METABOLISM AND NUTRITION Performance and Physiological Parameters of Broiler Chickens Subjected to Fasting on the Neonatal Period E. Gonzales,*,1 N. Kondo,* É. S. P. B. Saldanha, M. M. Loddy,* C. Careghi, and E. Decuypere *School of Veterinary Medicine and Animal Production, UNESP, São Paulo, Brazil Institute of Animal Production, São Paulo, Brazil and Catholic University of Leuven, Leuven, Belgium ABSTRACT Broiler chicks aged 12 h after hatching to the Pearson test, BW of birds at 21 and 42 d were were allotted according to a block design in a 7 2 factorial schedule of 14 treatments and four replications of 50 chicks each one. The main experimental factors were fasting for 0, 6, 12, 18, 24, 30, and 36 h after chick placement and sex. Independent of sex, fasting had a negative linear effect on weight and productivity of broilers at market age (42 d) without affecting feed conversion or mortality index. Groups subjected to 18 and 36 h of fasting after placement, corresponding to 30 and 48 h posthatching fasting, had lower biometrical values for small intestine (length, weight, and size; villus height; and crypt depth) than chicks fed immediately after placement. According significantly correlated to BW at 7 d (r = 0.77) and 21 d (r = 0.45), respectively. Males performed better than females but had higher mortality rates. Fasting did not influence serum concentrations of corticosterone or sexual steroid hormones. Nevertheless, early signs of sexual dimorphism arose from the high estradiol (E 2 ) concentration on female serum. Heterophil:lymphocyte ratio was not different among treatments, indicating that early fasting did not seem to be a stress factor 21 or 42 d after fasting. The results suggested a maximum fasting of 24 h after hatching in order to preserve broiler productivity at market age. (Key words: chick, fasting, neonatal, performance, physiology) 2003 Poultry Science 82:1250 1256 INTRODUCTION Continuous genetic selection of broilers for fast growth has resulted in reducing the rearing period necessary for broilers to reach the same live weight. Because of this continuous decline, 1 d of fasting corresponds to an even increasing period in the lifetime of a chick. Chick growth before exogenous feeding depends on the nutritional elements absorbed from the residual yolk sac. This yolk, enclosed in the abdominal cavity, is the first nutrient source for the hatchlings. Early feeding after hatch, compared to delayed feeding, appears to stimulate yolk utilization (Noy and Sklan, 1998a; Speake et al., 1998) but also more importantly the development of the gastrointestinal system. This produces an initial enhancement in BW (Noy and Sklan, 1999b; Sklan and Noy, 2000). Early in development, the relative growth rate is the highest due to the marked increase in the weight of the gastrointestinal tract (Nitsan et al., 1991), which is much higher as compared to that of the rest of the body (Nir et al., 1993) 2003 Poultry Science Association, Inc. Received for publication December 2, 2002. Accepted for publication March 24, 2003. 1 To whom correspondence should be addressed: elisa.gonzales@ uol.com.br. A delay in the placement of chicks may cause productive losses later on in the life of birds from strains selected for rapid growth (Gonzales et al., 1999, 2000), indicating that nutrient supply via yolk sac is not sufficient to sustain the extreme growth of the broiler chick after hatching. The yolk also contains valuable maternal antibodies that are better used for passive immunity than as a source of amino acids, which is what happens when no exogenous feed is given (Dibner, 1999). Also in hatchlings, development of the gastrointestinal tract and maturation of the secretion of digestive enzymes are impaired when feed is restricted (Noy and Sklan, 1999a; Sklan and Noy, 2000). However, even in males, weights of 42- or 49-d-old birds subjected to some type of feed restriction early posthatch rarely equals those of birds fed ad libitum. Variation in environmental temperature, nutritional levels employed, amount of feed intake in the period after feed restriction, genetic line, sex, and severity of restriction are some of the causes for the observed differences (Yu and Robinson, 1992). Recently, more attention has been given to the effect of the time of feeding on the performance of chicks, but the physiological basis remains to be elucidated (Noy and Pinchasov, 1993; Pinchasov and Noy, 1993; Noy and Sklan, 1998b; Sklan and Noy, 2000; Bigot et al, 2001). Abbreviation Key: B = corticosterone; E 2 = estradiol; T = testosterone. 1250
