• Users Online: 111
  • Print this page
  • Email this page

 Table of Contents  
Year : 2021  |  Volume : 8  |  Issue : 5  |  Page : 224-228

Risk factors of severe dengue in children: A nested case-control study

Department of Pediatrics, Government Kilpauk Medical College, Chennai, Tamil Nadu, India

Date of Submission22-Jul-2021
Date of Decision21-Aug-2021
Date of Acceptance24-Aug-2021
Date of Web Publication28-Sep-2021

Correspondence Address:
Dr. Sridevi A Naaraayan
Government Kilpauk Medical College, Poonamalle High Road, Kilpauk, Chennai - 600 010, Tamil Nadu
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/jpcc.jpcc_59_21

Rights and Permissions

Background: Dengue is a mosquito-borne viral disease whose severity varies from a mild undifferentiated fever to severe fatal form. The objectives of this study were to determine the prevalence and risk factors of severe dengue.
Subjects and Methods: This nested case–control study was done in a medical college hospital in South India from October 2019 to January 2020. Children between 1 month and 12 years of age who presented with fever without focus and tested positive for dengue serology were included. Demographic data, anthropometry, clinical features, and laboratory parameters were noted down from all children. Severity of illness was classified according to the World Health Organization dengue guidelines 2009. Prevalence of severe dengue was expressed in proportion with 95% confidence interval (CI). Bivariate analysis was performed using Chi-square test, Fisher's test, and Student's t-test, and a P < 0.05 was considered statistically significant.
Results: Out of 370 children, 192 (51.9%) had dengue without warning signs, 168 (45.4%) had dengue with warning signs, and 10 had severe dengue. The prevalence (95% CI) of severe dengue was 2.7% (1.3%–4.9%). Abdominal pain/tenderness (odd's ratio [OR] - 7.429), facial puffiness (OR - 4.333), hepatomegaly (OR - 3.857), and pleural effusion (OR - 13.222) were the significant risk factors for severe dengue.
Conclusions: The significant risk factors for severe dengue were abdominal pain/tenderness, facial puffiness, hepatomegaly, and pleural effusion.

Keywords: Children, dengue shock, risk factors, severe dengue

How to cite this article:
Naaraayan SA, Sundar KC. Risk factors of severe dengue in children: A nested case-control study. J Pediatr Crit Care 2021;8:224-8

How to cite this URL:
Naaraayan SA, Sundar KC. Risk factors of severe dengue in children: A nested case-control study. J Pediatr Crit Care [serial online] 2021 [cited 2021 Oct 16];8:224-8. Available from: http://www.jpcc.org.in/text.asp?2021/8/5/224/326867

  Introduction Top

Dengue is a mosquito-borne viral infection which is found in tropical and sub-tropical climates worldwide, mostly in urban and semiurban areas. Around 100–400 million infections are estimated to occur worldwide each year.[1] During 2019, National Vector Borne Disease Control Program reported more than 150,000 laboratory-confirmed cases of dengue in the country.[2] The severity of dengue ranges from subclinical disease to severe life-threatening illness. The World Health Organization (WHO) 2009 guidelines of dengue classify the illness as dengue without warning signs, dengue with warning signs, and severe dengue.[3]

Clinical symptoms and signs such as severe abdominal pain/tenderness, persistent vomiting, facial puffiness, mucosal bleed, hepatomegaly, lethargy, and laboratory parameters such as increase in hematocrit count with rapid drop in platelet count have been identified as warning signs in dengue infection.[2] It is noteworthy that only some but not all patients who develop these warning signs progress to severe dengue. It is essential to identify the risk factors for severe dengue so that children with these risk factors may be closely monitored.

The prevalence of severe dengue varies depending on the setting, the type of hospital, and the population it caters to. The primary objective of this study was to determine the prevalence of severe dengue in a tertiary care institution, and the secondary objective was to identify the risk factors for severe dengue in children.

  Materials and Methods Top

This nested case–control study was done in the pediatric intensive care unit and ward of a government multispecialty hospital attached to a medical college in South India from October 2019 to January 2020. The study was commenced after obtaining approval from the Institutional Ethics Committee (No. 259/2019 dated November 7, 2019). All children of age 1 month to 12 years who were admitted for fever were methodically examined and those without septic focus were tested for IgM dengue by enzyme-linked immunosorbent assay on or after day 5 of fever. This is the standard unit protocol followed during postmonsoon epidemic season of the year. Those children who had fever without focus and positive dengue serology were included in the study. Children who were discharged before clinical recovery and those with coinfections were excluded. Since the prevalence of severe dengue will vary depending on the type of healthcare facility and population it caters to, the pilot study was done to calculate the sample size. Sample size calculation was done using Epi Info software CDC Epi info. With a prevalence of severe dengue of 5% observed in pilot study and with an error margin of 2.2% to calculate 95% confidence interval (CI), the number of samples needed was 370.

