Dengue & Dengue Haemorrhagic Fever

Meet the patient

A 9-year-old boy arrives in the October monsoon, day 4 of fever that has just settled. Instead of improving he is cold, restless, and clutching his belly; he has vomited twice. His obs are T 37.1, HR 132, BP 100/88 (pulse pressure 12 mmHg), capillary refill 4 seconds, and he has not passed urine since morning. Platelets are 40,000, and the haematocrit has climbed from 38 to 47 per cent.^[1]

The registrar is relieved - "the fever's gone, he's turning the corner." That sentence is the recurring trainee error in dengue, and the whole of this topic is the answer to the examiner's question it hides: is this recovery, or the critical phase? The dropping fever is the start of the danger, not the end of the illness.^[1]

The leak, not the bleed - the one idea that frames dengue

Dengue is a disease of capillary endothelial leak, not haemorrhage. Hold that single idea and the rest of the topic arranges itself: the severe spectrum (DHF, DSS) is driven by plasma leaving the vasculature, which is why haematocrit - not haemoglobin - is the monitoring variable, why fluids are the only specific therapy, and why the patient deteriorates at defervescence, the moment the fever falls.^[2]

The agent is a positive-sense single-stranded RNA virus of the genus Flavivirus, family Flaviviridae - a sibling of yellow fever, Japanese encephalitis, Zika and West Nile - existing as four antigenically distinct serotypes (DENV-1 to DENV-4). Infection with one confers lifelong immunity to that serotype only and only transient cross-protection against the others, which is the seed of the ADE story below. It reaches humans through the bite of Aedes aegypti, the principal urban vector, and less often Aedes albopictus.^[1]

The vector you must be able to draw: Aedes aegypti is a small dark mosquito with a white lyre on the thorax and white-banded legs, day-biting (peaks early morning and late afternoon), peri-domestic, anthropophilic, breeding in small collections of clean standing water - flowerpots, discarded tyres, coconut shells, water-storage jars - and biting mainly below the knee. It is nervous and low-flying. A. albopictus, the Asian tiger mosquito, is a secondary, more rural, cold-tolerant vector that has crept into temperate regions.^[1]

Etymology for viva gold: dengue is thought to come from the Swahili ka-dinga pepo - a sudden cramp-like seizure blamed on evil spirits - which the Spanish trimmed to dengue, "fastidious" or "careful", describing the stiff gait of sufferers in severe bone pain. The same illness carries two older names still worth knowing: break-bone fever (for the myalgia) and dandy fever.^[1]

How common, and who is at risk: dengue is the most important arboviral disease of humans, endemic across the tropics, with roughly 390 million infections a year (~96 million symptomatic) and about half the world's population at risk across more than 100 endemic countries (WHO estimates 100-400 million infections each year). The burden sits heaviest in South and Southeast Asia (India, Indonesia, Vietnam, the Philippines, Thailand, Bangladesh, Sri Lanka), the Americas (Brazil, Mexico, the Caribbean), the Western Pacific and sub-Saharan Africa. In India all four serotypes circulate, dengue is endemic to every state and union territory, and the monsoon (June-November) drives the annual surge.^[1]

^[1]

Two classifications, one ward - 1997 and 2009

Hold both classifications at once: the WHO 1997 scheme names the disease, the WHO 2009 scheme runs the ward. Both are examinable, both are used in India, and an examiner will ask you to reproduce each. The 1997 scheme (still the Indian exam standard) divides disease into DF, DHF (grades I-IV) and DSS; the 2009 revision (level-of-severity, used for triage and management) splits it into dengue without warning signs, dengue with warning signs, and severe dengue.^[1]

WHO 1997 - DF, DHF and DSS (the exam standard)^[1]

Dengue fever (DF) is an acute febrile illness with two or more of headache, retro-orbital pain, myalgia, arthralgia, rash, a haemorrhagic manifestation (petechiae, positive tourniquet test) and leucopenia - in the absence of plasma leakage. It is usually self-limiting.^[1]

Dengue haemorrhagic fever (DHF) requires all four of:^[1]

1. Fever - acute, high, lasting 2-7 days.
2. Haemorrhagic tendency - a positive tourniquet test, OR spontaneous petechiae/purpura/ecchymoses, OR mucosal or GI bleeding, OR haematemesis/melaena.
3. Thrombocytopenia - platelet count under 100 x 10^9/L.
4. Plasma leakage - haematocrit rise of 20% or more above baseline (or population mean for age/sex), OR a 20% or greater fall in haematocrit after fluids, OR clinical leakage (ascites, pleural effusion - classically right-sided - hypoproteinaemia).^[1]

DHF is graded I-IV by the presence and severity of shock:^[1]

Grades III and IV are dengue shock syndrome (DSS). The pulse-pressure threshold of 20 mmHg is the single most tested number in dengue - a BP of 100/88 (pulse pressure 12) is already DSS Grade III, and the systolic number can still look reassuringly normal.^[1]

WHO 2009 - dengue, warning signs, and severe dengue (runs the ward)^[3]

The 2009 revision simplified the scheme to two clinical levels to drive triage, and is the system the bedside and trial world now uses:^[3]

• Dengue without warning signs - a febrile illness with non-specific features (nausea/vomiting, rash, aches and pains) and no warning signs, no plasma leakage and no organ impairment - managed as an outpatient.
• Dengue with warning signs - fever plus any warning sign. The multicentre DENCO study (Alexander et al., seven countries, 2259 patients) that underpinned the revision found a small proportion of patients (5%) progress to severe disease while under observation, and a systematic review and meta-analysis of 39 studies found persistent vomiting, hepatomegaly over 2 cm, and a rising haematocrit with falling platelets among the features associated with severe dengue - these patients are admitted for close observation.^[10][13]
• Severe dengue - any one of shock or severe vascular leakage, severe bleeding, or severe organ dysfunction. The DENCO study concluded that including exactly these "readily discernible complications" was necessary for a classification that identified patients requiring major intervention, and proposed the two-tier scheme of Dengue and Severe Dengue that the WHO adopted in 2009.^[10][13][3]

