NARRATIVE BACKGROUND TO CANADA'S ASSESSMENT OF AND RESPONSE TO THE BSE OCCURRENCE IN ALBERTA

July 2003

INTRODUCTION

As a signatory to the charter of the Office international des épizooties (OIE) and supporter of the International Animal Health Code chapter on bovine spongiform encephalopathy (BSE), Canada adheres to, or surpasses, the guidelines established for assessing and managing associated risks of BSE and other transmissible spongiform encephalopathies (TSEs).

In the most recent iteration of its assessment of exogenous and indigenous BSE risk, Canada in 2002 acknowledged the risk of its being exposed to the disease as very low and the potential for its amplification and transmission as negligible. This mirrored the findings and conclusions of assessments of Canada's status conducted by other countries in recent years. The conclusion was reached on the basis of BSE risk management actions underway since the early 1990s to limit the potential for its entry and establishment within not only Canada but North America, as described in the special edition (April, 2003) of the OIE Revue Scientifique et Technique entitled "Risk analysis of prion diseases."

BSE surveillance has played a key role in Canada's risk management strategy for more than a decade, engaging both federal and provincial governments. In recent years, in concert with its OIE peers, Canada's surveillance regimen has markedly improved in the form of enhanced case definitions and industry penetration, improved diagnostics, and increased volumes which have far exceeded the OIE's minimal guidance. Alberta has led its provincial peers in these areas.

According to the BSE Chapter of the International Animal Health Code, on the basis of its risk assessment and management actions, Canada qualified as provisionally free of BSE risk and was scheduled to attain free status following two additional years of the meat and bone meal (MBM) feed ban introduced in July 1997. To safeguard that status, Canada limited its imports of TSE-susceptible animals and products to nations similarly recognized, with the U.S. being its preferred source.

In the mid to late 1980s and early 1990s, like many OIE member countries, Canada found itself in possession of ruminants imported from the U.K. and other European nations during the interval that preceded their recognition as BSE-affected. In 1993, a beef cow imported from the U.K. expressed BSE clinically in the province of Alberta, as a downer animal within the importing herd. The detection of the animal was the direct result of Canada's active efforts dating to 1990. At that time, the disease had been made officially notifiable, and a tracing and monitoring program had been instituted to identify and periodically examine all animals previously imported into Canada from the U.K. prior to the suspension of live cattle imports in 1989. Canada's rapid despatch of the herd of residence of the 1993 index case, the infected animal's progeny, and the balance of its fellow U.K. imports is chronicled in the documents described earlier in this report.

As did the U.S. and many other trading nations, Canada determined in 1993 that some members of the imported cohorts had been re-exported or had already made their way into the domestic feed chain via natural attrition. In accordance with its international obligations, Canada provided notification to any country receiving such animals.

In the wake of the Spongiform Encephalopathy Advisory Committee report of 1996 on the apparent zoonotic potential of BSE, Canada and the U.S. invoked an indigenous mammalian-to-ruminant feed ban that exceeded the measures recommended by the World Health Organization to augment existing risk management of BSE from exogenous sources. The federally regulated systems target renderers and feed mills in recognition of their elevated positions in the pyramidal clusters which typify the renderer-feed mill-producer relationship in North America. One renderer characteristically supplies approximately 20 feed mills, which have perhaps 4,000 customers. The most cost-effective enforcement approach would dictate that inspection concentrate on achieving the impact of full compliance at the renderer and feed mills in each cluster.

On January 31, 2003, a beef cow (which had a previous unrevealed period of abnormal behaviour) was found recumbent and unable to rise, was hoisted onto a truck and delivered to a provincially inspected abattoir in the Peace River District of Northern Alberta. On the movement manifest provincially required for transport, it was described only as a "black cow," with no additional distinguishing natural or artificial features recorded. The inexperienced producer had not sought veterinary attention and shipped the animal in an attempt to salvage meat for personal use. The animal qualified for BSE surveillance under the active national program and its head was forwarded to the Province of Alberta for assessment and incineration. The carcass was condemned because of pneumonia. Under Canadian program requirements, any carcass destined for human consumption that is subjected to TSE testing must be held, pending reporting of the test results. Since the carcass was condemned and it was not allowed to enter the human food chain, the carcass was not held and was instead sent to rendering, where it entered into the animal feed chain.

