Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
2003 Poultry Science Association, Inc.
Nutrient Compostion of Feed-Grade
and Pet-Food-Grade Poultry
By-Product Meal
W. A. Dozier, III,*
,1
N. M. Dale,† and C. R. Dove‡
*Poultry Science Department, University of Georgia, Rural Development Center,
PO Box 1209, Tifton, Georgia 31793; †Poultry Science Department,
University of Georgia, Athens, Georgia 36849-5416; and ‡Department of Animal
and Dairy Science, University of Georgia, Tifton, Georgia 31793
Primary Audience: Nutritionists and Researchers
SUMMARY
Pet-food manufacturers have specified to renderers the need for poultry by-product meal (PBM)
to be manufactured without lower quality by-product fractions, such as feathers and heads, leading
to a higher protein product than conventional feed-grade PBM. One result is that nutritionists
are faced with greater nutrient variation among PBM sources. Thirty-six PBM samples (26 = feed
grade and 10 = pet-food grade) were collected from commercial feed mills during a 3-mo period
to assess nutrient composition and its variation. Pet-food-grade PBM had higher protein, less ash,
and lower calcium than feed-grade PBM. Amino acid analyses indicated the pet-food-grade PBM
had higher lysine and methionine, and the amino acids in pet-food-grade PBM exhibited higher
digestibility compared with those in the feed-grade samples. Nutrient variability was more
pronounced in the feed-grade PBM.
Key words: animal by-product, feedstuff, ingredient, poultry by-product meal
2003 J. Appl. Poult. Res. 12:526–530
DESCRIPTION OF PROBLEM
Poultry by-product meal (PBM) is a popular
protein source for poultry feeds. In the past,
PBM was of reasonably consistent nutrient com-
position, with nutrient levels similar to those
reported by the NRC [1]. However, in recent
years, the pet food industry, which is willing to
pay a premium for PBM of a defined composi-
tion, has placed increased demand for higher
protein PBM. Meal manufactured for this market
is usually termed pet-food grade, whereas the
resulting segregation of materials has led to in-
creased variation in feed-grade PBM. It is been
widely suggested that feed-grade PBM, which
1
To whom correspondence should be addressed: bdozier@uga.edu.
contains more variety of processing residues, is
more variable than in previous years. This has
decreased the economic value of the ingredient,
as nutritionists must include wider margins of
safety in their nutrient matrixes to reduce the
likelihood that manufactured feed will have a
less than intended nutrient composition.
This study examined nutrient composition
and its variability in currently available feed-
grade and pet-food-grade sources of PBM. Mea-
surements were extended beyond proximate
composition to include amino acid profiles, an
estimate of amino acid digestibility, and min-
eral analyses.
DOZIER ET AL.: FEED-GRADE VS. PET-FOOD-GRADE MEAL 527
TABLE 1. Proximate composition of feed-grade and pet-food-grade poultry by-product meal
Feed grade
A
Pet-food grade
B
Item, % Maximum Minimum Average SD Maximum Minimum Average SD
Crude protein 63.7 49.3 58.1 3.2 69.3 63.0 66.1 1.9
Ether extract 24.5 10.5 14.4 3.1 15.1 10.9 12.6 1.6
Moisture 5.6 1.7 4.2 1.3 7.2 2.2 4.1 1.4
Ash 20.6 12.8 17.1 2.4 18.5 10.7 15.1 1.6
A
Values are from 26 subsamples collected from feed mills in the southeastern United States.
B
Values are from 10 subsamples collected from feed mills in southeastern United States.
MATERIALS AND METHODS
Thirty-six samples (26 = feed grade and 10 =
pet-food grade) were obtained from commercial
feed mills located in Alabama, Delaware, Geor-
gia, North Carolina, Tennessee, and Virginia.
Subsamples were analyzed for proximate com-
position, mineral content, amino acid composi-
tion, and amino acid digestibility.
Subsamples were prepared for proximate
analysis by grinding each sub-sample in a stain-
less steel blade grinder until caking was ob-
served. Percentage moisture was determined by
drying 2 g of feed at 135°C for 2 h [2]. Ash
content was measured by igniting 2 g of feed at
600°C for 2 h in a preweighed porcelain crucible
and weighing the residue [2]. Crude protein com-
position was determined with 0.2 g of the sub-
sample to measure the recovery of nitrogen [3].
Ether extract percentage was measured gravi-
metrically by extracting 1 g of feed dried at
105°C for 1 h with boiling petroleum ether using
glass fiber thimbles [4]. The extracted fat was
collected in preweighed aluminum cups, dried
at 105°C, and weighed.
