Deoxynivalenol (DON) is a mycotoxin produced by fungi which may contaminate cereal grains, including barley and wheat.  The contamination is especially problematic when wet, warm conditions prevail during the growing season.  The ingestion of grain that is severely contaminated by DON will cause overt symptoms such as vomiting (hence the common name “vomitoxin”).  Less dramatic, but more frequently observed symptoms, reduced feed intake and growth, will result when pigs consume feed with a lower concentration of the mycotoxin.  The Canadian Feed Inspection Agency suggests that 1 ppm mycotoxin in feed is a safe upper limit for swine.

There are several feed additives available which reportedly reduce the impact of the mycotoxin on the pig.  Modes of action vary, and include; binding the mycotoxin in the gut and preventing absorption, chemically transforming the toxin to decrease its toxicity, or enhancing immune system function. 

The overall objective of this experiment was to determine the effect of these feed additives on the performance of nursery pigs fed diets contaminated with DON.

We used 5 nurseries for this experiment, 24 pens per nursery and 4 pigs per pen.  Pigs were fed starter diets for 14 days before being offered the treatment diets (BW 9.02 ± 0.36 kg) for the next 14 days.  All starter diets contained in-feed antibiotics. 

Treatment diets were formulated to meet or exceed all requirements for pigs of this age.  A positive control diet contained no contaminated corn, while the negative control diet was formulated with contaminated corn but no feed additives.  Samples of corn which were pre-analyzed and shown to contain DON were used for 70% of the corn (35% in the final diet) in diets 2 to 12 to provide 2 ppm DON in the final diet. This concentration was chosen because a preliminary experiment indicated this amount would cause a measurable reduction in feed intake but would not be fatal.

Performance results are shown in Table 1.  Pigs on the positive control tended to be heavier than those on the negative control by day 22 (0.50 kg, P = 0.09).  Overall, pigs consuming diets contaminated with DON had reduced ADG and ADFI compared to those consuming the positive control diet free of DON (P < 0.001).  Weekly measurements of body weight and feed intake showed that the decline in feed intake preceded the decline in growth (data not shown).

Average daily gain and ADFI of pigs on the positive control was superior to those consuming the DON contaminated diet, regardless of the feed additive used.  None of the feed additives ameliorated the effects of DON on feed intake or gain.  Feed efficiency was unaffected by treatment (P > 0.05).

Based on a literature search and our preliminary experiment which indicated that 2 ppm would elicit a detectable decrease in feed intake but was non-fatal, we formulated the treatment diets to this level.  Analyses of the diets indicated a mean concentration in the DON containing diets of 1.99 ppm, however, the individual diet concentrations ranged from 1.57 to 2.61 ppm. 

The 1 tonne totes of contaminated corn were initially sampled from about 10 different locations within each tote to a depth of about 1 metre.  These samples, composited by tote, were sent to two different labs for analyses for DON and moulds.  The results were extremely variable, within and between the labs.  Results from lab “A” ranged from 2.4 to 5.5 ppm with a mean of 4.5 while the results from lab “B” were 2.2 to 9.6 ppm and a mean of 6.9.  We didn’t use the totes which displayed the most variability, however, the DON concentrations in our diets were still quite variable (Table 1). 

The above illustrates the difficulty of working with mycotoxins.  Obtaining representative samples for mycotoxin testing is very difficult, however it is imperative that a good sample is obtained or the results will be irrelevant.  It has been estimated that almost 90% of the error associated with mycotoxin testing can be attributed to the method used to obtain the original sample.  Because contamination within a field may be localized, a truck-load which has come directly from a field at harvest is likely to contain only discrete areas of contamination.  Moreover, mycotoxin contaminated grains are heavier, thus within a truckload or during storage, some stratification may occur. 

The “Grain Inspection, Packers and Stockyards Administration (GIPSA) of the USDA only recognizes samples which have been obtained using a probe.  Moreover, at least 4 samples should be taken from each lot, preferably 7 to 9, depending on the size and thickness of the trailer.  A 2000 to 2500 gram sample should be obtained.  This sample should be ground and then subsampled to obtain the approximately 100 gram sample required by the lab.  Producers are advised to contact the laboratory they will be using for the analyses to obtain specific sampling procedures and amounts required.

In summary, when nursery pigs were fed diets contaminated with approximately 2 ppm DON, feed intake declined by 10 % and growth by 7%.  None of the feed additives mitigated this response, however, actual concentrations of DON in the test diets varied. This variability is an illustration of the difficulties inherent in correct sampling and analysis for mycotoxins. 

Acknowledgements

Strategic funding was provided by Sask Pork, Alberta Pork, Manitoba Pork Council and Saskatchewan Agriculture and Food Development Fund.

Pork Insight was developed to address producer and industry needs for timely and accurate information related to pork production and is designed to help you find the information to help you fine -tune your operation. The Pork Insight database can be found online at www.prairieswine.com 

Table 1:  Analyzed concentrations of DON in treatment diets and effect on performance of nursery pigs (initial BW 9.02 kg).

^aDay 22 of the experiment, day 36 post-weaning.

^bUsed exclusively non-contaminated corn.

^cNegligible

^dFormulated to contain 2 ppm DON

^eDifferent from Trt 1, (positive control; P < 0.05).

^fDifferent from Trt 2, (negative control; P < 0.05).