© 2002 The American Society for Nutritional Sciences
J. Nutr. 132:1771S-1773S, June 2002
Supplement: Waltham International Symposium
A Method to Estimate Digestible Energy in Horse Feed
Annette Zeyner2 and
Ellen Kienzle*
Institute of Animal Nutrition, Nutritional Diseases and Dietetics, University of Leipzig, Leipzig, Germany and
* Institute of Physiology, Biochemistry and Animal Nutrition, Ludwig-Maximilians-University Munich, Munich, Germany
2To whom correspondence should be addressed. E-mail: Zeyner{at}aol.com.
KEY WORDS: symposium digestibility horses digestible energy
EXPANDED ABSTRACT
Predicting digestibility is a basic step for energy evaluation in horse feed. In horses, as in all species where microbial fermentation plays an important role in digestion, interactions between feed ingredients are even more important than those in other species. In horses suggestions for systems of energy evaluation usually refer to all feedstuffs and they are not limited to certain life stages or groups. It is quite common that the same mixed feed may be used for a leisure horse and a performance horse but in very different rations. Under these conditions it is not surprising that interactions between feedstuffs are widely neglected in predictive equations for digestible energy (DE). Several equations have been suggested that work quite well in some situations and lead to considerable errors in others (1,2). In our investigation we wanted to establish a predictive equation for DE on the basis of proximate nutrients with defined limitations for validity. The equation was derived mathematically from equations based on physiological considerations.
 |
MATERIALS AND METHODS
|
|---|
A total of 287 digestion trials was available for the development and the validation of the predictive equation. One hundred seventy trials (120 from Leipzig, 16 from Munich, and the rest from the literature) were used to develop a predictive equation (data pool A). Eighty-eight unpublished results and newer data from recent literature were used to validate the equation (data pool B). The relevant literature has been reviewed by Zeyner (1). In addition, results from other literature (314) were included. Rations with >8% crude fat in dry matter (DM) or with >35% of crude fiber in DM were not used, because interactions between nutrients preclude precise prediction of digestibility in such rations (1,15). Rations with significant percentages of straw treated with ammonia or NaOH were also excluded from the calculations. The first steps were to calculate descriptive statistics such as the distribution of the data (Table 1) and correlations between proximate nutrients (Table 2). Nutritive uniformity of each nutrient was tested by calculation of a linear regression between the nutrient and the digestible nutrient (16). Uniformity was confirmed for protein, fat and NFE within the range that is relevant for horse feed (Figs. 13). The resulting regression equations were included into the predictive equations. Fiber was much less uniform than the other nutrients (Fig. 4). In addition, feedstuffs with lower fiber content like cereals may contain a type of fiber that is less digestible than fiber from hay; however, this is not a general rule because fiber from straw is less digestible than that from hay. Therefore the mean digestibility of fiber was used to predict the digestible fiber from fiber content of the feed. However, fiber may decrease the digestibility of other nutrients. This was taken into account by a correction factor for fiber, calculated after the predictive equations, for all four digestible nutrients, which were multiplied by the gross energy content (Table 3), whereupon the equations were summed up.
View this table:
[in this window]
[in a new window]
|
TABLE 2 Bivariate correlations (Pearson, two-way) between nutrients in data pool A (n = 170) used to develop the predictive equation
|
|

View larger version (15K):
[in this window]
[in a new window]
|
FIGURE 1 Prediction of digestible crude protein (DCP) by crude protein (CP), data pool A, n = 170. ***P < 0.001 (16).
|
|

View larger version (19K):
[in this window]
[in a new window]
|
FIGURE 3 Prediction of digestible nitrogen-free extract (DNFE) by nitrogen-free extract (NFE), data pool A, n = 170. ***P < 0.001 (16).
|
|

View larger version (20K):
[in this window]
[in a new window]
|
FIGURE 4 Relationship between digestible crude fiber (DCF) and crude fiber (CF) (16), data pool A, n = 170. ***P < 0.001 (16). Mean digestibility of crude fiber = 47.7%.
|
|
 |
RESULTS AND DISCUSSION
|
|---|
The following equation was obtained: DE (MJ/kg DM) = -3.60 + 0.211 x CP + 0.421 x AEE + 0.015 x CF + 0.189 x NFE (nutrients in % DM). Figure 5 shows the relationship between experimentally determined and predicted DE in data from data pools A and B. There was no difference between the agreement in either data pool. There was no systematic deviation in high- or low-energy rations, provided that the limitations for rations (<35% CF, <8% AEE in DM) were considered. The equation can be applied to mixed feed with satisfactory results on the precondition that the total ration that contains the mixed feed is within the above-mentioned limitations. Rations with added fat and a total fat content of >5% as well as rations with a high content of highly fermentable fiber tended to be somewhat underestimated. To increase the predictive precision in such cases, it may probably be necessary to include more sophisticated methods of fiber analyses, such as detergent fiber, into the equation. However, predictive equations are mostly applied to mixed feed and detergent fiber analysis is not required for feed declaration in many countries. Evidence of the interaction of fat with other nutrients is contradictory and not yet fully understood (6,15). Therefore further research is needed for predictive equations including high-fat rations.

