998; 352: 971–77 35 Henchal EA, McCown JM, Seguin MC, Gentry MK, Brandt WE.
7 WHO. Strengthening implementation of the global strategy for Rapid identification of dengue virus isolates by using monoclonal
dengue fever and dengue haemorrhagic fever, prevention and antibodies in an indirect immunofluorescence assay. Am J Trop Med
control. Report of the informal consultation, 18–20 October 1999. Hyg 1983; 32: 164–69.
Geneva: WHO, 1999. 36 Vorndam V, Kuno G. Laboratory diagnosis of dengue virus
8 WHO. Scientific working group on dengue. Meeting report, infections. In: Gubler DJ, Kuno G, eds. Dengue and dengue
Geneva, Switzerland, 3–5 April 2000. Geneva: WHO, 2000. haemorrhagic fever. New York: CAB International, 1997: 313–33.

40 THE LANCET Infectious Diseases Vol 2 January 2002

For personal use. Only reproduce with permission from The Lancet Publishing Group.
 Dengue
Review

37 Rodriguez R, Alvarez M, Guzman MG, Morier L, Kouri G. Isolation 60 Hammon WM. Dengue hemorrhagic fever—do we know its cause?
of dengue 2 in C636/HT cells by rapid centrifugation/shell vial Am J Trop Med Hyg 1973; 22: 82–90.
assay. Comparison with conventional virus isolation method. J Clin 61 Halstead SB. Observations related to pathogenesis of dengue
Microbiol 2000; 38: 3508–10. hemorrhagic fever.VI: hypotheses and discussion. Yale J Biol Med
38 Lanciotti RS, Calisher CH, Gubler DJ, Chang GJ, Vorndam AV. 1970; 42: 350–62.
Rapid detection and typing of dengue viruses from clinical samples 62 Sangkawibha N, Rojanasuphot S, Ahandrik S, et al. Risk factors in
using reverse transcriptase-polymerase chain reaction. J Clin dengue shock syndrome: a prospective epidemiologic study in
Microbiol 1992; 30: 545–51. Rayong, Thailand. Am J Epidemiol 1984; 120: 653–69.
39 Morita K, Meomoto T, Honda D, et al. Rapid detection of virus 63 Rosen L. Disease exacerbation caused by sequential dengue
genome from imported dengue fever and dengue hemorrhagic fever infections: myth or reality? Rev Infect Dis 1989; 11: S840–42.
patients by direct polymerase chain reaction. J Med Virol 1994; 44: 64 Halstead SB. Antibody, macrophages, dengue virus infection,
54–58. shock, and hemorrhage: a pathogenic cascade. Rev Infect Dis 1989;
40 Harris E, Roberts G, Smith L, et al. Typing of dengue virus in 11(suppl 4): S830–39.
clinical specimens and mosquitoes by single-tube multiplex reverse 65 Kouri GP, Guzman MG, Bravo JR. Why dengue haemorrhagic fever
transcriptase PCR. J Clin Microbiol 1998; 36: 2634–39. in Cuba? 2: an integral analysis. Trans R Soc Trop Med Hyg 1987; 81:
41 Rosario D, Alvarez M, Díaz J, et al. Rapid detection and typing 821–23.
of dengue viruses from clinical samples using reverse 66 Halstead S. Pathophysiology and pathogenesis of dengue
transcriptase-polymerase chain reaction. Pan Am J Public Health hemorrhagic fever. In Thongcharoen P, ed. Monograph on
1998; 4: 1–5. dengue/dengue haemorrhagic fever. WHO Regional Publication,
42 Rico-Hesse R. Molecular evolution and distribution of dengue SEARO no 22, 1993: 80–103.
viruses type 1 and 2 in nature. Virology 1990; 147: 479–93. 67 Halstead SB, Nimmannitya S, Yamarat C, Russell PK.
