|Year : 2021 | Volume
| Issue : 4 | Page : 186-191
A clinico epidemiological analysis of dengue deaths in children during outbreaks in the year 2013 and 2017: A retrospective observational study from a tertiary care teaching hospital in South India
Sheeja Sugunan, A Santhosh Kumar, R Rekha Krishnan, Reshma Manayankath
Department of Pediatrics, SAT Hospital, Government Medical College, Thiruvananthapuram, Kerala, India
|Date of Submission||09-Mar-2020|
|Date of Decision||24-May-2021|
|Date of Acceptance||30-May-2021|
|Date of Web Publication||10-Jul-2021|
Dr. Sheeja Sugunan
Department of Pediatrics, SAT Hospital, Government Medical College, Thiruvananthapuram, Kerala
Source of Support: None, Conflict of Interest: None
Background: Majority of published studies have looked at the risk factors for mortality in dengue. There are limited data on critical analysis of the causes of death in children with dengue when they are treated according to the World Health Organization guidelines. This study was conducted to analyze the immediate causes and epidemiology of deaths in children in two different outbreaks of dengue.
Subjects and Methods: This was a retrospective observational study conducted at tertiary care teaching hospital during outbreaks in the years 2013 and 2017. All children who died due to Dengue during outbreaks were included in the study. A retrospective chart review was done using predesigned pro forma.
Results: A total of 29 children died due to Dengue in two outbreaks. There were Eighteen deaths in the year 2013 and eleven deaths in the year 2017 outbreak. Myocarditis (27%), Encephalitis (24%), Massive hemorrhage (13%), and severe capillary leak (13%) were major immediate causes of death. Myocarditis (45%) was the leading cause of death in 2017 while Dengue encephalitis and massive bleed (23%) each in 2013. Hypoalbuminemia at admission was seen in 26 (89%) patients at admission among all deaths. Aspartate transaminase (AST) >1000 IU/L was seen in 10 (91%) Children and had a sensitivity of 91% and specificity of 50% in diagnosing myocarditis.
Conclusions: The primary cause of death in fatal dengue had shown changing epidemiology during two outbreaks. Hypoalbuminemia can predict the beginning of the critical phase better than the defervescence of fever. AST >1000 IU/L may point towards myocarditis.
Keywords: Dengue deaths, encephalitis, epidemiology, immediate causes, myocarditis
|How to cite this article:|
Sugunan S, Kumar A S, Krishnan R R, Manayankath R. A clinico epidemiological analysis of dengue deaths in children during outbreaks in the year 2013 and 2017: A retrospective observational study from a tertiary care teaching hospital in South India. J Pediatr Crit Care 2021;8:186-91
|How to cite this URL:|
Sugunan S, Kumar A S, Krishnan R R, Manayankath R. A clinico epidemiological analysis of dengue deaths in children during outbreaks in the year 2013 and 2017: A retrospective observational study from a tertiary care teaching hospital in South India. J Pediatr Crit Care [serial online] 2021 [cited 2021 Sep 19];8:186-91. Available from: http://www.jpcc.org.in/text.asp?2021/8/4/186/321099
| Introduction|| |
Dengue is the most rapidly spreading mosquito-borne disease. In the last 50 years, the incidence of dengue has increased 30-fold worldwide. It is estimated that globally 390 million dengue infections occur per year with 90 million clinical cases Southeast Asia has the highest reported incidence of Dengue with epidemics occurring every 3–5 years. An estimated 500,000 people with severe dengue require hospitalization each year and about 2.5% of those affected die.
Dengue is endemic in India and is a leading cause of hospitalization and death in children. The reported case fatality rate for the Southeast Asia region is 1%, but outbreaks in India have reported a case fatality rate of 3%–5% World Health Organization (WHOs) global strategy for prevention and control of Dengue aims at reducing Dengue mortality by 50% by 2020. Severe dengue is a disease with protean manifestations and varying degrees of multi-organ involvement. There is a paucity of data on critical analysis of causes of death in dengue when children are treated according to the WHO protocol. This study was conducted to analyze the immediate causes and epidemiology of deaths in children in two different outbreaks of dengue.
| Materials and Methods|| |
This was a retrospective observational study conducted in Pediatric Intensive Care Unit at a tertiary care teaching hospital. The study was approved by the institutional ethics committee. All children who died due to Dengue during the Dengue outbreaks of the year 2013 and year 2017 were included. A retrospective chart review of confirmed Cases were done using predesigned Proforma.
