52-3-10

7/28/2019 52-3-10

1/6

Paediatrica IndonesianaVOLUME 52 NUMBER 3May

Original Article

Paediatr Indones, Vol. 52, No. 3, May 2012 175

Peripheral blood examination to assess bleeding risk in

children with dengue infections

Irene R Huwae, Kurniawan T Kadafi

AbstractBackground Dengue viral infection may cause mild to severeclinical manifestations, with or without bleeding. A number of

factors may cause bleeding in patients with dengue. However,

health providers may be unable to perform the examinations

required to sufficiently predict the risk of bleeding.

ObjectiveTo find risk factors for bleeding using peripheral bloodexaminations in children with dengue infection.

Methods This cross-sectional study was conducted at the3HGLDWULF:DUGRIWKH'U6DLIXO$QZDU*HQHUDO+RVSLWDO

0DODQJIURP-DQXDU\WR'HFHPEHU:HLQFOXGHG

FKLOGUHQDJHGWR\HDUVZLWKGHQJXHYLUDOLQIHFWLRQDV

FRQILUPHGE\WKH:+2FULWHULDDQGVHURORJ\3HULSKHUDO

EORRGH[DPLQDWLRQVZHUHPDGHGDLO\GHSHQGLQJRQWKHSDWLHQWVcondition. We classified the bleeding status into non-bleeding,

petechial bleeding (mild hemorrhage), and mucosal bleeding

VHYHUHKHPRUUKDJH:HUHFRUGHGVXEMHFWVEOHHGLQJVWDWXVDW

WKHWLPHRIWKHLUKLJKHVWSDFNHGFHOOYROXPH3&9DQGUHFRUGHG

their leukocyte and platelet counts at that time. We computed

WKHSDUDPHWHUVPHGLDQVDQGFRPSDUHGWKHPWREOHHGLQJVWDWXV

E\&KLVTXDUHWHVW)RUVLJQLILFDQW3DVVRFLDWLRQVZH

FDOFXODWHGWKH25RGGVUDWLRZLWKDFRQILGHQFHLQWHUYDO

$OOSDWLHQWVZHUHWUHDWHGDFFRUGLQJWRWKH:+2GHQJXH

guidelines.

Results 7KHUHZHUHDOOVXEMHFWVZLWKGHQJXHDQGVXEMHFWVKDGFRPSOHWHGDWDZLWKEOHHGLQJSHWHFKLDO

PXFRVDOEOHHGLQJDQGZLWKRXWEOHHGLQJ7KHPHGLDQ

3&9ZDVZKLOHPHGLDQSODWHOHWFRXQWZDV/DQGPHGLDQOHXNRF\WHFRXQWZDV/7KH25RI3&9

!IRUEOHHGLQJZDV&,WR7KH25

RISODWHOHWFRXQW/IRUEOHHGLQJZDV&,

WRFRPSDUHGWRSODWHOHWFRXQW!/7KH

25RISODWHOHWFRXQW/IRUPXFRVDOEOHHGLQJZDV

&,WR&KLVTXDUHDQDO\VLVIRUOHXNRF\WH

count showed it was not associated with bleeding in dengue

3

)URPWKH'LYLVLRQRI,QIHFWLRXV'LVHDVHVDQG7URSLFDO3HGLDWULFV

'HSDUWPHQWRI&KLOG+HDOWK0HGLFDO6FKRRO%UDZLMD\D8QLYHUVLW\'U

6DLIXO$QZDU*HQHUDO+RVSLWDO0DODQJ,QGRQHVLD

Reprint requests to: Irene Ratridewi Huwae, Division of Infectious

'LVHDVHVDQG7URSLFDO3HGLDWULFV'HSDUWPHQWRI&KLOG+HDOWK0HGLFDO

6FKRRO%UDZLMD\D8QLYHUVLW\'U6DLIXO $QZDU*HQHUDO+RVSLWDO-O

-DNJXQJ6XSUDSWR0DODQJ,QGRQHVLD7HO)D[

([email protected]

Dengue viral infection is the most prevalent

arboviral disease worldwide. It is caused by

infection with any one of four dengue virus

'(19VHURW\SHV'(19'(19

infections cause illness in tens of millions each year

throughout the tropics and subtropics and severe

morbidity in approximately 2 million persons/year,

ZLWKDSSUR[LPDWHO\GHDWKV\HDUIn Indonesia,

Conclusion 7KH 3&9 OHYHO!LQFUHDVHGWKH ULVNIRUEOHHGLQJE\WLPHVIRUERWKSHWHFKLDODQGPXFRVDOEOHHGLQJ

3ODWHOHWFRXQW/LQFUHDVHGWKHULVNIRUEOHHGLQJ

times and for mucosal bleeding by 3.4 times. Leukocytes count

was not associated with bleeding. Basic laboratory examinations

RI3&9DQGSODWHOHWFRXQWPD\WKHUHIRUHEHXVHGDVDSUHGLFWRU

of bleeding in children with dengue infection. [Paediatr Indones.

