Feeding and Excreta Collection Techniques in Metabolizable Energy Assays for Ducks1

O. ADEOLA,2 D. RAGLAND, and D. KING Department of Animal Sciences, Purdue University, West Lafayette, Indiana 47907 ABSTRACT Feeding and excreta collection techniques, lasting 102 h, for the determination of ME in feed ingredients for ducks are described. Eight and 32 h after feed withdrawal, all ducks received 30 g of dextrose in 100 g of water by orogastric administration. By orogastric administration, ducks received 30 g of test ingredients or dextrose (for ducks used in estimation of endogenous losses of energy and nitrogen) in 100 g of water at 48 and 54 h after feed withdrawal. The collection of excreta involved suturing a threaded plastic retainer ring to the vent and screwing a Whirl-Pak plastic bag, mounted on the top portion of a Playtex baby nurser set plastic bottle cut off 3 cm below the threads, to the retainer ring. Excreta were collected by replacing the Whirl-Pak bags at 54, 60, 72, 84, 96, and 102 h after feed withdrawal. In each of two experiments, ducks with an average weight of 3.7 kg were assigned to treatments. In Experiment 1, four ducks were assigned to each of three test ingredients (corn, dehulled oats, and wheat) and four ducks were assigned to be deprived of feed for estimation of endogenous losses of nitrogen and energy. In Experiment 2, six ducks were assigned to each of two test ingredients (corn and sorghum) and six ducks were assigned to be deprived of feed. Ducks lost an average of 537 g (Experiment 1) and 462 g (Experiment 2) during the 102-h experimental period and all the lost weight was regained within 7 d of return to full feed. Losses of nitrogen (milligrams per duck per 54 h) were 292 (Experiment 1) and 461 (Experiment 2) and energy (kilocalories per duck per 54 h) were 12.12 and 22.26 in feed-deprived group. The determined AMEn and TMEn for corn were 3.245 and 3.407, and 3.210 and 3.517 kcal/g for Experiments 1 and 2, respectively. For dehulled oats, wheat, and sorghum, the determined AMEn and TMEn were 3.464 and 3.625, 3.150 and 3.312, and 3.363 and 3.670 kcal/g, respectively.

(Key words: duck, excreta collection, metabolizable energy assay, harness) 1997 Poultry Science 76:728732

INTRODUCTION
Diet formulation for the duck employs ME values obtained from nutritional studies of domestic chicken. This formulation occurs because limited information is available on the duck and problems are normally encountered in collecting highly liquid excreta in duck ME assays. Assays for AME in feed ingredients for birds commonly rely on total collection of excreta in trays placed under the birds housed in cages. Collection of highly liquid excreta in trays placed under the ducks is subject to error due to splatter arising from contact of forcefully ejected excreta with trays and contamination with feed, dander, or scales. A collection device suitable for total excreta collection in chickens that appeared to be suitable for ducks was

reported by Revington et al. (1991). We made several unsuccessful attempts to apply the excreta collection technique reported by Revington et al. (1991) in duck ME assays. This communication reports the results of two experiments in which techniques were developed for tube-feeding and collecting excreta in ME assays with ducks.

MATERIALS AND METHODS

The excreta collection apparatus was fabricated using materials from the Playtex3 baby nurser set. Threaded plastic retainer rings from the nurser set with a 4.3-cm diameter hole in the center were modified by drilling 12 2-mm holes around the periphery as in the 12 points on a clock. Sixteen-week-old White Pekin male ducks were surgically fitted with the modified plastic retainer rings. During surgical fixation of the retainer rings, ducks were restrained in a Plexiglas box (8-mm wall thickness, 15 cm 17 cm 30 cm) and a 5-cm zone of feathers adjacent to the vent was removed to expose the skin. The skin was then sanitized with a dilute solution of chlorhexidine diacetate (Nolvasan4). 728

Received for publication September 27, 1996. Accepted for publication December 16, 1996. 1Journal paper Number 15223 of the Purdue University Agricultural Research Programs. 2To whom correspondence should be addressed. 3Playtex Products, Dover, DE 19901. 4Fort Dodge Laboratories, Inc., Fort Dodge, IA 50501.

