OBJECTIVETo analyze the results of nine-round environmental specimen surveillance programs in five live poultry markets pre-, during and post the 'closing days' and to evaluate the effects of 'closing days' on live poultry markets regarding the control against avian influenza pollution.

METHODSIn January 2014, control measures including culling poultry, completely cleaning and disinfecting and a 'three-day-closing' measure were conducted in five live poultry markets which were found positive for H7N9 nucleic acid in the 1(st) round environmental specimen surveillance program. Second surveillance program was conducted after a thorough disinfection campaign was launched. Several times surveillance were conducted in one week, after the markets were reopened. RT-PCR was used to test the nucleic acid of HA, H5, H7 and H9 viruses.

RESULTS654 specimens from the environment were collected and tested. During the first round surveillance program, positive rates for influenza A and H5/H7/H9 nucleic acid of poultry stalls appeared to be 94.44% and 61.11% respectively. The positive rates of poultry stalls reduced to 0 after the disinfection campaign but increased again after the markets reopened. The positive rate for influenza A of poultry stalls slightly increased from 50.00% in the third surveillance to 72.22% in the ninth surveillance (P > 0.05). The positive rate for H5/H7/H9 of poultry stalls showed a significantly increasing trend, from 0 in the third surveillance to 44.44% in the ninth surveillance (P < 0.01). The positive rates for influenza A and H5/H7/H9 nucleic acid of specimens were 28.89% and 17.78% respectively. The positive rate of specimens reduced to 0 after disinfection while increased again after reopening of the markets. The positive rate for influenza A of specimens slightly increased from 19.67% in the third surveillance to 27.54% in the ninth surveillance programs (P > 0.05). The positive rate for H5/H7/H9 of specimen showed a significant increasing trend, from 0 in the third surveillance to 8.70% in the ninth-round surveillance programs (P < 0.01). The positive rate for influenza A was the highest for slaughter- related specimens of 22.4% (35/156). The positive rates for influenza A from sewage and drinking water being collected on the later stage after the markets reopened (25.9%, 12.4%)were higher than those on the early stage (8.3%, 8.6%) (P > 0.05). The positive rate for influenza A of poultry stalls with overnight poultry storage (91.7%) was significant higher than that of poultry stalls without the overnight storage (33.3%). The positive rate for influenza A of poultry stalls in which simultaneously selling different kinds of poultry (85.7%) was significant higher than that of poultry stalls in which selling only one kind of poultry at one time (25.0%) (P < 0.05).

CONCLUSIONSlaughter in live poultry markets posed a large risk of pollution diffusion. Sewage and drinking water showed an accumulation effect for avian influenza virus. Overnight poultry storage and selling different kinds of poultry at one time at the poultry stalls seemed the risk factors for avian influenza virus transmission. Complete cleaning, disinfecting and several 'closing days' for live poultry markets seemed effective in eliminating avian influenza virus. Once the markets were reopened, they seemed to be soon polluted again.