The 1997 scheme is more specific but clumsy (many cases fall indeterminate); the 2009 scheme triages better but is less specific for severe dengue in non-Asian populations. India's NVBDCP still reports in 1997 categories while the bedside uses 2009 - know both, and know they describe the same patient.^[1]

^{[2] [13] [8] [15]}

Enlarge

The day the fever falls - febrile, critical, recovery

The danger begins the moment the fever drops. Dengue is a phasic illness, and the central bedside skill is recognising which phase the patient is in - and where in that phase the danger lies. The classic, paradoxical sign of severe dengue is that the patient deteriorates as the fever settles: the leaking window opens at defervescence, and the patient who "feels worse as the fever goes" is leaking.^[1]

Febrile phase (days 1-7)^[1]

The illness begins abruptly with high fever (often 39-40 C), carrying two or more of the WHO clinical features:^[1]

• Headache - typically bifrontal.
• Retro-orbital pain - pain behind the eyes, worsened by eye movement; classical.
• Myalgia and arthralgia - severe, the "break-bone" pain; back and muscle pain dominate.
• Rash - common. Early flushing/erythema of face, neck and chest ("dengue facies"); then a maculopapular or morbilliform rash on trunk and extensors, sometimes with islands of sparing; petechiae and a positive tourniquet test mark the bleeding tendency.
• Nausea, vomiting, anorexia, altered taste.
• Injected conjunctivae and pharyngeal injection, cervical lymphadenopathy and periorbital oedema may be noted.
• Hepatomegaly - present in many children; tender hepatomegaly suggests progression.^[1]

Fever typically lasts 2-7 days. A biphasic/saddle-back course is classical (settling on day 3-5 then recurring) but is not universally present.^[1]

Critical phase - defervescence and the warning signs^[1]

The transition occurs at defervescence (day 3-7), when the fever drops. The WHO warning signs (2009) mark impending severe disease and mandate admission with IV fluid therapy - learn them as the WARNING mnemonic:^[1]

^[1]

Clinical shock (DSS) - once plasma loss is severe enough, the patient develops cold clammy skin, prolonged capillary refill, weak rapid pulse, narrow pulse pressure (under 20 mmHg), hypotension (a late sign), oliguria, restlessness progressing to lethargy and impaired consciousness. Grade IV = undetectable pulse and BP. Hypotension is a late, pre-arrest sign in dengue - assess perfusion and pulse pressure, not the pressure number alone.^[1]

Recovery phase (48-72 hours)^[3]

With endothelial recovery, leaked plasma is reabsorbed, appetite returns, urine output rises (a diuresis), the platelet count recovers and the haematocrit falls. The convalescent rash - confluent petechiae with round white islands of normal skin, "isles of white in a sea of red" - is characteristic, often with generalised pruritus and sinus bradycardia.^[3]

Everyone forgets: the recovery phase is also where fluids turn poisonous. The patient who was leaking is now reabsorbing several litres, and continuing the IV rate from the critical phase floods them into pulmonary oedema - an entirely iatrogenic disaster. Pull the fluids down the moment the haematocrit falls and the urine picks up.^[1]

Atypical presentations (deliberately tested)^[2]

• Infants and young children - non-specific febrile illness, febrile seizures, refusal of feeds, vomiting; severe primary dengue occurs in infants of immune mothers - passively transferred maternal antibody increases DHF/DSS risk in the infants for a certain period of time (WHO).
• Elderly - blunted fever, comorbidity-driven severity (diabetes, hypertension, CKD), higher rates of severe bleeding, AKI, encephalopathy and mortality; shock may be masked by beta-blocker or diuretic therapy.
• Pregnant women - higher risk of severe dengue, miscarriage, preterm labour, fetal distress and vertical transmission; close fetal monitoring is essential.
• Diabetics, hypertensives, those with sickle-cell trait, asthma, CKD, obesity - a more severe course.
• Primary infection in adults may present with severe hepatitis, encephalitis, myocarditis or AKI without the classical warning signs (expanded dengue syndrome).
• "Felt better on day 1, worse on day 5" - the patient who improves then deteriorates is in the critical phase; never discharge on fever pattern alone.^[2]

Why the second bite kills you - ADE and the cytokine storm

A secondary infection with a different serotype is markedly more likely to cause DHF/DSS, and antibody-dependent enhancement is why. Dengue pathogenesis runs on three intertwined arms - viral entry and replication, the host cytokine storm, and the ADE phenomenon - and all three converge on a single end-result: capillary endothelial dysfunction with plasma leakage, the lesion that defines DHF/DSS.^[2]

Enlarge

The cascade, in molecular detail:^[1][2]