2012;52:175-80].

Keywords: peripheral blood, dengue virus, bleeding

7/28/2019 52-3-10

2/6

Irene R Huwae et al: 3HULSKHUDOEORRGH[DPLQDWLRQWRDVVHVVEOHHGLQJULVNLQGHQJXHLQIHFWLRQV

176Paediatr Indones, Vol. 52, No. 3, May 2012

dengue viral infections have spread through all

SURYLQFHVDQGYLOODJHVVLQFHZLWKWKHKLJKHVW

LQFLGHQFHLQZLWKDERXWSDWLHQWV2

Dengue virus infections may be subclinical or

manifest as dengue fever (DF) or dengue hemorrhagic

IHYHU'+)'+)LVD V\QGURPHZLWKGD\VRI

fever, headache, myalgia, and rash, accompanied byleukopenia and varying degrees of thrombocytopenia.3

An early clinical diagnosis of DHF is difficult because

the World Health Organization (WHO) clinical and

laboratory criteria for DHF may be manifested only

in the late phase of acute illness.4 The latest WHO

FULWHULDUHSRUWHGPXFRVDOEOHHGLQJDVDZDUQLQJ

sign of dengue, leading to severe dengue when it

presents as severe bleeding.

The pathogenesis of bleeding in dengue viral

LQIHFWLRQHQFRPSDVVHVPHFKDQLVPVYDVFXORSDWK\

thrombopathy, and coagulopathy. These 3 mechanisms

are caused by interconnected complement activation

cascades.3-5 Measuring vasculopathy, thrombopathy,

and coagulopathy is complex, expensive, and cannot

be done in all health care settings. The routine

follow-up for dengue patients includes a simple and

low-cost complete blood examination, especially

in the post-febrile defervesence period, which is a

particularly critical phase. The aim of this study

was to determine the predictive value of simple

laboratory examination on bleeding tendency in

dengue patients. The examinations include packed

FHOOYROXPH3&9OHXNRF\WHFRXQWDQGSODWHOHWcount, all of which are available in limited resource

settings.

Methods

This cross-sectional study was undertaken from

-DQXDU\ WR'HFHPEHU :H LQFOXGHG DOO

FKLOGUHQ WR \HDUVROGDGPLWWHG WR'U6DLIXO

Anwar General Hospital, with dengue viral infection

only. The dengue diagnoses were made according to

:+2FULWHULDDQGFRQILUPHGE\VHURORJ\$V

suggested in the guidelines, hematologic examinations

ZHUHSHUIRUPHGGDLO\GHSHQGLQJRQSDWLHQWVFOLQLFDO

FRQGLWLRQLQFOXGLQJ3&9OHYHOVSODWHOHWFRXQWV

DQGOHXNRF\WHFRXQWV7KHKLJKHVWPHGLDQ3&9ZDV

recorded with the assumption that it represented the

critical phase (leakage period). Median leukocyte

counts and platelet counts were also taken at that

time. We did not measure or observe bleeding volumes

nor outcomes of the patients. Data of patients

requiring blood or other transfusion components was

taken only up to the time of transfusion.

Laboratory examinations were performed in the

&OLQLFDO3DWKRORJ\/DERUDWRU\RI'U6DLIXO$QZDUGeneral Hospital. We did not collect informed consent

in this study because the laboratory examination was

part of the dengue management.

