Avian Pathologist
BC Ministry of Agriculture, Food & Fisheries
Animal Health Centre, Abbotsford, Canada
In early February 2004, an H7N3 avian influenza virus of low pathogenicity was isolated from a 52-week-old broiler breeder flock in the Fraser Valley near Abbotsford, British Columbia. This flock of 9,000 birds had experienced a sudden onset of lowered feed consumption rate, slight increase in mortality (0.5%) and a 20% drop in egg production. Suspecting a feed toxicity, the referring veterinarian submitted dead birds to the provincial veterinary diagnostic lab (Animal Health Centre - AHC) in Abbotsford for diagnostic work-up. Immediate feed replacement did not result in resolution of clinical signs. At necropsy, gross lesions were non-specific and included patchy congestion of skeletal musculature and lungs that were edematous with a "meaty" texture. The tracheas were slightly red and granular. A viral respiratory work-up was initiated that included virus isolation and PCR testing for IBV, ILT, NDV and AIV. Concurrently the suspected load of feed was being tested for mycotoxins and feed mixing errors. By the time the preliminary diagnostic test results were completed the mortality, feed consumption rate and egg production in this flock were beginning to return to normal.
On February 13, 2004 lung samples positive for ORT and AI by PCR were reported to the referring veterinarian who immediately contacted the producer; both initiated "self-quarantine" until the virus could be isolated and subtyped. On February 16, 2004 an AI virus was isolated in embryonated eggs and, as legally required, the laboratory notified the district veterinarian for the Canadian Food Inspection Agency (CFIA). By this time the affected barn was returning to normal production so the samples were not considered urgent. That night the adjacent barn of 9,000 24-week-old broiler breeders experienced a sudden spike in mortality (~100 birds), thought at that time to be related to a recent "barn disturbance". When the mortality in the second barn continued to climb dead birds from this younger flock were submitted to the AHC for necropsy. The gross lesions in the younger birds were similar to the older flock but more pronounced with severe hemorrhagic tracheitis and pulmonary edema. Cultured virus from the older flock and affected lung tissue from the younger flock were handed over to the CFIA veterinarian on February 17, 2004, who then expedited the samples to the National Centre for Foreign Animal Disease (NCFAD) Laboratory in Winnipeg, Manitoba for confirmatory virus identification and pathogenicity testing.
Subsequent test results from this index farm indicated that the AI virus isolated from the older flock was an H7N3 of low pathogenicity (LPAI) but the AI virus recovered from the younger flock was a highly pathogenic (HPAI) based on intravenous pathogenicity testing (IVPI= 2.87) and gene sequencing of the hemagglutinin cleavage site which contained a 21 base insertion sequence. It appears that the LPAI from the older flock had mutated into HPAI by gene insertion at some point between the older and younger flocks. It is important to note that although the massive mortality in the younger flock (25% in 48hrs) was highly suggestive of HPAI, this was not confirmed until March 9, 2004, a full month after the laboratory submission from the first flock.
During the 4-week time period between suspicion of AI and laboratory confirmation of HPAI several key events occurred. The self-quarantine on the index farm was initially considered to be effective since there were no new infected premises detected until March 9, 2004, although there had continued to be routine feed deliveries and egg collection services on and off the index farm up until the time of depopulation. On February 19, faced with an H7 subtype with an as yet unknown pathogenicity isolated from one barn and growing mortality in an adjacent barn, the CFIA made an extremely important precedent-setting decision to depopulate the index farm based solely on the H7 subtype. This was a direct result of the current OIE discussions making all H5 and H7 AIV subtypes subject to eradication regardless of pathogenicity since recent global occurrences of HPAI have indicated the ease with which H5 and H7 LPAI can mutate into HPAI.
