Study population

This pilot study was conducted among healthy adults aged 18–65 who donated blood at thirteen randomly selected donation-sites in the West Austrian federal state of Tyrol during the time period of August to November 2016. Participants provided written informed consent before they filled out a short questionnaire and gave a 5 mL EDTA-blood sample. The questionnaire provided information on history of travel to Latin America or Asia within the past eighteen months and/or a history of travel to established A. albopictus endemic areas in Europe in the time period between April to October 2016 before blood donation. Additionally, information on vaccination for tick-borne encephalitis virus (TBEV), Yellow fever (YFV) as well as Japanese encephalitis viruses (JEV) was obtained. History of known previous infection with any of the structurally relevant flaviviruses, namely TBE, YFV, JEV, West Nile Virus (WNV) and Dengue fever Virus (DEV) was also captured. The questionnaires (in German and English) are provided as supplementary materials (S1 File, S2 File).

Anti-ZIKV ELISA

ZIKV specific immunoglobulins were detected using Anti-Zika Virus ELISA (IgG / IgM) according to the recommendations of the manufacturer (EUROIMMUN, Medizinische Labordiagnostika, Germany) [12–13]. Diluted patient samples (1:101) were incubated in microplate wells coated with highly purified ZIKV non-structural protein (NS1). We conducted separate analyses for IgM and IgG antibodies using separate secondary anti-human antibodies directed at human IgM and IgG, respectively. In order to improve the specificity of IgM antibodies, samples were pre-incubated with buffer containing rheumatoid factor absorbent. The secondary anti-human antibodies are peroxidase-labelled and lead to color reactions in case of bound anti-Zika antibodies despite thorough washing. Optical density (OD) was measured as described elsewhere [13] using BEP III system (Siemens Healthcare, Munich, Germany). Results were given semiquantitatively as corrected OD by calculating a ratio of the OD of the patient sample to the OD of calibrator (provided in the test kit). Results were interpreted as recommended by the manufacturer: positive if the OD ≥1.1, negative if OD <0.8 and borderline if OD ≥0.8 to <1.1.

In order to rule out possibly cross reacting antibodies, serum samples which tested positive for ZIKV antibodies, were additionally tested for TBEV (Enzygnost Anti-FSME-Virus IgG/IgM, Siemens), DEV, WNV and JEV (in-house ELISA tests developed at the Arbovirus reference center, Vienna) antibodies. All anti-ZIKV IgM antibody positive samples were additionally tested for Epstein-Barr-Virus (EBV) (EBV IgM Architect i2000S2, Abbott) in order to exclude positivity due to polyclonal B-cell stimulation.

Neutralization test (NT)

ZIKV NT was performed essentially as described previously for West Nile virus at the National Arbovirus Reference Center in Vienna (Medical University of Vienna, Center for Virology) [14]. Briefly, serial dilutions of samples were mixed with 30–60 TCID50 Zika virus (strain H/PF 2013) [15], and incubated for one hour at 37°C. Vero cells were added and incubated further for 3 to 4 days. The presence of infectious (non-neutralized) virus in the supernatant was assessed by microscopic evaluation of cytopathic effects. Lack of cytopathic effect at titers ≥20 was interpreted as positive indicating the presence of specific neutralizing antibodies preventing the infection of Vero cells.

Nucleic acid amplification test (NAAT)

ZIKV nucleic acid extraction was conducted in a fully automated manner according to the manufacturers instruction (NucliSense® easyMAG®, Biomerieux France). 200 μl EDTA-plasma was eluted into 110μl of purified RNA. For the amplification process, we used the recommended volume of 10μl of RNA extract with 20μl primer-enzyme-mix provided by the test kit to make a final reaction volume of 30 μl (RealStar® Zika Virus RT-PCR Kit 1.0 Altona Diagnostics, Germany) [16]. The test is based on real-time technology with reverse transcription of target RNA (NS1 gene region) to cDNA followed by amplification of target cDNA with simultaneous detection of amplified product using fluorescence labelled probes. The analytical sensitivity of the test l0.61 copies/μl of nucleic acid extracts 95% CI: (0.39–1.27 copies/μl). The kit is equipped with a system that controls for the quality of extracted sample (internal control) as well as positive and negative controls. The sample is considered negative, if there is no fluorescence signal detected up to the fortieth amplification cycle.

Ethics

This study was approved by the Ethical Committee of the Medical University of Innsbruck (AN2016-0113 363/4.3).

The issue of blood safety

The major challenge in preventing transfusion-associated transmission of ZIKV is the high rate of asymptomatic infections. Previous studies on screening of donated blood for viral RNA in ZIKV-affected regions show a prevalence of 2–3%, which is non-negligible [20–22]. With a sample size of approximately 1000 healthy donors, our study was powered to detect ZIKV prevalences as low as 0.4–1.6% and hence excludes with 95% confidence ZIKV viremia among donors in an extent observed in ZIKV endemic regions.

ZIKV seroprevalence

Since viremia following ZIKV acquisition is limited to 5–7 days in the infection life cycle, a mere absence of ZIKV nucleic acid in blood samples does not by far rule out existence of the virus and its local transmission. Therefore, in addition to NAAT we searched for serological evidence of ZIKV activity. Although our study region is particularly known for high immunization coverage (over 80%) against TBEV [23], only one out of over 800 (0.13%) TBEV vaccinated donors was anti-Zika IgG positive confirming the high suitability of this ELISA kit in regions with high TBEV prevalence or high immunization coverage. With only 0.3% false positive IgM antibodies, which were negative for all other structurally related flaviviruses, it may be recommendable–at least in central Europe–to adopt Euroimmun ELISA as a stand-alone serological method in an acute and convalescent phase sera without necessarily going through a time consuming neutralization assay.

Strengths and limitations of the study

Our study has strengths and some limitations. A major strength is the fact that we searched for both serological and molecular evidence of ZIKV infection using highly sensitive and validated test kits (CE marked). The availability of data on history of vaccination against and/or history of infection with other related flaviviruses made our data more valuable in terms of understanding the role of antibody cross reactivity–particularly with that of TBEV against which a great majority of the population possesses antibodies. A major limitation of the study is the lack of convalescent sera from ELISA positive samples (except for one), which would have significance in supporting or excluding a recent infection through seroconversion or rise in antibody titer. However, this diagnostic gap is narrowed to a certain extent through concomitant testing of these samples for the presence of nucleic acid and for specific neutralizing antibodies. The modest sample size, which rendered the study less powered in detecting isolated sporadic cases of ZIKV in the region, may also be a limitation. Another pitfall of the study may be the subjectivity of microscopic detection of cytopathic effects in the NT assay, which may be probe to some degree of bias. However, the complete lack of neutralizing antibodies coupled with the fact that our ELISA did not cross react with TBEV makes the interpretation of the NT assay in our study rather unequivocal.

Conclusion

Looking into data from previous studies in endemic regions, which estimated ZIKV prevalence to reach up to 73% [24] once introduced in the area, our finding of no serological and molecular evidence among the study population may be a true reflection of the absence of autochthonous ZIKV transmission in the region. This, combined with the end of the warm season in the continent that leads to attenuation of mosquito activities, may suggest the end of ZIKV threat for now and end of the public health concern in Europe. With this result, we may dare to claim that routine screening of donated blood samples for ZIKV is most probably not a major priority in the study region. This, however, does not warrant the complete cessation of monitoring and surveillance of the virus particularly in A. albopictus endemic regions due to its disastrous complications once introduced. We recommend that efforts should be made to conduct similar studies in a larger scale in order to detect or exclude possible isolated sporadic occurrences.