Importance of vitamin A supplementation for performance of Sonali chickens under smallholder farm conditions in a tropical climate

A R Bhuiyan, C Lauridsen, A R Howlider^* and K Jakobsen

Department of Animal Nutrition and Physiology, Danish Institute of Agricultural Sciences,
Research Centre Foulum, 8830 Tjele, Denmark
kirsten.jakobsen@agrsci.dk
^* Department of Poultry Science, Bangladesh Agricultural University,
Mymensingh, Bangladesh

Abstract

A total of 600 Sonali chickens (Rhode Island Red x Fayoumi) were reared at 3 farms in two dietary groups (Control and VitA) for 10 weeks in the rural area of Bangladesh. The control received a basal diet without added vitamin A until 42 days of age. Then the chickens were treated orally with 500 IU vitamin A daily for 5 days, and thereafter received 500 IU vitamin A/kg feed until the end of the experiment. The VitA group received 1500 IU vitamin A/kg feed during the whole experiment. Vitamin A was added as retinol acetate.

Vitamin A deficiency symptoms occurred in the control group within days 18 to 39 with different severity at the three farms. Overall mortality was 51.0% in the control group and 5.7% in the VitA group. Feed consumption, body weight gain and feed conversion ratio were significantly poorer in the control compared with the VitA group.  Retinol levels in blood plasma and liver at 42 days of age in the control group were very variable and were not detectable in 10 out of 12 birds. The oral supplementation of vitamin A to the control group increased the liver concentration of vitamin A to a higher level than for the VitA group, while the retinol concentration in blood plasma was similar in the two groups (0.16 to 0.39 µg/ml).

It is concluded that the diets must be supplemented with vitamin A, at least with 1500 IU/kg feed. However, the dietary level and the stability of vitamin A need to be determined under relevant farming conditions using local breeds.

Key words:  Deficiency symptoms, chickens, growth, mortality, retinol, vitamin A

Introduction

Poultry are very rapidly affected by vitamin A (retinol) deficiency, which will seriously affect growth rate, feed utilization, development of bone, movements, vision, reproduction, resistance against diseases, and mortality. Deficiency symptoms occur usually within 3 to 4 weeks. Early signs of vitamin A deficiency are loss of appetite and decreased growth rate followed by general weakness, staggering gait, and ruffled plumage. Birds become more susceptible to infection, and both egg production and hatchability are markedly reduced. The eyes are affected with an abnormal exudate and epithelial keratinisation (for review see Olson 1991).

Vitamin A deficiency may be caused by several factors. In natural feedstuffs vitamin A is only present in animal products, notably liver, eggs, fish meal, and fish oil. Pro-vitamins A, of which β-carotene is the most common vitamin A source are only present in plants. According to NRC (1994) poultry are as efficient as the rat in converting β-carotene into vitamin A, and 1mg of β-carotene equals 1667 IU retinol. However, recent findings  suggest that the vitamin A activity of β-carotene may be overestimated. In broiler chickens it was found that 1mg of β-carotene was equal to 400 IU retinol (Johannsen et al 1998), in older geese (3-4 months old) to 1200 IU and in young geese (1-3 weeks of age) only to 60 IU (Jamroz et al  2002). The commercial source of vitamin A is an ester of retinol, either retinol acetate or palmitate. These are susceptible to light, oxygen, heat, moisture and pressure during processing (Roche 1994; Barua and Olsen 2000). Thus, in tropical climate zones with high temperature and humidity, the storage stability of premix products containing vitamins may be limited. Feed ingredients may also indirectly induce oxidation. Soya-beans for example contain a lipoxidase that readily destroys the carotenoids present in the feed unless quickly inactivated (Roche 1994).

In Bangladesh, the mortality of chickens is high and performance is low especially in flocks of small poultry holders. The birds are very susceptible to diseases and show poor response to immunization. Feed premixes are the main dietary source of vitamin A for the first 8-10 weeks of life, when the chickens are kept indoors. Thereafter, they are allowed to scavenge, whereby carotenoids may be available as a potential vitamin A source.

The purpose of the present study was to investigate if the natural content of vitamin A and β-carotene in the feed ingredients could sustain performance and health of the chickens or if a supply equivalent to the level (1500 IU/kg feed) recommended by NRC (1994) would improve vitamin A status, performance and disease resistance in chickens of the Bangladeshi local crossbreed "Sonali".

