Artikel Penelitian

Predictor of Dengue Shock Syndrome

Among Pediatric Dengue Infection in
Limited Resource Setting

Cindy Cecilia,* Julius Albert Sugianto**

*
Sampang General Hospital, Sampang, Madura, Indonesia
**
Faculty of Medicine, Universitas Airlangga, Surabaya, Indonesia

Abstract
Introduction: Indonesia has the most dengue outbreaks in the ASEAN region. Lack of
diagnostic modality and poor sanitation especially in rural area hinders proper and
prompt management of Dengue Fever.
Aim: Finding the early, simple, and reliable predictor of Dengue Shock Syndrome
(DSS) in Indonesia.
Materials and Methods: A retrospective study reviewing the medical record of pediatric
patient with dengue infection from 1 January 2016 to 31 March 2016 were conducted.
Age, gender, dengue signs, and simple laboratory result were collected. The data ac-
quired were then analyzed using logistic regression for multivariate analysis (P<0,05)
to find the severity predictor.
Results: 80 cases of dengue infection were included, mean age was 7,7 ± 3.4 years old,
the average length of hospital stay was 4.59 days. 28.75% of patients had DSS. From the
multivariate analysis, it is found that lethargy (p= 0.00; OR 21.23), bleeding (p= 0.01;
OR 0,04), cold extremities (p=0.00;OR=22.35), and increased haematocrit level
(p=0.01;OR=4.72) has significant relationship with DSS while other indicator did not.
Conclusion: Patients presenting with lethargy, cold extremities, and increase in
haematocrit level should be treated promptly and intensively as they are more likely to
develop DSS compared to patients who did not.
Keywords: dengue, dengue severy predictor, dengue in pediatrics.

Korespondensi: Julius Albert Sugianto

E-mail: [email protected]

178 J Indon Med Assoc, Volum: 69, Nomor: 4, April 2019

 Predictor of Dengue Shock Syndrome Among Pediatric Dengue Infection

Tanda Awalan Infeksi Dengue pada Pasien Pediatri untuk

Memprediksi Dengue Shock Syndrome di Daerah dengan
Modalitas Pemeriksaan Terbatas

Cindy Cecilia,* Julius Albert Sugianto**

*
Rumah Sakit Umum Daerah Sampang, Madura, Indonesia
**
Fakultas Kedokteran, Universitas Airlangga, Surabaya, Indonesia

Abstrak
Pendahuluan: Indonesia memiliki jumlah outbreak dengue terbesar di daerah
ASEAN. Kurangnya modalitas diagnosis dan sanitasi terutama di daerah
terpencil menjadi faktor utama yang mempersulit penanganan segera dan tepat
untuk demam dengue.
Tujuan: Menemukan tanda awal, sederhana, dan reliabel untuk memprediksi
Dengue Shock Syndrome (DSS) di Indonesia.
Metode: Dilakukan sebuah studi retrospektif melihat rekam medis pasien
pediatri dengan infeksi dengue antara 1 Januari 2016 hingga 31 Maret 2016.
Umur, jenis kelamin, tanda awalan, dan hasil laboratorium sederhana dicatat.
Data yang didapat kemudian dianalisis menggunakan uji regresi logistik untuk
analisis multivariat (P<0,05) untuk menemukan tanda yang dapat memprediksi
keparahan.
Hasil: Sebanyak 80 kasus sampel dengan rata-rata umur adalah 7,7 ± 3.4
tahun, lama rawat inap 4.59 hari, dan sebanyak 28.75% pasien mengalami
DSS. Dari analisis multivariat didapatkan bahwa letargi (p= 0.00; OR 21.23 ),
anggota gerak dingin ( p=0.00;OR=22.35), perdarahan(p= 0.01; OR 0,04),
dan peningkatan hematocrit( p=0.01;OR=4.72) memiliki hubungan yang
signifikan dengan DSS. Indikator lain tidak berhubungan secara signifikan
dengan DSS.
Kesimpulan: Pasien yang datang dengan letargi, perdarahan, dan peningkatan
hematokrit perlu diberikan penanganan segera dan intensif karena
berkemungkinan mengalami DSS menjadi lebih besar dibanding yang tidak
memiliki tanda tersebut.
Kata Kunci: dengue, prediksi DSS, pediatri

