Changing pattern of symptoms in dengue patients over the years:A review and meta-analysis

Objective:To describe the pattern of vomiting,diarrhoea and abdominal pain among dengue patients over the years in different age groups.Methods:Electronic databases were used to include studies that reported symptoms of vomiting,diarrhoea or/and abdominal pain.Confirmation of dengue was conducted by laboratory diagnosis.The exclusion criteria were case reports and studies which only described specific settings,focused only on dengue haemorrhagic fever or severe dengue and co-infection.Meta-analysis was performed on the overall prevalence of vomiting,diarrhoea and abdominal pain which was then stratified according to years and different age groups.Results:Out of the 2002 studies,32 studies were included.All studies were conducted between 1995 and 2014.The studies were conducted in 18 countries with a total sample size of 69466.The predominant symptom was abdominal pain before 2004 but became vomiting thereafter.The prevalence of vomiting was constant throughout the years(20%-40%),but the prevalence of abdominal pain reduced from 30%-50%to 20%-30%.The prevalence of diarrhoea increased steadily from 5%to 30%.Abdominal pain was the predominant symptom in children(75.03%)[95%confidence interval(CI):64.89%-85.17%],while diarrhoea was predominant in adults(32.42%)(95%CI:23.96%-40.88%).For studies which included all ages,the predominant symptom was vomiting(45.18%)(95%CI:33.87%-56.49%).Conclusions:The knowledge of symptomatic pattern change could assist clinicians in diagnosing and managing dengue patients.Future studies could address the occurrence of this change and the difference in symptoms between children and adults.

Biomarkers in differentiating clinical dengue cases:A prospective cohort study

Objective:To evaluate five biomarkers(neopterin,vascular endothelial growth factor-A,thrombomodulin,soluble vascular cell adhesion molecule 1 and pentraxin 3)in differentiating clinical dengue cases.Methods:A prospective cohort study was conducted whereby the blood samples were obtained at day of presentation and the final diagnosis were obtained at the end of patients’follow-up.All patients included in the study were 15 years old or older,not pregnant,not infected by dengue previously and did not have cancer,autoimmune or haematological disorder.Median test was performed to compare the biomarker levels.A subgroup Mann-Whitney U test was analysed between severe dengue and non-severe dengue cases.Monte Carlo method was used to estimate the 2-tailed probability(P)value for independent variables with unequal number of patients.Results:All biomarkers except thrombomodulin has P value<0.001 in differentiating among the healthy subjects,non-dengue fever,dengue without warning signs and dengue with warning signs/severe dengue.Subgroup analysis for all the biomarkers between severe dengue and non-severe dengue cases was not statistically significant except vascular endothelial growth factor-A(P<0.05).Conclusions:Certain biomarkers were able to differentiate the clinical dengue cases.This could be potentially useful in classifying and determining the severity of dengue infected patients in the hospital.

The accuracy of newly proposed warning signs in the third edition of Malaysian guideline on the management of dengue infection in adult

Objective:To evaluate the accuracy of persistent vomiting and persistent diarrhoea proposed in the Malaysian dengue guideline.Methods:The eligibility criteria were dengue patients with vomiting or/and diarrhoea at the earliest day of illness without other warning signs.Vomiting and diarrhoea categorised by three or more episodes over 24 h were considered as warning signs.Sensitivity,specificity,positive predictive value and negative predictive value were calculated.Results:There were a total of 479 patients,of whom 36(7.5%)got severe dengue and 443(92.5%)got dengue with warning signs.The overall accuracy ranged from 33.3%to 72.7%for sensitivity,28.8%to 56.5%for specificity,1.8%to 14.5%for positive predictive value and 88.5%to 96.3%for negative predictive value.There were 16 false negative.Conclusions:The newly proposed warning signs in the third edition of Malaysian dengue guideline may not be effective in predicting severe dengue.