National surveillance initiatives, when aimed only at reconfirming what is already believed to be disease freedom, are ranked in priority second to other demands which may entail direct and immediate public health consequences. They also compete with the multiple other diagnostic urgencies which confront national and sub-national veterinary authorities engaged in active disease control programs. For that reason, it was not until May 16, 2003, that a tentative diagnosis of BSE was made at the sub-national level. Federal review followed immediately, with the Weybridge Central Veterinary Laboratory in the U.K. employed within 48 hours as the ultimate confirmation. The OIE General Session, then in progress, was notified immediately.

The discovery launched a multi-pronged epidemiological investigation into the "micro" event and its immediate management, elucidation of its most probable origins, and the longer-term risk management of the "macro" situation to which the index case was sentinel. Major policy transitions are underway in response to the event, towards further clarifying and managing whatever residual BSE risk remains in Canada following six years of MBM feed ban application. The following paragraphs describe all aspects of the event in greater detail.

MICRO INVESTIGATION - Animal Health Aspects

The index case was of the Angus breed and a member of a herd of 80 cows established in the interval 2001-2002 from two streams of cattle. It is most probable, based on the investigation to June 24, 2003, that she was six years of age. Her expression of BSE at that advanced age offers the incident's first epidemiological insight, that into the probable low level of BSE that she had contacted, based on dose-response incubation curves established in the U.K. for exposures as low as 1, 0.1 and, most recently, 0.01 grams of infected CNS material.

Before coming together in the affected herd, the two streams had independently flowed through 8 defined production units. There were a number of animals recorded in the course of flow of the two streams that might have been the "black cow" described in the shipping manifest that accompanied the index case to slaughter on January 31, 2003. By late May 2003, a possible 80 animals had been reduced to a most probable 7, through an exhaustive and repetitive examination of natural and artificial identification, provincially regulated shipping manifests, breed registries, and the memories of owners and service sector workers. The search included the identification and testing of an exhumed index herd carcass, the confirmation of inadvertent mixing of the index herd with a neighbouring herd of cattle and the independent interviews of four persons associated with the index case during the key interval of its illness and slaughter.

The investigation respected the most recent recommendations coming from the OIE's BSE Ad Hoc Working Group and buttressed them with a "precautionary perimeter" of additional measures aimed at compensating for human nature, cattle identification limitations, and natural variability within BSE epidemiology. It followed parallel animal health and feed investigatory paths, with extensive and independent questioning of each producer by epidemiologists, field veterinarians, and feed experts.

On the animal-health side, the limits of memory and artificial identification were augmented by the application of DNA testing throughout the stream of animals determined the most probable (95% probability) source of the index cow. DNA from the index case's brain was compared with semen, blood from living cattle and breed registry records of dead ones. The testing to June 24, 2003, has tentatively ruled out several of the herds as the animals of origin, but only to the degree that owners have been able to advance specific cattle as the putative sires and dams of the index animal.

While the examinations continue, a DNA-driven depopulation decision tree has already led to the destruction of epidemiologically associated stock. While potentially subject to criticism in respect of the strictest interpretation of BSE epidemiology, the approach reflects Canada's commitment of zero tolerance for all TSEs (OIE Revue Scientifique et Technique, April 2003) and the limitations faced in assessing an exposure that in all probability took place at least six years before.

An eradication template - based on current understanding of the disease and the best possible knowledge of the cohorts which might have accompanied the index case during its interval of BSE susceptibility - was applied to all epidemiologically related herds. An additional template was created for cohort members that had moved from the herds prior to the identification of the occurrence.

During the investigation, an implicated herd was determined to have sold to slaughter an offspring of a cohort member. A decision template based on DNA confirmation of the offspring's dam and the assessment of her BSE status as negative on testing was applied to the event and precluded the need for an extensive feed investigation.

In total, 15 premises were quarantined as part of the tracing-back and tracing-forward investigations. An additional 25 herds were scrutinized in the tracing-out of single animals or cohorts from the primary line of enquiry, which included the identification and notification of the export of five animals to the U.S. in 1997. The application of the above-described eradication templates resulted in the culling of more than 2,700 cattle. Among them, the more than 2,000 animals 24 months of age or older were tested and found negative to both Prionics Western Blot and immunohistochemistry techniques.