Introduction
Dengue is an arthropod-borne virus which has 158.192 cases in 2009.5
been long believed to come from genus Flaviviridae The considerable incidence is partly if not mostly,
with four serotypes: from DENV-1 until DENV-4 but caused by the lack of equal distribution of healthcare
new studies have found new dengue serotypes is in Indonesia. In a rural area such as Sampang, a small
DENV-5, with its main host is human.1,2 Dengue fe- city in Madura island, the diagnostic modality for
ver incidence has been on the rise in recent decades. Dengue is limited to only simple laboratory tests and
While the exact numbers are underreported, an esti- clinical observation. Other advanced and sophisticated
mated 3.9 billion people are at risk of infection annu- examinations such as NS1, Ultrasonography, Blood
ally. Reported cases exceed 100 million with 25.000 Gas Assay, and X-ray are either very limited or un-
deaths annually. Main contributors of those numbers available.
are from tropical and sub-tropical countries including Moreover, low level of education and lack of
Indonesia.3,4 In Indonesia, the Health Ministry has health knowledge amongst the population also con-
reported that the number of dengue outbreaks in Indo- tributed to the difficulty of diagnosis because most of
nesia was the largest in ASEAN region and there is a the patient presenting at the outpatient clinic is already
significant rise in dengue cases from 58 in 1968 to at a higher grade of Dengue Fever. Record by the lo-