Pertinent data regarding patient demographics, clinical presentation, prereferral fluid therapy, type and quantity of fluid received after admission, blood and blood products received, serial laboratory parameters, number of organ involved, immediate and underlying causes of death and treatment provided were recorded. Fluid overload percentage was calculated for all children who survived for more than 24 h. The duration of illness was determined from the onset of fever. In all children blood cultures, dengue immunoglobulin M (IgM) and NS1 were sent at admission. Organ functions, serum electrolytes, and serum albumin were sent daily in all cases. Co-infection with Leptospirosis and Scrub was ruled out by IgM ELISA in all cases who presented with fever of more than 5 days duration. Causes of death were analyzed by two independent investigators and consensus reached after deliberations.
Patients were considered to have Dengue with warning signs if they had abdominal pain, persistent vomiting, clinical fluid accumulation (CFA), mucosal bleed, lethargy or restlessness, liver enlargement >2 cm, or hematocrit rise of more than 20% with rapid fall in platelet count. The WHO 2009 definitions were used to identify children with severe Dengue and classify stages. Minor changes in definitions were done for better clarity for study purposes.Severe capillary leak was considered when rise in hematocrit of >20% with hypotension (blood pressure <5th centile for age) or narrow pulse pressure of <20 mm of Hg and pleural effusion with respiratory distress (respiratory rate [RR] >60/mt in <2 months, >50/mt in 2 months to 12-month, RR >40 in children 1–5 years age group and >30/min in children more than 5 years of age) or spo2 <92% in room air. Acute liver failure was defined with Aspartate transaminase (AST) or Alanine transaminase (ALT) >1000 IU/L and acute renal injury defined as doubling of creatinine from baseline or decline in GFR by more than 50%.
Dengue myocarditis was defined as a reduction in left ventricular ejection fraction, evidence of arrhythmias, or elevated cardiac enzymes like Trop T or CPK MB with clinical evidence of cardiac failure. Dengue encephalitis was defined as persistently altered sensorium with Glasgow Coma Scale (GCS) <12, seizures with persistently altered sensorium for more than 1 h, presence of abnormal computed tomography (CT) or magnetic resonance imaging findings or cerebrospinal fluid (CSF) dengue IgM or polymerase chain reaction positivity.
The ascertainment of the Cause of death was done according to the guidelines recommended by the WHO. The primary cause of death was considered severe capillary leak if patients had a refractory shock with high hematocrit at the time of death. Myocarditis was considered as the primary cause of death if patients had two-dimensional echocardiography evidence of cardiac dysfunction (ejection fraction <50%) with refractory shock or if the cause of death was arrhythmia. Encephalitis was considered as the primary cause in patients who had a GCS <3, at the time of death after ruling out other causes for such low GCS.
Categorical variables are expressed as proportions and quantitative variables as mean and standard deviation. Nominal categorical data between the groups were compared using the Chi-square test or Fisher's exact test as appropriate. For all statistical tests, P < 0.05 was considered significant.
| Results|| |
A total of 29 children died due to dengue in two outbreaks of the study period. Eighteen out of 952 (1.8%) and 11 out of 1474 (0.74%) children died due to Dengue in the year 2013 and 2017 respectively. There was 59% reduction in the case fatality rate.
Children, more than 5 years of age (median age 10 years) accounted for 23 (79%) deaths, with a male-to-female ratio of 1.9:1. Duration of illness before admission was 3 days in 4 (14%) cases, 4 days in 5 (17%) cases, 5 days in 15 (52%) cases, and 6 days in 4 (14%) cases. Fourteen (48%) children were febrile at the time of admission. Vomiting was the commonest warning sign which was present in 18 (62%) patients. Warning signs were present in all cases who died due to massive hemorrhage, severe capillary leak, or myocardial dysfunction, and 5 (71%) cases who died due to Dengue encephalitis. Twelve (41%) children presented in profound shock with absent peripheral pulses [Table 1].
|Table 1: Clinical and laboratory profile of patients who died due to dengue|
Click here to view
Thrombocytopenia <20,000 cells/mm3 was rare in children who died due to Dengue. Nine (31%) cases had platelet count >50,000 cells/mm3, 17 (59%) had platelet count between 20,000–50,000 cells/mm3, while 3 (10%) had platelet count <20000 cells/mm3. After the fluid bolus, in 20 (69%) cases the change in hematocrit was <20% while only in 2 (7%) cases the fall in hematocrit by more than 20% was found and in 7 (24%) cases there was a rise in hematocrit.
