Evaluation of a Peanut Skin-Supplemented Corn/Soy Diet on Broiler Meat Quality

Experimental design and feed information

A controlled experiment was conducted with 90 Ross 708 male broilers, which were randomly assigned at hatch to 2 dietary treatments supplemented with or without 5% ground peanut skins (45 birds/treatment, 3 pen/treatment, 15 birds/pen). Hatchlings were provided with feed and water ad libitum for 6 wk and housed in floor pens lined with pine shaving bedding. Broilers were fed a 3-phase corn/soy mash diet regimen. Diets were isocaloric and isonitrogenous, with the starter providing 3,000 kcal/kg and 22% crude protein, the grower providing 3100 kcal/kg and 21% protein, and the finisher providing 3200 kcal/kg and 19% protein. A PN skin-supplemented diet was prepared for half of the broilers by adding 5% ground PN skins to the basal diet at each phase. Collective pen body weights and feed intake were recorded weekly. At 6 wk of age, 5 birds were sampled from each pen (15 birds/treatment, pen = experiment unit) for processing. All animal experimental procedures conducted during the feeding trials were approved by the Institutional Animal Care and Use Committee (IACUC) of North Carolina State University (IACUC # 24-134-02 and IACUC # 22-402-05).

Meat processing and sampling

Broiler chickens were processed at the Chicken Education Unit of the Prestage Department of Poultry Science at North Carolina State University. Upon arrival, live weights were recorded, and birds were placed on the processing line for low-voltage electrical stunning (∼15 V for 10 s), followed by a neck cut to allow bleeding for approximately 1.5 min. Birds were then scalded at 62.78°C for about 60 s and mechanically defeathered. Evisceration was performed manually, and hot carcass weights along with initial pH values were recorded. Carcasses were submerged in a pre-chiller at 14°C for 30 min, then transferred to a chiller maintained between 0 and 4°C for 90 min. Water temperatures were monitored using a THERMOWORKS thermometer (Model: RT301WA-N). After chilling, carcasses were weighed and transferred to a cutting room at 10°C, where they were cut into breasts, tenders, wings, thighs, and drumsticks. The right breast was used to measure pH at 2 hr postmortem, and individual cut weights were recorded. Carcass parts were then immersed in a 150 ppm peracetic acid solution for 10 seconds and stored in pre-labeled Ziploc bags for traceability. Samples were kept in a walk-in cooler at 4–10°C for 24 hr before further analysis. Overall, carcasses were chilled after 15 min post-slaughter, deboned after 2 hr of chilling, and kept at refrigeration temperature after deboning for 24 hr to take the pH, color, and weight measurements.

Weight measurements

For both treatment groups, individual body weights were recorded upon arrival. After the first processing, chilled carcass weights were also measured for birds in both groups. Whole breast and right breast muscle weights were measured at 24 hr postmortem. All weight measurements were obtained using a calibrated digital scale (Model: Champ Scale BASE With CD-11 Weighing Indicator Terminal, Ohaus Corporation, Parsippany, NJ, USA).

pH measurement and color evaluation

The pH of the breast muscle (n = 15 per treatment) was assessed using a calibrated pH meter (Model HI98263, Hanna Instruments Inc., Woonsocket, RI, USA) equipped with a penetrating probe featuring an integrated temperature sensor (FC2323, stainless steel blade tip, optimized for meat). Measurements were taken at 15 min, 2 hr, and 24 hr postmortem, as well as following 2.5 mo of frozen storage after subsequent thawing. The pH probe was inserted into the thickest point in the cranial region of the fillet for consistent sampling. Color attributes of breast fillets from both peanut skin-supplemented and control treatments (n = 15 per treatment) were evaluated at 24 hr postmortem and after 2.5 mo of frozen storage in thawed samples. Measurements were recorded at 3 anatomical locations (cranial, medial, and caudal) using a handheld Minolta chromameter (CR-400, Konica Minolta Sensing Americas Inc., Ramsey, NJ, USA), configured with SpectraMagic^™ NX CM-S100w software. The instrument was standardized with a D65 illuminant, 2° observer angle, and minimal surface reflectance. Calibration was performed using the manufacturer-provided white calibration plate, with Y, x, and y values set at 93.5, 0.3114, and 0.3190, respectively. Color values were expressed according to the Commission Internationale de l'Éclairage (CIE) Lab* system, representing lightness (L*), redness (a*), and yellowness (b*). Each color parameter was measured in triplicate across the 3 regions, and mean values for L*, a*, and b* were subsequently calculated.