1. Inoculation. An infective A. aegypti female probes the dermis and inoculates virus-laden saliva with anticoagulant into the capillary plexus.
2. Skin replication and primary viraemia. Dermal Langerhans cells and dendritic cells bind virus through DC-SIGN (CD209), the principal attachment receptor; infected cells mature, migrate to draining lymph nodes, and spill into the blood as primary viraemia peaking around day 2-4 of fever. Other receptors include heparan sulphate, the mannose receptor and the phosphatidylserine receptors TIM/TAM (which recognise the "eat-me" signals virions borrow from apoptotic cells).
3. Systemic dissemination. Virus seeds monocytes, macrophages, liver Kupffer cells and hepatocytes, splenic and bone-marrow macrophages. Hepatocyte infection drives transaminitis (AST higher than ALT); bone-marrow infection causes transient megakaryocyte suppression, contributing to thrombocytopenia.
4. Antibody-dependent enhancement - the central concept. Infection with one serotype induces lifelong serotype-specific neutralising antibody plus transient cross-reactive, non-neutralising antibody against the others. When a different serotype arrives months to years later, that cross-reactive IgG binds the virion but fails to neutralise it; the virus-IgG complex is taken up by Fc-gamma receptors on monocytes/macrophages, which internalise and replicate virus far more efficiently than naive cells. The result is a 10- to 100-fold higher viral load, a heavier cytokine storm, and a markedly greater risk of DHF/DSS.
5. Cytokine storm. Infected monocytes/macrophages release TNF-alpha, IL-1beta, IL-6, IL-8, IL-10, IL-12, IL-18, IFN-gamma, MCP-1, RANTES and VEGF; complement releases C3a and C5a; mast cells release tryptase, chymase and TNF-alpha. The combined effect is capillary endothelial dysfunction: gaps open between endothelial cells and albumin and plasma leak into the interstitium and serous cavities (peritoneum, pleura - mostly right-sided on chest X-ray, a constant finding whose extent tracks severity). Plasma leakage, not haemorrhage, is the dominant lesion of DHF/DSS.
6. Thrombocytopenia and bleeding. Multifactorial: bone-marrow megakaryocyte suppression, peripheral immune destruction (anti-platelet antibodies, complement), platelet activation and consumption, and splenic sequestration. Platelet dysfunction plus mild disseminated intravascular coagulation (prolonged APTT, low fibrinogen, raised D-dimer) and hepatic synthetic compromise produce the bleeding tendency. Severe bleeding is more often a late (recovery-phase) event driven by prolonged shock and DIC than the cause of shock itself.
7. The critical phase. Around defervescence (day 3-7) the leaking patient enters the critical phase (24-48 hours). Haematocrit climbs as plasma leaves the vasculature; if leakage is severe and unchecked, hypovolaemic shock (DSS) supervenes. The same cytokine milieu can rarely drive organ impairment (hepatitis, encephalopathy, myocarditis, AKI, ARDS) - expanded dengue syndrome.
8. Recovery. After 48-72 hours the endothelium recovers, leaked fluid is reabsorbed (which can cause transient haemodilution and pulmonary oedema if IV fluids were over-given), a diuresis appears, the platelet count recovers, and the characteristic convalescent rash (confluent petechiae with round white islands) appears, often with bradycardia and pruritus.^[3] Transmission in one breath: human to mosquito to human, no animal reservoir for the urban cycle. The mosquito acquires infection by biting a viraemic human (patients are infectious to mosquitoes from up to 2 days before symptoms to about 2 days after the fever settles; most are viraemic for about 4-5 days, occasionally up to 12); after an extrinsic incubation of 8-12 days it is infectious for life, and transovarial transmission of the virus within mosquitoes has also been recorded (WHO). Vertical (maternal-fetal) spread can occur, with risk linked to the timing of infection in pregnancy; transfusion, transplant and needle-stick transmission are rare.^[1]

ADE is the key that unlocks three otherwise-puzzling facts: it explains why secondary dengue is more severe than primary, why infants born to dengue-immune mothers can develop severe primary dengue (passively transferred maternal antibody raises DHF/DSS risk in the infants for a certain period of time - WHO), and - later in this page - why the Dengvaxia vaccine harmed seronegative recipients (it behaved like a silent primary infection, priming them for ADE on later wild-type exposure).^[2]

Risk factors for severe dengue (DHF/DSS) - run through these on autopilot:^[1][2]

• Secondary (heterologous) infection - the single most important host risk factor; on meta-analysis, secondary infection is among the features associated with WHO-classified severe dengue.
• Young age - infants of immune mothers and children aged 5-15 years have the highest DHF rates in hyperendemic regions.
• Viral serotype/genotype - DENV-2 and DENV-3, particularly Asian genotypes, are more virulent than DENV-1 or DENV-4.
• Host genetics - HLA class I and II alleles; polymorphisms in DC-SIGN, Fc-gamma receptor, vitamin D receptor and TNF genes.
• Female sex, high BMI, comorbidity (diabetes, hypertension, asthma, sickle-cell trait, CKD), pregnancy, and infants under 1 year with maternally derived antibody.
• Climate - the monsoon expands vector breeding and drives the seasonal surge.^[1]

Transmission in one breath: human to mosquito to human, no animal reservoir for the urban cycle. The mosquito acquires infection by biting a viraemic human (patients are infectious to mosquitoes from up to 2 days before symptoms to about 2 days after the fever settles; most are viraemic for about 4-5 days, occasionally up to 12); after an extrinsic incubation of 8-12 days it is infectious for life, and transovarial transmission sustains the virus between epidemics. Vertical (maternal-fetal) spread can occur, with risk linked to the timing of infection in pregnancy; transfusion, transplant and needle-stick transmission are rare.^[1]

The bedside round - vitals, perfusion, the tourniquet test

Bedside assessment in dengue is not about making the diagnosis (NS1/IgM does that) - it is about detecting leakage and shock before the blood pressure collapses. It rests on vital signs and perfusion (especially pulse pressure), a focused fluid-balance and abdominal exam, and the tourniquet test - none of which needs a laboratory.^[1]

History. Establish the tempo (abrupt fever, day of illness - the number that decides NS1 versus IgM), residence or travel to a dengue-endemic area, monsoon exposure to Aedes breeding sites, prior dengue (raises ADE risk), comorbidity (diabetes, hypertension, CKD, asthma, sickle-cell, pregnancy, immunosuppression), and any NSAID/aspirin/anticoagulant use. Ask specifically for warning signs: abdominal pain, persistent vomiting, bleeding, lethargy, restlessness, oliguria, and dizziness on standing.^[1]

Vital signs and perfusion - every 1-4 hours through the critical phase:^[1]

• Temperature - a dropping fever with a worsening patient = the critical phase.
• Pulse rate and pulse pressure - pulse pressure under 20 mmHg (e.g. 100/80) defines DSS Grade III; undetectable BP = Grade IV.
• Blood pressure - a late sign; shock is present before hypotension in dengue.
• Capillary refill time, peripheral temperature, urine output - markers of perfusion.
• Respiratory rate - a rising rate may signal pleural effusion, pulmonary oedema, or the metabolic acidosis of shock.^[1]

Tourniquet test (Hess test). Inflate a sphygmomanometer cuff on the upper arm to a pressure midway between systolic and diastolic, hold it for 5 minutes, then count petechiae distal to the cuff in a 2.5 cm (1 inch) square. Twenty or more petechiae = positive (the WHO standard). It supports - but never confirms - the diagnosis; the test may be negative or only mildly positive during profound shock and usually becomes positive after recovery from shock (WHO 1997).^[1]