View larger version (21K):
[in this window]
[in a new window]
|
FIGURE 5 Relationship between experimentally determined and predicted DE: x, data pool A (used for development of equation); , data pool B (for validation). r = 0.626; ***P < 0.001; SE of regression equation = ±1.012; n = 236.
|
|
 |
FOOTNOTES
|
|---|
1 Presented as part of the Waltham International Symposium: Pet Nutrition Coming of Age held in Vancouver, Canada, August 67, 2001. This symposium and the publication of symposium proceedings were sponsored by the Waltham Centre for Pet Nutrition. Guest editors for this supplement were James G. Morris, University of California, Davis, Ivan H. Burger, consultant to Mars UK Limited, Carl L. Keen, University of California, Davis, and DAnn Finley, University of California, Davis. 
3 Abbreviations used: AEE, acid ether extract; CP, crude protein; CF, crude fiber; DAEE, digestible acid ether extract; DCP, digestible crude protein; DCF, digestible crude fiber; DE, digestible energy; DM, dry matter; DNFE, digestible nitrogen-free extract; NFE, nitrogen-free extract; SD, standard deviation. 
 |
LITERATURE CITED
|
|---|
1. Zeyner, A. (1995) Ermittlung des Gehaltes an verdaulicher Energie im Pferdefutter über die Verdaulichkeitsschätzung. Übers. Tierernährg. 23:55-104.
2. Fehrle, S. (1999) Untersuchungen zur Verdaulichkeit von Mischfutter beim Pferd in. Abhängigkeit von der Rauhfutteraufnahme. Thesis 1999 Ludwig-Maximilians-University Munich, Germany.
3. Ahlswede, L. (1977) Untersuchungen über Pferdealleinfutter in Form von Briketts. Dtsch. Tierärztl. Wschr. 84:132-135.
4. Barth, K. M., Williams, J. W. & Brown, D. G. (1977) Digestible energy requirements of working and non-working ponies. J. Anim. Sci. 44:585-589.
5. Hitter, E. & Iben, C. (2001) The application of estimating equations in horse feeding depending on the rations. Proc. Soc. Nutr. Physiol. 10:49.
6. Jansen, W. L., van der Kuilen, J., Geelen, S.N.J. & Beynen, A. C. (2000) The effect of replacing nonstructural carbohydrates with soybean oil on the digestibility of fibre in trotting horses. Equine Vet. J. 32:27-30.[Medline]
7. Kaden, C. (2000) Einfluß der Verabreichungsform von ß-Carotin und des Fettgehaltes der Ration auf den Serumresponse von ß-Carotin und
-Tocopherol. Thesis 2000 Ludwig-Maximilians-University Munich, Germany.
8. Kossila, V. & Ljung, G. (1977) Value of whole oat plant pellets in horse feeding. Ann. Agric. Fenniae 15:316-321.
9. Lensing, A. (1998) Eine Pilotstudie zum Einfluß der Fütterung auf Knochenmarker beim Pferd. Thesis 1998 Ludwig-Maximilians-University Munich, Germany.
10. Ott, E. A., Feaster, J. P. & Lieb, S. (1979) Acceptability and digestibility of dried citrus pulp by horses. J. Anim. Sci. 49:983-987.[Abstract/Free Full Text]
11. Parkins, J. J., Snow, D. H. & Adams, S. (1982) The apparent digestibility of complete diet cubes given to thoroughbred horses and the use of chromic oxide as an inert faecal marker. Br. Vet. J. 138:350-355.[Medline]
12. Pulse, R. E., Baker, J. P. & Potter, G. D. (1973) Effects of cecal fistulation upon nutrient digestion and indicator retention in horses. J. Anim. Sci. 37:488-492.
13. Schramm, A. (1988) Untersuchungen zum Einfluß der Rationsgestaltung auf die Verdaulichkeit ausgewählter. Rohnährstoffe und den Energiegehalt in Rationen zur Pferdefütterung Diplomarbeit 1988 University of Leipzig Leipzig, Germany.
14. Wolter, R. & Valette, J. P. (1985) Etude experimentale de linfluence de regimes hyperlipidiques sur les aptitudes sportives dequides en effort dendurance. J. Recherce Cheval. 11:122-136.
15. Zeyner, A. (2002) Ernährungsphysiologische Wirkungen eines Austauschs von stärkereichen Komponenten durch Sojaöl in der Reitpferdeernährung 2002 Habilitationsschrift, Georg-August-University Göttingen, Germany.
16. Lucas, H. L., Smart, W.W.G. & Cippoloni, M. A. (1961) Relation between digestibility and composition of feeds and foods, Report no. S-45 1961 North Carolina State University Raleigh.