METABOLIZABLE ENERGY ASSAYS FOR DUCKS

TABLE 1. Feeding and collection schedule Hours after feed withdrawal (h) 0 8 32 48

729

Day 1 1 2 3

Time (h) 0700 1500 1500 0700

Operation Food withdrawn Ducks fed dextrose solution (30 g/100 g water) Ducks fed dextrose solution (30 g/100 g water) Ducks fed (30 g/100 g water) appropriate feedstuff Ducks from which fasting energy loss is determined fed dextrose solution (30 g/100 g water) Whirl-pak bags placed through the bore of plastic bottle, screwed to retainer rings sutured to the vents Excreta collected and frozen by replacing Whirl-Pak bags Ducks fed (30 g/100 g water) appropriate feedstuff Ducks from which fasting energy loss is determined fed dextrose solution (30 g/100 g water) Excreta collected and frozen by replacing Whirl-Pak bags Excreta collected and frozen by replacing Whirl-Pak bags Excreta collected and frozen by replacing Whirl-Pak bags Excreta collected and frozen by replacing Whirl-Pak bags Excreta collected and frozen by replacing Whirl-Pak bags

3 3 4 4 5 5

1300 1900 0700 1900 0700 1300

54 60 72 84 96 102

The area in which the retainer ring was to be sutured was then infused in the dorsal, ventral, and lateral quadrants around the vent with 2% lidocaine hydrochloride to desensitize the skin for suturing. The retainer rings were then sutured to the vent area using a continuous suture pattern with the retainer rings anchored in place by passing the needle and suture through 2-mm holes drilled in the periphery of the retainer rings. Ducks were used in experiments approximately 72 h after suturing retainer rings to the vents. A plastic bottle of the nurser set was cut to a length of 3 cm below the threads on the bottle. During collection, Whirl-Pak5 bags were then placed through the bore of the bottle and the flaps of the bag overlaid to the sides of the bottle covering the threads. The bottle and WhirlPak bag were then screwed onto the modified retainer ring attached to the bird with the threads of the ring and bottle securing the bag in place. The Whirl-Pak bags containing excreta were changed as indicated in Table 1. Tube-feeding apparatus consisted of a 60-mL catheter-tip syringe and a 35-cm long Nalgene6 tubingwith an inside diameter of 8 mm. Forty-eight hours prior to feeding the test ingredients, feed was withdrawn from all ducks. All test ingredients were ground through a 0.5-mm screen prior to feeding. Feeding was done by mixing 30 g of test ingredient with 80 g of deionized water in a 125-mL beaker. Ducks were intubated and the gruel was poured into the 60-mL syringe and pumped into the crop with a plunger. The beaker was rinsed with 20 g of deionized water, poured into the syringe and pumped into the crop. The feeding and excreta collection schedule is presented in Table 1. At 48 and 54 h after feed

5Nasco, Fort Atkinson, WI 53583. 6Fishers Scientific, Itasca, IL 60143. 7LECO Corp., St. Joseph, MI 49085. 8Parr Instrument Co., Moline, IL 61265.

withdrawal, the ducks assigned to the feed deprived group for estimation of endogenous losses were tube-fed 30 g of dextrose in 100 g of water. All ducks were fitted with their respective collection vessels at the time of the first feeding of test ingredients and excreta was collected for 54 h as shown in Table 1 (Hours 48 to 102). The feeding, surgical, and collection protocols were approved by the Purdue University Animal Care and Use Committee. Ducks were weighed and sorted according to weight and placed in stainless-steel cages (0.66 m 0.66 m) such that the average weight in each treatment was similar. Ducks were housed in a facility in which a temperature of approximately 25 C was maintained and 24 h light /d was provided. In Experiment 1, four ducks were assigned to each of three test ingredients (corn, dehulled oats, and wheat) and four ducks were assigned to be deprived of feed for estimation of endogenous losses of nitrogen and energy. In Experiment 2, six ducks were assigned to each of two test ingredients (corn and sorghum) and six ducks were assigned to be deprived of feed for estimation of endogenous losses of nitrogen and energy. The group of ducks used in Experiment 1 was different from that used in Experiment 2. An excreta sample from each duck was dried at 55 C for 48 h and ground through a 0.5-mm screen prior to analysis. Dry matter of the test ingredients and all excreta samples (previously dried at 55 C for 48 h) was determined by drying the samples at 110 C for 24 h. Nitrogen content of test ingredients and excreta was determined by the combustion method using the Model FP2000 nitrogen analyzer.7 Energy content of test ingredients and excreta was determined by bomb calorimetry using an adiabatic calorimeter.8 The AME, AMEn, TME, and TMEn of the test ingredients were calculated as follows: AME = (EI EO) FI; AMEn = AME (8.22 ANR FI); TME = AME + (FEL FI); TMEn = AMEn + (FEL FI) (8.22 FNL FI), where EI is gross energy intake (kilocalories); EO is

