Classification, quality characteristics, sensory perception and texture prediction of wooden breast myopathy in broilers from Argentina

Abstract

The objective of this study was to characterize the wooden breast (WB) myopathy in the poultry industry, and establish degrees of severity by analyzing the composition and important characteristics of meat quality. 175 chickens from the COBB-500 commercial line were analyzed, and three WB categories were established: normal, moderate, and severe. The prevalence was: 4.00% severe, 46.29% moderate, and 49.71% normal, and it was affected by sex and weight. The WB characterization was carried out on 7 breasts of each WB degree. A decrease in protein and ashes, an increase in fat and loss of water by dripping, and color changes were observed as the degree of severity advanced. The texture was evaluated by instrumental, sensory, and image analysis techniques. Severe samples showed alterations in compression test, cohesiveness and juiciness, together with significant differences on the parameters, “contrast” and “energy”. A characterization of WB myopathy was achieved for the first time in Argentina. The decline in meat quality could lead to industrial losses. Image analysis proved to be a promising technique for differentiating the severity of WB myopathy in raw chicken. Differences between raw and cooked samples were detected, thus both types of meat should be studied in detail.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13197-022-05608-9.

Introducción

In recent years, reports of breast myopathies from broilers have increased, indicating that they have become an important issue for the meat industry due to the economic losses generated (Zanetti et al. 2018). Myopathies prevalence has been reported in the European Union, US, Canada, Australia, New Zealand, Japan, and Brazil. In Argentina, these myopathies are evidenced by customer complaints, as well as by the veterinary inspection of several poultry companies. However, there are no scientific reports regarding myopathies on broilers from Argentina.

Among these conditions, the wooden breast (WB) is widely known. Sihvo et al. (2014) performed the first complete pathological description of WB. This condition manifests itself in broiler chickens, showing a hardened and pale muscle, in which the pectoralis major, and occasionally the pectoralis minor are affected (Sihvo et al. 2018). This lesion appears as a local focus at approximately 2 weeks of age, and then it spreads as a generalized fibrotic lesion characterized by the replacement of muscle with connective tissue, water, and fat (Velleman et al. 2018; Brothers et al. 2019). Factors such as genotype, sex, diet, and slaughter weight could also be associated with this myopathy (Zanetti et al. 2018).

Owens (2016) reported that incidence cases of WB disease can affect 10–40% of the production. These alterations directly affect some sensory attributes of texture, such as tenderness and juiciness due to the loss of the capacity to retain water (Soglia et al. 2017; Xing et al. 2017). Overall, this disease affects the chicken breasts in several aspects like appearance, technological and nutritional quality, as well as consumer acceptance.

To assess the prevalence of this myopathy, evaluation methods based on manual palpation and visual inspection are used (Tijare et al. 2016). However, due to the increase in severity observed in recent years, many different studies have been reported (Petracci et al. 2019). Additionally, several authors reported a significant reduction in protein, and increased humidity and lipid levels in the WB muscles compared to their normal counterparts (Soglia et al. 2017; Wold et al. 2017). Samples with a severe degree of WB myopathy produce more exudate, which can be detrimental to the processing of chicken meat (Oliveira et al. 2021). Many works worldwide have carried out an exhaustive analysis of the changes in the composition and quality of WB meat, however, it is important to include other studies like image analysis that may allow a quicker classification of WB degree. In this sense, texture parameters can be calculated based on the grey level co-occurrence matrix (GLCM) applied to images (Pieniazek et al. 2018).

It is evident that in the last decades, the preferential consumption of chicken meat has increased over other types of meat (Petracci et al. 2019). In Argentina, various commercial chicken lines are employed, being the Cobb 500 an efficient fattening chicken, with high feed conversion and low cost of breeding, which is important at a productive level for developing countries.

Producing a high-quality meat supply is an important objective for any exporting country, therefore, this study was designed to characterize WB myopathy in the poultry industry in Argentina, in which no studies have been previously recorded, and determine degrees of severity, performing a study of the composition and important characteristics of meat quality, as texture, employing different analytical techniques.

Materials and methods

Training and criteria classification

A training procedure of a professional team from the poultry industry Bonnin Hnos. (Colón, Entre Ríos, Argentina) was employed to achieve a unified classification criteria for WB myopathy. For the training, 623 muscles of the left pectoralis major were taken from broiler chickens (commercial line COBB-500). The training lasted three months, and included palpation and visual observation techniques (Sihvo et al. 2014; Petracci et al. 2019), together with the use of a 200 g weight applied for 30 s on the cranial part of the muscle (Owens 2016).

3 categories for WB myopathy were established (Fig. 1):

• Normal No detectable hardness, flexible, normal looking. The weight compresses the sample and leaves a very pronounced mark.

• Moderate punctual and marked rigidity in the cranial area, remains flexible. The weight compresses and leaves a mark.

