Prevalence of Zika virus in blood donations: a systematic review and meta-analysis.

backgroundTransfusion-Transmitted Zika virus (TT-ZIKV) has become an emerging threat to world blood banks due to the fast spread of ZIKV epidemics and high rate of asymptomatic infections. For the risk assessment of ZIKV infection in blood products, relevant studies in blood donations or blood donors tested for ZIKV were collected and analyzed systematically. The overall prevalence of ZIKV infection were estimated through meta-analysis and potential risk factors were detected. The results will provide important clues for the protocol design of blood screening tests.

methodsRelevant articles about the rate of ZIKV detected in blood samples were identified from PubMed, Scopus and Web Of Science using key terms search strategy until October 7, 2017. Eligible articles were screened following inclusion and exclusion criteria. Meta-analysis and subgroup analyses were performed by software R3.4.1. Overall postdonation and posttransfusion follow-ups were analyzed.

resultsTen literatures (528,947 blood samples) were included for meta-analysis. The overall pooled prevalence of ZIKV (RNA and antibody) in blood donations was 1.02% (95%CI 0.36-1.99). The pooled prevalence of ZIKV RNA in blood donations was 0.85% (95%CI 0.21-1.88) less than the pooled prevalence of anti-ZIKV antibodies 1.61% (95%CI 0.03-5.21), however the difference was not statistically significant (p = 0.52). The prevalence varied significantly in different geographical regions (p < 0.001). Blood donations were more than two times likely to be infected by ZIKV in Zika epidemic period (1.37, 95%CI 0.91-1.91) than in non-epidemic period (0.61, 95%CI 0-2.55). The prevalence of anti-ZIKV antibodies (1.61, 95%CI 0.03-5.21) was almost twice as much as ZIKV nucleic acid detected in blood donations (0.85, 95%CI 0.21-1.88). However, statistically significant differences were not observed. A total of 122 ZIKV positive blood donors were followed, of which 48 (39%) reported symptoms postdonation, but none of the 13 followed recipients reported any clinical symptoms related to Zika infection posttransfusion.

conclusionThe pooled prevalence of Zika infection in blood donations was 1.02%. The prevalence varied greatly and reached to high-risk level in most of the situations. The results suggest that nucleic acid tests (NAT) for blood screening and pathogen reduction/inactivation technology (PRT) should be implemented in Zika-endemic areas and appropriate strategies should be designed according to different conditions. More studies are needed in the future to provide more evidence.