Abdomen. Look for hepatomegaly (often tender, with mild transaminitis), right-upper-quadrant tenderness, ascites (shifting dullness, fluid thrill) and the bowel-bladder pattern - abdominal pain is a warning sign. Gallbladder wall thickening is best seen on ultrasound, but tender hepatomegaly and a positive Murphy sign (without stones) point to it.^[3]

Skin and other systems. Examine for petechiae, ecchymoses, purpura, rash and mucosal bleeding (gingival, nasal, conjunctival); bleeding from venepuncture sites marks significant coagulopathy. Assess for pleural effusion (dullness, reduced breath sounds, typically right-sided), encephalopathy, dehydration, and shock.^[3]

Monitoring. In the admitted patient through the critical phase, watch hourly pulse, BP, pulse pressure, capillary refill and urine output; 4-6 hourly haematocrit and platelets; and a daily ultrasound for effusion or ascites if available. Rising haematocrit with falling platelets is the bedside signature of plasma leakage.^[1]

Exclude malaria first - the tropical-fever mimics

Dengue sits at the centre of the acute undifferentiated febrile illness of the tropics - a syndrome shared with malaria, chikungunya, leptospirosis, typhoid, scrub typhus and influenza. Always exclude malaria first (a film or antigen test is rapid and changes management), and exclude bacteraemia or meningococcaemia in the toxic patient. The diseases that look most like dengue - malaria and chikungunya - are also mosquito-borne, and the three can coexist; a mixed infection is more severe.^[1]

^[1]

The exam trap: a returned traveller with fever, headache, myalgia and thrombocytopenia should have malaria excluded first (film or antigen), then dengue (NS1/IgM) and chikungunya (IgM). A single negative malaria slide or Widal does not exclude either - repeat the films, because mixed infections occur and are more lethal.^[1]

NS1 then IgM - test by the day of fever

There is no single perfect test - the test depends on the day of fever. Viraemia, antigenaemia and the antibody response follow a predictable time course, and the day of illness decides whether you send NS1 or IgM. Get the day wrong and you send the wrong test.^[1]

^{[18] [1]}

Timing summary - the most examinable table in the topic:^[1][18]

The practical rule: in the first week send NS1 plus FBC; in the second week send IgM plus FBC; combine NS1 and IgM when the timing is uncertain.^[1]

Supportive bloods and imaging^[3]

• Full blood count - thrombocytopenia and an increased haematocrit with decreased platelets are the key laboratory correlates of severe dengue: a systematic review and meta-analysis of 39 studies found both among the features significantly associated with WHO-classified severe dengue. Laboratory markers - haematocrit, platelet count, liver enzymes and coagulation tests - are the ones to follow for early detection of progression.

• Coagulation - coagulation tests are part of the core marker set for early detection of severe dengue.

• LFTs - elevated AST and ALT are associated with severe dengue on meta-analysis.

• Ultrasound - gallbladder wall thickening on imaging is associated with severe dengue, and the same review suggests it may help refine triage of severe dengue patients.^[13][12]

• Blood cultures and a malaria film - to exclude bacterial sepsis and malaria (always exclude malaria first in any tropical fever).^[3]

Criteria reproduced verbatim - WHO 1997 DHF criteria^[1]

A diagnosis of DHF requires all four of:^[1]

1. Fever - acute, high, biphasic, lasting 2-7 days.
2. Haemorrhagic manifestations - at least one of positive tourniquet test, petechiae, ecchymoses/purpura, epistaxis, gum bleeding, haematemesis, melaena, haematuria (a positive tourniquet test alone counts if other features are absent).
3. Thrombocytopenia - platelet count under 100 x 10^9/L.
4. Plasma leakage - haematocrit rise of 20% or more above baseline (or above population mean for age/sex), OR a 20% or greater fall in haematocrit after fluid therapy, OR clinical/biochemical leakage (ascites, pleural effusion, hypoproteinaemia or hypoalbuminaemia).^[1]

DHF grading, and the threshold for DSS:^[1]

• Grade I: fever plus non-specific symptoms, the only haemorrhagic manifestation a positive tourniquet test and/or easy bruising.
• Grade II: grade I signs plus spontaneous bleeding (skin, mucosa, GI tract).
• Grade III: circulatory failure - weak rapid pulse, narrow pulse pressure (under 20 mmHg), hypotension, cold/clammy skin, restlessness.
• Grade IV: profound shock - pulse and blood pressure undetectable.^[1]

Grades III and IV are dengue shock syndrome (DSS). The pulse-pressure threshold of 20 mmHg is the single most tested number in dengue.^[1]

WHO 2009 severe-dengue criteria^[3]

Severe dengue is defined by any one of:^[10]

1. Shock or severe vascular leakage (the category that captures dengue shock syndrome and substantial fluid accumulation).
2. Severe bleeding (as evaluated by a clinician) - e.g. haematemesis, melaena, intracranial bleed.
3. Severe organ dysfunction - meta-analysis associates impaired consciousness and dyspnoea with WHO-classified severe dengue; myocarditis is a recognised cardiac manifestation.^[10][13][15]

Fluids are the cure - WHO Groups A, B and C

There is no antiviral for dengue. The entire treatment is fluid titration, and it is the single intervention that cuts severe-dengue case fatality below 1%. Everything else - oxygen, transfusion, organ support - is supportive; the specific therapy is matching the fluid dose to the phase of illness and the degree of leakage. Too little and the patient shocks; too much and they flood into pulmonary oedema.^[1]

Enlarge

^[3]

ABCDE first. Secure the airway, give oxygen to any patient in shock or respiratory distress, establish IV access, and assign the WHO triage group (A, B or C) at presentation from the warning signs and severity.^[1]

A word on antibiotics. In suspected or established DSS, take blood cultures, send lactate, and begin fluid resuscitation immediately - but not broad-spectrum antibiotics as routine. They are reserved for proven or strongly suspected bacterial co-infection, which is more common in severe dengue than historically appreciated.^[1]

Fluid resuscitation in shock (WHO Group C):^[1]

