BMJ | 21 noveMBer 2009 | voluMe 339 1189

CliniCal ReviewFor the full versions of these articles see bmj.com

Dengue has been reported in almost 70 countries, with about five million cases reported between 2000 and 2007 (figs 1 and 2).1 In 2002, at the peak of the cur-rent pandemic, over 1.2 million cases were reported worldwide. Until the end of the 1980s South East Asia and the Western Pacific region were the regions most affected by dengue. In recent years, however, the inci-dence of dengue has increased in the Central and South America region, which now accounts for 70% of all cases reported worldwide. Mortality from dengue has varied greatly across countries, but the World Health Organization estimates that about 22 000 deaths are associated with dengue every year.1 2 Despite efforts to control the main vector (the mosquito Aedes aegypti), the incidence of the disease has not shown any tendency to decrease.

Dengue is predominantly a childhood disease in South East Asia but an adult disease in most North and South American countries.3 4 However, since 2007,

Brazil (the country responsible for more than 70% of cases of dengue in the Americas) has experienced an unexpected increase in the incidence of the disease in people aged under 15 years.4 5

This article reviews the diagnosis and treatment of dengue and discusses the importance of the organisa-tion of healthcare systems in reducing deaths from den-gue in areas where the disease is endemic.

what causes dengue and how is it transmitted? The dengue virus, a single stranded RNA virus belonging to the Flaviridae family, has been classi-fied into four serotypes, DENV-1, DENV-2, DENV-3, and DENV-4, which are genetically and antigenically different; infection with one serotype produces life-long immunity only to that particular serotype.6 The main vector of the dengue virus is the mosquito Aedes aegypti. The disease is more readily transmitted in urban environments.

Diagnosis and management of dengueMaria Glória Teixeira, Mauricio L Barreto

Instituto de Saúde Coletiva, Federal University of Bahia, Rua Basílio da Gama s/n, Canela, 40110-040 Salvador, Bahia, BrazilCorrespondence to: M G Teixeira magloria@ufba.br

Cite this as: BMJ 2009;339:b4338doi: 10.1136/bmj.b4338

SummaRy pointS Dengue affects over 70 countries in four continents and results in about 22 000 deaths annuallyPrompt, adequate clinical management reduces deaths from dengue haemorrhagic feverDiagnostic testing for the detection of the NS1 antigen facilitates the diagnosis of dengue in patients with feverDuring an epidemic, additional resources and procedures are needed to enable people with fever to access health care easily, otherwise the care of routine patients will be affected

Box 1 | WHO’s case definitions for dengue fever9

Probable caseAn acute febrile illness with two or more of the following manifestations:

Headache, retro-orbital pain, myalgia, arthralgia, rash, haemorrhagic manifestations, •leucopenia; plus Supportive serology test results (a reciprocal haemagglutination-inhibition antibody titre ≥1280, a comparable IgG enzyme linked immunosorbent assay (ELISA) titre or a positive IgM antibody test on a late acute or convalescent-phase serum specimen); or Occurrence at the same location and time as other confirmed cases of dengue fever

Confirmed caseA case confirmed by laboratory criteria: •Isolation of the dengue virus from serum or autopsy samples Demonstration of a fourfold or greater change in reciprocal IgG or IgM antibody titres to one or more dengue virus antigens in paired serum samples Demonstration of dengue virus antigen in autopsy tissue, serum, or cerebrospinal fluid samples by immunohistochemistry, immunofluorescence, or ELISA Detection of dengue virus genomic sequences in autopsy tissue serum or cerebrospinal fluid samples by polymerase chain reaction

Box 2 | WHO’s case definition for dengue haemorrhagic fever9

The following criteria must all be present: Fever (or history of acute fever) lasting two to seven days, •occasionally biphasicHaemorrhagic tendencies (positive tourniquet test*; •petechiae; ecchymoses or purpura; bleeding from the mucosa, gastrointestinal tract, injection sites, or other locations; haematemesis or melena); Thrombocytopenia • (≤100 000×106 cells/l).† Evidence of plasma leakage caused by increased vascular permeability and manifested by at least one of the following: a rise in the packed cell volume ≥20% above average for age, sex, and population; a drop of ≥20% in the packed cell volume after volume replacement treatment; signs of plasma leakage such as pleural effusion, ascites, and hypoproteinaemia