730

ADEOLA ET AL.
TABLE 2. Initial and final weights of ducks, Experiments 1 and 2 Item Experiment 1 Feed-deprived Corn Dehulled oats Wheat SEM Experiment 2 Feed-deprived Corn Sorghum SEM
acMeans

Initial weight

Final weight (g)

Weight gain

3,726 3,746 3,742 3,729 154 3,767 3,771 3,747 160

3,168 3,093 3,311 3,223 142 3,281 3,344 3,265 137

558b 653a 431c 507b 75 476 427 483 45

4 4 4 4

6 6 6

with no common superscript differ significantly (P 0.05).

gross energy output in the excreta (kilocalories); FI is feed intake (grams); ANR is apparent nitrogen retention (grams) calculated as the difference between nitrogen intake and nitrogen output; FEL is the fasting energy loss from the group of the feed-deprived ducks (kilocalories); and FNL is fasting nitrogen loss from the group of feed-deprived ducks (grams). Data from the two experiments were subjected to the General Linear Models (GLM) procedures of SAS (SAS Institute, 1990) appropriate for a completely randomized design. Means were separated using the least significant difference test (Steel and Torrie, 1980).

RESULTS AND DISCUSSION

Surgical attachment of the modified retainer rings to the ducks was critical for collection of contaminant-free excreta and the accurate estimation of nitrogen and energy output. The use of physical restraint and local anesthetic during attachment of the retainer rings minimized stress and discomfort. The procedure was done under the most hygienic condition possible and no infections or cellulitis were observed after surgical attachment. During the experiments, ducks adjusted very well to the collection apparatus and there was no appearance of any discomfort or impaired mobility. The ME assay used in the current experiments followed the standard techniques devised by Sibbald (1976) and the modifications suggested by McNab and Blair (1988). The initial 48-h period of feed deprivation used in the current studies followed a report by McNab and Blair (1988) that the residue remaining in the digestive tract of cockerels after 48 h was much less than after 24 h of feed deprivation (0.17 0.08 vs 1.59 0.56 g). In using the assay for ducks, it became necessary to modify the techniques. We increased the amount of test ingredient or dextrose tube-fed from 50 to 60 g, thus maintaining the feeding level at approximately 1.5 % body weight. Feeding dextrose to birds from which endogenous losses are collected decreases excessive weight loss and reduces the variability in endogenous losses (McNab and Blair, 1988). During collection, the test ingredient

