In this study, we compared cross-bred dairy cows in the Susa Valley (Piedmont, northern Italy), reared either near a high-temperature steel production plant (Farms A and B) or in an industry-free area (control). Exposed cows (n = 36) were selected based on mean bulk milk toxic equivalent values of polychlorodibenzodioxins (PCDDs) and dioxin-like (DL) polychlorobiphenyls (PCBs) and polychlorodibenzofurans (PCDFs) equal to 18.56 pg/g fat and 8.56 pg/g of fat in dairy cows from Farms A and B, respectively, exceeding both those permitted by the legislation in force (6 pg/g fat PCDDs and DL-PCDFs/PCBs), and those measured in dairy cows (n = 19) of the farm used as control (1.75 pg/g of fat PCDDs and DL-PCDFs/PCBs). Two types of peripheral blood cell cultures were performed: without (normal cultures for the chromosome abnormality (CA)-test: gaps, chromatid breaks, chromosome breaks and fragments) and with addition of bromodeoxyuridine [for the sister chromatid exchange (SCE)-test]. Both tests revealed a significant (P ≤ 0.05) higher chromosome fragility in the exposed cattle compared to controls: CA/cell mean values (without gaps) were 0.65 ± 0.91, 0.51 ± 0.81 and 0.13 ± 0.39 in Farms A, B and controls, respectively, while SCE/cell mean values were 7.00 ± 2.88, 6.39 ± 2.80 and 5.29 ± 2.51. Although the role of other pollutants (e.g. heavy metals) in the genesis of the recorded chromosome alterations cannot be ruled out, our results confirm the findings of previous research into dioxin-exposed sheep.