*Performed by inflating a blood pressure cuff on the upper arm to a point midway between the systolic and diastolic pressures for five minutes. A test is considered positive when ≥20 petechiae per 2.5 cm area square are observed. The test may be negative or mildly positive during the phase of profound shock. It usually becomes positive, sometimes strongly positive, if the test is conducted after recovery from shock†This value represents a direct count using a phase-contrast microscope (normal is 200 000-500 000×106/l). For outpatients, an approximate count from a peripheral blood smear is acceptable. In normal people, 4-10 platelets per oil-immersion field (100×; the average of the readings from 10 oil-immersion fields is recommended) indicates an adequate platelet count. An average of ≤3 platelets per oil-immersion field is considered low (that is, <100 000×106/l).

1190 BMJ | 21 noveMBer 2009 | voluMe 339

CliniCal Review

what are the clinical features of dengue? Dengue virus infection may be symptomless, particu-larly in children, or may present as acute fever that is self limiting in most cases. It is often confused with other viral or bacterial illnesses. Acute infection may progress to severe forms of the disease and eventually death. A seminal cohort study of children in Thai-land (1962-1964) showed that dengue haemorrhagic fever (one form of severe dengue) is characterised by extraordinary third space fluid accumulation from leakage of plasma into the extravascular space, which leads to haemoconcentration, hypoalbuminaemia, and cavitary infarction.7 This study was essential in the development of classification of the clinical forms of dengue and guidelines for treating the disease.8 9

Dengue is classified as dengue fever, dengue haemor-rhagic fever, or dengue shock syndrome (boxes 1, 2, and 3), depending on its severity and presenting features.

Dengue fever is an acute febrile illness with two or more of the following manifestations: headache, retro-orbital pain, myalgia, arthralgia, rash, leucopenia, and mild haemorrhagic symptoms. Warning signals—such as spontaneous or provoked bleeding, vomiting, intense abdominal pain, painful hepatomegaly, breathing dis-comfort, lethargy, and cavitary infarction (pleural, pericardial, ascites)—usually precede the severe mani-festations of dengue (box 4). These warning signals, which appear between the third and seventh days after the onset of acute dengue fever (usually between the fourth and fifth days) when fever subsides, are clinically important as the patient is at increased risk of develop-ing dengue hemorrhagic fever (box 2). The additional signs of circulatory failure indicate dengue shock syn-drome (box 3).9

Dengue hemorrhagic fever presents in a similar fashion both in children and adults, with slight differences in the frequency of manifestations. Whereas petechiae, melena, headache, retro-orbital pain, myalgia, nausea, and vomit-ing are more common in adults, epistaxis, oliguria, and hepatomegaly are more frequent in children.10

which haematological and biochemical abnormalities occur in dengue?Patient with dengue fever usually present with leuco-penia, with or without lymphocytosis and lymphocyte atypia. If the clinical condition develops into dengue haemorrhagic fever, a gradual increase in packed cell volume occurs, accompanied by thrombocytopenia (platelet counts <100 000×106/l), increase in pro-thrombin, partial thromboplastin and thrombin times, and reduction in fibrinogen, factor VIII, factor XII, antithrombin, antiplasmin, and serum albumin. Albu-minuria may be observed.9 11 Hepatic manifestations such as hepatomegaly and increased serum levels of aspartate aminotransferase and alanine aminotrans-ferase are often observed in dengue fever and dengue haemorrhagic fever. Increased serum levels of bilirubin, alkaline phosphatase, and γ-glutamyltransferase are observed in variable frequency.12-14

neurological manifestations in dengueNeurological manifestations have been reported in prospective and retrospective clinical studies that used different diagnosis variables; the findings are not stand-ardised. Reported neurological manifestations include altered consciousness, lethargy, somnolence, and, more rarely, agitation and generalised seizures. Focal upper neurone signs, extrapyramidal features, and transverse myelitis have been reported.15 Normal or moderately raised pressure and lymphocytosis (5-500×106 cells/l) in the cerebrospinal fluids can be observed.15 16

making a definitive diagnosisDengue virus may be isolated in cell culture or detected by using reverse transcription polymerase chain reac-tion from the blood of febrile patients in the acute stage