TABLE 2. Mineral composition of feed-grade and pet-food-grade poultry by-product meal
Feed grade
A
Pet-food grade
B
Item Maximum Minimum Average SD Maximum Minimum Average SD
Calcium, % 6.50 3.10 5.17 1.10 6.40 2.40 4.61 1.39
Phosphorus, % 3.20 1.50 2.50 0.44 3.40 1.70 2.59 0.59
Potassium, % 0.64 0.33 0.51 0.08 0.75 0.62 0.69 0.04
Magnesium, % 0.21 0.11 0.15 0.02 0.19 0.12 0.15 0.02
Manganese, ppm 19 6 19 4 14 5 9 3.30
Iron, ppm 4,626 406 1,830 1,480 960 129 352 302
Copper, ppm 50 12 22 8 430 4 57 131
Zinc, ppm 114 65 94 14 119 74 94 15
Sodium, ppm 6,660 2,961 4,608 996 5,897 3,681 4,635 719
A
Values are from 26 subsamples collected from feed mills in the southeastern United States.
B
Values are from 10 subsamples collected from feed mills in the southeastern United States.
Inductively coupled plasma emission spec-
troscopy (ICP) was performed to determine min-
eral composition. Sample preparation included
grinding subsamples in a stainless steel blade
grinder until caking was observed. After grind-
ing, 2 g of feed were digested with 10 mL of
concentrated nitric acid and 5-mL of concen-
trated perchloric acid on a hot plate. The diges-
tion was continued until the temperature in-
creased to the boiling point of perchloric acid,
approximately 200°C. After this digestion step,
10 mL of 25% vol/vol hydrochloric acid was
added, and the digests were diluted to 100 mL
with distilled-deionized water. Mineral concen-
trations were determined from the digested sam-
ples by using ICP [5].
Amino acid concentrations of the subsam-
ples were determined in triplicate, after 22 h of
hydrolysis (6 N HCl) by ion exchange chroma-
tography [6]. In addition to measuring total
amino acid concentrations of PBM subsamples,
amino acid digestibility was estimated by Novus
International using an IDEA analysis [7]. PBM
subsamples were prepared for IDEA analysis by
grinding to a fine powder and passing through
JAPR: Field Report528
TABLE 3. Amino acid composition of feed-grade and pet-food-grade poultry by-product meal
Feed grade
A
Pet-food grade
B
Concentration,
C
% Maximum Minimum Average SD Maximum Minimum Average SD
Lysine 3.32 1.77 2.75 0.41 3.65 2.48 2.92 0.38
Methionine 1.12 0.46 0.77 0.17 1.02 0.63 0.84 0.10
Cystine 0.35 0.07 0.20 0.08 0.28 0.14 0.18 0.05
Threonine 2.10 1.16 1.85 0.24 2.33 1.61 1.82 0.23
Arginine 4.78 2.51 3.63 0.65 4.34 2.94 3.49 0.48
Isoleucine 1.72 0.80 1.36 0.23 1.68 1.12 1.30 0.18
Tryptophan . . . . . . . . . . . . . . . . . . . . . . . .
Valine 2.34 1.14 1.86 0.30 2.33 1.45 1.74 0.28
Histidine 3.70 0.80 1.65 0.76 3.23 0.88 1.62 0.89
Leucine 3.80 1.88 3.16 0.47 3.87 2.61 3.00 0.42
Phenylalanine 2.07 0.90 1.57 0.27 1.84 1.13 1.52 0.23
Glycine 5.65 3.00 4.48 0.62 5.67 3.75 4.68 0.55
Glutamic acid 7.33 3.73 5.84 0.76 6.77 5.18 5.90 0.48
Digestibility,
D
%
Lysine 86.6 41.1 62.9 11.8 93.3 74.5 85.2 8.0
Methionine 88.5 58.7 72.1 7.6 95.5 79.5 87.0 5.6
Cystine 69.3 19.5 41.5 12.6 82.6 53.7 68.3 5.6
Threonine 85.4 57.6 70.0 7.0 92.2 76.9 84.7 4.3
Arginine 89.2 77.8 82.3 2.8 93.9 77.8 89.1 3.0
Isoleucine 88.8 60.8 73.6 7.1 96.9 60.8 88.1 4.8
Tryptophan 91.2 79.0 84.8 3.1 91.2 87.9 90.8 1.8
Valine 87.1 63.7 73.0 5.8 96.9 78.4 87.0 6.1
Histidine 82.3 57.4 67.2 6.2 92.7 73.0 82.1 6.5
Leucine 89.6 64.6 75.9 6.4 95.5 82.1 89.1 4.5
Phenylalanine 91.1 75.0 81.4 4.0 97.9 85.1 91.0 4.2
Glycine 87.1 73.6 79.0 3.3 92.8 82.1 87.0 3.5
Glutamic acid 86.0 64.1 73.1 5.3 95.0 78.0 85.9 5.6
A
Values are from 26 subsamples collected from feed mills in the southeastern United States.
B
Values are from 10 subsamples collected from feed mills in the southeastern United States.
C
Analysis was conducted in triplicate.
D
Amino acid digestibility was estimated using an immobilized digestive enzyme assay (Novus, IDEA system) in duplicate.
a 1-mm mesh screen. The ground samples were
solubilized in 50 mM KH
2
PO
4
, 0.1% NaN
3
, and
50 mM EDTA, pH 6.2, at a final concentration of
8 mg/mL. Duplicate samples (2.5 mL), including
any insoluble material, were transferred to an
enzyme kit [7]. Digestion was carried out on an
end-to-end rotator for2hina37°C incubator.
