Professional Documents
Culture Documents
WHO Handbook For Management of Dengue 2012
WHO Handbook For Management of Dengue 2012
Management of Dengue
WHO Library Cataloguing-in-Publication Data
Publications of the World Health Organization are available on the WHO web site (www.
who.int) or can be purchased from WHO Press, World Health Organization, 20 Avenue
Appia, 1211 Geneva 27, Switzerland (tel.: +41 22 791 3264; fax: +41 22 791 4857; e-mail:
[email protected]).
Requests for permission to reproduce or translate WHO publications – whether for sale or
for noncommercial distribution – should be addressed to WHO Press through the WHO
web site (http://www.who.int/about/licensing/copyright_form/en/index.html).
The designations employed and the presentation of the material in this publication do
not imply the expression of any opinion whatsoever on the part of the World Health
Organization concerning the legal status of any country, territory, city or area or of its
authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on
maps represent approximate border lines for which there may not yet be full agreement.
The mention of specific companies or of certain manufacturers’ products does not imply
that they are endorsed or recommended by the World Health Organization in preference
to others of a similar nature that are not mentioned. Errors and omissions excepted, the
names of proprietary products are distinguished by initial capital letters.
All reasonable precautions have been taken by the World Health Organization to verify
the information contained in this publication. However, the published material is being
distributed without warranty of any kind, either expressed or implied. The responsibility
for the interpretation and use of the material lies with the reader. In no event shall the
World Health Organization be liable for damages arising from its use.
ii
2.3.2
Fluid overload ............................................................................................... 44
2.3.3
Co-infections and nosocomial infections ...................................................... 49
2.3.4
Haemophagocytic syndrome ........................................................................ 49
2.3.5
Supportive care and adjuvant therapy .......................................................... 49
2.3.6
References ................................................................................................... 51
2.4
Treatment of dengue in specific risk groups ........................................................ 52
2.4.1
Dengue in adults – some specific issues ..................................................... 52
2.4.2
Dengue in the elderly.................................................................................... 53
2.4.3
Dengue in cases with co-morbidities ............................................................ 54
2.4.4
References ................................................................................................... 58
2.4.5
Dengue in pregnancy ................................................................................... 59
2.4.6
References ................................................................................................... 62
2.4.7
Dengue in paediatric cases .......................................................................... 63
2.4.8
References ................................................................................................... 67
3.
Pitfalls in the management of dengue and solutions .................................................. 69
3.1
Frontline physicians ............................................................................................. 69
3.2
At the emergency department ............................................................................. 73
3.3
Shock and its many manifestations ..................................................................... 73
3.4
Parenteral fluid therapy........................................................................................ 75
3.5
Urine output ......................................................................................................... 77
3.6
Blood transfusions for severe bleeding ............................................................... 79
3.7
References .......................................................................................................... 82
4.
Case studies ............................................................................................................... 83
4.1
Case study one .................................................................................................... 83
4.2
Case study two .................................................................................................... 87
4.3
Case study three.................................................................................................. 93
4.4
Case study four.................................................................................................... 98
5.
Annex ........................................................................................................................ 102
5.1
Textbox A: Good and bad clinical practice ........................................................ 102
5.2
Textbox B: Differential diagnosis of dengue fever ............................................. 103
5.3
Textbox C: Warning signs.................................................................................. 104
iii
5.4
Textbox D: Haemodynamic assessment:
continuum of haemodynamic changes .............................................................. 104
5.5
Textbox E: Admission criteria ............................................................................ 105
5.6
Textbox F: Discharge criteria ............................................................................. 105
5.7
Textbox G: Home-care card for dengue (for patient or adult caregiver) ............ 106
5.8
Textbox H: Calculations for normal maintenance of intravenous fluid infusion . 107
5.9
Textbox I: Hourly maintenance fluid regime based on ideal body weight .......... 107
5.10
Textbox J: Estimated ideal body weight for overweight or obese adults ........... 108
5.11
Textbox K: Example of a monitoring chart for dengue:
Vital signs and hourly fluid monitoring of dengue patients ................................ 109
5.12
Textbox L: Choice of intravenous fluids for resuscitation .................................. 111
iv
Foreword
Since publication of the new edition of Dengue: Guidelines for diagnosis, treatment,
prevention and control by the World Health Organization (WHO) in 20091, the need to
provide more training to health-care workers in this area has become increasingly evident.
Existing training materials need to include more detail to help clinicians recognize the
evolution of the course of dengue disease in its various forms of severity, and to enable
them to apply the knowledge and principles of management accordingly.
With this aim in mind and following previous successful collaborations, the WHO
Department of Control of Neglected Tropical Diseases (WHO/NTD) and the Special
Programme for Research and Training in Tropical Diseases (WHO/TDR), set out to
develop new training materials.
This handbook has been produced to be made widely available to health-care practitioners
at all levels. Aspects of managing severe cases of dengue are also described for
practitioners at higher levels of health care. Additional and more specific guidance on the
various areas related to clinical management of dengue (from other sources in WHO and
elsewhere) are cited in the reference sections.
Contributions and reviews, by many experts both within and outside WHO, have facilitated
the preparation of this publication through consultative and peer review processes. We are
most grateful to all contributors who are listed in the acknowledgements section.
This handbook is not intended to replace national treatment training materials and
guidelines but it aims to assist in the development of such materials produced at a local,
national or regional level. All information is up-to-date at the time of writing, to the best
knowledge of the authors.
1
Available from: http://whqlibdoc.who.int/publications/2009/9789241547871_eng.pdf, last accessed July 2012.
v
Methodology
Writing team
Each chapter was allocated to one lead writer who received a small fee for their work. Declarations
of interest were obtained from all lead writers and no conflicting interests were declared as a result.
The lead writers were chosen because of their expertise in the field and their willingness to
undertake the work.
Peer review
All the chapters were submitted for peer review. The group of peer reviewers was determined by
the coordinator and lead writers in consensus, not excluding any potential peer reviewer for a
particular view. The peer reviewers were not paid for their work. Declarations of interest were
obtained from all peer reviewers and no conflicting interests were declared.
For each chapter, resolution of disputed issues arising from the comments of the peer reviewers
was achieved by electronic mail discussion within the group of lead writers.
Acknowledgements
Dr Lucy Chai See Lum of the University of Malaya, Kuala Lumpur, Malaysia was the contract holder
for the development of the handbook. Funds were made available by the World Health
Organization’s Department of Control of Neglected Tropical Diseases (WHO/NTD). Dr Lum was
responsible for coordinating the lead authors of all chapters.
Dr Silvia Runge-Ranzinger coordinated the finalization of the handbook with the help of
Dr Olaf Horstick (WHO/TDR) and Dr Raman Velayudhan (WHO/NTD).
The following individuals reviewed and edited the comments of peer reviewers:
Dr Lucy Chai See Lum; Dr Lian Huat Tan; Dr Silvia Runge-Ranzinger.
vi
Abbreviations
vii
LVEDD Left ventricular end-diastolic diameters
MAP Mean arterial pressure
NS1 Ag Non-structural protein 1 antigen
NSAID Non-steroidal anti-inflammatory agent
NT Neutralization test
ORS Oral rehydration solution
PaCO2 Partial pressure of carbon dioxide
PCR Polymerase chain reaction
PEEP Positive end-expiratory pressure
PICU Paediatric intensive care unit
PLT Platelets
PR Pulse rate
PT Prothrombin time
PTT Partial thromboplastin time
RA Ringer’s acetate
RBC Red blood cell
RL Ringer’s lactate
RR Respiratory rate
RT-PCR Reverse transcriptase polymerase chain reaction
SARS Severe acute respiratory syndrome
SD Standard deviation
SpO2 Oxygen saturation
TWBC Total white blood count
URTI Upper respiratory tract infection
WBC White blood cell
WHO World Health Organization
WHO/TDR Special Programme for Research and Training in Tropical Diseases
WHO/NTD Department of Control of Neglected Tropical Diseases
YF Yellow fever
viii
1. Overview, differential diagnosis and dengue diagnostics
1
1.1.1 Febrile phase
Patients typically develop a high-grade fever suddenly. This acute febrile phase usually
lasts 2−7 days and is often accompanied by facial flushing, skin erythema, generalized
body ache, myalgia, arthralgia, retro-orbital eye pain, photophobia, rubeliform exanthema
and headache (1). Some patients may have a sore throat, an injected pharynx, and
conjunctival injection. Anorexia, nausea and vomiting are common.
It can be difficult to distinguish dengue clinically from non-dengue febrile diseases in the
early febrile phase. A positive tourniquet test in this phase indicates an increased
probability of dengue (3, 4). However, these clinical features do not predict the severity of
disease. Therefore it is crucial to monitor for warning signs and other clinical parameters
(Textbox C) in order to recognize progression to the critical phase.
Mild haemorrhagic manifestations such as petechiae and mucosal membrane bleeding
(e.g. of the nose and gums) may be seen (3, 5). Easy bruising and bleeding at
venepuncture sites is present in some cases. Massive vaginal bleeding (in women of
childbearing age) and gastrointestinal bleeding may occur during this phase although this is
not common (5). The liver may be enlarged and tender after a few days of fever (3). The
earliest abnormality in the full blood count is a progressive decrease in total white cell
count, which should alert the physician to a high probability of dengue (3). In addition to
these somatic symptoms, with the onset of fever patients may suffer an acute and
progressive loss in their ability to perform their daily functions such as schooling, work and
interpersonal relations (6).
1.1.2 Critical phase
During the transition from the febrile to afebrile phase, patients without an increase in
capillary permeability will improve without going through the critical phase. Instead of
improving with the subsidence of high fever; patients with increased capillary permeability
may manifest with the warning signs, mostly as a result of plasma leakage.
The warning signs (summarized in Textbox C) mark the beginning of the critical phase.
These patients become worse around the time of defervescence, when the temperature
drops to 37.5−38°C or less and remains below this level, usually on days 3–8 of illness.
Progressive leukopenia (3) followed by a rapid decrease in platelet count usually precedes
plasma leakage. An increasing haematocrit above the baseline may be one of the earliest
additional signs (7, 8). The period of clinically significant plasma leakage usually lasts
24−48 hours. The degree of plasma leakage varies. A rising haematocrit precedes changes
in blood pressure (BP) and pulse volume.
The degree of haemoconcentration above the baseline haematocrit reflects the severity of
plasma leakage; however, this may be reduced by early intravenous fluid therapy. Hence,
frequent haematocrit determinations are essential because they signal the need for
possible adjustments to intravenous fluid therapy. Pleural effusion and ascites are usually
only clinically detectable after intravenous fluid therapy, unless plasma leakage is
significant. A right lateral decubitus chest radiograph, ultrasound detection of free fluid in
the chest or abdomen, or gall bladder wall oedema may precede clinical detection. In
addition to the plasma leakage, haemorrhagic manifestations such as easy bruising and
bleeding at venepuncture sites occur frequently.
2
If shock occurs when a critical volume of plasma is lost through leakage, it is often
preceded by warning signs. The body temperature may be subnormal when shock occurs.
With profound and/or prolonged shock, hypoperfusion results in metabolic acidosis,
progressive organ impairment, and disseminated intravascular coagulation. This in turn can
lead to severe haemorrhage causing the haematocrit to decrease in severe shock. Instead
of the leukopenia usually seen during this phase of dengue, the total white cell count may
increase as a stress response in patients with severe bleeding. In addition, severe organ
involvement may develop such as severe hepatitis, encephalitis, myocarditis, and/or severe
bleeding, without obvious plasma leakage or shock (9).
Some patients progress to the critical phase of plasma leakage and shock before
defervescence; in these patients a rising haematocrit and rapid onset of thrombocytopenia
or the warning signs, indicate the onset of plasma leakage. Cases of dengue with warning
signs will usually recover with intravenous rehydration. Some cases will deteriorate to
severe dengue (see Section 1.1.4).
Warning signs of dengue
Warning signs usually precede the manifestations of shock and appear towards the end of
the febrile phase, usually between days 3–7 of illness. Persistent vomiting and severe
abdominal pain are early indications of plasma leakage and become increasingly worse as
the patient progresses to the shock state. The patient becomes increasingly lethargic but
usually remains mentally alert. These symptoms may persist into the shock stage.
Weakness, dizziness or postural hypotension occur during the shock state. Spontaneous
mucosal bleeding or bleeding at previous venepuncture sites are important haemorrhagic
manifestations. Increasing liver size and a tender liver is frequently observed. However,
clinical fluid accumulation may only be detected if plasma loss is significant or after
treatment with intravenous fluids. A rapid and progressive decrease in platelet count to
about 100 000 cells/mm3 and a rising haematocrit above the baseline may be the earliest
sign of plasma leakage. This is usually preceded by leukopenia (≤ 5000 cells/mm3) (4).
1.1.3 Recovery phase
As the patient survives the 24−48 hour critical phase, a gradual reabsorption of
extravascular compartment fluid takes place in the following 48−72 hours. General well-
being improves, appetite returns, gastrointestinal symptoms abate, haemodynamic status
stabilizes, and diuresis ensues. Some patients have a confluent erythematous or petechial
rash with small areas of normal skin, described as “isles of white in the sea of red” (10).
Some may experience generalized pruritus. Bradycardia and electrocardiographic changes
are common during this stage. The haematocrit stabilizes or may be lower due to the
dilutional effect of reabsorbed fluid. The white blood cell count usually starts to rise soon
after defervescence but the recovery of the platelet count is typically later than that of the
white blood cell count. Respiratory distress from massive pleural effusion and ascites,
pulmonary oedema or congestive heart failure will occur during the critical and/or recovery
phases if excessive intravenous fluids have been administered. Clinical problems during
the different phases of dengue are summarized in Table 1.
3
Table 1. Medical complications seen in the febrile, critical and recovery phases of
dengue
1 Febrile phase Dehydration: high fever may cause neurological disturbances and febrile seizures
in young children
2 Critical phase Shock from plasma leakage: severe haemorrhage; organ impairment
3 Recovery phase Hypervolaemia (only if intravenous fluid therapy has been excessive and/or has
extended into this period) and acute pulmonary oedema
4
Patients who have dengue and are in compensated shock often remain conscious and
lucid. The inexperienced physician may measure a normal systolic pressure and a normal
pulse oximetry (SpO2 95–100%) in a conscious patient and underestimate the critical state
of the patient.
Worsening hypovolaemic shock manifests as increasing tachycardia and peripheral
vasoconstriction. Not only are the extremities cold and cyanosed but the limbs become
mottled, cold and clammy. By this stage the breathing becomes more rapid and increases
in depth − a compensation for the metabolic acidosis (Kussmaul’s breathing). Finally, there
is decompensation, both systolic and diastolic BPs disappear suddenly and dramatically,
and the patient is said to have hypotensive or decompensated shock.
At this time the peripheral pulses disappear while the central pulse (femoral) will be weak.
Hypotension develops when physiologic attempts to maintain systolic BP and perfusion are
no longer effective. One key clinical sign of this deterioration is a change in mental state as
brain perfusion declines. The patient becomes restless, confused and extremely lethargic.
Seizures may occur and agitation may alternate with lethargy. On the other hand, children
and young adults have been known to have a clear mental status even in profound shock.
Adults have been known to be able to work until the stage of profound shock is reached.
The failure of infants and children to recognize, focus or make eye contact with parents
may be an early ominous sign of cortical hypoperfusion, as is the failure to respond to
painful stimuli such as venepuncture. Parents may be the first to recognize these signs −
but they may be unable to describe them, other than to say something is wrong. Listen to
parents! Hypotension is a late finding and signals an imminent total cardiorespiratory
collapse.
Prolonged hypotensive shock and hypoxia lead to severe metabolic acidosis, multiple
organ failure and an extremely difficult clinical course (12) (Textbox D). It may take a few
hours for patients to progress from warning signs to compensated shock and another few
hours for compensated shock to progress to hypotensive shock, but only minutes for
hypotensive shock to progress to cardiorespiratory collapse and cardiac arrest.
Hypotension is associated with prolonged shock which is often complicated by major
bleeding (12). Patients with severe dengue have varying degrees of coagulation
abnormalities, but these are usually not sufficient to cause major bleeding (13). When
major bleeding does occur, it is almost always associated with profound shock since this, in
combination with thrombocytopenia, hypoxia and acidosis, can lead to multiple organ
failure and advanced disseminated intravascular coagulation.
Massive bleeding may occur without prolonged shock in instances when acetylsalicylic acid
(aspirin), ibuprofen, or corticosteroids have been taken. Bleeding may occur in patients with
previous peptic or duodenal ulcers (14, 15). Acute liver and renal failure and
encephalopathy may be present in severe shock; these have been described even in the
absence of severe plasma leakage or shock (16–21). Cardiomyopathy and encephalitis
have also been reported in a few dengue case series (22–25). However, most deaths from
dengue occur in patients with profound and prolonged shock resulting from plasma leakage
and complicated by bleeding and/or fluid overload.
Patients with severe plasma leakage may not have shock if prompt fluid replacement has
been carried out. Instead, they manifest with respiratory distress due to massive pleural
effusion and ascites, which can also be exacerbated by unguided intravenous fluid therapy.
5
1.1.5 References
1. Rigau-Pérez JG et al., Dengue and dengue haemorrhagic fever. Lancet, 1998, 352:971–977.
2. Yip WCL. Dengue haemorrhagic fever: current approaches to management. Medical Progress, 1980, 7:13.
3. Kalayanarooj S et al., Early clinical and laboratory indicators of acute dengue illness. Journal of Infectious
Diseases, 1997, 176:313–321.
4. Cao XT et al., Evaluation of the World Health Organization standard tourniquet test in the diagnosis of
dengue infection in Vietnam. Tropical Medicine and International Health, 2002, 7:125–132.
5. Balmaseda A et al., Assessment of the World Health Organization scheme for classification of dengue
severity in Nicaragua. American Journal of Tropical Medicine and Hygiene, 2005, 73:1059–1062.
6. Lum LCS et al., Quality of life of dengue patients. American Journal of Tropical Medicine and Hygiene, 2008,
78(6):862–867.
7. Srikiatkhachorn A et al., Natural history of plasma leakage in dengue hemorrhagic fever: a serial ultrasonic
study. The Pediatric Infectious Disease Journal, 2007, 26(4):283−290.
8. Nimmannitya S et al., Dengue and chikungunya virus infection in man in Thailand, 1962−64. Observations on
hospitalized patients with haemorrhagic fever. American Journal of Tropical Medicine and Hygiene,
1969, 18(6):954−971.
9. Martinez-Torres E, Polanco-Anaya AC, Pleites-Sandoval EB. Why and how children with dengue die?
Revista cubana de medicina tropical, 2008, 60(1):40−47.
10. Nimmannitya S. Clinical spectrum and management of dengue haemorrhagic fever. Southeast Asian
Journal of Tropical Medicine and Public Health, 1987, 18(3):392−397.
11. Pediatric Advanced Life Support (PALS) Provider Manual, Dallas, American Heart Association, 2006.
12. Lum LCS et al., Risk factors for hemorrhage in severe dengue infection. Journal of Pediatrics, 2002,
140:629–631.
13. Wills BA et al., Coagulation abnormalities in dengue hemorrhagic fever: Serial investigations in 167
Vietnamese children with dengue shock syndrome. Clinical Infectious Diseases, 2002, 35:277–285.
14. Tsai CJ et al., Upper gastrointestinal bleeding in dengue fever. American Journal of Gastroenterolgy, 1991,
86(1):33–35.
15. Chiu YC et al., Endoscopic findings and management of dengue patients with upper gastrointestinal
bleeding. American Journal of Tropical Medicine and Hygiene, 2005, 73(2):441–444.
16. Lum LC et al., Fulminant hepatitis in dengue infection. Southeast Asian Journal of Tropical Medicine and
Public Health, 1993, 24(3):467–471.
17. Nguyen TL, Nguyen TH, Tieu NT. The impact of dengue haemorrhagic fever on liver function. Research in
Virology, 1997, 148(4):273–277.
18. Poovorawan Y et al., Dengue virus infection: a major cause of acute hepatic failure in Thai children. Annals
of Tropical Paediatrics, 2006, 26(1):17–23.
19. Ooi ET et al., Gastrointestinal manifestations of dengue infection in adults. Medical Journal of Malaysia,
2008, 63(5):401–405.
20. Kumar R et al., Prevalence of dengue infection in north Indian children with acute hepatic failure. Annals of
Hepatology. 2008, 7(1):59–62.
21. Trung DT et al., Liver involvement associated with dengue infection in adults in Vietnam. American Journal
of Tropical Medicine and Hygiene. 2010, 83(4):774–780.
22. Lum LCS et al., Dengue encephalitis: a true entity? American Journal of Tropical Medicine and Hygiene,
1995, 54(3):256–923.
23. Miagostovich MP, Ramos RG, Nicol AF. Retrospective study on dengue fatal cases. Clinical
Neuropathology, 1997, 16:204–208.
24. Solomon T et al., Neurological manifestations of dengue infection. Lancet, 2000, 355:1053–1059.
25. Pancharoen C, Thisyakorn U. Neurological manifestations in dengue patients. Southeast Asian Journal of
Tropical Medicine and Public Health, 2001, 32(2):341–345.
6
1.2 Dengue case classification
Changes in the epidemiology of dengue, especially with an increasing number of cases in
adults (with and without co-morbidities) and the expansion of dengue into other regions of
the world, has led to problems with the use of the existing WHO classification.
This clinical guide uses three categories for case management (A, B, C), based on the
model of case classification that follows (Figure 2) after a patient has fulfilled the criteria for
probable dengue.
Fig. 2. Dengue case classification by severity
WHO/TDR 2009
dengue
(important when no sign of plasma * Requiring strict observation consciousness
leakage) and medical intervention • Heart and other organs
ALT = alanine aminotransferase; AST = aspartate aminotransferase; CNS = central nervous system; DSS =
dengue shock syndrome; HCT = haematocrit
8
1.2.2 References
1. Guha-Sapir D, Schimmer B. Dengue fever: new paradigms for a changing epidemiology. Emerging Themes
in Epidemiology, 2005, 2:1.
2. Deen JL et al., The WHO dengue classification and case definitions: time for a reassessment. Lancet, 2006,
368:170–173.
3. Rigau-Perez J. Severe dengue: the need for new case definitions. Lancet Infectious Diseases, 2006, 6:297–
302.
4. Bandyopadhyay S, Lum LC, Kroeger A. Classifying dengue: a review of the difficulties in using the WHO
case classification for dengue haemorrhagic fever. Tropical Medicine and International Health, 2006,
11(8):1238–1255.
5. Santamaria R et al., Comparison and critical appraisal of dengue clinical guidelines and their use in Asia and
Latin America. International Health, 2009, 1:133–140.
6. Dengue. Guidelines for diagnosis, treatment prevention and control, Geneva, World Health Organization,
2009, WHO/HTM/NTD/DEN/2009.
9
1.2.3 Differential diagnoses of dengue
A number of infectious and non infectious diseases mimic dengue and severe dengue. It is
thus necessary for clinicians to be familiar with the epidemiological characteristics of febrile
diseases in the locality. Clinical manifestations associated with fever, epidemiological
information and virological tests (if available) are particularly useful in patients with acute
undifferentiated fever (1, 2).
Conditions that present with an influenza-like syndrome, such as influenza itself, measles,
chikungunya, infectious mononucleosis and human immunodeficiency virus (HIV)
seroconversion illness, may mimic the febrile phase of dengue (Textbox B). Upper
respiratory symptoms such as rhinitis and cough are always present in influenza, in
addition to fever, headache and body pains which are also commonly seen in dengue.
Patients with dengue usually have gastrointestinal symptoms (i.e. abdominal discomfort,
vomiting and sometimes diarrhoea) during the febrile phase. Coinfections with both dengue
and influenza viruses make the differential diagnosis more difficult. Only rhinitis and/or
nasal congestion are frankly prominent in influenza cases (3).
Infection with adenoviruses may cause fever associated with rash, abdominal pain,
leukopenia, lymphopenia and organ impairment (liver, heart). Bleeding can occur, but is not
frequent. Rhinitis or pharyngitis, cough and other respiratory symptoms are present,
associated with cervical adenopathies in most patients (4). The diagnosis of severe acute
respiratory syndrome (SARS) is difficult early in the illness. When large SARS outbreaks
occurred in dengue-endemic countries, laboratory features that were highly predictive of
dengue diagnosis were leukopenia and low platelet count (5).
While fever, arthralgia, rash, malaise and leukopenia are common in both chikungunya and
dengue, symmetric arthritis of small joints is pathognomonic of the former. A bleeding
tendency and pronounced thrombocytopenia are more frequent in dengue (6, 7).
Primary infection with HIV may mimic dengue with high fever, malaise, rash and
generalized adenopathies (8). Splenomegaly and prolonged fever should prompt the
consideration of malaria and typhoid in the differential diagnoses. Fever, malaise, vomiting,
liver enlargement and elevated liver enzymes may be misdiagnosed as infectious hepatitis,
and vice-versa (9). Evidence of plasma leakage during defervescence and
thrombocytopenia are more in keeping with dengue.
The rash associated with measles and rubella has a particular distribution from the head to
the trunk and extremities, but in dengue the rash usually first appears on the trunk and later
extends to the face and extremities (10). Although both diseases may have common signs
and symptoms, including myalgia and arthralgia, measles patients always have a cough,
rhinitis and conjunctivitis. Fever, rash and adenopathies may be seen in dengue, rubella,
erythema infectiosum caused by parvovirus B19 (11), and herpes virus type 6 (12). Other
exanthems such as enteroviruses, infectious mononucleosis, scarlet fever and Kawasaki
disease are associated with particular characteristics.
Sepsis and meningococcal disease should be considered in shock patients because of the
need for urgent treatment with specific antibiotics. Common symptoms and signs with
dengue cases are fever, rash, petechiae, bleedings and shock associated with leukopenia
(particularly in severe gram-negative sepsis and poor prognosis meningococcaemia) and
thrombocytopenia (13). In septic shock, the temperature is usually high, although it could
be subnormal in the late stages. Bounding pulses with warm extremities are present in
10
early septic shock. In dengue patients, shock usually occurs after defervescence; hence
the temperature is often subnormal or normal, pulse volume is small, and pulse pressure is
narrowed with the patient having cold extremities. Clinical and radiographic signs of plasma
leakage and progressive haemoconcentration in severe dengue cases are useful
distinguishing features. Another helpful tool to differentiate dengue from these other
diseases is to determine the sequence of signs and symptoms, including warning signs
during defervescence that frequently announce severe dengue.