Thaw loss and cook loss measurement

Breast fillets (n = 8/treatment) were thawed for approximately 24 hr at 4°C in the refrigerator following removal from frozen storage. After thawing, the samples were taken out of vacuum-sealed packaging and weighed to determine thawing loss, expressed as a percentage of the initial raw weight, using the following formula:

Breast samples were cooked in a preheated oven at 177°C until reaching a final internal temperature of approximately 74°C. The Taylor food thermometer (Taylor Bright LED kitchen thermometer, Model: 5265465KHL) was used to check the internal temperature of the samples. Following thermal processing, the samples were cooled to ambient temperature (22 ± 2°C) and reweighed to determine the cooked weight. Cooking loss was expressed as a percentage and calculated using the following equation:

Blunt Meullenet–Owens razor shear and Warner–Bratzler shear force determination

The texture analysis was conducted on right breast fillets following 2.5 mo of frozen storage. Shear force assessments were performed on both raw and cooked samples using the blunt blade configuration of the blade-meat razor shear (BMORS) method (Bowker and Zhuang, 2019). Samples were cooked as described in the above section. Measurements were done on the cranial portion of the ventral surface of each fillet and were carried out with a TA-XT Plus C texture analyzer (Texture Technologies Corp., South Hamilton, MA, USA) equipped with a 2 kg load cell. To minimize errors due to sample preparations, the tests were conducted on the whole intact fillets placed over a duty platform with an aluminum plate. The blunted razor blade was applied perpendicularly to the muscle fiber orientation. The BMORS test was conducted with a 5-kg load cell, a trigger force of 5 g, and a distance of 20 mm. The pre-test and post-test speeds were set at 1 mm/s and 10 mm/s, respectively, while the probe test speed was maintained at 2 mm/s. Results of BMORS shear readings, including peak shear force (N), total shear energy (N·mm), and the frequency of shear peaks, were recorded. Each treatment underwent 6 shear tests on 6 different samples in both raw and cooked conditions, with each sample comprising 3 consecutive shears within the cranial region. For Warner–Bratzler (W–B) shear force analysis, the cooked samples (6 samples per treatment) were shortened by excising 3 consecutive strips measuring 1.5 cm in width, 1.5 cm in thickness, and 2 cm in length from the cranial region of each breast fillet, oriented parallel to the muscle fibers. Shear testing was performed perpendicular to the fiber direction using a W–B knife with a guillotine mounted on a TA-XT Plus C texture analyzer. Warner–Bratzler shear force measurements were performed with a 5-kg load cell, a trigger force of 20 g, and a distance of 30 mm. The pre-test and post-test speeds were set at 3.3 mm/s and 5 mm/s, respectively, while the probe test speed was maintained at 3.3 mm/s. Shear force readings were reported as the peak force (N) and toughness (N·mm) required to break the samples.

Meat composition analysis

Three replicate samples of right-side breast muscle from each treatment group were thawed overnight at 4°C, then equilibrated for 1 hr at room temperature. The samples were minced and ground using a food processor (Model: FP4200B-RF, Black+Decker, Towson, MD, USA) for subsequent compositional analysis. Near-infrared (NIR) spectra were obtained using the FoodScan^™ Meat Analyzer equipped with ISIScan^™ Nova software (FoodScan^™ 2, FOSS Analytical A/S, Hillerød, Denmark). Ground samples were loaded into the sample cup and analyzed in transmission mode across the 850–1100 nm wavelength range, with each scan lasting approximately 15–20 s. 3 spectra were collected from different regions of each sample to ensure representativeness. Measurements included fat, moisture, protein, collagen, and ash content.