After obtaining written informed consent from parent and assent from the child, children satisfying inclusion criteria were recruited in the study. The demographic data including age and gender, complaints including fever and its duration, persistent vomiting for more than 48 h, and abdominal pain were noted. The height/length of the child and his/her weight were noted down using standardized methods. Anthropometry was interpreted using the WHO 2006 growth chart for children up to 5 years of age and the Indian Academy of Pediatrics 2015 growth charts for children above 5 years of age.[4],[5] Physical examination findings including facial puffiness, petechiae, abdominal tenderness, and liver span were documented. Any child with liver span more than 2 cm above the normal value for age was considered to have hepatomegaly. All children were subjected to detailed physical examination including all vital parameters, and the variables were fitted in pediatric assessment triangle and their physiological status was determined.[6] The relevant investigations recorded were total count, platelet count, hematocrit, chest X-ray, and transaminases level. Case management was in line with the WHO dengue guidelines 2009. Patients were followed up every day for development of new symptoms, signs, or complications till discharge/death. Blood counts were repeated every 48 h till defervescence of fever and 24 h after defervescence. Blood counts done 24 h after defervescence were considered for analysis.

Subsequently, patients were grouped into three categories – those with dengue without warning signs, dengue with warning signs, and severe dengue as per the WHO dengue guidelines 2009. The primary outcome was presence or absence of severe dengue, and the secondary outcome was risk factors for severe dengue. Data were entered in Microsoft excel and analyzed using IBM SPSS version 23. Categorical data were presented as proportion and numerical data as mean and standard deviation (SD). The prevalence of severe dengue was expressed in proportion with 95% CI. Patients with severe dengue were considered as cases. For each case of severe dengue, the next two patients of dengue with warning signs were considered as controls. Thus, a nested case–control study was carried out with cases and controls in the ratio of 1:2.[7] The presence of risk factors in these two groups was compared and odds ratio (OR) was calculated. Since the objective was to identify which warning signs are more commonly associated with severe dengue, demographic variables and the warning signs of dengue were considered as risk factors. The risk factors considered included age, sex, nutritional status, duration of fever, abdominal pain/tenderness, persistent vomiting, facial puffiness, petechiae, mucosal bleeds, hepatomegaly, pleural fluid, hematocrit >40%, platelet count <100,000, total count <4000, and elevated transaminases (more than twice the reference values). Distribution of numerical variables was checked by ShapiroWilk test. The association between numerical variables and outcome variables was determined using Student's t-test and categorical variables and outcome variables was determined by Chi-square test/Fisher's exact test. A p value < 0.05 was considered statistically significant.

  Results Top

A total of 945 children were admitted with fever during the given period. Out of them, 370 dengue patients were included in the study. The patient flow is given in [Figure 1]. The mean (SD) age of the patients was 7.63 (3.14) years. Two hundred and sixty-six (72.5%) patients were aged more than 5 years. There were 187 (50.5%) males. Two hundred and eighty-two (76.2%) children had normal body mass index while 78 (21.1%) had under-nutrition and 6 (1.6%) were overweight. All children (100%) had fever with a mean (SD) fever duration of 5.59 (2.37) days. One hundred and seventeen (31.7%) children had abdominal pain/tenderness, while 77 (20.8%) had persistent vomiting. A total of 38 (10.3%) children had hepatomegaly, 6 (1.6%) each had facial puffiness and petechiae, and 3 (0.8%) had mucosal bleed in the form of melena. The proportion of children with various warning signs is depicted in [Figure 2]. Two hundred and eight (56.2%) children had leukopenia, while 12 (3.2%) of them had leukocytosis. Platelet count was normal in 65 (17.6%) children, while 62 (16.8%) had mild thrombocytopenia, and 243 (65.7%) had platelet count <1 lakh. A total of 60 (16.2%) had raised hematocrit and 40 (10.8%) showed chest X-ray findings, suggestive of pleural effusion. Raised serum glutamic-oxaloacetic transaminase (SGOT) was observed in 156 (42.1%) children, raised serum glutamic pyruvic transaminase (SGPT) was observed in 56 (15.1%), while both were raised in 53 (14.3%) children. All children were discharged after clinical recovery, and no mortality was noted. The mean (SD) length of stay was 5.3 (1.8) days.
Figure 1: Patient flow

Click here to view
Figure 2: Patients with various danger signs

Click here to view

Out of 370 patients with dengue, 192 (51.9%) patients had dengue without warning signs, 168 (45.4%) patients had dengue with warning signs, and 10 patients had severe dengue; all the patients with severe dengue presented with shock. No other forms of severe dengue namely pulmonary edema, severe bleeding, and multiorgan failure were observed in the study population. The prevalence (95% CI) of severe dengue was noted to be 2.7% (1.3%–4.9%).