Eight (27%) had Myocarditis, 7 (24%) had Encephalitis (24%) 4 (13%) each had massive hemorrhage and severe capillary leak, 3 (10%) had resorption pulmonary edema, 3 (10%) had secondary infection including 2 (7%) with ARDS The major immediate causes of deaths [Table 2]. and comparison of causes of death is shown in [Figure 1].
|Figure 1: Primary causes of death in children with Dengue during 2013 and 2017|
Click here to view
Eleven (38%) children had evidence of myocarditis, but it was considered as the primary cause of death in 8 (28%) cases. In the rest of the 3 cases of myocarditis, blood pressure and cardiac contractility had improved by day 3 of admission. All 3 had renal failure and died due to pulmonary edema during the resorptive phase. Four (36%) patients had arrhythmia (1 ventricular tachycardia, 1 junctional rhythm, 1 severe sinus bradycardia, and 1 ventricular fibrillation). Nine (82%) patients had ejection fraction <50% including 7 with ejection fraction <35%. Cardiac enzymes like Troponin T, Troponin I or CPK MB were found to be elevated only in 4 (36%) cases with clinical evidence of myocardial involvement. The median troponin T level was 0.51 ng/ml (interquartile range [IQR] 0.33–3.26) and the mean CPK-MB level was 25.08 (severe dengue [SD] 15.1) IU/L. 10 (91%) children with clinical myocarditis had AST >1000 IU/L and AST >1000 IU/L had a sensitivity of 91% and specificity of 50% in diagnosing myocarditis.
Seven (24%) children died because of encephalitis. Six Out of these seven children became comatosed, with GCS 3 within 3 h of admission. Four children also developed diabetes insipidus. Two had hypertension, 2 had normal blood pressure and 3 had hypotension at admission. CT scan was done in four cases. Two cases showed evidence of brain edema without any evidence of hemorrhage or infarct, while one had bilateral thalamic involvement, one had infarct in the bilateral cerebellar hemisphere and left anterior temporal region. CSF study was done only in 2 cases. CSF Dengue IgM were positive in both, with no cells and normal biochemistry.
Seventeen (59%) deaths occurred within 48 h of admission, which includes four cases of massive hemorrhage and severe capillary leak each, five cases (71%) of encephalitis, and four cases (50%) of myocarditis. Refractory shock was observed in 13 cases including 4 cases of massive hemorrhage, 4 cases of severe capillary leak, and 5 cases of myocarditis.
All 4 children who died of massive bleed and 6 (75%) children who died of myocarditis presented in profound shock with absent peripheral pulses at admission and developed massive pulmonary and GI bleed during the administration of fluid bolus by push-pull method. None of the children who died due to severe hemorrhage had received prereferral intravenous fluid treatment. The median fluid overload percentage (FO%) was 8.17%. FO % was least for children with neurological complications (3.76%) and highest for children with severe capillary leak (16%). Children with myocarditis had a median FO% of 8.15%. Children who died due to pulmonary edema of varying etiology (severe capillary leak, cardiogenic pulmonary edema, resorption pulmonary edema with or without renal failure) had a higher FO% ([9.3% vs. 6.6%] P = 0.13) compared to those who did not have pulmonary edema.
Three children had secondary infection including 1 case of Methicillin-resistant Staphylococcus aureus septicemia secondary to abscess in the nail bed and 2 cases of ARDS. Both children with ARDS had initially improved with normalization of blood pressure and improvement in serum albumin and organ functions by day 3. But later developed a hospital-acquired pneumonia and succumbed. Three children died of resorption pulmonary edema. All three had evidence of acute kidney injury (AKI). One child was on peritoneal dialysis, one was on slow continuous ultrafiltration, while the third child had AKI stage 2. All the 3 children were off intravenous fluids, had shown stabilization of vital signs with improvement in platelet count and serum albumin, and had new-onset pulmonary edema at the time of death.
The predicted mortality rate by PIM 3 score at admission was low for children who subsequently died due to severe capillary leak and encephalitis (11.5% and 16.5%). Children who died due to myocarditis had a high predicted mortality rate of 65%.
| Discussion|| |
Dengue is a disease with protean manifestations with varying immediate and underlying causes of death. Consistent with published data, in our study, the median age of children who died due to dengue was 10 years (IQR5) with a male to female ratio of 1.9:1.