Clinical distinction between leptospirosis and dengue may be challenging, particularly when
both epidemics are concurrent (14). Delayed antibiotic therapy is associated with mortality
in leptospirosis. Jaundice is more often associated with leptospirosis, but ocular pain,
arthralgia and diarrhoea could be present as well (15). Leptospirosis is frequently
associated with professional activity (e.g. working with garbage or in agriculture) or with a
history of certain pastimes (e.g. trekking to waterfalls or playing water sports). Pulmonary
haemorrhage is a particular form of leptospirosis without jaundice that has some common
signs and symptoms with severe dengue; these include fever, thrombocytopenia, shock
and massive bleeding in the lungs (16). Pulmonary haemorrhage is uncommon in dengue;
evidence of plasma leak such as pleural effusion or ascites would suggest the diagnosis of
dengue.
Leukopenia and thrombocytopenia, with or without bleeding, may be clinical manifestations
of infectious diseases such as malaria, leptospirosis, typhoid, typhus, bacterial sepsis and
acute HIV-seroconversion illness. Leukopenia and thrombocytopenia may be present in
non-infectious diseases such as systemic lupus and other systemic auto-immune diseases;
acute leukaemia (17) and other haematological disorders such as Henoch-Schönlein
purpura and thrombocytopenic purpuric syndromes, mainly thrombotic thrombocytopenic
purpura and immunological thrombocytopenic purpura.
During the critical phase of dengue, patients with plasma leakage or shock may present
with severe abdominal pain when the fever subsides. The severe abdominal pain may
mimic an acute abdominal condition such as in acute appendicitis. Ultrasound studies in
these patients have shown fluid collection around the appendix. Other abdominal signs
such as right iliac fossa tenderness and rebound tenderness disappear after a few days of
conservative management (18).
Another misdiagnosis is acute (alithiasis) cholecystitis, with the abdominal ultrasonograph
showing thickening/oedema of the wall of the gallbladder. This is associated with pain in
the subhepatic region, mainly during defervescence. Plasma leakage, not inflammation, is
responsible for these clinical features. Patients who underwent surgery as a result of
misdiagnosis with an acute surgical abdominal condition have been found to have life-
threatening bleedings; some went on to die. Failure to recognize severe abdominal pain as
a warning sign that heralds severe dengue has led to the misdiagnosis of renal lithiasis and
delayed intravenous fluid treatment for dengue shock. A differentiating feature of an acute
surgical abdomen and the severe abdominal pain of dengue shock is that the abdomen in
dengue shock is soft and the pain subsides with fluid resuscitation. Other medical
emergencies presenting with intense and continuous abdominal pain include diabetic
ketoacidosis, renal failure and lactic acidosis. Again, evidence of plasma leakage
(especially after intravenous fluid therapy), thrombocytopenia and bleeding tendencies help
to distinguish dengue from other medical or surgical conditions.
11
The group of viral haemorrhagic fever diseases have bleeding, thrombocytopenia and
shock in common. All have the monocyte/macrophage as the target cell (19). These
diseases are present in different geographic areas, have different vectors (Table 2), have
causative agents that belong to different viral families (20–22), and have different means of
transmission. The main pathogenic difference is that the severity of dengue disease is
mediated by an immunological disorder that enhances viral infection, causing patients to
have complications after the viraemic (febrile) phase, during defervescence or 24 hours
later. An analysis of the particular clinical, epidemiological and vectoral information will lead
to the respective diagnosis.
Yellow fever (YF) is now considered a re-emergent disease in Africa and the Americas.
Aedes aegypti is the common vector of YF and dengue and some clinical features are
common to both diseases. The classic description of YF covers two phases: the febrile
phase and the toxic phase. During the toxic phase, liver damage causing jaundice, renal
insufficiency and central nervous system impairment are life-threatening characteristics.
The disease in adults has been associated with high mortality, a relatively short duration,
symptoms such as headache, back pain, fever, vomiting and nausea, jaundice,
haemorrhages and unconsciousness (23).
Severe dengue with noncardiogenic pulmonary oedema (fluid overload) and pulmonary
distress has clinical signs common to the Hantavirus pulmonary syndrome (HPS). In severe
dengue, and not in HPS, the pulmonary oedema is usually preceded by prolonged or
recurrent shock, associated with bleedings in the lung and other sites, together with signs
of plasma leakage and fluid overload. The initial picture of HPS is very similar to that of
influenza, with fever, myalgia, vomiting and cough associated with dyspnoea at the end of
the first week, and leukocytosis, neutrophilia, thrombocytopenia and elevated haematocrits
(24). Children with HPS may have severe abdominal pain and liver enlargement, but
bleeding is not frequent and they do not present with pulmonary haemorrhage, but with
interstitial and alveolar oedema (25).
12
1.2.4 References
1. Phuong HL et al., Acute undifferentiated fever in Binh Thuan province, Vietnam: imprecise clinical diagnosis
and irrational pharmaco-therapy. Tropical Medicine and International Health, 2006, 11(6): 869–879.
2. Thai KT et al., Clinical, epidemiological and virological features of dengue virus infections in Vietnamese
patients presenting to primary care facilities with acute undifferentiated fever. The Journal of Infection,
2010, 60(3): 229–237.
3. Morens DM. Dengue Outbreak Investigation Group. Dengue in Puerto Rico, 1977: public health response to
characterize and control an epidemic of multiple serotypes. American Journal of Tropical Medicine and
Hygiene, 1986, 35:197–211.
4. Fisher RG, Boyce TG. Nonstreptococcal pharyngitis. In: Moffet´s Pediatric Infectious Diseases. A problem-
oriented approach, 4th ed. Philadelphia, PA, Lippincott Williams and Wilkins, 2005: 34–35.
5. Wilder-Smith A, Earnes A, Paton NI. Use of simple laboratory features to distinguish the early stage of
severe acute respiratory syndrome from dengue fever. Clinical Infectious Diseases, 2004, 39(12):1818–
1823.
6. Kularatne SA et al., Concurrent outbreaks of Chikungunya and Dengue fever in Kandy, Sri Lanka, 2006-07: a
comparative analysis of clinical and laboratory features. Postgraduate Medical Journal, 2009,
85(1005):342–346.
7. Yamamoto K et al., Chikungunya fever from Malaysia. Internal Medicine, 2010, 49(5):501–505.
8. Cabie A et al., Dengue or acute retroviral syndrome? Presse Med, 2000, 29(21):1173–1174.
9. Martinez E. Diagnóstico Diferencial. In: Dengue. Rio de Janeiro, FIOCRUZ, 2005: 189−195.
10. Dietz VJ et al., Diagnosis of measles by clinical case definition in dengue endemic areas: implications for
measles surveillance and control. Bulletin of the World Health Organization, 1992, 70(6):745–750.
11. Oliveira MJC et al., Frequency of measles, rubella, dengue and erythema infectiosum among suspected
cases of measles and rubella in the State of Pernambuco between 2001 and 2004. Revista da
Sociedade Brasileira de Medicina Tropical, 2008, 41(4):338–344.
12. Oliveira SA et al., The aetiology of maculopapular rash diseases in Niteroi, state of Rio de Janeiro, Brazil:
implications for measles surveillance. Epidemiology and Infection, 2001, 127(3):509–516.
13. Martinez E. Meningococcal disease: physiopathology, clinical picture and prognosis. Revista de Hospital de
Niños de Buenos Aires, 1994, 36(158-159):204–214.
14. Flannery B et al., Referral pattern of leptospirosis cases during a large urban epidemic of dengue. American
Journal of Medicine and Tropical Hygiene, 2001, 65(5):657–663.
15. Sanders EJ et al., Increase of leptospirosis in dengue-negative patients after a hurricane in Puerto Rico in
1996. American Journal of Tropical Medicine and Hygiene, 1999, 61(3):399–404.
16. Muñoz F et al., Outbreak of acute febrile illness and pulmonary hemorrhage-Nicaragua 1995. Journal of the
American Medical Association, 1995, 274(21):1668.
17. Gawoski JM, Ooi WW. Dengue fever mimicking plasma cell leukaemia. Archives of Pathology and
Laboratory Medicine, 2003, 127(8):1026–1027.
18. Premaratna R et al., Dengue fever mimicking acute appendicitis. Transactions of the Royal Society of
Tropical Medicine and Hygiene, 2007, 101(7):683-685.
19. Halstead SB. Viral hemorrhagic fevers. Journal of Infectious Diseases, 1981, 143(1):127–129.
20. Salas R et al., Venezuelan haemorrhagic fever. Lancet, 1991, 338:1033–1036.
21. Zaki A. Isolation of a flavivirus related to the tick-borne encephalitis complex from human cases in Saudi
Arabia. Transactions of the Royal Society of Tropical Medicine and Hygiene, 1997, 91(2):179–181.
13
22. Madani TA. Alkhumra virus infection, a new viral hemorrhagic fever in Saudi Arabia. The Journal of
Infection, 2005, 51(2):91–97.
23. Huamán JH et al., Yellow Fever: clinical, epidemiological and serologic study, 1995-1998, Pichanaki’s Rural
Hospital, Peru. 2000. 9th International Congress on Infectious Diseases. Buenos Aires, Argentina, April
10-13, 2000. Abstract 85.018, p. 200.
24. Duchin JS et al., Hantavirus Pulmonary Syndrome: a clinical description of 17 patients with a newly
recognized disease. New England Journal of Medicine, 1994, 330(4):949–955.
25. Soza G et al., Síndrome pulmonar por hantavirus en población infantil. Chile: regiones IX y X. 1998-2000.
Revista chilena de infectología, 2000, 17(3):248–257.
14
1.3 Dengue diagnostics for clinicians
The objectives of dengue laboratory diagnosis are (i) to confirm the clinical diagnosis and
(ii) to provide information for epidemiological surveillance. Laboratory diagnosis is not
necessary for clinical management except in atypical cases or when carrying out differential
diagnosis with other infectious diseases.
Laboratory diagnosis of dengue is made by detecting the virus and/or any of its
components (infective virus, virus genome, dengue antigen) or by investigating the
serological responses present after infection (specifically IgM and IgG levels) (Table 3) (1–
4).
Dengue viruses are RNA viruses belonging to the family flaviviridae, genus flavivirus. The
four dengue viruses (DEN-[1–4]) are serologically related but antigenically and genetically
distinctive (5–7).
Three main aspects should be considered for an adequate dengue diagnosis:
• virological and serological markers in relation to the time of dengue infection;
• type of diagnostic method in relation to clinical illness;
• characteristics of the clinical samples.
15
Virological and serological markers in relation to time of dengue infection (Figure 3)
An incubation period of 4–10 days occurs after the mosquito bites, resulting in an
asymptomatic or symptomatic dengue infection. During this period the virus replicates and
an antibody response is developed. In general, viraemia is detectable in most dengue
cases at the same time that symptoms appear, and is no longer detectable at the time of
defervescence. The development of IgM antibody is coincident with the disappearance of
fever and viraemia (8). Virological and serological markers differ in time evolution and titre
response and according to whether the infection is primary or secondary.
In a primary infection (i.e. when an individual is infected for the first time with a flavivirus),
viraemia develops from 1–2 days before the onset of fever until 4–5 days after.
Accordingly, anti-dengue IgM specific antibodies can be detected 3−6 days after fever
onset. On average, IgM is detected in 50% of cases by days 3–5 after the onset of illness,
this figure increasing to 95–98% for days 6−10. Low levels of IgM are still detectable
around one to three months after fever. In addition, the primary infection is characterized by
slowly increasing but low levels of dengue-specific IgG, becoming elevated at days 9−10.
Low IgG levels persist for decades, an indication of a past dengue infection (1–4, 9, 10).
A totally different picture is observed during a secondary infection, with a rapid and higher
increase of anti-dengue specific IgG antibodies and slower and lower levels of IgM. High
IgG levels remain for 30–40 days. A short-lasting but higher viraemia level characterizes
the secondary infection compared to the primary infection (1–4, 9, 10).
Fig. 3. Virological and serological markers of dengue infection according to time of
illness
IgG
Viremia Viremia
IgM
16
Febrile phase (day 1 to days 4–5 of fever)
The infective virus can be isolated in serum by inoculation in tissue culture (mosquito cell
cultures) and mosquitoes. This method allows for identification of the viral serotype.
Virus genome detection using reverse transcriptase polymerase chain reaction (RT-PCR)
and real-time RT-PCR confirms an acute dengue infection. Both methods have a high
sensitivity and allow serotype identification and quantification of genome copies (1–4, 11–
13). Some studies suggest the presence of a higher number of copies in severe dengue
cases (14–16).
NS1 Ag is a marker of acute dengue infection. Both enzyme-linked immunosorbent assay
(ELISA) and rapid commercial tests are available for NS1 Ag detection. The sensitivity and
specificity of commercial kits in different serotype infections and days of illness are being
evaluated (17–19).
17
Characteristics of the clinical sample
Similar to other enveloped viruses, dengue virus is labile and readily inactivated at
temperatures above 30°C, so care should be taken during transportation and storage of
samples. Serum samples collected during the first 4 days of fever are useful for virus,
genome and dengue antigen detection, thus confirming a dengue infection. Samples
should be rapidly transported at 4°C to the laboratory and be processed as soon as
possible. Sterile serum without anticoagulant is useful. If specimen delivery cannot be
performed in the first 24–48 hours, freezing at –70°C is recommended.
Sera for serological studies should be stored at 4°C for short time periods and at –20°C for
a longer time periods. When serum collection or transportation is not possible, blood
collected on filter papers represents an opportunity for IgM and IgG determination and also
for RNA detection (24, 25).
Tissue specimens collected from fatal cases are useful for virus, genome and antigen
detection. Liver, spleen and lymph nodes are the tissues of choice (26, 27). Tissue
samples should be collected immediately after death and be immediately frozen at –70°C,
or rapidly transported at 4°C to the laboratory for sample processing. Fresh tissues are also
suitable for virus isolation (26–29).
Besides general patient information a summary of clinical and epidemiological data, such
as the date of fever onset, method of sample collection and the type of sample, should
accompany clinical samples (1).
The usefulness of available diagnostic tests depends on the level of health care (see Table
4). At primary-care level, rapid tests for NS1 Ag detection (suggestive of an acute dengue
infection) as well as rapid tests for IgM determination (suggestive of a recent infection), are
useful. As patients access care independent of the period of infection suffered – some
early, some late – a combination of both NS1 Ag and IgM markers is advisable. At district
health centres, both antigen-based tests and serology can be performed using ELISA and
rapid tests. All diagnostic methods should be available at reference centres, including virus
isolation, nucleic acid detection, diagnostics for tissues samples and all serological
techniques (1–4, 30).
18
Table 4. Recommended diagnostic tool according to laboratory service level
IHA Yes
19
1.3.1 References
1. Dengue. Guidelines for diagnosis, treatment prevention and control. Geneva, TDR/WHO, 2009.
WHO/HTM/NTD/DEN/2009.
2. Guzman MG, Rosario D, Kouri G. In: Kalitzky M and Borowski P, eds. Diagnosis of dengue virus infection.
Molecular Biology of the flaviviruses. Horizon Bioscience, UK, 2009.
3. Buchy F et al., Laboratory tests for the diagnosis of dengue virus infection. Geneva, TDR/Scientific Working
Group, 2006. TDR/SWG/08.
4. Guzman MG, Kouri G. Dengue diagnosis, advances and challenges. International Journal of Infectious
Diseases, 2004, 8:69–80.
5. Burke DS, Monath TP. Flaviviruses. In: Knipe DM, Howley PM eds. Fields virology. London-New York-Tokyo,
Lippincott Williams & Wilkins, 2001:1043 –1125.
6. Lindenbach BD, Rice CM. Flaviviridae: The Viruses and Their Replication. In: Knipe DM, Howley PM eds.
Fields virology. London-New York-Tokyo, Lippincott Williams & Wilkins, 2001: 991–1042.
7. Chambers TJ et al., Flavivirus genome organization, expression, and replication. Annual Review of
Microbiology, 1990, 44: 649–688.
8. Vaughn DW et al., Dengue viremia titer, antibody response pattern, and virus serotype correlate with disease
severity. The Journal of Infectious Diseases, 2000, 181:2–9.
9. Dengue and dengue hemorrhagic fever in the Americas: Guidelines for prevention and control. Washington
DC, Pan American Health Organization, 1994: 548.
10. Vazquez S et al., Kinetics of antibodies in sera, saliva, and urine samples from adult patients with primary or
secondary dengue 3 virus infections. International Journal of Infectious Diseases, 2007, 11:256–262.
11. Kumaria R, Chakravarti A. Molecular detection and serotypic characterization of dengue viruses by single-
tube multiplex transcriptase-polymerase chain reaction. Diagnostic Microbiology and Infectious Disease,
2005, 52:311–316.
12. Rosario D et al., Rapid detection and typing of Dengue viruses from clinical samples using Reverse
Transcriptase- Polymerase Chain Reaction. Pan American Journal of Public Health, 1998, 4:1–5.
13. Chien LJ et al., Development of real-Time reverse Transcriptase PCR assays to detect and serotype
dengue viruses. Journal of Clinical Microbiology, 2006, 44:1295–1304.
14. Libraty DH et al., High circulating levels of the dengue virus nonstructural protein NS1 early in dengue
illness correlate with the development of dengue hemorrhagic fever. Journal of Infectious Diseases,
2002, 186:1165–8.
15. Libraty DH et al., Differing influences of virus burden and immune activation on disease severity in
secondary dengue-3 virus infections. Journal of Infectious Diseases, 2002, 185:1213–21.
16. Wang WK et al., High levels of plasma dengue viral load during defervescence in patients with dengue
hemorrhagic fever: implications for pathogenesis. Virology, 2003, 305(2):330–8.
17. Dussart P et al., Evaluation of two new commercial tests for the diagnosis of acute dengue virus infection
using NS1 antigen detection in human serum. PLoS Neglected Tropical Diseases, 2008, 2:e280.
18. Hang VT et al., Diagnostic accuracy of NS1 ELISA and lateral flow rapid tests for dengue sensitivity,
specificity and relationship to viraemia and antibody responses. PLoS Neglected Tropical Diseases,
2009, 3:e360.
19. Kumarasamy V et al., Evaluation of a commercial dengue NS1 antigen-capture ELISA for laboratory
diagnosis of acute dengue virus infection. Journal of Virological Methods, 2007, 140(1-2):75–79.
20. Blacksell SD et al., The comparative accuracy of 8 commercial rapid immunochromatographic assays for
the diagnosis of acute dengue virus infection. Clinical Infectious Diseases, 2006, 42:1127–1134.
21. Hunsperger EA et al., Evaluation of commercially available anti-dengue virus immunoglobulin M tests.
Emerging Infectious Diseases, 2009, 15:436–440.
20
22. Vazquez S et al., Serological markers during dengue 3 primary and secondary infections. Journal of Clinical
Virology, 2005; 33:132–137.
23. Alvarez M et al., Dengue hemorrhagic fever caused by sequential dengue 1-3 virus infections over a long
time interval: Havana Epidemic, 2001-2002. American Journal of Tropical Medicine and Hygiene. 2006,
75:1113–1117.
24. Vazquez S et al., Detection de IgM contra el virus del dengue en sangre entera absorbida en papel de filtro.
Rev Panam Salud Publica [Pan American Journal of Public Health], 1998, 3:174–178.
25. Prado I et al., PCR detection of dengue virus using dried whole blood spotted on filter paper. Journal of
Virological Methods, 2005, 125:75–81.
26. Guzman MG et al., Fatal dengue hemorrhagic fever in Cuba 1997. International Journal of Infectious
Diseases, 1999, 3:130–135.
27. Rosen L et al., Detection of dengue virus RNA by reverse transcription-polymerase chain reaction in the
liver and lymphoid organs but not in the brain in fatal human infection. American Journal of Tropical
Medicine and Hygiene, 1999, 61(5):720–724.
28. Hall WC et al., Demonstration of yellow fever and dengue antigens in formalin-fixed paraffin-embedded
human liver by immunohistochemical analysis. American Journal of Tropical Medicine and Hygiene,
1991, 45:408–417.
29. Limonta D et al., Apoptosis in tissues from fatal dengue shock syndrome. Journal of Clinical Virology, 2007,
40:50–54.
30. Vorndam V, Kuno G. Laboratory diagnosis of dengue virus infections. In: Gubler DJ, Kuno G, eds. Dengue
and dengue hemorrhagic fever. New York, NY, CAB International, 1997: 313–333.
21
2. Recommendations for clinical management
22
The investigation (refer also to Section 1.3)
If facilities are available, a full blood count should be done at the first visit (it may be
normal); and this should be repeated daily until the critical phase is over. The haematocrit
in the early febrile phase could be used as the patient’s own baseline. Decreasing white
blood cell and platelet counts make the diagnosis of dengue very likely. Leukopenia usually
precedes the onset of the critical phase and has been associated with severe disease. A
rapid decrease in platelet count, concomitant with a rising haematocrit compared to the
baseline, is suggestive of progress to the plasma leakage/critical phase of the disease.
These changes are usually preceded by leukopenia (≤ 5000 cells/mm3). In the absence of
the patient’s baseline, age-specific population haematocrit levels could be used as a
surrogate during the critical phase.
If facilities for a full blood count are not available or if resources are limited, such as in
outbreak settings, a full blood count or microhaematocrit should be done at the first visit to
establish the baseline. This should be repeated after the 3rd day of illness and in those with
warning signs and risk factors for severe disease.
Dengue-specific laboratory tests should be performed to confirm the diagnosis. However, it
is not necessary for the acute management of patients, except in cases with unusual
manifestations.
Additional tests should be considered in patients with co-morbidities and severe disease as
indicated. These may include tests of liver function, glucose, serum electrolytes, urea and
creatinine, bicarbonate or lactate, cardiac enzymes, electrocardiogram (ECG) and urine-
specific gravity.
2.1.3 Step III – Disease notification and management decision (Groups A–C)
Disease notification
In dengue-endemic countries, cases of suspected, probable and confirmed dengue should
be notified early so that appropriate public-health measures can be initiated. Laboratory
confirmation is not necessary before notification, but it should be obtained. In non-endemic
countries, usually only confirmed cases will be notified.
Management decisions
Depending on the clinical manifestations and other circumstances, patients may either be
sent home (Group A); be referred for in-hospital management (Group B); or require
emergency treatment and urgent referral (Group C). Figure 4 shows a summary of
management decisions.
23
Fig. 4 Dengue case management algorithm (2)
negative positive
Classification
Co-existing conditions
positive
Social circumstances
negative
at maintenance rate 2-3 ml/kg/hr or less according to clinical 5 ml/kg/hr for 2-4 hr, then to 2-3 ml/kg/hr for 2-4 hr and then reduced
Patients with stable Hct can be sent home response further depending on haemodynamic status
Reassess clinical status and repeat Hct o I . V. fluids can be maintained for up to 24 - 48 hours
o I f Hct remains the same or rises only If patient still unstable:
minimally -> continue with 2-3 ml/kg/hr o C h eck Hct after first bolus
for another 2-4 hours o If Hct increases/ still high (>50%), repeat a second bolus of crystalloid
o I f worsening of vital signs and rapidly solution at 10-20 ml/kg/hr for 1 hr.
rising Hct -> increase rate to 5-10 o I f improvement after second bolus, reduce rate to 7-10 ml/kg/hr for 1-2 hr,
ml/kg/hr for 1-2 hours continue to reduce as above.
Reassess clinical status, repeat Hct and o If Hct decreases, this indicates bleeding and need to cross-match and
review fluid infusion rates accordingly transfuse blood as soon as possible
o R e duce intravenous fluids gradually Treatment of hypotensive shock
when the rate of plasma leakage o I nitiate I.V. fluid resuscitation with crystalloid or colloid solution at 20
decreases towards the end of the critical ml/kg as a bolus for 15 min
phase. If patient im p r o v e s
This is indicated by: o G i v e a crystalloid / colloid solution of 10 ml/kg/hr for 1 hr, then reduce
o A d e quate urine output and/or fluid intake gradually as above
o H ct decreases below the baseline value in If patient still unstable
a stable patient o R e view the Hct taken before the first bolus
Monitoring Monitoring Monitoring o If Hct was low (<40% in children and adult females, < 45% in adult males)
o D a i l y review for disease progression: o T emperature pattern o V ital signs and peripheral perfusion (1-4 this indicates bleeding, the need to crossmatch and transfuse (see above)
! D e creasing WBC o V olume of fluid intake and hourly until patient is out of critical o If HCT was high compared to the baseline value, change to I.V. colloids at
! D e f ervescence losses phase 10-20 ml/kg as a second bolus over to 1 hour; reassess after second bolus
! W a r ning signs (until out of critical o U rine output – volume and o U rine output (4-6 hourly) o If improving reduce the rate to 7-10 ml/kgt/hr for 1-2 hours, then back to
period) frequency o H ct (before and after fluid replacement, I.V. crystalloids and reduce rates as above
o A d vice for immediate return to o W a r ning signs then 6-12 hourly) o I f condition still unstable, repeat Hct after second bolus
hospital if development of any warning o H ct, white blood cell and o B l o o d glucose o If Hct decreases, this indicates bleeding, see above
signs platelet counts o O t h er organ functions (renal profile, liver o If Hct increases/ remains high (> 50%), continue colloid infusion at 10-20
o W r i t t e n advice of management (e.g. profile, coagulation profile, as indicated) ml/kg as athird bolus over 1 hr, then reduce to 7-10 ml/kg /hr for 1-2 hours,
home care card for dengue) then change back to crystalloid solution and reduce rate as above
Treatment of haemorrhagic complications:
o G i v e 5-10 ml/kg of fresh packed red cells or 10-20 ml/kg fresh whole blood
Discharge criteria: o No fever for 48 hours o I n c r e asing trend of platelet count o S t able haematocrit without intravenous fluids
-> all of the following criteria must be o Improvement in clinical o No respiratory distress
present picture
24
2.2 Treatment according to Groups A–C (1)
2.2.1 Group A
These are patients who may be sent home (see the home-care card for dengue in Textbox
G).