Muscle myopathies

The right breast muscle fillets (n = 15/treatment) were assessed after 2 h post-slaughter for the incidence of woody breast (WB), white striping (WS), and spaghetti meat (SM) myopathies. Breast fillets were assessed using manual palpation for WB and visual inspection for WS and SM. WB and WS were scored on a scale from 0 to 3. A score of 0 indicated normal fillets with full flexibility and either no visible striping or only faint striping in the cranial region. A score of 1 denoted mild myopathy, characterized by firmness localized to the cranial portion and the presence of white striations less than 1 mm in thickness. A score of 2 indicated moderate severity, with reduced flexibility in the midsection and pronounced striping between 1–2 mm. A score of 3 represented severe myopathy, where the fillet was rigid throughout and exhibited striping greater than 2 mm in thickness. SM myopathy was assessed on a separate 0 to 2 scale. A score of 0 indicated normal muscle integrity with no observable fiber separation. A score of 1 denoted moderate SM, evident by visible fiber separation on the surface, while a score of 2 indicated severe SM, characterized by prominent fiber separation extending below the surface.

Scanning electron microscopy and sarcomere length

Frozen right breast muscle samples were thawed for approximately 24 hr at 4°C in the refrigerator following removal from frozen storage and then sectioned into uniform blocks measuring approximately 20 mm × 20 mm × 10 mm (n = 2 per treatment group). The tissues were fixed in a solution of 4% paraformaldehyde and 1% glutaraldehyde (4F1G) prepared in 0.1 M sodium cacodylate buffer at ambient temperature. Following fixation, the samples were rinsed 3 times for 15 min each in 0.1 M sodium cacodylate buffer and then subjected to a graded ethanol series for dehydration. Critical point drying was performed using a Samdri-795 dryer (Tousimis, Rockville, MD, USA). The dried specimens were mounted on aluminum stubs and sputter-coated with a gold/palladium alloy for 2 min using a Cressington sputter coater. Scanning electron microscopy (SEM) images of longitudinal sections of breast muscles were collected using a Hitachi SU3900 variable pressure SEM operated at 10 keV at the Analytical Instrumentation Facility, North Carolina State University, Raleigh, NC, USA. Sarcomere length was quantified by calculating the Euclidean distance between adjacent sarcomere boundaries, using the high-magnification SEM images to identify and reference the parallel striations of each sarcomere. An example of SEM images of the breast meat samples from each group is shown in Figure 1. Sarcomere lengths were determined by measuring the distance between the parallel Z-lines (dark bands) visible in the figure. Sarcomere lengths were measured using ImageJ software for our study and recorded approximately 30 times per sample to enhance measurement reliability. The mean value of the measurements was used as the sarcomere length for that sample to improve the accuracy of the final sarcomere length estimation.

Figure 1.

Scanning electron micrographs showing longitudinal sections of chicken breast meat from broilers fed a conventional diet (a) and a PN skin diet (b).