• The WHO 1997 regimen for initial resuscitation in DSS is an isotonic crystalloid (Ringer's lactate, Ringer's acetate or 5% dextrose in saline) given as a rapid bolus of 10-20 mL/kg in under 20 minutes, repeatable to 20-30 mL/kg, with plasma/plasma substitute 10-20 mL/kg if shock persists and the haematocrit keeps rising. The WHO 2009 handbook phrases the same escalation as: compensated shock - isotonic crystalloid 5-10 mL/kg/hour over 1 hour then reassess; hypotensive shock - crystalloid or colloid 20 mL/kg as a bolus over 15 minutes, then stepwise reduction.
• The WHO-guideline regimen as tested in a randomised trial of early resuscitation in children is Ringer's lactate 20 mL/kg over 15 minutes, followed by stepwise decreasing rates of 10, 7, 5 and 3 mL/kg/hour over the next 12 hours - and the trial explicitly notes that following the WHO guideline this way usually required large volumes of Ringer's lactate that might induce secondary fluid overload, so reassess continuously.
• Which fluid? A randomised double-blind comparison of four fluids in the first hour of DSS in 230 children (dextran, gelatin, lactated Ringer's, and normal saline) found no clear advantage to any one of the four, though recovery times were longest with lactated Ringer's.
• Colloid may act faster: in an earlier randomised trial in 50 children with DSS, colloids (dextran 70 or gelafundin) restored cardiac index and blood pressure and normalised haematocrit more rapidly than crystalloids (Ringer's lactate or 0.9% saline), with dextran 70 the most rapid.
• Maintenance fluid once perfusion is restored; carefully titrate the rate down as the patient stabilises to avoid overload.
• Do not give prophylactic platelets for a low count alone (below).^[6][8][7]

Antipyretics and analgesia. Consensus guidelines from the World Health Organization and the US CDC recommend paracetamol (acetaminophen) to manage pain and fever, and contraindicate non-steroidal anti-inflammatory drugs because of potentially increased bleeding risk on a background of thrombocytopenia. The WHO handbook caps paracetamol at not more than 4 g/day in adults (adjusted accordingly in children) with a dosing interval of not less than 6 hours; tepid sponging is a safe adjunct. Avoid NSAIDs and aspirin, and monitor closely through the critical phase.^[14]

WHO Group A - outpatient (mild, no warning signs)^[1]

• Encourage oral fluids and maintain hydration.
• Paracetamol for fever and pain (adults max 4 g/day, interval not less than 6 hours - WHO); no NSAIDs, no aspirin - WHO/CDC consensus guidelines contraindicate them in dengue because of bleeding risk with thrombocytopenia.
• Review daily for warning signs and the platelet and haematocrit trend.
• Safety-net: return immediately if any warning sign appears - abdominal pain, persistent vomiting, bleeding, lethargy, restlessness, oliguria, cold hands or feet.^[14][3]

WHO Group B - ward (warning signs, plasma leakage without shock)^[1]

• Admit. Establish IV access and monitor closely through the critical phase.
• Judicious, quota-based fluid replacement is the mainstay during the critical phase of DHF: guideline-based practice uses a calculated fluid quota of maintenance plus 5% deficit (M + 5% deficit) to prevent organ hypoperfusion.
• What adults actually need: in a prospective study of 115 adults with DHF, the mean fluid requirement during the critical phase was about 5.3 L over 48 hours - and most patients in practice exceeded the recommended quota, which is the route to fluid overload.
• Reduce the rate the moment the patient stabilises - continuing critical-phase rates into recovery is what causes the fluid overload and pulmonary oedema.^[9]

WHO Group C - ICU (severe dengue, DSS)^[1]

^{[7] [8] [6] [5] [15]}

Platelet transfusion - when, and when NOT, to give^[1]

Prophylactic platelet transfusion for thrombocytopenia alone is NOT supported by evidence. The randomised trial that settled this recruited adults with confirmed dengue and platelets at or under 20 000 per microlitre, none with persistent or severe bleeding, and found:^[5]

• No significant reduction in bleeding - clinical bleeding by day 7 or discharge occurred in 21% of transfused patients versus 26% of controls (risk difference -4.98%, RR 0.81, p = 0.16).
• More harm - adverse events occurred in 13 transfused patients versus 2 controls.
• The bottom line for the ward: a low platelet count alone is not an indication to transfuse; reserve platelets for genuinely significant bleeding, on haematology advice.^[5]

The drugs to AVOID - the high-yield contraindication list^[1]

• NSAIDs (ibuprofen, diclofenac, mefenamic acid, naproxen, aspirin) - platelet inhibition plus gastric irritation drives severe bleeding. The single most tested "do not" in dengue. In children, aspirin also risks Reye syndrome.
• Intramuscular injections - haematoma risk; route every medication IV or oral.
• Corticosteroids - not recommended routinely: a Cochrane review of eight randomised trials (948 participants) found no evidence of benefit on death, transfusion need, or disease progression in dengue-related shock or early dengue, and the WHO handbook notes there is little or no evidence in favour of steroids. May be considered for haemophagocytic lymphohistiocytosis or refractory shock, case by case.^[16]
• Prophylactic platelet transfusions (above).
• Carbaspirin calcium-containing "aspirin-combination" antipyretics - read the labels in India.^[1]

^[1]

Discharge criteria^[1]

Discharge when the patient is clinically well with no fever for 48 hours (without antipyretics), improving clinical status (general well-being, appetite, haemodynamic stability, urine output, no respiratory distress), an increasing trend of platelet count, and a stable haematocrit without intravenous fluids (WHO 2009 discharge criteria). Counsel on the recovery-phase rash, pruritus and bradycardia.^[1]

The subtypes that bite

Primary versus secondary is the fork that predicts severity. Secondary infections - high IgG early, an IgM/IgG ratio under 1.2, or a documented prior episode - carry a markedly higher risk of DHF/DSS through ADE, with a more rapid course, more profound leakage, and a more dangerous critical phase.^[2][13]