These patients are able to tolerate adequate volumes of oral fluids, pass urine at least once
every six hours and do not have any of the warning signs (particularly when fever
subsides).
The key to the success of ambulatory (outpatient) management is to give clear, definitive
advice on the care that the patient needs to receive at home: i.e. bed rest and frequent oral
fluids. Patients with ≥ 3 days of illness should be reviewed daily for disease progression
(indicated by decreasing white blood cell and platelet counts and increasing haematocrit,
defervescence and warning signs) until they are out of the critical period. Those with stable
haematocrit can be sent home but should be advised to return to the nearest hospital
immediately if they develop any of the warning signs and to adhere to the following action
plan:
• Adequate oral fluid intake may reduce the number of hospitalizations (3). Encourage
oral intake to replace fluid loss from fever and vomiting. Small amounts of oral fluids
should be given frequently for those with nausea and anorexia. The choice of fluids
should be based on the local culture: coconut water in some countries, in others rice
water or barley water. Oral rehydration solution or soup and fruit juices may be given
to prevent electrolyte imbalance. Commercial carbonated drinks that exceed the
isotonic level (5% sugar) should be avoided. They may exacerbate hyperglycaemia
related to physiological stress from dengue and diabetes mellitus. Sufficient oral fluid
intake should result in a urinary frequency of at least 4 to 6 times per day. A record of
oral fluid and urine output could be maintained and reviewed daily in the ambulatory
setting.
• Give paracetamol for high fever if the patient is uncomfortable. The recommended
dose is 10 mg/kg/dose, not more than 3−4 times in 24 hours in children and not more
than 3 g/day in adults). Sponge with tepid water if the patient still has a high fever. Do
not give acetylsalicylic acid (aspirin), ibuprofen or other non-steroidal anti-
inflammatory agents (NSAIDs) or intramuscular injections, as these aggravate
gastritis or bleeding.
• Instruct caregivers that the patient should be brought to hospital immediately if any of
the following occur: no clinical improvement, deterioration around the time of
defervescence, severe abdominal pain, persistent vomiting, cold and clammy
extremities, lethargy or irritability/restlessness, bleeding (e.g. black stools or coffee-
ground vomiting), shortness of breath, not passing urine for more than 4−6 hours.
Admission during the febrile period should be reserved for those who are unable to
manage adequate oral hydration at home, infants, and those with co-existing conditions.
Refer to Textbox C and Section 1.1.2 for a description of the warning signs.
25
Ambulatory patients should be monitored daily for temperature pattern, volume of fluid
intake and losses, urine output (volume and frequency), warning signs, signs of plasma
leakage and bleeding and complete blood counts.
2.2.2 Group B
These are patients who should be admitted for in-hospital management for close
observation as they approach the critical phase. These include patients with warning signs,
those with co-existing conditions that may make dengue or its management more
complicated (such as pregnancy, infancy, old age, obesity, diabetes mellitus, hypertension,
heart failure, renal failure, chronic haemolytic diseases such as sickle-cell disease and
autoimmune diseases), and those with certain social circumstances (such as living alone,
or living far from a health facility without reliable means of transport). Rapid fluid
replacement in patients with warning signs is the key to prevent progression to the shock
state.
If the patient has dengue with warning signs or signs of dehydration, judicious volume
replacement by intravenous fluid therapy from this early stage may modify the course and
the severity of disease. The action plan should be as follows and applies to infants, children
and adults:
• Obtain a reference haematocrit before intravenous fluid therapy begins. Give only
isotonic solutions such as 0.9% saline, Ringer's lactate or Hartmann's solution. Start
with 5−7 ml/kg/hour for 1−2 hours, then reduce to 3−5 ml/kg/hour for 2−4 hours, and
then reduce to 2−3 ml/kg/hour or less according to the clinical response (see
Textboxes H, J and K).
• Reassess the clinical status and repeat the haematocrit. If the haematocrit remains
the same or rises only minimally, continue at the same rate (2−3 ml/kg/hour) for
another 2−4 hours. If the vital signs are worsening and the haematocrit is rising
rapidly, increase the rate to 5−10 ml/kg/hour for 1−2 hours. Reassess the clinical
status, repeat the haematocrit and review fluid infusion rates accordingly.
• Give the minimum intravenous fluid volume required to maintain good perfusion and
an urine output of about 0.5 ml/kg/hour. Intravenous fluids are usually needed for only
24−48 hours. Reduce intravenous fluids gradually when the rate of plasma leakage
decreases towards the end of the critical phase. This is indicated by urine output
and/or oral fluid intake improving, or the haematocrit decreasing below the baseline
value in a stable patient.
• Patients with warning signs should be monitored by health-care providers until the
period of risk is over. A detailed fluid balance should be maintained. Parameters that
should be monitored include vital signs and peripheral perfusion (1−4 hourly until the
patient is out of the critical phase), urine output (4−6 hourly), haematocrit (before and
after fluid replacement, then 6−12 hourly), blood glucose and other organ functions
(such as renal profile, liver profile, coagulation profile, as indicated).
If the patient has dengue with co-existing conditions but without warning signs, the action
plan should be as follows:
• Encourage oral fluids. If not tolerated, start intravenous fluid therapy of 0.9% saline or
Ringer’s lactate with or without glucose at the appropriate maintenance rate (Textbox
H). Use the ideal body weight for calculation of fluid infusion for obese and
overweight patients (Textboxes J and K). Patients may be able to take oral fluids after
26
a few hours of intravenous fluid therapy. Thus, it is necessary to revise the fluid
infusion frequently. Give the minimum volume required to maintain good perfusion
and urine output. Intravenous fluids are usually needed only for 24−48 hours.
Patients should be monitored by health-care providers for temperature pattern, volume of
fluid intake and losses, urine output (volume and frequency), warning signs, haematocrit,
white blood cell and platelet counts (Textbox K). Depending on the clinical picture and the
facilities of the hospital or health centre, other laboratory tests (such as liver and renal
functions tests) can also be carried out.
2.2.3 Group C
These are patients with severe dengue who require emergency treatment and urgent
referral because they are in the critical phase of the disease and have:
• severe plasma leakage leading to dengue shock and/or fluid accumulation with
respiratory distress;
• severe haemorrhages;
• severe organ impairment (hepatic damage, renal impairment, cardiomyopathy,
encephalopathy or encephalitis).
All patients with severe dengue should be admitted to a hospital with access to blood
transfusion facilities. Judicious intravenous fluid resuscitation is the essential and usually
sole intervention required. The crystalloid solution should be isotonic and the volume just
sufficient to maintain an effective circulation during the period of plasma leakage. Plasma
losses should be replaced immediately and rapidly with isotonic crystalloid solution: in the
case of hypotensive shock, colloid solution is preferred (Textbox L). If possible, obtain
haematocrit levels before and after fluid resuscitation.
Continue replacement of further plasma losses to maintain effective circulation for 24−48
hours. For overweight or obese patients, the ideal body weight should be used for
calculating fluid infusion rates (see Textboxes J and K). All shock patients should have their
blood group taken and a cross-match carried out. Blood transfusion should be given only in
cases with established severe bleeding, or suspected severe bleeding in combination with
otherwise unexplained hypotension.
Fluid resuscitation must be clearly separated from simple fluid administration. This is a
strategy in which larger volumes of fluids (e.g. 10−20 ml/kg boluses) are administered for a
limited period of time under close supervision, to evaluate the patient’s response and to
avoid the development of pulmonary oedema. These fluids should not contain glucose. The
degree of intravascular volume deficit in dengue shock varies. Input is typically much
greater than output, and the input/output ratio is of no help in judging fluid resuscitation
needs during this period.
The goals of fluid resuscitation include:
• improving central and peripheral circulation – i.e. decreasing tachycardia, improving
BP and pulse volume, warm and pink extremities, a capillary refill time < 2 seconds;
• improving end-organ perfusion – i.e. achieving a stable conscious level (more alert or
less restless), and urine output ≥ 0.5 ml/kg/hour or decreasing metabolic acidosis.
27
2.2.3.1 Treatment of shock (4–8)
The action plan for treating patients with compensated shock is as follows (see algorithms
in Figures 5 and 6):
• Obtain a reference haematocrit before starting intravenous fluid therapy. Start
intravenous fluid resuscitation with isotonic crystalloid solutions at 5−10 ml/kg/hour
over one hour in adults and 10−20 ml/kg/hour over one hour in infants and children.
Then reassess the patient’s condition (vital signs, capillary refill time, haematocrit,
urine output).
• If the adult patient’s condition improves, intravenous fluids should be gradually
reduced to 5−7 ml/kg/hour for 1−2 hours; then 3−5 ml/kg/hour for 2−4 hours and
finally 2−3 ml/kg/hour which can be maintained up to 24−48 hours. Consider reducing
intravenous fluid earlier if oral fluid intake improves. The total duration of intravenous
fluid therapy should not exceed 48 hours.
• If the condition in the infant or child improves, intravenous fluids should be reduced
to 10 ml/kg/hour for 1−2 hours; then to 7 ml/kg/hour for 2 hours; 5 ml/kg/hour for 4
hours and then to 3 ml/kg/hour, which can be maintained for up to 24−48 hours.
Consider reducing intravenous fluid earlier if oral fluid intake improves. The total
duration of intravenous fluid therapy should not exceed 48 hours.
• (See Textboxes H and J for a more appropriate estimate of the normal maintenance
requirement based on ideal body weight).
• If vital signs are still unstable (i.e. shock persists), check the haematocrit after the
first bolus.
o In adults:
If the haematocrit increases or is still high (e.g. haematocrit > 50%), repeat a
second bolus of crystalloid/colloid solution at 10−20 ml/kg/hour for one hour.
After this second bolus, if there is improvement continue with crystalloid solution
and reduce the rate to 7−10 ml/kg/hour for 1−2 hours, then continue to reduce
as above.
If haematocrit decreases compared to the initial reference haematocrit
(especially if the repeat haematocrit is below the baseline, for example
< 35−40% in adult females, < 40−45% in adult males), and the patient still has
unstable vital signs, this may indicate bleeding. Look for severe bleeding.
Cross-match fresh whole blood or fresh packed red cells and transfuse if there
is severe overt bleeding, If there is no bleeding, give a bolus of 10−20 ml of
colloid, repeat clinical assessment and determine the haematocrit level. A senior
staff member should carry out a review to consider blood transfusion (see
Section 2.2.3.3, Treatment for haemorrhagic complications).
Compensated shock
(Systolic pressure maintained + signs of reduced perfusion )
Yes No
IMPROVEMENT*
HCT↑
IV crystalloid, reduce or High HCT↓
gradually Check
5−7 ml/kg/hr for 1−2 hours HCT
3−5 ml/kg/hr for 2−4 hours
2−3 ml/kg/hr for 2−4 hours
Crystalloid (2nd bolus) or
colloid** Yes
10-20 ml/kg/hr for 1 hour No
As clinical improvement Severe overt
is noted, reduced fluids bleed
accordingly
IMPROVEMENT*
No Urgent Colloid 10-20
Further boluses may be blood ml/kg/hr
needed for the next Yes transfusion Evaluate to
24−48 hours consider
blood
transfusion if
Reduce IV crystalloids no clinical
Stop IV fluids at 48 7-10 ml/kg/hr for 1−2 improvement
hours hours
*Reassess the patient’s clinical condition, vital signs, pulse volume, capillary refill time and temperature of extremities.
**Colloid is preferable if the patient has already received previous boluses of crystalloid
-IV: intravenous, HCT: haematocrit, ↑: increased, ↓: decreased.
29
Fig. 6. Algorithm for fluid management of compensated shock: in infants and
children
Compensated shock
(Systolic pressure maintained + signs of reduced perfusion)
Yes No
IMPROVEMENT*
HCT↑
IV crystalloid, reduce gradually or high HCT↓
10 ml/kg/hr for 1−2 hours Check
7 ml/kg/hr for 2 hours HCT
5 ml/kg/hr for 4 hours
3 ml/kg/hr
Crystalloid (2nd bolus) or
colloid
10-20 ml/kg/hr for 1 hour
As clinical improvement is
noted, reduced fluids Severe Overt
accordingly Bleed
Yes No
IMPROVEMENT
* Urgent Colloid 10-20
Further boluses may be No
needed for the next 24−48 blood ml/kg/hr
hours transfusion Evaluate to
Yes
consider blood
transfusion if no
Reduce IV crystalloids clinical
Stop IV fluids at 48 hours 7-10 ml/kg/hr for 1−2 hours improvement
^Colloid is preferable if the patient has already received previous boluses of crystalloid
*Reassess the patient’s clinical condition, vital signs, pulse volume, capillary refill time and temperature of
extremities.IV = intravenous; HCT = haematocrit; ↑ = increased; ↓ = decreased
30
Treatment of profound shock (hypotensive; undetectable pulse and BP)
All patients (infants, children and adults) with hypotensive shock should be managed
more vigorously. The action plan for treating patients with hypotensive shock is outlined
below (also see Textbox D and Figure 7).
For all patients (infants, children and adults), initiate intravenous fluid resuscitation with
crystalloid or colloid solution at 20 ml/kg as a bolus given over 15−30 minutes to bring the
patient out of shock as quickly as possible. Colloids may be the preferred choice if the BP
has to be restored urgently, i.e. in those with pulse pressure less than 10 mmHg. Colloids
have been shown to restore the cardiac index and reduce the level of haematocrit faster
than crystalloids in patients with intractable shock (4–6). The intra-osseous route should be
attempted if peripheral venous access cannot be obtained.
• If the patient’s condition improves:
o In adults, give a crystalloid/colloid infusion of 10 ml/kg/hour for 1 hour. Then
continue with crystalloid infusion and gradually reduce to 5−7 ml/kg/hour for 1−2
hours, then to 3−5 ml/kg/hour for 2−4 hours, and finally to 2−3 ml/kg/hour (or less),
which can be maintained for up to 24−48 hours (Textbox H). Consider reducing
intravenous fluid earlier if oral fluid intake and urine output improve. The total
duration of intravenous fluid therapy should not exceed 48 hours.
o In infants and children, give colloid infusion of 10 ml/kg/hour for 1 hour. Then
continue with crystalloid 10 ml/kg/hour for 1 hour, then to 7.5 ml/kg/hour for 2 hours,
to 5 ml/kg/hour for 4 hours and to 3 ml/kg/hour, which can be maintained for up to
24–48 hours. Consider reducing intravenous fluid earlier if oral fluid intake and urine
output improve. The total duration of intravenous fluid therapy should not exceed 48
hours.
• For all patients, if vital signs are still unstable (i.e. shock persists), review the
haematocrit obtained before the first bolus.
o If the haematocrit was normal or low (< 30–35% in infants, < 35–40% in children
and adult females, < 40–45% in adult males), this may indicate bleeding. Look for
severe bleeding. Cross-match fresh whole blood or fresh packed red cells and
transfuse if there is severe overt bleeding. If there is no bleeding, give a second
bolus of 10–20 ml/kg of colloid over 30 minutes to 1 hour, repeat clinical
assessment and haematocrit level plus a review by senior staff to consider blood
transfusion (see treatment for haemorrhagic complications).
o If the haematocrit was high compared to the baseline value (if not available, use
population baseline), change intravenous fluids to colloid solutions at 10−20 ml/kg
as a second bolus over 30 minutes to 1 hour. After the second bolus, reassess the
patient. If the condition improves, reduce the rate to 7−10 ml/kg/hour for 1−2 hours,
then change back to crystalloid solution and reduce the rate of infusion as
mentioned above.
• If the condition is still unstable, repeat the haematocrit after the second bolus.
o If the haematocrit decreases compared to the previous value (< 35% in infants,
< 40% in children and adult females, < 45% in adult males), this indicates bleeding
and the need to cross-match and transfuse blood as soon as possible (see 2.2.3.3
for haemorrhagic complications).
o If the haematocrit increases compared to the previous value or remains very high
(> 50%), continue colloid solutions at 10−20 ml/kg as a third bolus over 1 hour. After
this dose, reduce the rate to 7−10 ml/kg/hour for 1−2 hours, then change back to
31
crystalloid solution and reduce the rate of infusion as mentioned above when the
patient’s condition improves. If the condition is still unstable, repeat the haematocrit
after the third bolus.
• Further boluses of fluids may need to be given during the next 24 hours. The rate and
volume of each bolus infusion should be titrated to the clinical response. Patients with
severe dengue should be admitted to the high-dependency or intensive care area
and be managed by senior staff.
o Clinicians who take care of dengue shock infants should remember that an infant
with a low baseline haematocrit of 30%, presenting with dengue shock and a
haematocrit of 40%, is relatively more haemoconcentrated than another child with a
baseline value of 42% and haematocrit of 50% at the time of shock.
Fig. 7. Algorithm for fluid management in hypotensive shock – infants, children and
adults
Hypotensive shock
Try to obtain an HCT level before fluid resuscitation
Yes No
IMPROVEMENT*
^Colloid is preferable if the patient has already received previous boluses of crystalloid
*Reassess the patient’s clinical condition, vital signs, pulse volume, capillary refill time and temperature of extremities.
IV = intravenous; HCT = haematocrit; ↑ = increased; ↓ = decreased
Patients with dengue shock should be monitored frequently until the danger period is over.
A detailed fluid balance of all inputs and outputs should be maintained.
Parameters to be monitored include: alertness and comfort levels, vital signs and peripheral
perfusion (every 15−30 minutes until the patient is out of shock then 1−2 hourly). In
general, the higher the fluid infusion rate, the more frequently the patient should be
32
monitored and reviewed in order to avoid fluid overload while ensuring adequate volume
replacement. If previously not detectable, pleural effusion and ascites should be detectable
after fluid boluses. Monitor their effects on breathing.
If resources are available for blood gas and/or lactate analysis, capillary or venous blood
should be sampled for repeated analysis to monitor changes in the circulation during fluid
replacement. An arterial line has certain advantages but its placement is hazardous
because of the attendant bleeding from failed attempts. The advantage of an arterial line is
that in shock states, estimation of BP using a cuff is commonly inaccurate. The use of an
indwelling arterial catheter allows for continuous and reproducible BP measurements and
frequent blood sampling to base decisions regarding therapy. Monitoring of ECG and pulse
oximetry should be available in the intensive care unit.
Urine output should be checked regularly (each hour until the patient is out of shock, then
every 1−2 hours). A continuous bladder catheter enables close monitoring of urine output.
The first urine volume after bladder catheterization should be discarded because the
duration in the bladder is unknown. Thereafter, an acceptable urine output would be about
0.5 ml/kg/hour. Haematocrit should be monitored (before and after fluid boluses until
stable, then 4−6 hourly). In addition, there should be monitoring of: blood glucose (before
fluid resuscitation and repeat as indicated); arterial or venous or capillary blood gases;
lactate; total carbon dioxide/bicarbonate (every 30 minutes to 1 hour until stable, then as
indicated); and other organ functions (such as renal profile, liver profile, coagulation profile)
before resuscitation and as indicated.
Interpretation of haematocrit
The patient’s baseline haematocrit on the first three days of illness is a useful reference
point. It is important to note that during fluid therapy, blood samples for haematocrit should
be timed such that they are taken before or after the infusion of a known volume of
intravenous fluid. The interpretation will be most meaningful if the corresponding
haemodynamic state, or response to fluid therapy and the acid-base balance, are known at
the time of blood sampling. Used in this “3-dimensional” way, changes in the haematocrit
are the most useful guide to decision-making about fluid therapy. Random haematocrit
levels, such as random blood glucose in diabetes mellitus, may not be meaningful for
interpretation of the real-time clinical situation.
A rising or persistently high haematocrit together with unstable vital signs (such as
narrowed pulse pressure) indicates active plasma leakage and the need for a further bolus
of fluid replacement. However, a rising or persistently high haematocrit together with stable
haemodynamic status and adequate urine output does not require extra intravenous fluid.
In the latter case, continue to monitor closely and it is likely that the haematocrit will start to
fall within the next 24 hours as plasma leakage stops.
A decrease in haematocrit (for example from 50% to 40% or below the patient’s known
baseline) together with unstable vital signs (narrowed pulse pressure, tachycardia,
metabolic acidosis and poor urine output), may indicate major haemorrhage. If there is
severe haemorrhage, urgent blood transfusion should be given. If there is no clinical sign of
bleeding, then a further bolus of 10–20 ml/kg of colloid should be given, followed by repeat
clinical assessment and haematocrit level determination plus a review by senior staff to
consider blood transfusion. Concealed bleeding may take several hours to become
33
apparent and the patient’s haematocrit will continue to decrease without achieving
haemodynamic stability.
On the other hand, a decrease in haematocrit together with stable haemodynamic
status and adequate urine output, indicates haemodilution and/or reabsorption of
extravasated fluids In this case intravenous fluids must be discontinued immediately to
avoid pulmonary oedema.
34
• have any form of trauma, including intramuscular injection.
Patients with haemolytic conditions are at risk of acute haemolysis with haemoglobinuria
and may require blood transfusion (see Section 2.4.3).
Severe and occult bleeding is the most common cause of profound/refractory/prolonged
shock, but can be difficult to recognize. This is because bleeding usually occurs after a
period of prolonged shock in dengue (9). The preceding plasma leakage causes the
haematocrit to rise to very high levels. When bleeding occurs the haematocrit will then drop
from this high level and as a result haematocrit levels may not be as low as in the absence
of plasma leakage. Even in severe bleeding, the haematocrit remains above the baseline
and only drops to normal or low levels after several fluid boluses.
Severe bleeding should be recognized in the following situations:
• persistent and/or severe overt bleeding in the presence of unstable haemodynamic
status, regardless of the haematocrit level;
• a decrease in haematocrit after boluses of fluid resuscitation together with unstable
haemodynamic status;
• refractory shock that fails to respond to consecutive fluid resuscitation of 40–
60 ml/kg;
• hypotensive shock with inappropriately low/normal haematocrit;
• persistent or worsening metabolic acidosis in patients with a well-maintained systolic
BP, especially in those with severe abdominal tenderness and distension.
Blood transfusion is life-saving and should be given as soon as severe bleeding is
suspected or recognized. However, blood transfusion must be given with care because of
the risk of fluid overload. Do not wait for the haematocrit to drop too low before deciding on
blood transfusion. Note that for the reasons stated above a haematocrit of < 30% as a
trigger for blood transfusion, as recommended in the Surviving Sepsis Campaign Guideline
(10), is not applicable to dengue.
The action plan for the treatment of haemorrhagic complications is as follows:
• If possible, attempts should be made to stop bleeding if the source of bleeding is
identified e.g. severe epistaxis may be controlled by nasal adrenaline packing.
• If blood loss can be quantified, this should be replaced. If not, give aliquots of
5−10 ml/kg of fresh -packed red cells or 10−20 ml/kg of fresh or fairly fresh whole
blood (FWB) at an appropriate rate and observe the clinical response. It is important
that fresh whole blood or fresh red cells are given. Oxygen delivery at tissue level is
optimal with high levels of 2,3 diphosphoglycerate (2,3 DPG). Stored erythrocytes
lose 2,3 DPG, low levels of which impede the oxygen-releasing capacity of
haemoglobin, resulting in functional tissue hypoxia. A good clinical response includes
improving haemodynamic status and acid-base balance.
• Consider repeating the blood transfusion if there is further overt blood loss or no
appropriate rise in haematocrit after blood transfusion in an unstable patient.
• There is no evidence that supports the practice of transfusing platelet concentrates
and/or fresh-frozen plasma for severe bleeding in dengue (11). Observational studies
35
show that transfusions of platelet concentrates and fresh frozen plasma in dengue
were not able to sustain the platelet counts and coagulation profile. However, in the
case of massive bleeding, they often exacerbate the fluid overload.
• Nevertheless, in certain situations such as obstetrical deliveries or other surgeries,
transfusions of platelet concentrates with or without fresh frozen plasma should be
considered in anticipation of severe bleeding.
• In gastrointestinal bleeding, H-2 antagonist and proton pump inhibitors have been
used, but their efficacy have not been studied.
• Great care should be taken when inserting a nasogastric tube or bladder catheters
which may cause severe haemorrhage. A lubricated orogastric tube may minimize
the trauma during insertion. Insertion of central venous catheters should be done with
ultra-sound guidance or by an experienced person.
• It is essential to remember that blood transfusion is only indicated in dengue patients
with severe bleeding. Unnecessary blood transfusions cause the haematocrit to rise
sharply, thus giving a false impression of haemoconcentration and severe plasma
leakage leading to unwarranted fluid therapy (see Chapter 3, pitfall 30).
36
2.2.3.5 Electrolyte and acid-base imbalances
Hyponatraemia is a common observation in severe dengue; the underlying mechanism is
not fully understood. It could be related to gastrointestinal losses through vomiting and
diarrhoea or the use of hypotonic solutions for resuscitation and correction of dehydration.
The use of isotonic solutions for resuscitation will prevent and correct this condition.
Hyperkalaemia is observed in association with severe metabolic acidosis or acute renal
injury. Appropriate volume resuscitation will reverse the metabolic acidosis and the
associated hyperkalaemia. Life-threatening hyperkalaemia, in the setting of acute renal
failure should be managed with Resonium A and infusions of calcium gluconate and/or
insulin-dextrose. Renal support therapy may have to be considered.
Hypokalaemia is often associated with gastrointestinal fluid losses and the stress-induced
hypercortisol state; it is usually encountered towards the later part of the critical phase. It
should be corrected with potassium supplements in the parenteral fluids.
Serum calcium levels should be monitored and corrected when large quantities of blood
have been transfused or if sodium bicarbonate has been used.
37
2.2.4 References
1. Martinez E. Preventing deaths from dengue: a space and challenge for primary health care. Pan American
Journal of Public Health, 2006, 20:60−74.
2. Dengue. Guidelines for diagnosis, treatment prevention and control, Geneva, World Health Organization,
2009, WHO/HTM/NTD/DEN/2009.
3. Harris E et al., Fluid intake and decreased risk for hospitalization for dengue fever, Nicaragua. Emerging
Infectious Diseases, 2003, 9:1003−1006.
