Between 1996 and 2004, tissue samples from 3 630 slaughtered pigs were examined by gross examination, microscopy after the Ziehl-Neelsen (ZN) staining of homogenised tissues for the detection of acid-fast rods (AFR) and by culture for the presence of mycobacteria and Rhodococcus equi: 1 781 head lymph nodes (ln), 1 123 mesenteric ln, 54 pulmonary ln, 32 inguinal ln, 562 non-identified ln and 78 samples of tissues from parenchymatous organs ( liver, spleen and kidneys). Tuberculous/tuberculoid lesions were not detected in 249 (6.9%) animals slaughtered due to a positive response to avian tuberculin. Various gross lesions were detected in 3 381 (93.1%) animals as follows: adenopathy in 150 (4.1%), tuberculous lesions with caseation in 2 026 (55.8%) and tuberculous lesions with calcification in 1 205 (33.2%) of them. AFR were found in tissues from 2 047 (56.4%) animals: in 36 (14.5%) animals free from gross lesions, in 28 (18.7%) animals with adenopathy, in 801 (39.5%) animals with caseation and in 913 (75.8%) animals with calcified tuberculous lesions. Mycobacteria were isolated from the tissues of 289 (15.8%) of 1 852 animals without detected AFR and from the tissues of 1 290 (72.5%) of the 1 778 animals with detected AFR of various intensities. Of 1 579 mycobacterial isolates 1 493 (94.6%) were classified as M. avium complex ( MAC) members: 469 (29.7%) M. a. avium (IS901+, serotypes 1, 2, and 3) and 891 ( 56.4%) M. a. hominissuis (IS901-) isolates of serotypes 4 (n = 1), 8 ( n = 643), 9 ( n = 74) and non-typed ( n = 173). The other 52 (3.3%) isolates were members of other mycobacterial species: M. chelonae ( n = 35), M. smegmatis ( n = 4), M. xenopi ( n = 3), M. terrae ( n = 7), M. aurum ( n = 1), M. scrofulaceum ( n = 1), M. fortuitum ( n = 1) and biochemically non-identified mycobacteria ( n = 34). By examination of ZN stained homogenised tissues, AFR were detected significantly more frequently (P < 0.01) in samples from animals with caseated and/or calcified tuberculous lesions than in tissues from animals without tuberculous lesions. The detection rate of isolates from tissues with tuberculous lesions was likewise significantly higher (P < 0.01) than from tissues without tuberculous lesions.