43 Deubel V, Nogueira R, Drouet MT, Zeller H, Reynes JM, Ha DQ. Haemorrhagic fever in Thailand: newer knowledge regarding
Direct sequencing of genomic cDNA fragments amplified by the etiology. J Med Sci Biol 1967; 17: 96–103.
polymerase chain reaction for molecular epidemiology of dengue-2 68 Kliks SC, Nisalak A, Brandt WE, Wahl L, Burke DS. Antibody-
viruses. Arch Virol 1993; 129: 197–210. dependent enhancement of dengue virus growth in human
44 Chungue E, Cassar O, Drouet MT, et al. Molecular epidemiology of monocytes as a risk factor for dengue hemorrhagic fever. Am J Trop
dengue-1 and dengue-4 viruses. J Gen Virol 1995; 76: 1877–84. Med Hyg 1989; 40: 444–51.
45 Lanciotti RS, Gubler DJ, Trent DW. Molecular evolution and 69 Morens DM. Antibody-dependent enhancement of infection and
phylogeny of dengue-4 viruses. J Gen Virol 1997; 78: 2279–86. the pathogenesis of viral disease. Clin Inf Dis 1994; 19: 500–12.
46 Balmaseda A, Sandoval E, Perez L, Gutierrez CM, Harris E. 70 Guzman MG, Kouri, Bravo J, Soler M, Vazquez S, Morier L.
Application of molecular typing techniques in the 1998 dengue Dengue hemorrhagic in Cuba, 1981: a retrospective
epidemic in Nicaragua. Am J Trop Med Hyg 1999; 61: 893–97. seroepidemiologic study. Am J Trop Med Hyg 1990; 42: 179–84.
47 Tolou H, Coussinier-Paris P, Mercier V, et al. Complete genomic 71 Jatanasen S, Thongcharoen P. Dengue haemorrhagic fever in south-
sequence of a dengue type 2 virus from the French West Indies. east Asian countries. In Thongcharoen P, ed. Monograph on
Biochem Bioph Res Comm 2000; 277: 89–92. dengue/dengue haemorrhagic fever. WHO Regional Publication,
48 Gubler DJ, Suharyono W, Tan R, Abidin M, Sie A. Viraemia in SEARO no 22, 1993: 23–30.
patients with naturally acquired dengue infection. Bull World 72 Guzman MG, Kouri G, Morier L, Fernandez A. A Study of fatal
Health Organ 1981; 59: 623–30. haemorrhagic dengue cases in Cuba 1981. Bull PAHO 1984; 18:
49 Sudiro TM, Zivny J, Ishiko H, et al. Analysis of plasma viral RNA 213–20.
levels during acute dengue virus infection using quantitative 73 Kliks SC, Nimmanitya S, Nisalak A, Burke DS.Evidence that
competitor reverse transcription-polymerase chain reaction. J Med maternal dengue antibodies are important in the development of
Virol 2001; 63: 29–34. dengue haemorrhagic fever in infants. Am J Trop Med Hyg 1988; 38:
50 Kuno G, Gomez I, Gubler DJ. An ELISA procedure for the 411–19.
diagnosis of dengue infections. J Virol Methods 1991; 33: 101–13.
74 Gamble J, Bethell D, Day NPJ, et al. Age-related changes in
51 Miagostovich MP, Nogueira RMR, Santos FB, Schatzmayr HG, microvascular permeability: a significant factor in the susceptibility
Araujo ESM, Vorndam V. Evaluation of an IgG enzyme-linked of children to shock? Clinical Science 2000; 98: 211–16.
immunosorbent assay for dengue diagnosis. J Clin Virol 1999; 14:
183–89. 75 Kurane I, Ennis FA. Cytokines in dengue virus infections: role of
cytokines in the pathogenesis of dengue hemorrhagic fever. Semin
52 Nawa M, Takasaki T, Yamada KI, Akatsuka T, Kurane I. Virology 1994; 5: 443–48.
Development of dengue IgM-capture enzyme-linked
