Salmonella infections on hog farms are much more dynamic than ever was thought, according to University of Minnesota's Thomas Blaha, DVM.

To gain control of these infections, producers will have to do two things: 1. Do a salmonella-targeted (very thorough) job of cleaning and disinfecting between groups of pigs, and

2. Adopt daily working conditions to prevent the "exchange" of environmental and herd salmonella strains.

Blaha is the Allen D. Leman Chair and coordinator of a long-term grant project, "Salmonella Abatement in the Pork Chain" for the National Pork Producers Council.

Salmonella dynamics are characterized by changing prevalence and serotype patterns. Researchers have identified up to 18 different serotypes of salmonella on a single hog farm, explains Blaha.

Fecal-oral spread of salmonella is well known as the most common means of on-farm spreading. But what is not so well known is the complex interaction between salmonella-infected animals and their contaminated environment and the role it plays in the pattern of infection in a herd. This dynamic interaction must be identified and understood before control measures can be put in place to reduce the amount of salmonella carried into the food chain by market hogs, stresses Blaha.

In Europe, Blaha's research has shown how salmonella-free starter poultry flocks developed the same level of salmonella prevalence after placement as conventional health flocks without salmonella elimination. Environmental contamination was identified as the source of infection. Salmonella control was only achieved by permanently reducing the organism from the environment.

"Salmonella is not like pseudorabies, for example, where we can eradicate it and be done with it. Something has to be added to a quality assurance system which has salmonella reduction as one of its primary tenets," points out Blaha.

Researchers at the University of Minnesota have confirmed the role of the salmonella-contaminated environment in infecting hog herds. About 10% of environmental samples taken on four Minnesota hog farms were salmonella-positive.

Work at USDA's Agricultural Research Service (ARS) and Iowa State University showed that salmonella elimination can be achieved by segregated early weaning with pigs reared in isolation. Later on, two groups were created, one stayed in isolation, the other was fed out in a traditional facility. The animals in isolation remained salmonella-free, while those raised conventionally became positive for salmonella.

In the Minnesota project, which begun in 1996, scientists followed one complete production cycle of two herds. One herd had high prevalence (HP) for salmonella and one had low prevalence (LP) for the disease in their slaughter hogs.

Objectives were: 1. To identify farm "salmonella-specific" critical areas and practices;

2. Judge the use of the Danish ELISA test for detection of infected herds; and

3. Use salmonella subtyping methods to identify factors linked to salmonella infections in swine.

In the first objective, 400 pigs were repeatedly sampled for fecal salmonella shedding. The rectal swabs of some 15 sows and their offspring from HP and LP farms and 809 environmental samples were tested for salmonella strains.

For results of the first objective, some 69 of 205 pigs from the HP farm tested positive for salmonella at least once during the entire production cycle in the rectal swab study (Table 1). Thirteen pigs were positive at two sampling periods and two pigs on three. Just one sample from the LP farm was positive.

In Tables 2 and 3, the impact of environment is detailed. Samples were categorized as:

EI - direct contact with study group animals (pen surface, feeders, waterers, etc.);

EII - indirect contact with the study animals but nearby (water, feed, ventilation and heating equipment, walkways); and

EIII - indirect contact with the study animals, outside hog raising areas (organic matter outside buildings, walkways, feed trucks).

Specifically, the results of cleaning procedures on each farm were checked by taking post-cleaning environmental samples, shown in Table 3.

The latter study revealed two critical areas for improvement.

First, clean and disinfect barns before bringing in pigs. On the HP farm, 28% of the post-wash samples were salmonella positive; at the LP farm, only 4% of these samples were positive.

And, Blaha suggests, daily working procedures are more important to influencing the "exchange" of environmental and herd salmonella strains than the working system of the farm.

Key examples include maintaining barn-specific boots and immediately cleaning up outside feed spills.

In the second objective, 1,000 blood samples collected during the study and meat pieces collected at slaughter were checked for salmonella antibodies using the mixed ELISA test from BI/NOBL Laboratories.

The majority of those blood and meat juice samples proved negative for salmonella. Only five samples from the HP farm and three samples from the LP farm were positive.

These results confirm that serological testing can only be used as a diagnostic tool on a herd basis. It is not an individual indicator of salmonella herd status, explains Blaha.

The third objective went beyond testing the two Minnesota herds. It involved testing 25 herds in the state. Detected were 384 salmonella strains (28 serotypes) that were tested for resistance patterns by the ARS Laboratory in Athens, GA. Several different clones of the same strain were found in both the animals and the environment of the study farms, says Blaha.

Also, 92% of the Salmonella typhimurium strains found in pigs and environment of four study farms are common for pigs and humans.

That brings out the importance of understanding on-farm epidemiology and zoonotics (communicable diseases from animals to people) of the salmonella strains. DNA fingerprinting will be useful in this regard, says Blaha.

The majority of test isolates showed resistance to one or more antibiotics. But resistance to four antibiotics was rare, Blaha reports, noting the need for more research in this area.

"The fact that there is no significant difference in the frequency of resistance between strains from lymph nodes and strains from the environment, shows that the phenomenon of antimicrobial resistance in salmonella species is much more complex than often discussed as direct consequence of the use of antimicrobials in animals, which urgently needs further research."

Blaha stresses it is very important to understand the pork industry will never be rid of salmonella. Permanent reduction controls must be an integral part of an on-farm, pork quality assurance program.

Researcher: Thomas Blaha, DVM, University of Minnesota. Phone Blaha at (612) 625-8290 or e-mail blaha002@