Dengue infection is one of the most common mosquito-borne diseases of public health significance. Currently dengue infection affects about 1% of the world’s population annually and is associated with high morbidity.1–4 The clinical manifestations of dengue infection range from asymptomatic to undifferentiated fever, an influenza-like symptom known as dengue fever (DF) and a severe, sometimes fatal disease characterized by massive plasma leakage, hemorrhage and shock known as dengue hemorrhagic fever (DHF) and dengue shock syndrome (DSS). Due to advanced comprehensive care technology, the mortality rate among patients with DHF in Thailand has progressively declined from 13.7% in 1958 to 0.13% in 2018, and has remained unchanged (below 0.2%) during the last decade.5 Nevertheless, high-risk patients with multi-organ failure involving acute liver failure, acute respiratory failure, acute kidney injury, active massive bleeding and fluid overload are at risk of lethal DHF/DSS with a higher mortality rate up to 15% to 44%.1,6–12 and almost 50% to 100% in DSS with massive uncontrolled bleeding or multi-organ failure.11 In addition, studies among Thai children have found that infants and obese children had a higher risk of having severe dengue infection.13–16 Also, patients with underlying diseases such as thalassemia and hemophilia are prone to aggravated hemolytic anemia17,18 and bleeding manifestations,19 respectively.
The clinical diagnosis of DHF is based on four major characteristic manifestations19: (1) high continuous fever lasting for 2 to 7 days; (2) hemorrhagic tendency such as positive tourniquet test, petechiae, epistaxis or intestinal bleeding; (3) thrombocytopenia (platelet count <100,000 µ/L) and (4) evidence of plasma leakage due to increased vascular permeability manifested by hemoconcentration (an increase in hematocrit >20%) and pleural effusion or ascites. The three stages of DHF are febrile, toxic and defervescence. The febrile stage lasts 2 to 7 days followed by an abrupt fall to normal or subnormal levels of temperature; the toxic stage lasts 24 to 48 h; and finally, rapid clinical recovery without sequelae in the defervescence stage. The toxic stage is the most critical period from prominent plasma leakage leading to DSS and requiring intensive supportive care. Optimal fluid therapy is essential to maintain vital organ functions during the critical period. According to the 1997 World Health Organization (WHO) committee classification of dengue infection20 varying from DF to DHF grade I–IV was precisely concluded after the defervescence emphasizing the pathophysiologic alteration of the clinical manifestations. In the 2009 WHO classification, dengue infection was defined as either dengue with or without warning signs, or severe dengue including severe plasma leakage, severe hemorrhage and severe organ impairments.21 It has been used to aid appropriate management in decision making regarding outpatient care and hospitalization criteria.
This study was designed to develop a daily dengue severity score to assess severe manifestations among patients with dengue infection in a hospitalized setting.
SUBJECTS AND METHODS
The medical records of hospitalized children admitted for dengue by serologic confirmation between 2004 and 2018 were retrospectively reviewed. The study consisted of 2 phases. Phase I established the daily dengue score system and assessment tool for severe manifestations, followed by phase II testing the accuracy of the severity score. This study was approved by the Ramathibodi Hospital Ethics Committee for Human Research.
In all, 355 children 1 month to 18 years of age with serologic confirmation of dengue infection (DF,102; DHF grade I, 88; DHF grade II, 103; DHF grade III, 46; and DHF grade IV, 16) were hospitalized at the Department of Pediatrics, Faculty of Medicine, Ramathibodi Hospital, Mahidol University from 2004 to 2013. All patients were treated by an expert multidisciplinary team using Ramathibodi Clinical Practice Guidelines.22–24 The complete medical records of 191 cases (DF, 35; DHF grade I, 53; DHF grade II, 50; DHF grade III, 37; and DHF grade IV 16) from 355 patients were included in the phase I study to establish daily dengue severity score. Then, 51 additional hospitalized patients with DHF grades II, III, and IV from 2014 to 2018 were included in phase II study to test the accuracy of the severity score.
