Lack of Evidence for Human-to-Human Transmission of Avian Influenza A (H9N2) Viruses in Hong Kong, China 1999

Timothy M. Uyeki, Centers for Disease Control and Prevention, Atlanta, Georgia, USA; Yu-Hoi Chong, Department of Health, Hong Kong Special Administrative Region of China; Jacqueline M. Katz, Centers for Disease Control and Prevention, Atlanta, Georgia, USA; Wilina Lim, Yuk-Yin Ho, Department of Health, Hong Kong Special Administrative Region of China; Sophia S. Wang, Thomas H.F. Tsang, Centers for Disease Control and Prevention, Atlanta, Georgia, USA; Winnie Wan-Yee Au, Shuk-Chi Chan, Department of Health, Hong Kong Special Administrative Region of China; Thomas Rowe, Jean Hu-Primmer, Jensa C. Bell, William W. Thompson, Carolyn Buxton Bridges, Nancy J. Cox, Centers for Disease Control and Prevention, Atlanta, Georgia, USA; Kwok-Hang Mak, Department of Health, Hong Kong Special Administrative Region of China; Keiji Fukuda, Centers for Disease Control and Prevention, Atlanta, Georgia, USA


Emerging Infectious Diseases. 2002;8(2) 

In This Article


These cohort studies suggest that influenza A (H9N2) viruses were not transmitted from the two H9N2-infected children to family and household members or HCWs who were exposed to the H9N2 patients during their acute illness infectious periods. As described for the avian influenza A (H5N1) viruses[15,20], a combination of serologic assays was effective in detecting H9 virus-specific antibodies in two pediatric cases of H9N2 infection. However, the same serologic assays did not detect H9 antibodies in family members or HCWs exposed to the H9N2 patients. Only two known exposed persons, an HCW and a family member of one H9N2 patient, declined to participate in the studies. The HCW who tested seropositive for antibodies to H9N2 had no known exposure to a patient with confirmed H9N2 infection or contact with poultry or swine. The timing of H9N2 infection in this HCW could not be determined.

Evidence for influenza A (H9N2) infection as the cause of acute illness in the two patients includes the direct isolation of H9N2 viruses from nasopharyngeal aspirate specimens during the acute phase of illness[1] and the detection of H9-specific IgM antibodies, suggesting recent infection with an H9 virus. No other bacterial or viral pathogens were identified except for isolation of adenovirus type 3 from patient 1. The significance of the latter finding is unknown since this patient did not have typical signs of adenovirus type 3 infection, such as conjunctivitis. Isolation of adenovirus in this patient could represent acute atypical infection, acute subclinical infection, or persistent viral shedding from previous adenovirus infection.

The apparent lack of human-to-human transmission of avian H9N2 viruses and the low transmissibility of avian H5N1 viruses among humans have several possible explanations[14,15]. The genomes of the H9N2 and H5N1 strains that were isolated from humans were derived entirely from avian influenza viruses; no reassortment with circulating human influenza A viruses had occurred. It is possible that the avian virus genome limits viral spread among humans. The molecular basis of influenza virus transmission among humans and other species remains poorly understood. However, following the introduction of an avian virus into humans, alterations in receptor-binding specificity of the HA are likely necessary for effective human-to-human transmission[22]. Alternatively, the children may not have shed H9N2 virus in titers sufficient to facilitate transmission to other persons. Neither H9N2-infected child had coughing or sneezing that would have enhanced transmission to persons who had close contact with them.

To improve specificity for detecting antibody for H9N2 over that of the hemagglutination-inhibition antibody assays used previously[3], we used a combination of confirmatory tests and an adsorption step to reduce cross-reactivity with antibodies to other influenza viruses. Sera testing positive by neutralization test were then tested by Western blot assay. Sera positive for both these assays were further tested by neutralization assay following adsorption of sera with influenza A (H3N2) viruses. Sera that were negative for antibodies to H9N2 by neutralization assay were not tested by Western blot because of resource limitations. However, all sera from children who were contacts of the H9N2 patients, as well as the patients themselves, were also tested by an H9-specific ELISA. Both patients but none of the exposed children tested positive for H9 antibodies.

Because of insufficient sera, the H9N2 patients were not tested for antibodies to neuraminidase (NA). The N2 NA of the H9N2 viruses isolated from patients is antigenically distinct from that of recent H3N2 human viruses, although some cross-reactivity with human H2N2 and early H3N2 viruses has been reported[8]. However, additional studies from our laboratory indicate that the apparent cross-reactive antibodies that could be removed from some human sera by adsorption with H3N2 viruses was not due to cross-reactivity between the N2 NAs, since these sera also reacted with a reassortment H9N7 virus (CDC, unpub. data).

Because only two H9N2 cases were identified, we did not conduct a case-control study to identify risk factors for H9N2 infection. Thus, the sources and modes of acquisition of H9N2 for the two infected children are unknown. The Hong Kong Department of Health found that one H9N2 patient had very brief exposure to live chickens 11 days before onset of illness but did not directly touch the birds. No other contacts with live poultry, swine, or other animals for either H9N2 patient were found. There was no known contact or common exposure between the two H9N2 patients.

During the 1997 FLUAV (H5N1) outbreak in Hong Kong, a case-control study found that visiting a poultry stall or market with live poultry during the week preceding illness was the main risk factor for H5N1 infection[12]. During that outbreak, the Hong Kong Department of Health enhanced its active surveillance for influenza-like illness and influenza viruses in hospitals, general outpatient clinics, and physicians' offices. This enhanced surveillance system detected the two novel H9N2 infections.

We were able to obtain only one convalescent-phase blood specimen from study participants, which limited our ability to document seroconversion. However, none of the exposed persons were seropositive for H9N2. Currently, there are no seroprevalence data on rates of H9N2 infection in children or the general population. One study of a cohort of poultry workers in Hong Kong found that approximately 30% were seropositive for antibodies to H9N2[5]. Ongoing surveillance and availability of H9N2-specific reagents should facilitate timely identification of H9N2 infection and allow collection of paired sera for further studies of person-to-person transmission.

In addition to H9N2, other avian influenza viruses have been isolated from specimens collected from Hong Kong poultry since 1997, including H6, H4, and H11 viruses[23]. During April and May 2001, highly pathogenic avian influenza A (H5N1) viruses were again isolated from live poultry in Hong Kong markets[24]. After chicken deaths were observed in some markets, the Hong Kong government temporarily closed all wholesale and retail live poultry markets for cleaning, stopped importing poultry from China, and slaughtered approximately 1.3 million birds during May 2001. The poultry markets reopened in June 2001. No human illnesses attributed to avian influenza viruses have been identified since the two H9N2 cases in 1999. However, these recent events have heightened the need to understand the public health risk of H5N1, H9N2, and other avian influenza viruses.

These limited studies suggest that avian influenza A (H9N2) viruses were not transmitted from the two infected children to exposed household members, relatives, or HCWs in Hong Kong. However, H9N2 viruses are widely distributed in avian populations, can infect humans, and could evolve or undergo genetic reassortment with potential for increased pathogenicity and transmissibility in humans. The recent emergence of human infections with avian influenza A (H9N2) and (H5N1) viruses highlights the need to improve surveillance for influenza viruses in poultry, swine, and humans, especially in Asia. Further studies to assess the health risks posed by H9N2 and other avian influenza viruses are warranted.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: