Passive Fever Surveillance for Dengue in Nha Trang City, Vietnam: Association between Clinical and Laboratory-Confirmed Diagnoses of Dengue

Abstract

Dengue is a leading cause of morbidity in Vietnam with gradual increase from 32.5 per 100,000 population in 2000 to 120.0 per 100,000 population in 2009. In Vietnam, almost all cases of dengue virus (DENV) infection are diagnosed clinically based on the WHO guideline (2009), but it is commonly under-reported and many cases are misclassified due to its broad spectrum of symptoms shared with other febrile diseases. On behalf of the Dengue Vaccine Initiative (DVI), the International Vaccine Institute (IVI) assessed the strength of association between clinical and laboratory-confirmed diagnoses of dengue infection from a passive hospital-based surveillance study in Nha Trang City, in Khanh Hoa province, Vietnam from July 2014 to December 2015. The study was designed to determine the true burden of dengue, symptomatic infection of all fever cases among children and adults between 1 and 55 years of age in a defined population. Clinical signs and symptoms of 553 dengue patients and 1,152 non-dengue patients (laboratory-confirmed by IgM and IgG capture ELISA or real-time PCR) were analyzed. Leucopenia (OR = 6.03), thrombocytopenia (OR = 1.97), rash (OR = 1.90), headache (OR = 1.75), arthralgia (OR = 1.75), petechiae (OR = 1.71), and nausea/vomiting (OR = 1.42) were highly associated with laboratory-positive dengue cases in the absence of respiratory signs and symptoms such as rhinorrhea (OR = 0.41) and expectoration (OR = 0.39). A comparison between adults and children of laboratory-confirmed dengue cases revealed that the frequency of clinical signs and symptoms was different between two age groups. DENV infected adults were more likely to present flushed face (OR = 2.13), headache (OR = 2.10), rash (OR = 1.96), thrombocytopenia (OR = 1.91), and arthralgia (OR = 1.83) which were not apparent to DENV infected children. Overall sensitivity and specificity of clinical diagnoses were 57.7% and 92.0% with diagnostic accuracy of 80.9%. Once validated, those key clinical features for children and adult patients with dengue infection may improve the future clinical outcomes, especially in the resource-poor dengue endemic countries, as they would allow more closely monitoring of selected patients and it may affect the sensitivity of clinical diagnosis.