Area infested by Aedes aegyptiRisk of dengue transmission

Presence of dengue worldwide and areas infested by the main vector, Aedes aegypti

Box 3 | WHO’s case definition for dengue shock syndrome9 All of the criteria for dengue haemorrhagic fever (box 2) must be present, plus evidence of circulatory failure manifested by:

Rapid and weak pulse and narrow pulse pressure (<20 mm •Hg (2.7 kPa)) orHypotension for age, and cold, clammy skin and •restlessness

Box 4 | Warning signals* of dengue haemorrhagic fever9

Sudden disappearance of feverIntense and continuous abdominal painPersistent vomitingOrthostatic hypotensionLow blood pressure Painful hepatomegalyMajor bleeding (haematemesis and/or melena)Cold extremities, cyanosisRapid and weak pulseAgitation or lethargy, or bothDecreased diuresisSudden decrease of body temperature or hypothermiaSudden increase in packed cell volume Respiratory distress* Patients with one or more warning signs and evidence of haemoconcentration and thrombocytopenia should be rehydrated and remain under medical observation until clinical remission (see table)

BMJ | 21 noveMBer 2009 | voluMe 339 1191

CliniCal Review

of the disease. Recently, the development of tests to detect the non-structural protein NS1 (an antigen local-ised on the surface of cells infected with dengue virus that is common to the four serotypes and is detectable between the first and ninth days after the appearance of fever) has improved diagnosis of this viral infection. The sensitivity and specificity of this method are high,17 costs are reasonable, and results are almost instantane-ous, which makes the tests useful to physicians attending patients with fever in areas where dengue is endemic or patients who have recently visited such areas. Tests for the detection of IgM (Mac-ELISA) are cheap and easily performed in clinical laboratories, and their sen-sitivity and specificity are over 93%.18 19 However, IgM antibodies reach their peak only after the sixth day after the onset of symptoms, which means that, although the results of these tests are invaluable for epidemiological surveillance systems, they are not helpful for the clini-cal diagnosis of dengue during the most critical phase of the disease.9 17 20

what predisposes a patient to developing dengue haemorrhagic fever?A controversial theory based on epidemiological and bio-logical evidence suggests that the great majority of cases of dengue haemorrhagic fever occur after a secondary infection by a different viral serotype, which produces the phenomenon of antibody dependent enhancement (the infecting virus forms a complex interaction with non-neutralising antibodies, thus enhancing phagocyto-sis by mononuclear cells). Sequential circulation of the four serotypes, irrespective of the order, establishes a complex interaction of heterotypic antibodies, producing increased viral replication in the host, a phenomenon that is thought to play an important role in triggering the physiopathological mechanisms that determine the severity of the disease.21 The enhanced virulence of some strains of dengue virus has also been suggested as a risk

factor for occurrence of dengue haemorrhagic fever.22 Case series have suggested that individuals with some comorbidities (diabetes, allergies, or hypertension) are more susceptible to dengue haemorrhagic fever and that black individuals seem to be more resistant to this severe form of dengue.23 24 A case-control genetic study in Brazil showed an inverse association between African ancestry and occurrence of dengue haemorrhagic fever.25

How to treat dengue fever?Treatment for dengue fever is supportive as no specific curative treatment exists. Fluid replacement is the only recognised form of intervention for most patients with dengue haemorrhagic fever and dengue shock syn-drome. Early diagnosis and optimal clinical management reduces the likelihood of death in both paediatric and adult patients.10 The therapeutic approach to preventing deaths from dengue haemorrhagic fever must be defined by the dynamic evolution of the disease. Therapeutic regimens proposed in early clinical studies7 26 informed the World Health Organization’s 1997 guidelines.9 The table presents an adaptation of these guidelines that is based both on the old evidence7 26 and on newer clini-cal data that compare different and more aggressive

Number of cases and notifying countries affected, 1955-2005

No

of c

ases

(mil

lion

s)

