Dietary Crude Protein Reduction with Addition of Crystalline Amino Acids in Growing Pekin Ducks Housed in Cascading Cages: Influence on Growth Performance, Carcass Traits, and Apparent Nutrient Digestibility

Abstract

The intensive raising model for meat duck production has widely adopted cascading cages. However, investigations on low-protein diets for Pekin ducks under this model are scarce. Thus, a 3 × 6 factorial experiment was conducted using six dietary crude protein (CP) levels (17.61%, 16.25%, 15.34%, 14.55%, 13.38%, and 12.64%, analyzed) and three cage layers (lower, middle, and upper). The study aimed to examine the effects of dietary CP reduction with crystalline amino acid supplementation on growth performance, carcass traits, and apparent nutrient digestibility in growing Pekin ducks from 21 to 42 days of age housed at different cage layers. A total of 1620 21-day-old Pekin ducks were randomly divided into 18 groups, each with six replicates of 15 ducks per replicate. Ducks were distributed across different cage layers and were fed one of six experimental diets, respectively, each varying in dietary CP levels but maintaining similar dietary energy and amino acid profiles. The results showed that the reduction in dietary CP levels affected growth performance (p < 0.05), while weight gain, feed intake, and feed/gain of Pekin ducks had no difference among 17.61%, 16.25%, and 15.34% CP groups (p > 0.05). As dietary CP decreased, abdominal fat yield increased (p < 0.05), but breast and leg meat yields were unaffected (p > 0.05). CP intake, nitrogen intake, and nitrogen excretion decreased linearly (p < 0.05) with decreasing dietary CP levels, while apparent nutrient digestibility in Pekin ducks increased proportionally (p < 0.05). Additionally, body weight, weight gain, feed intake, CP intake, nitrogen intake, and nitrogen excretion of Pekin ducks decreased (p < 0.05) due to different cage layers, while carcass characteristics remained unaffected by cage layer (p > 0.05). In summary, it is feasible to reduce the dietary CP level with crystalline amino acid supplementation in growing Pekin ducks from 21 to 42 days of age housed in cascading cages, while the cage layers affected growth performance of ducks. Based on broken-line models, it was determined that the optimal dietary CP requirement for achieving maximum weight gain and feed efficiency in growing Pekin ducks is approximately 15%.

1. Introduction

Duck meat, a primary source of animal protein, is an integral part of human diets. With the growing demand for duck meat, meat duck production has been increasing annually, the population of which reached approximately 5.6 billion worldwide in 2022 [1]. Consequently, significant advancements have been made in genetic breeding to maximize duck performance, driving the meat duck industry over the past decades. To optimize production efficiency, the majority of meat duck farming is conducted in intensive rearing houses with cascading cages on large-scale farms.

While the intensive rearing model meets consumer demand for duck products, it has led to severe environmental pollution due to ammonia emissions and nitrogen excretion in parallel. Formulating low-protein diets supplemented with crystalline amino acids is a viable strategy to mitigate this pollution. Additionally, global feed ingredient shortages have hindered the stable development of the duck industry. Thus, it is essential to economize feed ingredients by formulating low-protein diets. Previous studies performed on broilers [2], ducks [3,4], and quails [5] have shown that low-protein diets (a marginal reduction in dietary crude protein (CP) content) with supplementation of crystalline amino acids had no negative influences on growth performance and carcass characteristics, and caused low-protein intake and a reduction in nitrogen excretion. However, these studies were conducted under conditions of single-layer cages, floors, or pens, making the data potentially inapplicable to modern intensive rearing models with cascading cages on large-scale farms. Therefore, further investigation into low-protein diets for Pekin ducks under intensive rearing with cascading cages is warranted.

The cascading cage rearing system is a promising, environmentally friendly breeding mode that should be widely encouraged. However, in actual production, the distribution of environmental factors within the house is more complex than in traditional flat-breeding duck houses, potentially affecting the growth performance of Pekin ducks at different spatial locations. The layer of rearing cages must be considered in cascading cage rearing houses due to variations in environmental factors, including temperature, humidity, air velocity, particulate matter, and NH₃ and CO₂ concentrations, which may influence the growth performance of ducks at different cage layers. To our knowledge, little research has investigated the impact of different cage layers on growth conditions in meat duck production. Therefore, this study aims to evaluate the growth performance, carcass traits, and apparent nutrient digestibility of growing Pekin ducks fed different CP diets under a cascading cage rearing model and to determine the influence of different cage layers on these parameters.