FASTING OF NEONATAL BROILER CHICKS 1251 MATERIALS AND METHODS Bird Management and Experimental Design We used a total of 3,933 chicks derived from eggs with an average weight of 66.2 g produced by the same flock of Avian Farms broiler breeders of 52 wk of age in this experiment. Chicks were removed from the hatchers at 0430 h. Still in the hatchery, and after 2 h, they were winged-sexed, and vaccinated for Marek s and Gumboro diseases. The chicks left the hatchery at 0800 h, arrived at the farm at 1200 h, and were all placed at 1600 h of the same day. On the farm, chicks were assigned to pens according to a randomized block design, with a factorial arrangement of 2 (two sexes) 7 (seven feeding schedules: fasting for 0, 6, 12, 18, 24, 30, and 36 h after placement) and 4 replicates (blocks) of 70 to 72 birds each, with a total of 14 treatments and 56 experimental units. A control group received feed and water ad libitum 30 min after placement, 12 h after the chicks were removed from the hatchers. The other groups received water ad libitum but were fed only 6, 12, 18, 24, 30, or 36 h after placement, corresponding to 18, 24, 30, 36, 42, or 48 h posthatching. After this period, all birds were fed ad libitum until 42 d of age. Birds were reared in a broiler experimental facility built to house up to 75 birds per pen (12 birds/m 2 ). During the total rearing period (42 d), birds received 23 h of light (23L:1D). At 10 d of age, birds were vaccinated via drinking water for Newcastle disease, LaSota strain. Feeding was divided into two phases: starter (1 to 21 d of age) and grower (22 to 42 d of age). Feeds were in meal form, formulated to supply the nutritional specifications of broilers (Gonzales et al., 1998), containing 3,050 and 3,150 kcal/kg AME and 21.0 and 19.0% CP in the starter (1 to 21 d) and grower (22 to 42 d) phases, respectively. Performance and Mortality Mortality was checked daily, and date of death and BW recorded. In addition, the following data were collected and calculated on d 1, 7, and 42 of the experimental period: average weight, average weight gain, average feed intake, feed conversion, and mortality percentage. The production index [PI = (daily weight gain, g livability, %)/(feed conversion 100] was calculated until 42 d of age. Ten birds per replicate of each experimental group were identified by a numbered wing band fit on placement day. These birds were individually weighed at placement and at 7, 21, and 42 d of age. The obtained data were used to evaluate the CV and the correlation among the weights obtained during the indicated periods. 2 Video Plan, Kontron Elektronik, Carl Zeiss Vision, Göttingen, Germany. Yolk Sac Utilization Immediately before placement, 28 male and 28 female chicks were weighed and euthanized by acute blunt cranial trauma. At 6, 12, 18, 24, 30, 36, 48, and 72 h after placement, two birds per pen were also weighed and euthanized by the same method to measure yolk residue, if present. Differential Leukocyte Counting Smears at d 21 and 42 of the experiment for each treatment group were made without an anticlotting agent, covered with methanol for 5 min, washed with distilled water, covered with Wright-modified Rosenfeld dye for 10 min, and then washed with distilled water. The Wright-modified Rosenfeld dye was produced by solubilizing 1 g of May-Grumwald, 2 g of Giemsa, and 1 g of Wright in 1 L of methanol through mixing and filtering and was stored in a dark flask. Cell counts (heterophils and lymphocytes) on each slide were performed by counting 100 leukocytes under common optical microscope. Small Intestine Evaluation Before placement (0 h) and after fasting periods of 18 and 36 h, four birds per treatment (nonfasting and fasting birds; male or female) were euthanized for small intestine weight, small intestine