OBJECTIVES: Oxidative stress in dengue viral infection has been suggested and severity of it was found to be associated with progress of illness. Hence assessing oxidative stress mediated changes in plasma proteins can be an early biomarker for prediction of severe dengue infection. DESIGN AND METHODS: Thirty two dengue hemorrhagic fever (DHF), 21 dengue shock syndrome (DSS), 27 dengue fever (DF) and 63 age and sex matched controls, were included in this study. Blood samples were collected on the 3rd day of fever. Protein carbonylation (PCOs) and protein-bound sulphydryl (PBSH) group levels were determined by spectrophotometric method and analyzed as predictor of dengue hemorrhagic fever and dengue shock syndrome. RESULTS: About 80-84% of cases presented with no signs of DHF/DSS at the time of sampling. Dengue infected individuals had significantly elevated PCOs and low PBSH group levels than the controls. Using one-way ANOVA we found a significant difference with high PCOs and low PBSH group levels between DHF and DSS when compared with DF (P<0.001). However, no difference was observed in PBSH group levels between DHF and DSS. A significant difference in PCOs to PBSH ratio was observed among DF, DHF and DSS (P<0.001). Linear regression analysis revealed that duration of hospitalization is dependent on PCOs and PBSH group levels. Receiver operator curve (ROC) analysis indicated that 5.22nmol/mg protein PCOs; 1.08 PCOs to PBSH group levels ratio were optimal cutoff value for predicting DHF with sensitivity and specificity of 87.5% and 74.1%; 96.9% and 81.5%, respectively. For DSS prediction, 6.13 nmol/mg protein PCOs; 1.16 PCOs to PBSH group levels ratio were found as effective cutoff with sensitivity and specificity of 81% and 71.9%; 95.2% and 56.2%, respectively. CONCLUSION: Oxidative stress has been observed to develop since early days of onset of dengue infection. Plasma PCOs, PCOs to PBSH group ratio were found to very well predict DHF/DSS.