2. Materials and Methods

2.1. Birds and Housing

All experimental procedures in the present study were approved by the animal care and welfare committee of Institute of Feed Research, Chinese Academy of Agricultural Sciences (approval number: CAAS-IFR-20200519). A total of 1800 1-day-old Pekin ducks (Type Z) obtained from one commercial hatchery were raised with common starter diets (metabolizable energy (ME): 2900 kcal/kg; CP: 20%) in cascading cages until 21 d of age. At 21 d of age, all ducklings were weighed individually, and the birds with lowest and largest body weight were removed. Finally, the remaining 1620 birds (1079 ± 40 g/bird, male and female are equally divided) were randomly allocated to 18 treatments with 6 replicates and 15 ducklings in each replicate. Ducks were raised in three-layer cascading cages (upper, middle, and lower; 200 cm × 120 cm × 40 cm) from 21 to 42 d of age. The heights from ground to lower, upper, and middle layer cages were 0.6, 1.2, and 1.8 m, respectively.

All ducks had free access to water (drip-nipple water supply lines) and feed (pellet form), and received continuous lighting each day. The temperature of the duck house was kept at 32 °C from 1 to 3 d of age, reduced gradually to approximately 25 °C until 14 d of age, and was kept approximately 18 to 22 °C from 21 to 42 d of age. The relative humidity was 20% from 1 to 3 d of age and then gradually increased to approximately 65% thereafter.

2.2. Experimental Design and Diets

A 3 × 6 factorial experiment, using 3 cage layers (lower, middle, and upper) and 6 dietary CP levels (17.61%, 16.25%, 15.34%, 14.55%, 13.38%, and 12.64%, analyzed), was conducted to investigate the effects of dietary CP levels and cage layer on growth performance, carcass characteristics, and apparent nutrient digestibility of growing Pekin ducks from 21 to 42 days of age.

The experimental diets containing different CP levels were formulated based on corn, soybean meal, and rapeseed meal, and the crystalline amino acids were added to the diets to obtain similar total amino acid profiles (

Table 1. Composition and nutrient contents of the experimental diets for Pekin ducks from 21 to 42 days of age (as-fed basis).

Item	Dietary CP Levels (%)
	17.5	16.5	15.5	14.5	13.5	12.5
Ingredients (%)
Corn	66.10	69.16	72.22	75.28	78.34	81.40
Soybean meal	19.30	17.44	15.58	13.72	11.86	10.00
Rapeseed meal	8.00	6.80	5.60	4.40	3.20	2.00
Soybean oil	2.00	1.60	1.20	0.80	0.40	0.00
Dicalcium phosphate	1.30	1.39	1.48	1.57	1.66	1.75
Limestone	1.20	1.18	1.16	1.14	1.12	1.10
Salt	0.30	0.30	0.30	0.30	0.30	0.30
DL-methionine	0.15	0.16	0.17	0.18	0.19	0.20
L-lysine•HCl	0.15	0.20	0.26	0.31	0.37	0.42
L-threonine	-	0.03	0.07	0.10	0.14	0.17
L-tryptophan	-	0.01	0.02	0.03	0.04	0.05
L-arginine	-	0.06	0.12	0.18	0.24	0.30
L-isoleucine	-	0.03	0.06	0.10	0.13	0.16
L-valine	-	0.04	0.08	0.12	0.16	0.20
L-Serine	-	0.05	0.10	0.14	0.19	0.24
Glycine	-	0.04	0.08	0.13	0.17	0.21
Chromium sesquioxide	0.50	0.50	0.50	0.50	0.50	0.50
Premix 1	1.00	1.00	1.00	1.00	1.00	1.00
Total	100.00	100.00	100.00	100.00	100.00	100.00
Nutrient levels 2
Metabolizable energy (MJ/kg)	12.17	12.17	12.17	12.17	12.16	12.16
Crude protein (%)	17.61	16.25	15.34	14.55	13.38	12.64
Calcium (%)	0.86	0.86	0.86	0.86	0.86	0.86
Total phosphorus (%)	0.62	0.62	0.62	0.62	0.62	0.62
Non-phytate phosphorus (%)	0.35	0.36	0.38	0.40	0.41	0.43
Methionine (%)	0.44	0.44	0.44	0.43	0.43	0.42
Methionine + cystine (%)	0.78	0.76	0.73	0.71	0.71	0.70
Lysine (%)	0.94	0.94	0.93	0.92	0.92	0.91
Threonine (%)	0.69	0.68	0.67	0.66	0.65	0.64
Tryptophan (%)	0.20	0.20	0.19	0.19	0.18	0.18
Arginine (%)	1.02	1.01	1.00	0.99	0.98	0.97
Valine (%)	0.77	0.77	0.76	0.76	0.75	0.74
Isoleucine (%)	0.62	0.61	0.60	0.59	0.58	0.57
Serine + glycine (%)	1.58	1.58	1.56	1.56	1.54	1.54
Phenylalanine (%)	0.85	0.80	0.75	0.70	0.65	0.60
Phenylalanine + tyrosine (%)	1.46	1.38	1.29	1.20	1.13	1.04

¹ Supplied per kilogram of total diet: vitamin A, 4500 IU; vitamin D₃, 2500 IU; vitamin E, 25 IU; vitamin K₃, 2.5 mg; thiamin, 1.8 mg; riboflavin, 12 mg; pyridoxine hydrochloride, 4.5 mg; cobalamin, 0.02 mg; calcium-d-pantothenate, 25 mg; nicotinic acid, 50 mg; folic acid, 1.5 mg; biotin, 0.2 mg; choline chloride, 1200 mg; Cu, 7.5 mg; Fe, 50 mg; Zn, 50 mg; Mn, 90 mg; Se, 0.25 mg; I, 0.45 mg. ² The dietary crude protein level represents analyzed values, and others are calculated values.