4. Dung NM, Day NP, Tam DT. Fluid replacement in dengue shock syndrome: a randomized, double-blind
comparison of four intravenous fluid regimens. Clinical Infectious Diseases, 1999, 29:787−794.
5. Ngo NT, Cao XT, Kneen R. Acute management of dengue shock syndrome: a randomized double-blind
comparison of 4 intravenous fluid regimens in the first hour. Clinical Infectious Diseases, 2001,
32:204−213.
6. Wills BA et al., Comparison of three fluid solutions for resuscitation in dengue shock syndrome. New
England Journal of Medicine, 2005, 353:877−889.
7. Hung NT et al., Volume replacement in infants with dengue hemorrhagic fever/dengue shock syndrome.
American Journal of Tropical Medicine and Hygiene, 2006, 74:684-691.
8. Wills BA. Management of dengue. In: Halstead SB, ed. Dengue. London, Imperial College Press, 2008: 193
−217.
9. LCS Lum et al., Risk factors for hemorrhage in severe dengue infection. Journal of Pediatrics, 2002,
140:629−631.
10. Dellinger RP et al., Surviving sepsis campaign: International guidelines for management of severe sepsis
and septic shock. Critical Care Medicine, 2008, 36:296−327. [Published correction appears in Critical
Care Medicine, 2008, 36:1394−1396.]
11. Lye DC et al., Lack of efficacy of prophylactic platelet transfusion for severe thrombocytopaenia in adults
with acute uncomplicated dengue infection. Clinical Infectious Diseases, 2009, 48:1262−1265.
12. Burdett E, Roche AM, Mythen MG. Hyperchloremic acidosis: Pathophysiology and clinical impact.
Transfusion Alternatives in Transfusion Medicine, 2003, 5(4):424−430.
38
2.3 Complications and intensive care management
Many of the complications seen in dengue are preventable if clinical team members are
alert to the physiological problems of the three different phases. When hypovolaemic shock
is adequately managed, patients appear to “sail out” of the critical phase with mere
parenteral fluids. But that belies the effort that has been invested in the monitoring and
careful titration of intravenous fluid therapy, guided by frequent clinical and haematocrit
evaluation.
Causes of complications in dengue include:
• missed diagnosis at the frontline;
• inadequate monitoring and misinterpretation of vital signs;
• inadequate monitoring of fluid intake and urine output;
• late recognition of shock leading to profound and/or prolonged shock;
• late recognition of severe bleeding;
• too much or too little intravenous fluids i.e. not following/understanding the treatment
guidelines;
• careless attitude towards aseptic techniques.
Outcome: These lead to a life-threatening situation characterized by one or a combination
of the following:
• prolonged and/or profound shock;
• severe bleeding with severe disseminated intravascular coagulopathy;
• fluid overload;
• respiratory distress and failure;
• multi-organ dysfunction of liver, kidneys and neurological system;
• irreversible shock and death.
This group of patients should be referred to a hospital with intensive care facilities wherever
possible.
The clinical evaluation of these patients should include:
A general inspection of the mental state: remember that shock patients remain quiet but
alert until hypotension sets in. An irritable combative dengue patient indicates severe shock
with declining cortical perfusion. Confusion, lethargy, seizures and coma will set in very
quickly. Other possible causes of change in mental state are fulminant hepatic failure (1, 2),
hypoglycaemia, electrolyte abnormalities and rarely, intracranial bleeding and encephalitis
(3−5).
Airway and breathing: There is usually no problem with the airway. Breathing should be
assessed carefully. The patient in early compensated shock has quiet tachypnoea (the
lungs are clear). As shock progresses, metabolic acidosis sets in; the breathing then
becomes deeper and faster. The lungs remain clear with good air entry. There might be
clinically detectable pleural effusion but the question should be: Is this enough to cause the
tachypnoea? If no, then this could be Kussmaul’s breathing. Patients with Kussmaul’s
39
breathing will prefer to lie horizontal because of hypovolaemia, whereas patients with true
respiratory distress from pulmonary causes will most probably be sitting up for optimal
chest wall mechanics.
For pleural effusion to be the cause of the respiratory distress, it should be substantial
and easily detectable. The diaphragm may be splinted by tense ascites. This scenario
occurs after unrestrained intravenous fluid therapy, not before. Wheezing and rhonchi
indicate pulmonary oedema and hypervolaemia, not the presence of asthma. Fine
crepitations of pulmonary oedema may not be audible if the breathing effort is poor. Again,
this occurs after aggressive or unguided intravenous fluid therapy.
Occasionally a combination of metabolic acidosis and fluid overload may be seen in
patients who have received inappropriate fluid therapy. Thus they remain in severe shock
and have fluid overload at the same time. These patients cannot get comfortable lying
down or sitting up and are very restless. The clinical evaluation should be corroborated with
radiological findings and blood gas analysis. Diminished air entry in a drowsy patient
indicates impending respiratory failure. Cyanosis and gasping respiration is seen in an
imminent respiratory arrest.
Circulation: The peripheral perfusion should be assessed by examination and palpation of
the extremities and the radial pulse – pulse volume, capillary refill time, temperature, colour
of extremities and pulse pressure (see Section 1.1.4).
Examine for bleeding (orogastric tube insertion, if indicated), jaundice, evidence of plasma
leakage and abdominal tenderness and distension.
Monitoring: Continuous heart rate (ECG), respiration and pulse oximetry should be
monitored. With the latter it may be difficult to get a signal in severe shock. Intermittent BP
(measured every 5−30 minutes), peripheral perfusion and four-hourly temperature checks
should be maintained for these critically ill patients. Finger-on-the-pulse may be necessary
in patients with feeble pulses.
Insert a bladder catheter for continuous bladder drainage and hourly urine output
measurements.
An arterial line may be inserted (if available) when the radial pulse is stronger, but in the
meantime use venous or capillary samples for intermittent blood gas and lactate analysis,
and frequent haematocrit timed to before or/and after fluid boluses. Lactate levels are
higher in venous and capillary samples rather than arterial ones. A micro-haematocrit
centrifuge in the ward facilitates the management of severe dengue.
Evaluate the history and prior treatment:
• What was the date of onset of fever?
• Which phase of dengue illness is the patient in? Is the patient in the critical phase?
The countdown may begin from the time of defervescence (temperature below 38°C),
or appearance of warning signs, or increase in haematocrit, or sudden decrease in
platelet count. For those patients who present with shock, the beginning of the critical
period could be a few hours before presentation to the hospital. The period of
intravenous therapy should therefore be less than 48 hours in those who present to
the hospital with shock.
40
• Is there evidence of plasma leakage? How many more hours of plasma leakage do
you envisage? These timelines may be approximations but will help guide your
decision about stopping intravenous fluid therapy.
• Review the fluid intake, output and balance since the start of parenteral fluid therapy.
What types of fluids have been administered – hypotonic, glucose solutions, isotonic
crystalloids (0.9% sodium chloride solution or Ringer’s lactate or Hartman’s solution),
colloids, blood, and blood products? When was the last urine output? What is the
amount of urine? Is the urine volume appropriate to the haemodynamic state? How
much is the positive fluid balance? A large positive balance of fluid is reflected in
large pleural effusions, ascites, respiratory distress and generalized oedema.
• Review the haemodynamic response to the volume and type of intravenous fluid
therapy, and the corresponding trends of haematocrit. You will obtain a better
understanding of the clinical situation and the dynamics if you view the process in a
“3-dimensional” way.
Additional investigations:
• Full blood count, haematocrit, blood glucose, blood gas analysis and lactate should
be done for patients with tachypnoea and shock. Serum electrolytes and calcium,
liver function tests, blood urea and creatinine is indicated in all cases of severe
dengue shock. Coagulation profile – prothrombin time (PT), partial thromboplastin
time (PTT), international normalized ratio (INR), fibrinogen degradation products and
fibrinogen level is indicated in severe bleeding.
• Chest radiography – a small heart shadow indicates intravascular depletion. A large
heart (cardiothoracic ratio > 0.55) together with prominent pulmonary vascular
markings, “bat’s wings” appearance ± Kerley B lines are suggestive of intravascular
hypervolaemia and pulmonary oedema. The lung volume may be reduced by the
pleural effusion and upward displacement of the diaphragm by gross ascites.
• Determine blood group and match for fresh whole blood or fresh packed cells, at
least 10−20 ml/kg aliquot.
• Blood culture and sensitivity (if indicated)
• Dengue diagnostic tests (see Section 1.4)
41
• Is the patient in the beginning or nearing the end of the critical phase or in the
recovery phase?
• Does the patient have intravascular and extravascular fluid overload?
• Is the shock prolonged? Has it been missed? You could track shock by the
tachypnoea, tachycardia and/or the period of anuria/oliguria and/or metabolic
acidosis.
• Is there multi-organ dysfunction? This is seen in prolonged shock.
• Is there hyperglycaemia? If there is, is it related to stress, inappropriate fluids or
undiagnosed or uncontrolled diabetes mellitus? Please refer to the information on
hyperglycaemia in Section 2.2.3.4.
• Is there electrolyte disturbances? Please refer to section on electrolyte disturbances
in Section 2.2.3.5.
• Does the patient have severe overt bleeding? An orogastric tube may reveal fresh
bleeding or coffee grounds.
• Does the patient have severe internal/occult bleeding? There are several ways to
answer this question. The most direct method: Is the patient’s baseline haematocrit
known i.e. the haematocrit on days 1−3 of fever? If yes, then it is valuable
information. If not, the population haematocrit may be used, although this will not be
as precise as the patient’s own baseline. Many conditions such as iron deficiency,
mild haemolytic disease, chronic smoking, and chronic obstructive sleep apnoea will
affect the patient’s baseline haematocrit.
o For other ways to answer this very important question, refer to Section 2.2.3.3 on
the recognition of severe bleeding.
o The earlier this critical condition is recognized, the earlier definitive treatment can
be given, saving the patient from receiving several ineffective bolus fluids, most of
which will be redistributed to the third spaces.
Prolonged/profound shock:
Prolonged/profound shock is characterized by severe metabolic acidosis ± multi-organ
failure.
Scenario 1
If the patient is quietly alert and cooperative, an urgent haematocrit will guide further fluid
therapy. If the analysis indicates severe bleeding and matched fresh whole blood (FWB) is
available, blood transfusion should be started as soon as possible. However, the following
applies if blood is not available:
• If the patient has received < 2 boluses of resuscitation fluid, a colloid solution of
10−20 ml/kg over 15−30 minutes should be used (refer to the algorithm in Figures 5–
7). If the patient has received more than 2 boluses of resuscitation fluid, fluids should
be switched to a colloid solution of 10−20 ml/kg over 30 minutes for hypotensive
shock, and over 1−2 hours for compensated shock.
If severe overt bleeding is apparent (haematemesis, malaena or hypermenorrhoea), the
colloid bolus should be followed urgently by transfusion of 10−20 ml/kg FWB, regardless of
42
the haematocrit level. After transfusion of FWB, some degree of haemodynamic stability is
usually achieved together with improvement of metabolic acidosis. Further colloid infusions
may be necessary if the haematocrit rises again. A repeat transfusion of FWB will be
required if bleeding continues. Bleeding will usually slow down towards the end of the
critical phase.
There is no evidence that transfusion of platelet concentrates or the disseminated
intravascular coagulopathy (DIVC) regime is effective. This practice will contribute to third
space losses and expose the patient to multiple blood donors. Prolonged stay in the
intensive care unit (ICU) is also expected.
If no overt bleeding is seen after the colloid bolus, a repeat clinical evaluation and
haematocrit level should be performed. A decrease in haematocrit together with clinical
improvement means there is restoration of circulatory volume with colloids. However, a
decrease in haematocrit, not accompanied by clinical improvement should prompt the
suspicion of severe internal/occult bleeding.
Scenario 2
The patient is restless and combative. This patient has prolonged severe shock and is at
high risk of dying soon. This scenario is often associated with severe bleeding. In addition
to the management as in scenario 1, one should make preparations for tracheal intubation
and mechanical ventilation. The patient should be given a high flow oxygen mask, or the
airway and breathing should be supported with mask ventilation. Fresh, safe, group “O”
blood should be considered if cross-matched blood will take > 1 hour to become available.
Refer to mechanical ventilation later in Chapter 2.
Scenario 3
If the patient does not respond to the above measures, a concomitant nosocomial infection
should be suspected and appropriate antibiotics started.
44
• the chest X-ray which shows cardiomegaly, pleural effusion, upward displacement of
the diaphragm by the ascites and varying degrees of “bat’s wings” appearance ±
Kerley B lines, suggestive of fluid overload and pulmonary oedema;
• ECG to exclude ischaemic changes and arrhythmia;
• echocardiogram for assessment of left ventricular function and left ventricular end-
diastolic diameters (LVEDD) and regional wall dyskinesia that may suggest
underlying ischaemic heart disease. The LVEDD is a reliable measure of the filling
volume of the left ventricle and is increased in hypervolaemia. Other indirect
measures of the intravascular compartment status are the sizes of internal jugular
vein and inferior vena cava;
• cardiac enzymes.
Action plan:
• Oxygen therapy should be given immediately.
• The further action plan for the treatment of fluid overload is dependent on the
patient’s haemodynamic stability, intravascular volume status and the timing of this
event with respect to the timeline of the critical phase.
Scenario 1
Strong pulses with warm extremities are positive indications to stop (if ≥ 48 hours of plasma
leakage) or reduce (if ≤ 48 hours of plasma leakage) intravenous fluids. The haematocrit is
typically low; however it may remain elevated in some patients. The acid-base balance and
lactate level are almost normal. If the patient has difficulty in breathing because of
excessive third space fluid accumulation, it is all the more imperative to stop fluid therapy.
Small doses of furosemide 0.1−0.5 mg/kg/dose twice or thrice daily or a continuous
infusion of furosemide 0.1 mg/kg/hour may be indicated for patients who are out of the
critical phase. Monitor serum potassium and correct the ensuing hypokalaemia. A high
creatinine level suggests that the kidneys may only respond to higher doses of furosemide.
Watch out for hypertension and treat during the recovery phase otherwise hypertensive
encephalopathy may occur. Respiratory support may be indicated depending on the
severity of respiratory distress. Recognizing when to decrease or stop intravenous fluids is
crucial to preventing fluid overload (see Section 2.2.3.2).
Scenario 2
The patient has stable haemodynamic status but is still within the critical phase, < 48 hours
of plasma leakage. The acid-base balance and lactate level are almost normal. Reduce the
intravenous fluid accordingly or change to a colloid solution at 1–2 ml/kg/hour and reduce
accordingly. Avoid diuretics during the plasma leakage phase because they may precipitate
intravascular volume depletion. Refer to scenario 1 when the patient is out of the critical
period.
Scenario 3
Patients who remain in shock with elevated haematocrit levels but show excessive fluid
accumulation are likely to have received rapid infusions of crystalloid or hypotonic solutions
or blood products. In these patients a carefully titrated bolus of colloid solution at
5−10 ml/kg/hour for 1–2 hours may be helpful; reduce accordingly. All fluids should be
45
stopped at or before the 48 hours of the critical period. Furosemide therapy may be
required as in scenario 1.
Scenario 4
Patients who remain in shock with low or normal haematocrit levels but have excessive
fluid accumulation, are most likely to have severe occult bleeding (see Section 2.2.3.3). If
the BP is low, a dopamine infusion should be started. Further infusion of large volumes of
intravenous fluids will lead to a poor outcome. Careful transfusion of FWB, at least
10 ml/kg, should be initiated as soon as possible at a rate titrated to a clinical response,
blood gases and lactate. Respiratory support may be required; refer to section on
mechanical ventilation.
Scenario 5
A combination of scenarios 3 and 4 may play out in some cases. Patients with both
intravascular and extravascular fluid overload remain in shock with metabolic acidosis
because the severe bleeding has been replaced with non-fresh whole blood. Refer to
Chapter 3, pitfall 28. A combination of FWB or fresh packed red cells transfusion and
careful colloid infusion may help the patient out of the plasma leakage phase with minimal
worsening of the respiratory condition. Dopamine should be commenced for hypotension
and the patient should be monitored carefully for respiratory failure.
Pulmonary oedema and acute respiratory distress syndrome (ARDS)
These two conditions will cause life-threatening hypoxaemia. Pulmonary oedema is more
common than ARDS. Both are aggravated by rapid infusion of large volumes of fluid during
the critical phase.
The goals of therapy are to optimize oxygenation and ventilation with respiratory support
and stabilize the haemodynamic situation.
• Apart from increasing the fractional inspired oxygen, positive end-expiratory pressure
(PEEP) is essential to maintain adequate oxygenation and reduce the work of
breathing. PEEP can be delivered through non-invasive ventilation such as
continuous positive airway pressure (CPAP) and mechanical ventilation.
• Patients who are alert, cooperative and haemodynamically stable with no, or mild,
metabolic acidosis may benefit from non-invasive ventilation (6).
• If the patient is out of the plasma leakage phase and has stable haemodynamics,
intravenous fluid therapy should be discontinued and diuretic therapy can be
commenced cautiously. Please refer to Section 2.3.2.
• Indications for mechanical ventilation include:
o patients who have shock and are restless, combative or confused;
o respiratory failure from acute pulmonary oedema/ARDS ± shock;
o patients who fail to respond to non-invasive ventilation.
Undertaking a tracheal intubation in a dengue shock patient is a risky procedure for the
following reasons:
• Sedation and induction agents will precipitate hypotension and cardiac arrest in a
hypovolaemic patient.
46
• Unless the tidal volume and respiratory rate (RR) are increased to match those of the
patient’s own respiratory effort, muscle relaxants will worsen the metabolic acidosis
by removing the respiratory compensation. This will further precipitate hypotension
and cardiac arrest.
• The compliance of the respiratory system will be diminished by the pleural effusion
and ascites, so higher pressures will be required to achieve adequate ventilation and
oxygenation.
• Trauma to the airway will cause bleeding which will block the tracheal tube.
Possible benefits of sedation and mechanical ventilation are:
• easier monitoring of the haemodynamic status;
• adequate respiratory support will decrease the oxygen demand of the body.
The first step in carrying out this procedure is to inform the caregiver and the patient (if still
cognizant). Explain the risks and take measures to ensure the patient’s safety. These
include full monitoring of the ECG, pulse oximetry and BP, careful choice of sedation and
induction agents, and the availability of 500 ml of colloidal fluid and FWB if the patient is
evaluated to have hypovolaemia or have severe bleeding. A syringe of dilute dopamine
should be ready to be started if needed. An adequate size cuffed tracheal tube should be
selected in anticipation of decreased compliance of the respiratory system.
The role of each team member should be rehearsed. Intubation should be undertaken by
experienced personnel. Pre-oxygenation will be necessary with mask-bag ventilation,
synchronized to the patient’s breathing effort. Before administration of sedation, ensure that
the BP is within the normal range. The BP will decrease after sedation. You may start the
colloid infusion/blood transfusion and/or dopamine if the BP before sedation is on the low
side. Sedate the patient with midazolam or fentanyl or ketamine which may help increase
the BP. A fraction of the usual dose may be all that is required to put the patient to sleep,
so titrate to the conscious level so that the BP does not crash. The BP should be monitored
continuously. If appropriate equipment is not available, finger-on-radial/femoral pulse will
indicate when fluids and inotropes will be required to support the BP. Even mild
hypotension should be treated.
When the patient is sedated and their breathing becomes shallow, you should try to take
over their breathing by simulating the pre-sleep breathing pattern. This will prevent
worsening of metabolic acidosis. Cricoid pressure and a PEEP should be applied. A muscle
relaxant such as succinyl choline should be used if there is no hyperkalaemia, otherwise
use agents such as atracurium or rocuronium. Atropine should be used to prevent
bradycardia. Oral-tracheal intubation should be performed with minimal trauma to the
airway. Once the position of the tracheal tube has been verified, the cuff should be inflated
and cricoid pressure released. An orogastric tube should be inserted to remove stomach
contents.
Manual ventilation breaths should simulate the patient’s pre-sleep breathing pattern to
enable respiratory compensation of metabolic acidosis, thus maintaining the pH within the
normal range. It is essential to remember this principle: the immediate goal is to achieve
adequate oxygenation and ventilation to a near-normal pH of about 7.35 rather than a
normal partial pressure of carbon dioxide (PaCO2). In metabolic acidosis, the PaCO2 level
has to be low to maintain a normal pH. Ventilator settings should be adjusted to attain this
47
goal. As the patient’s circulation improves with blood transfusion and other fluids, the
metabolic acidosis will subside. Likewise, ventilator settings have to be scaled back;
otherwise the pH will increase to the alkalosis range.
PEEP will be required to restore lung volume compressed by pleural effusion and ascites.
Remember that PEEP will decrease venous return: this may be desired in intravascular
hypervolaemia, but may precipitate hypotension in intravascular hypovolaemia. However, it
is best to maintain PEEP ≤ 10 cm H2O to avoid increasing intra-abdominal pressure that
will in turn reduce urine output. Adjust the fractional inspired oxygen accordingly. If large
pleural effusions or ascites interfere with ventilation and oxygenation, a careful aspiration of
the thorax with a small bore needle may be performed, but there is a high risk of bleeding
from the procedure.
The next immediate task is to stabilize the haemodynamic state. Refer to the key questions
again. Verify the state of the intravascular volume by clinical examination, jugular venous
pressure, chest radiography or echocardiography and blood gases. The central venous
pressure may not accurately predict the intravascular volume, particularly if PEEP is used.
Arterial blood gas analysis and serial lactate will add valuable information. Lactic acidosis
and a low/normal haematocrit indicate severe bleeding and the need for urgent transfusion
of FWB. The haemodynamic state and urine output should be evaluated frequently
throughout the process of blood transfusion. Any inotropes that have been used to support
the BP during tracheal intubation should be decreased to the minimum possible level. If this
is not possible, there could be unrecognized hypovolaemia and/or unrecognized severe
bleeding and/or myocarditis, verified with echocardiography.
Once the haemodynamic state has been stabilized, intravenous fluid therapy should be
reduced accordingly and stopped when the point of 48 hours of plasma leakage has been
reached. A low infusion of glucose-isotonic fluid may be required to maintain euglycaemia
in young children with liver impairment. The urine output should gradually increase. A urine
volume of ~0.5 ml/kg/hour is adequate. Furosemide may be indicated in those with massive
positive fluid balances and those with increased creatinine levels.
The urine volume is a reliable indicator of end-organ perfusion as fluid repletion gets
underway. Urine output should not be used as an indicator of organ perfusion under the
circumstances listed below:
• blood glucose exceeds the renal threshold of 10 mmol/L;
• there is acute kidney injury, as manifested by elevated serum creatinine levels;
• there is underlying chronic renal failure, or uncontrolled hypertension or diabetes
mellitus;
• moderately high PEEP is used;
• intra-abdominal pressure is elevated;
• administration of furosemide or hyperoncotic fluids.
48
2.3.3 Co-infections and nosocomial infections
Co-infections with gram-negative bacteria have been reported in patients with diabetes
mellitus and renal failure. Other tropical diseases such as leptospirosis, typhus, malaria,
chikungunya and enteric fever may occur concomitantly. A high index of suspicion is
necessary to recognize this, especially in those with atypical presentations such as
prolonged fever, pulmonary haemorrhage, unexplained renal failure or liver failure in the
absence of shock.
It is not uncommon for patients to acquire a nosocomial infection, especially those with
severe dengue and when intravenous therapy has been prolonged. Careful attention to
aseptic techniques is necessary in procuring and accessing intravascular devices. Prompt
and appropriate antibiotic therapy will be crucial to prevent morbidity and mortality.
2.3.4 Haemophagocytic syndrome
Evidence of haemophagocytosis in dengue was alluded to by the presence of numerous
macrophages that have phagocytosed erythrocytes and lymphocyte phagocytosis in the
spleen (7, 8). The unusual incidence of phagocytic reticulum cells which phagocytosed all
blood elements has been reported (9−11).
The clinical significance of reactive haemophagocytosis in dengue has not been studied.
However, case reports of prolonged fever in dengue patients have been attributed to this
phenomenon. The clinical picture is characterized by persistent high fever, variable
cytopenia and multi-organ failure associated with macrophage activation,
haemophagocytosis and hypercytokinaemia. Serum ferritin levels are markedly elevated.
Definitive diagnosis is made by bone marrow biopsy which demonstrates haemophagocytic
activity. Response to methyl-prednisolone and immunoglobulin has been reported to be
dramatic (12). However, supportive treatment leading to spontaneous recovery has also
been reported (13 and 14).
2.3.5 Supportive care and adjuvant therapy
Supportive care and adjuvant therapy may be necessary in severe dengue and includes:
Vasopressor and inotropic therapy
The use of vasopressor and inotropic therapy should be limited to the following clinical
situations:
• As a temporary measure to prevent life-threatening hypotension in dengue shock and
during induction for intubation, while correction of intravascular volume is being
vigorously carried out. The recommended choice of vasopressor is dopamine which
should be titrated to maintain mean arterial BP of 65 mmHg in adults. Vasopressor
therapy should be weaned off as intravascular volume is restored and end-organ
perfusion re-established.
• Evidence of cardiogenic shock due to myocarditis or ischemic heart disease.
Dobutamine is the recommended choice.
In concomitant septic shock, dopamine or norepinephrine are the vasopressors of choice.
Vasopressor therapy should be carefully monitored since dengue shock is primarily a
hypovolaemic shock caused by plasma leakage ± haemorrhage. The most essential and
effective strategy is the correction of intravascular volume with the appropriate types of
49
fluids. Vasopressors, by further increasing the peripheral vascular resistance, may be able
to maintain the central BP but without improving end-organ perfusion. Paradoxically,
vasopressors exacerbate tissue hypoxia and lactic acidosis when the intravascular volume
has not been restored. The correct use of vasopressor therapy is reflected in an increased
BP concomitant with a decreased tachycardia. If both the BP and tachycardia increase,
then repletion of intravascular volume should be considered as the urgent alternative
strategy.
Central venous pressure monitoring
The use of central venous pressure (CVP) to guide fluid therapy in severe dengue with
profound or prolonged shock has been described (15). Although the CVP provides an
additional parameter to gauge the intravascular volume, its reliability in predicting left
ventricular filling volume and hemodynamic response to fluid challenge has been contested
(16, 17). The risks of severe bleeding and pneumothorax due to placement of the central
venous line could be minimized with the use of ultrasound guidance. A more reliable
method to evaluate the intravascular volume status in patients with severe dengue is
echocardiography.
