Assessment of producer conducted antemortem inspection of market pigs in Australia.

OBJECTIVE: A field trial was conducted to evaluate the effectiveness of on-farm antemortem inspection performed by Australian pig producers in terms of proficiency in detecting unhealthy animals (suspects), the impact on food safety and animal welfare outcomes. ANIMALS: A total of 64 trial groups of market pigs (n = 10,703) were dispatched from 20 herds to three abattoirs in three states. PROCEDURE: All producers and abattoir antemortem inspectors were trained in antemortem inspection prior to the trial, using a standardised set of antemortem criteria. Study 1 consisted of three comparisons: on-farm versus abattoir antemortem inspections on the same pigs, transport effects in suspect and normal pigs, and meat rejection and carcase disposition in suspect and normal pigs. Study 2 compared Salmonella spp. status of suspect and normal pigs. Studies 3 and 4 were conducted to assist the interpretation of the results of on-farm versus abattoir inspection. Antemortem inspection results of three producers and one abattoir antemortem inspector were compared with those of a reference inspector (Gold Standard). RESULTS: Of the pigs examined, 2.12% (95% CI: [1.84%, 2.42%]) were classified as suspect at on-farm antemortem inspection compared to only 0.14% at abattoir antemortem inspection. Forty one percent of pigs classified as suspect on-farm had locomotor problems (arthritis, lameness and foot abscess). Compared to normal pigs, suspect pigs had a relative risk of 67.7 (P < 0.0001) of suffering transport injuries. Meat rejection amounted to the equivalent of 9.68% of suspect carcases, compared to the equivalent of 0.35% of normal carcases. The isolation of Salmonella species from caecal content from 20.9% of 67 suspect pigs and 18.8% of 133 control pigs (two matched healthy pigs from the same batch for each suspect pig) were not significantly different. When compared with the reference inspector on-farm, three producers classified abnormalities in 2495 pigs with high specificity (0.98), acceptable sensitivity (0.82) and kappa (0.57). CONCLUSIONS: With training, producer sensitivity in antemortem inspection may be high and more proficient than abattoir antemortem inspection. Use of producer inspections may have benefits for animal welfare and chain efficiency, but not food safety.

Assessment of producer conducted antemortem inspection of market pigs in Australia.

OBJECTIVE: A field trial was conducted to evaluate the effectiveness of on-farm antemortem inspection performed by Australian pig producers in terms of proficiency in detecting unhealthy animals (suspects), the impact on food safety and animal welfare outcomes. ANIMALS: A total of 64 trial groups of market pigs (n= 10,703) were dispatched from 20 herds to three abattoirs in three states. PROCEDURE: All producers and abattoir antemortem inspectors were trained in antemortem inspection prior to the trial using a standardised set of antemortem criteria. Study 1 consisted of three comparisons: on-farm versus abattoir antemortem inspections on the same pigs, transport effects in suspect and normal pigs, and meat rejection and carcase disposition in suspect and normal pigs. Study 2 compared Salmonella spp status of suspect and normal pigs. Studies 3 and 4 were conducted to assist the interpretation of the results of on-farm versus abattoir inspection. Antemortem inspection results of three producers and one abattoir antemortem inspector were compared with those of a reference inspector (Gold Standard). RESULTS: Of the pigs examined, 2.12% (95% CI: 1.84%, 2.42%) were classified as suspect at on-farm antemortem inspection compared to only 0.14% at abattoir antemortem inspection. Forty one percent of pigs classified as suspect on-farm had locomotor problems (arthritis, lameness and foot abscess). Compared to normal pigs, suspect pigs had a relative risk of 67.7 (P < 0.0001) of suffering transport injuries. Meat rejection amounted to the equivalent of 9.68% of suspect carcases, compared to the equivalent of 0.35% of normal carcases. The isolation of Salmonella species from caecal content from 20.9% of 67 suspect pigs and 18.8% of 133 control pigs (two matched healthy pigs from the same batch for each suspect pig) were not significantly different. When compared with the reference inspector on-farm, three producers classified abnormalities in 2,495 pigs with high specificity (0.98), acceptable sensitivity (0.82) and kappa (0.57). CONCLUSIONS: With training, producer sensitivity in antemortem inspection may be high and more proficient than abattoir antemortem inspection. Use of producer inspections may have benefits for animal welfare and chain efficiency, but not food safety.

The adequacy of sample type/weight and incubation period on detection of Salmonella spp. in slaughter cattle.

The purpose of this study was to examine the adequacy of different sample types (fecal and rumen content), rumen-content sample weight (1, 10, and 25 g), and incubation period on the detection of Salmonella spp. in grass-fed beef cattle at slaughter. The culture technique was the same for all samples and followed the Australian Standard (AS 1766.2.5-1991). Sample adequacy was defined as the ratio between the overall prevalence, as obtained from samples identified as positive by any sample type/weight, and the estimated prevalence, as obtained from samples identified as positive by a particular sample type/weight. Sample adequacy reflects the likelihood of a sample of a particular type and weight to contain the organism of interest and hence is related to the sensitivity of the diagnostic test. It was found that sample adequacy differed between sample types and weights: 37.5% for both a 10-g fecal sample and a 1-g rumen sample, 77.1% for a 10-g rumen sample, and 79.2% for a 25-g rumen sample. On this basis, it is strongly recommended that sample type and weight be considered in the design of studies that aim to quantify Salmonella contamination in cattle.