J Indon Med Assoc, Volum: 69, Nomor: 4, April 2019 179

Predictor of Dengue Shock Syndrome Among Pediatric Dengue Infection

cal health department showed that Sampang had an Table 1. Grading Dengue Infection Severity 13
increasing dengue fever incidence over the last four
DF/DHF Symptoms Laboratory
years. From 271 in 2012 up to 639 in 2015.6 Such events
lead to significant morbidity and mortality. DF* Fever with two or more Leukopenia, trombosi-
Although most of the dengue cases are asymp- of the following sign: topenia, mab be present
tomatic and self-limiting, without proper anticipation headache, retro-orbital no plasma leakage
pain, myalgia, atralgia,
and early diagnosis, Dengue could rapidly develop into
plus positive tourniquet
Dengue Shock Syndrome (DSS). This condition po- test
tentially leads to significant morbidity and mortality.7 DHF I DF sign + positive tourniquet Thrombocytopenia
Despite the similar studies available which assess test <100,000/ µL, hematoc-
rit rise >20%
severity predictor of dengue fever, dengue fever is
DHF II Above sign + spontaneus Thrombocytopenia
known to vary widely between countries because dif- bleeding <100,000/ µL, hematoc-
ferent population which undertook the disease (par- rit rise >20%
ticular host genetics are associated with increased DHF III Above signs plus circu- Thrombocytopenia
latory failure (weak pulse, <100,000/ µL, hematoc-
susceptibility to certain infections) and different den-
hypotension, restlessness) rit rise >20%
gue serotype that commonly infect the local popu- DHF IV Profound shock with un- Thrombocytopenia
lace.8–11 Moreover, there is lack of studies available detectable blood pressure <100,000/ µL, hematoc-
in Indonesia regarding the matter. With an established and pulse rit rise >20%
early predictor in DSS, clinicians may give appropri- *
DHF III and DHF IV consider as DSS
ate treatment beforehand, hindering the disease’s pro- (Dengue Shock Syndrome)
gression and its ability to cause death.12
The potential mortality that it may cause without version 17. Measures of frequency, mean, median,
prompt treatment, the significant number of incidence maximum and minimum value was used to summa-
in Indonesia, and the lack of studies available regard- rize and describe all of the data. P value < 0.05 was
ing the topic in Indonesia summed the importance of accepted as significant. To find the predictor of se-
this study to treat dengue patient better. Therefore, vere dengue infection in pediatric patients, basic de-
this study aims to find the essential dengue signs and mographic data, simple laboratory results, and warn-
simple laboratory test’s values that can be used as a ing signs were collected as an independent variable
predictor of dengue fever’s progression into DSS. We and analyzed using logistic regression for multivari-
hope this study’s result can be used as an early, simple, ate analysis to find its relation with DSS and non-DSS
and reliable predictor of dengue shock syndrome, es- group. Simple laboratory results that will be analyzed
pecially in limited resources setting. are: leucopenia (WBC < 3500), thrombocytopenia (Plt
< 100.000), and High Hematocrit (Hct> 41%). Basic
Materials and Methods demographic data that will be analyzed using multi-
This is a retrospective cross-sectional study re- variate analysis are: gender (male or female) and age
viewing the medical record of pediatric patient with (<5 years and >5 years).
dengue infection from 1 January 2016 to 31 March
2016 in Sampang General Hospital. In Sampang gen- Results
eral hospital, maximum age for patients to be admit- A total of 80 pediatric patients with dengue infec-
ted to the pediatric ward is 14 years old. Data that tion were found. All of the necessary data were col-
were collected are basic demographic data (age and lected and analyzed. Characteristics of the patients
gender), diagnosis (Dengue Infection Severity and DSS are elaborated in Table 2. The sex group was distrib-
or non-DSS), dengue symptom, and simple labora- uted evenly between males and females, 40 patients
tory results (hematocrit, white blood cell (WBC) for each gender. The mean age of the subjects were
count, and platelet count). The ‘dengue symptom’ as- 7,7 ± 3.4 years old, ranging from 0.5 - 14 years old.
sessed in this study are based on the warning signs Most of the subject was 6 years old (n= 6; 15%). To
elaborated by WHO 2009 (Abdominal pain, vomiting portray the severity of the dengue infection, dengue
with low intake, lethargy, bleeding sign, and decrease fever grades were also described in table 2. Most of
in urine production) and other important signs of den- the patients were diagnosed with Dengue
gue fever (cold extremity, arthralgia or myalgia, and Haemorrhagic Fever (DHF) (n= 74; 92.5%) instead
headache).12 Simple laboratory signs that were as- of Dengue Fever (n= 6; 7.5%) and, overall, most pa-
sessed are also based on the critical laboratory values tients were diagnosed with DHF grade II (n= 38;
for dengue grading.13 47.5%). Subjects that had Dengue Shock syndrome
The data extracted were analyzed using SPSS were 23 (28.75%).

180 J Indon Med Assoc, Volum: 69, Nomor: 4, April 2019

 Predictor of Dengue Shock Syndrome Among Pediatric Dengue Infection

Table 2. Patient Demographic Table 3. Patient’s Blood Profile

Parameters Value (n = 80) Parameter Mean Minimum Maximum

(n=80) Value Value
Age distribution 7,7 ± 3.4 years (0.5 - 14 years) (n=80) (n=80)
Sex distribution Male = 40 (50%)
Female = 40 (50%) Smallest Hct 34.35 26.6 42.3
Dengue Fever grades Dengue Fever = 6 (7.5%) Largest Hct 41.3 29.6 51.8
DHF Grade 1 = 13 (16.3%) Days of Largest Hct 4.36 2 7
DHF Grade 2 = 38 (47.5%) Initial Platelet Counts 39,853 3,650 111,000
DHF Grade 3 = 20 (25.0%) Platelet at Discharge 99,250 44,000 280,000
DHF Grade 4 = 3 (3.8%) Initial WBC count 4,938 1,600 14,100
Dengue Shock Syndrome non-DSS = 57 (71.25%) Length of Hospitali- 4.59 2 8
DSS = 23 (28.75%) zation