weight relative to bird live weight, and small intestine length measurements. Sections of these samples of the small intestine (duodenum, jejunum, and ileum) were collected to measure villus height and crypt depth of the mucosa with the aid of light microscopy. Each small intestine section was washed with phosphate buffer solution (0.1 M, ph 7.4), open through the mesenteric edge, and extended by the serosa. Then, the sections were fixed in Bouin solution for 24 h, washed in ethanol at 70% Gay-Lussac, dehydrated in a rising series of alcohol, diaphinized in xylol, and finally processed by the paraffin standard method. Samples were cut into 5 µm thick sections with a microtome. Section were fixed on slides and dyed according to Harris hemotoxylin-eosin (HE) technique. In each slide, 30 villus height and crypt depth measurements were made using an image analyzer system, 2 with 230 amplification. Hormone Determinations Blood samples were collected directly from hearts of chicks used to evaluate yolk sack residue at placement and 36 h after placement. At the end of d 21 and 42, blood samples were collected by wing vein puncture of two birds per pen. Blood samples were centrifuged (1,000 rpm, 10 min), plasma was removed, and the samples were stored at minus 20 C for later determinations of testosterone (T), estradiol (E 2 ), and corticosterone (B) hormones in order to follow stress susceptibility of delayed feeding and its possible interference with precocial or
1252 GONZALES ET AL. delayed dimorphism interfering with broiler productivity. Sexual steroid hormones were determined by RIA using a commercial kit. 3 Plasma quotas (100 µl) were extracted with di-chlor-methane (900 µl), and the dried residue was processed for corticosterone RIA. 4 Statistical Analysis Statistical analyses of performance, mortality, production index, differential blood cell counts, and plasma levels of hormones data followed a randomized block design, with a factorial arrangement of 2 7, taking into consideration sex (male or female) and feed fasting (fasting for 0, 6, 12, 18, 24, 30, and 36 h) and an equal number of replicates. Within the feed fasting factor, the decomposition of degrees of freedom of the treatments in orthogonal polynomials up to the fifth degree was analyzed. Statistical evaluation of yolk residue retraction was made considering an unequal number of replicates at 0, 6, 12, 18, 24, and 30 experimental h of the different groups at the time of evaluation. Therefore, at 6 h, there were only four experimental groups the nonfasted controls and the 6-h fasted birds, male and female. At 12 h, there were six experimental groups, and so on, until 36 h, when the 14 experimental groups were under test. Only the sex factor was considered for the chicks before placement (experimental d 0). The microscopic measurements of small intestine sections (18 or 36 h posthatching) were statistically analyzed in a completely randomized design, with a factorial arrangement of 2 2 (control = 0 h vs. fasting = 18 or 36 h; male vs. female) and four replicates. When necessary, differences among means were evaluated by Tukey s test (P < 0.05). The correlation among individual weights at 0, 7, 21, and 42 d was tested by the coefficient of Pearson, according to Cochran and Cox (1957). Statistical tests were performed with the aid of the general procedure of the SAEG software (SAEG, 1999). RESULTS Performance and Mortality At exit from the hatchery, 330 h after chicks were removed from the hatchers, males weighed 49.4 g and females 48.9 g. Twelve hours after removal from the hatchers, considered as 0 h of placement, males weighed 48.1 ± 0.13 g, and females weighed 47.8 ± 0.13 g. Therefore, the loss of BW between exit from the hatchery and placement at the farm was 2.6 and 2.3% for males and females, respectively. Live weights of all chicks at placement were 47.96 ± 0.09 g, which was homogenous and compatible to breeder age. 3 Byk-Sangtec Diagnostica GMbH, Dietzenbach, Germany. 