Renal replacement therapy
Renal replacement therapy may be indicated in acute kidney injury. It should be
commenced after haemodynamic stability has been achieved and maintained without
further fluid resuscitation, usually after the critical period of plasma leakage. The preferred
choice of renal replacement therapy is continuous veno-venous haemodialysis (CVVH).
Peritoneal dialysis may be considered if CVVH is not available, but there is a risk of
bleeding. However, commencement of CVVH during the critical phase when hypovolaemia
has not been corrected will lead to technical problems. The placement of a large dialysis
catheter in a hypovolaemic patient may cause unnecessary trauma and severe bleeding.
CVVH is not possible in the hypovolaemic and hypotensive state. When renal replacement
therapy is not available or cannot be performed yet, the ensuing hyperuricaemia,
hyperkalaemia and hyperphosphataemia should be managed with allopurinol, Resonium A
and calcium carbonate respectively.
Other organ impairment
Drug toxicity resulting from the use of paracetamol or acetaminophen should be suspected
if liver enzymes have increased disproportionate to the severity of shock (18). Paracetamol
should be discontinued in patients with liver enlargement or raised liver enzymes. Further
treatment of organ impairment, such as severe hepatic involvement, encephalopathy or
encephalitis may be needed − otherwise cardiac abnormalities, such as conduction
abnormalities, may occur (the latter usually not requiring interventions). The most critical
issue for recovery is stabilization of the haemodynamic state; without this there can be no
recovery of any organ. Once the critical period is over and stability of the haemodynamic
state attained, it is essential to stop or reduce intravenous fluids to the minimum and to
maintain euglycaemia. The body will heal itself remarkably over the next few days to
weeks. Excessive fluid is cleared by the kidneys and normal liver function returns gradually,
coagulation abnormalities and platelet counts return to normal. During this period any
suspected nosocomial sepsis should be treated vigorously, but without adding further
insults to the kidneys or liver. Supportive treatment (for the liver and kidneys) and enteral
nutrition is all that is required in most cases.
50
2.3.6 References
1. Lum LC et al., Fulminant hepatitis in dengue virus infection. Southeast Asian Journal of Tropical Medicine of
Public Health, 1993, 24(3):467−471.
2. Venkataraman S, Suresh S, Anjelivelil JJ. Dengue haemorrhagic fever and fulminant hepatic failure.
Digestive Diseases and Sciences, 2005, 50(6):1146−1147.
3. Lum LC et al., Dengue encephalitis: a true entity? American Journal of Tropical Medicine and Hygiene,
1996, 54(3):256−259.
4. Miagostovich MP, Ramos RG, Nicol AF. Retrospective study on dengue fatal cases. Clinical
Neuropathology, 1997, 16:204−208.
5. Solomon T et al., Neurological manifestations of dengue infection. Lancet, 2000, 355:1053−1059.
6. Cam BV et al., Randomized comparison of oxygen mask treatment vs. nasal continuous positive airway
pressure in dengue shock syndrome with acute respiratory failure. Journal of Tropical Pediatrics, 2002,
48(6):335−339.
7. Aung-Khin M et al. Changes in the tissues of the immune system in dengue haemorrhagic fever. Journal of
Tropical Medicine and Hygiene, 1975, 78:256.
8. Bhamarapravati N, Tuchinda P, Boonyapaknavig V. Pathology of Thailand haemorrahgic fever: A study of
100 autopsy cases. Annals of Tropical Medicine and Parasitology, 1967, 61:500.
9. Kho LK, Wilbur H, Himawan T. Blood and bone marrow changes in dengue haemorrhagic fever. Paediatrica
Indonesia, 1972, 12:31−39.
10. Na-nakorn S et al., Bone marrow studies in Thai haemorrhagic fever. Bulletin of the World Health
Organization, 1966, 35:54−55.
11. Nelson ER, Bierman HR, Chulajata R. Hematologic phagocytosis in postmortem bone marrows of dengue
hemorrhagic fever. American Journal of Medical Science, 1966, 252:68−74.
12. Tanomsri Srichaikul et al., Hemophagocytic syndrome in dengue hemorrhagic fever with severe multiorgan
complications. Journal of the Medical Association of Thailand, 2008, 91(1):104−109.
13. Lu PL et al., Dengue virus-associated hemophagocytic syndrome and dyserythropoiesis: A case report. The
Kaohsiung Journal of Medical Sciences, 2005, 21:34−38.
14. Jain D, Singh T. Dengue virus related hemophagocytosis: a rare case report. Hematology, 2008,
13(5):286−268.
15. Hung NT et al., Volume replacement in infants with dengue hemorrhagic fever/dengue shock syndrome.
American Journal of Tropical Medicine and Hygiene, 2006, 74(4):684−691.
16. Kumar A et al. Pulmonary artery occlusion pressure and central venous pressure fail to predict ventricular
filling volume, cardiac performance, or the response to volume infusion in normal subjects. Critical Care
Medicine, 2004, 32:691−699.
17. Marik PE, Baram M, Vahid B. Does central venous pressure predict fluid responsiveness? A systematic
review of the literature and the tale of seven mares. Chest, 2008, 134(1):172−178.
18. Dumortier C et al., Factors associated with increased serum alanine aminotransferase levels during the
French Guiana dengue epidemic of 2005–2006. Infectious Diseases in Clinical Practice, 2010,
18:41−45.
51
2.4 Treatment of dengue in specific risk groups
The rapid increase in extent of dengue infections around the globe has resulted in a shift in
dengue disease burden to older children and adults. Age-related differences in dengue
disease and its severity have been observed.
2.4.1 Dengue in adults – some specific issues
While there are many similarities in the disease course, there are minor but significant
differences in clinical manifestations and laboratory findings between adults and children
with dengue.
Clinical manifestations and laboratory findings
Some clinical manifestations such as petechiae, melaena, headache, retro-orbital pain,
joint pain, myalgia, nausea and vomiting may be more common in adults, while epistaxis,
oliguria and liver enlargement are more common among children (1). Adults have higher
baseline haematocrit levels than children. Among adults, males have a higher baseline
haematocrit than females. Among the bleeding manifestations, haematuria and
menorrhagia may be more prevalent in adults whereas haematemesis and melaena are
more often seen in children (2). Infants, followed by children and then adults are most
susceptible to plasma leakage and thrombocytopenia.
Issues in management
• Recognition of plasma leakage
Due to its name “dengue haemorrhagic fever”, a general clinician may become overly
concerned about the risks of bleeding. The name “dengue haemorrhagic fever” may also
create panic in some doctors and the general public. The critical phenomenon of plasma
leakage that leads dengue patients to the initial shock is often overlooked. Refer to Chapter
3, pitfall 13. Clinicians should familiarize themselves with the warning signs of severe
dengue and early signs of shock.
• Recognition of shock
Young adult patients with normal cardiorespiratory function can compensate for
hypovolaemic shock for several hours. As with children, their mental status is clear and
they may even be able to work until the stage of profound shock. Other manifestations of
shock such as tachycardia, tachypnoea, cold peripheries, delayed capillary refill, reduced
urine output (with concentrated urine), are subtle but very important parameters that should
alert the physician to the early recognition of shock and prompt fluid correction. A pulse
pressure of ≤ 20 mmHg that defines shock in children often indicates a far more advanced
state of shock in adults.
Patients self-medicate with analgesics such as paracetamol, NSAIDs, anti-emetics and
other drugs that worsen liver and platelet functions. Adult patients should be advised to
take no more than 3 g per day of paracetamol (3). In addition, some adult patients may not
even be aware of the fever themselves and hence may delay seeking medical attention.
52
• The impact of increased co-morbidities
Refer to Section 2.4.3).
• Dual-infection
Concurrent bacteraemia or dual-infections have been reported in adult dengue. Due to the
overlapping clinical manifestations, concurrent bacteraemia can be overlooked in a dengue
endemic setting. Prolonged fever (> 5 days) and acute renal failure may predict the
presence of dual infection (4). Failure to make a timely diagnosis of dual infection and to
start timely and appropriate antibiotic therapy will lead to increased morbidity and mortality.
53
2.4.3 Dengue in cases with co-morbidities
Co-morbidities such as diabetes mellitus (see Table 8), hypertension (see Table 7) and
renal insufficiency are significantly associated with severe dengue and higher fatality rates
(6).
Haemolytic anaemias – sickle-cell anaemia, thalassaemias, hereditary spherocytosis,
glucose-6-phosphate dehydrogenase (G6PD) deficiency and other
haemoglobinopathies (10, 11 and 12)
Acute dengue illness may precipitate intravascular or extravascular haemolysis in patients
with haemolytic anaemias. The patient appears pale and jaundiced. Reticulocytosis, a
normal response of the bone marrow to haemolysis may be absent in these dengue
patients because of bone marrow suppression.
Other markers for haemolysis should be checked. Haemoglobinuria (black urine) is an
indicator of severe intravascular haemolysis while plasma haptoglobin will be low or absent
if significant intravascular or extravascular haemolysis has occurred.
Low baseline heamatocrit in thalassaemias and other chronic haemolytic anaemias
Without knowledge of the low baseline haematocrit, haemoconcentration (raised
haematocrit) during plasma leakage will be missed in anaemic patients with dengue.
Despite severe plasma leakage the haematocrit may seem to be normal. Hypochromic and
microcytic red blood cells may alert the clinician to the underlying baseline anaemia.
Clinicians should interpret the haematocrit data in the light of the patients’ clinical status
and make reference to the baseline haematocrit level whenever available, particularly
during the plasma leakage phase. Transfusion with fresh packed red cells or fresh whole
blood should be given if significant haemolysis is suspected.
Haemoglobinuria and acute renal injury
Haemoglobinuria following intravascular haemolysis may cause rapid deterioration of renal
function resulting in acute renal injury. The renal hypoperfusion of dengue shock
exacerbates this organ impairment.
Adequate hydration is essential to prevent acute renal injury from haemoglobinuria. Prompt
and appropriate blood transfusion is necessary if there is evidence of significant
haemolysis. Diuretics and alkalinization normally used to counter the effects of
haemoglobinuria have to be instituted carefully and in a timely way. Diuretics should be
avoided during the plasma leakage phase when hypovolaemia has yet to be corrected.
54
Hypertension
Table 7. Challenges when managing dengue patients with pre-existing hypertension
Interpretation of BP Hypotension is a late sign of shock. However, in patients with uncontrolled
hypertension a BP reading that is considered normal for age may, in reality, be low
for patients with uncontrolled hypertension. Similarly, what is considered as “mild”
hypotension may in fact be profound. Patients with chronic hypertension should be
considered to be hypotensive when the mean arterial pressure (MAP) declines by
40 mmHg from the baseline, even if it still exceeds 60 mmHg. (For example, if the
baseline MAP is 110 mmHg, a MAP reading of 65 mmHg should be considered as
significant hypotension.) Look for other manifestations of shock (see Section
2.2.3.1).
The heart rate It is essential to know the specific antihypertensive agent a patient is taking for the
response: following reasons.
Bradycardia: ß-blockers, a common antihypertensive medication, cause bradycardia and may
block the tachycardic response in shock. The heart rate should not be used as an
assessment of perfusion in patients on ß-blockers.
Tachycardia: Antihypertensive agents such as calcium channel blockers may cause tachycardia.
Tachycardia in these patients may not indicate hypovolemia. Knowing the baseline
heart rate before the dengue illness is helpful in the haemodynamic assessment.
The impact on The continuation of antihypertensive agents during the acute dengue illness should
hypotension: be evaluated carefully during the plasma leaking phase. The BP lowering effects of
these agents and diuretic therapy may exacerbate the hypotension and
hypoperfusion of intravascular volume depletion.
End-organ damage Heart failure and renal failure are common complications of chronic uncontrolled
from chronic hypertension. Clinicians should be aware if there is pre-existing or new onset of end-
hypertension: organ damage. Interpretation of urine output as a marker of renal perfusion has to
be revoked in these situations.
55
Management
Dengue patients with known diabetes mellitus should be admitted for closer monitoring of
the diabetic as well as dengue states. If the patient has gastrointestinal disturbances, blood
glucose should be controlled with intravenous short-acting insulin during the dengue
illness.
A validated protocol for insulin dose adjustments to a target glucose level of < 150 mg/dl
(8.3 mmol/L) should be used. A source of glucose may be maintained once the target is
achieved while receiving intravenous insulin. Blood glucose should be monitored every 1–2
hours until glucose values and insulin rates are stable and then every 4 hours thereafter.
Electrolyte imbalances
Hyponatraemia, hypokaleamia, hyperkalaemia or hypomagnesaemia are common
electrolyte imbalances observed during dengue illness. These imbalances may be more
severe and require prompt attention and appropriate actions.
Chronic renal failure
Dengue patients with chronic renal failure (CRF) have a significantly higher risk of severe
dengue and mortality. The outcome correlates with the renal function (13, 14).
Warning signs in severe dengue versus the presentation of uraemia:
The warning signs of severe dengue are similar to those of uraemia in CRF. Ascites and/or
pleural effusion, and signs of plasma leakage in dengue, are not uncommon findings in
patients with CRF and fluid retention. The ambiguity of these symptoms and signs could
delay the recognition of plasma leakage and severe dengue.
Low baseline haematocrit and platelet count:
Patients with CRF have a low baseline haematocrit. Refer to 2.4.3 above: Low baseline
haematocrit in thalassaemias
A low baseline platelet count is not an uncommon finding in dialysis patients. This could be
another challenge in the early recognition of dengue infection.
Challenges in fluid management:
• Narrow window of fluid tolerance.
Patients with CRF have limited fluid tolerance. Frequent assessments of the
haemodynamic state and frequent fluid regime adjustments are mandatory to avoid
fluid overload or under-fill. Adequate fluid replacement is necessary to prevent
worsening of renal function during the critical phase.
• Urine output
The urine output should not be used as an indicator of the intravascular volume
status because patients with CRF can have either low or high urine-output renal
failure. Low urine output in CRF contributes to the risk of fluid overload whereas high
urine output may aggravate hypovolaemia.
• Limited effect of diuretics
Diuretics have a limited effect in CRF, making patients more susceptible to fluid
overload. Dialysis may be required.
56
Acid base balance and electrolyte balance
Patients with CRF are at risk of metabolic acidosis and electrolyte imbalance which will
become worse during dengue shock. If these persist after adequate fluid replacement,
dialysis may be considered after haemodynamic stability is achieved (see Section 2.3.5).
Platelet dysfunction
Platelet dysfunction, well recognized in CRF together with severe thrombocytopenia ±
coagulopathy, predispose the dengue patient to severe bleeding that may be difficult to
control.
Chronic heart disease with or without heart failure
Congenital or acquired cardiac lesions such as valvular heart disease or ischaemic heart
disease, especially the latter, are common co-morbidities in adults or the elderly. Patients
may be well compensated under normal circumstances. However, an infection such as
dengue that causes high fever, tachycardia and increased metabolic demands may
precipitate decompensation of cardiac functions. Such patients have limited ability to
compensate for hypovolaemia or hypervolaemia. Therefore, fluid therapy should be guided
by frequent clinical assessments, haematocrit and blood gas determinations. Patients with
cyanotic heart diseases have polycythemia and a high baseline haematocrit.
• Positive pressure ventilation
Non-invasive positive pressure ventilation should be considered to support patients
with cardiac decompensation. Failing this, mechanical ventilation should be instituted.
• The use of loop diuretics
Loop diuretics should be used cautiously and in a timely way: after achieving
haemodynamic stability when intravenous fluid therapy has been discontinued or
reduced and in patients with fluid overload.
57
2.4.4 References
1. Kittigul L et al., The differences of clinical manifestations and laboratory findings in children and adults with
dengue virus infection. Journal of Clinical Virology, 2007, 39(2):76−81.
2. Hammond SN et al., Differences in dengue severity in infants, children and adults in Nicaragua. American
Journal of Tropical Medicine and Hygiene, 2005, 73(6):1063–1070.
3. Thomas L et al., Predictors of severe manifestations in a cohort of adult dengue patients. Journal of Clinical
Virology, 2010, 48(2):96−99.
4. Lee IK, Liu JW, Yang KD. Clinical characteristics and risk factors for concurrent bacteraemia in adults with
dengue haemorrhagic fever. American Journal of Tropical Medicine and Hygiene, 2005, 72(2):221−226.
5. Garcia-Rivera EJ, Rigau-Perez JG. Dengue severity in the elderly in Puerto Rico. Pan American Journal of
Public Health, 2003, 13:362–336.
6. Lee MS et al., Clinical characteristics of dengue and dengue hemorrhagic fever in a medical center of
southern Taiwan during the 2002 epidemic. Journal of Microbiology, Immunology and Infection, 2006,
39(2):121−129.
7. Guzman MG et al., Effect of age on outcome of secondary dengue 2 infections. International Journal of
Infectious Diseases, 2002, 6:118−124.
8. Lee IK, Liu JW, Yang KD. Clinical and laboratory characteristics and risk factors for fatality in elderly
patients with dengue hemorrhagic fever. American Journal of Tropical Medicine and Hygiene, 2008,
79(2):149–153.
9. Bravo JR, Guzmán MG, Kouri GP. Why dengue haemorrhagic fever in Cuba? Individual risk factors for
dengue haemorrhagic fever/dengue shock syndrome (DHF/DSS). Transactions of the Royal Society of
Tropical Medicine and Hygiene, 1987, 81: 816–820.
10. Pongtanakul B et al., Dengue hemorrhagic fever in patients with thalassemia. Journal of the Medical
Association of Thailand, 2005, 88 Suppl.8:S80−85.
11. Jetsrisupard A et al., Development of severe anemia during fever episodes in patients with hemoglobin E
trait and hemoglobin H disease combinations. Journal of Pediatric Hematology/Oncology, 2006,
28(4):249−253.
12. Tanphaichitr VS et al., Effect of red blood cell glucose-6-phosphate dehydrogenase deficiency on patients
with dengue hemorrhagic fever. Journal of the Medical Association of Thailand, 2002, 85 Suppl
2:S522−529.
13. Kuo MC et al., Difficulty in diagnosis and treatment of dengue hemorrhagic fever in patients with chronic
renal failure: report of three cases of mortality. American Journal of Tropical Medicine and Hygiene,
2007, 76:752–756.
14. Mei-Chuan Kuo et al., Impact of renal failure on the outcome of dengue viral infection. Clinical Journal of the
American Society of Nephrology, 2008, 3:1350–1356.
58
2.4.5 Dengue in pregnancy
Introduction
In the recent decade more cases of dengue in pregnancy are being reported. The clinical
manifestations, treatment and outcome of dengue in pregnant women are similar to those
of non-pregnant women but with some important differences (1, 2).
Misdiagnosis or delayed diagnosis are not uncommon due to some of the overlapping
clinical and/or laboratory features with the better recognized conditions of pregnancy.
These include eclampsia or pre-eclampsia, haemolysis, elevated liver enzymes and low
platelet count (HELLP) syndrome, pneumonia, pulmonary embolism, various obstetric
causes of per-vaginal bleeding and other infectious diseases (Table 9).
Haemolysis - - +
DIVC = disseminated intravascular coagulopathy; FBC = full blood count; HELLP = haemolysis, elevated liver
enzymes and low platelet count; WBC = white blood cell
In order to recognize and diagnose dengue disease early in pregnancy, clinicians need to
maintain a high index of suspicion when dealing with pregnant women who present with
febrile illness after travelling to, or living in dengue-endemic areas.
59
Impact of dengue on pregnancy:
• Adverse pregnancy outcome (2–6)
It is still uncertain whether dengue is a significant factor for adverse pregnancy
outcomes such as preterm birth, low-birth weight and caesarean deliveries, as most
of the published data were based on hospitalized patients.
• Risk of vertical transmission (2, 6–10)
The risk of vertical transmission is well established among women with dengue
during the perinatal period (see Section 2.4.5).
• Significant impact of dengue at parturition (5,11)
Severe bleeding may complicate delivery and/or surgical procedures performed on
pregnant patients with dengue during the critical phase, i.e. the period coinciding with
marked thrombocytopenia with or without coagulopathy and vasculopathy.
Management of dengue during pregnancy:
• Early admission for close monitoring is recommended, especially for women close to
full-term/labour.
• Conservative medical and obstetrical management is the treatment of choice (12).
Challenges in recognition of dengue disease and plasma leakage in pregnancy
• Symptoms of hyperemesis during the first trimester of pregnancy resemble the
warning signs of severe dengue and this may delay the recognition of severe dengue.
• After the second trimester of pregnancy it is normal to see an increase in circulating
blood volume with generalized vasodilatation, resulting in an increased baseline heart
rate and lower baseline BP, as well as a lower baseline haematocrit. This can
confuse the diagnosis of dengue and therefore clinicians need to be alert to the
following:
60
• There is no difference in fluid therapy compared with the non-pregnant state (see
section on fluid management in Section 2.2). However it is important to note that the
growing gravid uterus may result in narrower tolerance of fluid accumulation in the
peritoneal and pleural cavity from plasma leakage. Hence excessive fluid
replacement should be avoided.
• The increased baseline heart rate and a lower baseline BP are normal physiological
changes in late pregnancy. Targeting an inappropriate heart rate and “normal” levels
of BP could result in fluid overload and respiratory distress.
• The presence of wounds or trauma during the critical phase of dengue with marked
thrombocytopenia, coagulopathy and vasculopathy creates a substantial risk of
severe haemorrhage.
• If severe haemorrhage occurs, replacement with transfusion of fresh whole
blood/fresh packed red cells should be promptly instituted (see Section 2.2.3.3).
• Prophylactic platelet transfusion is not recommended unless obstetrically indicated.
• Delivery should take place in a hospital where blood/blood components and a team of
skilled obstetricians and a neonatologist are available.
• Tocolytic agents and measures to postpone labour to a suitable time may be
considered during the critical phase of dengue illness. However there is currently a
lack of evidence on this practice.
Inevitable delivery during critical phase
• If delivery is inevitable, bleeding should be anticipated and closely monitored.
• Blood and blood products should be cross-matched and saved in preparation for
delivery.
• Trauma or injury should be kept to the minimum if possible.
• It is essential to check for complete removal of the placenta after delivery.
• Transfusion of platelet concentrates should be initiated during or at delivery but not
too far ahead of delivery, as the platelet count is sustained by platelet transfusion for
only a few hours during the critical phase.
• Fresh whole blood/fresh packed red cells transfusion should be administered as soon
as possible if significant bleeding occurs. If blood loss can be quantified, it should be
replaced immediately. Do not wait for blood loss to exceed 500 ml before
replacement, as in postpartum haemorrhage. Do not wait for the haematocrit to
decrease to low levels.
• Ergotamine and or oxytocin infusion as per standard obstetrical practice should be
commenced to contract the uterus after delivery to prevent postpartum haemorrhage.
• Refer to Section 2.2.3.3 for management of severe bleeding.
Post-delivery
• Newborns with mothers who had dengue just before or at delivery, should be closely
monitored in hospital after birth in view of the risk of vertical transmission (7, 8), see
Section 2.4.5.
61
o At or near-term/delivery, severe foetal or neonatal dengue illness and death may
occur when there is insufficient time for the production of protective maternal
antibodies.
o Clinicians should be aware that presentation in either maternal or neonatal
disease may be atypical and confound diagnosis.
• Congenital infection could eventually be suspected on clinical grounds and then
confirmed in the laboratory.
2.4.6 References
1. Waduge R et al. Dengue infections during pregnancy: a case series from Sri Lanka and review of literature.
Journal of Clinical Virology, 2006, 37(1):27−33.
2. Tan PC et al., Dengue infection in pregnancy: prevalence, vertical transmission, and pregnancy outcome
Obstetrics & Gynecology, 2008, 111(5):1111−1117.
3. Carles G et al., Dengue fever in pregnancy. A study of 38 cases in French Guiana. European Journal of
Obstetrics & Gynecology and Reproductive Biology (Paris), 2000, 29(8):758−762.
4. Seneviratne SL, Perera J, Wijeyaratne C. Dengue infections and pregnancy: caution in interpreting high
rates of premature deliveries and maternal mortality. Southeast Asian Journal of Tropical Medicine and
Public Health, 2007, 38(1):195−196.
5. Basurko C et al., Maternal and fetal consequences of dengue fever during pregnancy. European Journal of
Obstetric Gynecology and Reproductive Biology, 2009, 147 (1):29−32.
6. Pouliot SH et al., Maternal dengue and pregnancy outcomes: a systematic review. Obstetrical &
Gynecological Survey, 2010, 65(2):107−118.
7. Perret C et al., Dengue infection during pregnancy and transplacental antibody transfer in Thai mothers. The
Jounal of Infection, 2005, 51(4):287−293.
8. Sirinavin S et al., Vertical dengue infection: case reports and review. The Paediatric Infectious Disease
Journal, 2004, 23(11):1042−1047.
9. Kerdpanich A et al., Perinatal dengue infection. The Southeast Asian Journal of Tropical Medicine and
Public Health, 2001, 32(3):488−493.
10. Fernandez R et.al., Study of the relationship dengue-pregnancy in a group of Cuban-mothers. Revista
Cubana de Medicina Tropical, 1994, 46:76−78.
11. Thaithumyanon P et al., Dengue infection complicated by severe hemorrhage and vertical transmission in a
parturient woman. Clinical Infectious Diseases, 1994, 18(2):248−249.
12. Carroll ID, Toovey S, Van Gompel A. Dengue fever and pregnancy - a review and comment. Travel
Medicine and Infectious Disease, 2007, 5:183−188.
62
2.4.7 Dengue in paediatric cases
Dengue virus infections affect human populations of all age groups worldwide. In some
parts of the world, dengue is mainly a paediatric health problem. The vast majority of
dengue cases occur in children < 15 years of age and around 5% of all severe dengue
cases occur in infants (1–4). In one dengue-endemic area, the incidence of dengue
infection exceeded 10% in infants aged 2–15 months (5). Most infants acquire primary
dengue virus infections (1, 5–6). Clinical manifestations, the course of dengue illness and
management of dengue in older children and adults are well documented in the previous
sections of this handbook. In this section clinical manifestations and management of
dengue in infants will be addressed and compared to clinical manifestations and
management of dengue in older children and adults. Vertical transmission and dengue in
neonates are also covered briefly.