No

of c

ount

ries

0

0.4

0.6

0.8

1.0

1.2

1.4

0.2

0

20

30

40

50

60

70

10

No of cases

No of countries

19551955

19571957

19591959

19611961

19631963

19651965

19671967

19691969

19711971

19731973

19751975

19771977

19791979

19811981

19831983

19851985

19871987

19891989

19911991

19931993

19951995

19971997

19991999

20012001

20032003

20052005

Treatment guidelines according to level of dengue severityFeatures Treatment

Level APatient with signs and symptoms of dengue fever Oral hydrationNo warning signs Antipyretics and analgesics*Level BPatient with signs and symptoms of dengue fever Vigorous oral hydrationSpontaneous haemorrhagic manifestations or positive tourniquet test; absence of other warning signs

Antipyretics and analgesics*

Level CPatient with signs and symptoms of dengue fever Keep the patient under strict observation at a health care unit with enhanced oral hydrationRaised packed cell volume (until 10% above baseline value) with or without thrombocytopenia (between 50×106/l and 100 000×106/l); with or without other warning signs

Antipyretics and analgesics*

Level DHaemoconcentration: raised packed cell volume (>10% above baseline value). In the absence of this information consider the following values for increased haemoconcentrations—children, >42%; women, >44%; men, >50%; with or without other warning signs

Intravenous hydration with crystalloid solution and plasma expander in a healthcare unit under medical supervision for at least 24 hours; provide sufficient fluids to prevent shock; antipyretics and analgesics*; clinical re-evaluation and measurement of packed cell volume after hydration

Level EShock (pulse pressure <10 mm Hg) Ideally in an intensive healthcare unit; intravenous colloid solution (hydroxy ethyl starch 6%,

molecular weight 200,000; Dextran 70) is mandatoryHaemorrhagic manifestations present or absent Transfusion of fresh whole blood or fresh frozen plasma in case of severe gastrointestinal

bleedingEssential laboratory tests: complete blood count, determination of serum albumin, chest radiography. Other tests might be needed: glucose, urea, creatinine, electrolytes, serum aspartate aminotransferase and serum alanine aminotransferase, blood gas analysis, and ultrasonography of abdomen and chest.*Do not use aspirin or anti-inflammatory drugs.

1192 BMJ | 21 noveMBer 2009 | voluMe 339

CliniCal Review

schedules for fluid resuscitation in the early phase of treatment.27-29

Treatment schemes are organised in five levels according to dengue severity. When the patient is in severity level A (no warning signs) or B (no warning signs but mild spontaneous bleeding events or a posi-tive tourniquet test) admission to hospital is not usually necessary. Treatment centred on adequate oral hydra-tion and control of fever with antipyretics (preferably paracetamol, as aspirin and anti-inflammatory drugs are contraindicated) can be provided in outpatient clinics. Admission to hospital is required to enable fast and monitored hydration in the event of haemo-concentration (a rise in packed cell volume to a level ≤10% above the patient’s own baseline) or thrombo-cytopenia, with or without warning signs (indicating progression to severity level C). An increase in packed cell volume to a level >10% above the patient’s own baseline strongly suggests third space plasma leak-age, the main feature of dengue haemorrhagic fever (severity level D). At this stage, restoring circulatory volume by rapid infusion with an intravenous crys-talloid and a plasma expander is mandatory.7 9 26- 30 A patient who seems to be haemodynamically com-promised also requires continuous monitoring of vital signs, diuresis, and fluid balance and repeated meas-urements of packed cell volume and serum albumin and urea. Intensive supportive care could prevent the patient from developing dengue shock syndrome. If the patient reaches severity level E (dengue shock syn-drome with pulse pressure <10 mm Hg) transfer to an intensive care unit for rapid intravenous colloids solu-tion is considered the ideal management strategy.27--30 Severe gastrointestinal haemorrhage is a rare event that requires the administration of fresh blood or fresh frozen plasma.9 26