4 IDS Ltd., Boldon, England. Table 1 shows live weight at 7 and 42 d and feed intake, feed conversion, mortality, and production index of birds at 42 d of age. There was no interaction (P 0.05) between the factors fasting and sex for any of the productive traits during the studied periods. Thus, the results are presented as a function of the main effects (fasting and sex). Independent of sex, fasting at the beginning of the rearing period negatively influenced performance traits evaluated at 7 and 42 d of age, except for feed conversion (Table 1). At 7 d of age, there was a nonlinear trend (P 0.05) of birds to present a lower weight as the fasting period increased (Y pm = 143.4 0.345 x 0.0098 x 2 ;R 2 = 0.96). The comparison of treatment means showed that live weight at 7 d of age was not influenced when fasting had a maximum duration of 12 h. However, at 42 d of age, there was a linear and negative relationship between fasting period and weight (Y 42 = 2077 2.671 x; R 2 = 0.94). The discrimination of means by the test of Tukey showed that fasting longer than 18 h significantly impaired the final weight of the broilers. The production index was linearly affected as the fasting period increased (Y = 237.7 0.4013 x; R 2 = 0.91). Feed conversion and bird mortality rates were not influenced by the used fasting periods. Taking into consideration the effect of sex, males showed better performance than females, with higher weights at the end of 7 and 42 d of rearing and lower feed conversion but with higher mortality (Table 1). By means of the Pearson rank correlation, it was possible to demonstrate that live weight at 21 d of age was positively correlated (P < 0.05) with live weight at 7 d of age (r = 0.77), and the same occurred with weight at 42 d of age relative to weight at 21 d (r = 0.45). Weights at 7, 21, or 42 d of age were not correlated (P > 0.05) to chick weight at placement (d 0). Time of feeding after hatching did not influence the coefficient of variation of weights (data not shown) in any of the analyzed periods (7, 21, and 42 d), indicating that this procedure did not impair flock uniformity. However, the coefficient of variation of weights was higher for males than for females at 7 and 21 d. Leukocyte Differential Count The heterophil:lymphocyte ratio changed from 1.25 immediately after hatch, to 0.25 and 0.55, respectively, at d 21 and 42 of age but was not significantly affected by fasting or sex in any of the evaluated periods and is therefore not presented. Small Intestine Evaluations The results of the analysis of variance of villus height and crypt depth, and absolute/relative weights of intestines, and lengths of intestines of chicks subjected to 18 and 36 h of fasting after allotment and compared to fed animals are presented in Tables 2 and 3. Villus height at the duodenum was affected by fasting and by sex in the evaluations made at 18 and 36 h, with an interaction between factors in the first evaluation (18
FASTING OF NEONATAL BROILER CHICKS 1253 TABLE 1. Body weight at 7 and 42 d, and feed intake (FI), feed conversion (FC), mortality, and production index (PI) at 42 d of male and females broilers subjected to neonatal fasting Body weight, g Mortality, 42 d FI, g FC PI 3 Fast 1 h 7 d 42 d 42 d 42 d % Arsin 2 42 d 0 142 aq (100.0) 4 2,065 al (100.0) 3,877 al 1.93 4.36 0.308 237 L 6 142 a (100.0) 2,069 a (100.2) 3,878 a 1.94 4.39 0.306 237 12 140 a (98.6) 2,056 ab (99.6) 3,851 a 1.94 5.73 0.328 233 18 132 b (93.0) 2,019 bc (97.8) 3,781 b 1.94 5.77 0.330 227 24 127 c (89.4) 2,019 bc (97.8) 3,794 b 1.95 5.29 0.324 228 30 126 c (88.7) 1,999 c (96.8) 3,740 b 1.94 5.08 0.321 227 36 118 d (83.1) 1,975 c (95.6) 3,664 c 1.93 6.23 0.340 223 Sex Male 138 a 2,114 a 3,901 a 1.91 b 6.39 0.340 a 241 a Female 127 b 1,944 b 3,695 b 1.96 a 4.13 0.305 b 220 b Mean 133 2,029 3,798 1.94 0.322 231 ± SEM 1.5 13.3 18.6 0.01 0.01 2.00 a d Means followed by different letters within the same column are statistically different (P < 0.05). 2 Arcsin transformations. 3 Production index = (daily weight gain, g livability, %)/(feed conversion 10). 