Manifestations of dengue in infants
As in older children and adults, dengue virus can cause a spectrum of outcomes in infants,
ranging from asymptomatic infection to mild or clinically significant, severe disease (5). The
burden of severe dengue lies predominantly in infants 4–9 months of age (1, 4, 6).
Infants with dengue typically have high fever that usually lasts 2–7 days; the same as in
older children. Upper respiratory tract symptoms (cough, nasal congestion, runny nose,
dyspnoea), gastrointestinal symptoms (vomiting, diarrhoea), and febrile convulsions are
more common in infants with dengue compared to older children (3, 5, 6). Differentiation
between dengue and other common infections in infants (such as pneumonia, bacterial
sepsis, meningoencephalitis, hand foot and mouth disease, measles, rotavirus infections,
etc.) is often not possible at the febrile stage. The presence of a febrile seizure, macular
rash, petechiae and lower platelet counts early in the illness are significantly associated
with dengue among infants with acute undifferentiated febrile illness (5).
In the majority of dengue infants, an increase in capillary permeability, in parallel with
increasing haematocrit levels, becomes apparent around the time of defervescence (which
usually falls on days 3–6 of illness). The period of clinical plasma leakage lasts 24–48
hours (3, 6, 7). During this critical phase, clinical features and laboratory findings of infant
dengue become more prominent. Skin bleeding such as petechiae, mucosal membrane
bleeding (e.g. of the nose and gums), and gastrointestinal bleeding may occur.
Hepatomegaly is usually noted (2, 6, 7, 8). Splenomegaly is seen in almost 10% of dengue
infants, seven times more frequently than in older children (2, 6, 8).
Shock occurs when a critical volume of plasma is lost through leakage. As with older
children, it is often preceded by warning signs. The body temperature may be subnormal
when shock occurs. However, some infants may still have fever at the onset of shock; in
these patients a differential diagnosis of septic shock should be kept in mind (6). With
prolonged shock, the consequent organ hypoperfusion results in multiple organ
dysfunction, metabolic acidosis and disseminated intravascular coagulation.
The degree of increase above the baseline haematocrit often reflects the severity of
plasma leakage. Haemoconcentration, manifested by an increase in haematocrit of ≥ 20%
above the baseline haematocrit may be seen (6, 9). The normal value of haematocrit in
infants 2–12 months of age is relatively low (28–42%) (10) and may be even lower in iron-
deficiency anaemia. The mean maximal haematocrit values in dengue infants vary from
31.1−40.8% (range = 30–60%) (2, 6, 11). Thrombocytopenia and leukopenia are often
63
observed in this phase. Liver involvement and/or dysfunctions, as indicated by mean
aspartate aminotransferase/alanine aminotransferase (AST/ALT) elevation and prolonged
prothrombin time compared to children, are found more frequently in infants (3).
During the recovery phase, progression of infants with dengue is the same as that of
children and adults (see Section 1.1.3).
Management of dengue in infants
Severe dengue is less common in infancy but when it does occur the risk of dying is higher
than in older children and adults (3, 6, 11). Infants with dengue should be referred for in-
hospital management.
Management of dengue infants without warning signs
Treatment is supportive. Oral rehydration should be encouraged with oral rehydration
solution (ORS), fruit juice and other fluids containing electrolytes and sugar, together with
breastfeeding or formula feeding and/or solid food. Febrile seizures are more frequent in
infants and young children with dengue than in older patients. Parents or caregivers should
be instructed about fever control with antipyretics and tepid sponging. Advise the parent or
caregiver to bring the infant back to the nearest hospital immediately if the infant has any of
the warning signs.
Management of dengue infants with warning signs (see Section 2.2.2 for more details)
Intravenous fluid therapy is indicated when the infant has dengue with warning signs.
Judicious volume replacement of lost plasma by intravenous fluid therapy from this early
stage may modify the course and severity of the disease. Only isotonic crystalloid solutions
such as Ringer's lactate (RL), Ringer's acetate (RA), or 0.9% saline solution should be
used. Start with 5–7 ml/kg/hour for 1–2 hours and then adjust the rate according to the
patient's clinical response. A recent study indicates that 79.6% of non-shock dengue infants
require intravenous fluid therapy with the mean volume of 102.1 ml/kg over a mean time of
25.9 hours (11). The percentage of dengue infants requiring parenteral fluid therapy
(79.6%) is higher than that of non-shock dengue children older than 1 year of age (25–
33%) (11). Parenteral fluid therapy is only required for 24–48 hours in most infant patients
since the capillary leak resolves spontaneously after this time (3, 11).
Management of infants with severe dengue
Treatment of shock (see Section 2.2.3.1 for more details)
Volume replacement in infants with dengue shock is a challenging management problem.
The strategy for fluid resuscitation in infants is similar to that in children and adults. The
recommended regimen for the treatment of infants with dengue shock is as follows:
• Immediate and rapid replacement of the plasma loss with isotonic crystalloid
solutions or, in the case of profound shock, colloid solutions.
• Continued replacement of further plasma losses to maintain effective circulation for
24–48 hours.
• Correction of metabolic and electrolyte disturbances.
• Blood transfusion – only to cases with severe bleeding.
64
Treatment of compensated shock in infants: Please refer to Section 2.2.3.1
Treatment of profound shock (hypotensive; undetectable pulse and BP) in infants:
Please refer to Section 2.2.3.1
Clinicians who take care of dengue-shock infants should remember that an infant with a low
baseline haematocrit of 30%, presenting with dengue shock and a haematocrit of 40%, is
relatively more haemoconcentrated than another child with a baseline value of 42% and a
haematocrit of 50% at the time of shock.
In patients with profound, recurrent or prolonged shock, a central venous catheter may be
inserted through the antecubital basilic vein or internal jugular vein to guide intravenous
fluid therapy. This should be done by an experienced member of staff using ultrasound to
guide the insertion (if available).
In infants intravenous fluids must be administered with special care to avoid fluid overload.
Fluids account for a greater proportion of body weight in infants than children and minimum
daily requirements are correspondingly higher. Infants have less intracellular fluid reserve
than older children and adults. Moreover, capillary beds are intrinsically more permeable
than those of older children or adults. Both early cardiovascular compromise and significant
fluid overload are more likely if capillary leaks occur in these circumstances (12).
Monitoring dengue-shock infants
Dengue-shock infants should be under close observation around the clock until it is certain
that danger has passed. Generally, the duration of intravenous therapy should not exceed
24−48 hours after the infant is out of shock. Stop intravenous fluid therapy when the infant's
condition has been stable for more than 24 hours or if there is any sign or symptom of fluid
overload (1, 6−8); see Sections 2.2.3.2 and 2.3.2.
Treatment of haemorrhagic complications, hyponatraemia, and metabolic acidosis
Blood transfusion is only indicated in dengue infants with severe bleeding (see Section
2.2.3.3). Hyponatraemia and metabolic acidosis can occur in severe cases. Electrolyte
levels and blood gases should be determined periodically in severe cases. Early volume
replacement will usually correct the metabolic acidosis and generally result in a favourable
outcome. Sodium bicarbonate may be considered in severe metabolic acidosis.
Although dengue in infants comprises around 5% of all paediatric cases, mortality rates are
higher in infants than in older children (3, 6, 13). Fluid replacement in infants with severe
dengue is a challenge to good clinical management. This involves following established
procedures for use of colloid solutions and blood transfusions. To further reduce case
fatalities, special emphasis needs to be given to dengue infants who have severe
complications or who go on to develop them.
Vertical transmission and neonatal dengue
Pregnant women with dengue virus infection can transmit the virus to their foetus and
vertical dengue transmission has been described. Questions about dengue in pregnant
women relate to the effect of the pregnancy on the disease process; the effect of the
disease on the pregnancy; possible effects on the foetus and the neonate; and how the
pregnant woman and the newborn might best be managed (14). Answers to questions
about the effect of the disease on the pregnancy are discussed in Section 2.4.5 of this
65
handbook. Here follows a review of the effects of dengue on the neonate, clinical
manifestations and management of neonatal dengue.
Dengue virus can be vertically transmitted to the foetus in utero or to the infant at
parturition. Results of a recent systematic review indicate 16 cases of vertical transmission
among 25 (64.0%) neonates reported from 12 case reports, and vertical transmission
occurring in 4 of the 10 case series, in 18 of 143 (12.6%) of cases (15). One comparative
study that tested 64 umbilical cord serum samples for dengue IgM from 63 women who
were found to be IgM positive at the time of delivery, found a vertical transmission rate of
1.6% (16).
In the vertical transmission cases, some newborns may be asymptomatic (16). Clinical
manifestations of vertically infected neonates vary from mild illness such as fever with
petechial rash, thrombocytopenia and hepatomegaly, to severe illness with clinical sepsis,
pleural effusion, gastric bleeding, circulatory failure, massive intracerebral haemorrhage
and death (17−24). Clinical presentation in the newborn infant does not appear to be
associated with maternal disease severity or dengue immune status, or mode of delivery
(19, 24, 25). However, timing of maternal infection may be important; peripartum maternal
infection may increase the likelihood of symptomatic disease in the newborn. A review of
17 mother–infant pairs with dengue infection found that the time intervals between the
mothers’ onset of fever and that of their neonates, were 5–13 days (median, 7 days); fever
in neonates occurred at 1–11 days of life (median, 4 days), and the duration of fever in
neonates was 1–5 days (median, 3 days) (24).
Passive transfer of maternal dengue antibodies to the foetus influences the occurrence of a
severe development of the disease (9). Antibodies to the dengue virus in the dengue-
infected mother can cross the placenta and can cause severe dengue in newborn infants
(1, 26).
Management of neonatal dengue
When a pregnant or parturient woman develops signs consistent with dengue, the
diagnosis of dengue should be considered in her neonate even if the neonate appears well
in the first several days of life. Remember that some neonates have become ill as long as
11 days after birth (24). The diagnosis of neonatal dengue could eventually be suspected
on clinical grounds and then confirmed in the laboratory, but initial presentation may be
confused with bacterial sepsis, birth trauma and other causes of neonatal illness.
Symptomatic and supportive treatment under close observation is the mainstay of
treatment (14, 20, and 24).
66
2.4.8 References
1. Halstead SB et al., Dengue hemorrhagic fever in infants: research opportunities ignored. Emerging
Infectious Diseases journal, 2002, 8:1474–1479.
2. Pancharoen C, Thisyakorn U. Dengue virus infection during infancy. Transactions of the Royal Society of
Tropical Medicine and Hygiene, 2001, 95:307–308.
3. Kalayanarooj S, Nimmannitya S. Clinical presentations of dengue hemorrhagic fever in infants compared to
children. Journal of the Medical Association of Thailand, 2003, 86(Suppl 3):S673–S680.
4. Hammond SN et al., Differences in dengue severity in infants, children, and adults in a 3-year hospital-
based study in Nicaragua. American Journal of Tropical Medicine and Hygiene, 2005, 73:1063–1070.
5. Capeding RZ et al., The incidence, characteristics, and presentation of dengue virus infections during
infancy. American Journal of Tropical Medicine and Hygiene, 2010, 82(2):330–336.
6. Hung NT et al., Dengue hemorrhagic fever in infants: a study of clinical and cytokine profiles. Journal of
infectious Diseases, 2004, 189:221–232.
7. Witayathawornwong P. Dengue hemorrhagic fever in infants, late infants and older children: a comparative
study. Southeast Asian Journal of Tropical Medicine and Public Health, 2005, 36:896–900.
8. Kabilan L et al., Dengue disease spectrum among infants in the 2001 dengue epidemic in Chennai, Tamil
Nadu, India. Journal of Clinical Microbiology, 2003, 41:3919–3921.
9. Martínez E et al., Dengue fever and hemorrhagic dengue in infants with a primary infection. Revista cubana
de medicina tropical. 1993, 45(2):97–101.
10. Simpkin PS, Hinchliffe RF. Reference values. In: Arceci RJ, Hann IM, Smith OP, eds. Pediatric Hematology,
3rd ed. Oxford, Blackwell Publishing Ltd, 2006: 792–995.
11. Hung NT et al., Volume replacement in infants with dengue hemorrhagic fever/dengue shock syndrome.
American Journal of Tropical Medicine and Hygiene, 2006, 74:684–691.
12. Wills BA. Management of dengue. In: Halstead SB, ed. Dengue. London, Imperial College Press, 2008:
193–217.
13. Hadinegoro SR, Purwanto SH, Chatab F. Dengue shock syndrome: clinical manifestations, management
and outcome – a hospital-based study in Jakarta, Indonesia. Dengue Bulletin, 1999, 23:105–106.
14. Carroll D, Toovey S, Gompel AV. Dengue fever and pregnancy - a review and comment. Travel Medicine
and Infectious Disease, 2007, 5:183–188.
15. Pouliot SH et al., Maternal dengue and pregnancy outcomes: a systematic review. Obstetrical &
gynecological survey. 2010, 65(2):107–118.
16. Tan P et al. Dengue infection in pregnancy. Obstetrical & Gynecological Survey, 2008, 111:1111–1117.
17. Chotigeat U, Kalayanarooj S, Nisalak A. Vertical transmission of dengue infection in Thai infants: two case
reports. Journal of the Medical Association of Thailand, 2003, 86 (suppl 3):S628–S632.
18. Ahmed S. Vertical transmission of dengue: first case report from Bangladesh. Southeast Asian Journal of
Tropical Medicine Public Health, 2003, 34:800–803.
19. Chye JK et al. Vertical transmission of dengue. Clinical Infectious Diseases, 1997, 25:1374–1377.
20. Janjindamai W, Pruekprasert P. Perinatal dengue infection: a case report and review of literature. Southeast
Asian Journal of Tropical Medicine and Public Health, 2003, 4:793–796.
21. Kerdpanich A et al. Case report: perinatal dengue infection. Southeast Asian Journal of Tropical Medicine
and Public Health, 2001, 32:488–493.
22. Maroun S et al. Case report: vertical dengue infection. Journal of Pediatric Gastroenterology and Nutrition,
2008, 84:556–559.
67
23. Petdachai W et al., Neonatal dengue infection: report of dengue fever in a 1-day-old infant. Southeast Asian
Journal of Tropical Medicine and Public Health, 2004, 35:403–407.
24. Sirinavin S et al., Vertical dengue infection: case reports and review. Pediatric Infectious Disease Journal,
2004, 23:1042–1047.
25. Thaithumyanoa P et al., Dengue infection complicated by severe hemorrhage and vertical transmission in a
parturient woman. Clinical Infectious Diseases, 1994, 18:248–249.
26. Kliks SC et al., Evidence that maternal dengue antibodies are important in the development of dengue
hemorrhagic fever in infants. American Journal of Tropical Medicine and Hygiene, 1988, 38:411–419.
68
3. Pitfalls in the management of dengue and solutions
The clinical management of dengue is more fraught than that of most other infectious
tropical diseases. The uninitiated physician who has managed uncomplicated dengue
cases may be lulled into believing that dengue is a “mild disease of thrombocytopenia” that
requires no more than intravenous fluid therapy and platelet transfusions for a couple of
days. When faced with severe dengue, these same physicians may be unprepared for the
changing clinical, biochemical and haematological profiles that accelerate after the first few
days of fever and therefore will not step up their vigilance during the critical period.
In this chapter we identify some of the common pitfalls in dengue management and include
some explanations for the misconceptions together with suggested solutions.
69
The earlier the FBC is done, the more likely it is that the result will seem normal. An
uninitiated physician may misinterpret this and rule out dengue.
Solution: The purpose of an early FBC is to establish the patient’s baseline haematocrit.
The baseline haematocrit will become useful to guide clinical management when
haematocrit levels increase during the critical phase. The early and progressive decrease
in white cell and platelet counts during the febrile period is a useful indicator of dengue.
Pitfall 4: Surge of hospital admissions during a dengue outbreak
During a dengue outbreak there may be a surge in hospital admissions because of fear
and panic among patients and frontline physicians, or institutional directives to admit all
suspected dengue patients.
Explanation: In a small group of dengue patients the disease could progress very rapidly
during the critical phase resulting in severe shock and death. However, admission of all
suspected dengue cases will overwhelm the hospital system with many uncomplicated
dengue cases that could be taken care of in the ambulatory system.
Solution: Clear guidelines for admissions need to be available. Integrating an out-patient
management protocol within the hospital system can ensure effective gate-keeping in a
dengue outbreak (1, 2). Please refer to Section 2.2.1 and Integrated Management of
Childhood Illness (IMCI) guidelines (3, 4, 5).
Admission during the febrile period should be for those who are unable to manage
adequate oral hydration at home, infants and those with co-morbid conditions. Refer to
Textbox C for warning signs.
The chart presented in Table 10 is to assist frontline physicians in the early recognition and
management of dengue patients, and is in accordance with IMCI guidelines.
70
Table 10. Chart for case management of dengue for frontline physicians*
Assessment Classification Management
*for a patient from a dengue endemic locality, with fever, or history of fever, which lasts 2–7 days, and fulfils
criteria for probable dengue (refer to Figure 2: Dengue classification)
The frontline physicians (health-centre staff, private doctors and primary health-care
workers), should use the chart (Table 10), to assess, identify, classify and manage patients
with fever or history of fever which lasts 2–7 days. The chart can also be used by staff at
the hospital outpatient clinic with some modifications, e.g. direct hospitalization of patients
with severe dengue or dengue with warning signs.
How to use the chart:
1. The patient comes from a dengue endemic locality and fulfils criteria for probable
dengue. Start with the top row. Assess the patient to discern whether they have signs of
shock (cold, clammy extremities, prolonged capillary refill time and weak pulse), or signs of
severe bleeding or impaired consciousness.
1.1. If the answer is YES, classify the case as severe dengue and refer the patient
URGENTLY to hospital.
72
3.2 At the emergency department
Pitfall 6: Triage based on fever
The afebrile dengue patient gets a non-urgent triage and has to wait for medical attention
or be sent home.
Explanation: A subnormal temperature is seen in patients with established plasma
leakage or dengue shock. At this stage the clinical outcome is extremely time-sensitive; any
delay in fluid resuscitation will lead to profound, prolonged shock and a stormy course for
the disease.
Solution: Triage personnel should be taught assessment of perfusion (refer to Pitfall 8),
especially when the BP is measured by a digital machine (see Textbox D). Additionally, a
quick targeted history for nausea, vomiting and lethargy will indicate the severity of illness.
Pitfall 7: Waiting for full blood count results (for thrombocytopenia)
Waiting for FBC results to make decisions about admission and intravenous fluid therapy
will cause further delay in recognition and treatment of dengue shock.
Solution:
Intravenous fluid therapy should be initiated when warning signs or clinical evidence of
shock are present. However, it is useful to sample blood for FBC and haematocrit before
commencement of intravenous fluid therapy.
3.3 Shock and its many manifestations
Pitfall 8: Vital signs are “stable”
Explanation: This phrase means that the patient is alert and has BP within the normal
range. These two conditions are present in compensated shock. With respect to the “ABC”
of resuscitation, patients with airway and breathing problems are “visible” – in other words
they are sitting up and struggling to breathe noisily. The patient who is in shock is initially
quietly tachypnoaeic; alert but lying down. This is more difficult to recognize. Unless you
touch the patient, you will not be able to detect the cold extremities, feeble peripheral
pulses and prolonged capillary refill time that are the earliest changes in shock. The patient
in shock remains in this quiet alert state until cerebral perfusion diminishes, and then
develops a “sudden” shortness of breath and restlessness or seizures followed quickly by a
“sudden” collapse.
Solution: Make it a habit to touch and assess the peripheral perfusion of every patient, not
only dengue patients. It takes less than 30 seconds to assess the four parameters that
inform you whether the patient’s life may be in danger: pulse volume, capillary refill time,
colour and temperature of the extremities. Only trained humans can perform this vital
function, no machine can do it.
Pitfall 9: Symptomatic treatment with antispasmodic agents or antacid is being given
to patients with abdominal pain
Abdominal pain is an early sign of plasma leakage and becomes more severe as
hypovolaemia progresses. The pain could be in the epigastrium and, together with
vomiting, may be misinterpreted as gastritis. The patient will progress into full-blown shock
if definitive treatment for shock is not given promptly.
73
Solution: A history of transitioning from fever to no-fever and the development of
gastrointestinal symptoms such as nausea, vomiting or diarrhoea, should suggest dengue
with warning signs or dengue shock. Patients with these symptoms should be carefully
assessed for signs of dehydration and hypovolaemic shock.
Pitfall 10: Patients who present with severe abdominal pain are referred to the
surgical team or for ultrasound studies.
Explanation: Plasma leakage is associated with abdominal pain that may occur anywhere
in the abdomen. Guarding and rebound tenderness in the right iliac fossa have been
described. This may be due to serosal oedema and lymphoid hyperplasia (6) of the
intestines. The abdominal pain becomes more severe in the shock state due to splanchnic
vasoconstriction. Referrals or ultrasound studies further delay initiating fluid resuscitation
and may compound the morbidity and clinical outcome.
Solution: Evaluate the patient’s history for fever that preceded the onset of abdominal
pain. Assess the patient for signs of shock. Haemoconcentration and thrombocytopenia
should raise the suspicion of dengue diagnosis. A bolus of intravenous fluid (5−10 ml/kg
over 1 hour) may lead to resolution of abdominal pain, making the diagnosis of a surgical
abdomen unlikely.
Pitfall 11: Shortness of breath misinterpreted as fluid overload or due to pleural
effusion or pneumonia.
As plasma leakage progresses undetected and untreated, the patient develops metabolic
acidosis (lactic acidosis) that initially appears as quiet tachypnoea. This represents the
respiratory compensation to maintain the acid-base balance. With further deterioration to
the shock state, the breathing becomes deeper and gives the impression of shortness of
breath. (See Section 2.3.1, Acute respiratory distress and failure).
Solution: It is essential to be familiar with this breathing pattern, which is seen in dengue
shock as well as diabetic ketoacidosis. Do not be distracted by the patient’s dramatic
breathing pattern. Look for features of shock: if unsure, a blood gas (+ lactate) analysis
should be done. Start intravenous fluid resuscitation immediately after sampling blood for
haematocrit determination.
Pitfall 12: Seizures or altered sensorium – diagnosed as meningitis or
meningoencephalitis.
Seizures could occur during the febrile phase and during the critical phase. The first group
will usually affect young children who present as febrile seizures during the viraemic febrile
phase of dengue. At this time the full blood count may suggest a viraemic illness. The child
should be treated as for febrile seizures with vigilance for warning signs and plasma
leakage during the transition from the febrile to the afebrile phase.
The second group of patients present during the critical phase with no fever, or low-grade
fever, but who are in severe shock. The brain is one of the last organs compromised by
perfusion in hypovolaemic shock. When seizures or loss of consciousness occur during the
critical phase, it is associated with prolonged and profound shock that causes acute
cerebral ischaemia. At this stage the clinical outcome is extremely time-dependent; any
delay in fluid resuscitation such as a detour to the computer tomography (CT) scan will lead
to cardiorespiratory collapse and irreversible shock.
74
Solution: A targeted history is essential to confirm that the patient had fever preceding the
seizures and might not be febrile at the time of seizures. The evidence of shock, such as
narrowed pulse pressure and tachycardia, should alert the physician that urgent fluid
resuscitation should be initiated, regardless of diagnosis. A CT scan is not necessary if the
patient’s neurological state recovers together with a stable haemodynamic state.
Pitfall 13: Bleeding versus plasma leakage
The physician who is overly concerned about the risks of bleeding, focuses on monitoring
the trend of platelet counts alone to the exclusion of the bigger picture of plasma leakage.
These physicians may administer platelet transfusions in addition to the maintenance of
intravenous fluid. Fluid overload or inadequate fluid replacement may ensue, depending on
the severity of the plasma leakage and the phase of disease at the time of platelet
transfusion. Inadequate fluid replacement leads to prolonged and profound shock
characterized by bleeding, the very phenomenon that these physicians are trying to
prevent. It is important to note that the platelet count may remain low/very low during the
first one or two days of recovery and that platelet transfusion during this time will cause
fluid overload (7, 8).
Solution: The physician should be alert to the onset of this unique syndrome of dengue
vasculopathy that causes plasma leakage and hypovolaemic shock when the fever
subsides. Monitoring should be focused on manifestations of plasma leakage, bleeding and
shock. The trend of haematocrit level together with haemodynamic status is the main guide
to fluid management.
The usefulness of trending of the platelet count during the febrile phase is to identify its
rapid decrease, which marks the beginning of plasma leakage. Beyond this phase, trending
of platelet counts has no bearing on fluid management.
Pitfall 14: Pulse rate and systolic and diastolic pressures
Physicians frequently focus on the systolic BP and ignore the rising diastolic pressure and
pulse rate (PR). (See Section 1.1.4, Severe dengue).
Solution: Patients with compensated shock may be critically ill despite being alert and
having a “normal” systolic BP. Furthermore, automated BP devices can provide normal BP
readings (i.e. falsely high BP results) in severe shock and hypotension. In general, if you
cannot palpate the radial or brachial pulse and the central (femoral) pulses are weak or
absent, assume that the patient is hypotensive. Do not rely on automated BP devices in
patients with clinical signs of shock.
3.4 Parenteral fluid therapy
Pitfall 15: During the febrile stage the tachycardia and cold extremities could be
misinterpreted as shock and fluid resuscitation commenced.
Vasoconstriction resulting in cold extremities is part of the body’s normal mechanism to
conserve heat to increase core body temperature.
Solution: The assessment of the intravascular volume of a patient should not be based on
one or two parameters only. During the early febrile phase in addition to the high body
temperature, the haematocrit and platelet count, pulse volume and urine output will be
normal. Some patients may continue to have fever during the plasma leakage phase. In
these patients, there may be tachycardia and cool peripheries despite a good pulse
75
volume. Other parameters of intravascular volume should be sought, e.g. capillary refill
time, pulse pressure, urine output and/or acid-base balance.
Pitfall 16: Starting intravenous fluids in the febrile phase (because NS1 Ag is
positive)
This practice should not be routine. Studies have shown that plasma leakage is detectable
as early as day 2 of fever onset (9). Early intravenous fluid therapy will exacerbate
accumulation of third space losses before the critical phase when plasma leakage
increases significantly.