• Dengue haemorrhagic fever (DHF) - plasma leakage is the main pathophysiological hallmark that distinguishes DHF from DF; thrombocytopenia, a rising haematocrit with falling platelets, elevated AST and ALT, and gallbladder wall thickening are the features meta-analysis associates with severe dengue.
• Dengue shock syndrome (DSS) - the end of the leakage spectrum: severe plasma leakage can result in hypovolaemic shock, and fluid resuscitation to counteract it is the mainstay of treatment.
• Expanded dengue syndrome - atypical severe organ involvement; severe organ dysfunction sits alongside shock/severe vascular leakage and severe bleeding as the severe-dengue criteria, and myocarditis is a recognised cardiac manifestation (systematic review and meta-analysis).
• Dengue with comorbidity - comorbidity was one of the features associated with higher risk of severe dengue on meta-analysis; expect a more dangerous course and a lower threshold to admit.
• Secondary infection - pre-existing dengue antibodies are among the factors reported to correlate with development of DHF (the ADE story above).^[2][13][15]

When dengue breaks its pattern - expanded syndrome and pitfalls

The complications cluster by mechanism, and the mechanism names them. Plasma leakage gives DSS, pleural effusion (usually right-sided, occasionally bilateral), ascites, hypoproteinaemia and electrolyte disturbance (hyponatraemia, hypokalaemia).^[1]

Bleeding runs from petechiae, purpura and ecchymoses through epistaxis, gum bleeding, menorrhagia and haematuria to haematemesis and melaena (from gastric mucosal microvascular injury and stress ulceration), intracranial haemorrhage (rare but devastating, especially with concomitant NSAIDs), retroperitoneal bleed and haemothorax.^[1]

Organ impairment (expanded dengue syndrome):^[1]

• Hepatic - transaminitis (AST over ALT), jaundice, acute liver failure (rare).
• Neurological - encephalopathy (metabolic, often reversible), encephalitis (rare, virus isolated from CSF), seizures, Guillain-Barre syndrome, transverse myelitis, hypoglycaemic brain injury in children.
• Cardiac - myocarditis, arrhythmias, bradycardia (also a recovery-phase sign).
• Renal - acute kidney injury (shock, rhabdomyolysis, HLH).
• Pulmonary - ARDS, pulmonary haemorrhage, pneumonitis.
• Haematological - disseminated intravascular coagulation, haemophagocytic lymphohistiocytosis (HLH).^[1]

The iatrogenic complications - most of them are fluid or NSAID errors:^[1]

• Fluid overload - the single most common iatrogenic complication: pulmonary oedema, pleural effusion, respiratory distress, especially when fluids continue into the recovery phase. Treat with furosemide once stable.
• Transfusion-transmitted infection, allergic reaction, TRALI from unnecessary blood products.
• Aspirin in children - Reye syndrome as well as bleeding.
• NSAID-induced renal injury and GI bleed.^[1]

The classic pitfalls every candidate must name:^[1]

• Treating "a fever that has settled" as recovery - the dropping fever is the start of the critical phase, not the end of the illness. The patient who feels worse as the fever goes is leaking.
• Using haemoglobin instead of haematocrit to monitor leakage - haematocrit is the marker; haemoglobin is unaffected by plasma leakage.
• Giving NSAIDs, aspirin or IM injections - the most preventable cause of severe bleeding.
• Transfusing platelets for a low count alone - no benefit, real harm.
• Over-resuscitating with crystalloid - flooding into pulmonary oedema.
• Missing a co-pathogen - malaria, leptospirosis, bacterial sepsis are common and lethal if untreated.
• Treating a single Widal or malaria slide as definitive exclusion - repeat the films; mixed infections occur.^[1]

Prognosis and who goes where

Treated dengue rarely kills; untreated DSS often does, and the entire difference is fluid.^[1]

• Uncomplicated dengue fever - case fatality less than 1% (WHO 1997); full recovery in 7-10 days.
• DHF without shock - mortality is low with appropriate fluid therapy (national dengue CFRs of well-run programmes sit around or below 1%, e.g. below 0.2% in Thailand's mature programme).
• DSS - untreated shock is typically lethal within 12-24 hours (WHO 1997); prompt volume replacement cuts the case fatality of severe dengue below 1% (WHO).
• Severe dengue with major bleeding, expanded dengue syndrome, or in the elderly, comorbid or pregnant - higher mortality; hepatic failure, ARDS and intracranial haemorrhage carry the worst outcomes.^[2]

Predictors of a poor outcome: secondary infection, age extremes, comorbidity (diabetes, CKD, hypertension, obesity), pregnancy, late presentation, delayed fluid resuscitation, NSAID or aspirin use, major bleeding and organ impairment.^[2]

Disposition:^[1]

• Outpatient (Group A) - clinically well, no warning signs, tolerating oral fluids, platelet trend monitored, safety-net given.
• Ward (Group B) - any warning sign, plasma leakage, comorbidity, pregnancy, infants, or social reasons.
• ICU (Group C) - DSS, severe bleeding, severe organ impairment, refractory shock, or a need for vasoactive support, ventilation or renal replacement.^[1]

Follow-up is short - most patients recover fully within 1-2 weeks. Counsel on post-dengue fatigue (may last weeks), the convalescent rash and pruritus, and mosquito avoidance for the febrile week, because the patient is infectious to mosquitoes.^[1]

Special populations - thresholds and danger shift^[1]

• Children - non-specific febrile illness, febrile seizures, refusal of feeds, vomiting; higher DHF/DSS rates in hyperendemic regions; hypoglycaemia is common and dangerous; weight-based fluid and paracetamol dosing; paracetamol only - never aspirin (Reye syndrome); close monitoring of capillary refill and urine output. The tourniquet test and gallbladder wall thickening are especially useful in children.
• Pregnancy - higher risk of severe dengue, miscarriage, preterm labour, fetal distress, low birth weight, vertical transmission (causing neonatal thrombocytopenia, fever and rash) and peripartum haemorrhage; close fetal monitoring; platelet transfusion before surgical delivery; coordinate obstetrics and infectious disease.
• Elderly - blunted fever, comorbidity-driven severity, higher rates of severe bleeding, AKI, encephalopathy and mortality; shock may be masked by beta-blockers; lower threshold to admit and to use ICU.
• Diabetes, hypertension, CKD, asthma, sickle-cell trait, obesity, immunosuppression - more severe disease; adjust fluid and drug doses; monitor glucose and renal function closely.
• Anticoagulated or antiplatelet patients - temporarily stop warfarin, DOAC, aspirin or clopidogrel during acute dengue; bridge with heparin only if thrombotic risk is very high and after haematology advice; give vitamin K and FFP if the INR is prolonged.
• Healthcare workers and laboratory staff - standard precautions; viraemic blood is infectious by needle-stick.^[1]

