The consequences of a change from a traditional meat inspection procedure, including manual handling, palpation and incision, to an entirely visual postmortem meat inspection procedure in Danish slaughter pigs were assessed by a comparative study of the two methods in 183,383 slaughter pigs. Out of 58 lesion codes (selected with a prevalence > or = 5.5 x 10(-5)), 26 (45 per cent) were assessed either as merely aesthetic or as the healed stage of an earlier lesion and nine (15 per cent) as active, but local processes, occurring only in non-edible tissue. Five lesion codes (9 per cent) were assessed as active, non-abscessal processes occurring in edible tissue, caused by swine-specific pathogens and 10 (17 per cent) were abscessal or pyaemic lesions occurring in edible tissue. Seven lesion codes (12 per cent) may be associated with consumer health hazards (two frequently and five rarely), and one with occupational health hazards. It was estimated that per 1000 carcases, an additional 2.5 with abscessal or pyaemic lesions (in edible tissue) containing Staphylococcus aureus, 4 x 10(-4) containing ochratoxin, 0.2 with arthritis due to Erysipelothrix rhusiopathiae, 0.1 with caseous lymphadenitis, 0.7 faecally contaminated with Salmonella species, and 3.4 faecally contaminated with Yersinia enterocolitica would remain undetected as a result of changing from the traditional to the visual inspection procedure. Two valuable reasons for implementing a visual control system are the potential for decreased cross-contamination (no handling, cutting and incision) and reduced inspection costs. The resources released as a result may be reallocated to hygiene and surveillance programmes.

Summary The prevalence of fibrinous pericarditis detected at slaughter in Danish slaughter pigs is approximately 0.02%. The microbiology and pathology of this disorder was studied through 46 field cases collected at slaughter from May 1994 to August 1995. Mycoplasmas (Mycoplasma hyopneumoniae, M. hyosynoviae and M. hyorhinis) were isolated from the pericardium in 38 cases and from the joints in six. M. Hyopneumoniae dominated with 33 isolates from the pericardium. Actinobacillus pleuropneumoniae was isolated from the pericardium in three cases (as mono‐infection in one and together with M. hyopneumoniae in two cases) and Actinomyces pyognes was isolated from the pericardium twice (as mono‐infection in one case and together with M. hyopneumoniae in another). A pyogenes was also present in the lung, liver and kidney in one of these cases. Streptococcus suis was isolated together with mycoplasmas from the pericardium in three Cases. The pericardium was sterile in three cases and contained a low grade mixed flora in two; the latter is believed to be a result of the slaughter procedure rather than an infection. No chlamydiae were found in connection with me microbiological examination. Parvovirus was found in one case in the pericardium and in two cases in me spleen. Forty‐five cases were subacute to chronic. The gross pathological examination showed mat bronchopneumonia, chronic pleuritis and synovitis or a low grade arthritis occurred in most cases. Only the arthritic lesions are believed to be pathogenetically concurrent with the pericarditis. Slight evidence of stasis of the liver and/or periangiolar edema was present in six cases. The histopathological examination confirmed the above mentioned findings. Bacterial colonies were found within the fibrinous layer on the pericardium in all cases from which either A pleuropneumoniae, A pyognes or S. suis was isolated. It is concluded that in this study mycoplasmas, particularly M. Hyopneumoniae, are the more likely cause of fibrinous pericarditis in slaughter pigs.

Summary The pathology and microbiology of non‐purulent arthritis in Danish slaughter pigs were studied in order to draw up an updated picture of the disease. Samples of one or more joints, their regional lymph nodes and the heart were collected from 101 arthritic pigs, totalling 137 joints, 86 lymph nodes and 96 hearts (group I). For another selected group including 25 pigs (group II), the preliminary diagnosis of mycoplasmic arthritis, based on the appearance of the corresponding regional lymph node, was assessed versus the appearance of the joint fluid. A total of 44 samples were examined. More than 70 % of all joint samples in both groups were sterile. Erysipelothrix rhusiopathiae was isolated from 7 % of the joints and Mycoplasma hyosynoviae from 9 %. A contamination flora was isolated in 17 % of the samples. Very few of the lymph nodes (5 %) were infected, and none of the hearts were. In group‐II M. hyosynoviae was isolated from 8 % of the joints. From the remainder the isolates were contamination flora or low numbers of non‐specific bacteria. The gross pathological changes of the lymph nodes and joints of the E. rhusiopathiae cases differed notably from those of the M. hyosynoviae cases. The pathological features of the lymph nodes, in particular, were type specific. This suggests that these features may be used to differentiate between the two types of arthritis. Considering the high number of sterile samples, it is necessary to test this hypothesis in greater detail. A programme is outlined to elucidate the zoonotic and hygienic consequences of a proposed alternative handling and removal procedures for the two types of arthritis at slaughter.