Solution: During the febrile phase hydration should be maintained by the oral route (see
Section 2.2.1). Only in exceptional cases should intravenous infusion be used and then it
should be kept at a minimal infusion rate and be discontinued as soon as the patient can
take oral fluids.
Pitfall 17: Administering intravenous fluid at high rates for several hours without
review of the clinical state.
The management of dengue shock is centred on intravenous fluid management. In dengue,
unlike acute gastroenteritis, there is no loss of fluid from the body. Instead, fluid is merely
lost from the intravascular to the extravascular compartments through leaky capillary
endothelium. This state will continue for 24−48 hours during the critical phase of illness.
Solution: Thus it is important to consider that in a “leaky” capillary state, beyond a certain
hydrostatic pressure, the more fluids are infused, the more they will leak into the “third
space”; also that all fluids administered will be reabsorbed and fluid overload may result.
Massive pleural effusion and ascites are the consequences of unrestrained intravenous
fluid administration that leads to respiratory distress 12−48 hours later. Thus, it is essential
to administer the minimum intravenous fluid necessary to maintain a “just-enough”
circulation. Placing a micro- haematocrit centrifuge on the ward for use by trained staff
facilitates the use of haematocrit determinations to manage fluid requirements. Titration of
intravenous fluid as in the algorithm is recommended to avoid such situations (see Section
2.2.3.1, Treatment of shock).
Pitfall 18: Increased fluid therapy in a well perfused patient with persistently elevated
haematocrit.
The haematocrit may remain elevated towards the end of the critical phase (24−48 hours of
plasma leakage) despite adequate fluid therapy. A patient who remains well perfused i.e.
with warm extremities, stable BP, PR and pulse volume as well as urine output, is
compensating well for the ongoing plasma leakage.
An attempt to correct the haematocrit level by increasing intravenous fluid therapy in this
scenario will often lead to fluid overload.
Solution: Continue close monitoring of this patient. Do not increase the intravenous fluid
therapy; if the patient continues to be well perfused in the next 2–3 hours then intravenous
fluids should be gradually reduced and even stopped, regardless of the haematocrit level.
The haematocrit will gradually reduce even when intravenous fluid is discontinued
signifying the re-absorption phase. This is illustrated in case study 2 (see Section 4).
76
Pitfall 19: Intravenous fluid therapy is continued into the recovery phase because the
patient is not drinking.
A careful examination of the patient and the fluid intake and output chart should convince
the physician that the patient has a big, positive balance of fluid accumulated in the third
spaces over the past 48 hours of intravenous therapy. The only way for the pleural effusion
and ascites to be reabsorbed into the circulation is to stop intravenous therapy. The
patient’s own thirst mechanism will kick in after the diuresis.
3.5 Urine output
Pitfall 20: During the febrile phase the urine output is often ignored and not factored
into the evaluation of adequate oral fluids.
The urine output is a useful marker of the hydration state during the febrile phase.
Solution: Adequate oral intake of fluids should result in a urinary output frequency of 4−6
times per day.
Pitfall 21: During the critical phase the urine output of 1 ml/kg/hour is used as a
criterion of adequate intravenous therapy or might not be monitored at all.
During the critical phase the urine output is a useful indicator of the intravascular volume
state but should not be used as the sole criterion.
Solution: In severe shock, urine output should be monitored hourly using an indwelling
catheter. The expectation of what constitutes an adequate urine output should be scaled
down to ~ 0.5 ml/kg/hour to avoid fluid overload. On the other hand, a urine output
exceeding that may be an indication to reduce intravenous fluid therapy.
Pitfall 22: Use of furosemide for oliguria during the critical phase.
Oliguria in this phase is due to “pre-renal”, i.e. inadequate tissue-organ perfusion. The use
of loop diuretics such as furosemide will produce some urine but will exacerbate the shock
state.
Solution: Adequate peripheral perfusion should be the main end-point of intravenous fluid
therapy. Patients with oliguria should continue to receive intravenous fluid therapy until
peripheral perfusion is established.
Pitfall 23: The practice of a fluid bolus challenge followed by furosemide may
produce urine; the physician may think they have averted an oliguric renal failure.
This practice may be detrimental in dengue with plasma leakage. During the infusion of a
large fluid bolus, fluid will inevitably be redistributed to the “third space”. However with
furosemide administration fluid can only be drawn out of intravascular compartment. The
net effect is that the “third space” gains fluids while the intravascular compartment remains
depleted and the patient loses electrolytes such as potassium.
Solution: Urine will be produced by the kidneys when there is adequate renal perfusion.
However, if established acute kidney injury sets in, as manifested in anuria and high serum
creatinine levels, volume infusion should continue until the circulation is established. The
physician should recognize that urine output as a parameter of adequate perfusion is very
limited in this situation.
77
Pitfall 24: Worsening or persistent abdominal pain/tenderness appearing after 24
hours of large volumes of fluid replacement, misinterpreted as “warning sign for
shock”.
After more than 24 hours’ administration of large volumes of fluid, tense ascites would have
developed in patients with severe plasma leakage. These patients often complain of
abdominal discomfort or pain. This may be due to tense ascites and/or enlarged and
congested liver from the fluid overload.
Solution: This symptom and sign should be interpreted cautiously. The intravascular
volume should be thoroughly assessed based on peripheral perfusion, pulse pressure and
volume, urine output (~0.5 ml/kg/hour) and acid-base balance; not just by abdominal
symptoms and signs alone. Intravenous fluid should be changed to colloids or be reduced
and discontinued. Further fluid replacement will not only increase the abdominal symptoms
but also aggravate the respiratory distress.
Pitfall 25: Abdominal compartment syndrome.
Abdominal compartment syndrome is the result of massive ascites accumulated in a short
time, and hence is under tension. It may be precipitated by using moderately high PEEP
during mechanical ventilation of dengue patients. Pressure on the retroperitoneal structures
such as the kidneys, renal veins and inferior vena cava, will cause reduced renal perfusion
with consequent oliguria even though peripheral perfusion may have been adequate.
Solution: The physician should realize this and be cautious about using urine output as a
parameter of adequate perfusion. (See Textbox D). If it is possible, the patient may be put
in a reclining or lateral position to relieve pressure on the retroperitoneal structures.
However with time and especially when intravenous fluid therapy is discontinued at the end
of the critical period, the intra-abdominal tension will ease and urine may start flowing
gradually, spontaneously or with the help of an infusion of furosemide. The use of
furosemide should be commenced in the recovery period when the patient is stable without
intravenous fluid infusion. Premature use of furosemide may precipitate a shock state (10).
Pitfall 26: Use of renal replacement therapy such as continuous venous
haemodialysis during the critical phase when the patient has severe metabolic
acidosis and oliguria.
There is a misconception that lactic acidosis can be “washed out” by using lactate-free
solutions in dialysis. Severe metabolic acidosis and oliguria are the consequences of
uncorrected and prolonged shock. The haemodialysis machine will face technical problems
with drawing blood from the hypovolaemic patient. Invariably your attention will be diverted
to sorting out the machine rather than the patient.
Solution: Renal replacement should be reserved for patients who are already in the
recovery phase, who are haemodynamically stable, are fluid overloaded and have oliguria
not responding to furosemide therapy. In this situation the haemodialysis machine works
very well and is effective in preventing life threatening pulmonary oedema.
Pitfall 27: Large volumes of dilute urine in a patient with unstable haemodynamics
could mislead a physician into judging that the patient is stable.
Not correcting the hyperglycaemic state would exacerbate the shock state. (See Section
2.2.3.4.)
78
3.6 Blood transfusions for severe bleeding
Pitfall 28: Not transfusing blood until the haematocrit has decreased to low levels in
unstable patients.
Severe but occult bleeding can be difficult to recognize (see Section 2.2.3.3, Treatment of
haemorrhagic complications).
The haematocrit will only decrease to low levels after several boluses of fluid resuscitation
(especially with colloids), or if the bleeding has been very severe. Throughout the period of
fluid resuscitation the patient remains unstable while the “third space” fluid accumulation
increases.
Solution: It is critical to recognize that the lower the haematocrit in dengue shock (8, 9),
the more likely the significant bleeding, and therefore it is vital to change strategy from
crystalloid/colloid infusion to blood transfusion. This change in strategy will prevent
excessive infusion of non-blood fluids and accumulation of “third space” fluid. Bleeding
should always be considered as a possible cause of prolonged/profound shock. Some
authors have stated that “haematocrit is not useful in bleeding patients” (10). On the
contrary, haematocrit is very useful in recognizing severe bleeding, provided that it is
interpreted together with other clinical parameters. Please refer to interpretation of
haematocrit Section 2.2.3.1.
Pitfall 29: DIVC regime for severe bleeding.
The causes of severe bleeding in dengue are not well understood. These patients have
thrombocytopenia, deranged coagulation and disseminated intravascular coagulopathy. It
is inevitable that these laboratory parameters become the targets of therapy with multiple
transfusions of fresh frozen plasma, platelet concentrates and cryoprecipitate.
Unfortunately, the clinical and laboratory responses are poor (7 ); meanwhile bleeding and
plasma leakage continue unabated. Unless transfusion with fresh whole blood is given
urgently, the patient will succumb to haemorrhagic shock with massive third space
accumulation.
Solution: See Section 2.2.3.3. It may not be possible to arrest the bleeding or correct the
thrombocytopenia or coagulopathy during the critical phase, but it is essential to keep
abreast of blood loss with repeated blood transfusions if necessary. This should be guided
by clinical assessment, haematocrit levels, blood gas and lactate analysis. Bleeding will
usually decrease once the critical phase is over, as manifested by decreasing tachycardia
and improved peripheral perfusion.
It is important to request fresh whole blood, otherwise nearly expired blood may be issued
by the blood bank. Stored blood does not have the same properties as fresh whole blood
and may not reverse the clinical situation.
Pitfall 30: Blood transfusion when the haematocrit is on the rise in a shocked patient.
Fluid boluses that are administered aggressively in unwarranted situations, such as in
compensated shock (refer to Figure 6), may give the impression of refractory shock and
prompt the physician into changing treatment from crystalloid fluid to blood transfusion. The
subsequent haematocrit will rise even more sharply after the blood transfusion and prompt
79
the physician to give more crystalloids even though the haemodynamic situation is stable.
The patient will become fluid overloaded with such an approach.
Solution: If haematocrit continues to rise despite fluid resuscitation, the physician should
consider using a colloid bolus at 10 to 20 ml/kg. If the shock patient has a normal systolic
BP, there is no need to rush the fluid resuscitation. Aggressive fluid resuscitation over 15
minutes should be reserved for those with hypotension (see Figure 7).
Pitfall 31: Blood transfusion in a stable patient with low haematocrit.
This usually occurs during the recovery period when re-absorption causes haemodilution. A
blood transfusion at this stage may precipitate life-threatening pulmonary oedema.
Solution: If the patient has a stable circulation they do not need blood transfusion. The
haematocrit will increase after the excess fluid has been diuresed.
Pitfall 32: Severe shock and inotropes.
Physicians who have administered 40−60 ml/kg of fluid and have failed to reverse the
shock state will usually start inotropes such as dobutamine or dopamine and
norepinephrine to support the BP (10).
Refractory shock is the most common cause of death within 24−48 hours of admission.
There is usually a combination of plasma leakage and bleeding following prolonged
uncorrected shock and this is accompanied by renal and liver impairment.
Solution: After every bolus infusion of 10−20 ml/kg, there should be a re-evaluation of the
haemodynamic state. If there is no clinical improvement, a repeat haematocrit should be
done. (see Figure 7). A rising haematocrit or an haematocrit that is still high, would suggest
ongoing plasma leakage. This should prompt a change to colloid solution. A decrease in
haematocrit suggests significant bleeding and the physician should proceed to transfuse
matched fresh whole blood as soon as possible (see Section 2.2.3.3). Colloids and
inotropes should be used as a temporary measure.
Pitfall 33: Liver enzymes and multi-organ involvement may prompt multi-specialty
referrals that result in several “single organ” treatments.
There are many parameters that can be deranged in severe dengue – raised liver
transaminases, muscle enzymes, e.g., creatine kinase, high blood creatinine levels,
hyperglycaemia, thrombocytopenia, coagulopathy – all serving to distract the physician’s
attention. The multi-organ involvement is mainly due to shock with tissue hypoperfusion
and diminished oxygen delivery.
Solution: Specialty consultations must not distract the physician from prioritizing
achievement of haemodynamic stability and improved tissue oxygenation by fluid, colloid,
or blood administration until the patient is out of the critical period. All other outstanding
clinical issues (except hypoglycaemia and electrolyte imbalances) should await
achievement of haemodynamic stability.
Pitfall 34: Last but not least, dengue shock treated “aggressively” as in septic shock
The main haemodynamic elements of septic shock are maldistribution of blood volume
resulting from an increased vascular capacitance and myocardial suppression, while
dengue shock is hypovolaemia with decreased vascular capacitance resulting from plasma
leakage. The increased diastolic pressure in compensated shock preserves myocardial
80
perfusion. The clinical picture is dissimilar; septic shock is characterized by warm flushed
extremities, decreased diastolic pressure and increased pulse pressure; the latter on the
other hand is by cold vasoconstricted extremities, increased diastolic pressure with
decreased pulse pressure (see Textbox D). Thus, the strategy of aggressive fluid
resuscitation of septic shock is not applicable to severe dengue with plasma leakage.
Aggressive fluid resuscitation may indeed be harmful and should be limited to dengue
shock with hypotension. Once the peripheral pulses are established the rate of parenteral
fluid replacement should be reduced, as per frequent and simultaneous haemodynamic
and haematocrit assessment.
In summary, training in the clinical management of dengue is essential if the clinician is to
be able to navigate the patient through the three phases of the illness. Training is needed,
first, to understand the disease course and second, to be alert to the physiological
problems.
Intravenous fluid therapy is life-saving in dengue shock. However, there is a “narrow
therapeutic index”. In other words, fluids have to be given timely, at the appropriate volume,
rate, of the appropriate type (crystalloids, colloid and/or blood) and for the appropriate
duration. Therein lies the challenge to physicians who are not familiar with the important
practice of fluid titration through frequent and meticulous assessment. Progression of the
disease through the critical phase should be tracked in hours of plasma leakage.
Recognizing the cues to discontinue intravenous fluid therapy is just as important as
knowing when to start it. Given time and haemodynamic stability, other issues such as
thrombocytopenia, coagulopathy and raised liver enzymes will recover spontaneously or
with supportive care.
81
3.7 References
1. Integrated management of childhood illnesses. Geneva, World Health Organization, 2008.
2. Dengue, dengue haemorrhagic fever and dengue shock syndrome in the context of the integrated
management of childhood Illness. Geneva, World Health Organization, 2005. WHO/FCH/CAH/05.13.
2005.
3. Phuong CXT, The Dong Nai/WHO Study Team. Evaluation of an algorithm for integrated management of
childhood illness in an area of Vietnam with dengue transmission. Tropical Medicine & International
Health, 2004, 9:573−581.
4. Anh NN, Tram TT. Integration of primary health care concepts in a children’s hospital with limited
resources. Lancet, 1995, 346:421−424.
5. Ingram PR, Mahadevan M, Fisher DA. Dengue management: practical and safe hospital-based
outpatient care. Transactions of the Royal Society of Tropical Medicine and Hygiene, 2009,
103(2):203−205.
6. Premaratna R et al., Dengue fever mimicking acute appendicitis. Transactions of the Royal Society of
Tropical Medicine and Hygiene, 2007, 101:683−685.
7. Lum LC et al., Preventive transfusion in dengue shock syndrome: Is it necessary? Journal of Pediatrics,
2003, 143:682-684.
8. Lye D et al., The benign nature of acute dengue infection in hospitalized older adults in Singapore.
International Journal of Infectious Diseases, 2010, 14(5): e410−413.
9. Srikiatkhachorn A et al., Dengue hemorrhagic fever: the sensitivity and specificity of the world health
organization definition for identification of severe cases of dengue in Thailand, 1994-2005. Clinical
Infectious Diseases, 2010, 50(8): 1135−1143.
10. Ranjit S, Kissoon N, Jayakumar I. Aggressive management of dengue shock syndrome may decrease
mortality rate: a suggested protocol. Pediatric Critical Care Medicine, 2005, 6(4):412−419.
82
4. Case studies
23:00 He was admitted to the ICU after the CT scan of the brain which showed no
intracranial bleed or oedema. He had spontaneous breathing but was deeply sedated. The
lung fields were noted to be clear. His BP was 85/50 mmHg, PR 116/min, RR 27/min and
83
SpO2 100%. The arterial blood pH was 6.97, bicarbonate 7.3 mmol/L, base excess -23.7
and lactate 18.50 mmol/L. Refer to Table 11 for other investigations. A central venous
catheter was inserted and a pressure of 6 cm H2O was obtained. He was given 5 ml/kg of
colloid over 1 hour, followed by 3 ml/kg/hour of crystalloids. Dopamine was administered at
5 µg/kg/min.
16 May
03:00 The patient was unresponsive except to pain. By then he had received 500 ml of
fluid and a second 500 ml was in progress. His BP was 88/46 mmHg, PR 110/min. His
urine output was minimal. Refer to Table 12 for serial blood results.
07:00 The patient’s abdomen was distended, soft and tender. His BP was 87/52 mmHg
with PR 105/min and SpO2 100%. Dopamine was administered at 10 µg/kg/min,
Investigations carried out at 03:00 were reviewed (Table 12). One unit of DIVC regime was
ordered − 4 units of cryoprecipitate, 4 units platelet concentrates, 4 units FFP (fresh frozen
plasma). Dopamine was increased to 20 µg/kg/min and noradrenaline was added. The
patient was referred to the nephrology and surgical team.
08:00 A nephrology assessment showed the labile BPs, systolic pressure 70−80 mmHg,
diastolic pressure 50−60 mmHg, double inotropes, CVP 12−15 cm H2O, heart rate (HR)
106/min, SpO2 100% on ventilation, RR 18/min; auscultation showed bi-basal crepitations
in the lungs, his abdomen was distended but soft and non tender. Bowel sounds were
present. The total fluid balance by this time was intake/output: 3207/95 ml.
The problems identified were:
1. Septic shock with possible perforation of viscus or ischaemic gut.
2. Coagulopathy with thrombocytopenia – to transfuse one cycle DIVC regime.
3. Acute renal failure with severe metabolic acidosis and hyperkalaemia for renal
replacement therapy.
4. Meningoencephalitis.
Noradrenaline was increased and titrated to maintain the MAP (mean arterial
pressure)> 65 mmHg. A second cycle of DIVC regime transfusion was initiated.
10:00 Surgical review noted unstable BPs, tachycardia and distended but soft abdomen
with no guarding. Per-rectal examination showed fresh blood. The nasogastric tube did not
yield any blood.
10:00 to 18:00 The patient’s condition continued to deteriorate despite increasing inotropes
infusions and transfusions of FFP, platelets and cryoprecipitates. Sodium bicarbonate
infusions were started for metabolic acidosis, calcium gluconate infusions for
hyperkalaemia. The CVP continued to increase from 12 cm to 20 cm H2O. His BP was
120/80 mmHg and his SpO2 100%.
Ultrasound study of abdomen showed that the liver was mildly enlarged with normal liver
echogenicity. No focal liver lesions were noted. The biliary tree was not dilated. The gall
bladder was distended with thickened walls and no calculus within. There was
pericholecystic fluid. There were no focal lesions within the normal-sized spleen. The
pancreas was obscured by bowel gas. Both kidneys were normal in echogenicity and size
84
with no focal lesion. The urinary bladder was empty. Generalized ascites and a right pleural
effusion were noted.
18:00 Investigations done at 15:00 (Table 12) were reviewed. He was started on
continuous renal replacement therapy and one unit of packed cells was transfused together
with a third cycle of DIVC regime.
Despite all these treatments, the general condition continued to deteriorate and blood
lactate continued to increase as the BP remained low. The patient died on 17 May at 14:00.
The results of laboratory diagnostic tests were as follows:
• dengue serology: positive for IgG and IgM;
• leptospira serology: negative;
• blood culture and sensitivity: no growth
Post-mortem examination:
Head & neck:
The brain was normal, weighing 1240 g. No oedema was noted.
Thorax:
Both lungs were congested and there were bilateral pleural effusions. There was no
pericardial effusion.
Abdomen:
There was ascites. Patchy ecchymoses were present all over the surface of the liver and of
the large and small bowels. No perforations were noted. The spleen, kidneys, gall bladder,
ureter and bladder were congested.
85
Table 12. Summary of serial laboratory results: fever began 10 May
15 May 16 May
19:00 23:00 03:00 15:00
Hb (g/L) 184 111 83
HCT (%) 57 35 26
9
PLT (x 10 /L) 22 17 48
9
WBC (x 10 /L) 13.6 8.2 6.8
pH 6.972 6.906
Although it may not be easy to be certain about the underlying diagnosis of either severe
sepsis or dengue shock, what is important is the recognition of shock. Not once was
peripheral perfusion assessed. Fluid resuscitation should still be the primary action,
whether due to sepsis, dengue or some other cause, as compromised hemodynamic status
is recognized.
The severity of the compromised hemodynamic status was also underestimated. Rapid and
shallow breathing in a young patient with the absence of significant lung findings and
adequate SpO2 without O2 supplement should prompt the suspicion of possible acidotic
86
breathing; i.e. metabolic acidosis, and hence strengthen the suspicion of profound and/or
prolonged shock. Similarly, the poor urine output and seizures are all pointing towards the
same conclusion. Inotropes have limited value in the management of hypovolaemic shock
although they make actual BP readings appear better,
Significant bleeding was recognized late. By 03:00 on 16 May, the haematocrit had
decreased to 35% while the pH and lactate remained in severe acidosis. In the context of
dengue, this signifies severe haemorrhage and the need for urgent transfusion with fresh
whole blood. The main pitfall here was that the clinicians looked only at the Hb/ haematocrit
level to determine the possibility of bleeding and to decide on blood transfusion. This
practice would result in no, or late, recognition of severe bleeding and no, or delayed, blood
transfusion. It is also important to note that overt bleeding may not be present at the
beginning. In practice, the patient’s poor hemodynamic status and the need for large
volumes of fluid resuscitation, with or without oliguria and/or metabolic acidosis, are
important clues to the possibility of significant bleeding. Given the missed diagnosis of
shock at presentation in this patient and the progression to prolonged and profound shock,
severe bleeding should have been anticipated.
The other main pitfall is the practice of following a DIVC transfusion regime in managing
bleeding in dengue. Red cell transfusion is always the last to be given in this scenario. By
the time transfusion of other blood products has been completed, much time has been lost
and Hb/ haematocrit would have dropped even further. Hence it was not surprising that this
patient continued to deteriorate and bleed even though the abnormal coagulation profile or
platelet count level were partially corrected with aggressive transfusion of blood products.
This is because adequate tissue oxygen delivery needs sufficient levels of red cells,
particularly fresh red cells.
Another lesson is that if a patient continues to deteriorate while undergoing treatment, the
provisional diagnosis should be re-visited and the patient re-evaluated, perhaps by a fresh
team of clinicians who may be able to see the real picture. This case study demonstrated
the inability of clinicians to re-evaluate the clinical diagnosis even when the patient was not
responding to the prescribed therapy.
88
16:00 (22 hours post defervescence), she complained of abdominal pain and vomited
about 200 ml of coffee ground fluid. Her PR was 140–150/min, her peripheries were cold
and there was a small pulse volume. The systolic BP was 96–100 mmHg while the diastolic
BP was 78–88 mmHg and pulse pressure 15–20 mmHg (refer to clinical chart, page 95).
The bedside microcentrifuge showed an haematocrit of 53% and a blood glucose level of
8.8 mmol/L. Intravenous fluid was increased to 7 ml/kg/hour and she was started on
intravenous ranitidine (H-2 blocker) and cross-matched for fresh whole blood. (She was not
given the blood transfusion.)
18:00 She had no more abdominal pain or vomiting but clinically she had not improved;
her pulse volume was still small, her PR 140–145/min, and CRT 3–4 sec. Her pulse
pressure was narrow (10–15 mmHg). The bedside haematocrit was 48% (laboratory
haematocrit 50%). Fluid therapy was changed to colloid solution at 10 ml/kg/hour for one
hour.
19:00–24:00 Progressive clinical improvement was observed, refer to clinical chart: HR
120–130/min, pulse volume improved, BP 100/80 mmHg, pulse pressure 20–25 mmHg,
urine output 1–2 ml/kg/hour. Colloid infusion was reduced every hour from 8–7–
6 ml/kg/hour then changed to crystalloid at 5 ml/kg/hour. At 24:00 total fluid intake/output
since admission was 3466/1077 ml (positive balance of 2389 ml).
26 July
00:00–07:00 (30–37 hours post-defervescence), she showed improvement in the
haemodynamic state. Intravenous fluid therapy was reduced from 5–4–3–2–1 ml/kg/hour,
every 1–2 hours. By 07:00 she had warm peripheries and a good pulse volume, moderately
severe periorbital oedema, bilateral pleural effusion and ascites. The BP systolic was 100–
108, diastolic was 82–86 mmHg, HR 110–120/min, RR 40–45/min, SpO2 > 95% on air. Her
urine output was 1–1.5 ml/kg/hour. Her haematocrit remained about 49−48%. Intravenous
fluid therapy was discontinued at 07:00 on 26 July. At 16:00 her haematocrit was 43%.
Spontaneous diuresis ensued and her appetite improved. She was discharged from the
PICU on 27 July. Dengue IgM was positive. Table 13 shows a summary of serial CBC and
intravenous fluid therapy.
Lessons learnt and analysis of disease course:
At the ED medical consultation, a patient with fever, cough and running nose will usually be
given a diagnosis of URTI. However, the reduced oral intake, absence from school and
weakness and dehydration should prompt a more serious diagnosis. If one is unsure, a
CBC should be done. Another clue to the diagnosis of dengue is that the patient’s condition
worsened at a time when the temperature was becoming normal.