and dextrose was fed in two equal portions 6 h apart (at Hours 48 and 54, Table 1) and excreta collection was extended from a total of 48 to 54 h; thus, the experimental period was increased from 96 to 102 h. This modification became necessary because in preliminary experiments, ducks regurgitated generous portions of the test ingredients when 50 g was tube-fed at one time. In an earlier report, Mohamed et al. (1986) suggested the necessity to feed the test materials in two portions to avoid regurgitation. The collection technique we developed was inspired in part by the technical note published by Revington et al. (1991), in which specimen container caps with 3.5-cm center holes were secured to the vents of chickens using cotton tapes tied up over the back just posterior to the wings and around the base of the tail; and rigid specimen containers were attached to serve as a collection vessel. This approach was unsuitable due to displacement of the specimen container cap and container when the ducks assumed a squatting position. With watery excreta, samples spilled from the collection cups. For collection apparatus, we later devised the use of materials from the Playtex baby nurser set. Surgical attachment of a collection apparatus to the vent of the duck was considered a more suitable method because it provided better security against excreta loss. Cutting the plastic bottles to a length of 3 cm below the threads on the bottle prevented undue tension on the sutures and displacement of the collection apparatus when the ducks assumed a squatting position because the bottle did not make contact with the cage floor. Excreta samples were stored in the same plastic bags in which they were collected, thus preventing losses that might occur during transfer to other containers. In Experiment 1, the analyzed nitrogen of corn, dehulled oats, and wheat were 1.12, 1.74, and 2.08%, respectively, and gross energy of corn, dehulled oats, and wheat were 3.991, 4.092, and 3.890 kcal/g, respectively. In Experiment 2, analyzed nitrogen and gross energy of corn and sorghum were 1.12% and 3.984 kcal/ g, and 1.76% and 4.181 kcal/g, respectively. The way ducks were assigned to the fasting treatment and test ingredients ensured that initial weight was similar across treatments in each experiment (Table 2). In

METABOLIZABLE ENERGY ASSAYS FOR DUCKS

TABLE 3. Fasting losses of nitrogen1 and energy2 for feeddeprived ducks, Experiments 1 and 23 Item Experiment 1 Fasting losses of Nitrogen Energy Experiment 2 Fasting losses of Nitrogen Energy
1Milligrams 2Kiloocalories

731

Mean

SD

Range

292 12.12

112 3.63

142 to 444 9.36 to 18.74

461 22.26

326 2.84

258 to 1,139 17.59 to 28.05

per duck per 54 h. per duck per 54 h. 3Mean weights (initial, final, and loss) are indicated for feeddeprived ducks in Table 2.

Experiment 1, ducks that were fed 60 g of corn lost more (P < 0.05) weight than either the feed-deprived or wheat-fed ducks. Ducks that received dehulled oats lost the least (P < 0.05) weight. In the second experiment, there was no difference in weight loss across treatments, with ducks losing an average of 462 g over the 102-h experimental period (Table 2). When returned to full feed after the 102-h experiment, ducks regained all the weight lost within 7 d in both experiments. Although different groups of ducks were used in both of the experiments reported here, based on our experience in subsequent studies, the ducks could be used repeatedly after recovery from lost weight without any apparent problems or loss of retainer rings sutured to the vents. The fasting losses of energy and nitrogen used in calculating TME and TMEn were much higher in Experiment 2 than in Experiment 1 (Table 3). Fasting losses of nitrogen and energy are only reasonably approximate and will vary from time to time and from bird to bird (McNab and Blair, 1988). In experiments with adult cockerels, fasting energy losses of between 14 and 21 kcal/48 h were reported (McNab and Blair, 1988; Yalcin and Onol, 1994), which are similar to 12 and 22

kcal/54 h observed in Experiments 1 and 2, respectively. Nitrogen retention was higher (P < 0.05) in ducks that were fed wheat than in those fed dehulled oats, which in turn was higher (P < 0.05) than in those fed corn in Experiment 1 (Table 4). In the second experiment, nitrogen retention was lower (P < 0.05) in ducks fed corn than in those fed sorghum. The differences in nitrogen retention presumably are reflective of the nitrogen intake. Energy voided in the excreta was not different across treatments in either experiment. In Experiment 1, the AME, AMEn, TME, and TMEn for corn and wheat were similar to (P > 0.05) but lower than (P < 0.05) those of dehulled oats (Table 5). Corn and sorghum had similar energy values as determined in Experiment 2. The NRC (1994) TMEn value for corn in cockerel assays is 3.470 kcal/g, which lies in the range of 3.407 and 3.517 kcal/g obtained for corn in Experiments 1 and 2, respectively. The TMEn values from ducks determined in the current experiments vs NRC (1994) values from cockerel assays for dehulled oats, wheat, and sorghum experiments were 3.625 vs 2.625 (hulled oats), 3.312 vs 3.167, and 3.670 vs 3.376 kcal/g, respectively. Thus, the use of these values in formulating duck diets would make dehulled oats, wheat, and sorghum competitive in replacing corn, and values from chicken AME assays may not be appropriate for use in formulating duck diets. Correction of TME for nitrogen resulted in a 2 to 5% reduction in TME values of ingredients examined in both experiments, an observation similar to the 2 to 4% reduction reported by McNab and Blair (1988) in TME of ingredients for cockerels. Furthermore, the TMEn values obtained were higher than the AMEn values, which agrees with previous reports of 9 to 18% higher TME than AME in a variety of feed ingredients (Sibbald and Price, 1977; Baidoo et al., 1991). The tube-feeding and excreta collection methods described in this communication offer means of precisely feeding known amounts of ingredients and accurately collecting contaminant-free voided excreta