In our study, 14% of children did not have warning signs. Woon et al. also observed the absence of warning signs in 21.9% of fatal Dengue cases, while Thein et al. reported absence of warning signs in 58% of patients with severe Dengue. The difference was attributed to different patient groups as the study by Thein et al. also included survivors of Severe Dengue. We observed that warning signs were present in all children who died of massive bleed, severe capillary leak or myocarditis, while 5 (71%) of those with encephalitis had warning signs (P = 0.08). Further studies are needed to identify more sensitive warning signs in Dengue Encephalitis.
Vomiting was the commonest warning sign identified (63%). Thein et al. in their study found persistent vomiting to have high specificity (93%) in predicting severity. Carrasco et al. and Sreenivasan P et al. also found persistent vomiting as an important predictor for severity. Vomiting usually occurs due to abdominal wall edema, hepatic involvement, mesenteric ischemia due to shock, electrolyte imbalance, or CNS involvement. Fifty-two percent of children had CFA at admission. CFA is a late warning sign and it manifests after significant plasma leakage has occurred either due to the disease itself or augmented by the administration of IV fluids. Thein et al. reported a high specificity (98%) for CFA in predicting SD. Thanachartwet et al. also reported high sensitivity (75%) and specificity (90%) of CFA in diagnosing SD.
The critical phase is usually characterized by the defervescence of fever with a rise in hematocrit and a fall in platelet count. In our study, 48% of children were febrile at admission, while 89% had hypoalbuminemia and 59% had packed cell volume more than 40% with platelet count <1 lakh/mm3 at admission. Hypoalbuminemia was found to be a better predictor of the onset of the critical phase (P < 0.01). Branco et al. also observed hypoalbuminemia in all patients who died of severe Dengue infection. The study by Ong et al. from Singapore found all patients with fatal Dengue infection to be febrile at admission. Further studies are needed to identify if the presence of fever after the onset of the leaky phase is a risk factor for mortality.
In our study, 91% (10) of children with Dengue myocarditis had AST >1000 IU/L. AST >1000 IU/L had a sensitivity of 91% and specificity of 50%. In the study by Li et al. prevalence of myocarditis in hospitalized Dengue was 11.28%, while Dengue with warning signs and severe Dengue had a prevalence of 46.67%. They also observed a significantly higher AST and ALT in children with myocarditis compared to patients without myocarditis. Six (85.7%) children with Dengue encephalitis were comatosed with GCS 3 within 3 h of admission. The two cases of dengue encephalitis leading to death reported by Osnaya et al. also had evidence of uncal herniation following seizures at admission. In children with Dengue who present with seizure, the possibility of dengue encephalitis leading to severe brain edema and herniation should be anticipated and supportive measures to lower raised ICP should be provided early. The low predicted mortality rate by PIM 3 score for children with Dengue encephalitis compared to those who died due to Dengue myocarditis highlights the varied disease progression trajectory of Dengue.
When were compared two outbreaks, there has been a 59% reduction in the case fatality rate in 2017 compared to 2013. There was an improvement in the stabilization rate at the local hospital of children admitted in shock (47% vs. 56%) over the years, but the difference was not statistically significant. There was also a significant reduction in death within 8 h of admission between the two outbreaks (0% vs. 22%). Health education activities during previous outbreaks might have contributed to better stabilization rates in 2017. The proportion of death due to severe capillary leaks has remained the same during the two outbreaks. Myocarditis (28%) and encephalitis (25%) were the leading primary causes of death. There was an increase in the number of deaths due to myocarditis (17% vs. 45%) in the year 2017, while deaths due to massive hemorrhage (22% vs. 0%) and secondary infections (17% vs. 0%) showed a decreasing trend. Whether the change in Dengue serotype and genotype also contributed to change in presentation is not known. Dengue serotypes 2 and 3 were the predominant serotype in 2013. Investigations had revealed a genetic shift from the erstwhile predominant DENV2/DENV3 serotypes to a newly introduced DENV1 Asian genotype in the year 2017.
In 2017, fluid boluses were not administered by the push-pull method instead allowed to run in over 30 min to 1 h. This may have contributed to a decrease in the incidence of massive bleed in 2017 due to decreased shear pressure on the capillary endothelium. Early presentation and better stabilization at the peripheral hospital may have also contributed to a decrease in the incidence of massive bleed in the 2017 outbreak. Whether serotype change has also contributed to the decrease in the incidence of massive bleed is not known.