Blood profile of dengue patient is in Sampang age, 4.59 days, the least was 2 days, and the longest
General hospital are elaborated in Table 3. Smallest was eight days.
Hematocrit (Hct) value on the patient’s had a mean Complete data of dengue’s symptom frequency in
of 34.35% with minimum and maximum value of dengue patient is elaborated on table 4. The percent-
26.6% and 42.3% respectively. The largest Hct value age elaborated in the table is the percentage of patient
had a mean of 41.3% with a minimum and maximum with the symptom among DSS/non-DSS patient divided
value of 29.6% and 51.8% respectively. Hct reached by those without the symptom among DSS/non-DSS
its peak for an average of 4.36 days, minimum and patient. From the data acquired, the most common sign
maximum values are two and seven days respectively. among all pediatric dengue patients was vomiting and
Initial platelet count upon admission had a mean of low intake (n= 53; 66,25%) and abdominal pain (n=
39,853 cells/mm3 with minimum and maximum value 41; 51,25%). Meanwhile, among DSS patient, the most
of 3,650 cells/mm3 and 111,000 cells/mm3 respec- common symptom was lethargy (n= 15; 65,21%) and
tively. Platelet at discharge was much better averag- vomiting (n=10; 43,74%). Other presenting signs were
ing at 99,250 cells/mm3, and minimum and maximum uncommon among DSS patient (less than 50%). Among
value were 44,000 cells/mm3 and 280,000 cells/mm3 non-DSS patients, the most common symptom was
respectively. Meanwhile, initial WBC count averaged vomiting and low intake in 43(75,43%) patients. From
at 4,938 cells/mm3 with minimum and maximum this research, we also found that most patients do not
value of 1,600 cells/mm3 and 14,100 cells/mm3 re- present with any cold extremities. Cold extremities
spectively. Length of Hospitalization was, on aver- only happen in 7 out of 80 patients (8,8%) but the symp-

Table 4. Dengue’s Sign and Symptom Frequency

Frequency (percentage)
Signs and Among DSS Among non- p Odds 95% CI
Symptom Patient DSS Patient Value Ratio Lower Upper

Demographic Age (<5 years) 1 (4,34%) 14 (24,56%) 0,07 0,14 0,26 1,91
Gender (Male) 10 (43,47%) 30 (52,63%) 0,49 0,70 0,02 1,14
Vomiting and 10 (43,47%) 43 (75,43%) 0,74 0,58 0,02 14,94
Low Intake
Abdominal Pain 7 (30,43%) 34 (59,64%) 0,79 1,53 0,07 33,90
Lethargy 15 (65,21%) 4 (7,01%) 0,00* 21,33 6,89 66,08
Bleeding 6 (26,08%) 46 (80,7%) 0,01* 0,04 0,00 0,39
Symptom Cold Extremities 6 (26,08%) 1 (1,75%) 0,00* 22,35 7,97 25,05
Arthralgia 1 (4,34%) 7 (12,28%) 0,64 2,44 0,06 101,84
Headache 4 (17,39%) 16 (28,07%) 0,25 5,79 0,29 116,66
Decreased Urine 0 (0%) 2 (3,51%) 1,00 0,00 0,00 0,00
Production
Lab Result Severe Thrombo- 18 (78,26%) 27 (47%) 0,28 1,99 0,57 6,88
cytopenia (<50.000)
Leucopenia 3 (13,04%) 25 (43,85%) 0,09 0,29 0,07 1,19
(<3.500)
High Hematocrit 18 (78,26%) 21 (36,84%) 0,01* 4,72 1,46 15,27
(>41%)

* = significantly related to DSS (p< 0,05)

