I always had a special relationship with enzymes! After all, they were involved in the first trials I conducted as a student, and they were the theme of one of my papers. Enzymes have been a relative recent development, and new advances continue to progress their effectiveness. Yet, many misunderstandings remain regarding their use under practical conditions. It is thus a little surprising that I continue to get frequent questions on how best to use commercial enzyme products in pig formulas. So, let's set the record straight!
First, we're talking here about enzymes that attack the non-starch polysaccharide (NSP) fraction of cereals; those highly diverse and hard if not impossible for the animal to digest components of all cereals! Different cereals have different NSP. For example, wheat is quite rich in arabinoxylans (pentosans), while barley is rich in beta-glucans. Maize is relatively 'innocent', while rice can be considered virtually free of NSP for all practical purposes. Rye and oats have a blend of both classes of major NSP. The chemist purists will certainly object to this simplified view providing page-length lists of NSP in each and every cereal, but we're talking here about feeding pigs in the farm, so this simplified view will suffice for now.
When it comes to selecting a commercial enzyme it is always best to use a product specifically designed for the formula to be used. Thus, if the formula contains mainly maize, and only less than 10% other cereals, then it makes little economic sense to use any enzymes at all! But, if the formula is based on wheat, then a xylanase (pentosanase) is required for maximal efficacy. The opposite is true in barley-based diets where a beta-glucanase is required. Sometimes, the same enzyme (one of the two described above) is used regardless of the major cereal in the diet, which of course is not ideal, and in most cases a pure waste of money.
When a diet is based on a mix of wheat and barley, in about equal proportions, then ideally a blend of both types of enzymes (a cocktail) should be used. If wheat or barley make the majority contribution whereas the other cereals make up a small proportion of the total, then again only a cereal-specific enzyme is required. And, as far as I am aware of, there is no strong background to suggest pigs can benefit for the inclusion of an amylase (for the breakdown of starch), as they produce enough of this enzyme themselves.
A most frequent question is how much benefit should be expected by the use of enzymes. Scientific studies have always been difficult to interpret into practical recommendations because cereal quality tends to be quite variable. In brief, wheat low in pentosans will hardly benefit from the addition of any enzyme. Only marginal results should be expected with average quality wheat, but when quality is really low then one should expect good results from pentosanases. Barley is a bit more problematic and for this even average quality batches benefit from beta-glucanases.
As for a quantitative assessment of the benefit achieved from the use of enzymes under proper conditions, one might expect on average the release of about 50 kcal metabolizable energy per kg cereal. Claims on improved protein digestibility have some scientific merit but they remain largely without strong proof, and for this it is best to assume a conservative position and ignore this effect. It is hard some times to get even an energy response to enzymes let alone one in protein!
And, a closing remark. These 50 kcal ME better be cheaper coming from enzymes, because enzymes cost money, than by the addition of a bit of oil or fat! This is likely in most cases, but one should always keep cost-profit calculations updated.