immunosorbent assay with higher sensitivity using monoclonal 76 Kurane I, Zeng L, Brinton MA, Ennis FA. Definition of an epitope
detection antibody. J Virol Methods 2001; 92: 65–72. on NS3 recognized by human CD4+ cytotoxic T lymphocyte clones
cross-reactive for dengue types 2, 3 and 4. Virology 1998; 240:
53 Jelinek T, Wastlhuber J, Proll S, Schattenkirchner M, Loscher T.
169–74.
Influence of rheumatoid factor on the specificity of a rapid
immunochromatographic test for diagnosing dengue infection. 77 Spaulding AC, Kurane I, Ennis FA, Rothman AL. Analysis of
Eur J Clin Microbiol Infect Dis 2000; 19: 555–56. murine CD8+ T-cell clones specific for the dengue virus NS3
54 Cussubo AJ, Vaughn DW, Nisalak A, et al. Comparison of protein: flavivirus cross-reactivity and influence of infecting
PanBio dengue Duo IgM and IgG capture ELISA and Venture serotype. J Virol 1999; 73: 398–403.
technologies Dengue IgM and IgG Dot Blot. J Clin Virol 2000; 78 Rothman AL, Ennis FA. Toga/flaviviruses: immunopathology. In:
16: 135–144. Cunningham MW, Fujinami RS, eds. Effects of microbes on the
55 Chakravarti A, Gur R, Berry N, Mathur MD. Evaluation of three immune system. Philadelphia: Lippincott Williams & Wilkins,
commercially available kits for serological diagnosis of dengue 1999: 473–87.
haemorrhagic fever. Diagnostic Microbiol Infect Dis 2000; 36: 79 Hober D, Nguyen TL, Shen L, et al. Tumor necrosis factor alpha
273–74. levels in plasma and whole-blood culture in dengue infected
56 Gubler DJ. Surveillance for dengue and dengue hemorrhagic fever. patients: relationship between virus detection and pre-existing
Bull PAHO 1989; 23: 397–404. specific antibodies. J Med Virol 1998, 54: 210–18.
57 Pelegrino JL, Arteaga E, Rodriguez AJ, Gonzalez E, 80 Green S, Pichyangkul S, Vaughn DW, et al. Early CD69 expression
Frontela MC, Guzman MG. Normalizacion de tecnicas on peripheral blood lymphocytes from children with dengue
inmunohistoquimicas para la detección de antigenos del virus hemorrhagic fever. J Infect Dis 1999; 180: 1429–35.
dengue en tejidos embebidos en parafina. Rev Cubana Med Trop 81 Kontny U, Kurane I, Ennis FA. Gamma interferon augments FC_
1997; 49: 100–07. receptor-mediated dengue virus infection of human monocytic
58 Malergue F, Chungue E. Rapid and sensitive streptavidin-biotin cells. J Virol 1988; 62: 3928–33.
amplified fluorogenic enzyme-linked immunosorbent-assay for 82 Mukerjee R, Chatuverdi UC, Dhawan R. Dengue virus-induced
direct detection and identification of dengue viral antigens in human cytotoxic factor: production by peripheral blood leucocytes
serum. J Med Virol 1995; 47: 43–47. in vitro. Clin Exp Immunol 1995; 102: 262–67.
59 Young PR, Hilditch PA, Bletchly C, Halloran W. An antigen 83 Anderson R, Wang S, Osiowy C, Issekutz AC. Activation of
capture enzyme-linked immunosorbent assay reveals high levels of endothelial cells via antibody-enhanced dengue virus infection of
the dengue virus protein NS1 in the sera of infected patients. J Clin peripheral blood monocytes. J Virol 1997; 71: 4226–32.
Microbiol 2000; 38: 1053–57. 84 Avirutnan P, Malasit P, Seliger B, Bhakdi S, Husmann M. Dengue