Data collection included baseline characteristics (age, sex, duration of fever, and duration of hospitalization), complete blood cell counts and bleeding sites. The 14 items related to clinical parameters and managements, shown in Table, Supplemental Digital Content 1, http://links.lww.com/INF/D712, were entered in the computerized program using Microsoft Access to obtain the daily severity score during hospitalization until discharge or death.
According to the pathophysiology of dengue infection, the day of defervescence was designated as Day 0 (D0). One and 2 days before defervescence were designated as Day −1 (D −1) and Day −2 (D −2) and so on. One and 2 days after defervescence were designated as Day +1(D +1) and Day +2 (D +2). Designating the day of illness related to the day of defervescence is useful in recruiting patients in the same pathophysiology in the same group because the febrile stage of dengue infection lasts 2 to 7 days.
The dengue nonstructural protein antigen 1 (NS1) was identified by strip test in 125 patients of a total of 242 patients and was positive in 84 of 125 patients (67.2%). All patients had positive serologic studies of dengue-specific IgM/IgG performed by enzyme-linked immunosorbent assay (ELISA).
WHO Dengue Classification
Dengue infection was classified as DF and DHF according to the WHO (1997) criteria by two physicians (KT, AC) from clinical data, laboratory investigation and/or a right lateral view of chest radiograph obtained after defervescence.
Database was analyzed using the Statistical Package for Social Sciences Software (SPSS, Version 12.0). Categorical variables were analyzed using χ2 or Fisher exact tests. Daily dengue severity score was assessed using the Area under the Receiver Operating Characteristic (AUROC) curves. Continuous severity scores were compared using multilevel analysis. The predictive ability of a score for subsequent DSS was analyzed and presented with sensitivity, specificity and positive likelihood ratio.
The baseline characteristics of the studied patients from 2004 to 2013 in the phase I study are shown in Table 1. Their mean (SD) age was 10.7 (3.8) years with 102 males and 89 females. Patients with DHF grades I and II were significantly older than those with DF, DHF grades III and IV. Duration of fever in each group did not significantly differ but DHF grades III and IV had significantly longer hospitalizations than those of the other groups. The initial complete blood counts at admission showed higher white blood cell counts and lower platelet counts among patients with DHF grades III and IV compared with those of the other groups. However, the levels of initial hematocrit did not significantly differ between patients with DF and DHF. The levels of hematocrit among patients with DHF gradually increased to levels of more than 20% of their baseline except for those with massive bleeding. Importantly, patients with DHF had significantly lower platelet counts during hospitalization compared with those of patients with DF (29,000–51,000/µL vs. 86,000/µL, P = 0.002).
Concerning bleeding manifestations, patients with DHF exhibited frequent bleeding episodes while those with DF rarely exhibited bleeding except for petechiae. The logistic regression analysis showed that epistaxis was commonly found among patients with DHF grade II (odd ratio 4.47, 95% CI 1.45–15.58, P = 0.01) but gastrointestinal hemorrhage was commonly found among patients with DHF grades III and IV (odd ratio 102, 95% CI 10.35–1005.23, P = 0.01) compared with those of patients with DF. The case-fatality rate was 4.7% (9/191) found among patients with DHF grade IV only. All fatal patients had experienced prolonged shock with uncontrolled massive bleeding and multi-organ failure before being referred for comprehensive management at the Pediatric Intensive Care Unit, Faculty of Medicine Ramathibodi Hospital.
A total of 593 daily dengue severity scores from Day −4 to Day +2 among patients with DF and DHF grades I to IV were recorded and analyzed. The median number of daily score was 3.0 with an interquartile range of 2 to 4. The mean daily dengue severity score during hospitalization related to defervescence is shown in Figure 1. The analysis revealed that the range of mean daily dengue severity scores among patients with DHF grade III (10–20) and DHF grade IV (31–47) were significantly higher than those of DF (5–13), DHF grade I (2–10) and DHF grade II (6–11) with P-values of 0.0001. The ability of dengue severity score to assess the subsequent threatened shock and profound shock in DHF grades III and IV known as DSS using AUROC curves was the score on Day −1 compared with other days as shown in Table 2. Using the validity test, dengue severity score of ≥12 was used as an assessment tool for the severe manifestations of DSS with a sensitivity of 86.2%, specificity of 84.3% and positive likelihood ratio of 5.7 as shown in Table 3.