The methods of euthanasia and depopulation of the two barns on the index farm, as well as the subsequent 2nd infected farm, had never been attempted before in Canada and posed significant challenges. On the index farm, barns were not completely sealed off as carbon dioxide gas was pumped in and fans continued to ventilate both before and after euthanasia, which may have assisted any airborne spread of the virus onto adjacent roadways. With the initial intention for on-farm composting, the dead birds were ground up uncovered in a mobile industrial feed mixer and open loads were moved by tractor into an outside dairy feed bunker. Capacity was reached at 60% of the birds so the remaining carcasses were placed into sealed totes in reefer trucks and transported 200km away for incineration. An alternative method of euthanasia was utilized on the second affected farm, 1.5km away, which became positive approximately 2 weeks following the depopulation efforts on the index farm. The second farm, with 4 barns containing a total of 24,000 broiler breeders, was depopulated by conveying birds outside in the open air into a portable electrical stunning chute designed to dispatch spent fowl prior to rendering. This method was unsatisfactory due to the awkward inconsistent kill and the significant feather-fly generated as evidenced by the feather debris settling on the nearby-parked cars. This method of euthanasia was consequently abandoned with a return to the more efficient in-barn CO2 gassing.
The CFIA declared HPAI to the OIE on March 10, 2004 and established a 5 km High Risk Region (HRR) around the index farm with an additional circumferential 10km Surveillance Region (SR). The City of Abbotsford assisted in road closures and funneling any exiting HRR traffic through decontamination stations but only the tires of vehicles that admitted to have actually visited poultry farms were disinfected, the rest were waived through. In retrospect, if there had been roadway contamination of virus secondary to the depopulation activities on the first two farms, selective disinfection would have been an ineffective way to contain the virus to the HRR. Within two weeks HPAI had spread to a total of 6 commercial premises and 1 backyard flock [This backyard flock of 30 chickens had tested negative for any AI during HRR surveillance but became clinically affected 3 days following the depopulation of the 3rd infected premises that was 400m down and across the road. This was the only backyard flock from which HPAI was ever recovered (550 backyard flocks were killed & tested). Compelling evidence that biocontainment was not achieved during the depopulation of the positive HRR commercial premises.] within the HRR. On March 30, 2004 HPAI was diagnosed by the Animal Health Centre (AHC) in a commercial table egg flock approximately 15km away from the HRR in a dense cluster of layer farms (as well as being within 2km of an international airport).
On March 11, 2004, by Ministerial Order, the entire Fraser Valley of British Columbia was defined as a Control Area (CA) with significant restrictions imposed on the movement of poultry and poultry products. This had a serious immediate economic impact and within one week all Control Area processing storage capacity was reached at 10M kg with 500K kg per day continuing to come in. The Control Area contains approximately 17-18 million commercial poultry birds at any time and produces 84% of the domestic consumer requirement for the province.
The CFIA, under the Health of Animals Act is mandated to eradicate foreign animal disease with the support and assistance of provincial agencies and the local poultry industry based on pre-determined animal disease emergency response plans. Throughout the outbreak there were huge operational challenges to many aspects of the eradication effort which, while taking up valuable time to sort through, allowed the disease outbreak to progress at an alarming rate. Better defined disease response plans may have allowed less time to be wasted in debate and the chain-of-command decision relay since they could have provided more clearly outlined role definitions, designated responsibilities, protocols, the decision-making hierarchy and responsive contingency plans.
The AAVLD-accredited provincial veterinary diagnostic laboratory, the Animal Health Centre, is a full service diagnostic lab with resident expertise in avian diseases and avian virology. The lab location is within the Control Area, 10km from the HRR. This lab made the initial diagnosis of AI in the index, 2nd and 3rd infected farms and as well as identified the first positive flock outside the Surveillance Region. Despite the capability of an AI-PCR result turnaround time of 6 hours, the CFIA insisted that all diagnostic samples had to be confirmed at the NCFAD in Winnipeg before a decision to depopulate a flock could be made. This decision resulted in a crucial delay in testing and response (up to 10 days in the case of the third infected farm in the HRR), and possibly allowed so much virus to be generated in these heavily stocked barns that subsequent biocontainment would be almost impossible to achieve. It wasn't until well into the outbreak that CFIA agreed to accept the results of the AHC and allow them to assist in the diagnostic testing.
