The nutritive value of industrial by-products for poultry

The metabolizable energy (ME) and partly also the digestibility of several industrial by-products, mainly originating from grain, were determined in trials with adult cockerels. Five experiments were performed using a conventional total collection procedure, and test ingredients were substituted for 25-40 % of the basal diet. The nitrogen-corrected apparent (AME n and true (TME n ) metabolizable energy values of distillers’ dried grains (DDG) and distillers’ dried grains with solubles (DDGS) derived from barley, wheat and rye varied from 7.85 to 11.8 and from 7.98 to 12.0 MJ/ kg DM, respectively. The highest ME n values were observed in the dehulled barley-derived products. The digestibility of crude protein (CP) was 62-83 and that of carbohydrates 25-50 %, the latter being lowest for feeds originating from whole barley grain. The results indicated that distillers’ feeds derived from dehulled grain, especially barley, are satisfactory energy sources for poultry diets. Dried barley protein (BP 1, CP 38 %), oats protein (CP 44 %) and distillers’ solids (CP 58 %) from the integrated ethanol starch process, as well as dehulled barley and oats proved to be excellent energy feeds containing AMEn 14.9, 18.6, 15.0, 14.2 and 14.6 MJ/kg DM, respectively. The TMEn values for BPI and dehulled grains were 15.3, 14.4 and 14.8 MJ/kg DM. Supplementation with enzymes did not significantly improve the ME value of dehulled barley. The AMEn values for barley protein (BP2, CP 27 %), barley bran and barley molasses obtained from the starch process, and wheat feed meal from the milling industry, were 13.3, 8.52, 13.5 and 10.6 MJ/kg DM, respectively. The lowest ME values were obtained for samples of wheat bran, dried potato pulp and malt sprouts: AMEn 6.09, 6.23 and 6.01; TMEn 6.15, 6.07 and 6.01 MJ/kg DM, respectively.


Introduction
Numerous by-products are produced in the process- es of the food and alcohol industries.Many of these are potential sources of energy and protein also for poultry diets, and new processing techniques such as the integrated ethanol-starch process (Lehmus-  saari and Ham 1987) are being developed enabling more specific fractionation of raw materials.This report describes trials to determine the nutritive value of products from the Finnish food and alco- hol industries using grain as a principal raw mate- rial including their ME content and digestibility.

Animals and housing
The birds included in this study were 30-52 weeks old White Leghorn males.They were kept in indi- vidual wire cages under housing conditions describ- ed earlier by Kiiskinen and Huida (1984).Before the start of the experiments the cocks were weighed and devided into groups of six birds in such a way that the average live weights varied as little as poss- ible from one group to another.

Test ingredients
The distillery products assayed in Experiments 1 and 2 (Table 1) were received from the Finnish alcohol industry (ALKO Ltd).Distillers' dried grains (DDG) and distillers' dried grains with solubles (DDGS) were produced in the conventio- nal distillery process utilizing dehulled barley (BDDG, BDDGS), hulled barley (BDDGH, BDDGSH), wheat (WDDGS) or rye (RDDGS) as raw material.Two components from the alcoholstarch process described by Näsi (1988), barley protein (BP 1) and barley fibre, were also included in Experiment 1 (Table 2).Later an other sample of BPI and samples of distillers' solids (a fraction obtained from the distillation of B-starch) and oats protein from the same process were examined (Expt 5).The barley starch process developed by the Raisio Group also produces a barley protein product (BP 2).A sample of BP 2 and samples of barley bran and barley molasses from the same process were studied in Experiment 3 (Table 2).In addition to the above test materials, several pro- ducts from the milling industry (wheat feed meal, wheat bran, dehulled barley and oats, malt sprouts), as well as dried potato pulp from the potato starch industry (Hämeen Peruna Ltd) were included in the present study (Expts 1 and 4).

Experimental procedures
Five separate trials were conducted to determine the ME values of the dietary ingredients.A total excreta collection procedure was applied for the difference trials.The test ingredients generally re- placed 25-40 % (molasses 10 %) of the basal diet, which normally consisted of 5 % fish meal, 91 % barley, 1 % rapeseed oil, 1 % dicalcium phosphate, 1 % calcium carbonate, 0,3 % sodium chloride and 0,7 % vitamin-trace mineral premixes.The diets were pressed into 4 mm pellets with a laboratory pelleting machine (Amandus Kahl Laborpresse L 175) and the daily ration per bird varied from 80 to 110 gin the experiments.The adaptation and col- lection periods lasted 4-5 days each.Samples from the basal diets were taken before and after pelleting, and from the test diets after pelleting and two days' drying in room temperature.The excreta were collected from the plates once per day and placed into plastic cans to be kept in a freezer (-20°C) to await analysis.In order to calculate the true ME values (TME, TME n , Expts 1 and 4) excreta of six birds were totally collected during the last 48 hours of the 72 hours' fasting period in order to determine the endogenous energy loss (EEL, EEL n ).