MICRO INVESTIGATION - Feed Aspects

Parallel investigation determined two potential MBM exposure routes within the epidemiologically associated herds from the interval of concern. One took the form of a feed concentrate. The other was a high energy feed block employed to supplement pasture grazing. Investigation of feed mill records and compounding formulae confirmed that MBM incorporation in both products was curtailed in 1997 upon implementation of the MBM feed ban.

The carcass of the index case was traced through the abattoir-renderer-feed mill- producer continuum to its direct allocation into pet food and poultry meal and its additional retail distribution across 1,800 farm sites. As earlier described, the associated cluster is typical of the pyramidal feed production and distribution relationship in Canada. Visits to the renderer and feed mills confirmed adherence to the MBM feed ban legislation on product receipt, segregation, labelling and distribution.

The median infective dose (ID50) per tonne of each exposed commercial product was determined as part of the assessment. While the levels were low and subject to incremental dilution in subsequent feeding regimens, acceptance of the possibility of non-random clustering of infected particles prompted a stratified, statistically based on farm study of the risk distribution across 600 bulk feed recipients and 1,200 bagged feed consumers and the associated site-specific risk levels. The survey suggested that 99% of sites experienced either no exposure (96%) or incidental exposure (3%). The exposures representative of the other 1% of the population were expressed through the likes of a leaking mixer, sheep with fence-row access to swine feed, a feed pile break-in by cattle and a goat with access to a feed bag. The recent nature of the exposure, the limited cross-population exposure distribution and site-specific risk levels, in the face of the ensuing specified risk material ban and feed ban enhancements, denied the need for remedial action beyond those taken against the specific infractions cited.

Through this process, three additional farms were quarantined when investigation could not preclude the exposure of 63 head of cattle to feed destined for poultry. The animals were culled and tested negative with the diagnostic regimen described above.

MACRO INVESTIGATION - RETROSPECTIVE - Animal Health Aspects

The immediate occurrence in hand, investigation turned to the more encompassing significance of the sentinel finding. As a first step, employing the best estimate of age and origin in accordance with feed distribution clusters and other information provided in preceding paragraphs, a polygon was constructed to describe the tentative geographic and temporal cattle cohort to whose BSE exposure the index case was sentinel.

Without specific reference to the population size encompassed, epidemiologists defined the probability distributions of its cattle classes in 1997, the most probable year of exposure of the index case, based on demographic studies previously conducted in western Canada. From the same and additional sources, they then plotted the net contribution of those same classes to the same geographic cohort in June 2003. While the composite remaining cohort constitutes approximately 17% of the population of the area today, the age class of particular interest, that which by virtue of its age was particularly BSE-susceptible in 1996-1997 and potentially exposed to feed sources in advance of the 1997 feed ban, now constitutes an estimated more diminutive 5.9%.

Of parallel interest in respect of the macro aspects of the investigation were the interim dispositions from the geographic and temporal cohort. Western Canadian demographic studies indicated that 5% of the adult population was slaughtered each year in the U.S., 5.7% within Canada, and 1 to 1.5% was exported for breeding purposes, primarily to the U.S. As an assistance to that major trading partner, the investigators plotted the detailed paths of distribution into that country of slaughtered cohort members during two widely spaced years of interest. The other 1 to 2% of adults from the temporal and geographic cohort died on the farm, constituting the targeted BSE surveillance group advocated by the OIE, from which derived the index case itself.

The epidemiologists next addressed the possible means of exposure to which the index case was sentinel, by reference to the North American cattle cycle. Were the animal born and raised entirely within Canada, its most probable potential exposures would have included its own dam; feed contaminated by the U.K. cohort; domestic or sylvatic cervid chronic wasting disease (CWD) or TSEs resident within other indigenous or imported ruminants alluded to earlier; or imported MBM.

While not fully discounted, direct exposure to cervid CWD is considered a lesser possibility. Were any portion of the BSE-susceptible interval of the index case's life spent outside Canada, the same exposure could have occurred in the other country(s) involved, depending on its indigenous and exogenous exposures to the same influences. Were the index case in either instance infected by its dam, the case would be sentinel to the same exposure possibilities, but backdated by the two or more years of its dam's geographic area(s) of maturation prior to eventually calving what became in 2003 the index case. Spontaneous development was assessed and ostensibly ruled out on the basis of a molecular genetic evaluation of the index case conducted by the Canadian Food Inspection Agency in conjunction with Health Canada.