Sensory evaluation-quantitative descriptive analysis

Sensory evaluation was conducted using a paper-based Quantitative descriptive analysis (QDA) approach, in accordance with Institutional Review Board protocol #27894 titled “Meat Quality (Sensory Evaluation) of Broiler and Layer Meat Supplemented with Peanut or Peanut Skin.” Frozen right breast samples were thawed at 4°C for approximately 24 hr and subsequently cooked in a preheated oven at 177°C until the internal temperature reached ∼74°C. Doneness was confirmed using a calibrated Taylor LED kitchen thermometer (Model: 5265465KHL). Cooked fillets were arranged on labeled stainless steel trays and coded with randomized three-digit identifiers to ensure blinding. Each fillet was then portioned into uniform 1.5 cm × 1.5 cm × 1.5 cm cubes and served to panelists on disposable paper plates. At the start of each session, panelists were provided with palate cleansers (apple juice, water, unsalted crackers), evaluation forms, pencils, napkins, forks, and expectoration cups. Prior to tasting each sample, panelists were instructed to cleanse their palate using the provided items. Each sensory session included 7 trained panelists, each with over 30 hr of experience evaluating cooked breast fillets, including a faculty member with over 12 years of meat sensory evaluation experience. During each session, panelists evaluated 2 samples per treatment group, ensuring that samples were sourced from the same anatomical location within the breast muscle to minimize sensory variation. Panelists underwent one month of training across 18 sessions to calibrate their evaluation of textural attributes (juiciness, springiness, tenderness, cohesiveness, gumminess, chewiness), flavor notes (brothy, meaty, rancid, cardboardy, metallic, off-flavor), and basic tastes (sweetness, umami, sourness, saltiness, bitterness). Sensory attribute definitions were adopted from Meilgaard et al. (1999) and Zhang et al. (2020). Each attribute was rated on a 15-point sensory intensity line scale, where 0 indicated a barely perceptible attribute, 3 indicated slight intensity, 7 indicated moderate intensity, 11 indicated high intensity, and 15 indicated extreme intensity. Individual evaluation forms were used for data collection and retrieved at the end of each session.

Lipid oxidation analysis

Lipid oxidation in the samples (n = 3 per treatment) was evaluated using a modified thiobarbituric acid reactive substances (TBARS) assay, adapted from the protocols of Buege (1978) and Luque et al. (2011). A standard curve was generated using 3M 1,1,3,3-tetraethoxypropane, prepared in conjunction with a trichloroacetic acid and thiobarbituric acid (TCA/TBA) reagent mixture. Additionally, a butylated hydroxyanisole (BHA) solution was formulated by dissolving 10 g of BHA in 90% ethanol and adjusting the volume to 100 mL. At predetermined time points (day 0, 14, and 28) during retail display, meat samples were rapidly frozen in liquid nitrogen and pulverized into a fine powder using a high-speed grinder (SP-7412, Secura Inc., Lake Forest, CA, USA). A 5.0 ± 0.1 g aliquot of each powdered sample was mixed with 15 mL of deionized water, vortexed, and subjected to centrifugation at 1850 × g for 10 min. Subsequently, 2 mL of the resulting supernatant was transferred to clean tubes and combined with 4 mL of the TCA/TBA reagent and 100 μL of the BHA solution. After vortexing for 1 min, the tubes were incubated in a 100°C water bath for 15 min to facilitate the reaction, followed by immediate cooling in an ice bath for 10 min. The samples were then centrifuged again under identical conditions. From each sample, 200 μL aliquots were pipetted in duplicate into a 96-well microplate. Absorbance at 531 nm was measured using a multi-mode microplate spectrophotometer (SpectraMax i3x Multi-Mode Microplate Reader, Molecular Devices LLC., San Jose, CA, USA). The concentration of malondialdehyde (MDA), a marker of lipid peroxidation, was quantified based on the standard curve and expressed in mg/kg.

Statistical analysis

A complete random design was employed for this study, with 15 birds assigned per treatment group (5 birds per pen; 3 pens per treatment; pens = experimental units). Individual dependent variables, including live weight, carcass and breast muscle weights, pH, color, cook loss, thaw loss, shear force, meat composition, lipid oxidation, sarcomere length, and myopathy scores, were analyzed using one-way analysis of variance via the PROC GLM procedures in SAS statistical software (SAS version 9.4, SAS Inc., Cary, NC, USA). Treatment means were separated using Duncan’s New Multiple Range test, with statistical significance determined at P < 0.05.