THE LANCET Infectious Diseases Vol 2 January 2002 41

For personal use. Only reproduce with permission from The Lancet Publishing Group.
 Review Dengue

virus infection of human endothelial cells leads to chemokine 97 Morens DM, Halstead SB. Disease severity-related antigenic
production, complement activation and apoptosis. J Immunol differences in dengue 2 strains detected by dengue 4 monoclonal
1998; 161: 6338–46. antibodies. J Med Virol 1987; 22: 169–74.
85 King CA, Marshall JS, Alshurafa H, Anderson R. Release of 98 Worobey M, Rambaut A, Holmes EC. Widespread intra-serotype
vasoactive cytokines by antibody-enhanced dengue virus infection recombination in natural populations of dengue virus. Proc Natl
of a human mast cell/basophil line. J Virol 2000; 74: 7146–50. Acad Sci USA 1999; 96: 7352–57
86 Chungue E, Poli L, Roche C, Gestas P, Glaziou P, Markoff LJ. 99 Holmes E, Burch SS. The causes and consequences of genetic
Correlation between detection of plasminogen cross-reactive variation in dengue virus. Trends Microbiol 2000; 8: 74–77.
antibodies and hemorrhage in dengue virus infection. J Infect Dis 100 Halstead SB. Pathogenesis of dengue: challenges to molecular
1994; 170: 1304–07. biology. Science 1988; 239: 476–81.
87 Wang S, He R, Patarapotikul J, Innis BL, Anderson R. Antibody- 100 Guzman MG, Kouri G, Halstead S. Do escape mutants explain
enhanced binding of dengue-2 virus to human platelets. Virology rapid increases in dengue case-fatality rates within epidemics?
1995; 213: 254–57. Lancet 2000; 355: 1902–03
88 Yang KD, Wang CL, Shaio MF. Production of cytokines and 102 Bhamarapravati N. Pathology in dengue infections. In: Gubler DJ,
platelet activating factor in secondary dengue virus infections. Kuno G, eds. Dengue and dengue haemorrhagic fever. New York:
J Infect Dis 1995; 172: 604. CAB International, 1997: 115–132.
89 Lin CF, Lei HY, Liu CC, et al. Generation of IgM anti-platelet 103 Wu SJ, Grouard-Vogel G, Sun W, et al. Human skin Langerhans
autoantibody in dengue patients. J Med Virol. 2001 ;63: 143–49. cells are target of dengue virus infection. Nat Med 2000; 6:
90 Puri B, Nelson WM, Henchal EA, et al. Molecular analysis of 816–820.
dengue virus attenuation after serial passage in primary dog 104 Libraty DH, Pichyangkul S, Ajariyakhajorn C, Endy TP, Ennis FA.
kidney. J Gen Virol 1997; 78: 2287–29. Human dendritic cells are activated by dengue virus infection:
91 Leitmeyer KC, Vaughn DW, Watts DM, et al. Dengue virus enhancement by gamma interferon and implications for disease
structural differences that correlate with pathogenesis. J Virol 1999; pathogenesis. J Virol 2001; 75: 3501–08.
73: 4738–47. 105 Chambers TJ, Tsai TF, Pervikov Y, Monath TP. Vaccine
92 Pandey BD, Igarashi A. Severity-related molecular differences development against dengue and Japanese encephalitis: report of a
among nineteen strains of dengue type 2 viruses. Microbiol World Health Organization meeting. Vaccine 1997; 15: 1494–1502.
Immunol 2000; 44: 179–88. 106 Bhamarapravati N, Sutee Y. Live attenuated tetravalent dengue
93 Guzman MG, Deubel V, Pelegrino JL, et al. Partial nucleotide and vaccine. Vaccine 2000; 18: 44–47.
amino acid sequences of the envelope and the 107 Alvarez M, Guzman MG, Pupo M, Morier L, Bravo J,
envelope/nonstructural protein-1 gene junction of four dengue-2 Rodriguez R. Study of biologic attributes of Cuban dengue 2 virus
virus strains isolated during the 1981 Cuban epidemic. Am J Trop after serial passage in primary dog kidney cells. Int J infect Dis
Med Hyg 1995; 52: 241–46. 2001; 5: 35–39.
94 Watts DM, Porter KR, Putvatana P, et al. Failure of secondary 108 Guirakhoo F, Weltzin R, Chambers TJ, et al. Recombinant
infection with American genotype dengue 2 to cause dengue chimeric yellow fever–dengue type 2 virus is immunogenic and
haemorrhagic fever. Lancet 1999; 354: 1431–33. protective in nonhuman primates. J Virol 2000; 74: 5477–85.
95 Sariol C, Pelegrino JL, Martinez A, Arteaga E, Kouri G, 109 Kochel TJ, Raviprakash K, Hayes CG, et al. A dengue virus
Guzman MG .Detection and genetic relationship of dengue virus serotype 1 DNA vaccine induces virus neutralizing anibodies and
sequences in seventeen-year old paraffin embedded samples from provides protection from viral challenge in Aotus monkeys.
Cuba. Am J Trop Med Hyg 1999; 61: 994–1000. Vaccine 2000; 18: 3166–73.
96 Briseño B, Gomez H, Argott E, et al. Potential risk for dengue 110 Raviprakash K, Porter KR, Kochel TJ, et al. Dengue virus type 1
hemorrhagic fever: the isolation of serotype dengue-3 in Mexico. DNA vaccine induces protective immune responses in rhesus
Emerg Infect Dis 1996; 2: 133–35. macaques. J Gen Virol 2000; 81: 1659–67.
A list of references for further reading is available on The Lancet Infectious Diseases website, http://infection.thelancet.com

42 THE LANCET Infectious Diseases Vol 2 January 2002

For personal use. Only reproduce with permission from The Lancet Publishing Group.