J Indon Med Assoc, Volum: 69, Nomor: 4, April 2019 181

Predictor of Dengue Shock Syndrome Among Pediatric Dengue Infection

tom manifested mostly in DSS patient (n=6; 26,08%) in the age range of 5-9 years old.15
compared to non-DSS patients (n= 1; 1,75%). Each country or specific area commonly accen-
tuated different parameter, because the manifestation
From the multivariate analysis with confidence of dengue itself varied and is associated with host-
interval of 95%, we divide all the factors into 3 major agent-environment triad which differs in every re-
group: demographic, symptom, and laboratory result. gion.8–11,16 Other studies have found that pregnancy,17
Demographic group were divided into aged below 5 two extremes of age,15,18 secondary infection,19 nutri-
years old (p=0.07;OR=0.14) and gender (p=0.49; tion status,20 and presence of diabetes or hypertension
OR=0.07), no significancy with DSS were found in in adults as risk factors for more severe dengue dis-
this group. Vomiting and low intake (p=0.79; OR=0.58), eases.21 Compared to this study, pregnancy, extremes
abdominal pain (p=0.79; OR=1.53), lethargy(p=0.00; of age, and the presence of diabetes or hypertension
OR=21.23), bleeding (p=0.01; OR=0.04), cold extremi- are irrelevant because this study focuses on the pedi-
ties (p=0.00; OR=22.35), arthralgia (p=0.64; OR=2.44), atric patient which mostly are not affected by those
headache (p=0.25; OR=5.79), and decreased urine factors. Nutrition status and secondary infection can
production (p=0.00; OR=0.00) were inside symptom’s be assessed, but the data is not available on the medi-
group. We found that lethargy, cold extremities, and cal record in this study.
bleeding was significantly related to DSS. It is also Further study on the pediatric patient should in-
important to note that by reading the OR, lethargy and clude those data whenever possible. In Singapore, they
cold extremities can be used as the predictor of DSS used clinical bleeding, low serum protein, high serum
because its OR is higher than one. Meanwhile, bleed- urea, and low lymphocyte proportion as an indicator
ing is a predictor of non-DSS because its OR is less for progression to DHF but this is applied to adult pa-
than one. tients only.22,23
The laboratory result composed of severe throm- Similar to this study, the authors identified two
bocytopenia (p=0.28; OR=1.99), leucopenia (p=0.09; other similar kinds of research that also assessed the
OR=0.29), and high hematocrit (p=0.01; OR=4.72). dengue signs that are related to the outcome of pediat-
Based on this data we found that high haematocrit ric patients. Phuong et al. in 2004 assessed 700 pedi-
was significant as DSS predictor. atric patients in Vietnam. The study found gastrointes-
tinal symptom, hepatomegaly and rise in hematocrit
Discussion until more than 50% was a predictive sign of DSS.
This study found that there is an equal amount of The result differed from this research’s result. This
male and female infected by dengue. This record por- phenomenon can be caused by a difference in popula-
trayed the similar possibility of both gender for being tion, dengue serotype, and environment. Potts JA et
infected by dengue. Such finding is similar to the re- al24 assessed 1348 pediatric patients in Thailand and
sult of a study by Anker M et al.14 The study assessed found that WBC count, monocyte percentage, hemat-
male-female differences in the number of reported ocrit, and platelet count for identifying severe cases
incident dengue fever cases in six Asian countries over of Dengue with 97% sensitivity but only 48% speci-
ten years. For pediatric patients (aged less than 14), ficity. The result of this study showed that the signs
they found the inconsistent difference and small dif- that significantly relate to DSS are lethargy, bleeding,
ferences for both genders on 5 out of 6 countries stud- cold extrimities, and increase haematocrite level which
ied. The result is also not surprising because Aedes differed from the two similar studies which may be
aegypti, the dengue virus primary vector, are active caused due to the difference host-agent-environment
during the daytime and pediatric patients tend to have triad as mentioned previously.
similar activities during the day between both sexes This study would help to enlighten the problem of
which is going to school. Therefore, exposing both predicting DSS among pediatric dengue patient but
genders on a similar risk of being infected by dengue. predicting the outcome of dengue infection remains a
The majority of pediatric patients in Sampang challenge. Further studies are necessary especially
General Hospital are either DHF grade 2 (47.5%) or those that could cover the limitation of this study. We
grade 3 (20%). Severity in dengue cases in pediatric hope this preliminary result could help in early diag-
patients is typical because they have less capability nosis, timely treatment, and prevention of deaths in
to compensate for plasma leakage compared to pediatric dengue patient.
adults.12 From comparing the subject’s age, we found
that the majority of pediatric dengue patient was 6-7 Conclusion
years old (n=23; 28.1%). Other studies also find a simi- Dengue infection is common in children especially
lar result: most severe dengue infection were found around the age of 7 years old with a similar incidence

