366
Evaluation of NE prediction equations for corn distillers dried grains with solubles (DDGS) for growing-finishing pigs

Monday, March 16, 2015: 2:45 PM
302-303 (Community Choice Credit Union Convention Center)
F. Wu , Department of Animal Science, University of Minnesota, St Paul, MN
L. J. Johnston , West Central Research and Outreach Center, University of Minnesota, Morris, MN
P. E. Urriola , Department of Animal Science, University of Minnesota, St. Paul, MN
G. C. Shurson , Department of Animal Science, University of Minnesota, St. Paul, MN
Abstract Text:

Empirical equations based on analyzed chemical composition have been developed to estimate NE content of complete diets and feed ingredients, but have not been validated for use in DDGS. The objective of this study was to determine the NE content of 4 DDGS sources fed to growing-finishing pigs, and to evaluate the precision and accuracy of 5 published NE equations and NE estimates from Illuminate®. A total of 432 pigs (9 pigs/pen; 12 replications/diet) were fed 4 corn-soybean meal diets containing 40% DDGS from 4 different sources. Pig growth responses were determined every 2 wk, and least square means of BW and G:F were used as factors to determine diet NE using the NRC (2012) model. Pig growth potential was defined by BW, ADFI, and protein deposition rate of 12 pens of pigs (n = 108) fed corn-soybean meal diets (diet NE was calculated using NRC recommended values). For each time period, diet NE was estimated by adjusting diet NE inputs until the model-predicted G:F matched the observed G:F. Next, NE of DDGS was calculated by subtracting NE content (NRC, 2012) of corn and soybean meal from diet NE, and adjusting for the percentage of DDGS in the diet. The mean NE value for each DDGS source was defined as the average among 6 periods, weighted for feed disappearance in each period. Error and bias of predicting NE compared with model estimates for the 4 DDGS were calculated for each equation. Illuminate® estimates provided the least bias and moderate error, while estimated NE from EvaPig® using equation 1 had the least error and moderate bias. In conclusion, these data suggest that current NE prediction equations need to be revised for better NE prediction among sources of DDGS.

 

 

DDGS

 

 

Item

A

B

C

D

Error

Bias

Overall G:F

0.362

0.353

0.385

0.385

 

 

NE, kcal/kg (as-fed)

  NRC (2012) model estimates

2,377

1,924

2,612

2,513

 

 

  Illuminate® estimates

2,083

2,255

2,469

2,743

259

31

  EvaPig® 1

 

 

 

 

 

 

    Equation 1

2,248

2,195

2,337

2,368

216

-69

    Equation 2

2,747

2,756

2,853

2,912

511

460

    Equation 3

2,193

2,203

2,309

2,365

238

-89

  NRC (2012)

 

 

 

 

 

 

    Equation 1-7

2,149

2,177

2,242

2,281

277

-144

    Equation 1-8

2,194

2,204

2,309

2,366

237

-88

1Crude fiber, NDF, and ADF are factors for equations 1 to 3, respectively.

Keywords: DDGS, growing-finishing pigs, NE, prediction equations