Materials and Methods

Birds and Housing

612 day-old Sonali birds (crossbreed of ♂ Rhode Island Red x ♀ Fayoumi) of both genders were collected from the Government poultry farm. Twelve day-old birds were randomly selected and killed to determine vitamin A status in blood and liver. The remaining 600 chickens were divided equally among three farms. The experiment at each farm was carried out in two replicates each with two groups of animals. The birds were reared by the farmers for 70 days. On day 4 of age, the birds were weighed within the group, and each bird was tagged. From day 7 and thereafter the birds were weighed individually every week. Weekly feed consumption was registered for each group from day 8 (week 2) to day 70 (week 10) for calculation of feed intake and feed conversion ratio (FCR). The birds were vaccinated against Newcastle disease (ND) at days 3, 24 and 60 and infectious bursal disease (IBD) at days 14 and 21. Dead birds were collected and post mortem examination was performed for assessment of possible cause of death.

The farmers were selected by the Upazilla co-ordinator of Muktagacha working area. These farmers were working with Proshika and the Participatory Livestock Development Project, managed by Directorate of Livestock Services, Bangladesh. The farmers were selected on the basis of working schedule of the project and their willingness to participate in research work.

Each farmer had a poultry shed (3.0 x 4.6 m) made of bamboo and grass. The floor was made of split bamboo, and each shed was divided into four equal sized rooms. Rice husk, collected from local auto rice mills and sun dried, was used as litter.

Diet

A basal ration was prepared based on ingredients produced locally. After collection of the ingredients they were analysed at the Danish Institute of Agricultural Sciences (DIAS), Department of Animal Nutrition and Physiology ( Table 1).

A premix containing all minerals and vitamins, but without Vitamin A was obtained from a local Bangladeshi Producer (Rampart Power, Bangladesh). Vitamin A was added separately as retinol acetate (Rovimix A 500^TM type P, Hoffmann La Roche, Basel, Switzerland). Two levels of vitamin A were used for this experiment.

The control group received 500 IU/kg diet of vitamin A from day 42 until the end of the experiment and the VitA group received 1500 IU Vitamin A/kg of feed from the beginning to the end of the experiment. At the beginning of the experiment, the control group received the basal diet without added vitamin A, because soybean meal was supposed to contain 1282 IU vitamin A/kg of feed according to Narahari (1997). However, at day 43 of the experiment severe mortality and morbidity of the control group was observed, and 500 IU of vitamin A were therefore supplied orally with a plastic dropper for 5 days followed by addition of 500 IU Vitamin A/kg feed until the end of the experiment. Vitamin A was added to the feed in a two-step mixing procedure. Diets were thoroughly mixed by hand by experienced technicians. No coccidiostat was added to the diets The birds had free access to feed and water. The experimental diets were used from the second day of the experiment. On the first day, coarsely ground wheat was provided as feed.

Sampling

Initially, 12 birds were sacrificed for sampling of blood and liver for determination of vitamin A. A second sampling was done at day 43 when nearly all the control group birds showed morbidity and mortality. Two birds were selected randomly from the control group of each farm. At the end of the experiment, 4 birds both from the control group and the VitA group of each farm were selected randomly for sampling. They were decapitated, blood was collected in heparinised tubes, kept on ice just after collection, and centrifuged at 1500 rpm for separation of plasma. The liver was obtained, weighed, and kept in polyethylene bags, and plasma and liver were stored at -20 °C. The samples were transported to DIAS in Denmark, in a cool box with ice for analysis of vitamin A.

Analyses

Retinol and β-carotene were determined in liver and plasma after these materials had been cleaned up, saponified, and extracted with heptane modified with 2-propanol and quantified by using HPLC, with a 4.0 x 125 mm Perkin Elmer HS-5-Silica column. Fluorescence detection was performed with an excitation wavelength of 344 nm and emission wavelength of 472 nm. Details of the procedures are described by Jensen et al (1998). For analysis of dietary retinol and β-carotene, 40 g feed were suspended in 250 ml 2-propanol and stirred in the dark over night. Aliquots were treated as described for plasma and liver.