182 J Indon Med Assoc, Volum: 69, Nomor: 4, April 2019

Predictor of Dengue Shock Syndrome Among Pediatric Dengue Infection

between males and females. Vomiting and low intake diagnosis, treatment, prevention and control: new edition
[Internet]. World Health Organization; 1997. Available from:
is the most common dengue sign amongst overall den-
http://whqlibdoc.who.int/publications/1997/
gue patient and also amongst non-DSS patients. Den- 9241545003_eng.pdf
gue signs of lethargy, cold extremities, and increase 14. Anker M, Arima Y. Male-female differences in the number
in haematocrit level were significant in predicting of reported incident dengue fever cases in six Asian coun-
tries. West Pacific Surveill Response [Internet]. 2011 Jul
dengue shock syndrome in the pediatric patient.
5;2(2):e1–e1. Available from: http://ojs.wpro.who.int/ojs/
index.php/wpsar/article/view/118/37
Acknowledgments 15. Hammond SN, Balmaseda A, Pérez L, Tellez Y, Saborío SI,
Mercado JC, et al. Differences in dengue severity in infants,
All authors have seen and approve the final manu-
children, and adults in a 3-year hospital-based study in Nica-
script. CC designed the outline concept of the research ragua. Am J Trop Med Hyg [Internet]. 2005 Dec;73(6):1063–
and wrote the initial draft. JAS revised and expanded 70. Available from: http://www.ncbi.nlm.nih.gov/pubmed/
the manuscript. 16354813
16. Yacoub S, Wills B. Predicting outcome from dengue. BMC
Med [Internet]. 2014;12(1):147. Available from: http://
References bmcmedicine.biomedcentral.com/articles/10.1186/s12916-
1. Bhatt S, Gething PW, Brady OJ, Messina JP, Farlow AW, 014-0147-9
Moyes CL, et al. The global distribution and burden of den- 17. Machado CR, Machado ES, Rohloff RD, Azevedo M, Cam-
gue. Nature [Internet]. 2013;496(7446):504–7. Available from: pos DP, de Oliveira RB, et al. Is Pregnancy Associated with
http://dx.doi.org/10.1038/nature12060. Severe Dengue? A Review of Data from the Rio de Janeiro
2. Centers for Disease Control and Prevention. Dengue and the Surveillance Information System. Halstead SB, editor. PLoS
Aedes aegypti Mosquito [Internet]. 2012 [cited 2018 Oct 8]. Negl Trop Dis [Internet]. 2013 May 9;7(5):e2217. Available
Available from: https://www.cdc.gov/dengue/resources/ from: http://dx.plos.org/10.1371/journal.pntd.0002217
30Jan2012/aegyptifactsheet.pdf. 18. Lee M-S, Hwang K-P, Chen T-C, Lu P-L, Chen T-P. Clinical
3. Gubler DJ. Dengue and dengue hemorrhagic fever. Clin characteristics of dengue and dengue hemorrhagic fever in
Microbiol Rev [Internet]. 1998 Jul;11(3):480–96. Available a medical center of southern Taiwan during the 2002 epi-
from: http://www.ncbi.nlm.nih.gov/pubmed/9665979. demic. J Microbiol Immunol Infect [Internet]. 2006
4. World Health Organization (WHO). The Dengue Strategic Apr;39(2):121–9. Available from: http://www.ncbi.nlm.nih.
Pain for the Asia Pacific Region 2008-2015. 2008;50. Avail- gov/pubmed/16604244
able from: http://www.searo.who.int/entity/vector_borne_ 19. Guzman MG, Alvarez M, Halstead SB. Secondary infection
tropical_diseases/documents/WP-DP8-15/en/ as a risk factor for dengue hemorrhagic fever/dengue shock
5. Kementerian Kesehatan RI. Buletin Jendela Epidemiologi. syndrome: an historical perspective and role of antibody-
