Seroprevalence of avian influenza A (H5N1) virus among poultry workers in Jiangsu Province, China: an observational study

Since 2003 to 06 Jan 2012, the number of laboratory confirmed human cases of infection with avian influenza worldwide was 576 and 339 were fatal. In China, the number was 41 and 27 were fatal [1]. In December 2007, a family cluster of 2 individuals infected with highly pathogenic avian influenza (HPAI) A (H5N1) virus was identified in Jiangsu Province, China. One case died [2]. However, the official estimate of the H5N1 case-fatality rate has been described by some as an over estimation since there may be numerous undetected asymptomatic/mild cases of H5N1 infection [3]. One particularly important hypothesis is that cases are being missed because current surveillance primarily detects severe infections [4]. This study was designed to better understand the real infection rate of H5N1. It investigated the potential for the zoonotic spread to humans and evaluates the risk factors associated with subclinical infection among poultry workers in Jiangsu, China.

A total of 306 serum samples were collected and tested. One hundred samples were from Gaochun, 76 samples from Jianhu and 130 samples from Gaoyou. The median and quartiles of ages of poultry workers were 58.00 (45.75-64.25). The sex ratio of workers was 1:1.59. All the serum samples were negative for antigen of A/Hubei/1/10. Thus the test results for antigen of A/Anhui/1/05 were used for analyses. The overall seropositive rate of H5N1 was 2.61% (95%CI, 1.14%-5.09%) in this survey. The positive rates of samples collected from Gaochun, Jianhu and Gaoyou were 0, 1.32% (95%CI, 0.03%-7.11%) and 5.38% (95%CI, 2.19%-10.78%), respectively, with significant difference (P= 0.029, Kruskal-Wallis Test).

The median and quartiles of raising poultry numbers were also found significantly different among 3 counties (P< 0.0001, Kruskal-Wallis Test). The numbers were 5.00 (3.00-7.00), 14.50 (7.25-24.50) and 18.00 (13.00-28.00) in Gaochun, Jianhu and Gaoyou, respectively.

In correlation analyses, the seropositive rates of H5N1 virus significantly correlate with the medians of raising poultry numbers (r = 1.00 and P< 0.01 for spearman test) (Figure 1). And the antibody titers against H5N1 virus also significantly correlate with the raising poultry numbers (r = 0.181, P= 0.001).

Several serological studies have been conducted to assess the transmission of H5N1 virus in poultry workers. The reported seropositive rates vary from 0 to 10% [9–11]. In the present study, the overall positive rate we found was 2.61% and in different sampling districts the rates varied from 0 to 5.38%. Many factors could be responsible for the variability, such as the H5N1 infection rate and vaccination rate of the poultry, the precautions and job duties of the workers. In our study, poultry number was identified as a novel risk factor associated with human infection with avian H5N1 virus, which could be accounted for the distinction of seropositive rate among districts and studies.

Poultry workers were expected to have the highest level of exposure to avian H5N1virus and more intensive poultry exposure was associated with having anti-H5N1 antibody [11]. Compared with poultry workers from large scale poultry farms, backyard workers may have an even higher risk due to the lack of preventive measures and healthy environment. Wang et al. conducted a serologic surveillance study in Guangzhou, China, founding workers in scale poultry farms all seronegative for H5N1 virus [9]. Certain job duties such as butchering poultry, feeding poultry and preparing poultry for restaurants may also be associated with the infection risk [11]. In our study, enrolled poultry workers were all from backyard poultry farms. Their job duties were almost identical, mainly including feeding poultry, collecting eggs and cleaning poultry stalls.

High poultry density could raise the infection risk of poultry with avian influenza H5N1. Tiensin et al. conducted an ecologic investigation on risk factors of clusters of avian H5N1 virus infection. Fighting cock flock density, meat and laying duck flock density were found significantly higher in case subdistricts than in control subdistricts [12]. However, the association between increasing poultry number and risk of humans infected with avian influenza H5N1 was identified for the first time in this study. This finding could help public health officers recognize high-risk population and institute countermeasures.

Because up to 40% of the poultry workers enrolled in our study were ≥60 years old and the microneutralization and western blot assays were found to be less specific for persons ≥60 years old [5], the horse RBC HI assay, which has high reliability and good agreement with MN assay results without age limit was used in this study [7].

Our findings suggested that avian-to-human transmission of influenza H5N1 virus remains low in China. Workers associated with raising larger poultry flocks have a higher risk of seropositivity. There were several potential reasons for that, such as high poultry density could raise the infection risk of poultry with avian influenza H5N1 [12], and more intensive poultry exposure was associated with having anti-H5N1 antibody for poultry workers [11]. As we didn't know the infection rate of the poultry and the moderate sample size limited the statistical significance of our research, further studies are warranted to validate our results.

Avian-to -human transmission of avian H5N1 virus remained low. Workers associated with raising larger poultry flocks have a higher risk of seropositivity.