6XEMHFWVZLWKEOHHGLQJZHUHGLYLGHGLQWR

groups, i.e. mild bleeding represented by petechial

bleeding, and mucosal bleeding with the potential

IRUVHYHUHEOHHGLQJ7KH:+2FULWHULDUHSRUWHG

mucosal bleeding to be a warning sign which may

OHDGWRVHYHUHGHQJXH+RZHYHULQWKH:+2

criteria, a spontaneous bleeding tendency was defined

as dengue hemorrhagic fever grade II. The bleeding

tendency of each patient was recorded at the same

WLPHWKHKLJKHVW3&9GDWDZDVWDNHQ

:HFDOFXODWHGWKHRGGVUDWLR25RIWKH3&9

level, leukocyte count, and platelet count relationships

to non-bleeding, petechial bleeding, and mucosal

bleeding case status.

Results

7KHUHZHUHSDWLHQWVZLWKGHQJXHYLUDOLQIHFWLRQ

EXWRQO\KDGFRPSOHWHGDWD$PRQJWKHVHKDGEOHHGLQJSHWHFKLDOEOHHGLQJPXFRVDO

EOHHGLQJZKLOHKDGQREOHHGLQJ7KHPHGLDQ

3&9ZDVPHGLDQSODWHOHWVFRXQWZDV

/DQGPHGLDQOHXNRF\WHVFRXQWZDV/7KH

numbers of male and female patients with dengue

viral infection were not significantly different. The

incidence of petechial bleeding appeared greater than

mucosal bleeding in each of the above blood parameter

groups. The basic characteristics of subjects are shown

LQ7DEOH

Table 2 shows the OR of predicting the risk of

EOHHGLQJ 6XEMHFWVZLWK D3&9!KDG

WLPHVWKHULVNRIEOHHGLQJWKDQWKRVHZLWK3&9

7KHOHXNRF\WHVFRXQWZDVQRWVLJQLILFDQWO\

DVVRFLDWHGZLWKEOHHGLQJWHQGHQF\6XEMHFWVZLWK

SODWHOHWVFRXQW/KDGWLPHVWKHULVN

for mucosal bleeding than those with platelet count

!/

7/28/2019 52-3-10

3/6



Discussion

Dengue has a wide spectrum of clinical presentations,

often with unpredictable clinical evolution and

outcomes. While most patients recover following

a self-limiting, non-severe clinical course, a small

proportion progress to severe disease, mostly

characterized by plasma leakage with or without

hemorrhage.67KH:+2FULWHULDJXLGHOLQH

groups symptomatic dengue virus infections into three

FDWHJRULHVXQGLIIHUHQWLDWHGIHYHUGHQJXHIHYHU')

and dengue hemorrhagic fever (DHF). DHF is further

classified into four severity grades, with grades III and

IV being defined as dengue shock syndrome.

There are several difficulties in applying these

criteria in that some cases cannot be classified intoRQHRIWKRVHWKUHHGHQJXHFDWHJRULHV7KH:+2

criteria guideline is more practical for clinicians. The

:+2FULWHULDFODVVLI\GHQJXHFDVHVLQWRWKUHH

JURXSV*URXSFRPSULVHVSUREDEOHGHQJXHFDVHV

defined as a person who traveled or lived in an endemic

DUHDDQGKDGIHYHUZLWKRIIROORZLQJFULWHULDQDXVHD

Table 2.#UUQEKCVKQPQHDNGGFKPIVQ2%8NGWMQE[VGUEQWPVCPFRNCVGNGVUEQWPV

Non-

bleeding

n=80

Bleeding

n=202

OR

(95% CI)

Non-

bleeding

n=80

Petechial

bleeding

n=120

OR

Non-

bleeding

n=80

Mucosal

bleeding

n=82

OR

(95% CI)

PCV, n

> 36.8%

3,400/L

Chi-squareP=0.128

Platelet count, n

< 51,000/L

>51,000/L

28

52

46

34

30

50

112

90

97

105

118

84

2.31

(1.35 to 3.95)

2.34

(1.37 to 3.99)

28

52

46

34

30

50

67

53

53

67

63

57

2.35

(1.31 to 4.21)

1.84

(1.03 to 3.28)

28

52

46

34

30

50

45

37

44

38

55

27

2.26

(1.20 to 4.25)

3.4

(1.78 to 6.48)

Table 1.Subjects basic characteristics

CharacteristicsNon-bleeding

(n=80)Petechial bleeding

(n=120)Mucosal bleeding

(n=82)Total

(n=282)

