Effect of steam pelleting and inclusion of molasses in amaranth diets on broiler chicken performance, carcass composition and histopathology of some internal organs

L.W. Kabuagea*, P.N. Mbuguaa, B.N. Mitarua, T.A. Ngatiab

aDepartment of Animal Production,

bDepartment of Veterinary Pathology and Microbiology,

University of Nairobi, P.O. Box 29053, Nairobi, Kenya.

* Corresponding author: email address: kabuage@net2000ke.com


A study was carried out to determine the effect of pelleting grain amaranth diets with or without molasses on broiler chicken performance, carcass composition and histopathology of some internal organs. The grain was incorporated in eight diets at 20 or 40% levels. Four diets were in mash form while the other four were steam pelleted at 70oC. Molasses was included in four diets. The eight diets were fed to one week old broiler chicks up to eight weeks of age and compared to a maize-soyabean meal control diet. While pelleting improved (P<0.05) body weight, feed intake and feed efficiency, it also increased (P<0.05) carcass fat and reduced (P<0.05) carcass moisture. Molasses inclusion did not increase (P>0.05) feed intake, body weight or feed efficiency at 8 weeks of age. Chicks on 20% amaranth diets were heavier (P<0.05) than those on 40% inclusion at 4 weeks of age but the difference was not significant (P>0.05) at 8 weeks of age. The former group of chicks gave a more efficient (P<0.05) carcass protein retention. The pelleted diets resulted in lower (P<0.05) pancreas weights than the mash diets at four weeks of age but the difference was not significant (P>0.05) by eight weeks of age. Histopathology of the pancreas and liver showed moderate changes which could not be attributed to the feeding of amaranth. Although pelleting of amaranth diets was beneficial in improving growth, chicks on mash diets largely overcame the adverse effects of raw amaranth as they matured.

1. Introduction

Grain amaranth has an energy content similar to that of cereal grains and about twice the amount of protein, with a superior amino acid composition. The grain would hence serve as a suitable substitute for maize grain in poultry feed and partly replace the expensive animal protein ingredients. However, acceptability and utilization of raw grain amaranth by poultry is poor probably due to the presence of heat labile anti-nutritional factors. Increasing levels of the raw grain in the diet results in declining feed intake and body weight gain (Waldroup et al. 1985; Laovoravit et al, 1986; Kabuage, 1996). Some of the anti-nutritional factors reported include tannins, trypsin inhibitors, lectins and saponins (Cheeke and Bronson, 1980; Lorenz and Wright, 1984; Calderon et al. 1985; Koeppe et al, 1985; Koeppe and Rupnow, 1988; Kabuage, 1996).

Numerous metabolic processes occur in the body before ingested toxic plant materials (anti-nutritional factors) exert their negative effects or are excreted. These processes could be in the gastro-intestinal tract, in the liver or in other tissues (Cheeke and Shull, 1985). Certain lectins can cause liver necrosis and diminished levels of liver enzymes (Liener, 1980). Liver biopsies can be used to assess any sequential changes in liver histology (Cheeke and Shull, 1985). High levels of trypsin inhibitor are known to cause adverse effects on the pancreas which can result in hypertrophy of that organ (Cheeke and Shull, 1985; Huisman and Tolman, 1992). However, thermal processing by extrusion and autoclaving apparently improves the nutritional value of grain amaranth (Tillman and Waldroup, 1986; Kabuage, 1996). Owing to the high cost of extrusion and autoclaving, there is need to evaluate a conventional heat processing method such as steam pelleting, which is easily accessible to feed manufacturers. To counteract the poor palatability, it might also be worthwhile to include a flavouring agent such as molasses in raw grain amaranth diets. This study was designed to determine the effect of steam pelleting and inclusion of molasses in grain amaranth diets on broiler performance, carcass composition and histopathology of the pancreas and liver.