A patient’s condition may evolve quickly from a moderate to a severe stage of illness, taking only a few hours to develop shock. Close monitoring of clinical signs and laboratory measurements are therefore the keys for good clinical management. However, inap-propriate administration of intravenous fluid, espe-cially in children, could be dangerous and must be prevented.9 26

How might dengue be prevented? International efforts are under way to develop an effi-cacious vaccine against dengue, and some tetravalent vaccine candidates are in phase 2 or phase 3 trials.31 To date, the only alternative control strategy is to reduce the vector population, a strategy in general centred on the use of chemicals. However, this approach is expensive, poorly sustainable, and environmentally aggressive, with low or no effectiveness in reducing levels of dengue transmission.4 32 Other approaches have been tried without success. For example, a recent randomised trial evaluating a community involvement strategy embedded in a traditional control programme found that the intervention was effective in reducing the Aedes aegypti population, but the trial did not investigate the effect of the intervention on the transmission of the

SOurCeS aNd SeleCtiON CriteriaThis paper was based on a review of the literature available on Medline up to 30 July 2009. We searched articles in English, Spanish, and Portuguese using the key word “dengue”. In addition, we referred to chapters in books and used our personal experience and files

additiONal eduCatiONal reSOurCeS

For patientsFrequently Asked Questions (www.cdc.gov/dengue/•fAQFacts/index.html)—This website (of the Centers for Disease Control and Prevention) answers many questions about dengue, on topics such as transmission, symptoms, treatment, where dengue occurs, prevention

For healthcare professionalsDiagnosis, treatment, and prevention of dengue •(www.who.int/csr/resources/publications/dengue/Denguepublication/en/)—These guidelines contain useful information for health practitioners, laboratory staff, and vector-control workers

tiPS FOr NON-SPeCialiStS aNd tHe PubliC

For non-specialistsA patient’s condition may evolve quickly from a moderate to •a severe stage of illness, taking only a few hours to develop shock. Close monitoring of clinical signs and laboratory measurements are therefore the keys for good clinical management

For residents and travellersResidents and travellers in areas where dengue is endemic •(or travellers up to two weeks after return from such an area) who present with sudden fever must consider dengue as a diagnosis and are recommended to take the following action: To drink large amounts of fluids (water, juices, soups, milk) –To consult a doctor –In the case of high body temperature to use paracetamol, –with no more than four doses in 24 hours (age: <1 year, 60 mg/dose (or a quarter of a 250 mg tablet); 1-4 years, 120 mg/dose (or half a 250 mg tablet); 5-12 years, 250 mg/dose; 12 years, 1000 mg/dose). Never use aspirin or ibuprofenComplications associated with dengue fever usually •appear between the third and fifth day of illness. Go to an emergency clinic if any of the following manifestations appear: red spots on the skin; bleeding from the nose or gums; frequent vomiting; vomiting with blood; black stools; sleepiness; constant crying; abdominal pain; excessive thirst (dry mouth); pale, cold, or clammy skin; or difficulty in breathing When travelling to areas where dengue is endemic, use •insect repellents, wear protective clothes, and stay indoors where windows and doors have mesh screens, or in air conditioned environments

QueStiONS FOr Future reSearCH

What characteristics of the patient with dengue fever •contribute to or predict the risk of developing dengue haemorrhagic fever?What are the physiopathological mechanisms involved in •dengue fever developing into dengue haemorrhagic fever?Is the virulence of a particular strain of dengue virus an •independent risk factor for disease severity?

BMJ | 21 noveMBer 2009 | voluMe 339 1193

CliniCal Review

dengue virus.33 New control technologies have been developed (such as biologicals, traps for adult mosqui-toes), but none has so far shown a major impact on transmission levels.

How to reduce deaths during epidemics and in areas where dengue is endemic?The observed decrease in the death rates for dengue haemorrhagic fever has occurred as a consequence of the advances in the clinical management of the disease.34 The training of health professionals and the promotion of information to increase awareness of the warning signs of severe dengue are important; the latter measure would alert populations living in areas where dengue is endemic and where epidem-ics exist to seek medical care in the early stages of a potential severe disease.

