MATERIALS AND METHODS

Ethical approval

All procedures involving animals were reviewed and approved by the Animal Ethics Committee, Faculty of Veterinary Medicine, Universitas Brawijaya, Malang, Indonesia (Approval No. 57-KEP-FKHUB-2025). The study was conducted in accordance with national guidelines governing the care and use of animals in research and complied with institutional standard operating procedures for handling bulls. Throughout the experimental period, all animals were maintained under appropriate housing, feeding, and veterinary management to ensure their welfare. The study also complied with Indonesian regulations concerning animal welfare and internationally accepted principles for the ethical use of animals in scientific research.

Study period and location

The study was conducted from January to December 2025 and comprised both the adaptation and experimental periods, following the procedures described by Moran [15]. The study was conducted in Bangkalan Regency, East Java Province, Indonesia (7°01′51″S, 112°44′42″E), at an altitude of approximately 56 m above sea level. This region is characterized by smallholder-based cattle production systems in which forage availability is often limited and locally available agricultural by-products are widely utilized, making it an appropriate location for evaluating locally formulated concentrate rations. During the 21-day adaptation period, bulls were gradually acclimatized to the concentrate mixtures and communal housing facilities. Maize stover was offered during the first 3 days, followed by a gradual increase in concentrate allowance during the subsequent 18 days to facilitate ruminal adaptation [16]. The experimental period lasted for 90 days after completion of the adaptation phase. All animal-related procedures were performed in smallholder-managed communal stalls modified to permit individual feeding, thereby ensuring experimental accuracy without disrupting the traditional production environment. Nutrient analyses were conducted at the Nutrition Laboratory, Faculty of Animal Science, Universitas Brawijaya, Malang, Indonesia, under standardized laboratory conditions.

Experimental animals

Forty bulls, consisting of 20 Madura and 20 Madrasin animals aged 2.0–2.5 years, were used in the study. The breed-specific initial live weight (LW; mean ± standard deviation) was 288 ± 18.4 kg for Madura bulls and 322 ± 22.1 kg for Madrasin bulls, with an overall mean LW of 305 ± 23.6 kg. Before the experiment, all animals underwent clinical examination by a veterinarian and were confirmed to be healthy. No vaccinations were administered because no endemic disease outbreaks were reported during the study period. Animals were purchased from regional cattle markets in Bangkalan with documented farmer origin and health records to ensure traceability. Each bull was fitted with a neck collar tag for identification throughout the experiment.

Prior to the adaptation and feeding phases, internal parasites were controlled with Nitroxinil (Fluconix® 340; Interchemie Werken "De Adelaar" B.V., Castenray, the Netherlands) at a dose of 1.5 mL/50 kg LW. Subsequent mentions should refer to Fluconix® 340 or Interchemie Werken "De Adelaar" B.V. only. LW measurements were performed every 2 weeks in the morning before feeding to facilitate adjustment of feed allowances according to current LW.

Stall design and management conditions

Animals were housed in communal cattle stalls equipped with separators to enable individual feed intake measurements. Separate feed troughs were provided for forage and concentrate. The pens were roofed and had concrete flooring to minimize thermal and environmental stress. Fresh drinking water was provided ad libitum throughout the study.

Study design and feeding regimes

The experiment was conducted using a randomized complete block design with a 2 × 4 factorial arrangement, comprising two breeds (Madura and Madrasin) and four concentrate treatments. A total of 40 bulls (20 Madura and 20 Madrasin) were allocated to the treatments [17]. Initial LW served as the blocking factor, with five blocks established according to LW classes ranging from 260 to 350 kg. Animals were randomly assigned to one of four dietary treatments, resulting in five animals per breed × treatment combination. Thus, the block term in the statistical model represented the initial LW class. Individual animals were considered the experimental units because they were managed and fed individually and feed intake and refusals were measured separately. This experimental design was based on previous studies [18, 19].

All bulls received native grass at 0.5% of LW as the basal diet and one of four concentrate mixtures at 2.25% of LW on a DM basis. Native grass was harvested at the vegetative stage, chopped into 3–5 cm lengths, and offered fresh. The dietary treatments were as follows:

1. T1 = native grass (0.5% LW) + C1 (2.25% LW)

2. T2 = native grass (0.5% LW) + C2 (2.25% LW)

3. T3 = native grass (0.5% LW) + C3 (2.25% LW)

4. T4 = native grass (0.5% LW) + C4 (2.25% LW)

The ingredient composition of the concentrate mixtures is presented in Table 1. The principal difference among treatments was the progressive increase in cassava waste from 19% in C1 to 34% in C4, accompanied by a corresponding reduction in tofu waste from 20% to 5%. Rice bran, copra meal, broken corn, and mineral mixture remained constant among treatments. Concentrates were mixed daily using a mechanical mixer, stored in sealed plastic containers at room temperature, and offered twice daily at 08:00 and 15:00 h. Unlike many station-based studies that use pelleted or ensiled feeds, the present study employed fresh local by-products at inclusion levels representative of farmers' practices while maintaining near-isoenergetic concentrate formulations.