2. Materials and methods

Two hundred and eighty eight one-week old broiler chicks were divided into thirty six groups of eight chicks each. Each group was weighed, put in a 1m2 electrically heated pen and randomly assigned to one of nine diets made to NRC (1984) requirements (Table 1). The chicks were fed up to eight weeks of age. There were four replicates (pens) per dietary treatment in a completely randomised design. Grain amaranth (A. hypochondriacus) was included in 8 diets, half of them containing 20% level while the rest had 40%. Four of the diets had cane molasses. Diet 1 was a maize-soyabean meal control diet. Four amaranth diets were steam pelleted using a California Press type pelleting machine (made by Simon Barron Ltd., England). The pelleting temperature was 70oC and steam was applied at a pressure of 5.1 kg/cm2. The pellets were then crumbled to a smaller size suitable for chick feeding. Feed and water were provided ad libitum.

Data on weekly broiler performance and carcass composition at eight weeks of age was recorded. The percentage of nutrient retained in the carcass was calculated as:-

The diets were analysed for chemical composition using standard procedures (AOAC, 1984). True metabolisable energy (TME) was determined according to the method of Sibbald (1986). For each diet, 3 adult Isabrown cockerels were starved for 24 hours, then precision fed with 40 grams of the test feed using a tube funnel and plunger device inserted through the oesophagus to the crop. Each bird was placed in an alternate wire cage and an excreta collection tray placed underneath. Cockerels that were not fed during the experimental period served as negative controls, providing for an estimation of the metabolic and endogenous energy losses. Excreta voided by the cockerels for the next 48 hours was collected, dried in a draft oven at 60oC and assayed for gross energy using an adiabatic bomb calorimeter. At the end of the 4th and 8th weeks of age, 1 bird per replicate was weighed, killed by cervical dislocation and its pancreas removed and weighed. Sections of the pancreas and liver were then obtained, put in 10% formalin and thereafter processed for histopathological examination. At the end of 8 weeks of age, two birds in each replicate were picked at random, starved for 24 hours, sacrificed, defeathered and weighed. They were then cut up into small sections and minced thoroughly using a large manual mincer. After uniform mixing of the minced carcass, small samples were taken, minced fine in a blender and used for determination of proximate analysis following standard procedures (AOAC, 1984). Statistical analysis of data on the nine dietary treatments was done by analysis of variance procedures (SAS, 1988) and means compared using Tukey's multiple range test. This was followed by statistical analysis of the eight amaranth treatments as 2 x 2 x 2 (amaranth level x processing x molasses inclusion) factorial arrangement.

3. Results and Discussion

3.1 Broiler performance

i) 4 weeks of age

The mean body weight of chicks on the control diet was similar to that of chicks on pelleted diets except the one with 40% amaranth and no molasses. There were no differences (P>0.05) in feed intake between the nine diets. However, feed efficiency for the two non-pelleted 40% amaranth diets was poorer (P<0.05). Results of broiler performance on the eight amaranth diets are shown in Table 2. Chicks on pelleted amaranth diets grew faster (P<0.05) and had better (P<0.05) feed efficiency than those on mash diets. Similarly, chicks on diets containing amaranth at 20% level had a higher (P<0.05) body weight and better feed efficiency than those on 40% inclusion. There were however no significant differences (P>0.05) in feed intake between diets. Inclusion of molasses resulted in heavier (P<0.05) broilers at this age although it had no significant effect (P>0.05) on feed intake or feed efficiency. The interaction between level of amaranth and pelleting was significant (P<0.05) for body weight and feed efficiency. The pancreas weights for the nine diets at 4 weeks of age were in the range of 0.32-0.42% of body weight. For the eight amaranth diets, pelleting resulted in lower (P<0.05) pancreas weights than the mash diets.