Dietary dry matter (DM) and crude ash were determined according to AOAC (1990). Nitrogen was measured by the Kjeldahl method using a Kjell-foss 16200 autoanalyzer (Foss Electric, Hillerød, Denmark) and protein was calculated as N x 6.25. Fat was extracted with diethyl ether after hydrochloric acid hydrolysis (Stoldt 1952). Crude fibre was determined by the Weende method (Tecator, Höganas, Sweden). Nitrogen free extracts (NFE) were calculated as the difference between DM and the sum of crude protein, crude fat, crude ash and crude fibre. Calcium was determined by atomic absorption spectrophotometry as described by Milner and Whiteside (1981) and King (1984). Phophorus was determined by the ammonium molybdovanadate method according to Stuffins (1967). Apparent nitrogen-corrected metabolisable energy (AMEn) was calculated on the basis of the analysed values according to WPSA (1989).

Statistical analysis

Statistical analysis of weight gain, feed intake and feed conversion ratio was performed by means of analysis of variance using the General Linear Models procedure in the statistical package of SAS^(R) (1989). The model included treatments (a) and farm (f) as class variables, and were treated according to the following model:

Y_afi = m + a_a + b_f + e_afi, e_afi = N(0,s²), where Y_afi = the i^th observation in the a^th diet and f^th farmer, and m is the general mean, a_a and b_f  refer to the effect of dietary vitamin A and of farm, respectively, and e_afi is the residual effect. Results of the statistical analysis are presented for the effect of vitamin A treatment, being the main purpose of the present study.

Results

Clinical findings, disease frequency and mortality

At the beginning of the experiment, no vitamin A was added to the diet of the control group, because the purpose of the planned experiment was to observe if the birds could thrive only on natural feedstuffs. However, the birds of this group showed deficiency symptoms from the middle of the third week in Farm1, at the beginning of week 5 in Farm 2 and at the end of week 6 in Farm 3. Initial symptoms were weight loss and inappetance, almost no feed in the crop, followed by ataxia, and some birds were unable to walk.

Figure 1.  Vitamin A deficient chickens showing crooked tails

One feature was very common: swollen watery and white caseous mass in the eyes, and the tail was crooked to one side (Figure 1). Post mortem lesions were: swollen kidneys with white urate-like substances, excessive bile in the gall bladder, and epithelial metaplasia in the oesophagus and oropharynx when the birds survived for a few days after showing deficiency symptoms. Some apparently healthy birds died suddenly without showing any deficiency symptoms.

Mortality was quite high in Farms 1 and 2, but relatively low in Farm 3 (Table 2). Overall mortality after 10 weeks was 51.0% in the control group and 5.7% in the VitA group. Mortality was high in weeks 5, 6, and 7. Because of the severity of deficiency symptoms and high mortality, it was necessary to provide water-soluble vitamin A (retinol acetate) to the control group orally once a day (500 IU vitamin A/bird) for 5 days from day 43 of age. This resulted in reduction of mortality in weeks 8, 9, 10. From day 48 to the end of the experiment (day 70), 500 IU vitamin A (retinol acetate)/kg feed was supplemented to the control group. Chickens in Farm 1 showed yolk sac infection during the first week of the experiment. Because of heavy rain at the beginning of the experiment, it was difficult to maintain proper brooding conditions. At the end of week 3 (day 22) Farm 2 was infested with coccidiosis. The birds were treated with sulfaclozine sodium, 2.5 g/litre of water, the control group for 9 days and the VitA group for 3 days. The possible causes of mortality were evaluated at post mortem examination during the experiment. In the control group a total of 102 birds died from vitamin A deficiency, while no birds died from vitamin A deficiency in the VitA group. Coccidiosis was more profound in the dead birds of the control group than of the VitA group (32 vs 7) as was also yolk sac infection (13 vs 3). The cause of the death of the rest of the birds was not identified.

Performance

The average weekly feed intake did not differ between the two groups (Table 3). The variation within groups expressed as the coefficient of variation, (CV = 100*(SD/Mean) was, however, quite high ranging from 13.0 to 40.6% in the control group and from 4.7 to 48.6% in the VitA group. The total feed intake over the total experimental period of 10 weeks was 3568 ± 440 g (Mean ± SD) in the control group and 3546 ± 259 g in the VitA group. Body weight gain was significantly higher during weeks 4, 5 and 6 for the VitA group than for the control group. The final average body weight was 739 ± 56 g (Mean ± SD) for the control group and 873 ± 68 g for the VitA group corresponding to an average daily body weight gain of 10.6 g and 12.5 g, respectively,  for the whole experimental period.