). The contents of nutrients in all experimental diets, except dietary CP level, all met the NRC (1994) [6] recommendation of Pekin ducks from 2 to 7 wk of age. The dietary CP levels were determined according to the Chinese standard method (GB/T 6432-2018) [7], and CP levels of 6 experimental diets were analyzed, respectively (Table 1).

2.3. Growth Performance

At 42 d of age, after overnight feed deprivation, the total weight of all ducks from each cage was measured. Then, the residual feed in each cage was weighed. Subsequently, average body weight, average daily weight gain, average daily feed intake, and feed/gain of Pekin ducks from 21 to 42 d of age were determined. Average daily feed intake and feed/gain were all corrected for mortality.

2.4. Carcass Characteristics

At 42 d of age, after overnight feed deprivation, one bird selected randomly from each cage was weighed and euthanized by CO₂ inhalation and eviscerated manually. The breast muscle (pectoralis major and pectoralis minor), leg muscle (thigh and drumstick), and abdominal fat were all removed manually from carcasses and weighed. The removed breast and leg meat were all skinless and boneless. To compensate for the differences in size of ducks, the percentages relative to live body weight at processing were calculated.

2.5. Measurement of Nutrient Digestibility

The nutrient digestibility trial was performed at the end of the experimental phase using chromic oxide (Cr₂O₃, 0.5%) as an external indicator in experimental diet (Table 1). The excreta of ducks from each cage were collected for 2 consecutive days (41 d and 42 d of age). After collection, the feathers and litter in excreta samples were removed, and the excreta were immediately stored at −20 °C. The collected excreta samples in each cage over 2 days were blended, dried at 65 °C for 48 h, and then ground to homogenous powder for further analysis. The Cr₂O₃ content in diets and excreta was analyzed spectrophotometrically [8]. Gross energy, CP, and dry matter (DM) of diets and excreta samples were determined. Energy was determined using an adiabatic calorimeter (HR-15A adiabatic calorimeter, Changsha, China). CP (N × 6.25) was determined using the Kjeldahl method (1030-Auto-analyzer, Tecator, Sweden). DM was determined by oven drying at 105 °C for 24 h. The results are presented on a DM basis. Apparent digestibility of DM, CP, and gross energy were calculated according to our previous study [9], as follows:

$$\mathrm{Apparent} \mathrm{digestibility} (\%)=100-\left[100\times \left(\frac{\mathrm{C}{\mathrm{r}}_2{\mathrm{O}}_3 \mathrm{in} \mathrm{diet} (\%)}{\mathrm{C}{\mathrm{r}}_2{\mathrm{O}}_3 \mathrm{in} \mathrm{feces} (\%)}\times \frac{\mathrm{nutrient} \mathrm{in} \mathrm{feces} (\%)}{\mathrm{nutrient} \mathrm{in} \mathrm{diet} (\%)}\right)\right]$$

(1)

2.6. Statistical Analysis

Data were analyzed by two-way ANOVA with GLM procedure of SAS 9.4 software, and the model included the main effects of cage layers, dietary CP level, and their interactions. In the case of significant interaction between cage layers and dietary CP level (p < 0.05), the Tukey multiple comparison test among all groups was used to compare differences among means, and multiple comparison in the main effect was not performed. When there was no significant interaction (p > 0.05), the multiple comparison in the main effect was performed by the Tukey method, and multiple comparison among all groups were not performed. Polynomial contrast analysis was performed to assess linear or quadratic responses of dietary CP levels on all dependent variables. The variability of data was expressed by the pooled standard error of the means (SEM). In our study, the broken-line regression analysis [10] was used to estimate the dietary CP requirements of Pekin ducks in 3 cage layers or in 3 merged cage layers using the PROC NLIN procedure (SAS 9.4), and the broken-line model is presented as follows:

$$\begin{array}{c}\mathrm{Y}=\mathrm{L}+\mathrm{U} (\mathrm{R}-\mathrm{X}) \mathrm{X}<\mathrm{R}\\ \mathrm{Y}=\mathrm{L} \mathrm{X}\ge \mathrm{R}\end{array}$$

where Y = growth performance or nutrient digestibility; X = dietary CP level (%); R = the abscissa of the breakpoint, which was taken as the requirement of dietary CP level; L = the response at X = R; and U = slope of the model.