Then the phase II study recruited a total of 149 hospitalized patients (DF 77, DHF grade I, 21; II 29; III 18 and IV 4) from 2014 to 2018 treated by a multidisciplinary expert team using Ramathibodi Clinical Practice Guidelines. Their mean (SD) age was 10.1 (4.4) years. The case-fatality rate was 2.0% (3/149) found among patients with DHF grade IV only. A total of 190 daily dengue severity scores from Day −3 to Day +2 from patients with DHF grades II, III and IV (35 males, 16 females,) were recruited to test the accuracy of daily dengue severity scores. The median number of daily scores was 4 with an interquartile range of 3 to 5. The number of patients, who exhibited severe manifestations including DHF grades III and IV (n = 21) and DHF grade II with severe hemorrhage, severe plasma leakage and risk factor of age ≤1 year or obesity (n = 7) and having the daily score of ≥12, was significantly higher than those of DHF grade II without severe manifestations (n = 23). This included from Day −3 to Day +1 (Day −3, 3/5 vs. 0/5, P = 0.038; Day −2, 8/10 vs. 0/12, P < 0.0001; Day −1, 11/17 vs. 1/18, P < 0.0001, Day 0, 24/28 vs. 2/23, P < 0.0001 and Day +1, 11/26 vs. 0/21, P = 0.001). Therefore, the daily dengue severity scores of ≥12 was proven as an accurate assessment tool for severe manifestations among patients with DHF starting from the febrile to the toxic stage.
Dengue virus infection is a clinical diagnosis and is subsequently confirmed by virus isolation, the dengue NS1 antigen and serologic testing requiring a single sample or paired sera from acute and convalescent stages. Dengue virus isolation or dengue virus amplification will be positive during the first 1–2 days of fever requiring advanced technology and is unavailable in general hospital services. Recently, determining the dengue nonstructural protein 1 (NS1) antigen in the serum and urine is helpful to initially confirm dengue virus infection during the febrile stage.25,26 However, patients with positive dengue NS1 may exhibit mild manifestations of DF or the more severe manifestations of DHF.
Several attempts were made to find out a suitable marker to predict patients at risk of subsequent severe manifestations of bleeding, plasma leakage, threatened shock and profound shock during the febrile and toxic stages. In 2006, Butthep et al27 studied 111 patients (25 DF, 78 DHF and 8 with other febrile illnesses) 4 to 16 years of age revealed that elevated soluble thrombomodulin acted as an early predictor of DSS during the febrile stage. However, testing to determine soluble thrombomodulin is routinely unavailable in hospital services. The routine laboratory tests of complete blood counts and coagulation tests of activated partial thromboplastin time, prothrombin time and thrombin time were analyzed among the same group of 111 patients. The results revealed any 1 of the following abnormal laboratory findings during the febrile stage served as predictors for risk of DSS: increased hematocrit >25%, platelet count <40,000/µL, activated partial thromboplastin time >44 seconds, prothrombin time >14 seconds and thrombin time >16 seconds with the relative risk ranging from 6.7 to 13.6.28
In 2013, Pongpan et al29 developed a dengue infection severity score using the data of 777 patients with dengue infection (391 DF, 296 DHF, 90 DSS) from 3 hospitals located in northern Thailand. The results revealed that the significant clinical predictors of severe dengue at the initial assessment included age >6 years, hepatomegaly, hematocrit >40%, systolic pressure <90 mm Hg, white blood cell count >5,000/µL, and platelet count ≤50,000/µL with maximum scores of 1, 8.5, 1, 2, 1 and 4.5 for each item, respectively. The cutoff scores of >11.5 could predict DSS in 39 of 90 cases (43.3%).