Epidemics are a huge challenge for healthcare services to provide high quality care. For example, at the peak of the 2008 dengue epidemic over 1000 cases of dengue fever or dengue haemorrhagic fever were recorded in a single day in Rio de Janeiro, B razil, creating fear, despair, and chaos in the health system.4 35 No standard strategy exists for manag-ing an epidemic, especially those epidemics with a great proportion of dengue hemorrhagic fever. But experience has shown that additional resources and procedures are needed to enable people with fever to access health care easily, otherwise the care of routine patients will be affected. The idea is that the patient with fever would be guaranteed immediate care at healthcare units. But it is also important that healthcare professionals follow simple and reliable protocols for diagnosis and treatment based on the best current scientific knowledge, and patients must be sufficiently informed to be highly alert to the warning signs of dengue. In serious epidemics, all available staff need to be mobilised.We thank Rivaldo Venâncio Cunha for comments on an early draft of this manuscript.Contributors: MGT and MLB contributed equally to the preparation of this article, and both are guarantors. Competing interests: None declared.Provenance and peer review: Commissioned; externally peer reviewed.

World Health Organization. Dengue net. 2009. http://apps.who.int/1 globalatlas/default.aspWorld Health Organization. Impact of dengue. 2009. www.who.int/2 csr/disease/dengue/impact/en/index.html.Halstead SB, Dengue in the Americas and Southeast Asia: do they 3 differ? Rev Panam Salud Publica 2006;6:407-15.Teixeira MG, Costa MCN, Barreto F, Barreto ML. Dengue: twenty-five 4 years since reemergence in Brazil. Cad Saúde Pública 2009;25(suppl 1):S7-18.Teixeira MG, Costa MCN, Coelho G, Barreto M. Recent shift in age 5 pattern of dengue hemorrhagic fever, Brazil. Emerg Infect Dis 2008;14:1663.Beasley DW, Barrett ADT. The infectious agent. In: Halstead SB, ed. 6 Dengue: tropical medicine. Vol 5. Imperial College Press, 2008: 29-73. Cohen SN, Halstead SB. Shock associated with dengue infection. I: 7 clinical and physiologic manifestations of dengue hemorrhagic fever in Thailand, 1964. J Pediatr 1966;68:448-56.Technical Advisory Committee (TAC) on DHF for SE Asia and Western 8 Pacific regions (World Health Organization). Guides for diagnosis, treatment, surveillance, prevention and control of dengue haemorrhagic fever. Technical Report Series. Geneva: WHO, 1975.World Health Organization. Dengue haemorrhagic fever: diagnosis, 9 treatment, prevention and control. 2nd ed. Geneva: WHO, 1997.