She presented to the hospital on the fourth day of illness, just after defervescence (her
temperature decreased from 40oC at 17:00 to 37°C at 19:00 on 24 July) with warning signs
of poor oral intake, vomiting, lethargy, high haematocrit and enlarged tender liver. (Her
baseline haematocrit was estimated to be <35%). Although she had evidence of plasma
leak, pleural effusion was not clinically detectable and peripheral circulation was assessed
to be normal at presentation to the ED.
89
Despite intravenous fluid therapy between admission and 07:00 on 25 July, she had
tachycardia and increasing haematocrit levels. These suggest inadequate replacement of
ongoing plasma leak.
At 08:15 on 25 July, about 14 hours after defervescence, she developed severe abdominal
pain and hypotensive shock. Pleural effusion had become detectable by this time. The
tachypnoea could be related to metabolic acidosis. She was resuscitated with 30 ml/kg of
crystalloid; the ideal fluid would have been colloids. Intravenous fluid therapy was gradually
reduced upon transfer to the PICU.
However, at about 17:00 on 25 July she developed compensated shock, having a BP of
100/90 mmHg with rising heart and respiratory rates. Although she had gastrointestinal
bleeding, her elevated haematocrit of 48−53% suggested that the main cause of shock was
plasma leak. Hence a colloid solution of 10 ml/kg was given. Clinical improvement after the
colloid infusion supported the decision to withhold blood transfusion. From admission to
18:00 on 25 July the cumulative fluid balance was positive (1381 ml) and this was reflected
in the moderately severe right pleural effusion and ascites.
Her haemodynamic status stabilized thereafter and intravenous therapy was discontinued
at 07:00 on 26 July, about 36 hours after defervescence (with a cumulative positive fluid
balance > 2000 ml). Note that the haematocrit was elevated to 48% at this time but her
peripheral perfusion was good, suggesting that she could compensate well for the plasma
leak. She was already in respiratory distress which would become worse if intravenous fluid
therapy were continued. A decrease in haematocrit at 16:00 was evidence of the re-
absorption phase of the disease. It is therefore important to consider the patient’s
haemodynamic state rather than the laboratory value exclusively, particularly when there is
already a big positive fluid balance. No blood products or antibiotics were given. The
haematocrit was 32% at hospital discharge on 29 July.
90
Table 13. Summary of serial CBC and intravenous fluid therapy: fever began on 21
July
Date HCT Bedside IV fluid therapy WBC PLT
(lab)% HCT %
Time (x 109/L) (x 109/L)
24 July
26 July
00:00 49 48 3 ml//kg/hour (Hartmann’s solution)
02:00 49 54 2 ml//kg/hour (Hartmann’s solution) 30.3 6
06:00 48 52 1 ml//kg/hour for 1 hour then stopped 30.6 10
16:00 43 No IV, oral fluids 36.0 10
21:00 38 No IV, oral fluids
27 July 34 Oral fluids 32.1 27
29 July 32 Discharge
CBC = complete blood count; ED = emergency department; HCT = haematocrit; IV = intravenous; NS = normal
saline or 0.9% sodium chloride solution; PLT =platelets, WBC = white blood cell
91
Clinical chart
Clinical chart in the paediatric intensive care unit showing temperature, BP, heart rate (upper black line) mean arterial BP
(red line), RR (lower black line) and peripheral perfusion records (First row, temperature of extremities: c = cold; w =
warm; s/c = slightly cold. Second row, capillary refill time in seconds. Third row, pulse volume: w = weak; g = good. Fourth
row, colour of extremities. Note the increasing diastolic pressure and narrowing pulse pressure and tachycardia between
15:00 and 18:00 on 25 July and the improving pulse pressure and reducing heart rate after colloid infusion between 16:00
and 20:00 on 25 July.
92
4.3 Case study three
First section
A previously well 10-month-old girl was brought to the ED on 31 October with generalized
seizures during the preceding 30 minutes.
She had had a high fever since 27 October (5 days duration); this was associated with poor
oral intake of fluids (water and a total of 90–120 ml of milk per day). She was less active than
usual.
On 28 October she had a generalized rash and was brought to the health clinic where a blood
test was said to be normal. The rash subsided on the 29 October; however the fever persisted
and she developed vomiting and diarrhoea twice a day while her fluid intake continued to be
poor. She had passed urine but her mother was unsure of the volume.
On 31 October she was lethargic and at 11:00 she became limp with up-rolling eyeballs. This
was followed by generalized tonic-clonic seizures. She was then brought to the ED.
31 October
12:00 On arrival at the ED her vital signs were: temperature 35.5°C (axilla), PR 162/minute,
BP 110/50 mmHg, SpO2 100% in room air. She was unresponsive and had jerking of the
limbs. Her pupils were about 2–3 mm equal. She showed neck stiffness, a depressed anterior
fontanelle, and a dry coated tongue. Feeble pulses with cold extremities and prolonged
capillary refill time were present. Her breathing was labored, but air entry was clear bilaterally.
A soft abdomen with a 4 cm firm liver was palpated. Her estimated body weight was 7 kg.
The following investigations were carried out:
Complete blood count, renal function test, liver function test, random blood glucose, venous
blood gases, blood culture and sensitivity, blood group and screen and dengue serology.
She was administered with high flow oxygen, diazepam 3 mg intravenously followed by 200 ml
of 0.9% saline fluid bolus. Intravenous ceftriaxone was given.
Another 100 ml of 0.9% saline and intravenous infusion of phenytoin was given for the status
epilepticus. She was then referred to the PICU.
Interpret the clinical history and physical signs at the ED (answers are to be found after
third section).
93
Second section
The following were the results of investigations in ED, before fluid resuscitation.
Table 14. Laboratory findings, case study 3, day 5 of illness, 31 October
pH 6.949
Urea 10 mmol/L
Creatinine 52 µmol/L
Albumin 28 g/L
Table 15. Complete blood counts on day 2 (28 October) and day 5 of illness (31 October)
Day 2, 28 October Day 5, 31 October,
(Health centre) ED of hospital
Haemoglobin (g/L) 113 108
Haematocrit (%) 36 33
9
White blood cell (x 10 /L) 4.0 15.2 (63% lymphocytes, 24% neutrophils)
9
Platelet count (x 10 /L) 340 24
Interpret the blood investigations, Tables 14 and 15 on day 2 and day 5 of illness (see answers
after following section).
Third section
31 October
15:00 On arrival in the PICU the patient was pale, unconscious with occasional decerebrate
posturing, and her pupils were equal in size, about 1–2 mm. She responded only to pain. Her
heart rate was 130/min and she had a normal breathing effort, an axillary temperature of
36.5°C and cold extremities with feeble radial pulses, prolonged capillary refill time and a BP of
115/62 mmHg. The anterior fontanelle was sunken and petechial hemorrhages and bleeding
from venepuncture sites were noted. A nasogastric tube yielded about 70 ml of coffee-ground
aspirates. A right pleural effusion was detected.
94
Investigation results in PICU (after resuscitation in ED):
Repeat blood glucose 6.8 mmol/L
Repeat haematocrit after 300 ml saline 0.26
Haemoglobin 86 gm/L
Interpret the clinical picture and blood investigations on arrival in PICU.
What is the definitive management?
ANSWERS
First section
Interpret the clinical history and physical signs at the emergency department
The history was that of an acute febrile illness associated with decreased fluid intake and
increased fluid loss resulting in dehydration. Clinical signs of shock included tachycardia in the
absence of fever, and decreased peripheral perfusion.
Seizures could indicate acute ischemia to the brain or meningitis/encephalitis or electrolyte
imbalance. Although seizures with neck stiffness were suggestive of meningitis, the subnormal
temperature made this diagnosis less likely. The sub-normal temperature meant
defervescence and the critical phase of dengue marked by increased vascular permeability,
plasma leakage and hypovolaemic shock. The patient most probably had dengue shock
syndrome.
Second section
Interpret the blood investigations on days 2 and 5 of illness, Tables 14 and 15
The CBC on day 2 of illness showed that the baseline haematocrit was 36%; the total white
cell count suggested a viral infection. The platelet count is expected to be normal on day 2 of
dengue illness.
On the day of admission, the venous blood gas showed severe metabolic acidosis, consistent
with shock. Hyperglycemia was part of the stress response. A lower haematocrit (33%) than
the baseline (36%) at the time of shock presentation suggested significant bleeding. The low
platelet count indicated the critical phase of dengue. Although the total white count was
elevated, this was predominantly due to lymphocytes; hence it was less likely that the patient
had a bacterial infection. The coagulation disorder (with prolonged prothrombin time and
activated partial thromboplastin time) together with raised liver enzymes, were consistent with
severe dengue with shock and bleeding. The hyponatraemia could be related to the intake of
plain water; the hyperkalaemia could be due to metabolic acidosis and the increased blood
urea due to dehydration and shock.
95
Third section
Interpret the clinical picture and blood investigations on arrival in PICU.
96
Table 16. Summary of fluid therapy and blood investigation results: fever began on 27
October
Date HCT (lab) Bedside IV fluid therapy WBC PLT
9 9
Time (%) HCT (%) (x 10 /L) (x 10 /L)
28 Oct 36 (Health centre) 4.0 340
31 Oct
12:00 33 (ED) NS 30–40 ml/kg 15.2 24
15:00 26 22 Colloid 20 ml/kg followed by fresh whole blood 10 8.2 17
(PICU) ml/kg over 2 hours (17:30 to 19:30)
20:00 27 27 Fresh whole blood 5 ml/kg over 2 hours + Dextrose-
saline 4 ml/hour
22:00 35 35 Dextrose-saline 3 ml/hour + NS 4 ml/hour
1 Nov
02:00 32 Dextrose-saline 3 ml/hour + NS 2 ml/hour 7.6 29
06:00 Dextrose-saline 3 ml/hour
12:00 34 Dextrose 10% saline 3 ml//kg/hour
16:00 31 Dextrose 10% saline 3 ml/kg/hour plus breastfeeding 6.8 Clump
4 Nov 25 82
5 Nov 29 Discharge
HCT = haematocrit; IV = intravenous; NS = normal saline or 0.9% sodium chloride solution; PLT = platelets, WBC =
white blood cell
97
4.4 Case study four
24 August
A previously well 24-year-old female was referred by a general practitioner for fever lasting
three days, myalgia and weakness lasting for two days, tea-coloured urine and mild jaundice
on the day of referral. A liver function test showed elevated liver enzymes and the test for
hepatitis B surface antigen was negative. She was referred with a diagnosis of possible
hepatitis A.
14:40 She was registered at the ED.
15:00 Observations at the triage counter showed that her temperature was 36.0°C; PR
110/min, RR 20/min and BP100/50 mmHg. Based on these findings she was given a triage
rating of three (not urgent).
16:00 She was seen by an ED physician who took her history of fever lasting 3 days;
anorexia and vomiting for 3 days and severe myalgia and dark-coloured urine. The physician
noted the vital signs taken at 15:00. He further detected a tender epigastrium and a positive
Murphy’s sign. The diagnosis made by the ED physician was obstructive jaundice. Blood was
sampled for the following investigations: CBC, liver function test, renal profile and urine
microscopic examination. The patient was referred for a general surgery opinion.
16:15 She was reviewed by general surgery staff who re-took the history of epigastric pain for
five days with worsening abdominal pain. She had generalized malaise together with loose
stools and had vomited several times. She appeared distressed and dehydrated. In addition to
the tender epigastrium and a positive Murphy’s sign, the abdomen was soft and not distended
with no guarding or rebound tenderness. Bowel sounds were present and per-rectal
examination revealed no blood or mass. The vital signs at 15:00 and blood investigations
taken at 16:00 were noted (Table 17). The surgical doctor made a diagnosis of urinary tract
infection with a differential diagnosis of acute cholecystitis. Due to the unexplained
thrombocytopenia and polycythaemia, he requested an ultrasound study of the abdomen (to
exclude gall stones) and a referral to internal medicine.
Table 17. Results of blood sampled at 16:00, case study 4
Blood biochemistry Liver function tests
98
18:30 The patient was reviewed by a medical physician after the ultrasound study. The
history was re-taken. Date of fever onset: 19 August (day 6 of fever on 24 August). Other
complaints included vomiting, poor appetite, lethargy and a transient rash. On 24 August she
experienced intense abdominal pain. The physician noted that the patient was flushed with a
tinge of jaundice. Her temperature was 36.2°C, PR 120/min, BP 100/80 mmHg. She had cold
extremities with weak and thready pulses. The lung fields were clear and tenderness in the
epigastrium and petechiae in the lower limbs were detected. The physician made a diagnosis
of compensated dengue shock in critical phase of illness and moved the patient to the
resuscitation bay. Defervescence was estimated at 12:00 on 24 August.
Management:
The patient’s actual body weight was 64 kg but based on her height, the ideal weight was
estimated to be 50 kg. A blood sample was taken for group and cross-match. She was given a
bolus of 10 ml/kg (500 ml) of 0.9% saline over 30 minutes followed by a 7 ml/kg/hour for 2
hours.
20:00 After 20 ml/kg (1000 ml) of 0.9 saline solution, the patient’s abdominal pain resolved;
she felt much better and her pulse volume was good. Her heart rate was 90/min and her BP
was 110/73 mmHg. Her abdomen was soft and non-tender; Murphy’s sign was negative. CBC
was repeated together with other investigations (Table 18). She was admitted to the general
medical ward.
The ultrasound study results were reviewed. They showed normal hepatic parenchyma,
thickened gall bladder wall with wall oedema, no gall stones, no intrahepatic duct dilatation and
no ascites. The radiological diagnosis was acalculous cholecystitis.
22:00 Intravenous fluid rate was reduced to 5 ml/kg/hour for 2 hours
24:00 Intravenous fluid rate was reduced to 3 ml/hour for 4 hours
25 August
04:00 Intravenous fluid rate reduced to 2 ml/kg/hour
08:00 Her general condition improved. She had passed urine three times during the night and
it was lighter in colour. She was able to take oral fluids. The intravenous fluid was maintained
at 1.5 ml/kg/hour.
26 August
08:00 Intravenous fluid therapy discontinued.
27 August
She was discharged, to be followed up a week later. Dengue IgM was positive.
99
Table 18. Serial laboratory results
Fever started on 19 August. Fluid resuscitation (20 ml/kg) was given between 18:30 and 20:00
on 24 August, followed by a step-wise reduction of fluid therapy from then through 25 August.
Intravenous fluid therapy was discontinued on 26 August.
24 Aug 25 Aug 26 Aug 27 Aug
16:00 20:00 24:00 09:00 16:00 20:00 06:00
4h PD 8h PD 12h PD 21h PD 28h PD 32h PD 42h PD
HCT(%) 50 43 43 44 42 41 41 40
9
WBC (10 /L) 6.9 6.8 6.1 7.2 6.5 6.5 7.4 12.6
9
PLT (10 /L) 4 6 5 5 2 5 5 31
Urea (mmol/L) 8.4 8.4 8.5 9.8 4.7
Creatinine 138 158 159 159 80 68
(mmol/L)
ALT (IU/L) 106 104 81 582
AST (IU/L) 198 191 142 1681
100
A shock that was missed
Dengue shock in its early stage can be easily missed, especially in adults − as shown in this
case. This illustrates clearly that without careful assessment of peripheral perfusion (i.e.
peripheral temperature, peripheral pulse volume and capillary refill time), a state of
compensated shock can be overlooked as the patient's BP is still well maintained. It is
important for clinicians to remember that touching the patient's extremities makes a big
difference to the clinical assessment. The other important clue for shock here is the
tachycardia in the absence of fever and/or other causes. Additional clues for early recognition
of reduced perfusion include reduced urine output, concentrated urine and silent tachypnoea.
Once dengue shock is recognized, prompt fluid management as per the algorithm in Figure 5
stabilizes the patient’s condition together with resolution of the severe abdominal pain. Table
18 shows the changes in CBC, renal profile and liver enzymes during admission. Note the
rapid, followed by a more gradual decrease in haematocrit with fluid resuscitation and step-
wise reduction in intravenous fluid therapy respectively. Despite the severe thrombocytopenia,
she was not given any transfusion of blood products. Renal profile improved with fluid therapy
but AST continued to increase even during the convalescent phase.
101
5. Annex
1 Assessment and follow-up of patients with Sending patients with non-severe dengue home
non-severe dengue and careful instruction with no follow-up and inadequate instructions
about warning signs to watch out for
2 Administration of paracetamol for high fever if Administration of acetylsalicylic acid (aspirin) or
the patient is uncomfortable ibuprofen
3 Obtaining a haematocrit level before and after Not knowing when haematocrit levels are taken
fluid boluses with respect to fluid therapy
4 Clinical assessment of the haemodynamic No clinical assessment of patient with respect to
status before and after each fluid bolus fluid therapy
5 Interpretation of haematocrit levels in the Interpretation of haematocrit levels independent of
context of fluid administered and clinical status and fluid therapy
haemodynamic assessment
6 Administration of intravenous fluids for Administration of intravenous fluids to all patient
repeated vomiting or a high or rapidly rising with non-severe dengue
haematocrit
7 Use of isotonic intravenous fluids for severe Use of hypotonic intravenous fluids for severe
dengue dengue
8 Giving just sufficient intravenous fluid volume Excessive or prolonged intravenous fluid
to maintain effective circulation during the administration for severe dengue
period of plasma leakage for severe dengue
9 Avoiding intramuscular injections in dengue Giving intramuscular injections to dengue patients
patients
10 Intravenous fluid rate and frequency of Fixed intravenous fluid rate and unchanged
monitoring and haematocrit measurement frequency of monitoring and haematocrit
adjusted according to the patient’s condition measurement during entire hospitalization for
severe dengue
11 Close monitoring of blood glucose, i.e. tight Not monitoring blood glucose, unaware of the
glycaemic control hyperglycaemic effect on osmotic diuresis and
hypovolaemia
12 Discontinuation or reducing fluid therapy once Continuation and no review of intravenous fluid
haemodynamic status stabilizes therapy once haemodynamic status stabilizes.
13 Urine output is carefully recorded and included Urine output is not recorded or ignored in the
in assessment of haemodynamic status. assessment of fluid balance.
102
5.2 Textbox B: Differential diagnosis of dengue fever
Illnesses with a rash Rubella, measles, scarlet fever, meningococcal infection, chikungunya, drug
reactions
Diarrhoeal diseases Rotavirus, other enteric infections
Illnesses with neurological Meningoencephalitis
manifestations Febrile seizures
Conditions that mimic the critical phase of dengue infection
Infectious Acute gastroenteritis, malaria, leptospirosis, typhoid, typhus, viral hepatitis,
Acute HIV-seroconversion illness, bacterial sepsis, septic shock
Malignancies Acute leukaemia and other malignancies
Other clinical pictures Acute abdomen
Acute appendicitis
Acute cholecystitis
Perforated viscus
Diabetic ketoacidosis
Kawasaki syndrome
Lactic acidosis
Leukopenia, thrombocytopenia and bleeding
Platelet disorders
Renal failure
Respiratory distress (Kussmaul's breathing)
Systemic lupus erythematosus
103
5.3 Textbox C: Warning signs
Persistent vomiting
Lethargy, restlessness
Mucosal bleed
Liver enlargement > 2cm or tender enlarged liver
Laboratory Increase in haematocrit level concurrent with rapid decrease in platelet count
Conscious level Clear and lucid Clear and lucid (shock can be Change of mental state
missed if you do not touch (restless, combative)
the patient)
Capillary refill time Brisk (< 2 sec) Prolonged (> 2 sec) Very prolonged, mottled skin
BP Normal for age Normal systolic pressure but Hypotension (see definition
rising diastolic pressure below)
Normal pulse pressure
for age Narrowing pulse pressure Unrecordable BP
(≤ 20 mmHg in children)
Postural hypotension
Note: Definition of hypotension = systolic BP of < 90 mmHg or mean arterial pressure < 70 mmHg in adults or a
systolic BP decrease of > 40 mmHg or < 2 standard deviation (SD) below normal for age.
In children up to 10 years of age, the 5th centile for systolic BP can be determined by the formula: 70 + (age in
years x 2) mmHg.
104
5.5 Textbox E: Admission criteria
Signs and symptoms related to Dehydrated patient, unable to tolerate oral fluids
hypotension (possible plasma
Dizziness or postural hypotension
leakage)
Profuse perspiration, fainting, prostration during defervescence
Hypotension or cold extremities
Clinical
No fever for 48 hours
Improvement in clinical status (general well-being, appetite, haemodynamic status, urine output, no respiratory
distress)
Laboratory
105
5.7 Textbox G: Home-care card for dengue (for patient or adult caregiver)
Home-care card for dengue
The patient should take this card with them to the health facility for each visit
If any of following is observed, the patient should be immediately taken to the nearest hospital; these are
warning signs for danger:
• Bleeding:
− red spots or patches on the skin
− bleeding from nose or gums
− vomiting blood
− black-coloured stools
− heavy menstruation/vaginal bleeding
• Frequent vomiting or not able to drink
• Severe abdominal pain
• Drowsiness, mental confusion or seizures
• Pale, cold or clammy hands and feet
• Difficulty in breathing
• Postural dizziness
• No urine output for 4–6 hours
1st visit
Date
Haematocrit
Platelet count
106
5.8 Textbox H: Calculations for normal maintenance of intravenous fluid
infusion
Normal maintenance fluid per hour can be calculated on the basis of the following formula* (equivalent to Holliday-
Segar formula):
4 ml/kg/hr for first 10 kg body weight
*For overweight/obese patients calculate normal maintenance fluid based on ideal body weight (IBW), using the
following formula:
5.9 Textbox I: Hourly maintenance fluid regime based on ideal body weight
Estimated ideal body weight (IBW) Normal maintenance fluid based on Holliday-Segar formula
(kg)
ml/hour ml/kg/hour
5 20 4
10 40 4
15 50 3.3
20 60 3
25 65 2.6
30 70 2.3
35 75 2.1
40 80 2
50 90 1.8
60 100 1.7
70 110 1.6
80 120 1.5
For adults with IBW > 50 kg, 1.5−2 ml/kg can be used for quick calculation of hourly maintenance fluid regime.
For adults with IBW <50 kg, 2−3 ml/kg can be used for quick calculation of hourly maintenance fluid regime.
107
5.10 Textbox J: Estimated ideal body weight for overweight or obese adults
Height Estimated, IBW (kg) for adult males Estimated IBW (kg) for adult females
(cm)
150 50 45.5
160 57 52
170 66 61.5
180 75 70
108
5.11 Textbox K: Example of a monitoring chart for dengue: Vital signs and hourly fluid monitoring of dengue
patients
Date
Time*
HCT
(%)
Platelet
WBC
Temperature
(°C)
BP
Pulse rate
(PR)
Respiratory
rate
Crystalloids
ml/kg/h
Cum Vol
109
Colloids
ml/kg/h
Cum vol
Blood product
Type ml/kg/h
Cum vol
Oral quantity
Cum oral
Cum input
Hourly urine
Cum urine
Others
WBC = white blood cell; Cum vol = cumulative volume i.e. total volume since start of treatment; Cum oral = cumulative oral intake since start of treatment; Cum
input = cumulative intravenous and oral fluid input; Cum urine = cumulative urine i.e. total urine volume since start of treatment
110
5.12 Textbox L: Choice of intravenous fluids for resuscitation
Based on the three randomized controlled trials comparing the different types of fluid resuscitation regime in
dengue shock in children, there is no clear advantage to the use of colloids over crystalloids in terms of the
overall outcome. However colloids may be the preferred choice if the BP has to be restored urgently, i.e. in
those with pulse pressure less than 10 mmHg. Colloids have been shown to restore the cardiac index and
reduce the level of haematocrit faster than crystalloids in patients with intractable shock
An ideal physiological fluid is one that resembles the extracellular and intracellular fluid compartments closely.
However, when used in large quantities, each type of available fluid has its own limitations. Therefore it is
advisable to understand the limitations of these solutions to avoid their respective complications.
Crystalloids
0.9% saline is hypertonic at an osmolality of 308 mOsm/L and contains a high sodium and chloride level (154
mmol/L each). Normal plasma chloride ranges from 95−105 mmol/L. 0.9% saline is a suitable option for initial
fluid resuscitation, but repeated large volumes of 0.9% saline may lead to hyperchloraemic acidosis.
Hyperchloraemic acidosis may aggravate or be confused with lactic acidosis from prolonged shock. Monitoring
the chloride and lactate levels will help to identify this problem. When the patient’s serum chloride level exceeds
the normal range, it is advisable to change to other alternatives such as Ringer’s lactate.
Ringer’s lactate
Ringer’s lactate has lower sodium (131 mmol/L) and chloride (115 mmol/L) contents and an osmolality of 273
mOsm/L. It may not be suitable for resuscitation of patients with severe hyponatraemia. However, it is a suitable
solution after 0.9% saline has been given and the patient’s serum chloride level has exceeded the normal range.
Ringer’s lactate should probably be avoided in liver failure and in patients taking metformin where lactate
metabolism may be impaired.
Colloids
The types of colloids are: dextran-based, hydroxyethyl starch and gelatin-based solutions.
One of the biggest concerns regarding their use is their impact on coagulation. Theoretically, dextrans bind to
von Willebrand factor/Factor VIII complex and impair coagulation the most. However, this was not observed to
have clinical significance in fluid resuscitation in dengue shock. Of all the colloids, gelatine has the least effect
on coagulation but the highest risk of allergic reactions. Allergic reactions such as fever, chills and rigors have
also been observed with Dextran 70. Dextran 40 and starch colloidal solutions can potentially cause an osmotic
renal injury in hypovolaemic patients. There is an upper limit to the volume per kg body weight per day.
After two to three boluses of crystalloid without haemodynamic stability and particularly if the haematocrit is still
elevated, it is essential to consider the switch to colloid resuscitation which is more effective in lowering the
haematocrit. When the haematocrit is lowered and the patient’s haemodynamic state improves, it indicates the
restoration of the circulating volume. However, if there is no improvement in the haemodynamic state, it is
essential to consider if severe bleeding has occurred.
111
Neglected Tropical Diseases (NTD) TDR/World Health Organization
HIV/AIDS, Tuberculosis and Malaria (HTM) 20, Avenue Appia
20, Avenue Appia 1211 Geneva 27
ISBN 978 92 4 150471 3 1211 Geneva 27, Switzerland Switzerland