Protective measures and human antibody response during an avian influenza H7N3 outbreak in poultry in British Columbia, Canada

Abstract

Background: In 2004 an outbreak of avian influenza of the H7N3 subtype occurred among poultry in British Columbia, Canada. We report compliance with recommended protective measures and associated human infections during this outbreak.

Methods: We sought voluntary participation by anyone (cullers, farmers and their families) involved in efforts to control the poultry outbreak. Recruitment was by advertisements at the worker deployment site, in local media and through newsletters sent directly to farmers. Sera were tested for antibody to H7N3 by microneutralization assay. A subset of 16 sera (including convalescent sera from 2 unprotected workers with conjunctivitis from whom virus had been isolated) was further tested by Western blot and routine and modified hemagglutination inhibition assays.

Results: A total of 167 people (20% to 25% of all workers) participated between May 7 and July 26, 2004. Of these, 19 had experienced influenza-like illness and 21 had experienced red or watery eyes. There was no significant association between illness reports and exposure to infected birds. Among 65 people who entered barns with infected birds, 55 (85%) had received influenza vaccine, 48 (74%) had received oseltamivir, and 55 (85%), 54 (83%) and 36 (55%) reported always wearing gloves, mask or goggles, respectively. Antibody to the H7 subtype was not detected in any sera.

Interpretation: During the BC outbreak, compliance with recommended protective measures, especially goggles, was incomplete. Multiple back-up precautions, including oseltamivir prophylaxis, may prevent human infections and should be readily accessible and consistently used by those involved in the control of future outbreaks of avian influenza in poultry. Localized human avian influenza infections may not result in serologic response despite confirmed viral detection and culture.

Figures

Fig. 1: Chronology of the 2004 outbreak of avian influenza H7N3 in the Fraser Valley, British Columbia. aIndex farm included an older flock and a younger flock housed in 2 separate barns. bOn Feb. 4, the older flock in barn 1 of the index farm had reduced egg production and slightly increased mortality rate (to 0.5% over 72 h) that subsequently resolved. The virus was later identified (on Feb. 19) as low-pathogenicity avian influenza H7N3 (intravenous pathogenicity index [IVPI] = 0). cOn Feb. 16, the younger flock in barn 2 of the index farm had a sudden increase in mortality (to more than 25% in 48 h). Birds on the index farm were destroyed and quarantine was imposed on Feb. 20 to attempt containment. Virus from birds in barn 2 was later characterized (on Mar. 8) as high-pathogenicity avian influenza H7N3 (IVPI = 2.96). Conversion from low pathogenicity to high pathogenicity was attributed to recombination between hemgagglutinin and matrix genes., dEfforts at containment had failed: a second farm (located several kilometres from the index farm) was found to be infected with high-pathogenicity avian influenza H7N3. eThe Canadian Food Inspection Agency (CFIA) officially notified the Office International des Epizooties of the outbreak of high-pathogenicity avian influenza on Mar. 13. The United States closed its border to BC poultry and poultry products. fOnset of unilateral conjunctivitis in unprotected workers following direct conjunctival contact with infected birds during culling on Mar. 13 (case 1, farm 2) and Mar. 22-23 (case 2, farm 3). Virus isolated from the nose of case 1 was low-pathogenicity H7N3 and from the conjunctiva of case 2 was high-pathogenicity H7N3. gBy Apr. 5, 20 commercial flocks were infected. Pre-emptive depopulation of all poultry in the Fraser Valley was announced by the CFIA. Most infected commercial flocks (34/42 or 81%) were detected between Mar. 21 and Apr. 24, 2004. hBy Apr. 23, a total of 45 trading partners had taken action against Canada: 29 against all of Canada, 16 against live poultry and poultry products from British Columbia. iA commercial duck farm without signs of infection was identified preslaughter as being infected with low-pathogenicity H7N3. jIn total, 11 backyard flocks were deemed infected; the last was detected and culled on May 18. kIn total, 42 commercial flocks were deemed infected; the last was detected on May 18 and culled on May 20. lAfter the events depicted in this chronology, the CFIA declared all depopulated premises eligible for restocking on July 9, the United States lifted trade restrictions on Aug. 17, and the European Union lifted restrictions on Oct. 1.