The vaccines that learnt from Dengvaxia - CYD-TDV and TAK-003

The Dengvaxia story is the most important cautionary tale in modern vaccinology, and it is an ADE problem, not a manufacturing problem. Sanofi's post-marketing finding (2017) was that Dengvaxia (CYD-TDV), given to seronegative recipients, behaved like a silent primary infection: on later wild-type exposure those recipients showed an excess risk of severe dengue while seropositive recipients stayed protected. The 2018 WHO recommendation followed directly - Dengvaxia is given only to seropositive individuals.^[4]

• Dengvaxia (CYD-TDV, Sanofi) - the first licensed dengue vaccine: a live attenuated tetravalent chimeric vaccine on a yellow-fever backbone, given as three doses at 0, 6 and 12 months. WHO 2018: only for individuals with evidence of past dengue infection (seropositive), in high-burden settings, because seronegative recipients carry increased severe-dengue risk.^[4]
• Qdenga (TAK-003, Takeda) - a live attenuated tetravalent vaccine (attenuated DENV-1 to -4 strains, authorised EU-wide in December 2022), given as two doses 3 months apart. WHO-prequalified in May 2024 (the second dengue vaccine prequalified after Dengvaxia), recommended for children aged 6-16 years in settings with high dengue burden and high transmission intensity. Unlike Dengvaxia it shows efficacy in seronegative recipients - demonstrated against DENV-1 and DENV-2 (vaccine efficacy 53.5% against virologically confirmed dengue and 79.3% against hospitalised dengue over 4.5 years in seronegative participants), though not observed against DENV-3 - the nuance that tempers the Dengvaxia comparison.^[11][3]
• Other candidates - TV003/TV005 (NIH/LNI/Merck) and WRAIR pipelines; an Indian indigenous live-attenuated tetravalent vaccine is in development.^[1]

Steroids in DSS - a Cochrane review of eight RCTs found no evidence of benefit on death or disease progression, and steroids are not recommended routinely; they may be considered for HLH or refractory shock on specialist advice. No licensed antiviral has demonstrated clinical benefit - balapiravir failed its randomised trial, and clinical trials of balapiravir, celgosivir, chloroquine and ribavirin showed limited efficacy - so fluid therapy remains the only specific intervention.^[16][17]

Regional deltas in management:^[1]

• India (NVBDCP/ICMR) - NS1 plus IgM combination RDT at the periphery, paracetamol only, no NSAIDs/aspirin/IM, no prophylactic platelets, isotonic crystalloid for leakage; vector control by source reduction, larviciding and fogging; an indigenous vaccine in development.
• WHO TDR (2012 Handbook) - fluid titration by triage group: oral fluids in Group A, isotonic crystalloid in Group B, crystalloid-first bolus with colloid for refractory shock in Group C.
• Southeast Asia (Singapore, Thailand, Vietnam) - mature programmes with Wolbachia-infected Aedes aegypti release: Wolbachia-carrying mosquitoes have reduced DENV replication, and the World Mosquito Programme has shown large field reductions in dengue incidence.
• The Americas (Brazil, PAHO) - high DENV-2 and DENV-3 burden; integrated vector management; Dengvaxia and Qdenga rollout in Brazil.
• UK (UKHSA) - dengue virus is a laboratory-notifiable organism; NS1 then IgM; mandatory malaria exclusion; mosquito-avoidance advice for travellers.^[1]

How dengue patients come to harm - the preventable list

• Treating "settled fever" as recovery and discharging the patient on the day the fever drops - that is the day they are most likely to shock.^[1]
• Giving NSAIDs, aspirin or an intramuscular injection - the single most preventable cause of severe bleeding (and Reye syndrome in children given aspirin).^[1]
• Monitoring haemoglobin instead of haematocrit - missing the plasma leak entirely, because haemoglobin does not move with leakage.^[2]
• Transfusing prophylactic platelets for a low count alone - reactions, overload, TRALI, and no bleeding benefit.^[1]
• Over-resuscitating with crystalloid - flooding the leaking, then reabsorbing, patient into pulmonary oedema.^[1]
• Missing a co-pathogen - malaria, leptospirosis or bacterial sepsis, any of which can kill in hours while you watch a platelet count.^[1]
• Stopping fluids too early in shock - the patient who still has a narrow pulse pressure needs more colloid, not a ward bed.^[1]
• Anticoagulating or antiplatelet therapy left running through an acute dengue illness, ending in major haemorrhage.^[1]

The mnemonic, and the mantra

^[1]

The mantra: worse as the fever falls, haematocrit not haemoglobin, paracetamol only, fluids are the cure.^[1][2]

Ward-round test - three stems, thirty seconds each

Stem 1 - the child who deteriorates as the fever settles (answer)ShowHide

The 9-year-old from the top of the topic: day 4, fever now settled, cold and restless, BP 100/88 (pulse pressure 12), capillary refill 4 seconds, oliguric, platelets 40,000, haematocrit climbing. The registrar says he is improving. What is happening and what do you do in the next 30 minutes? Model: This is the critical phase progressing to dengue shock syndrome - a dropping fever with a worsening patient is the danger window, and "the fever has gone" is the trap, not the reassurance. Admit to HDU/ICU (WHO Group C), establish IV access, and start fluid resuscitation - fluid resuscitation to counteract massive plasma leakage is the mainstay of treatment in DSS, and the WHO-guideline regimen tested in children is Ringer's lactate 20 mL/kg over 15 minutes followed by stepwise rate reduction (10, 7, 5, 3 mL/kg/h). Reassess perfusion and haematocrit continuously, and beware secondary fluid overload from the large volumes. Monitor the haematocrit with the platelet count. No NSAIDs, no aspirin - paracetamol only.^[7][6][14]

