Background
Dengue is the most common vector-borne viral disease worldwide. Most cases are mild, but some evolve into severe dengue (SD), with high lethality. Therefore, it is important to identify biomarkers of severe disease to improve outcomes and judiciously utilize resources.
Methods/Principal Findings
One hundred forty-five confirmed dengue cases (median age, 42; range <1-91 years), enrolled from February 2018 to March 2020, were selected from an ongoing study of suspected arboviral infections in the Asunción metropolitan area. Cases included dengue virus types 1, 2, and 4, and severity was categorized according to the 2009 World Health Organization guidelines. Serologic and biomarker (lipopolysaccharide binding protein and chymase) testing were performed on acute-phase samples by ELISA; additional serologic testing was performed with the multiplex pGOLD assay. Complete blood counts and chemistries were performed at the discretion of the care team. Age, gender, and pre-existing comorbidities were associated with SD vs. dengue with/without warning signs in logistic regression with odds ratios (ORs) of 1.06 (per year; 95% confidence interval, 1.02, 1.10), 0.12 (female; 0.03,0.5), and 9.82 (presence; 1.92, 50.24) respectively. In binary logistic regression, for every unit increase in anti-DENV IgG in the pGOLD assay, odds of SD increased by 2.54 (1.19-5.42). Platelet count, lymphocyte percent, and elevated chymase were associated with SD in a combined logistic regression model with ORs of 0.99 (1,000/μL; 0.98,0.999), 0.92 (%; 0.86,0.98), and 1.17 (mg/mL; 1.03,1.33) respectively.
Conclusions
Multiple, readily available factors were associated with SD in this population. These findings will aid in the early detection of potentially severe dengue cases and inform the development of new prognostics for use in acute-phase and serial samples from dengue cases.