The previously conducted risk assessments and risk management papers described earlier address all known TSE exposure possibilities. Scrapie is not considered in the current document, in deference to the two to three log lower Canadian and U.S. magnitude of the composite quantitative risk factors that suggested its potential as an aetiological source of BSE in the UK. Wilesmith refuted the possibility that BSE would derive from scrapie in an environment beyond the exceptional circumstances experienced in the UK. Also not considered are the more limited importations of other European ruminants from nations of considerably lower BSE prevalence. Instead, attention is focused on the more compelling importation of MBM and two specific groups of ruminants (cattle and cervids).

Earlier reference was made to the entry into the feed system of members of the U.K. cattle cohort from which appeared Canada's 1993 case of imported BSE. Those animals entered the Alberta feed system in greater volume than elsewhere in Canada. Given the clustering of renderer catchment areas with recipient feed mills and producers described earlier, it is reasonable to believe that BSE introduced into the western Canadian cluster(s) involved would cycle and develop within it (them), until sentinelized by the index case that expressed clinically its exposure within the last cohort of susceptible calves (1997) to have emerged before implementation of the feed ban. Only to the degree that exposed animals and MBM-bearing compounds left them would the problem disseminate. In that regard, the nature of commerce in cattle in North America would predicate extension into the northwestern U.S. as opposed to eastern Canada.

The theory is being further explored in several ways. In 1993, Canada preserved the brains from the imported U.K. cohort. They are being reassessed to determine, with sensitive diagnostics unavailable at the time, the degree to which the BSE threat may have extended beyond the single clinical case within clinically normal animals. In addition, Canada has maintained a liaison with the U.K. to determine whether the herds of origin of the imported cohort, unaffected by BSE as of 1993, have since demonstrated the disease. The list is being revisited on a provincial basis to further assess the relative probability, test results notwithstanding, of latent BSE carriage among the entries. Finally, through the application of the Canadian Cattle Identification Program, samples derived from the enhanced surveillance program being now planned will be ascribed on the basis of the feeding clusters described for comparison with cluster-specific U.K. attritions.

In a general examination of the event, another source of imported cattle bears consideration, albeit not by virtue of its ostensible BSE risk. The U.S. shared Canada's provisionally BSE-free status on the date of diagnosis of the index case and is itself en route to free status by July 2005. By reason of physical proximity, and shared freedom from devastating livestock diseases and the equivalent control of others, Canada has traded TSE-susceptible species (cattle, sheep, cervids, mink) and products freely with the U.S. in a reciprocal acceptance of status equality. Of Canada's cattle imports, 99% are of U.S. origin. That nation's contribution of 1.3 million cattle, almost exclusively to the geographic area of interest and during the specific interval of epidemiologic concern, is well documented.

Particular emphasis falls upon a special importation of 0.5 million head, based on a seasonally regional harmonization of anaplasmosis, brucellosis, tuberculosis and bluetongue disease control status between the northwestern U.S. and Canada's western provinces. Supported by a formal risk assessment and consistent with the letter of the governing legislation, but potentially compromised by a liberal protocol interpretation, approximately 25,000 pregnant females were moved from feedlots, with the loss of their U.S. identification, in advance of the implementation of the Canadian Cattle Identification Program, and made their way into the western Canadian cattle population. Of these, 70 to 80% were described as being of the black Angus breed. By joining the feed distribution clusters described in previous paragraphs, they too would have entered the Canadian feed chain. Although DNA investigations continue, Canada cannot, to date, exclude the possibility that the index case itself derived through this huge, unique importation; nor were these the only TSE-susceptible ruminant imports.

During the 1980s, entrepreneurs in Alberta and Saskatchewan had imported considerable numbers of domestically raised elk from the northwestern U.S. Unrecognized at the time was the distribution, within their populations of origin, of two diseases. In 1991, an Alberta herd became sentinel to the presence of bovine tuberculosis within the North American cervid industry. Canada eradicated the disease through the depopulation of much of Alberta's imports. The Saskatchewan population was considerably spared in the findings of the associated epidemiological investigation.