Feed conversion ratio (FCR) was significantly lower during week 4 and 6 for the VitA group than for the control group. The average feed conversion ratio (Mean ± SD) during week 1-10 was 3.84 ± 0.39 g feed/g gain in weight for the control group and 3.27 ± 0.40 g/g for the VitA group (P < 0.01)

Vitamin A levels of diets, blood plasma and liver

Analysis of the diets after 20 weeks of storage revealed that the content of vitamin A and β-carotene in the feed had decreased to undetectable levels in both cases probably due to oxidation. The retinol concentration (Mean ± SD) of plasma and liver of the day-old birds was on average 0.18 ± 0.06 µg/ml plasma (range 0.12-0.35 µg/ml) and 5.27 ± 2.21 µg/g liver (range 0.75-8.2 µg/g).

At 43 days of age, the vitamin A level of the control group was below detection limit (0.01 µg/g). No vitamin A was detected in liver and plasma in ten birds out of twelve. One bird had 13.2 µg/g of liver and 0.03 µg/ml of plasma; another bird had 0.45 µg/g of liver but no detectable vitamin A in plasma. Finally, after receiving the soluble vitamin A orally and with 500 IU/kg feed (the control group), the retinol concentration in the liver increased to an average of 120 ± 83 µg/g (range 29-255 µg/g), and in the plasma to 0.30 ± 0.06 µg/ml (range 0.22-0.35 µg/ml). The retinol concentration in the VitA group was on average 17.7 ± 8.4 µg/g liver (range 7.2-29.5 µg/g and 0.28 ± 0.08 µg/ml plasma (range 0.16-0.39 µg/ml) at day 70.

Discussion

The results of the present experiment show that the birds could not maintain normal performance on the natural content of vitamin A in the feedstuffs, although the inclusion of 22% yellow soybean meal should provide 1282 IU of vitamin A/kg of diet according to Narahari (1997). The birds suffered from vitamin A deficiency at 18 days of age in Farm 1, at 30 days of age in Farm 2 and at 39 days of age in Farm 3. The feed ingredients did not contain vitamin A at the time of analysis, and the vitamin A level in blood plasma and liver declined below detection limit. The natural vitamin A content may have been destroyed during storage of the feed. Previous results have shown that day-old chickens, derived from hens with an adequate intake of vitamin A, but receiving a diet completely devoid of vitamin A, had marginally deficient levels of vitamin A from the third week (Nockels et al 1984) and showed signs of deficiency from the sixth week (Scott et al 1982; Nockels et al 1984). Turkey poults fed a diet without added vitamin A died between 18 and 22 days after hatching (Sklan et al 1995). Liver vitamin A level declined significantly after consumption of a low vitamin A diet for 3 weeks and were depleted after 5 weeks (Aye et al 2000). Similar findings were observed in the progeny of normal hens fed a diet without vitamin A (West et al 1992). We observed that early infection like coccidiosis, yolk sac infection or poor brooding condition may affect the onset of vitamin A deficiency symptoms. Coccidial infection may have disrupted the intestinal epithelium. Injury to the intestinal wall by the coccidia may therefore result in an impaired conversion of  β-carotene to vitamin A or in an impairment of absorption of vitamin A per se, or both, which would, in either case, result in a decreased liver storage of vitamin A (Erasmaus et al 1960). Interference with the absorption of the vitamin or conversion of the pro-vitamin would increase the requirement of vitamin A. On the other hand, vitamin A exerts a specific action on the formation, maintenance and regeneration of epithelial tissues which may also increase the dietary demand of vitamin A (Erasmaus et al 1960).