• Severe Extreme rigidity in all muscles, hardness from the cranial to caudal region, presence of viscous liquid. The weight does not compress or slightly compressed and does not leave a mark.

Fig. 1.

Classification in three categories of Wooden Breast (WB) myopathy

Sampling and classification of myopathy WB

In the summer season, 175 chickens from the COBB-500 commercial line were identified by numerical labels, taking into account animal health procedures. Traceability was applied to identify each animal throughout the slaughter process, taking individual weight, sex, and age (approx. 54 days) as control variables.

The left supreme of the 175 chickens was separated and classified using the methodology described in point 2.1. The severe category was found in 7 chickens of the 175 analyzed, therefore, for the rest of the analyses, 7 breasts of each WB degree were transported refrigerated to the laboratory.

The seven breasts of each category were divided into two, the left and right supremes. The left supremes were used to carry out the color and pH analysis, then, 3 pieces were cut from the different areas (cranial, middle, and caudal) of each sample to perform the assay of water loss by dripping, and finally, the remaining meat was chopped to carry out the analyzes of proximal composition. The right supremes were employed for the determination of shear force, the sensory analysis, and the electron microscopy studies, for which the pieces were cut as indicated in the corresponding methodology section.

Proximate composition

Proximate composition (moisture, protein, fat, and ash) of fillets was determined on the homogenized sample following AOAC methods: moisture (AOAC, 934.01, 2000); ashes (AOAC, 24.009, 2002), fat (AOAC, 960.39, 1990), and protein (AOAC, 976.05, 1990).

Meat color, pH measurements, and drip loss

The color was determined on the ventral surface of the breasts using a colorimeter (Hunter Lab MiniScan EZ, Virginia, US) (Aguirre et al. 2018). The pH value was determined 24 h after slaughter to estimate the intensity of acidification drop (pHu) (AOAC, 945.27, 2002; Petracci and Baéza, 2011).

The loss of water by dripping was measured according to Velleman et al. (2018). For this assay, samples were taken in triplicate from the muscle of each degree of WB severity (middle, caudal and cranial areas). Drip loss was determined at 24 and 48 h after slaughter.

Texture studies

Cooking

Breasts of each grade were placed in individual bags suitable for cooking, they were vacuum sealed and immersed in a water bath until they reached an internal temperature of 75 °C (Soglia et al. 2017). Internal temperatures were monitored in the thickest part of each fillet with a handheld digital thermometer.

Effect of wooden breast (WB) on instrument texture

Warner–Bratzler shear force

The Warner–Bratzler shear force (WBSF) test was performed following the technique reported by Mudalal et al. (2014). Samples were prepared as follows: A portion of the cranial area of each right supreme (7 severe, 7 moderate, and 7 normal samples) was cut perpendicularly in the direction of the muscle fibers, obtaining a parallelepiped of (4 cm × 1 cm × 1 cm), and was kept refrigerated at 4 °C until the analysis. The measurements of the maximum shear force (N) were carried out on a Universal Testing Machine Instron 3342 (Massachusetts, USA) connected by a computer to the Instron Bluehill Material Testing software. A 500 N cell was used, and the analysis was done at a speed of 8 mm/s.

Compression test

A cylindrical of raw meat (size: height 10 mm and diameter 23 mm, weighing approximately 5 g) was cut from the caudal part of each fillet. It was compressed to 40% of the initial height using a Universal Testing Machine Instron 3342 (Massachusetts, USA). The test speed of the probe was 1 mm/s. The compression value was recorded as the maximum force necessary to compress 40% of the initial height of the sample and was expressed in N following techniques reported by Mudalal et al., (2014).

Descriptive sensory analysis

Sensory evaluation of the chicken breast was carried out by eight judges (5 females and 3 males, 25–45 years old). All of them were members of the texture profile panel from the Facultad de Bromatología, Universidad Nacional de Entre Ríos (Archaina et al. 2019). In order to describe the sensory properties, the sensory profiling method was applied (ISO 13299, 1988), and it was made up of two stages:

An initial stage allowed validating the panelists and familiarizing them with the chicken meat using seven parameters: hardness, density, fracturability, juiciness, cohesiveness, adhesiveness, and elasticity (Table S3), employing semi-structured scales (Archaina et al. 2019; Aguirre et al. 2018; Hough et al. 1994).

In the second stage, the assessors carried out the sensory analysis of the chicken breasts with different classification criteria (normal, moderate, and severe) according to the area (cranial middle, and caudal). In this phase, 9 breast samples (three for each category, which were divided in the three different areas) were evaluated. The panelists received samples cut from the superficial part of the breast (a 1.5 cm³ cube), they had to assess the intensity of each parameter, and then rate them on semi-structured scales.

Scanning electron microscopy combined with image analysis techniques

Scanning electron microscopy

Image acquisition was carried out using a SEM Zeiss Supra 40 microscope (Zeiss, Oberkochen, Germany) with 1000X magnification. Brightness and contrast were maintained during the image acquisition process (Pieniazek et al. 2018).