Kittigul L, Pitakarnjanakul P, Sujirarat D, Siripanichgon K. The 10 differences of clinical manifestations and laboratory findings in children and adults with dengue virus infection. J Clin Virol 2007;39(2):76-81.Mairuhu ATA, MacGillavry MR, Setiati TE, Soemantri A, Cate H ten, 11 Brandjes DPM, et al. Is clinical outcome of dengue-virus infections influenced by coagulation and fibrinolysis? A critical review of the evidence. Lancet Infect Dis 2003;3(1):33-41.Kabra SK, Jain Y, Pandey RM, Madhulika, Singhal T, Tripathi P, et al. 12 Dengue haemorrhagic fever in children in the 1996 Delhi epidemic. Trans R Soc Trop Med Hyg 1999;93:294-8.Harris E, Videa E, Perez L, Sandoval E, Téllez Y, Pérez ML, et al. Clinical, 13 epidemiologic and virologic features of dengue in the 1998 epidemic in Nicaragua. Am J Trop Med Hyg 2000;63(1):5-11.Chairulfatah A, Setiabudi D, Ridad A, Colebunders R. Clinical 14 manifestations of dengue haemorrhagic fever in children in Bandung, Indonesia. Ann Soc Belg Med Trop 1995;75:291-5.Solomon T, Dung NM, Vaughn DW, Kneen R, Thao LT, Raengsakulrach 15 B, et al. Neurological manifestations of dengue infection. Lancet 2000;355:1053-9.Cam BV, Fonsmark L, Hue NB, Phuong NT, Poulsen A, Heegaard ED. 16 Prospective case-control study of encephalopathy in children with dengue hemorrhagic fever. Am J Trop Med Hyg 2001;65:848-51.Dussart P, Labeau B, Lagathu G, Louis P, Nunes MR, Rodrigues 17 SG, et al. Evaluation of an enzyme immunoassay for detection of dengue virus NS1 antigen in human serum. Clin Vaccine Immunol 2006;13:1185-9.Vazquez S, Hafner G, Ruiz D, Calzada N, Guzman MG. Evaluation of 18 immunoglobulin M and G capture enzyme-linked immunosorbent assay Panbio kits for diagnostic dengue infections. J Clin Virol 2007;39:194-8.Souza VA, Tateno AF, Oliveira RR, Dominguesb RB, Araujo ES, Kuster 19 GW, et al. Sensitivity and specificity of three ELISA-based assays for discriminating primary from secondary acute dengue virus infection. J Clin Virol 2007;39:230-3. Endy TP, Nisalak A, Vaughn DW. Diagnosis of dengue virus infections. 20 In: Halstead SB, ed. Dengue: tropical medicine. Vol 5. Imperial College Press, 2008:327-60. Morens DM, Halstead SB. The association between dengue 21 hemorrhagic fever and second dengue infections is simply coincidental [controversies]. In: Halstead SB, ed. Dengue: tropical medicine. Vol 5. Imperial College Press, 2008:436-9.Rosen L. La pathogenèse de la dengue hemorragique: 22 discussion critique des hypothèses Actuelles. Bull Soc Pathol 1986;79:342-9.Sierra BC, Koury G, Guzmán MG. Race: a risk factor for dengue 23 hemorrhagic fever. Arch Virol 2007;152:533-42.Cunha RV, Schatzmayr HG, Miagostovich MP, Barbosa AMA, Paiva FG, 24 Miranda RMO, et al. Dengue epidemic in the State of Rio Grande do Norte, Brazil, 1997. Trans R Soc Trop Med Hyg 1999;93:247-9.Blanton RE, Silva LK, Morato VG, Parrado AR, Dias JP, Melo PR, et al. 25 Genetic ancestry and income are associated with dengue hemorrhagic fever in a highly admixed population. Eur J Hum Genet 2008;13:15-8.Nimmannitya S. Clinical spectrum and management of dengue 26 haemorrhagic fever. Southeast Asian J Trop Med Public Health 1987;18:392-7.Smart K, Safitri I. Evidence behind the WHO guidelines: hospital care 27 for children: what treatments are effective for the management of shock in severe dengue? J Trop Pediatr 2009;55:145-8.Ngo NT, Cao XTP, Kneen R, Wills B, Nguyen VM, Nguyen TQ, et al. Acute 28 management of dengue shock syndrome: a randomized double blind comparison of 4 intravenous fluid regimens in the first hour. Clin Infect Dis 2001;32:204-13. Ranjit S, Kissoon N, Jayakumar I. Aggressive management of dengue 29 shock syndrome may decrease mortality rate: a suggestive protocol. Pediatr Crit Care Med 2005;6:412-9.Wills BA, Nguyen MD, Ha TL, Dong TH, Tran TN, Le TT, et al. Comparison 30 of three fluid solutions for resuscitation in dengue shock syndrome. N Engl J Med 2005;353:877-89.Halstead SB. Dengue vaccines are coming: are we prepared? Virus 31 reviews and research. Socied Bras Virol 2008;13:29-30.Ooi EE, Goh KT, Gubler DJ. Dengue prevention and 35 years of vector 32 control in Singapore. Emerg Infect Dis 2006;12:887-93.Vanlerberghe V, Toledo ME, Rodríguez M, Gomez D, Baly A, Benitez 33 JR, et al. Community involvement in dengue vector control: cluster randomised. BMJ 2009;338:1959b.Kalayanarooj S. Standardized clinical management: evidence of 34 reduction of dengue haemorrhagic fever case-fatality rate in Thailand. Dengue Bull 1999;23:10-7.Barreto ML, Teixeira MG. Dengue fever: a call for local, national and 35 international actions [letter]. Lancet 2008;372:205.