3. Results

3.1. Growth Performance

As shown in

Table 2. Effects of dietary crude protein and cage layer on growth performance of Pekin ducks from 21 to 42 days of age.

Cage Layer	Dietary CP Levels (%)	Body Weight (g/bird)	Weight Gain (g/d/bird)	Feed Intake (g/d/bird)	Feed/Gain (g:g)	Mortality (%)
Lower	17.61	2895	86.1	242.7	2.82	0
	16.25	2876	85.2	241.5	2.83	0
	15.34	2820	84.5	243.0	2.88	1.11
	14.55	2840	83.5	241.2	2.89	1.11
	13.38	2732	79.7	239.6	3.01	2.22
	12.64	2700	77.1	237.4	3.08	0
Middle	17.61	2955	89.2	252.0	2.82	0
	16.25	2967	88.9	252.0	2.83	0
	15.34	2958	88.8	251.8	2.84	0
	14.55	2890	86.8	250.4	2.89	1.11
	13.38	2877	86.2	253.0	2.94	0
	12.64	2751	81.2	243.8	3.01	1.11
Upper	17.61	2989	89.3	254.2	2.85	0
	16.25	2932	88.5	253.2	2.86	0
	15.34	2963	89.2	254.4	2.85	0
	14.55	2924	87.2	253.9	2.91	0
	13.38	2890	85.6	255.8	2.99	0
	12.64	2747	80.6	246.7	3.06	1.11
	Pooled SEM	22	0.9	1.4	0.03	0.67
Main effects
Cage layer	Lower	2811 b	82.7 b	240.9 c	2.92	0.19
	Middle	2900 a	86.9 a	250.5 b	2.89	0.37
	Upper	2908 a	86.7 a	253.0 a	2.92	0.74
	Pooled SEM	9	0.4	0.6	0.01	0.27
Dietary CP level (%)	17.61	2946 a	88.2 a	249.7 a	2.83 b	0
	16.25	2925 a,b	87.5 a,b	248.9 a	2.84 b	0
	15.34	2913 a,b	87.5 a,b	249.7 a	2.86 b	0.37
	14.55	2885 b,c	85.8 b,c	248.5 a	2.90 b	0.74
	13.38	2833 c	83.8 c	249.5 a	2.98 a	0.74
	12.64	2733 d	79.6 d	242.6 b	3.05 a	0.74
	Pooled SEM	13	0.5	0.8	0.02	0.39
p-value
Two-way ANOVA	Cage layer	<0.001	<0.001	<0.001	0.180	0.3505
	Dietary CP level	<0.001	<0.001	<0.001	<0.001	0.4987
	Cage layer × Dietary CP level	0.117	0.830	0.510	0.974	0.4009
Polynomial Contrast	Linear (CP)	<0.001	<0.001	0.010	<0.001	-
	Quadratic (CP)	<0.001	<0.001	0.053	<0.001	-

^a–d Means with different superscripts within a column differ significantly at p < 0.05. Values are the means of 6 replicates of 15 ducks.

, dietary CP levels significantly affected the growth performance. The main effect showed that as dietary CP levels decreased gradually from 17.61% to 12.64%, body weight and daily weight gain of Pekin ducks were linearly or quadratically decreased (p < 0.05) and feed/gain were linearly or quadratically increased (p < 0.05). However, no difference was observed (p > 0.05) in these parameters among high-dietary-CP groups (17.61%, 16.25%, and 15.34%). No significant differences were observed in average feed intake among the different dietary CP levels (p > 0.05), except for the 12.64% CP group. Additionally, we observed that body weight, weight gain, and feed intake of ducks in the upper and middle layers were significantly greater that those in the lower layer (p < 0.05). Mortality of Pekin ducks showed no differences among all groups (p > 0.05). Furthermore, no interaction effects between dietary CP level and cage layers on these growth performance parameters were observed (p > 0.05).

3.2. Carcass Traits

As shown in

Table 3. Effects of dietary crude protein and cage layer on carcass characteristics of Pekin ducks at 42 days of age.