TABLE 4. Nitrogen and energy balances of ducks, Experiments 1 and 21 Nitrogen intake Nitrogen output (mg) Experiment 1 Corn Dehulled oats Wheat SEM Experiment 2 Corn Sorghum SEM
a,b,cMeans 1Mean

Item

Nitrogen retention

Energy intake (kcal)

Energy output

675 1,051 1,316

460 477 431 62 524 546 74

216c 573b 884a 62 148 503 74*

239.18 245.63 233.21

42.69 33.10 36.96 4.83 44.86 45.34 5.60

4 4 4

671 1,048

239.13 250.78

6 6

in the same column and experiment with no common superscript differ significantly (P < 0.05). weights (initial, final, and loss) are indicated for ducks fed test ingredients in Table 2. *Treatment effect significant at P < 0.05.

732

ADEOLA ET AL.
TABLE 5. The AME, AMEn, TME, and TMEn of ingredients, Experiments 1 and 2 Item Experiment 1 Corn Dehulled oats Wheat SEM Experiment 2 Corn Sorghum SEM
a,bMeans

AME

AMEn

TME (kcal/kg)

TMEn

3.275b 3.542a 3.271b 0.080 3.230 3.432 0.093

3.245b 3.464a 3.150b 0.073 3.210 3.363 0.084

3.477b 3.744a 3.473b 0.080 3.600 3.802 0.093

3.407b 3.625a 3.312b 0.073 3.517 3.670 0.084

4 4 4

6 6

in the same column and experiment with no common superscript differ significantly (P < 0.05.

that are extremely vital for obtaining reliable values from ME assays. The technique provides a viable alternative to pan collection.

ACKNOWLEDGMENTS
The valuable technical assistance provided by Charles Thomas and Brian Ford, and the donation of ducks by Maple Leaf Farms, Milford, IN 46542 are thankfully acknowledged.

REFERENCES
Baidoo, S. K., A. Shires, and A. R. Robblee, 1991. Effect of kernel density on the apparent and true metabolizable energy value of corn for chickens. Poultry Sci. 70: 21022107. McNab, J. M., and J. C. Blair, 1988. Modified assay for true and apparent metabolizable energy based on tube feeding. Br. Poult. Sci. 29:697707.

Mohamed, K., M. Larbier, and B. Leclercq, 1986. A comparative study of the digestibility of soybean and cottonseed amino acids in domestic chicks and Muscovy ducklings. Ann Zootech. 35:7986. National Research Council, 1994. Nutrient Requirements of Poultry. 9th rev. ed. National Academy Press, Washington, DC. Revington, W. H., N. Acar, and E. T. Moran, Jr., 1991. Research note: Cup versus tray excreta collections in metabolizable energy assays. Poultry Sci. 70:12651268. SAS Institute, 1990. SAS Users Guide: Version 6. 4th ed. SAS Institute Inc., Cary, NC. Sibbald, I. R., 1976. A bioassay for true metabolizable energy in feedstuffs. Poultry Sci. 55:303308. Sibbald, I. R., and K. Price, 1977. True and apparent metabolizable energy values for poultry of Canadian wheats and oats measured by bioassay and predicted from physical and chemical data. Can. J. Anim. Sci. 57:365374. Steel, R.G.D., and J. H. Torrie, 1980. Principles and Procedures of Statistics. A Biometrical Approach. 2nd ed. McGrawHill Book Co., New York, NY. Yalcin, S., and A. G. Onol, 1994. True metabolizable energy values of some feedings. Br. Poult. Sci. 29:119122.