Sensory data were analyzed using the GLIMMIX procedure (Generalized Linear Mixed Models) in SAS (version 9.4, SAS Institute Inc., Cary, NC, USA). The statistical model included the treatment type (PN skin and conventional as fixed effects, while replication and panelist were treated as random effects. The Kenward-Roger method was used to estimate denominator degrees of freedom and adjust standard errors for fixed effects in the mixed model analysis (ddfm = kr). Least-Squares Means (LSMEANS) with Tukey’s adjustment for multiple comparisons (adjust = tukey) were used to determine significant differences between treatment means at a significance level of P < 0.05.

Growth performance and yield traits

Broilers fed with PN skin diet exhibited significantly lower body weight compared to the control group (Table 1, 2.21 kg vs 2.49 kg, respectively; P < 0.05), suggesting that peanut skin inclusion may have suppressed growth. The chilled carcass weight, as well as both whole and right breast weights, exhibited a similar trend, showing a reduction in broilers fed a peanut skin–based diet (P < 0.05). The reductions in yield parameters might be attributed to the high tannin content in peanut skin. Tannin is known to interfere with protein digestibility and nutrient absorption when added in high concentrations, particularly from purified sources or tannic acid (Redondo et al., 2014; Garcia et al., 2004). However, low-to-moderate tannin supplementation has shown antioxidant, antimicrobial, and anti-inflammatory effects in poultry, which can support gut health and may help meat quality (Choi and Kim, 2020; Toomer et al., 2024). As a result, the optimal inclusion of peanut skin in broiler diets may serve as a beneficial and sustainable feed additive in poultry production. Although it may lead to a slight reduction in body weight, this trade-off could be justified by the antimicrobial properties of PN skin inclusion (Toomer et al., 2024). Moreover, Hisasaga and Makagon (2024) reported that Ross 708 broilers at the target age of 6 wk exhibited body weights ranging from approximately 2.08 to 4.28 kg. The body weights observed in both treatment groups in the present study fall well within this established range, with no mortalities over the course of the 6 wk feeding trial. However, we acknowledge that being within this range does not necessarily overlook the observed reduction in body weight compared with the control group.

pH

Following slaughter, the accumulation of lactic acid in muscle tissue leads to a decline in pH for both diet groups. No difference in meat pH was observed at 15 min postmortem (Table 2, P > 0.05), but by 2 hr, the PN skin group had a lower pH than the control, indicating a slightly faster pH decline (5.81 vs 5.94, respectively; P < 0.05). However, by 24 hr postmortem, the pH values of the conventional and peanut skin diet groups were no longer distinct (P > 0.05) despite earlier differences observed at 2 hr. This observation can be attributable to the rapid postmortem conversion of muscle glycogen to lactic acid, which leads to a rapid decline in pH (Yue et al., 2024; Mir et al., 2017). Within 24 hr, this process reaches completion, and the muscle enters rigor mortis, resulting in a stable pH that is largely independent of upstream dietary influences (Liu et al., 2022). Notably, by 2.5 mo of frozen storage, PN skin samples had a lower pH compared to controls (5.75 vs 5.89, respectively; P < 0.05). Nevertheless, the pH values were inside the optimal range for both groups. The ultimate pH of broiler breast meat typically falls within the optimal range of 5.7 to 6.1 (Beauclercq et al., 2022; Bihan-Duval et al., 2018), with values below 5.7 indicative of acid meat and those exceeding 6.1 associated with dark, firm, and dry characteristics (Fletcher et al., 2000). Additional statistical analysis across time points within each diet (Table 2) indicated that the pH decline occurred mainly between 15 min and 2 hr (P < 0.05), with no change thereafter (P > 0.05). This result suggests that most of the postmortem acidification occurred within the first 2 hr after slaughter. Meat pH is also a major determinant of water-holding capacity (WHC), which influences meat juiciness, drip loss, and yield during storage and cooking (Huff-Lonergan and Lonergan, 2005). Rapid pH decline may lead to protein denaturation and decreased WHC (Huff-Lonergan and Lonergan, 2005). In contrast, meat reaching ultimate pH within the normal range, as observed here, tends to maintain acceptable WHC (Qiao et al., 2001). Although the PN skin group exhibited a lower pH at 2 hr and after frozen storage, the values remained within the optimal range of 5.7 to 6.1 (Beauclercq et al., 2022; Bihan-Duval et al., 2018), suggesting that WHC was not adversely affected.