ii) 8 weeks of age

The differences in chick body weight between the nine dietary treatments were not significant (P>0.05) at this age. Utilisation of unprocessed amaranth diets apparently improved as the chicks matured beyond 4 weeks of age. Results of broiler performance on the eight amaranth diets are shown in Table 3. A comparison of the pelleted and mash diets showed that the former diets resulted in faster (P<0.05) chick growth than the latter. The level of amaranth had no significant effect (P>0.05) on body weight at this age. The feed intake trend showed higher (P<0.05) consumption of the pelleted diets and those containing 20% amaranth. A similar pattern was observed with feed efficiency. Inclusion of molasses had no effect (P>0.05) on body weight, feed intake and feed efficiency. The pancreas weights were not affected by the amaranth level or pelleting of diets by this age. Unlike the case at 4 weeks of age, the pancreas weights for the nine diets ranged between 0.18-0.22% of body weight, demonstrating differential growth rate of various body tissues as the bird matured.

The above results of broiler performance at 4 and 8 weeks of age indicated that steam pelleting of amaranth diets improved their consumption and utilisation. The moist heating process was apparently effective in inactivating any heat labile anti-nutritional factors possibly present in grain amaranth hence increasing feed intake and growth rate. This was in agreement with results reported by Sengor and Bayne (1991) that pelleting of broiler feed increased energy intake, growth rate and feed efficiency. Steam pelleting improved the nutritional value of certain feedstuffs through enhanced nutrient availability or destruction of heat labile toxins (Slinger, 1973; Smits et al., 1993). Inclusion of molasses in the current study did not affect (P>0.05) feed intake or feed efficiency. Unlike steam pelleting, molasses failed to effectively increase acceptability of grain amaranth.

The mash diets might have had a substantial level of trypsin inhibitor arising from the presence of raw grain amaranth which influenced the enlargement of the pancreas at the early age. The older chicks probably adapted and overcame this adverse effect as demonstrated by similar (P>0.05) pancreas weights at 8 weeks of age. Grain amaranth (A. hypochondriacus) was reported to have a higher trypsin inhibitor content (0.52TIU/mg) than maize (0.35 TIU/mg) or wheat (0.10 TIU/mg) by Koeppe at al. (1985). Similarly, Kabuage (1996) obtained a trypsin inhibitor level of 0.6 TIU/mg in raw A. hypochondriacus grain and 0.20 TIU/mg after the grain was extruded. The results of the current study are further in agreement with those discussed by Chubb (1982) who reported that diets containing moderate levels of trypsin inhibitor depressed growth and produced a slightly enlarged pancreas (0.44% of body weight). This occurs due to the enhanced production of pancreatic proteolytic enzymes in an attempt to cope with inactivation resulting from the trypsin inhibitor (Cheeke and Shull, 1985; Huisman and Tolman, 1992). The lack of significant differences in body weight at 8 weeks of age indicated a higher sensitivity to different diets at the early age of 4 weeks compared to older birds. This was consistent with the hypothesis that young animals were more sensitive to anti-nutritional factors than the older ones and the threshold capacity to cope with the adverse effects increased with age (Huisman and Tolman, 1992). Similarly, Gous (1986) noted that results from a trial conducted during one stage of growth need not reflect the response at some other stage due to the systematic changes that took place as the bird aged. The improved performance of older birds on amaranth diets in the current study indicated that the grain might be more suitable in broiler finisher diets.