The results of bivariate analysis are presented in [Table 1]. Age, gender, nutritional status, duration of fever, persistent vomiting, petechiae, leukopenia, thrombocytopenia, increased hematocrit, and elevated transaminases were not found to be significantly associated with severe dengue. On the other hand, abdominal pain/tenderness (OR [95% CI] – 7.429 [1.226–45.005]), facial puffiness (OR [95% CI] – 4.333 [2.148–8.742]), hepatomegaly (OR [95% CI] – 3.857 [2.039–7.297]), and pleural effusion (OR [95% CI] – 13.222 [2.129–82.129]) emerged as significant risk factors of severe dengue in children.
Table 1: Risk factors in dengue

Click here to view

  Discussion Top

The prevalence of severe dengue in our center and its risk factors was identified. A recent systematic review reported a prevalence of severe dengue ranging from as low as 1.4% to as high as 97.4% in various studies, with overall prevalence being 28.9%.[8] The proportion depends on the study setting and the population it caters to. The prevalence reported in the study implies that though the hospital is a tertiary care center, it receives patients similar to the pattern seen in secondary care centers, with a majority of uncomplicated cases and few complicated ones. This pattern was observed since the center is a multidisciplinary hospital and not an exclusive pediatric center. All patients with severe dengue presented with shock rather than other complications, which is like the observation in previous studies that shock is more common than other severe forms of dengue.[9]

The median age of the patients in this study was around 8 years, and almost three-fourths of the patients were above 5 years of age. This is like that reported by a previous study done in pediatric population which reiterates the fact that dengue is a disease of school-going children rather than under-five children.[10] This is due to increased outdoor activity in this age group compared to younger children. A previous study has stated that age above 5 years is a risk factor for severe dengue, which differs from the finding of our study.[11] Neither the incidence nor the severity of the disease has any gender predilection as noted by our study as well as another study done in neighboring regions.[12] The finding that as much as one-fifth of children had under-nutrition and only a small proportion were overweight implied that majority of patients attending the hospital belonged to the lower socioeconomic group. With almost similar proportion of children with under-nutrition as our study, a previous study has observed that under-nutrition is not a risk factor for severe dengue in line with our study.[13]

All children (100%) presented with fever as noted in most studies.[14] The most common warning symptom noted in our study was abdominal pain like few other studies but contrary to some other studies, which observed vomiting to be the most common one.[12],[13] Persistent vomiting for more than 48 h, however, was found to be the second most common warning symptom. Hepatomegaly was noted in lesser number of children in our study compared to other studies paralleling lesser number of severe dengue cases.[12],[13],[14] Among liver enzymes, SGOT was increased in more children than SGPT in our study like others.[14] The proportion of children with pleural effusion was similar to that noted in previous studies.[15]

Abdominal pain/tenderness was found to be a warning sign, which is a risk factor for severe dengue in our study as well as previous study.[16] This symptom/sign may be considered as a manifestation of shock, occurring due to visceral hypoxia because of visceral ischemia. Clinical signs of capillary leak such as facial puffiness and pleural effusion were identified as risk factors of severe disease in previous studies too, in agreement with ours.[11],[14] In contrast, increased hematocrit which also signifies capillary leak was not found to be a risk factor which too is consistent with the previous study.[11] Hepatomegaly was found to be a significant risk factor for severe dengue in our study like few previous studies and is an important clinical sign.[10],[11],[16],[17]

The smaller number of patients with severe dengue is the major limitation of the study. Therefore, a minimum number of ten with the outcome of interest was not available for all risk factors. Due to the same reason of small numbers, logistic regression analysis was not done. However, this depends on the type of health facility and the population it caters to, as discussed previously. The effect of this shortcoming is offset, at least partly by the robust study design, i.e., a nested case–control design which is the ideal study design to study prognostic factors. Large multicenter studies can generate robust data and evidence.