The current dengue severity score has been used at the university hospital setting with comprehensive multidisciplinary experts among intensive care, infectious, hematology, renal, cardiology and paramedical personnel such as nurses, medical technologists and scientists to handle patients with severe dengue and various complications. It included comorbid risk factors of the patients individually such as age ≤1 year, aspirin or nonsteroidal drug ingestion, underlying disease such as hemolytic anemia and congenital heart disease. Additional vital signs, urine output, bleeding sites, amounts of the required crystalloid, colloid and blood components, inotropic drug administration, respiratory support and invasive procedures were included. The maximum scores of the top 3 parameters were 30 of 100 scores consisting of maximum pulse rate, lowest systolic blood pressure and pulse pressure, and maximum urine output. It reflected the adequacy of perfusion, cardiovascular status, severity of organ dysfunction and responsiveness to comprehensive management. It acted as a clinical monitoring for the subsequent complications of severe manifestations which is important for the readiness of the intensive care unit, staff on duty, essential medications, equipment and adequate blood components28 to manage these patients with serious clinical manifestations and organ failure.
A daily dengue severity score can be obtained using computerized programs or other equivalent tools. It can also be performed manually on a worksheet depending on the hospital facilities. It has been tested among hospitalized pediatric patients over the previous 5-year period. The cutoff score of ≥12 could accurately assess severe manifestations among pediatric patients with DHF. An increase in severity score reflected the severe manifestations requiring intensive management. However, a decrease in severity score alternatively revealed clinical improvement requiring less intensive therapy. Therefore, daily dengue severity score was a useful clinical outcome assessment and the cutoff of ≥12 served as an accurate assessment tool for severe manifestations.
The authors wish to thank all staff of the expert multidisciplinary team and nurses in pediatric wards and the pediatric intensive care unit of the Faculty of Medicine, Ramathibodi Hospital.
1. Rigau-Pérez JG, Clark GG, Gubler DJ, et al. Dengue and dengue haemorrhagic fever. Lancet. 1998;352:971–977.
2. Anderson KB, Chunsuttiwat S, Nisalak A, et al. Burden of symptomatic dengue infection in children at primary school in Thailand: a prospective study. Lancet. 2007;369:1452–1459.
3. Gulati S, Maheshwari A. Atypical manifestations of dengue. Trop Med Int Health. 2007;12:1087–1095.
4. Halstead SB. Halstead SB. Dengue: overview and history. In: Dengue Tropical Medicine: Science and Practice. 2008:London: Imperial College Press; 1–28.
6. Ong A, Sandar M, Chen MI, et al. Fatal dengue hemorrhagic fever
in adults during a dengue epidemic in Singapore. Int J Infect Dis. 2007;11:263–267.
7. Lee IK, Liu JW, Yang KD. Clinical and laboratory characteristics and risk factors for fatality in elderly patients with dengue hemorrhagic fever
. Am J Trop Med Hyg. 2008;79:149–153.
8. Wang CC, Liu SF, Liao SC, et al. Acute respiratory failure in adult patients with dengue virus infection. Am J Trop Med Hyg. 2007;77:151–158.
9. Laoprasopwattana K, Pruekprasert P, Dissaneewate P, et al. Outcome of dengue hemorrhagic fever
-caused acute kidney injury in Thai children. J Pediatr. 2010;157:303–309.
10. Chuansumrit A, Tangnararatchakit K, Lektakul Y, et al. The use of recombinant activated factor VII for controlling life-threatening bleeding in dengue shock syndrome
. Blood Coagul Fibrinolysis. 2004;15:335–342.
11. Laoprasopwattana K, Chaimongkol W, Pruekprasert P, et al. Acute respiratory failure and active bleeding are the important fatality predictive factors for severe dengue viral infection. PLoS One. 2014;9:e114499.