4 The numbers in brackets are relative index to control groups (0 h of fasting). L Linear effect (P < 0.05). Q Square effect (P < 0.05). h). Villus height was higher in males than females in the control birds, 0 h fasting after allotment. However, the largest difference was a decrease in villus height at the duodenum of birds subjected to fasting, as compared to the control birds, male or female. At the jejunum, the interaction between fasting and sex was also significant at 18 h. Within the control group, males had higher villus height than females. However, in birds subjected to fast- TABLE 2. Villus height and crypt depth of small intestine of male and female chick broilers subjected to 18 and 36 h of fasting after allotment Villus height, mm Crypt depth, mm Intestine Fast h 1 Fast and sex 0 h 18 h Mean 0 h 18 h Mean Duodenum Male 421.6 Aa 129.6 B 275.5 ± 56.6 105.4 47.1 76.3 ± 14.8 Female 311.6 Ab 164.8 B 238.2 ± 29.4 62.8 42.8 52.8 ± 4.8 Mean 366.6 ± 24.9 147.1 ± 10.2 84.1 ± 12.9 B 44.9 ± 2.6 A Jejunum Male 213.7 Aa 84.1 Bb 148.9 ± 24.5 56.0 Aa 29.9 B 42.9 ± 5.3 Female 163.9 b 126.5 a 145.2 ± 12.8 31.8 b 33.5 32.7 ± 2.4 Mean 188.8 ± 12.0 105.3 ± 11.2 44.0 ± 4.9 31.7 ± 2.5 Ileum Male 185.5 116.1 150.8 ± 14.2 43.1 43.8 43.5 ± 3.1 a Female 198.2 105.0 151.6 ± 10.2 41.1 28.9 34.7 ± 3.1 b Mean 191.9 ± 4.7 A 110.5 ± 10.2 B 42.2 ± 2.7 36.0 ± 3.8 0 h 36 h Mean 0 h 36 h Mean Duodenum Male 341.2 158.1 249.6 ± 37.1 a 72.5 Aa 31.8 B 52.2 ± 8.1 Female 245.1 136.9 191.0 ± 26.1 b 38.4 b 34.2 36.3 ± 3.0 Mean 293.1 ± 25.2 A 147.5 ± 12.8 B 55.5 ± 6.9 33.0 ± 3.0 Jejunum Male 179.7 99.8 139.8 ± 21.6 41.0 27.5 34.2 ± 4.7 Female 156.3 116.1 136.2 ± 17.0 34.6 39.0 36.8 ± 3.2 Mean 168.0 ± 18.7 A 107.9 ± 12.2 B 37.8 ± 3.7 33.2 ± 4.2 Ileum Male 129.7 98.8 114.2 ± 8.9 b 26.9 b 33.8 30.4 ± 1.7 Female 186.8 123.2 155.0 ± 13.5 a 41.7 a 28.7 35.2 ± 3.5 Mean 158.2 ± 13.1 A 111.0 ± 7.0 B 34.3 ± 3.3 31.2 ± 2.3 A,B Values within the same line, within fast effect having different letters are different (P < 0.05). a,b Values within the same column, within sex effect, having different letters are different (P < 0.05).
1254 GONZALES ET AL. TABLE 3. Absolute, relative weights, and length of small intestines of male and female chick broilers subjected to 18 and 36 h of fasting after allotment Absolute weight, g Relative weight, % Length, cm Fasting 1 Male Female Mean Male Female Mean Male Female Mean 18 h 0 h 2.75 2.25 2.50 ± 0.16 a 4.91 4.59 4.75 ± 0.26 a 51.75 49.75 50.75 ± 1.19 a 18 h 1.44 1.62 1.53 ± 1.07 b 3.18 3.31 3.24 ± 0.11 b 39.50 44.38 41.94 ± 1.76 b Mean 2.09 ± 0.27 1.94 ± 0.16 4.04 ± 0.39 3.95 ± 1.29 45.63 ± 2.78 47.06 ± 1.47 36 h 0 h 3.70 4.18 3.94 ± 0.18 a 6.79 6.83 6.83 ± 0.80 a 57.38 63.75 60.56 ± 2.95 a 18 h 2.93 3.42 3.18 ± 0.13 b 5.41 5.80 5.08 ± 0.19 b 51.25 53.25 52.25 ± 0.78 b 36 h 1.64 1.61 1.63 ± 0.02 c 3.57 3.65 3.65 ± 0.43 c 50.38 48.88 49.63 ± 1.17 b Mean 2.76 ± 0.27 3.07 ± 1.34 5.26 ± 0.41 5.59 ± 1.54 53.00 ± 1.54 55.29 ± 2.42 a c Means followed by different letters within the same column are statistically different (P < 0.05). ing, the opposite was observed. Villus height at the ileum was also lower in birds subjected to fasting but was independent of sex (Table 2). At 36 h, the negative effect of fasting on villus height was observed in all sections of the small intestine, independent of sex (Table 2). Fasting also influenced crypt depth, at 18 and 36 h, with a decrease of the crypt depth at the duodenum level of fasted birds. The crypts of the jejunum were slightly affected, with a more marked negative effect in fasted males. At the ileum, no effects of fasting were observed (Table 2). Absolute and relative weights and length of intestines were depressed by fasting the chicks, and the depression was more pronounced when fasting time was 36 h (Table 3). Yolk Sac Utilization No effect of fasting or sex was observed on yolk sac utilization (Table 4). The percentage of the yolk sac in the beginning of the experiment (0 h) was 11.6 and 11.9% of BW in males and females, respectively. Hormone Values No marked effect of fasting or sex was observed on the plasma levels of B. After an initial increase during the first days, from 2.34 ± 0.18 ng/ml up to 3.54 ± 0.20 ng/ ml, an age-dependent decrease in B levels was observed and reached 1.80 ± 0.07 ng/ml at 21 d and 0.63 ± 0.03 ng/ml at 42 d. For reproductive hormones, E 2 and T (Table 5), no influence of fasting was observed during any of the analyzed periods. However, E 2 plasma levels were always higher in