Surveillance of Avian H7N9 Virus in Various Environments of Zhejiang Province, China before and after Live Poultry Markets Were Closed in 2013-2014

Background To date, there have been a total of 637 laboratory-confirmed cases of human infection with avian influenza A (H7N9) virus across mainland China, with 28% (179/637) of these reported in Zhejiang Province. Surveillance of avian H7N9 virus was conducted to investigate environmental contamination during H7N9 outbreaks. We sought to evaluate the prevalence of H7N9 in the environment, and the effects of poultry market closures on the incidence of human H7N9 cases. Methods We collected 6740 environmental samples from 751 sampling sites across 11 cities of Zhejiang Province (China) between January 2013 and March 2014. The presence of H7N9 was determined by reverse transcription polymerase chain reaction, with prevalence compared between sites and over time. The relationship between environmental contamination and human cases of H7N9 infection were analyzed using Spearman's ranked correlation coefficient. Results Of the 6740 samples, 10.09% (680/6740) were H7N9-positive. The virus was found to circulate seasonally, and peaked during the spring and winter of 2013-2014. The prevalence of the virus decreased from the north to the southeast of the province, coinciding with the geographical distribution of human H7N9 cases. Compared with other sampling sites, live poultry markets (LPMs) had the highest prevalence of H7N9 virus at 13.94% (667/4784). Of the various sample types analyzed, virus prevalence was highest for chopping board swabs at 15.49% (110/710). The prevalence of the virus in the environment positively correlated with the incidence of human H7N9 cases (r(2) = 0.498; P < 0.01). Cities with a higher incidence of human H7N9 cases also had a higher prevalence of H7N9 among samples and at sampling sites. Following the closure of LPMs at the end of January 2014, the prevalence of H7N9 decreased from 19.18% (487/2539) to 6.92% (79/1141). This corresponded with a decrease in the number of human H7N9 cases reported. Conclusions The prevalence of H7N9 virus in environmental samples oscillated seasonally, regardless of whether LPMs were open. The presence of H7N9 in environmental samples positively correlated with the number of human H7N9 cases, indicating that eradication of the virus from the environment is essential in reducing the numbers of H7N9 cases and halting the spread of the virus.