Chemical analyses and calculation
A proximate analysis was performed on each test ingredient, basal diet, test diet and dried excreta samples, according to the standard methods used at the Institute of Animal Production of the Agricultural Research Centre.When digestibilities were determined (Expts 1, 2 and 4) the samples were hydrolyzed with 3N HCI before ether extraction.
The thawed excreta samples were weighed mixed and homogenized individually and subjected to an analysis of moisture, total nitrogen (Kjeldahl pro- cedure), urinary nitrogen (Terpstra and Hart  1974) and 3N HCI ether extract (Expts 1, 2 and 4) before drying in an air oven at 60°C for 16 hours.
The gross energy (GE) of the feeds and dried excreta samples was measured with an IKA adiaba- tic bomb calorimeter.The apparent ME values (AME) of the diets were calculated by subtracting the excreta energy from the GE intake and then dividing by total dry matter intake.The values for the feedstuffs were calculated from the differences in the values between the basal diet and the test diet by extrapolation to 100 % feedstuff.The TME values of test diets were calculated by subtracting the EEL of the collection period from the total excreta energy of that period.All ME values were corrected for a nitrogen retention of zero using the factor 34.4 kJ per g of nitrogen retained or lost (Hill and Andersson 1958).The amount of urine organic matter was calculated by multiplying the amount of uric acid N by the factor 4 (Hartfield 1973).The ME and digestibility data from Experiments 1, 2 and 5 were analyzed by the one-way analysis of variance, and the mean values of the test ingredients were compared using Scheffe's multiple range test.

Results and discussion
The crude protein (CP) content of the distillery feeds derived from barley varied from 29.3 to 37.5 % and the crude fibre (CF) content from 8.0 to 16.1 % (Table 1).The highest CP values and the lowest CF values were for products obtained from dehull- ed barley.The AME n value of the feeds from dehull- ed barley was 11.6-11.8and that of whole barley grain was 7.85-8.36MJ/kg DM (P0.05,Table 3).
The starved birds produced excreta 2.7 (Expt 1) and 3.6 g DM/bird/24 h (Expt 4).Endogenous energy losses (EEL and EEL n ) were 33 and 12 (Expt 1) and 42 and 9 kJ/bird/24h (Expt 4).The TME n contents for the products from dehulled barley were 11.8-12.0and for those from hulled barley 8.08-8.59MJ/kg DM. (Table 3).The low energy values of the distillery products from whole barley grain are mainly due to their high fibre content and the result- ing low digestibility of carbohydrates, 25-27 % in the present study (Table 5).The results obtained with BDDGS and BDDGSH are relatively well in agreement with those reported by Askbrant and  Thomke (1986).The corresponding digestibility values for the products derived from dehulled bar- ley were significantly (P<0.05)higher: 38 (BDDG) and 50 % (BDDGS).The difference (P<0.05) between BDDG and BDDGS can be considered logical because DDGS contains more available car- bohydrates than DDG and the digestibility trials with pigs have also given similar results (Näsi 1985).The CP digestibilities of BDDGH and BDDGSH were lower than those of BDDG and BDDGS and the values of DDG were higher than those of DDGS.This last result is in agreement with the results obtained with pigs (Näsi 1985).In the European table for poultry feedstuffs (WPSA 1989) the AME n value for BDDGSH is 11.05 MJ/kg DM, which is remarkably higher than that obtained in the present study.The CF value for