MaleFemale

Median PCV

Median leucocytes countMedian platelets count

PCV < 36.8%PCV > 36.8%

Leucocytes count< 3,400/LLeucocytes count>3,400/L

Platelets count< 51,000/LPlatelets count>51,000/L

Stage of dengue hemorrhagicfeveDHF IDHF IIDHF IIIDHF IV

4634

5228

46

34

30

50

6357

5367

53

67

63

57

4438

3745

44

38

55

27

153129

36.8%

3,400/ L51,000/ L

142140

143

139

148

134

12432366129

7/28/2019 52-3-10

4/6



vomiting, rash, aches and pain, positive tourniquet

test, leukopenia, or any warning signs. Group (2)

comprises dengue cases with warning signs, such as

abdominal pain and tenderness, persistent vomiting,

clinical fluid accumulation, mucosal bleeding, lethargy,

UHVWOHVVQHVVOLYHUHQODUJHPHQW!FPDQGLQFUHDVLQJ

3&9FRQFXUUHQWZLWKUDSLGO\GHFUHDVLQJSODWHOHWcount. Group (3) comprises severe dengue, defined as

cases with severe plasma leakage (leading to shock or

fluid accumulation with respiratory distress), severe

bleeding (evaluated by clinician), and severe organ

LQYROYHPHQWOLYHU$67!EUDLQGHFUHDVHG

level of consciousness, heart and/or other organs).

These criteria require a basic and simple hematologic

examination to classify each case as probable dengue

with or without warning signs. The hematological exam

VKRXOGLQFOXGHKHPRJORELQ+EOHYHO3&9OHXNRF\WH

and platelet counts. Monitoring for bleeding is required

in dengue cases with warning signs that require hospital

admission.

:HREVHUYHGWKDWSDWLHQWVZLWK3&9!

KDGDWLPHVJUHDWHUULVNIRUEOHHGLQJWKDQWKRVH

ZLWK3&9:HXVHGWKHKLJKHVW3&9YDOXH

for each subject in our calculations in this study with

the assumption that patients were in the critical phase

at that time. The critical phase is characterized by

LQFUHDVLQJ3&9FRQFXUUHQWZLWKGHFUHDVLQJSODWHOHW

count. Although a bleeding tendency usually occurs

as a manifestation of plasma leakage and shock,it may

occur in cases without plasma leakage (previouslyclassified as dengue fever), usually manifested as

thrombocytopenia.Among the bleeding cases,

WKHUHZDVQRVWDWLVWLFDOHYLGHQFHWKDW3&9!

was a potential risk factor for mucosal bleeding, only

compared to petechial bleeding.

,QFUHDVHG3&9UHSUHVHQWVWKHKHPRFRQFHQWUDWLRQ

in dengue, which is caused by vascular leakage. One

possible cause of vascular leakage is platelet activating

IDFWRU3$)ZKLFKFDQSURGXFHPDQ\IHDWXUHVRI

V\VWHPLFLQIODPPDWRU\ UHVSRQVHV\QGURPH6,56

such as hypotension, increased vascular permeability,

cytokine release, and shock.

Thrombocytopenia and plasma leakage are

the main characteristics of severe dengue disease. It

has been postulated that autoimmunity is involved

in dengue pathogenesis, especially through the

generation of autoantibodies against platelets,

endothelial cells, and coagulatory products. The cross-

reactive antibodies may cause platelet dysfunction,

endothelial cell damage, coagulation defects, and

macrophage activation.

The platelets count and the platelets usage for

sealing the vascular breakage may be considered as the

cause of bleeding.The OR of platelets in non-bleeding

and petechial cases (mild bleeding) was only 2.34,LQGLFDWLQJWKDWVXEMHFWVZLWKSODWHOHWFRXQW

/RQO\KDGWLPHVJUHDWHUULVNIRUEOHHGLQJERWK

petechial and mucosal) than subjects with platelet

FRXQW! /6XEMHFWVZLWKDSODWHOHWV FRXQW

/KDGDWLPHVJUHDWHUULVNRIKDYLQJ

PXFRVDOEOHHGLQJWKDQWKRVHZLWK!/DQG

VXEMHFWVZLWKDSODWHOHWVFRXQW/KDGD

times greater risk of having petechial bleeding than

WKRVHZLWK!/7KLVREVHUYDWLRQVXSSRUWV

the notion that petechial bleeding may be considered

as a milder and less dangerous bleeding tendencythan mucosal bleeding. However, we did not evaluate

other parameters to assess bleeding tendency in this

study, such as hemostatic states by activated partial

WKURPERSODVWLQWLPH$377DQG337

:HREVHUYHGVHYHUDOFDVHVQZLWK

SODWHOHWFRXQW/ZLWKVHYHUHKHPRUUKDJH

(mucosal bleeding). This hemorrhaging may be caused

by a thrombopathy (platelet dysfunction), in addition

to the decrease of platelets in peripheral blood. Thus,

the decrease in platelets count (thrombocytopenia)

may not be the only cause of bleeding, but may be

one risk factor for bleeding tendency during dengueinfection.