When dealing with a highly communicable disease such as HPAI, disease response plans should clearly dictate that flocks are to be depopulated immediately and without hesitation upon first suspicion and that the confirmatory testing that follows can then be used for the subsequent epidemiological analysis. The local expertise, including poultry industry leaders, accredited and experienced poultry veterinarians and emergency response logistics officers were not identified early or constructively used in the strategy or decision-making process until very late in the outbreak. On April 5, 2004 (Day 48), with 20 commercial flocks now positive, the Provincial Emergency Program (PEP) was activated. Working from an "incident command" emergency management system, which had not been utilized by the CFIA, the PEP team was able to efficiently manage the logistical challenges of the outbreak in a much more effective and expedient way. At the same time the poultry industry assisted the massive pre-emptive cull effort of the entire Fraser Valley declared by CFIA on April 5, 2004. Producers and processors worked together to systematically clear all test-negative barns through the normal processing channels (slaughter or rendering), which then freed the CFIA to concentrate on the disease control strategy and the management of the outbreak. All commercial barns and hatcheries in the Abbotsford area were emptied by May 21, 2004. Cleaning and disinfection of all infected premises, at a cost of $3 million Canadian to the producers, was completed and signed off by CFIA inspectors in mid-June. On July 9, 2004 the Federal Minister of Agriculture declared that restocking of Fraser Valley poultry barns could begin. It is estimated that the BC poultry industry will not experience full production recovery for at least 18 months.
The recovery aspect to this outbreak began even before the barns had been emptied. Teams comprised of members from industry, provincial and federal governments worked together to address the issues of greatest importance: compensation, restocking strategy, enhanced biosecurity and modification of current disease response plans based on the lessons learned from this outbreak.
The last positive farm was detected May 18, 2004, just over 81 days after initial detection, with a final total of 42 commercial premises being identified as positive in this outbreak. Through the combined efforts of governmental agencies and the poultry industry itself, this HPAI outbreak was contained within the Fraser Valley and the objective of eradication appears to have been achieved.
The original source of the LPAI for the index farm is unknown. Retrospective serology of banked broiler breeder sera combined with the intense industry-wide surveillance testing during the outbreak did not reveal appreciable amounts of endemic AI (H6 and H11 were detected in commercial domestic ducks). The majority of backyard flocks, which should serve as useful sentinels for environmental AI contamination by waterfowl, were negative. Since outbreak testing was focused on the Abbotsford area it is possible that the source of LPAI (H7N3) was from on outlying untested area and future surveillance for AI should be expanded to include these areas.
Lessons learned from the BC outbreak of HPAI
- Never underestimate an Avian Influenza virus.
- Eradicate any H5 or H7 AI regardless of pathogenicity.
- Targeted surveillance and routine diagnostics is the key to early detection.
- A producer "self-quarantine protocol" is essential to initial containment.
- Breaches in biosecurity, especially related to vehicle traffic on and off commercial farms, perpetuate the spread of infectious disease. In this outbreak backyard flocks were NOT part of the disease dynamic, possible because they have different contact & patterns of movement.
- Early in an outbreak identify local expertise and utilize them in the decision-making process.
- Avoid delays in testing or response:
Pre-accredit local veterinary diagnostic labs and allow them to do testing.
Depopulate upon initial suspicion, confirm later. - Do not move infected birds outside of barns or off the premises until heat or time treated to destroy the virus.
- Disinfect all vehicle traffic exiting a controlled area.
- Animal Disease Emergency Response Plans should:
Detail roles & responsibilities
Provide clear & specific protocols for:- Humane mass euthanasia
- On-farm carcass handling & disposal
- Management of infected litter
Have backup contingency plans. Play "what if?"
Consult with Public Health authorities to include provisions for working with potentially zoonotic diseases like Avian Influenza.
Consult with those who have experienced the challenges of an outbreak.
- Activate Emergency Management teams early.
- Communications issues are extremely important. Take advantage of the industry's own information infrastructure.
- Carefully document the outbreak events (there will be a review).
- Industry input & cooperation is critical!