Cage Layer	Dietary CP Levels (%)	Breast Meat		Leg Meat		Abdominal Fat
		Weight (g)	Yield (%)	Weight (g)	Yield (%)	Weight (g)	Yield (%)
Lower	17.61	290.5	10.02	246.5	8.52	44.7	1.54
	16.25	291.3	10.13	250.5	8.71	44.7	1.55
	15.34	271.0	9.61	239.7	8.50	45.0	1.60
	14.55	282.8	9.95	242.0	8.52	46.2	1.63
	13.38	276.2	10.12	239.0	8.74	46.7	1.71
	12.64	276.5	10.24	228.5	8.44	47.8	1.77
Middle	17.61	293.5	9.94	249.0	8.43	45.8	1.55
	16.25	307.7	10.37	257.0	8.67	47.0	1.59
	15.34	294.0	9.94	263.5	8.91	46.8	1.58
	14.55	296.5	10.26	246.0	8.51	48.0	1.66
	13.38	287.5	9.99	252.3	8.76	49.3	1.71
	12.64	280.7	10.21	243.5	8.86	49.0	1.78
Upper	17.61	309.0	10.34	264.0	8.83	46.7	1.56
	16.25	286.8	9.78	243.5	8.31	46.2	1.58
	15.34	294.7	9.96	245.5	8.30	47.7	1.61
	14.55	307.7	10.54	256.2	8.77	48.2	1.65
	13.38	292.3	10.12	239.5	8.29	48.7	1.68
	12.64	269.2	9.83	237.5	8.67	47.2	1.72
	Pooled SEM	8.3	0.30	8.2	0.29	1.3	0.05
Main effects
Cage layer	Lower	281.4 b	10.01	241.0	8.57	45.8 b	1.63
	Middle	293.3 a	10.12	251.9	8.69	47.7 a	1.65
	Upper	293.3 a	10.09	247.7	8.53	47.4 a,b	1.63
	Pooled SEM	3.4	0.12	3.4	0.12	0.5	0.02
Dietary CP levels (%)	17.61	297.7 a	10.1	253.2	8.59	45.7	1.55 c
	16.25	295.3 a	10.09	250.3	8.56	45.9	1.57 c
	15.34	286.6 a,b	9.83	249.6	8.57	46.5	1.60 b,c
	14.55	295.7 a	10.25	248.1	8.60	47.4	1.65 a,b,c
	13.38	285.3 a,b	10.07	243.6	8.60	48.2	1.70 a,b
	12.64	275.4 b	10.09	236.5	8.65	48.0	1.76 a
	Pooled SEM	4.8	0.17	4.7	0.17	0.8	0.03
p-value
Two-way ANOVA	Cage layer	0.020	0.808	0.074	0.606	0.039	0.862
	Dietary CP levels	0.016	0.693	0.170	0.999	0.097	<0.001
	Cage layer × Dietary CP	0.380	0.743	0.606	0.736	0.987	0.998
Polynomial Contrast	Linear (CP)	0.002	0.856	0.011	0.759	0.003	<0.001
	Quadratic (CP)	0.303	0.744	0.378	0.786	0.907	0.093

^a–c Means with different superscripts within a column differ significantly at p < 0.05. Values are the means of 6 replicates of 2 ducks.

, breast meat weight decreased linearly (p < 0.05) as dietary CP levels decreased. However, the yield (absolute weight/body weight) of breast meat and leg meat was not affected (p > 0.05) by dietary CP levels. Although abdominal fat yield increased linearly as dietary CP decreased from 17.61% to 12.64% (p < 0.001), there were no significant differences among the higher dietary CP levels (17.61%, 16.25%, 15.34%, and 14.55%) (p > 0.05). Additionally, we observed that Pekin ducks in the lower layer had the lowest weights of breast meat (p = 0.020), leg meat (p = 0.074), and abdominal fat (p = 0.039), while the yields (relative weight) showed no differences among ducks housed at different cage layers. No interaction effects between dietary CP level and cage layers were observed for these carcass traits (p > 0.05).

3.3. Nutrient Apparent Digestibility

As is shown in

Table 4. Effects of dietary crude protein and cage layer on CP intake, nitrogen intake, and excretion of Pekin ducks from 21 to 42 days of age.

Cage Layer	Dietary CP Levels (%)	CP Intake (g/d/bird)	Nitrogen Intake (g/d/bird)	Nitrogen Excretion (g/d/bird)
Lower	17.61	42.7	6.82	1.51 a
	16.25	39.2	6.24	1.35 c,d
	15.34	37.3	5.95	1.18 f
	14.55	35.1	5.59	1.12 f,g
	13.38	32.1	5.10	0.96 i
	12.64	30.0	4.75	0.81 j
Middle	17.61	44.4	7.09	1.51 a
	16.25	41.0	6.59	1.39 b,c
	15.34	38.6	6.20	1.26 e
	14.55	36.4	5.76	1.17 f
	13.38	33.9	5.47	1.08 g,h
	12.64	30.8	4.93	0.84 j
Upper	17.61	44.8	7.08	1.44 b
	16.25	41.1	6.50	1.32 d,e
	15.34	39.0	6.19	1.26 e
	14.55	36.9	5.99	1.18 f
	13.38	34.2	5.49	1.04 h
	12.64	31.2	5.09	0.96 i
	Pooled SEM	0.2	0.04	0.02
Main effects
Cage layer	Lower	36.1 c	5.74 b	1.16
	Middle	37.5 b	6.01 a	1.21
	Upper	37.9 a	6.06 a	1.20
	Pooled SEM	0.1	0.02	0.01
Dietary CP levels (%)	17.61	44.0 a	7.00 a	1.49
	16.25	40.4 b	6.44 b	1.35
	15.34	38.3 c	6.11 c	1.23
	14.55	36.2 d	5.78 d	1.15
	13.38	33.4 e	5.35 e	1.03
	12.64	30.7 f	4.92 f	0.87
	Pooled SEM	0.1	0.03	0.01
p-value
Two-way ANOVA	Cage layer	<0.001	<0.001	<0.001
	Dietary CP levels	<0.001	<0.001	<0.001
	Cage layer × Dietary CP	0.449	0.067	0.001
Polynomial Contrast	Linear (CP)	<0.001	<0.001	<0.001
	Quadratic (CP)	0.115	0.435	0.064

