Impact of dengue virus (DENV) co-infection on clinical manifestations, disease severity and laboratory parameters

Abstract

Background: The co-circulation of 4 DENV serotypes in geographically expanding area, has resulted in increasing occurrence of DENV co-infections. However, studies assessing the clinical impact of DENV co-infections have been scarce and have involved small number of patients. This study explores the impact of DENV co-infection on clinical manifestations and laboratory parameters.

Methods: This retrospective study involved consecutive hospitalized patients with non-structural protein 1 (NS1) antigen positivity during an outbreak (Jan to April 2014). Multiplex RT-PCR was performed directly on NS1 positive serum samples to detect and determine the DENV serotypes. All PCR-positive serum samples were inoculated onto C6/36 cells. Multiplex PCR was repeated on the supernatant of the first blind passage of the serum-infected cells. Random samples of supernatant from the first passage of C6/36 infected cells were subjected to whole genome sequencing. Clinical and laboratory variables were compared between patients with and without DENV co-infections.

Results: Of the 290 NS1 positive serum samples, 280 were PCR positive for DENV. Medical notes of 262 patients were available for analysis. All 4 DENV serotypes were identified. Of the 262 patients, forty patients (15.3 %) had DENV co-infections: DENV-1/DENV-2(85 %), DENV-1/DENV-3 (12.5 %) and DENV-2/DENV-3 (2.5 %). Another 222 patients (84.7 %) were infected with single DENV serotype (mono-infection), with DENV- 1 (76.6 %) and DENV- 2 (19.8 %) predominating. Secondary dengue infections occurred in 31.3 % patients. Whole genome sequences of random samples representing DENV-1 and DENV-2 showed heterogeneity amongst the DENVs. Multivariate analysis revealed that pleural effusion and the presence of warning signs were significantly higher in the co-infected group, both in the overall and subgroup analysis. Diarrhoea was negatively associated with co-infection. Additionally, DENV-2 co-infected patients had higher frequency of patients with severe thrombocytopenia (platelet count < 50,000/mm(3)), whereas DENV-2 mono-infections presented more commonly with myalgia. Elevated creatinine levels were more frequent amongst the co-infected patients in univariate analysis. Haemoconcentration and haemorrhagic manifestations were not higher amongst the co-infected patients. Serotypes associated with severe dengue were: DENV-1 (n = 9), DENV-2 (n = 1), DENV-3 (n = 1) in mono-infected patients and DENV-1/DENV-2 (n = 5) and DENV-1/DENV-3 (n = 1) amongst the co-infected patients.

Conclusion: DENV co-infections are not uncommon in a hyperendemic region and co-infected patients are skewed towards more severe clinical manifestations compared to mono-infected patients.

Keywords: Clinical manifestations; Co-infection; DENV; Dengue virus; RT-PCR; Serotype.

Fig. 1

Maximum likelihood phylogeny of DENV-1 and DENV-2 strains from South East Asia and Oceania. The tree was rooted against DENV-1 (Maroon branches). Filled black circles in front of taxon names indicate strains that were sequenced and reported in this study. Taxon names are abbreviated by their Genbank accession number followed by country and strain name. Values at nodes indicate bootstrap support. Genotype classification within serotype is indicated by genotype name beside each group, within a coloured-region. Genotype I, IV and V in DENV1; genotype Cosmopolitan, Asian American (A/A), Asian-II and Asian-I in DENV-2. For clarity, the branch leading to DENV-1 has been shortened and indicated with the real length. (Scale bar: average number of substitutions per site)