Color

At both 24 hr postmortem and after 2.5 mo of storage, no changes were observed in any of the CIE color parameters (L*, a*, b*) between the PN skin and control groups (Table 3, P > 0.05). This indicates that inclusion of 5% PN skin in the broiler diet did not affect meat appearance, a key quality trait for consumer acceptance. Lightness (L*) values, which reflect the brightness or pale appearance of meat, were similar between the control and PN skin groups at 24 hr (62.55 vs 62.75, respectively; P > 0.05). After 2.5 mo of storage, both groups exhibited a decline in L* values (58.55 vs 59.69, respectively; P > 0.05) likely due to pigment oxidation and moisture loss over time, a common phenomenon in stored meat (Ciobanu et al., 2016; Petracci and Fletcher, 2002). However, the L* values remained consistent, indicating no adverse impact of PN skin on color lightness. Redness (a*), which is critical for consumer perception of freshness, remained negative or close to zero for both groups at 24 hr, suggesting low myoglobin oxidation. After 2.5 mo, a* values were not different between the control and the PN skin group (−0.04 vs 0.4, respectively; P > 0.05).

Over the storage period, yellowness (b*) values did not change between the control group and PN skin group (7.25 vs 7.64, respectively, at 24 hr, 8.93 vs 8.99, respectively, after 2.5 mo, P > 0.05). Nevertheless, the steady b* values suggest no negative effect from dietary PN skin inclusion. Overall, our results indicate that dietary supplementation with peanut skin did not adversely affect instrumental meat color or appearance under the conditions tested.

Proximate analysis: Near-infrared reflectance

As shown in Table 4, no notable changes were found in fat, moisture, protein, collagen, or ash content between treatments (P > 0.05). These results confirmed that the inclusion of 5% PN skin does not alter the fundamental chemical composition of broiler breast muscle. Fat content remained consistent between the control group and the PN skin group (2.26 vs 2.27, respectively; P > 0.05). Likewise, moisture content showed negligible variation for the PN skin group and the control group (75.44 vs 75.65, respectively; P > 0.05). Protein levels were identical between the groups, indicating that dietary peanut skin did not alter protein deposition in muscle tissues. Collagen content, a key factor affecting meat tenderness, was also unaffected, showing uniform values for the PN skin and control groups, respectively. Similarly, ash content, reflecting the mineral composition of the meat, remained stable across treatments. These finding aligns with previous literature using different phenolic compounds as well as peanut skin powder supplemented diet (1.0 g, 2.0 g, and 3.0 g PS /kg basal diet) for broilers, which similarly reported no negative impact on carcass composition, meat moisture, protein, or fat content (Starčević et al., 2015; Abd El-Hack et al., 2017). Additionally, the unaltered protein and fat levels observed in the present study align with previous findings using polyphenol-rich feed additives for different animals. For instance, Alfaia et al. (2022) found that monogastric animals′ supplemented with grape pomace exhibited no change in the crude protein or fat content of meat. Similarly, Kara et al. (2016) reported that Japanese quail fed up to 2.5 g/kg green tea catechins showed no adverse effects on live weight or proximate meat composition, despite improvements in water-holding capacity and antioxidant markers. Overall, our meat composition analysis indicates that 5% PN skin supplementation did not alter proximate composition.

