This study was the first made of CLA in sheep herds in Brazil that examined data from serum samples along with management information from slaughterhouse sheep suppliers. The serological results from an earlier study in Minas Gerais  and this present study indicate maintenance of a high frequency of seropositive animals and properties; in the first study conducted in 2002, the animal prevalence was 75.8% and prevalence in herds was 95.9%. Our current study, conducted seven years later, showed a high frequency of CLA in herds placed in the same region, demonstrating that CLA is still overlooked by farmers, who have not been taking preventive measures. This lack of disease control constitutes one of the greatest sanitary barriers to the development of sheep husbandry. Among these same animals, 15.3% (data not shown) had caseous nodules in pre-scapular and submandibular lymph nodes, found during slaughter, and 43.7% were seropositive. This shows that most of the infected animals had no clinical signs of CLA. Caseous material from abscesses, the main source of infection, is usually spontaneously drained. Live infectious C. pseudotuberculosis has been isolated after five months from areas where abscesses were drained .
The estimate of the frequency of animals infected by C. pseudotuberculosis in Minas Gerais was within expected limits; the 43.7% prevalence rate in the animals, calculated from the ELISA data is within the estimated confidence interval for frequency, even though 11.7% of sheep owners claimed that they used CLA vaccine. Although sampling and blood collection were conducted properly, there were some problems with the transportation of sera, resulting in the loss of 18 serum samples. All of the sampled herds had at least one infected sheep, showing that CLA is widespread in Minas Gerais, which is the state with the second-largest sheep population in the southeastern region of Brazil .
All properties used extensive/semi-extensive management systems (Additional file 1), similar to what was found in 2002 ; this is common for farms dedicated to sheep meat production, in which the animals are raised exclusively on pasture during the day, with some protection at night; they graze most of the year. Consequently, the farmers tend to make few inspections of the animals.
In 2002, few farmers made individual identification of sheep (22.7%) . In our current study, only 70% of the farmers made individual identifications, even though it is required by the slaughterhouses. It is possible that the farms that we sampled in this study had a better technological level than the farms sampled in 2002 due to the sanitary requirements that have been adopted by slaughterhouses, such as slaughtering of young animals (less than one year of age), without clinical signs of CLA, in good health and with individual identification. The finding of 30% of flock owners who do not use individual identification of animals appears to be because they think that such control is unnecessary because the sheep will be slaughtered at a young age. However, it is impossible to control livestock or achieve good sanitary control without individual identification.
The high percentage of farmers without any veterinary assistance (Additional file 1) is one of the most serious obstacles impeding successful control of CLA. Only 41.2% of farms sampled in 2002 had veterinary assistance; in our current study the percentage of properties with such assistance was even lower (Additional file 1). Veterinary monitoring is an important determinant for the success of sheep husbandry and it is vital to educate farm workers and establish and maintain sanitary control programs. We can infer that the high seroprevalence of CLA is a consequence of management deficiencies found in 2002 and 2007, among them the low level of veterinary assistance, resulting in serious losses to sheep husbandry and industry.
Few facilities underwent routine disinfection (Additional file 1). Disinfection of facilities is very important because some pathogens, including C. pseudotuberculosis, can survive for long periods in the soil. Through experimental contamination of facilities, it was found that C. pseudotuberculosis can survive up to eight months at various temperatures . This bacterium has been isolated after five months in places where there has been contamination with pus . Consequently, environmental contamination due to leaking abscess is a real threat, and regular disinfection of facilities is an important measure to control CLA.
The frequent use of fences with barbed wire (Additional file 1), traditional in the state for cattle, is inappropriate for sheep because they tend to force passage between the wires of the fence, thereby suffering lesions in the skin, opening passage for the entry of many bacteria, including C. pseudotuberculosis. This is especially important because many of the breeds we sampled animals do not produce wool, including santa ines, morada nova and somalis. Other breeds used locally have only a short light coat of wool and hair, these being dorper and white dorper; there are also close-wool breeds, ile de france and lacaune .
Three commercial immunogens for C. pseudotuberculosis are available in Brazil; however, only 11.7% of farmers vaccinated their herds systematically against CLA. No farmer sampled in 2002  indicated that he vaccinated sheep against CLA; five years later, the frequency of vaccination remained very low; associated with other poor management practices, this has allowed continued spread of C. pseudotubercuslosis in sheep herds in Minas Gerais. Vaccination against CLA is one of the most important measures for disease control.
Very few farmers took notes of animals with clinical signs of CLA, opened, sanitized the abscesses and isolated the animals from the herd, which would be correct management strategy. Most of them let the abscess open by itself or opened it, but did not disinfect and properly drain the wound, and then returned the animal to the herd. Evidently, this procedure does not control, but rather accelerates the rate at which the agent is spread, because the caseous material is a source of contamination of facilities, including troughs, stalls and paddocks. Animals with recurrent abscesses (Additional file 2) should be culled from the herd, but 98.3% (59/60) did not do so. Although animals were individually identified, we found that identification was not used for sanitary control of the herd.
Few farmers were aware that the etiological agent of CLA can affect humans (Additional file 2); this ignorance further increases the risk of human infection. Moreover, only a small percentage of farmers burned and buried the material from abscesses or purulent material from manipulation of clinically-affected animals (Additional file 2), which can also be an important source of infection for other animals and humans. The veterinarian can be key to the orientation of the employees of these properties, teaching them to properly handle the abscesses and discard any material from such manipulations. All of the flock managers reported having information about losses due to CLA during slaughter (Additional file 2); so, they knew the disease is in their herds; but they did not relate this to losses in their farms and risk of human infection.