Yolk sac infection and poor brooding condition might exert an increasing utilisation of the limited store of vitamin A in the yolk sac during early weeks of life. Factors like disease, gastrointestinal parasites, environmental stress due to temperature and/or humidity may increase the requirement of vitamin A, which is a relatively unstable vitamin under tropical conditions (Christensen 1983). The initial symptoms of vitamin A deficiency were weight loss and inappetance resulting in virtual absence of feed in the crop, which was followed by incoordination, imbalanced posture and gait, and ruffled feathers. Swollen watery eyes with whitish exudate were very common in the present study, and this is in accordance with earlier reports on early signs of vitamin A deficiency (Olson 1991; Roche 1976). Some apparently healthy birds died suddenly without showing any symptoms, which is in accordance with the findings reported by West et al (1992) and Sebrell and Harris (1967). It might be the result of a lower resistance to infections (Scrimshaw et al 1968; Sporn et al 1984; Underwood 1984; Biesel 1988). Post mortem lesions were swollen kidneys with white urate-like substances. Epithelial metaplasia was found in the oesophagus and oropharynx when the birds survived for few days after showing deficiency symptoms. Similar lesions were reported in turkeys and chickens fed a vitamin A deficient diet (Aye et al 2000; West et al 1992). It was reported that the typical clinical symptoms only occur when the deficit persists (Biesalski and Seelert 1989; Beynen et al 1989). Excessive storage of bile in the gall bladder in deficient birds also may be an important post mortem sign of vitamin A deficiency. Crooked tail (chicken tail is curved to one side) was very common in the deficient group. Beynen et al (1989) reported that feather distribution may be a parameter of vitamin A deficiency, but to our knowledge crooked tail in vitamin A deficient chickens has not been reported before.

In the present studies it was observed that mortality and morbidity were much higher, and the course of the disease was prolonged when vitamin A deficient chickens were affected with coccidiosis, and the response to the anti-coccidial drug was also very poor. These findings confirm that dietary inadequacy of vitamin A apparently increases susceptibility and severity of coccidiosis in chickens (Wilgus 1977). Birds on low vitamin A diets were not able to respond to this antigenic stimuli as quickly as birds on normal vitamin A rations, and consequently, the coccidial life cycle progressed for a longer period of time (Coles et al 1970). However, the reason for a poor response to this treatment is not clear.

It was observed that oral supplementation of water-soluble vitamin A to the severely deficient birds reduced the mortality within 2-3 days, and feed intake became normal within one week. However, a milder form of ataxic symptoms persisted in some cases even after three weeks. It has been reported that squamous metaplasia arising from vitamin A deficiency is reversed through intake of vitamin A (Biesalski and Seelert 1989).

Feed utilisation was not significantly different between the groups except in week 4 and 6. This difference in week 4 might be attributed to the onset of deficiency symptoms in Farm 2. During week 5, Farm 1 and 2 were already affected with deficiency, so the birds of the control group consumed less feed than those of the VitA group. Overall body weight gain and feed efficiency were significantly better in the VitA group receiving 1500 IU/kg feed. These results agree with Davis and Sell (1983) who reported reduced body weight gain, and impaired growth of the Bursa of Fabricius and thymus in chicks fed a low vitamin A diet and vitamin A free diet when compared to birds raised on a supplemented vitamin A diet. Beynen et al (1989) reported 16% less average body weight for vitamin A deficient chickens than for those with adequate vitamin A. This result differed from what was found by Lessard et al (1997) who reported similar body weight gains of the birds maintained on 400 IU Vitamin A/kg feed and 1500 IU vitamin A/kg feed respectively. Friedman and Sklan (1989) reported that the body weights between two groups of chickens receiving either normal (1mg retinol equivalent/kg feed) or vitamin A depleted diets (no added vitamin A) were not significantly different which may be due to the short period of experiment (40 days). Uni et al. (2000) reported that the absence of vitamin A interferes with the normal growth rate in chickens because it influences the functionality of the small intestine by altering proliferation and maturation of cells in the small intestinal mucosa. Overall mortality in the present study was very high in the control group (51%) compared to the VitA group (5.7%). Beynen et al (1989) also reported higher incidence of mortality in vitamin A deficient chickens compared to adequately supplemented chickens. West et al (1992) stated that vitamin A deficiency is sometimes irreversible, and together with increased sensitivity to infection can lead to death.

We observed a wide range of variation in the vitamin A concentration in the liver and blood plasma of the day-old birds, and in the control group at day 43 on the onset of severe deficiency symptoms. This may be due to the variation in vitamin A in the egg yolk from the hen. West et al (1992) found that some birds were deficient after two weeks, while others had normal values after six weeks when fed vitamin A deficient diets. As we treated the birds of the control group with oral application of vitamin A, relatively higher concentrations of vitamin A were found in liver and plasma than in the birds of the VitA group at the end of the experiment. The concentration of retinol in plasma was, however, similar in the two groups. In accordance with several findings in the literature (Christensen et al 1978; Biesalski and Seelert 1989) plasma vitamin A concentrations are not indicative of vitamin A status of the body as long as liver stores of vitamin A are present.