7KHEOHHGLQJWHQGHQF\LQ'+)'66SDWLHQWV

involves the role of both vascular endothelial cells

and platelets, although the pathogenic mechanism

LVQRWIXOO\XQGHUVWRRG6WXGLHVRQRWKHUYLUXVHV

such as CMV, HIV, and hepatitis C virus has shown

the presentation of platelet autoantibodies to be

the cause of thrombocytopenia. In dengue viral

infection these auto anti-platelet antibodies induce

complement-mediated cell lysis, which may play

role in thrombocytopenia. During blood vessel

injury, activated platelets adhere to the injury site,

aggregating together through fibrin formation.

Carlos et al concluded that thrombocytopenia was

not the only cause of bleeding in dengue, but that the

SURORQJDWLRQRI$377DQGUHGXFWLRQLQILEULQRJHQ

concentration were strongly associated with severity

of vascular leakage, leading to hypotension, shock

7/28/2019 52-3-10

5/6



and finally bleeding.3 Other studies have reported

that viremia may play a role in determining clinical

parameters, as is the case of secondary infection.

Higher viral load may be used as an independent

predictor for more severe thrombocytopenia, but

may also be associated with a small increase in

hemoconcentration.Examinations to determineviral load are expensive and not available in all

settings.

The leukocyte count of our subjects was not

significantly associated with a bleeding tendency.

Hence, it should not be considered as a risk factor

for predicting bleeding. However, several studies have

reported contrasting result on leukocyte counts in

dengue. A Thai study showed that leukopenia was

a good predictor of dengue infection in children,

whilea study in Nicaragua stated that leukopenia was

significantly associated with dengue in adults, but not

LQFKLOGUHQ$VWXG\LQ3XHUWR5LFRZDVLQFRQFOXVLYH

about the role of leukopenia in dengue, but their

results showed that leukocyte counts in dengue cases

ZHUHORZHUPHDQ/WKDQWKRVHRIQRQGHQJXH

cases.$QRWKHUVWXG\RQULVNIDFWRUVIRU'66LQ

FKLOGUHQFRQFOXGHGWKDWOHXNRSHQLD/ was a

significant risk factor for shock in dengue. However,

the authors did not mention an association between

leukopenia and hemorrhage.

6XEMHFWVZLWK3&9OHYHOVRIKLJKHUWKDQ

KDGDWLPHVKLJKHUULVNIRUEOHHGLQJHLWKHU

petechial or mucosal. Furthermore, a platelet countRIOHVVWKDQ/VKRZHGDWLPHVKLJKHU

risk for mucosal bleeding. Leukocyte count was

notsignificantly correlated to bleeding in our dengue

cases. As such, a basic laboratory examination of

3&9DQGSODWHOHWFRXQWPD\EHXVHGDVDSUHGLFWRU

of bleeding in pediatric dengue cases.

References

.KULVQDPXUWL&.DOD\DQDURRM6&XWWLQJ0$3HDW5$

5RWKZHOO6:5HLG7- HWDO0HFKDQLVPVRI KHPRUUKDJH

in dengue without circulatory collapse. Am J Trop Med.

8..,QIHNVLGDQ3HGLDWUL7URSLV,QIHNVLYLUXVGHQJXH,Q

6RHGDUPR663*DUQD++DGLQHJRUR656DWDUL+,HGLWRUV

Buku ajar infeksi dan pediatri tropis. 2ndHG-DNDUWD%DGDQ

3HQHUELW,'$,S

&DUORV&&2LVKL.&LQFR07''0DSXD&$,QRXH6&UX]

DJM, et al. Comparison of clinical features and hematologic

abnormalities between dengue fever and dengue hemorrhar-

gic feveUDPRQJFKLOGUHQLQWKH3KLOLSSLnes. Am J Trop Med

+\J

6KX3

7/28/2019 52-3-10

6/6



'X\HQ+7/1JRF79+DGR7+DQJ977.LHX177

Documents

52-3-10