3.2 Carcass composition

The carcass protein content showed less variation than fat and moisture between the various dietary treatments. The broilers on the control diet had similar (P>0.05) carcass weight as those on amaranth diets except the two non-pelleted with 40% amaranth. There were no differences (P>0.05) in carcass moisture and protein between the nine diets. Protein retention was similar (P>0.05) between the 20% amaranth diets and the control. Results of carcass composition from the eight amaranth diets are shown in Table 4. The pelleted diets gave higher (P<0.05) carcass fat levels and lower (P<0.05) moisture and protein contents than the mash diets. This trend corresponded with the higher (P<0.05) body and carcass weights obtained with the pelleted diets. The faster growth rate of chicks on pelleted diets hence resulted in heavier but fatter (P<0.05) carcasses compared to those on mash diets. This was probably due to higher dietary energy consumption from the increased feed intake inducing higher fat deposition. The carcass composition was similar to that obtained by Keren et al. (1990) with 8 week old broiler chickens. These authors however reported slightly higher values for protein and moisture contents and lower ones for the lipid content since the carcasses in their study had their skin removed. The chicken skin contains considerable levels of fat. Increased dietary energy is known to produce heavier but fatter carcasses (Leenstra, 1989; Leclerq, 1986). Chicks on 20% amaranth diets had a higher (P<0.05) carcass weight and carcass protein than those on 40% inclusion. In addition, they had higher (P<0.05) carcass protein retention, which is a measure of protein conversion efficiency expressed as carcass protein: feed protein (Table 4). This reflected more efficient utilisation of low amaranth diets for lean tissue growth. The protein retention values were within the range reported by Fisher (1980). Low gross protein conversion efficiency as indicated by a low carcass protein growth is probably caused by an imperfect balance of amino acids in dietary protein (Fisher, 1980). In the current study, the higher carcass protein retention for 20% amaranth diets indicated a better amino acid balance for protein synthesis from these diets than from those containing a higher (40%) amaranth level. This was yet another pointer on the inefficient utilisation of high levels of raw grain amaranth in broiler diets. Inclusion of molasses produced leaner (P<0.05) carcasses with a higher (P<0.05) protein content and protein retention. These results were consistent with the generally lower metabolisable energy values obtained for the molasses containing diets. Replacement of maize with molasses apparently reduced the metabolisable energy content of the diets which might have consequently caused reduction in fat deposition.

3.3 Histopathology

Results of microscopic evaluation of liver and pancreas were as follows:-


The birds did not show acinar dissociation or necrotic changes at both 4 and 8 weeks of age. The secretory activity of the acini varied from moderate to maximum but this did not correspond to any pattern of the dietary treatments. Some lymphocytic infiltration was observed in all the treatments at 8 weeks of age except for the control diet.


Results of changes in this organ are summarised in Table 5. There were generally more degenerative changes in chicks at 8 weeks than at 4 weeks of age. Chicks on mash diets exhibited more of these changes compared to those on control or pelleted diets. For all diets, necrosis was absent. Lymophytic infiltration was higher with the amaranth diets at 8 weeks of age than with the control.

The differences in histopathological changes of the pancreas and liver between the various dietary treatments were not large enough to be significant. Hence, the lack of evidence attributing the changes to the diets.

4. Conclusion

Steam pelleting of grain amaranth diets was beneficial to broiler chicks through increased feed intake and faster growth but with higher fat deposition. Molasses was not effective in enhancing feed intake of amaranth diets. Utilisation of the grain improved with age of the birds demonstrating that raw grain amaranth would be more suitable as an ingredient for broiler finisher diets. Histopathological changes of the internal organs were generally moderate and could not be attributed to the feeding of amaranth. This study shows that steam pelleting is an effective method of processing amaranth diets for improved broiler performance.


AOAC, 1984. Association of Official Analytical Chemists, Official Methods of Analysis - Animal Feed Section.

Calderon de la Barca, A.M., J.L. Ochoa and M.E. Valencia, 1985. Effect of the extraction of a hemagglutinin on the nutritive value of Amaranthus leucocarpus seeds. J. Food Sci., 50: 1700-1702.

Cheeke, P.R. and L.R. Shull, 1985. Natural toxicants in Feeds and Poisonous Plants. AVI Publishing Company Inc., Westport, Connecticut, U.S.A. pp. 33-48; 235-242; 321-322.

Cheeke, P.R., and J. Bronson, 1980. Feeding trials with amaranthus grain, forage and leaf protein concentrates. In: Proceedings of the Second Amaranth Conference. Rodale Press, Emmaus, PA, U.S.A. pp. 5-11.