females than males. For T, the plasma level was only statistically higher in males at 42 d of age. Similar age-dependent changes were observed for T and E 2, after an initial decrease from 1-d-old chicks to 3 wk of age for both steroids in both sexes. The levels increased again in 6-wk-old male and female broilers, but this increase was more pronounced for E 2 in females and for T in males. DISCUSSION Previous experiments with newly hatched broiler chicks subjected to fasting for 6 to 96 h after placement showed a negative effect of fasting on the final performance at 42 d of age, when the fasting period was longer than 24 h, corresponding to 32 to 36 h after chicks were removed from the hatchers (Gonzales et al., 1999, 2000). The loss in ultimate performance due to feed deprivation in this critical stage of broiler chick development was also observed in the present experiment. It was also demonstrated that the maximum fasting period, which had no TABLE 4. Weight of viteline sac of male and female chick broilers subjected to neonatal fasting Age of evaluation Fasting 1 0h 6h 12h 18h 24h 30h 36h 48h 72h (g) 0 h 4.40 4.3 4.1 2.7 2.9 1.6 1.6 1.0 6 h 5.10 4.3 3.0 2.8 2.5 2.2 1.4 1.1 12 h 4.3 3.2 2.8 2.5 2.0 1.3 1.0 18 h 3.7 2.2 2.3 2.3 1.5 0.9 24 h 2.8 2.7 2.2 1.8 0.8 30 h 2.3 1.7 1.6 1.0 36 h 2.1 1.7 0.9 Sex Male 5.6 5.1 4.2 3.4 2.9 2.5 3.9 1.6 1.0 Female 5.7 4.9 4.4 3.8 2.8 2.5 3.7 1.5 0.9 Mean 5.6 5.0 4.3 3.6 2.8 2.5 2.0 1.6 0.9 ± SEM 0.16 0.14 0.12 0.13 0.11 0.09 0.08 0.08 0.04
FASTING OF NEONATAL BROILER CHICKS 1255 TABLE 5. Testosterone and estradiol levels (ng/ml) on broiler chicks, males and females, subjected to neonatal fasting Fasting time, h 1 Sex 2 Hormone levels, Bird ng/ml age 0 6 12 18 24 30 36 M F Mean ± SEM Number Testosterone 0 h 5.9 4.5 5.1 ± 0.52 46 36 h 6.5 8.3 4.5 8.3 4.3 5.7 2.8 6.4 5.1 5.8 ± 0.79 49 21 d 3.1 3.7 2.3 3.5 3.8 4.4 3.9 4.1 2.9 3.6 ± 0.46 45 42 d 7.0 16.2 19.4 10.7 18.6 23.4 17.8 22.0 a 11.5 b 15.8 ± 2.47 36 Estradiol 0 h 27.8 b 42.5 a 35.3 ± 2.44 55 36 h 37.5 35.6 34.2 36.5 35.9 35.2 35.9 25.7 b 46.0a 35.8 ± 1.82 56 21 d 26.5 27.6 28.3 26.7 24.8 31.9 26.2 18.5 b 36.3 a 27.4 ± 1.47 56 42 d 40.8 44.4 41.6 45.2 51.4 54.7 59.0 24.0 b 72.3 a 48.1 ± 4.0 56 a,b Means followed by different letters within the same line are statistically different (P < 0.05). 2 M = male; F = Female. significant negative effect on final weight, was 24 h after the chicks were removed from the hatchers. Considering the critical period of 18 h fasting after placement (30 h posthatching), in which a significant negative effect on bird performance could already be observed, we found that at the end of 7 d of rearing, there was a 7% lower rate of weight gain as compared to nonfasted birds, whereas at 42 d, there was 2.2% less weight gain. The highest relative growth rates between 7 and 42 d were observed in chicks fasted for 30 h posthatching as compared to their nonfasted peers (1,530 vs. 1,454%, respectively), showing that there was a physiological catch-up effort to reach the weight as determined by their genetic potential at a particular age. However, this compensatory growth capacity was limited, probably because the feed deprivation occurred too early, agreeing with the observations of Winick and Noble (1966). These authors suggested that the level of growth recovery, which occurs after a period of food deprivation, depends on the physiological state of the animal. When feed restriction is applied in the very early stages of development, cell hyperplasia is impaired. Taking into consideration the fast growth rate of modern broiler strains, particularly during the first week after hatching, it is possible to assume that this critical growth stage coincides with the period when the birds are kept with no feed, that is, the period between hatching and placement at the farm. The results of yolk sac residues of male and female chicks at the day of placement are in agreement with previous studies that showed a relative weight of the yolk sac of about 10 and 15% of the live weight (Noy and Sklan, 1999a). Similar to the results of Nitsan et al. (1991), about 84% of the yolk residue was used within 72 h of placement. However, in this study this