The rate of digestion was quantified by the reac-
tion of α-amino groups with o-phthalaldehyde
on initial and digested samples. An IDEA value
was calculated as the rate of digestion divided
by the crude protein of the sample. This value
was used to calculate the predicted amino acid
digestibilities estimated from equations supplied
in the kit, which are based upon correlations with
poultry true amino acid digestibility coefficients.
All samples were performed in duplicate. A stan-
dard sample of known digestibility was included
in the assay as the control.
RESULTS AND DISCUSSION
Proximate compositions of the feed-grade
and pet-food-grade PBM are presented in Table
1. Pet-food-grade PBM was higher in protein
content (66.1 vs. 58.1%) and had less variation
than the feed-grade sources. The average crude
protein content of feed-grade PBM is in close
agreement with that listed by the NRC [1], but
individual values ranged from 49.3 to 63.7%.
This extreme variation may be evidence of the
diverse residues included in the rendering pro-
cess. The ether extract content was also more
varied with the feed-grade PBM compared with
the pet-food-grade, which could indicate the in-
clusion of sludge in feed-grade PBM. Ash was
lower with pet-food grade than feed grade but
not as low as might have been expected. Average
moisture contents of the 2 sources were similar;
DOZIER ET AL.: FEED-GRADE VS. PET-FOOD-GRADE MEAL 529
FIGURE 1. Relation between moisture content and lysine digestibility in feed-grade poultry by-product meal.
however, the feed-grade source had several val-
ues that were considered to be extremely low
(<5%).
Mineral contents are noted in Table 2. In
agreement with ash composition, calcium con-
tent and its variation were lower for the pet-
food-grade PBM than feed-grade sources, but
the converse occurred with phosphorus. The
cause of the inverse relationship in calcium and
phosphorus of pet-food-grade PBM relative to
its ash content is elusive. Manganese and iron
contents were higher in the feed-grade PBM than
in the pet-food sources, and extensive variation
was observed for the feed-grade PBM. The large
disparity in iron contents between the 2 sources
may suggest impurities associated with the in-
clusion of sludge in rendering the feed-grade
source. Iron sulfate is frequently used as a floc-
culating agent in processing plant water purifi-
cation. The pet-food-grade source had higher
copper content than feed-grade PBM.
Amino acid composition indicated that pet-
food-grade PBM sources had higher total con-
CONCLUSIONS AND APPLICATIONS
1. Pet-food-grade PBM had a higher protein content, lower ether extract, and lower ash composition
than feed-grade product. The feed-grade sources were more variable in proximate composition.
centrations of lysine, methionine, glycine, and
glutamic acid than the feed-grade sources (Table
3). Lysine and methionine were lower in the
present study (regardless of source) than re-
ported by NRC [1] (lysine = 3.10% and methio-
nine = 0.99%). The digestibility of amino acids
was less for the feed-grade PBM, and variability
was more pronounced (Table 3). In comparison
with values reported by the NRC [1], average
amino acid digestibility coefficients were lower
for the feed-grade sources. Several feed-grade
PBM samples had a low moisture content (<5%),
which may indicate that some of the variability
associated with digestible lysine might have
been due to overprocessing during rendering.
The digestible lysine coefficient increased with
higher moisture in feed-grade PBM (Figure 1).
Lysine digestibility did not appear to be related
to moisture content in pet-food-grade PBM, be-
cause low moisture content (<5%) was not as
common with this source of PBM. Use of a
trend line revealed that the reduction in lysine
digestibility was not apparent with samples hav-
ing at least 5% moisture.
JAPR: Field Report530
2. Pet-food-grade PBM had lower calcium content and higher phosphorus composition compared
with the feed-grade source.
3. Lysine and methionine compositions were higher in pet-food-grade PBM. Pet-food-grade PBM
also had a higher amino acid digestibility. Lysine digestibility appeared to be influenced by the
moisture content of the feed-grade sources.
REFERENCES AND NOTES
1. National Research Council. 1994. Nutrient Requirements of
Poultry. 9th ed. National Academy Press, Washington, DC.
2. AOAC. 1996. Official Methods of Analysis of AOAC Inter-
national. 16th ed. AOAC International, Gathersburg, MD.
3. Combustion nitrogen analyzer, Elementar Americus Inc., Mt.
Laurel, NJ.
4. Tecator-Soxtec fat extractor. Tecator, Herndon, VA.
5. Maxfield, R., and B. Mindak. 1985. EPA Method Study
27, Method 200.7. EPA- 600/S4-85/05. Natl. Tech. Inf. Service,
Springfield, VA.
6. Beckman 6300 Amino Acid Analyzer, Beckman Instruments,
Palo Alto, CA.
7. PC IDEA Kit, Novus Int., St. Louis, MO.