No vitamin A was found in the diets after 20 weeks of storage. Vitamin A might have been oxidised during the storage of the feed samples. Reduced vitamin A levels in diets after storage have been described earlier by Fullerton et al (1982) and West et al (1992).

It appears that one cannot rely on feed tables as to the content of vitamin A. The natural content of vitamin A is not stable under humid and warm conditions as occur in a tropical climate. The synthetic vitamin A ester is probably also not stable for as long time as indicated on the batch and certainly not 2 to 3 years as claimed by the premix producer. Vitamin A ester must be added to the feed. In the present studies addition of 1500 IU vitamin A (retinol acetate) might have been marginal. However, more studies are needed to assess the stability of vitamin A in natural feedstuffs and the requirement of vitamin A in chickens under Bangladeshi production conditions.

Acknowledgements

This project was funded by the Danish International Development Agency (DANIDA) through the Network for Smallholder Poultry Development, Royal Veterinary and Agricultural University, Copenhagen, Denmark. The authors wish to thank Dr. Søren Krogh Jensen and Dr. Henry Jørgensen, Department of Animal Nutrition and Physiology, DIAS, for fruitful discussions, and the lab technicians Elsebeth Lyng Pedersen, Anna Stouby and Asnakech Tadesse Borgi for their excellent technical assistance. The authors are very grateful to IT-technician Ole Hartvig Olsen for his help to perform the statistical analysis of the data.

References

AOAC 1990 Official Methods of Analysis. 15^th ed. Association of Official Analytical Chemists, Washington DC.

Aye P P, Morishita Y T, Saif Y M, Latshaw D J, Harr S B and Cihla B F 2000 Induction of vitamin A deficiency in Turkeys. Avian diseases, 44: 809-817.

Barua B A and Olsen A J 2000 Vitamin A and Carotenoids, in: Modern Chromatographic Analysis of Vitamins. 3rd ed., pp 1-73.

Beynen A C, Sijtsma R S, Kiepurski A K, West C E, Baumans V, Herck Van H, Stafleu R F and Van Tintelen G 1989 Objective clinical examination of poultry as illustrated by the comparison of chickens with different vitamin A status. Laboratory Animals, 23: 307-312.

Biesalski K H and Seelert K 1989 Vitamin A deficiency. New knowledge on diagnosis, consequences and therapy. Ernährungswissenschaft, 28: 3-16.

Biesel W R 1988 Use of animals for study of relations between nutrition and infectious diseases. In: A.C. Beynen and C.E. West, (Eds), Comparative Animal Nutrition. Vol. 60, Use of Animal Models for Research in Human Nutrition, Basle, Karger, pp 33-35.

Christensen K 1983 The pools of Cellular Nutrients: The Vitamins. In: P.M. Riis (Ed.), Dynamic of Biochemistry of Animal Production, Elsevier, pp 215-273.

Christensen K, Daugaard J, Henneberg U, Hjarde W, Jensen K and Aalund O 1978 Vitamin A and Beta-carotene to cattle. Evaluation of requirements under Danish conditions. Beretning fra Statens Husdyrbrugsforsøg, no. 470. Landhusholdningsselskabets Forlag, Copenhagen (in Danish with English summary), 77 pp.

Coles B, Biely J and March B E 1970 Vitamin A deficiency and Eimeria Acervulina infection in chick. Journal of Poultry Science, 49: 1295-1301.

DavisC Y and Sell J L 1983 Effect of all-trans retinol and retinoic acid nutriture on the immune system of chicks. Journal of Nutrition, 113: 1914-1919.

Erasmaus J and Scott M L and Levine P P 1960 A relationship between coccidiosis and vitamin A nutrition in chickens. Poultry Science, 39: 565-571.

Friedman A and Sklan D 1989 Antigen-specific immune response impairment in the chick as influenced by dietary vitamin A. American Institute of Nutrition, 119: 790-795.

FullertonF R, Greenman D L and Kendall D 1982 Effects of storage conditions on nutritional qualities of semipurified (AIN-76) and natural ingredients (NIH-07) diets. Journal of Nutrition, 112: 567-573.

Jamroz D, Jakobsen K, Wertelecki T, and Jensen SK 2002. Utilization of dietary ß-carotene and retinol acetate by young and older geese. Acta Agriculturae Scandinavica, Section A, 52: 150-158.