12. Schmitz L, Prayag S, Varghese S, et al. Nonhematological organ dysfunction and positive fluid balance are important determinants of outcome in adults with severe dengue infection: a multicenter study from India. J Crit Care. 2011;26:441–448.
13. Kliks SC, Nimmanitya S, Nisalak A, et al. Evidence that maternal dengue antibodies are important in the development of dengue hemorrhagic fever
in infants. Am J Trop Med Hyg. 1988;38:411–419.
14. Kalayanarooj S, Nimmannitya S. Clinical presentations of dengue hemorrhagic fever
in infants compared to children. J Med Assoc Thai. 2003;86(suppl 3):S673–S680.
15. Kalayanarooj S, Nimmannitya S. Is dengue severity related to nutritional status? Southeast Asian J Trop Med Public Health. 2005;36:378–384.
16. Pichainarong N, Mongkalangoon N, Kalayanarooj S, et al. Relationship between body size and severity of dengue hemorrhagic fever
among children aged 0-14 years. Southeast Asian J Trop Med Public Health. 2006;37:283–288.
17. Pongtanakul B, Narkbunnam N, Veerakul G, et al. Dengue hemorrhagic fever
in patients with thalassemia. J Med Assoc Thai. 2005;88(suppl 8):S80–S85.
18. Tanphaichitr VS, Chonlasin R, Suwantol L, et al. Effect of red blood cell glucose-6-phosphate dehydrogenase deficiency on patients with dengue hemorrhagic fever
. J Med Assoc Thai. 2002;85(suppl 2):S522–S529.
19. Chuansumrit A, Tangnararatchakit K, Sirachainan N, et al. Dengue virus infection in haemophilic patients: aggravation of bleeding risk. Haemophilia. 2011;17:553–556.
20. World Health Organization. Dengue Hemorrhagic Fever
: Diagnosis, Treatment, Prevention and Control. 1997:2nd ed. Geneva: WHO; 12–47.
21. World Health Organization. Dengue Hemorrhagic Fever
: Guidelines for Diagnosis, Treatment, Prevention and Control. 2009:Geneva: WHO; 25–58.
22. Lolekha S, Varavithya W. Ruangkanchanasetr S, Chunharas A, Ruangdaraganon N. Dengue hemorrhagic fever
. In: Ambulatory Pediatric 2. 2004:2nd edition. Bangkok: Holistic Publishing; 306–311.
23. Ruangkanchanasetr S, Chongviriyaphan N, Sutabutra P; Ramathibodi Clinical Practice Guideline. Dengue hemorrhagic fever
. In: Pediatrics: Guideline for Diagnosis and Treatment Volume II. 2004:Bangkok: Bjorn Enterprise; 576–584.
24. Chuansumrit A, Tangnararatchakit K. Pathophysiology and management of dengue hemorrhagic fever
. TATM. 2006;8(suppl.1):3–11.
25. Chuansumrit A, Chaiyaratana W, Pongthanapisith V, et al. The use of dengue nonstructural protein 1 antigen for the early diagnosis during the febrile stage in patients with dengue infection. Pediatr Infect Dis J. 2008;27:43–48.
26. Chuansumrit A, Chaiyaratana W, Tangnararatchakit K, et al. Dengue nonstructural protein 1 antigen in the urine as a rapid and convenient diagnostic test during the febrile stage in patients with dengue infection. Diagn Microbiol Infect Dis. 2011;71:467–469.
27. Butthep P, Chunhakan S, Tangnararatchakit K, et al. Elevated soluble thrombomodulin in the febrile stage related to patients at risk for dengue shock syndrome
. Pediatr Infect Dis J. 2006;25:894–897.
28. Chuansumrit A, Puripokai C, Butthep P, et al. Laboratory predictors of dengue shock syndrome
during the febrile stage. Southeast Asian J Trop Med Public Health. 2010;41:326–332.
29. Pongpan S, Wisitwong A, Tawichasri C, et al. Prognostic indicators for dengue infection severity. Int J Clin Pediatr. 2013;2:12–18.