Ingredients | C1 | C2 | C3 | C4
Rice bran | 20 | 20 | 20 | 20
Cassava waste | 19 | 24 | 29 | 34
Tofu waste | 20 | 15 | 10 | 5
Copra meal | 30 | 30 | 30 | 30
Broken corn | 10 | 10 | 10 | 10
Mineral mix | 1 | 1 | 1 | 1
Total | 100 | 100 | 100 | 100

Feed ingredients and chemical composition analyses

The ingredients used in this study included native grass, rice bran, cassava waste, tofu waste, copra meal, broken corn, and a mineral mix. Samples of each ingredient were analyzed for dry matter (DM), organic matter (OM), crude protein (CP), ether extract (EE), and crude fiber (CF) according to the methods described by AOAC [20]. Total digestible nutrient (TDN) values were calculated according to the equation proposed by Moran [16]. Acid detergent fiber (ADF) and neutral detergent fiber (NDF) contents were determined following the procedure described by Van Soest et al. [21].

Feeding procedures and intake measurements

Feed allowances were adjusted weekly according to changes in LW to maintain the desired feeding rates. Forage and concentrate were offered separately to minimize selective feeding. Feed refusals were collected daily before morning feeding, weighed, and dried in a forced-air oven at 60°C for 72 h to determine DM content [19]. Dry matter intake (DMI) was expressed as kg/day, percentage of LW, and relative to metabolic LW (g/kg LW0.75). Nutrient intake values for OM, CP, ADF, and NDF were also calculated to evaluate treatment effects on nutrient consumption.

Digestibility assessment and analytical procedures

Digestibility was determined by collecting total fecal matter during weeks 5, 9, and 13 of the experiment, following the procedures described by Moran [16]. Feed intake and fecal output data were used to calculate apparent digestibility coefficients. Feces were collected daily from each bull, weighed, homogenized, and a 10% subsample was stored at −20°C. At the end of each 7-day collection period, subsamples were thawed, pooled by animal, and dried at 60°C for 72 h before analysis. Parameters evaluated included DM, OM, CP, ADF, and NDF digestibility. These measurements enabled assessment of nutrient utilization under the different concentrate formulations and followed methods previously used in feed evaluation studies in Bangkalan [22]. Digestible nutrient intake was also expressed relative to metabolic LW to account for differences in animal size.

Measurement of growth performance and economic indicators

LWG was calculated as:

LWG (Kg/day) = Final LW – Initial LW / Experimental period

Body weight was measured weekly in the morning before feeding, after fasting. Feed conversion ratio (FCR) was calculated as:

FCR = Daily feed intake / Daily weight gain

Economic performance was assessed using IOFC, as described by Ndaru et al. [11]. This parameter considered both the economic value of weight gain and the cost of feed consumed. The equation used was:

IOFC (IDR/day) = (LWG, Kg/day × Price of bulls/Kg LW in IDR) – (DMI, Kg × Feed price/Kg DM in IDR)

Statistical analysis and modeling approach

Data were analyzed using a mixed-model analysis of variance to evaluate the effects of treatment and breed on the measured variables. Treatment and breed were considered fixed effects, whereas block was treated as a random effect. Statistical analyses were performed using the MIXED procedure of SAS version 9.4 (SAS Institute Inc., Cary, NC, USA) [23]. Least squares means were calculated for each variable, and treatment × breed interactions were evaluated when appropriate. Means were separated using Tukey's adjustment when significant differences were detected. Statistical significance was declared at p < 0.05.

The statistical model used was:

Yijkl = µ + Bi + Tj + (B × T)ij + Blk + Eijkl

where:

1. Yijkl = observed value;

2. µ = overall mean;

3. Bi = effect of breed;

4. Tj = effect of dietary treatment;

5. (B × T)ij = breed × treatment interaction;

6. Blk = block effect based on initial body weight class;

7. Eijkl = residual error.

The breed × treatment interaction was tested and found to be nonsignificant (p > 0.05) for all major response variables, including intake, digestibility, LWG, FCR, and IOFC. Therefore, interaction effects were omitted from the final model and only the main effects are presented.