During the 2003 outbreak of avian influenza caused by highly pathogenic avian influenza (HPAI) virus A/H7N7 in the Netherlands, human infection with this virus occurred in unexpectedly high numbers. The first signs of human infection were found in a veterinarian who visited several farms with infected poultry and subsequently developed acute conjunctivitis caused by the A/H7N7 virus. Human influenza virus (mostly A/H3N2) was circulating in the community simultaneously. In view of the potential risk of co-infection and reassortment, active case finding among people exposed to infected poultry was implemented immediately.

A case of conjunctivitis was defined as a person who had been exposed to contaminated poultry or a confirmed human case of avian influenza A/H7 since 28 February in the Netherlands, with at least two of the following symptoms: flow of tears or red or itching or painful or burning eyes or pus or photophobia. A case of influenza-like illness (ILI) was defined as a person who had been exposed to contaminated poultry or a confirmed human case of avian influenza A/H7 since 28 February in the Netherlands with acute illness onset and fever with at least one of the following symptoms: cough, rhinorrhoea, sore throat, myalgia, or headache. Based on the initial findings, all those involved in the culling of poultry were advised to wear protective goggles and face masks, and to wash their hands, especially before returning home from work. In addition, vaccination with the current influenza vaccine and prophylactic treatment with neuraminidase inhibitors of all people exposed to infected poultry was initiated, as well as treatment of all patients reporting to the municipal health services who had been exposed.

As of 22 April, conjunctival and throat swabs were collected from 293 patients, of whom 260 fitted the case definition of conjunctivitis and/or ILI (see table 1). Seventy eight of 260 cases had H7 positive conjunctival swabs. Eight cases also had an H7 positive throat swab. One had an H7 positive throat swab only. One of the cases presenting with ILI only had an H7 positive conjunctival swab. An additional two patients who did not fit the case definition for conjunctivitis had H7 positive lab results. Both complained of 'burning eyes' only. Six cases had H3 positive throat swabs, but no simultaneous infection with H7 and H3 influenza virus was detected.

Most conjunctivitis cases occurred between 5 and 20 March, after which date the number of new cases dropped below five reports per day (figure 1). This was approximately five days after implementation of the more stringent precautionary measures, including prophylactic treatment with oseltamivir of all persons exposed to suspected or confirmed infected poultry. Two weeks later, however, the number of reported and confirmed cases showed a slight increase. Only eight of 25 new cases (32%) claim to have taken oseltamivir prophylaxis in accordance with the guidelines, although all have had prolonged exposure to suspect or confirmed infected poultry. A relevant finding was also the confirmation of A/H7 associated conjunctivitis in four poultry workers from Poland and one from Belgium, all involved in the culling. While the risk is low, they do present a possible vector for transmission of avian influenza virus to other areas. This finding was communicated to the health authorities in Poland, Germany, and Belgium.

There has been preventive depopulation of poultry flocks in Germany along the border with the Netherlands, although there have so far been no suspected or confirmed AI infections in German poultry. Precautionary measures regarding possible spread of the virus to other farms and among humans are being implemented.

Three household contacts of confirmed cases developed A/H7-associated conjunctivitis while they had no direct exposure to infected poultry. All three had conjunctivitis as the most prominent presenting symptom, while one (a 12 year old child) additionally developed ILI. These results strongly suggest person to person transmission of avian influenza A/H7N7 virus.

On April 17, a veterinarian who had visited a poultry farm infected with HPAI died as the result of acute respiratory distress syndrome, which developed following a one week episode of ILI. Conjunctivitis was not observed, and the course of illness was protracted with bilateral pneumonia unresponsive to treatment resulting in death two weeks later. The veterinarian had visited an infected flock for sampling, and was not taking prophylactic drugs for reasons that have not yet been clarified. There was no history of underlying illness. The diagnostic evaluation was complicated: throat and conjunctival samples collected after one week of illness tested negative for A/H7 in both labs involved in the testing. A bronchoalveolar lavage collected two days later was positive for A/H7. Differential diagnostic tests for legionella, mycoplasma, psittacosis and other bacterial pathogens were negative. All hospital contacts and the relatives of this person were given oseltamivir prophylaxis, and persons with symptoms were sampled for virological testing. So far no secondary transmissions have been detected.

Release of information about the death to the media resulted in an upsurge of reports of possible cases, and renewed discussion about the detail of the preventive measures. It was reinforced that poultry workers and farmers, including anyone visiting an infected flock, should wear protective clothing, masks to cover the mouth and nose, and eye protection, be vaccinated against influenza, and have prophylactic antiviral medication. Based on the existing information, the consensus is that these measures should prevent serious consequences of A/H7 transmission.

At present, a cohort study is being conducted among confirmed cases and their household contacts in order to ascertain the extent of person to person transmission of AI and to identify risk factors. Another cohort study among poultry workers and poultry farmers will be conducted, in order to study risk factors for transmission of AI from poultry to humans.

Our preliminary observations support the hypothesis that infected humans might function as a 'mixing vessel' for generating novel influenza viruses and strongly support a need for active surveillance and pandemic planning.

Table 1: Results of laboratory testing in cases of conjunctivitis, Influenza Like Illness (ILI) and other patients who had been exposed to suspected or confirmed AI infected poultry in The Netherlands.

Figure 1: Human cases of conjunctivitis after exposure to avian influenza A/H7 in the Netherlands according to date of onset and laboratory result.