Sequential dengue virus infections detected in active and passive surveillance programs in Thailand, 1994-2010.

Piraya Bhoomiboonchoo,Darunee Buddhari,In-Kyu Yoon,Anon Srikiatkhachorn,Siripen Kalayanarooj,Timothy Endy,Mathuros Thipayamongkolgul,Alan L Rothman,Ananda Nisalak,Sharone Green,Mammen P Mammen,Robert V Gibbons,Natkamol Chansatiporn

doi:10.1186/s12889-015-1590-z

Abstract

BackgroundThe effect of prior dengue virus (DENV) exposure on subsequent heterologous infection can be beneficial or detrimental depending on many factors including timing of infection. We sought to evaluate this effect by examining a large database of DENV infections captured by both active and passive surveillance encompassing a wide clinical spectrum of disease.MethodsWe evaluated datasets from 17 years of hospital-based passive surveillance and nine years of cohort studies, including clinical and subclinical DENV infections, to assess the outcomes of sequential heterologous infections. Chi square or Fisher’s exact test was used to compare proportions of infection outcomes such as disease severity; ANOVA was used for continuous variables. Multivariate logistic regression was used to assess risk factors for infection outcomes.ResultsOf 38,740 DENV infections, two or more infections were detected in 502 individuals; 14 had three infections. The mean ages at the time of the first and second detected infections were 7.6 ± 3.0 and 11.2 ± 3.0 years. The shortest time between sequential infections was 66 days. A longer time interval between sequential infections was associated with dengue hemorrhagic fever (DHF) in the second detected infection (OR 1.3, 95% CI 1.2-1.4). All possible sequential serotype pairs were observed among 201 subjects with DHF at the second detected infection, except DENV-4 followed by DENV-3. Among DENV infections detected in cohort subjects by active study surveillance and subsequent non-study hospital-based passive surveillance, hospitalization at the first detected infection increased the likelihood of hospitalization at the second detected infection.ConclusionsIncreasing time between sequential DENV infections was associated with greater severity of the second detected infection, supporting the role of heterotypic immunity in both protection and enhancement. Hospitalization was positively associated between the first and second detected infections, suggesting a possible predisposition in some individuals to more severe dengue disease.Electronic supplementary materialThe online version of this article (doi:10.1186/s12889-015-1590-z) contains supplementary material, which is available to authorized users.

Highlights

The effect of prior dengue virus (DENV) exposure on subsequent heterologous infection can be beneficial or detrimental depending on many factors including timing of infection
Sixty-three percent (643/1,018) of repeat infections were detected from public health samples or hospitalbased dengue studies through passive surveillance; 424 were from Queen Sirikit National Institute of Child Health (QSNICH) and 219 were from Kamphaeng Phet Provincial Hospital (KPPPH)
DENV infections were clinically categorized into four groups: subclinical, nonhospitalized dengue fever (DF), hospitalized DF, and dengue hemorrhagic fever (DHF)

Summary

Introduction

The effect of prior dengue virus (DENV) exposure on subsequent heterologous infection can be beneficial or detrimental depending on many factors including timing of infection. Many studies have shown that the risk of DHF in noninfant patients is greater when an initial DENV infection is followed by a second infection with a different serotype [2,3,4,5,6,7]. Bhoomiboonchoo et al BMC Public Health (2015) 15:250 individuals [9,10] One mechanism underlying this observation has been postulated to be antibodydependent enhancement (ADE) during the second infection mediated by non-protective heterotypic antibodies arising from the first infection. The timing of the second infection seems to be important since some degree of short-term protection may be conferred against subsequent heterologous infection by the preceding infection [11]. In a population model of children hospitalized with dengue in Bangkok, Thailand, the length of this short-term heterologous protection was estimated to be one to three years [12]. In an analysis of repeat DENV infections from a prospective cohort study of children in Kamphaeng Phet, Thailand, the ratio of symptomatic to subclinical infections was found to be higher when the time from first to second infection was longer [14]

Methods

Results

Discussion

Conclusion