Stem 2 - the patient given ibuprofen who bleeds (answer)ShowHide

A 28-year-old with day 2 of classic dengue fever was given ibuprofen by a chemist for the "break-bone" pain. He now has haematemesis and a platelet count of 18,000. What went wrong, and what is the immediate management? Model: NSAIDs are contraindicated in suspected dengue - WHO and CDC consensus guidelines recommend paracetamol for pain and fever and contraindicate NSAIDs because of potentially increased bleeding risk on a background of thrombocytopenia. Stop the NSAID immediately and switch to paracetamol, protect the airway if haematemesis is ongoing, and treat significant bleeding with blood products on haematology advice. What you do not do is transfuse platelets for the count alone - the randomised trial in adults with platelets at or under 20 000 per microlitre showed no significant reduction in bleeding (21% vs 26%, p = 0.16) and more adverse events (13 vs 2) with prophylactic transfusion. Exclude a co-pathogen.^[14][5]

Stem 3 - the narrow-pulse-pressure shock (answer)ShowHide

A 14-year-old, day 5 of dengue, is cold, clammy and confused. Pulse is 140 and thready, BP 90/78, no urine for 6 hours, haematocrit 52%. What grade is this, and what fluid, what dose, and how fast? Model: This is dengue shock syndrome - hypotension, cold peripheries, oliguria and altered consciousness on a climbing haematocrit are the shock state of severe plasma leakage. Resuscitate now: fluid resuscitation to counteract massive plasma leakage is the mainstay of treatment. A randomised comparison of four fluids in the first hour (dextran, gelatin, lactated Ringer's, normal saline) found no clear advantage to any one of them, while an earlier trial found colloids (dextran 70, gelafundin) restored cardiac index and blood pressure and normalised haematocrit more rapidly than crystalloids. The WHO-guideline paediatric regimen is Ringer's lactate 20 mL/kg over 15 minutes, then stepwise decreases over 12 hours - and the trial of that regimen warns that such volumes can induce secondary fluid overload, so taper aggressively the moment perfusion and haematocrit recover.^[8][7][6]

References18ShowHide

1. [1]Guzman MG, Gubler DJ, Izquierdo A, Martinez E, Halstead SB. Dengue infection Nat Rev Dis Primers, 2016.PMID 27534439
2. [2]Wang WH, Urbina AN, Chang MR, et al. Dengue hemorrhagic fever - A systemic literature review of current perspectives on pathogenesis, prevention and control J Microbiol Immunol Infect, 2020.PMID 32265181
3. [3]Wilder-Smith A, Ooi EE, Horstick O, et al. Dengue Lancet, 2019.PMID 30696575
4. [4]World Health Organization. Dengue vaccine: WHO position paper, September 2018 - Recommendations Vaccine, 2019.PMID 30424888
5. [5]Lye DC, Archuleta S, Syed-Omar SF, et al. Prophylactic platelet transfusion plus supportive care versus supportive care alone in adults with dengue and thrombocytopenia: a multicentre, open-label, randomised, superiority trial Lancet, 2017.PMID 28283286
6. [6]Somasetia DH, Setiati TE, Sjahrodji AM, et al. Early resuscitation of dengue shock syndrome in children with hyperosmolar sodium-lactate: a randomized single-blind clinical trial of efficacy and safety Crit Care, 2014.PMID 25189175
7. [7]Dung NM, Day NP, Tam DT, et al. Fluid replacement in dengue shock syndrome: a randomized, double-blind comparison of four intravenous-fluid regimens Clin Infect Dis, 1999.PMID 10589889
8. [8]Ngo NT, Cao XT, Kneen R, et al. Acute management of dengue shock syndrome: a randomized double-blind comparison of 4 intravenous fluid regimens in the first hour Clin Infect Dis, 2001.PMID 11170909
9. [9]Madanayake P, Jayawardena A, Wijekoon SL, et al. Fluid requirement in adult dengue haemorrhagic fever patients during the critical phase of the illness: an observational study BMC Infect Dis, 2021.PMID 33743614
10. [10]Alexander N, Balmaseda A, Coelho IC, et al. Multicentre prospective study on dengue classification in four South-east Asian and three Latin American countries Trop Med Int Health, 2011.PMID 21624014
11. [11]Tricou V, Saez-Llorens X, Yu D, et al. Long-term efficacy and safety of a tetravalent dengue vaccine (TAK-003): 4.5-year results from a phase 3, randomised, double-blind, placebo-controlled trial Lancet Glob Health, 2024.PMID 38245116
12. [12]Sivasubramanian K, Bharath R R, Vajravelu LK, et al. Key Laboratory Markers for Early Detection of Severe Dengue Viruses, 2025.PMID 40431672
13. [13]Htun TP, Xiong Z, Pang J. Clinical signs and symptoms associated with WHO severe dengue classification: a systematic review and meta-analysis Emerg Microbes Infect, 2021.PMID 34036893
14. [14]Kellstein D, Fernandes L. Symptomatic treatment of dengue: should the NSAID contraindication be reconsidered? Postgrad Med, 2019.PMID 30575425
15. [15]Sandeep M, Padhi BK, Yella SST, et al. Myocarditis manifestations in dengue cases: A systematic review and meta-analysis J Infect Public Health, 2023.PMID 37738692
16. [16]Zhang F, Kramer CV. Corticosteroids for dengue infection Cochrane Database Syst Rev, 2014.PMID 24984082
17. [17]Shamsabadi S, Dayhimi A, Salari M, Pazoki Toroudi H. Evaluating antiviral candidates for dengue virus infection: A review Parasite Epidemiol Control, 2026.PMID 42004976
18. [18]Pillay K, Keddie SH, Fitchett E, et al. Evaluating the performance of common reference laboratory tests for acute dengue diagnosis: a systematic review and meta-analysis of RT-PCR, NS1 ELISA, and IgM ELISA Lancet Microbe, 2025.PMID 40209729