^a–j Means with different superscripts within a column differ significantly at p < 0.05. Values are the means of 6 replicates of 2 ducks.

, CP intake, nitrogen intake, and nitrogen excretion decreased linearly (p < 0.05), while apparent nutrient digestibility (metabolizable energy, CP, and dry matter) of Pekin ducks gradually increased as dietary CP levels decreased (p < 0.05,

Table 5. Effects of dietary crude protein and cage layer on apparent nutrient digestibility of Pekin ducks from 21 to 42 days of age.

Cage Layer	Dietary CP Levels (%)	Metabolizable Energy (%)	Crude Protein (%)	Dry Matter (%)
Lower	17.61	84.9	77.8 h	82.0
	16.25	85.4	78.3 g,h	81.6
	15.34	85.5	80.2 b,c,d	82.5
	14.55	86.2	80.0 c,d,e	84.0
	13.38	86.3	81.1 b,c	83.8
	12.64	86.7	82.9 a	84.1
Middle	17.61	85.1	78.7 f,g,h	82.5
	16.25	85.5	78.9 e,f,g	83.1
	15.34	85.6	79.7 d,e,f	82.6
	14.55	86.0	79.7 d,e,f	83.3
	13.38	86.5	80.3 b,c,d	83.8
	12.64	86.9	83.0 a	84.2
Upper	17.61	84.8	79.6 d,e,f	82.4
	16.25	84.9	79.7 d,e,f	82.3
	15.34	85.7	79.6 d,e,f	82.8
	14.55	86.4	80.4 b,c,d	83.8
	13.38	86.1	81.0 b,c	83.8
	12.64	86.6	81.2 b	84.6
	Pooled SEM	0.3	0.3	0.3
Main effects
Cage layer	Lower	85.8	80.1	83.0
	Middle	86.0	80.1	83.3
	Upper	85.7	80.3	83.3
	Pooled SEM	0.1	0.1	0.1
Dietary CP levels (%)	17.61	84.9 d	78.7	82.3 b
	16.25	85.3 c,d	79.0	82.3 b
	15.34	85.6 b,c	79.8	82.6 b
	14.55	86.2 a,b	80.0	83.7 a
	13.38	86.3 a	80.8	83.8 a
	12.64	86.7 a	82.4	84.3 a
	Pooled SEM	0.1	0.2	0.2
p-value
Two-way ANOVA	Cage layer	0.360	0.522	0.163
	Dietary CP levels	<0.001	<0.001	<0.001
	Cage layer × Dietary CP	0.860	0.002	0.083
Polynomial Contrast	Linear (CP)	<0.001	<0.001	<0.001
	Quadratic (CP)	0.939	0.010	0.079

^a–h Means with different superscripts within a column differ significantly at p < 0.05. Values are the means of 6 replicates of 2 ducks.

). CP intake, nitrogen intake, and nitrogen excretion of Pekin ducks significantly increased (p < 0.05, Table 4) among different cage layers, while the apparent digestibility of metabolizable energy, CP, and dry matter showed no differences among ducks in different cage layers (Table 5). Additionally, interaction effects between dietary CP level and cage layers were observed for nitrogen excretion (p = 0.001, Table 4) and CP digestibility (p = 0.002, Table 5). According to the simple effect among all groups, the nitrogen excretion of Pekin ducks was gradually reduced as dietary CP decreased, regardless of cage layer (p < 0.05, Table 4). The simple effect of crude protein apparent nutrient digestibility showed a significant increase as dietary CP decreased (p < 0.05, Table 5).

3.4. Dietary CP Requirement

Dietary CP requirements of growing Pekin ducks were estimated using a regression model. Based on the linear broken-line model, the dietary CP requirements of growing Pekin ducks from 21 to 42 days of age for optimal weight gain were 15.07%, 15.08%, and 14.91%, and for optimal feed/gain, they were 15.76%, 15.57%, and 15.24% at different cascading cage layers, respectively (

Table 6. Dietary crude protein requirements of Pekin ducks from 21 to 42 days of age at different cage layer by linear broken-line models.