Identification of risk factors is important to prioritize level of care in resource-limited setting. This can pave a way for optimum utilization of resources in addition to appropriate case management.

  Conclusions Top

Abdominal pain/tenderness, facial puffiness, hepatomegaly, and pleural effusion are risk factors of severe dengue in children, and children with these findings require close monitoring in a high-dependency unit rather than general pediatric ward during the critical phase of illness.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

  References Top

World Health Organization. Newsroom/Factsheet/Dengue and Severe Dengue. Available from: http://who.int/news-room/factsheets/detail/dengue-and-severedengue. [Last accessed on 2021 Jul 03].  Back to cited text no. 1
National Vectorborne Disease Control Program, Directorate General of Health Services. Available from: http://nvbdcp.gov.in/DENGU1.html. [Last accessed on 2021 Jul 03].  Back to cited text no. 2
World Health Organization. Dengue. Guidelines for Diagnosis, Treatment, Prevention and Control; 2009. Available from: http://who.int/tdr/publications/documents/dengue-diagnosis.pdf. [Last accessed on 2021 Jul 03].  Back to cited text no. 3
World Health Organization. Child Growth Standards. Available from: https://www.who.int/tools/child-growth-standards/standards. [Last accessed on 2021 Jul 03].  Back to cited text no. 4
Indian Academy of Pediatrics Growth Charts Committee; Khadilkar V, Yadav S, Agrawal KK, Tamboli S, Banerjee M, et al. Revised IAP growth charts for height, weight and body mass index for 5- to 18-year-old Indian children. Indian Pediatr 2015;52:47-55.  Back to cited text no. 5
Pediatric Assessment Triangle. Available from: https://www.health.ny.gov/professionals/ems/pdf/pediatricreferencecard-04.pdf. [Last accessed on 2021 Jul 03].  Back to cited text no. 6
Ernster VL. Nested case-control studies. Prev Med 1994;23:587-90.  Back to cited text no. 7
Ganeshkumar P, Murhekar MV, Poornima V, Saravanakumar V, Sukumaran K, Anandaselvasankar A, et al. Dengue infection in India: A systematic review and meta-analysis. PLoS Negl Trop Dis 2018;12:e0006618.  Back to cited text no. 8
Rosenberger KD, Alexander N, Martinezn E, Lum LC, Dempfle CE, Junghanss T, et al. Severe dengue categories as research endpoints – Results from a prospective observational study in hospitalised dengue patients. PLoS Negl Trop Dis 2020;14:e0008076.  Back to cited text no. 9
Kedia N, Mahadevan S, Kumar RR. Abstract PD-011: Risk factors for severe dengue: Retrospective observational study from a PICU in southern India. Pediatr Crit Care Med 2018;19:31.  Back to cited text no. 10
Gupta V, Yadav TP, Pandey RM, Singh A, Gupta M, Kanaujiya P, et al. Risk factors of dengue shock syndrome in children. J Trop Pediatr 2011;57:451-6.  Back to cited text no. 11
Pothapregada S, Sivapurapu V, Banupriya K, Mahalakshmy T. Role of early warning signs in children with severe dengue infection. Int J Contemp Pediatr 2018;5:1423-30.  Back to cited text no. 12
Rathod NP, Ansari NJ, Singh DK. Clinical profile of children with dengue and factors associated with severe dengue and dengue with warning signs. Pediatr Oncall J 2018;15:1-5.  Back to cited text no. 13
Mishra S, Ramanathan R, Agarwalla SK. Clinical Profile of Dengue Fever in Children: A Study from Southern Odisha, India. Scientifica. 2016; http://dx.doi.org/10.1155/2016/6391594.  Back to cited text no. 14
Majumdar I, Mukherjee D, Kundu R, Niyogi P, Das J. Factors affecting outcome in children with dengue in Kolkata. Indian Pediatr 2017;54:778-80.  Back to cited text no. 15
Hegazi MA, Bakarman MA, Alahmadi TS, Butt NS, Alqahtani AM, Aljedaani BS, et al. Risk factors and predictors of severe dengue in Saudi population in Jeddah, western Saudi Arabia: A retrospective study. Am J Trop Med Hyg 2020;102:613-21.  Back to cited text no. 16
Pothapregada S, Kamalakannan B, Thulasingham M. Risk factors for shock in children with dengue fever. Indian J Crit Care Med 2015;19:661-4.  Back to cited text no. 17
[PUBMED]  [Full text]  


  [Figure 1], [Figure 2]

  [Table 1]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  In this article
Materials and Me...
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded52    
    Comments [Add]    

Recommend this journal