Distinguishing Zika and Dengue Viruses through Simple Clinical Assessment, Singapore

Abstract

Dengue virus and Zika virus coexist in tropical regions in Asia where healthcare resources are limited; differentiating the 2 viruses is challenging. We showed in a case-control discovery cohort, and replicated in a validation cohort, that the diagnostic indices of conjunctivitis, platelet count, and monocyte count reliably distinguished between these viruses.

Keywords: Singapore; Southeast Asia; Zika virus; conjunctivitis; dengue virus; platelets; viruses; zoonoses.

Figure 1

Univariate logistic regression model of clinical characteristics for patients in study of clinical assessments to distinguish Zika and dengue virus infections, Singapore. We analyzed early presentation (seeking treatment within 3 days of symptom onset), conjunctivitis, fever, myalgia, and headache as dichotomous variables, and laboratory findings (monocyte and platelet counts, ALT and AST levels) as continuous variables. For dichotomous variables, odds ratio (OR) >1 is predictive of Zika virus infection and <1 of dengue virus infection; for continuous variables, every unit increase in readout is predictive of Zika virus infection for OR >1 and dengue virus infection for OR <1. Error bars indicate 95% CIs. ALT, alanine aminotransferase; AST, aspartate aminotransferase.

Figure 2

Receiver operating characteristics for different models in study of clinical assessments to distinguish Zika and dengue virus infections, Singapore. AUC is shown for different models: conjunctivitis alone (model C), conjunctivitis with platelet count (model CP), and conjunctivitis with platelet and monocyte counts (model CPM). AUC, area under the curve.