Response Criteria	Cage Layer	Broken-Line Models 1	R2	p-Value	Dietary CP Requirement (%)
Weight gain	Lower	Y = 85.27 − 3.34 × (15.07 − X)	0.979	0.003	15.07 ± 0.28
	Middle	Y = 88.97 − 2.71 × (15.08 − X)	0.884	0.040	15.08 ± 0.69
	Upper	Y = 89.00 − 3.27 × (14.91 − X)	0.928	0.022	14.91 ± 0.48
Feed/gain	Lower	Y = 2.83 + 0.078 × (15.76 − X)	0.971	0.005	15.76 ± 0.35
	Middle	Y = 2.83 + 0.060 × (15.57 − X)	0.986	0.002	15.57 ± 0.23
	Upper	Y = 2.85 + 0.078 × (15.24 − X)	0.995	<0.001	15.24 ± 0.14

¹ Regression equation is Y = L + U × (R − X) when X ≤ R, and Y = L when X > R; Y = weight gain, or feed/gain, X = dietary CP level (%), R = CP requirement, L = the response at X = R, U = slope of the model.

). Considering only the dietary CP levels, we observed that the dietary CP requirements of growing Pekin ducks from 21 to 42 days of age for optimal weight gain and feed/gain were 15.03% and 15.24%, respectively (Figure 1).

4. Discussion

Nowadays, the intensive duck raising model has been widely adopted globally. Research on low-protein diets for poultry has become a significant area of study due to the scarcity of feed resources [3,5,11,12]. Generally, crystalline amino acids, such as methionine (Met), lysine (Lys), threonine (Thr), tryptophan (Trp), and other amino acids, should be supplemented in low-protein diets to maintain the growth performance and carcass yield of poultry. As dietary CP is reduced in quail diets, the dietary amino acids may become growth-limiting factors. However, it is challenging to determine which amino acids, besides Met, Lys, Thr, and Trp, are limiting [13]. In starter and growing broilers, low-protein diets were effectively formulated by supplementing them with crystalline valine (Val), isoleucine (Ile), leucine (Leu), and arginine (Arg), in addition to Met, Lys, Thr, and Trp [2]. Additionally, the research of low-protein diets in broilers revealed important interactions among these amino acids [14]. Therefore, in the present study, crystalline amino acids, including essential amino acids (Met, Lys, Arg, Trp, Thr, Ile, Leu, and Val), and non-essential amino acids (serine and glycine), were all considered and supplemented when formulating low-protein diets of meat ducks. Unfortunately, dietary phenylalanine and phenylalanine + tyrosine contents gradually decreased with CP reduction (Table 1). However, stunted growth was observed when Pekin ducks were fed 13.38% and 12.64% CP diets where dietary phenylalanine and phenylalanine + tyrosine did not meet the requirements. This suggests that phenylalanine and phenylalanine + tyrosine may be growth-limiting factors when formulating low-protein diets.

Commonly, growth performance, such as weight gain, feed intake, and feed/gain, are vital and direct measurements for growth of meat-type livestock and poultry. In our study, as dietary CP levels decreased gradually, body weight and weight gain exhibited a linear or quadratic decrease, while feed/gain increased (Table 2). Previous studies on growing Pekin ducks demonstrated an increase in feed efficiency with decreased dietary CP levels [3]. Studies on quails showed a decrease in weight gain but a linear increase in feed intake and feed efficiency with gradual reduction in dietary CP levels [5]. Additionally, the CP intake of Pekin ducks showed a linear decrease when dietary CP level decreased from 17.61% to 12.64% (Table 4). Inadequate CP intake likely contributed to lower body weight and weight gain in Pekin ducks fed low-protein diets, consistent with findings in broilers [15] and previous studies on Pekin ducks [4]. This suggests that excessively reducing dietary CP levels, even when balanced with critical amino acids, negatively impacts the growth of Pekin ducks. Interestingly, we observed that there was no significant difference in growth performance among high-dietary-CP groups (17.61%, 16.25%, and 15.34%), regardless of the increased CP intake of Pekin ducks (Table 2 and Table 4), which suggesting that CP intake was adequate when dietary CP levels reached approximately 15%. In the present study, we observed that while breast meat weight decreased linearly, breast meat yield (relative weight) remained unchanged with dietary CP reduction. This may be attributed to the lower body weight of Pekin ducks due to varying dietary CP levels. Although abdominal fat yield increased linearly with dietary CP reduction, there was no significant difference among high dietary CP levels (17.61%, 16.25%, and 15.34%). Previous studies on starter Pekin ducks [12] and quails [5] have reported similar findings, suggesting that dietary protein deficiency leads to excessive abdominal fat accumulation. Additionally, excess dietary CP intake does not continuously decrease fat accumulation in Pekin ducks.