effect was independent of the length of the posthatching fasting period. According to Noy and Sklan (1998a, 2001), the lack of feed in the very early stages of broiler chick development may cause a negative effect on initial performance due to an inadequate use of the yolk sac, which seems to be caused by low stimulation of the gastrointestinal tract. Nevertheless, in the present experiment, in which the maximum fasting period was 36 h (48 h after hatching), the retraction of the yolk sac was not affected by fasting. In a previous study (E. Gonzales, unpublished data), when chicks were subjected to a fasting period up to 96 h after placement (104 h posthatching), it was observed that yolk sac utilization was impaired by feed deprivation. Although there is an apparent discrepancy among the studies, it must be noted that there are some differences among the experimental protocols of the mentioned studies, such as total fasting period and bird strain. The fasting period was longer in the latter study than in ours, which could change the behavior of the absorption of the yolk sac. Also, when comparing that study (E. Gonzales, unpublished data) with the present, the used strains (Hubbard- Peterson and Avian Farms, respectively) presented differences in the behavior of the growth curve in the initial stage of rearing, which can influence gastrointestinal tract development and possibly yolk residue absorption. The use of yolk sac in young chicks posthatch, with or without feed, may be the result of conflicting interests (e.g., without feed, the yolk sac is the main, if not the only energy source, which should speed up its utilization). But on the other hand, the impaired development of the gastrointestinal tract, the lower metabolism and overall growth may reduce the energy needs so much that, even without feed, yolk sac utilization is lower or equal than in chicks with access to feed. This may however be strain dependent according to the initial growth curve form and length of the fasting period. Also, it is important to mention that in times of acute metabolic need muscle tissue may be catabolized to meet basic survival needs of the hatchling. Changes in the development of the small intestine are also indicated as causes of initial lower performance of broilers with possible repercussions on final performance at market age due to a decrease in small intestine weight, villus area, and crypt development, particularly at the duodenum and jejunum, when the chick is fasted after hatching (Geyra et al., 2001, Sklan, 2001). The biometric results of the small intestine obtained in the present experiment are completely compatible with these reports, with a significant effect of fasting (30 and 48 h posthatching) on small intestine weight and length and on the morpho-
1256 GONZALES ET AL. metric characteristics of the duodenum and the jejunum. Therefore, these observations confirm that broiler chicks selected for high growth rates depend on exogenous, extra-yolk-sac feeding very soon after hatching and that absence of feed in the gut in a very early posthatching period impairs development and, thus, function of the small intestine. This could be the main cause of performance depression of broiler performance at market age. Differences in absolute data of villus heights or crypt depths at 0 h between the two groups of data, where 0 h was compared with 18 or 36 h of fasting, may be related to the fact that the 0 groups (receiving food immediately after placement) corresponded to chicks that were removed from the hatchers 12 h before. Taking into account the decrease in these parameters as compared to 18 or 36 h, a slight delay (e.g., the 0 group of the 36 h evaluation) could explain why all parameters measured in one 0 group were lower then in the other 0 group (18 h). Moreover, the spread in hatching time may explain why the SE or variation in these parameters is generally more important in the 0 groups than 18 and 36 h later, certainly taking into account the small number of birds per group that were killed. Despite the negative effect on performance at market age, fasting did not cause stress as measured by B levels or heterophil:lymphocyte ratio in