ELISA for detection of dengue antibodies in serum of febrile patients was not very precise and was subject to inter-laboratory variability. In this study the use of whole blood collected on filter papers introduced an extra, mainly random, error, affecting the diagnostic accuracy for the diagnosis of patients suspected of dengue virus infection. The correlation between serum IgG and IgG collected from whole dried blood was better and this method can be used for dengue virus IgG sero-prevalence studies.
There is no conflict of interest. One of the co-authors, J. Groen, recently changed position, and now works for the company which produces the test kit which was used in this study. There has never been any contact between this company and the first author or the corresponding author. J. Groen, withdrew from this research project after his change of position and the results of this study and the contents of this paper were not available at the time that he changed position.
T.T. Nga: Fieldwork and data collection, laboratory tests, datamanagement, analysis and statistics, writing manuscript
Peter J. de Vries: Study design, fieldwork and data collection, datamanagement, analysis and statistics, writing manuscript.
Hoang Lan Phuong: Fieldwork and data collection, datamanagement
Phan T. Giao: Fieldwork and data collection, datamanagement
L.Q. Hung: Fieldwork and data collection, datamanagement
Tran Q. Binh: Fieldwork and data collection, datamanagement, logistics and local scientific supervision, review and corrections of manuscript.
V.T.C. Mai: Supervision of laboratory work, review and corrections of manuscript
Nguyen V. Nam: Fieldwork and data collection, review and corrections of manuscript
Nico Nagelkerke: Study design, data analysis and statistics, writing manuscript
Piet A. Kager: Study design, review and corrections of manuscript
Jan Groen: Study design, laboratory tests, writing manuscript
The study was supported by the Foundation for the Advancement of Tropical Research (WOTRO), the Netherlands.