In 1996, a Saskatchewan elk herd became similarly sentinel to the North American presence of CWD in commercially raised cervids. Although the disease had been recognized in wildlife research facilities in Colorado and Wyoming decades before, its ingress into commercial North American elk herds had remained unrecognized until the Canadian investigation signalled its presence. The disease was eliminated from multiple herds in its primary epidemiologic focus in Saskatchewan, concentrated, coincidentally, in the area considered to date as one of the most probable origins of the index case. The few sites considered heavily contaminated remain cleared of all livestock and wild cervids. An epidemiologic study of the outbreak's progression continues. To date, depending on owners' election to bury or recycle, the study has identified the possibility that as many as 110 CWD-infected elk carcasses might have entered the feed distribution clusters shared by the prospective herds of origin of the index case during the interval 1993-1999 before the regional renderers ceased acceptance of all TSE-susceptible materials.

As in the U.S., though at a much reduced prevalence, wild cervids near CWD-contaminated sites in Saskatchewan have been found infected. The possibility that they might transmit the disease directly to intensively commingling cattle has been explored, with negative results to date in protracted studies in the U.S. A very large prolonged natural experiment created by the cohabitation of the two species on rangeland in the northwestern U.S., at a higher CWD prevalence, has yet to produce a BSE diagnosis in the U.S. While intracerebral injection has shown the susceptibility of cattle to CWD, other in vitro and in vivo attempts, including those described, have had limited success. The prion within the index case was molecularly characterized at Weybridge as BSE and not CWD.

It is to imported MBM, however, that the World Health Organization and the Food and Agriculture Organization of the United Nations direct their strongest admonitions regarding international BSE risk management. On the basis of other OIE List A diseases, long before the recognition of BSE in the U.K., Canada had banned importations from areas of subsequent BSE concern. As described earlier, cost considerations and its BSE risk management efforts had restricted Canada's MBM imports, comprising half its annual usage, to sources within the U.S. That nation's parallel importation and indigenous TSE experience suggests that MBM from the U.S. would have suffered the same potential TSE exposures, in the nature of scrapie, CWD and the imported U.K. and European ruminant cohorts, as that produced in Canada. Towards elucidating the relationship, the investigating epidemiologists plotted actual import sources and volumes for consignments of MBM received by western Canada in 1997, a key interval of epidemiologic interest sentinelized by the index case.

The possibility of the U.S. being the origin of the index case or the source of contaminated feed remains among the multiple avenues of continuing study, and might be best defined by ongoing surveillance being conducted in the U.S. Reciprocally, like all OIE member states, Canada bears a responsibility to its trading partners, the U.S. in particular, in helping them to manage the risk of potentially contaminated animals and products from the temporal and geographic cohort sentinelized by the index case. Earlier reference was made to the general demographic flows of fed cattle, breeders and slaughter cows into the U.S. from western Canada. As a specific assistance, the epidemiologists established the site-specific distributions of slaughter cows during two pivotal years. Given the nature of the North American Free Trade Agreement (NAFTA) trading relationship, it would not be unreasonable to assume the redistribution into western Canada of MBM created from their carcasses. It is these "tidal" flows of TSE-susceptible animals and products and the feed distribution clusters involved that suggested the general geographic polygon of exposure earlier described.

It was the recognition of such commercial interdependence and shared TSE risk that prompted the co-ordination of TSE risk management across NAFTA, as described in the April 2003 special edition of the OIE Revue Scientifique et Technique. The trilateral group has already met towards adoption of a harmonized response to the occurrence in Alberta.

MACRO INVESTIGATION - RETROSPECTIVE - Feed Aspects

Recognition of the feed distribution clusters earlier described has permitted determination of the general geographic polygon of distribution of the MBM which most probably exposed the index case to BSE. When further defined, it will facilitate, via the Canadian Cattle Identification Program, BSE-exposed, cohort-specific surveillance tracking. In the interim, it has contributed to the identification, within the clusters involved, of the presence of two nutritional compounds that bore MBM in the interval of exposure. One took the form of high energy range blocks. The other comprised a high protein concentrate.

In an era of dioxin and other risks, excellent record systems retained by the feed industry for personal liability protection surpass those required by regulatory authority. Those sources permitted extensive characterization of the substances and their commercial histories. Inspectors confirmed cessation of MBM usage in both, upon the coming into force of the MBM ban in 1997.