Risk-based evaluation of postmortem inspection procedures for pigs in Australia.

The results of traditional (incision) and risk-based (visual) postmortem inspection procedures were compared on groups of approximately 30,000 pigs. The performance characteristics used as a basis for comparison included the non-detection rates of grossly detectable abnormalities, the microbiological contamination rates of carcases and boned product, the association of reactive lymph nodes with carcase condemnation and the achievement of 'finished product standards' for 'wholesomeness'. It was estimated that 6 per cent of all cases of abscessation and 28 per cent of all cases of arthritis were undetected by the traditional method, and the comparable figures for the risk-based procedure were 19 per cent and 39 per cent. However, when the rates of contamination of undetected abnormalities with foodborne hazards and other carcase contamination parameters were taken into account, it was concluded that both inspection systems were likely to result in a very similar level of consumer protection. Any increase in potential exposure to foodborne hazards in the abnormalities undetected by risk-based inspection would be insignificant in comparison with the potential exposure to foodborne hazards resulting from contaminated 'normal' lymph nodes and carcase surfaces. There were no statistically significant differences between the two procedures in the contamination rates of pre-chill carcases or boned retail products with Salmonella and Yersinia species.

Risk assessment of organoleptic postmortem inspection procedures for pigs.

A systematic quantification of foodborne hazards in abnormal and normal tissues of pig carcases was undertaken to provide a risk-based assessment of the effectiveness of traditional organoleptic meat inspection. A total of 36,059 pigs, representing all major pig-producing areas and systems in Australia, were inspected on a seasonal basis at three abattoirs over 12 months. The prevalence of grossly detectable abnormalities of possible food-borne disease significance was recorded. A subset of the grossly detectable abnormalities, together with tissues classified by inspection as normal (controls) were submitted for the detection of a broad range of food-borne hazards. The potential exposure of consumers to hazards in fresh pork was characterised as the number of carcases per 10,000 containing hazards in selected tissues. The results indicated that the level of exposure of consumers to microbiological hazards in fresh pork is unlikely to be reduced significantly by the detection and removal of gross abnormalities in the tissues examined. On the basis of carcase throughput, the rate of contamination of normal lymph nodes was commonly 100 times higher, and no hazards were isolated from two types of grossly abnormal nodes. While further processing, cooking and handling may alter the exposure characterisation, the study nevertheless identifies the proportional contribution of abnormal and normal tissues to risks to consumers and clearly identifies the need for consideration of 'visual only' inspection in the re-evaluation of traditional inspection procedures.

Using slaughter inspections to evaluate sarcoptic mange infestation of finishing swine.

Sarcoptic mange is one of the common swine diseases worldwide. Although mange-free populations can be established with caesarean derived stock, by herd repopulation programmes or by eliminating mange with ivermectin, mange remains prevalent in many countries. Field and experimental studies indicate that hypersensitive mange is detrimental to performance of growing pigs. Typically, producers tolerate mange infestation in their herds and control measures are often haphazard. This tolerance to mange infestation is attributable to the covert nature of the losses (reduced growth rate and feed efficiency without mortality) and to the fact that clinical signs of hypersensitive mange (pruritus) are usually viewed as normal. Lack of tools to evaluate mange severity in pigs and to demonstrate its importance has hindered the efforts of veterinarians to control the disease. Traditionally, veterinarians have used slaughter inspections to assess respiratory diseases such as enzootic pneumonia and atrophic rhinitis. Much of the value of slaughter inspections is as a tool with which veterinarians can educate and motivate their clients to improve disease control measures. The potential for evaluating hypersensitive mange by inspecting slaughtered pigs for lesions of papular dermatitis was recognised some time ago, but quantitative evaluation of the reliability of this approach has been lacking. We have conducted several studies in Australia, the USA, Canada, Europe and Latin America to evaluate associations between Sarcoptes infestation and the severity of papular dermatitis at slaughter, using a simple ordinal scale for classifying carcasses. Our initial field and experimental data in Australia indicated the specificity of localised dermatitis to be in the order of 75-80%, but that the generalised dermatitis was highly specific (> 98%) for mange. Subsequent studies in the US Midwest yielded almost identical results, and indicated that the method may also have some utility for surveillance of mange-free herds. Results from other locations invariably have shown significant associations between dermatitis lesions and mange infestation. Relative to other methods such as skin scrapings and monitoring pruritus, this method is simple and relatively objective, and should be considered for routine inclusion in slaughter inspection protocols.

Seasonality of sarcoptic mange in pigs in South Australia.

