Dengue is one of the most important human arboviral infections. The pathogenesis of dengue fever and dengue hemorrhagic fever is not well understood. The analysis of plasma proteins in search of biomarker candidates to monitor various diseases has been used for many years. The aim of this study was to compare the 2-D electrophoresis profiles of plasmas from patients with dengue hemorrhagic fever and from healthy donors, identifying proteins that are differentially expressed. Plasmas were collected in the Hospital Clínicas de Niterói (RJ) during the last dengue outbreak in the state of Rio de Janeiro (2001-2002). We have optimized the following experimental conditions using a pool of plasmas from healthy donors: a) the two most abundant proteins were removed from plasma using the “Albumin and IgG removal kit” from GE Healthcare; b) the use of cup-loading for sample application in the first dimension (IPG strips 4-7 / 18 cm) resulted in a better resolution of the spots when compared with sample in-gel rehydration; c) the ideal amount of protein to be applied in each gel was 550 mg; d) gel staining with either Coomassie R-250 or silver nitrate (MS compatible protocol) was equivalent in terms of sensitivity and a little less efficient than colloidal Coomassie. Gel images were compared using the ImageMaster 2D Platinum software (GE Healthcare) and several differentially expressed spots were detected in the plasmas of dengue patients and healthy donors. Some of them were excised, subjected to in-gel trypsin digestion and identified by MALDI-TOF/TOF MS as fibrin beta chain, vitamin D-binding protein, fibrinogen gamma chain, haptoglobin, transferrin and hemopexin. After combining the results from 12 gels (6 patients and 6 controls), none of these differences could be considered statistically significative. We concluded that, even after depleting albumin and IgG from plasma, the proteins that could be visualized by colloidal Coomassie are still highly abundant. They are probably masking the presence of low abundant proteins that could correlate with the disease. To overcome these limitations, we are now depleting the 6 most abundant proteins from plasma and analysing the gels using the 2-D Fluorescence Difference Gel Electrophoresis (DIGE) technique. This new approach may help us to identify important proteins in the pathogenesis of dengue hemorrhagic fever, contributing for a better understanding of the disease.