Jensen S K, Jensen C, Jakobsen K, Engberg R M, Andersen J O, Lauridsen C, Sørensen P, Skibsted L H and Bertelsen G 1998 Supplementation of broiler diets with retinol acetate, ß-carotene or canthaxanthin: Effect on vitamin status and oxidative status of broilers in vivo and meat stability. Acta Agriculturae Scandinavica, Section A. Animal Science, 48: 28-37.

Johannsen A K B, Jensen S K and Jakobsen K 1998 A note on vitamin A activity of ß-carotene in broilers. Acta Agriculturae Scandinavica, Section A. Animal Science, 48: 260-263.

King RD1984 Development in food analysis techniques. Chapter 10. Atomic Absorption Spectroscopy in Food Analysis. The Leatherhead Food Research Association, Randalls Road, Leatherhead, Surrey, UK, pp 293-314.

Lessard M, Hutchings D and Cave A N 1997 Cell-mediated and humoral immune response in broiler chickens maintained on diets containing different levels of vitamin A. Poultry Science, 76: 1368-1378.

Milner B A and Whiteside P J 1981 Introduction to atomic absorption spectrophotometry. Pye Unicom Ldt., York Street, Cambridge, England CB1 2PX.

Narahari D 1997Agro forestry products in livestock and poultry feeds. TamilNadu Veterinary and Animal Sciences University, Chennai-600 007, India, pp 134-142.

Nockels C F, Ewing D L, Phetteplace H, Ritacco K A and Mero K N 1984 Hypothyrodism: an early sign of vitamin A deficiency in chickens. Journal of Nutrition, 114: 1733-1736.

NRC 1994 Nutrient Requirements of Poultry. Ninth revised edition, National Academy Press, Washington DC, USA, pp 19-33.

Olson J A 1991 Vitamin A, In: L.J. Machlin (Ed), Handbook of Vitamins, 2^nd revised. Mercel Dekker, New York, pp 1-57.

Roche 1976 Vitamin A Compendium. Hoffmann-La Roche and Co., Basle, pp 29-32.

Roche 1994 Vitamin nutrition for ruminants. Hoffmann-La Roche and Co., Basle, pp 1-41.

SAS ^(R).1989 SAS/STAT ^(R) User's guide version, 6. Fourth edition. Volume 1 and 2. Cary NC: SAS^(R) Institute INC, 1848 pp.

Scrimshaw N S, Taylor C E and Gordon J E 1968 Interaction of Nutrition and Infection. World Health Organization Monograph Series no. 57. Geneva: WHO.

Scott M L, Nesheim M C and Young R J 1982 Nutrition of the chickens, 3rd editions Ithaca, USA, pp 34-56.

Sebrell W H Jr and Harris R S 1967 The vitamins, 2nd edition, vol 1, New York, Academic press, pp 232-235.

Sklan D, Melamed D and Friedman A 1995 The effect of varying dietary concentration of vitamin A on immune response in turkey. British Poultry Science, 36: 385-392.

Sporn M B, Roberts A B and Goodman D S 1984 The retinoids, vols 1 and 2. Orlando, Academic Press.

Stoldt W 1952 Vorschlag zur Vereinheitlichung der Fettbestimmung in Lebensmitteln (Suggestions to standardise the determination of fat in foodstuffs). Fette und Seifen, 54: 206-207.

Stuffins C B 1967 The determination of Phosphate and Calcium in feeding stuffs. Analyst, 92: 107-111.

Underwood B A 1984 Vitamin A in animal and human nutrition. In the Retinoids, vol.1. Orlando, Academic Press, pp 282-392.

Uni Z, Zaiger G, Gal-Garber O, Pines M, Rozenboim I and Reifen R 2000 Vitamin A deficiency interferes with proliferation and maturation of cells in the chickens small intestine. British Poultry Science, 41: 410-415.

West C E, Sijtsma S R, Peters H P F, Rombout J H W M and Van Der Zijpp A J 1992 Production of chickens with marginal vitamin A deficiency. British Journal of Nutrition, 68: 283-291.

Wilgus H S 1977 Disease, Nutrition-interaction. Poultry Science, 59: 772-781.

WPSA 1989 European Table of Energy Values for Poultry Feedstuff, 3^rd edition, Subcommittee energy of the working group no 2. Nutrition of the European Federation of branches of the World's Poultry Science Association, pp 11-28.