Retrospective analysis of records, from July 1987 to June 1991, of 1140 inspections of slaughtered pigs from 154 swine herds in the Pig Health Monitoring Scheme of South Australia was conducted to evaluate patterns of variation in the prevalence and severity of skin lesions typical of sarcoptic mange. An increase in the severity of sarcoptic mange in winter and spring was observed. There was a substantial reduction in the severity of mange over the period of monitoring. These observations support the role of monitoring at slaughter in the management of endemic disease. They provide some basis for refining control programs for mange to improve cost-effectiveness.

Sarcoptic mite hypersensitivity and skin lesions in slaughtered pigs.

The association between papular skin lesions in slaughtered pigs and hypersensitivity to Sarcoptes scabiei var suis was examined in experimental and field studies, and by the retrospective analysis of monitoring records obtained at slaughter. A causal role for S scabiei was indicated by the production of lesions in experimentally infested pigs, by increased clinical mange in herds having high lesion scores, and by a reduction in the severity of lesions after the implementation of more intensive mange control measures. The specificity of the lesions as an indicator of sarcoptic mite hypersensitivity was estimated to be 0.92 from experimental data and 0.79 from field data. Generalised lesions were rare in mange free pigs, indicating that such lesions were highly specific (1.0 from experimental data, greater than 0.98 from field data) for mange. The assessment of the severity of the lesions in samples of slaughtered pigs appears to be a useful aid to assessing the severity of sarcoptic mange in pig herds.

Enzootic pneumonia of pigs in South Australia--factors relating to incidence of disease.

Environmental and management conditions of 15 herds with a high-prevalence (greater than 70%) of enzootic pneumonia of pigs at slaughter were compared with 16 herds with a low-prevalence (less than 30%) to determine factors commonly predisposing pigs to enzootic pneumonia in South Australia. Comparisons were made of herds having greater than 100 sows (fully intensive units) and also small herds having 20 to 70 sows (sideline units). Half the herds were visited in summer and half in winter to detect seasonal factors. In small herds, factors commonly found associated with a high prevalence of enzootic pneumonia were larger numbers of pigs per shed section (p less than 0.001), larger group sizes (p less than 0.01) and draughty farrowing and weaner accommodation (p less than 0.01). In large herds, factors associated with a high prevalence were higher pen stocking rate (p less than 0.05) and airspace stocking rate (p greater than 0.05), and a trend toward higher atmospheric ammonia levels in summer (p less than 0.1).

Effect of enzootic pneumonia of pigs on growth performance.

Growing pigs were naturally infected with a field strain of Mycoplasma hyopneumoniae to assess the effect of enzootic pneumonia on production. Both the initial ("breakdown") and endemic stages of infection were evaluated. The pigs were reared under environmental and management conditions commonly found on commercial piggeries in South Australia. Growth rate of pigs held in-contact with inoculated pigs was reduced by 12.7% (p less than 0.01) between 50 to 85 kg bodyweight. In the second trial inoculated gilts were used to naturally infect piglets during suckling. Growth rate of infected pigs was reduced by 15.9% (p less than 0.001) between 8 to 85 kg bodyweight, while feed conversion was depressed by 13.8% (p less than 0.05) between 10 to 25 kg bodyweight. At current feed and production costs this reduced performance added approximately $2.80 to the cost of every pig produced. These losses were recorded in groups of pigs in which enzootic pneumonia was present. At slaughter, 40% of lungs contained gross lesions of enzootic pneumonia which were free of significant secondary bacteria. The nature of the infection was established by gross and microscopic pathology and confirmed by the detection of specific complement fixing antibody in infected pigs and the demonstration of M. hyopneumoniae by direct immunofluorescent staining of lung sections.

The pattern of endemic parvovirus infection in four pig herds.

Serological surveys were conducted on the gilts and adult sows in 4 herds endemically infected with porcine parvovirus. The study assessed the influence of the type of management of breeders on the spread of virus infection and the influence of endemic parvovirus infection on reproductive parameters of the herd. The practice of holding gilts and sows in groups did not reliably promote infection or maintain a 100% level of active immunity amongst adult sows in 2 of 3 group husbandry herds. In the 4 herds, the prevalence of adult sows (greater than 12 months) with active immune haemagglutination inhibition titres (greater than or equal to 256) ranged between 44% and 100%, while between 0% and 100% of gilts (6 to 12 months of age) had active immune titres. Fully susceptible gilts older than 9 months of age held in groups, failed to become infected by 12 months of age on farms endemically infected with PPV. In 2 herds a continued low infection rate of gilts resulted in increasing the potential of breeding animals becoming susceptible to parvovirus infection as infected sows were replaced by noninfected gilts. In both herds, epidemics of parvovirus infection followed, which were characterised by an increase in reproductive failure. Parvovirus infection during the first 70 days of pregnancy reduced the average number of piglets born alive per litter by 1.6 piglets (p less than 0.05). This was due to the combined effect of more piglets being born dead per litter and an overall reduction in litter size.