Jakarta; 2010. dependent enhancement of infection. Arch Virol [Internet].
6. Dinas Kesehatan Kabupaten Sampang. Laporan Bulanan 2013 Jul 8;158(7):1445–59. Available from: http://
Program P2M Demam Berdarah Dengue. Sampang; 2016. link.springer.com/10.1007/s00705-013-1645-3
7. 2010. Srikiatkhachorn. Dengue Hemorrhagic Fever the sen- 20. Kalayanarooj S, Nimmannitya S. Is dengue severity related
sitivity and specificity of WHO definition in identifying se- to nutritional status? Southeast Asian J Trop Med Public
vere dengue cases in thailand.pdf. Health [Internet]. 2005 Mar;36(2):378–84. Available from:
8. Lan NTP, Hirayama K. Host genetic susceptibility to severe http://www.ncbi.nlm.nih.gov/pubmed/15916044
dengue infection. Trop Med Health [Internet]. 2011;39 (4 21. Pang J, Salim A, Lee VJ, Hibberd ML, Chia KS, Leo YS, et
SUPPLEMENT):S73–81. Available from: http://joi.jlc. al. Diabetes with Hypertension as Risk Factors for Adult
jst.go.jp/JST.JSTAGE/tmh/2011-S08?from=CrossRef Dengue Hemorrhagic Fever in a Predominantly Dengue
9. Rico-Hesse R. Dengue Virus Virulence and Transmission Serotype 2 Epidemic: A Case Control Study. Halstead SB,
Determinants. In 2010. p. 45–55. Available from: http:// editor. PLoS Negl Trop Dis [Internet]. 2012 May 1;6(5):e1641.
link.springer.com/10.1007/978-3-642-02215-9_4 Available from: http://dx.plos.org/10.1371/journal.pntd.
10. Thammapalo S, Chongsuvivatwong V, Geater A, Dueravee 0001641
M. Environmental factors and incidence of dengue fever 22. Lee VJ, Lye DC, Sun Y, Leo YS. Decision tree algorithm in
and dengue haemorrhagic fever in an urban area, Southern deciding hospitalization for adult patients with dengue
Thailand. Epidemiol Infect [Internet]. 2008 Jan 15;136(1). haemorrhagic fever in Singapore. Trop Med Int Heal
Available from: http://www.journals.cambridge.org/ [Internet]. 2009 Sep;14(9):1154–9. Available from: http://
abstract_S0950268807008126 doi.wiley.com/10.1111/j.1365-3156.2009.02337.x
11. Trang NTH, Long NP, Hue TTM, Hung LP, Trung TD, 23. Lee VJ, Leo Y-S, Thein TL, Lye DC, Sun Y. Validation of
Dinh DN, et al. Association between nutritional status and Probability Equation and Decision Tree in Predicting Subse-
dengue infection: a systematic review and meta-analysis. quent Dengue Hemorrhagic Fever in Adult Dengue Inpa-
BMC Infect Dis [Internet]. 2016;16(1):172. Available from: tients in Singapore. Am J Trop Med Hyg [Internet]. 2011
http://bmcinfectdis.biomedcentral.com/articles/10.1186/ Nov 1;85(5):942–5. Available from: http://www.ajtmh.org/
s12879-016-1498-y content/journals/10.4269/ajtmh.2011.11-0149
12. World Health Organization. Dengue: guidelines for diagno- 24. Potts JA, Gibbons R V., Rothman AL, Srikiatkhachorn A,
sis, treatment, prevention, and control. Spec Program Res Thomas SJ, Supradish P on, et al. Prediction of dengue dis-
Train Trop Dis [Internet]. 2009;x, 147. Available from: http:/ ease severity among pediatric Thai patients using early clini-
/whqlibdoc.who.int/publications/2009/978924154 cal laboratory indicators. PLoS Negl Trop Dis. 2010;4(8):2–
7871_eng.pdf 8.
13. World Health Organization (WHO). Dengue guidelines for

J Indon Med Assoc, Volum: 69, Nomor: 4, April 2019 183