Chubb, L.G., 1982. Anti-nutritive factors in animal feedstuffs. In: Recent advances in animal nutrition. Ed. W. Haresign. Butterworths, UK. pp 21-38.

Fisher, C., 1980. Protein deposition in poultry. In: Protein deposition in animals. Ed. P.J. Buttery and D.B. Lindsay. Butterworths, U.K., pp. 251-270.

Gous, R.M. 1986. Measurement of response in nutritional experiments. In: Nutrient requirements of poultry and nutritional research. Ed. C. Fisher and K.N. Boorman, Butterworths, U.K., pp. 41-58.

Huisman, J. and G.H. Tolman, 1992. Antinutritional factors in the plant proteins of diets for non-ruminants. In: Recent advances in animal nutrition. Ed. Garnworthy, W., Haresign and D.J.A. Cole, pp. 3-31.

Kabuage, L.W. 1996. Nutritive evaluation of grain amaranth (Amaranthus spp.) in broiler chicken diets. Ph.D. Thesis, University of Nairobi, Kenya.

Keren-ZVI, S., I. Nir, Z. Nitsan and A. Cahaner, 1990. Effect of dietary concentrations of fat and energy on fat deposition in broilers divergently selected for high or low abdominal adipose tissue. Br. Poult. Sci., 31: 507-516.

Koeppe, S.J., J.H. Rupnow, C.E. Walker, and A. Davis, 1985. Isolation and heat stability of trypsin inhibitors in amaranth (Amaranthus hypochondriacus). J. Food Sci., 50: 1519-1521.

Koeppe, S.J., and J.H. Rupnow, 1988. Purification and characteristics of a lectin from the seeds of amaranth (Amaranthus cruentus). J. Food Sci. 53: 1412-1422.

Laovoravit, N., F.H. Kratzer and R. Becker, 1986. The nutritional value of amaranth for feeding chickens. Poultry Sci. 65: 1365-1370.

Leclercq, B. 1986. Energy requirements of avian species. In: Nutrient requirements of poultry and nutritional research. Ed. C. Fisher and K.N. Boorman, Butterworths, U.K. pp. 125-140.

Leenstra, F.R. 1989. Influence of diet and genotype on carcass quality in poultry and their consequences for selection. In: Recent developments in poultry nutrition. Ed. D.J.A. Cole and W. Haresign. Butterworths, U.K., pp. 131-144.

Liener, I.E.. 1980.. Toxic constituents of plant foodstuffs. 2nd Ed. Academic Press, New York, U.S.A. pp. 40-49.

Lorenz, K., and B. Wright. 1984. Phytate and tannin content of amaranth. Food Chem. 14: 27-34.

National Research Council (NRC) 1984. Nutrient requirements of poultry. National Academy of Sciences, Washington, D.C.

SAS Institute, 1988. SAS User's guide. Statistics. SAS Institute Inc., Cary, N.C.

Sengor, E. and B.E.F. Bayne, 1991. The effects of feed texture on performance in broilers. Nutr. Abstr. Rev., 61(10): 5322.

Sibbald, I.R., 1986. The T.M.E system of feed evaluation: methodology, feed composition data and bibliography. Technical Bulletin 1986 - 4E. Research Branch, Agriculture, Canada.

Slinger, S.J. 1973. Effect of pelleting and crumbling methods on the nutritional value of feeds. In: Effects of processing on the nutritional value of feeds. National Academy of Sciences, Washington D.C., pp. 48-45.

Smits, B., A.W. Jongbloed and L.B.L. Sebek, 1993. Factors affecting pig diet digestibility. Feed Mix. 1 (3): 42-45.

Tillman, P.B. and P.W. Waldroup, 1986. Processing grain amaranth for use in broiler diets. Poultry Sci. 65: 1960-1964.

Waldroup, P.W., H.M. Hellwig, D.E. Longer, and C.S. Endres, 1985. The utilisation of grain amaranth by broiler chickens. Poultry Sci. 64: 759-762.

Top Of Page