Screening and exclusion of Zika virus infection in travellers by an NS1-based ELISA and qRT-PCR

Objectives: Zika virus (ZIKV) infection during pregnancy may cause neurological abnormalities in the foetus, and therefore fast and accurate laboratory assays are critical for rapid diagnosis. ELISA based on ZIKV NS1 protein has been developed and shown to be sensitive and highly specific; however, its negative and positive predictive values have not been tested. In this study we evaluated the ability of the NS1-based ELISA to exclude ZIKV infection and serve as a first-line screening tool for travellers. Methods: We tested samples obtained during the peak of ZIKV infection from 1188 symptomatic and asymptomatic Israeli travellers using NS1-based IgG and IgM ELISA, real-time RT-PCR analysis and ZIKV neutralization. The Kaplan–Maier method was used to evaluate the duration of ZIKV RNA in whole blood and urine samples. Results: NS1-based ELISA identified 20 true-positive, five false-positive and four false-negative cases, resulting in sensitivity and specificity of 83.3% (95%CI: 62–94%) and 97.5% (95%CI: 94–99%) respectively, and positive and negative predictive values of 80% (95%CI: 59–92%) and 98% (95%CI: 95–99%) respectively. Based on 14 RT-PCR-positive cases, median time to detect ZIKV RNA in whole blood was 17.5 days (range 5–58 days) and in urine 10 days (range 5–26 days). Conclusions: The NS1-based ELISA and RT-PCR in whole blood are highly reliable for identification of ZIKV-negative and -positive cases, respectively. Combination of both assays minimizes the risk of false-negative results, and thus allows the exclusion of ZIKV infection in travellers returning from ZIKV-endemic countries, including those who are pregnant or wish for preconception screening.