Texture analysis

The effect of 5% ground PN skin inclusion on broiler breast meat texture was evaluated using the Blunt Meullenet–Owens Razor Shear (BMORS) method for both raw and cooked samples (Table 5), as well as the W–B method for cooked meat (Table 5). For raw meat (Table 5), peak shear force, shear energy, and the number of peaks did not change between 2 treatments (P > 0.05). However, for cooked meat (Table 5), broilers fed PN skin diet showed a greater peak force (23.58 N vs 13.15 N, respectively; P < 0.05) and shear energy (216.98 N·mm vs 153.39 N·mm, respectively; P < 0.05) than the broilers from the control diet, indicating firmer cooked meat in the PN skin group. Warner–Bratzler shear values (Table 5) also reflected consistent peak shear force between PN skin and control groups (28.10 N vs 25.66 N, respectively; P > 0.05). Many recent studies have reported shear force values for cooked broiler breast meat typically ranging between 10 N and 30 N (Bowker and Zhuang, 2019; Cai et al., 2018; Sun et al., 2022; Zhang et al., 2021; Pang et al., 2024; Sun et al., 2021). This range can vary based on several factors, including cooking temperature and method (Fabre et al., 2018), sample size and preparation (Zhuang and Savage, 2009), instrument setting, and so on. Although the instrumental texture of cooked meat from the PN skin group was firmer in our study, the shear force values are still within the range of cooked chicken meat texture according to the previous works.

Water-holding capacity

The water-holding capacity (WHC) of broiler breast meat, as reflected by both cook loss and thaw loss, was significantly influenced by dietary treatment (Table 6). Broilers fed a diet supplemented with 5% ground PN skin exhibited a notable reduction in both cook loss and thaw loss compared to the control group. Specifically, the average cook loss in the PN skin group was significantly lower than that of the control group (Table 6, 21.73% vs 25.58%, respectively; P < 0.05), indicating improved retention of moisture during thermal processing. Similarly, thaw loss was markedly reduced in the PN skin group compared to the control (Table 6, 12.72% vs 18.74%, respectively; P < 0.05). The observed improvement in WHC among broilers receiving the PN skin diet may be linked to subtle changes in muscle structure, fat distribution, collagen properties, or protein stability, even though the overall meat composition remained largely unchanged. Previously, Abd El-Hack et al. (2017) evaluated dietary supplementation with peanut skin powder (1.0 g, 2.0 g, and 3.0 g PS /kg basal diet) Cobb broiler chickens and found notable improvements in carcass traits and meat quality. Notably, they reported enhanced WHC, alongside lowered abdominal fat and improved sensory appearance. This reduction in moisture loss is consistent with findings in swine studies where dietary phenolics, especially from natural plant sources, enhanced WHC by stabilizing muscle proteins and inhibiting oxidative damage to cellular membranes (Muzolf-Panek et al., 2024). Comparable improvements in meat quality have been observed with other dietary phenolic compounds, for example, supplementation with soybean isoflavones (40 mg/kg) in broiler diets and quercetin (42 ppm) in cattle feed has been shown to enhance water-holding capacity and elevate postmortem muscle pH (Jiang et al., 2007; Valenzuela-Grijalva et al., 2017).

Lipid oxidation

The thiobarbituric acid reactive substances (TBARs) assay, which measures malondialdehyde (MDA) as an indicator of lipid peroxidation, showed no difference between treatments (Table 6, P > 0.05). MDA levels were 1.16 mg/kg in the PN skin group and 1.23 mg/kg in the control (P > 0.05). The absence of statistically significant differences may be attributed to the relatively low inclusion level (5%) of peanut skin and the extended 7 mo storage period, during which lipid oxidation processes often reach a plateau (Al-Dalali et al., 2022; Feng et al., 2022).