any of the evaluated times (21 and 42 d). Fasting did not affect plasma levels of gonadal steroid hormones (E 2, T). However, it was evident that sexual dimorphism was indicated by the higher levels of plasma estrogens in females than in males at a very early stage. This result may suggest that the differential growth of males and females was also related with E 2 levels, similar to results obtained by Kühn et al. (1996). The obtained results indicate that the maximum period that Avian Farms broiler chicks can be fasted after hatching, in order to preserve productivity at a market age of 42 d, is 24 h. The negative effect of fasting on performance traits is related to inadequate development of the gastrointestinal tract, particularly of the duodenum and jejunum, at a very early stage of the postnatal life. REFERENCES Bigot, K., S. Tesseraud, M. Taouis, M. Picard. 2001. Alimentation néonatale et développement précoce du poulet de chair. INRA Prod. Anim. 14:219 230. Cochran, W. G., and G. M. Cox. 1957. Experimental Designs. J. Wiley, New York. Dibner, J. 1999. Feeding hatchling poultry. Avoid any delay. Feed Int. 20(Dec):30 34. Geyra, A., Z. Uni, and D. Sklan. 2001. The effect of fasting at different ages on growth and tissue dynamics in the small intestine of the young chick. Br. J. Nutr. 86:53 61. Gonzales, E., J. Buyse, T. S. Takita, J. R. Sartori, and E. Deccuypere. 1998. Metabolic disturbances in male broilers of different strains. 1. Performance, mortality, and right ventricular hypertrophy. Poult. Sci. 77:1646 1653. Gonzales, E., N. S. M., Leandro, J. C. Varoli Júnior, T. S. Takita, and M. M. Loddi. 2000. O tempo de jejum do neonato e a restrição alimentar precoce influenciando a produtividade de frangos de corte na idade de abate. Revista Brasileira de Ciência Avícola. Suppl.2:4. (Abstr.) Gonzales, E., N. S. S. Mogyca, J. C. Varoli Júnior, T. S. Takita, and M. M. Loddi. 1999. O tempo de jejum do neonato afeta o desempenho do frango de corte na idade de abate. Revista Brasileira de Ciência Avícola. Suppl.1:12. Kúhn, E. R., V. M. Darras, C. Gysemans, E. Decuypere, L.R. Berghman, and J. Buyse. 1996. The use of intermittent lighting in broiler raising. Poult. Sci. 75:595 600. Nir, I., Z., Nitsan, and M. Mahagna. 1993. Comparative growth and development of the digestive organs and of some enzymes in broiler and egg type chicks after hatching. Br. Poult. Sci. 34:523 532. Nitsan, Z., E. A. Dunnington, and P. B. Siegel. 1991. Organ growth and digestive enzyme levels to fifteen days of age in lines of chickens differing in body weight. Poult. Sci. 70:2040 2048. Noy, Y., and Y. Pinchasov. 1993. Effect of a single posthatch incubation of nutrients on subsequent performance of broiler chicks and turkey poults. Poult. Sci. 72:1861 1866. Noy, Y., and D. Sklan. 1998a. Yolk utilization in the newly hatched poult. Br. Poult. Sci. 39:446 451. Noy, Y., and D. Sklan. 1998b. Metabolic response to early nutrition. J. Appl. Poult. Res. 7:437 451. Noy, Y., and D. Sklan. 1999a. Different types of early feeding and performance in chicks and poults. J. Appl. Poult. Res. 8:16 24. Noy Y., and D. Sklan. 1999b. Energy utilization in newly hatched chicks. Poult. Sci. 78:1750 1756. Noy, Y., and D. Sklan. 2001. Yolk and exogenous feed utilization in the posthatch chick. Poult. Sci. 80:1490 1495. Pinchasov, Y., and Y. Noy. 1993. Comparison of posthatch holding time and subsequent early performance of broiler chicks and turkey poults. Br. Poult. Sci. 69:1718 1723. SAEG (Sistemas de Análises Estatísticas). 1999. Fundação Arthur Bernardes, UFV, Viçosa. Brazil. Sklan, D. 2001. Development of the digestive tract of poultry. World s Poult. Sci. J. 57:415 428. Sklan, D., and Y. Noy. 2000. Hydrolysis and absorption in the small intestines of posthatch chicks. Poult. Sci. 79:1306 1310. Speake, B. K., R. Noble, A. Murray. 1998. The utilization of yolk lipids by the chick embryo. World s Poult. Sci. J. 54:319 334. Yu, M.W., and F. E. Robinson. 1992. The application of shortterm feed restriction to broiler chicken production: a review. J. Appl. Poult. Res. 1:147 153. Winick, M., and A. Noble. 1966. Cellular response in rats during malnutrition at various age. J. Nutr. 89:300 306.