Small sample size, retrospective nature of the study, single center observations, and absence of a matched cohort are the important limitations of this study.
| Conclusions|| |
The primary cause of death in fatal dengue cases showed changing epidemiology during different outbreaks. Hypoalbuminemia can predict the onset of the critical phase better than the absence of fever. Children with AST more than 1000 IU/l point toward myocarditis. Children with profound shock are more likely to die of massive bleed or myocardial dysfunction.
Early detection of warning signs and initiation of stabilization in peripheral hospitals is of paramount importance to decrease morbidity and mortality. The absence of warning signs in significant proportion of patients with dengue encephalitis highlights the need for the identification of new warning signs in this subgroup of patients. These findings need to be confirmed in larger prospective multicentric studies.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
World Health Organization. Global Strategy for Dengue Prevention and Control, 2012-2020. Geneva, Switzerland: World Health Organization; 2012.
Shepard DS, Undurraga EA, Halasa YA, Stanaway JD. The global economic burden of dengue: A systematic analysis. Lancet Infect Dis 2016;16:935-41.
WHO 2009. Dengue: Guidelines for Diagnosis, Treatment, Prevention and Control. Geneva: World Health Organization; 2009.
World Health Organization. Medical Certification of Cause of Death: Instructions for Physicians on Use of International form of Medical Certificate of Cause of Death. Geneva: World Health Organization; 1979.
Mishra S, Ramanathan R, Agarwalla SK. Clinical profile of dengue fever in children: A study from Southern Odisha, India. Scientifica (Cairo) 2016;2016:6. Article ID 6391594.
Woon YL, Hor CP, Hussin N, Zakaria A, Goh PP, Cheah WK. A two-year review on epidemiology and clinical characteristics of dengue deaths in Malaysia, 2013-2014. PLOS Negl Trop Dis 2016;10:e0004575.
Thein TL, Leo YS, Fisher DA, Low JG, Oh HM, Gan VC, et al
. Risk factors for fatality among confirmed adult dengue inpatients in Singapore: A matched case-control study. PLoS One 2013;8:e81060.
Carrasco LR, Leo YS, Cook AR, Lee VJ, Thein TL, Go CJ, et al
. Predictive tools for severe dengue conforming to World Health Organization 2009 criteria. PLoS Negl Trop Dis 2014;8:e2972.
Sreenivasan P, Geetha S, Sasikala K. Development of a prognostic prediction model to determine severe dengue in children. Indian J Pediatr 2018;85:433-9.
Thanachartwet V, Oer-Areemitr N, Chamnanchanunt S, Sahassananda D, Jittmittraphap A, Suwannakudt P, et al
. Identification of clinical factors associated with severe dengue among Thai adults: A prospective study. BMC Infect Dis 2015;15:420.
Branco Mdos R, Luna EJ, Braga Júnior LL, Oliveira RV, Rios LT, Silva Mdo S, et al
. Risk factors associated with death in Brazilian children with severe dengue: A case-control study. Clinics (Sao Paulo) 2014;69:55-60.
Ong A, Sandar M, Chen MI, Sin LY. Fatal dengue hemorrhagic fever in adults during a dengue epidemic in Singapore. Int J Infect Dis 2007;11:263-7.
Li Y, Hu Z, Huang Y, Li J, Hong W, Qin Z, et al
. Characterization of the Myocarditis during the worst outbreak of dengue infection in China. Medicine (Baltimore) 2016;95:e4051.
Osnaya-Romero N, Perez-Guille MG, Andrade-García S, Gonzalez-Vargas E, Borgaro-Payro R, Villagomez-Martinez S, et al
. Neurological complications and death in children with dengue virus infection: Report of two cases. J Venom Anim Toxins Incl Trop Dis 2017;23:25.
Ahamed SF, Rosario V, Britto C, Dias M, Nayak K, Chandele A, et al
. Emergence of new genotypes and lineages of dengue viruses during the 2012-15 epidemics in southern India. Int J Infect Dis 2019;84S:S34-43.
Kumar PN, Anish TS, Valamparampil MJ, Thomas AT, Abidha, Mathew J, et al
. Genotype shift of dengue virus (DENV1) during the 2017 outbreak of dengue fever in Thiruvananthapuram, Kerala, India. Indian J Exp Biol 2019;57:961-6.
[Table 1], [Table 2]