Sarcomere length: SEM

The sarcomere lengths of both treatment groups did not differ (Table 6, 1.17 μm vs. 1.15 μm, respectively; P > 0.05), which suggests that dietary PN skin supplementation did not alter the integrity of myofibrillar architecture during extended frozen storage. Preservation of sarcomere length during frozen storage reflects minimized postmortem shortening and structural degradation, which are often exacerbated by oxidative processes and proteolysis (Schulte et al., 2019). The antioxidant properties of phenolic compounds may contribute to the stabilization of muscle structure by mitigating oxidative damage. However, no difference has been observed for the inclusion percentage of PN skin considered in our study. The sarcomere length values for both diets are on the shorter end of the typical range reported for frozen broiler breast muscle, which usually falls between 1.6 and 2.0 μm (Wattanachant et al., 2005; Soglia et al., 2020; Sabikun et al., 2019). The comparatively shorter sarcomere lengths may be due to postmortem rigor contraction, cold shortening, or early-onset muscle stiffness, all of which are known to reduce sarcomere length even in uncooked samples (Ertbjerg and Puolanne, 2017). Overall, the lack of differences in sarcomere length between PN skin and control groups indicates that the inclusion of peanut skin in broiler diets does not adversely affect muscle structural integrity during long-term frozen storage, supporting its potential use as a functional feed additive.

Muscle myopathy

No significant differences were detected in WB, WS, or spaghetti meat (SM) scores between treatments (Table 7, P > 0.05). Toomer et al. (2019) evaluated the effects of a high-oleic peanut-enriched diet on the quality and sensory characteristics of broiler chicken meat. Their findings indicated no differences in WS and WB severity between the treatment groups (P > 0.05). This pattern aligns closely with the trends observed in our study regarding both WB and WS incidence. De Castilho Heiss et al. (2025) found that supplementing diets with a commercial polyphenol blend significantly reduced WB severity, though WS incidence remained unaffected. In general, high-energy and high-protein diets promote accelerated muscle hypertrophy, which increases oxidative stress in broilers and also increases the risk of WB, WS, or SM due to rapid growth (Zhang et al., 2011). However, antioxidants such as PN skin mitigate oxidative stress and can potentially reduce the risk of myopathy incidences (Wang et al., 2025). The distribution of macroscopic myopathy scores observed in the current study further supports these trends (Table 8). In the control group, 5 out of 15 fillets (33.3%) showed no evidence of any myopathy (WS0, WB0, SM0), with the remaining fillets primarily exhibiting mild WS or WB scores. Similarly, in the PN skin group, 4 fillets (26.7%) were free of myopathy, and most of the remaining samples exhibited only mild to moderate lesions. Overall, both groups had a relatively low prevalence of severe myopathy, with most fillets falling within the lower end of the scoring scales.

Sensory evaluation

Sensory evaluation (Table 9) revealed that most texture-related attributes, including juiciness, tenderness, cohesiveness, gumminess, and chewiness, did not differ significantly between the control and PN skin groups (P > 0.05). However, springiness was lower in the PN skin group compared to the control group (5.34 vs 6.13, respectively; P < 0.05), indicating a reduction in the meat’s elastic rebound capacity. Despite this, all attributes fell within moderate sensory intensity ranges (5-7), indicating general acceptability of meat texture even with the dietary inclusion of peanut skin. Moreover, since cohesiveness and gumminess scores did not change, the overall structural integrity of the meat was likely preserved, allowing it to remain palatable despite lower springiness. Furthermore, sensory flavor assessments (Table 9) showed no variations in brothy, chicken-like, rancid, cardboardy, metallic, or off-flavor attributes, reinforcing that the dietary addition of peanut skin did not impart adverse organoleptic effects. These results are supported by earlier work from Toomer et al. (2019), where broiler diets incorporating high-oleic peanut products did not alter sensory flavor profiles. Similarly, basic taste perceptions (Table 9), including sweetness, umami, sourness, saltiness, and bitterness, remained statistically unchanged (P > 0.05). Overall, these results suggest that while dietary peanut skin inclusion slightly reduced springiness, it did not compromise the broader sensory attributes of the meat, preserving favorable textural and flavor characteristics and supporting its feasibility as a functional feed ingredient. The results in this study also agree with earlier findings that phenolic compounds do not show a significant adverse impact on taste, flavor, or texture when included at moderate dietary levels (Kalogianni et al., 2020). Given that sensory traits were largely unaffected, PN skin inclusion should depend on a clearly demonstrated cost-benefit on a commercial scale.