The cage layer may serve as a factor to influence the growth of ducks in a cascading cage rearing system. In the present study, growth performance, including body weight, weight gain, and feed intake, was negatively affected when ducks were housed in lower-layer cages. This phenomenon might be explained by environmental factors, such as higher concentrations of NH₃ and CO₂ and lower temperatures at the lower layer of the duck house. This underscores the importance of considering cage layers, particularly the lower layers, to mitigate any adverse effects on growth in meat duck production. Carcass features, including breast muscle, leg muscle, and abdominal fat yields, are crucial indicators for assessing the nutritional status of poultry diets. Fortunately, no differences were observed in carcass traits between ducks housed at different cage layers, indicating that cage placement does not affect carcass characteristics in Pekin ducks. Moreover, the absence of significant effects of cage layers on carcass characteristics indicates that, while environmental factors at different cage levels influence growth performance, they do not substantially impact meat quality. This finding has practical implications for farm management, suggesting that uniform meat quality can be maintained across different cage layers, even if growth rates vary.

It is widely known that low-protein diets benefit environmental protection, especially by reducing CP and nitrogen intake and nitrogen excretion in the poultry industry. In our study, CP intake, nitrogen intake, and nitrogen excretion decreased linearly (Table 4), while apparent nutrient digestibility (ME, CP, and DM) of Pekin ducks gradually increased as dietary CP levels decreased (Table 5). This indicates that an appropriate reduction in dietary protein levels can reduce nitrogen intake and excretion while increasing nutrient digestibility in Pekin ducks. These results are consistent with that of Hernández et al. [16], who reported that reducing dietary protein by 1.5% or 3% decreased nitrogen excretion to the environment by 9.5% and 17% in male broilers and by 11.8% and 14.6% in female broilers, respectively. Additionally, the CP intake, nitrogen intake, and nitrogen excretion of Pekin ducks significantly decreased in lower-layer cages (Table 5), which might explain the decreased growth performance in Pekin ducks in lower-layer cages (Table 1). Furthermore, we observed interaction effects between dietary CP levels and cage layers on nitrogen excretion (Table 4) and CP digestibility (Table 5), indicating that appropriately reducing dietary CP levels is more significant for improving CP digestibility and decreasing nitrogen excretion in Pekin ducks housed in lower cage layers. Further research is needed to refine low-protein diet formulations and explore additional strategies, such as phase feeding or the use of alternative protein sources, to mitigate the negative impacts on growth performance while enhancing environmental sustainability [17].

Different regression models, including quadratic polynomial [4,18], linear [19], and quadratic broken-line models [20,21], have been previously fitted to assess nutrient requirements for poultry. Similarly, these models were fitted to determine the optimal dietary CP levels to formulate low-protein diets [3,5,12]. In the present study, as dietary CP decreased, the CP intake decreased linearly, while the weight gain and feed/gain of Pekin ducks were not impacted until dietary CP levels dropped to around 15%. Importantly, the decrease in CP intake resulted in reduced ammonia emissions and nitrogen excretion (Table 4), benefiting environmental pollution reduction. Therefore, predicting the minimum dietary CP requirement of growing Pekin ducks becomes crucial. In our study, linear broken-line regression was used to estimate the dietary CP requirement for optimal weight gain and feed/gain (vital indicators of growth performance) of growing Pekin ducks housed at different cage layers. According to the linear broken-line model, the dietary CP requirement of growing Pekin ducks for optimal weight gain was 15.07%, 15.08%, and 14.91%, and for optimal feed/gain, it was 15.76%, 15.57%, and 15.24% at different cage layers, respectively (Table 6). Considering only the dietary CP levels, the optimal dietary CP requirement for achieving optimal weight gain and feed efficiency in growing Pekin ducks was determined to be 15.03% and 15.24%, respectively (Figure 1). The optimal dietary CP requirement for achieving optimal feed efficiency (15.24%) was slightly higher than that for optimal weight gain (15.03%). Both values were lower than the CP recommendation of the NRC (1994) [6] for growing Pekin ducks aged 2 to 7 weeks (16%). These estimated response values align well with a previous study on the dietary CP requirement for growing Pekin ducks aged 14 to 35 days, which utilized a diet containing 15% CP supplemented with crystalline amino acids [3].

5. Conclusions

In summary, it is feasible to reduce the dietary CP level with crystalline amino acids supplementation in growing Pekin ducks from 21 to 42 days of age housed in cascading cages. Cage layers affected the growth performance of ducks, which suggested that cage location could be critical as a factor to consider in cascading cage rearing systems. According to the linear broken-line models, the dietary CP requirements of growing Pekin ducks for optimal weight gain were 15.07%, 15.08%, and 14.91%, and for optimal feed/gain, they were 15.76%, 15.57%, and 15.24% at different cascading cage layers (lower, middle, and upper), respectively. Taken together, the dietary CP requirement of growing Pekin ducks was approximately 15%.