52_
" See the explanations of the abbreviations below Table 1.
2) AME = apparent metabolizable energy, AME n = nitrogen corrected AME TME = true metabolizable energy, TME n = nitrogen corrected TME 3 -d g} Means in the same column with a different superscript letter are significantly (P<0.05)different.
BDDGSH in the above-mentioned table is, how- ever, lower than that of the examined sample.
In the WDDGS samples, CP varied from 31.8 to 41.5 and CF from 7.3 to 9.4 % in dry matter (Table 1).The corresponding values for the RDDGS sample were 33.9 and 8.2 %.The AME n value for the WDDGS samples which contained around 41 % CP was 10.5-10.7 MJ/kg DM (Table 3).The samples which contained less protein also had lower ME values (P0.05).The AME n value of RDDGS was 10.2 MJ/kg DM.According to these results, distillery feeds from dehulled grains seem to be satisfactory energy sources for poultry.The ME value of products derived from barley was nearly as high as that of their raw material.Näsi  (1990) used the same batches of BDDGS and WDDGS at a rate of up to 20 % in layers' diets wit- hout any significant effect on egg production.Dis- tillery feeds supplemented with methione replaced barley and soybean meal in the diets.Barley protein (BP 1) produced in the ethanol- starch process contains more protein (38 %) than that (BP 2) obtained in starch production (27 %, Table 2).Both products are relatively rich in meta- bolizable energy, BP 1 containing AME n 14.9 and BP 2 13.3 MJ/kg DM (Table 4).The values for the two BP 1 samples were very similar.The ME of barley proteins exceeds the table values of barley and in the case of BP 1 ME is similar to the value of wheat (Salo et al. 1990).The high ME value of BP 1 seems to be a result of its excellent digestibility which was 86 % for crude protein, 84 % for crude fat and 80 % for carbohydrates (Table 5).These results are in agreement with the results reported on pigs (Näsi 1989).The relatively high digestibility of carbohydrates is due to the low fibre content of Table 4.Total feed intake ofthe birds (g DM) and ME-values (MJ/kg DM) of the products from the ethanol-starch, starch and milling processes (SD = standard deviation).  1See the explanations of the abbreviations below Table 3.
21 an enzyme mixture (ALKO Ltd) A -B see the foot note below Table 3.
BP 1.According to NAsi (1988), BP 1 contains around 40 % of available carbohydrates (starch, sugars).It is relatively poor in methionine 1.8 and lysine 3.2 g/16 N (NAsi 1989).However, supplemented with the above-mentioned amino acids, BP 1 can be used at a rate of 5-10 % in the broiler finish- er diets (Kiiskinen 1988) and up to 20 % in layers' diets, replacing two thirds of soybean meal in the latter case (NAsi 1990).Oats protein (CP 44 %) which is obtained from the same process as BP 1 using oats as a raw material, has a very high energy value of 18.6 MJ/kg.The AME n of distillers' solids (CP 58 %) was approximately the same as that of BP 1.
The proximate analysis shows same kind ofcom- position for barley fibre and bran (Table 2).The AME n value of barley fibre, 9.06 MJ/kg DM was higher than that of barley bran (8.52) from the starch process or that of wheat bran (6.09) from the milling process (Table 4).According to Näsi  (1988) barley fibre contains 18 % available carbo- hydrates.In the present study the digestibility of total carbohydrates was found to be 48 % and that of wheat bran only 25 % (Table 5).Barley fibre and obviously also barley bran are potential feed ingredients for layers' diets.Barley fibre has success- fully been used at a rate of 20 % in layer's diets, replacing barley and soybean meal (NÄSI 1990).The AME of wheatbran was clearly lower than the table values of 7.3-7.S MJ/kg DM (WPSA 1989,   Salo et al. 1990).For wheat feed meal, approxi- mately the same ME value was obtained as given in the feedstuffs tables.
Dehulled grains proved to be rich in metaboli- zable energy, with barley containing AME n 14.2 and TME n 14.4 MJ/kg DM (Table 4).The corres- ponding values for oats were 14.6 and 14.8 MJ/kg DM.Supplementation with enzymes did not signi- 11 See the explanations of the abbreviations below Table 1 and 4 a-c See the footnote below Table 3 ficantly improve the ME values or the digestibility (Table 5) of dehulled barley.The ME value for dehulled oats was even better than that for wheat (Salo et al. 1990), and dehulled oats have been used as the sole grain in broiler diets (Kiiskinen 1986).
Compared with cane and beet molasses, the ash content of barley molasses from the starch process is only approximately one third (Table 2).This naturally improves the energy value of barley molasses.Its AME n value was 13.5 MJ/kg DM (Table 4), which is considerably higher than the 10.9-11.05MJ/kg DM for cane and beet molasses in the European tables (WPSA 1989).Wet droppings were not found in Experiment 3 although the content of molasses was 10 % in the diet.Barley molasses can be considered a good energy source for poultry, comparable with grains of the highest quality.
As energy feeds, dried potato pulp and malt sprouts rank in the same class with wheat bran (Table 2).The AME n of potato pulp was 6.23 MJ/kg DM.The digestibility of potato pulp carbo- hydrates was as high as 46 %, apparently due to starch left after the process (Table 5).The concen- trations of crude protein and crude fat were so low that their digestibility could not be determined with this method.Malt sprouts are rich in protein but the digestibility of this feedstuff was found to be very low and the AME n clearly lower than the table value 8.5 MJ/kg DM for poultry (Salo et al. 1990).

Table I .
Proximate analysis and gross energy (GE) content of the distillery feeds.

Table 2 .
Proximate analysis and GE values of the products from the integrated ethanol-starch, starch, and milling processes.

Table 3 .
Total feed intake of the birds (g DM) and ME values (MJ/kg DM) of the distillery feeds (SD = standard deviation).

Table 5 .
Apparent digestibility of the products in Experiments 1, 2 and 4 (SD = standard deviation).