MACRO INVESTIGATION - CURRENT

In the context of the clinical expression of BSE following six years of MBM ban and a decade of other risk management measures, a series of universal questions bears answering. Those regarding putative exposure sources have already been addressed. It is now important to examine why the disease should have been found when and where it was, in an animal of the breed described, as guidance to future risk management activity. The answers draw upon the nature of surveillance regimens, cattle demographics, the feed ban itself, and the underlying epidemiological characteristics of BSE.

The discovery can be attributed in part to the cumulative growth in BSE surveillance, with a pronounced increase in recent years. The enhanced case definition and diagnostic method can only have contributed positively. Alberta stands out among Canadian provinces in terms of the sample volumes it generates, while carrying an additional statistical advantage by virtue of its larger anticipated absolute volume of high risk animals. Alberta's numerical predominance within the Canadian cattle population, beef's predominance within that provincial herd, and the extended age demographic in beef cattle (compared to dairy) help rationalize the occurrence. The long incubation associated with low dose exposure, and the occurrence within the largest remaining pre-1997 MBM ban age cohort, which would have experienced the highest absolute level of BSE exposure, further support the finding.

All other things being equal, what remains of course, irrespective of the preceding, is the lingering question as to why the disease would not have first expressed in a dairy animal if, by common knowledge, that sector was exposed to exponentially higher volumes of MBM. The answer may reside in any of five areas. The first relates to the minimal exposure required to produce infection. It may be that beef animals, while exposed at a markedly lower dosage, might nevertheless have received the minimal threshold volume required for infection. Continuing research in the U.K. credits infection to such ever-diminishing challenges. The second relates to the effective calfhood window of vulnerability; dairy animals may be exposed to massive amounts of MBM later in life without incremental impact. The third relates to dairy demographics in North America. It might be that dairy animals exposed to admittedly larger doses were nonetheless slaughtered for other reasons prior to BSE expression. Fourthly, the dosage - response curve might explain the longer incubation in a beef animal which survived long enough, nevertheless, to express the disease. Finally, the prevalence of BSE may have been so low as to have defied, except by chance, detection against the background noise of the myriad other CNS ailments that afflict North American cattle.

MACRO INVESTIGATION - PROSPECTIVE

The question then arises as to what can be expected in the coming years. The investigating epidemiologists recognized that the Canadian epidemic curve would have paralleled that of the U.K. (irrespective of its absolute magnitude). As described in the Harvard Risk Assessment, they believe it would have peaked or already entered into decline at the time of discovery of the index case. They further believe that any additional cases uncovered in the enhanced surveillance now in planning will reflect the pre-MBM ban era and the low dosage exposure experienced.

They suggest that should one or more further animals be found, if surveillance patterns hold, it may well emerge again in western Canada among the beef breeds. To enhance Canada's ability to better describe the true epidemiological picture, they are structuring an enhanced surveillance approach prioritized across a spectrum of access points, which reflects the nature of the North American cattle cycle. At the same time, they have met with the Harvard Risk Assessment Group to facilitate its adjustment of the North American BSE exposure model in response to the occurrence.

At the same time, program adjustments are underway towards further safeguarding the human food and animal feed chains, through consideration of restrictions on specified risk materials prescribed by the OIE, along with enhancements to both the Canadian Cattle Identification Program and the MBM ban. All the preceding continues in recognition of the fact that, while clinical expression is still the principal means of identifying BSE presence, the clinical events signal a more encompassing epidemiological manifestation, against which these preventive measures are justified.

INTERNATIONAL SCRUTINY

In recent weeks the above findings were presented to an expert international team, comprised of distinguished representatives from three continents whose final report on the Canadian approach is now made public.

In their interim report, the expert international panel applauded the investigation as unprecedented in scope and rapidity and having been achieved because of the infrastructure already in place to respond to such unforeseen events.

The team noted that the depopulation activity had achieved its optimal utility and that efforts should focus on the further elaboration of amended priority policy options. It was the panel's view that the areas of specified risk material removal, enhanced targeted surveillance to further confirm the very low prevalence of the disease and benchmark the effectiveness of the control measures over time, and further consideration of feeding practices to preclude any potential cross contamination were indeed appropriate, based on the collective international experience.