These studies were performed on samples from the cranial areas of “normal” and “severe” breasts. This area of the breast was selected because more texture differences were evidenced upon WB development (Brambila et al. 2017). The samples were taken by making a cut on the surface of the ventral cranial side, then, they were subjected to a dehydration process with alcohol/acetone, placed on an aluminum support, and covered with platinum nanoparticles, keeping them in a vacuum desiccator until use (Pieniazek et al. 2018).

This analysis was performed for both raw and cooked breasts (using the cooking method explained above in point 2.5.1).

Texture image analysis

The texture properties were related to a set of gray level co-occurrence probability (GLCM) distribution matrices for a given image. GLCM shows the probability that a pixel of a particular gray level will occur at a specified direction and distance (d = 1) from its neighboring pixels. 5 characteristics of texture, correlation (COR), energy (ASM), contrast (CON), homogeneity (HOM) and entropy (ENT) were determined (Pieniazek and Messina, 2016). The CON parameter is related to hardness, the ENT parameter is used to characterize the roughness, the ASM parameter shows the homogeneity, and the HOM parameter measures the uniformity/smoothness (Pieniazek and Messina 2016). The five image texture features were calculated using MATLAB 8.4 (The Math Works, Inc., MA, USA) using the equations of Pieniazek et al. (2018).

Statistical analysis

Data analysis was carried out with ANOVA and a Tukey multiple range tests using the statistical package Statgraphics software®. Differences were considered significant at a P value of p < 0.05 and the results are presented as the mean and its standard deviation.

The integration of the textural image parameters was carried out by means of a principal component analysis (PCA) on samples of the cranial area of normal and severe raw and cooked breasts.

Results and discussion

Sampling and classification of myopathy WB

The training process developed in the poultry industry allowed a successful WB myopathy classification in 175 chickens (91 females and 84 males). The prevalence of WB found in this study was 4.00% severe, 46.29% moderate, and 49.71% normal. These values were higher than those reported by the Chinese poultry industry, which presented 30.8% for moderate and severe WB samples (Xing et al. 2020). Petracci et al. (2019) indicate that approx. 50% of chicken breasts in Italy, France, Spain, and Brazil are affected by WB, being these results similar to the ones obtained in the present study. Chicken breasts with myopathies appear in all countries where fast-growing hybrids are grown, and when the number of individuals is higher than the amount that the chicken industry should have.

The highest degree of severity was obtained in male birds (Fig. 2a) and supreme weights between 300 and 600 g (Fig. 2b). Regarding the moderate degree, there were breasts from both genders and the majority with a weight range between 300 and 400 g. These results coincide with the fact that breast weight, correlated with body weight, is one of the main factors involved in the appearance of WB myopathy (Xing et al. 2020). Furthermore, Dalle Zotte et al. (2017) established that breast fillets affected by this myopathy were heavier than control ones (505 vs. 377 g, respectively; P < 0.001).

Fig. 2.

Frequency of Wooden Breast (WB) myopathy according to sex (a) F = female M = male, and individual weight (b)

Proximate composition

As expected, the appearance of WB had an effect on some parameters related to the chemical composition of raw meat. Moisture did not show significant changes among the three WB categories. However, a significant decrease in protein and ashes content, and an increase in lipid content were found in severe breasts (Table 1). Some authors have argued about the nutritional value of WB breasts, and have highlighted that while there are no differences in the fatty acid profile, alterations are observed in the level of myofibrillar proteins, which may influence the capacity of the meat to retain water, emulsify lipids and form stable gels (Petracci et al. 2019). Other authors have also reported a decrease in protein and an increase in fat values with the appearance of WB myopathy. Cai et al. (2018) reported a decrease in protein content, of 23% for normal breasts and 21,7% for affected breasts, and an increased lipid content of 1.9% for WB breasts versus 1.2% for normal samples. Soglia et al. (2017) also established that severe wooden breasts showed lower protein content than normal. Wold et al. (2017) indicated that moderate and severe breasts had lower protein content than normal breasts, on the contrary in this study both normal and moderate breasts were not significantly different. For ash content, a significant decrease was found in breasts with severe WB compared to normal and moderate breasts (Table 1). As in the protein parameter, no significant differences were found between normal and moderate breasts. The found values are in the range of those reported by Soglia et al. (2017), who mentioned that the ash content in breasts with WB was 1.26% and for normal breasts 1.37%.

Table 1.

Effect of Wooden Breast (WB) on composition of raw meat

Parameter	Wooden Breast score
	Normal	Moderate	Severe
Moisture (%)	75,58 ± 2,76ab	73,58 ± 2,42a	76,99 ± 2,58b
Protein (%)	23,84 ± 0,54a	23,61 ± 0,57a	21,00 ± 0,40b
Fat (%)	1,37 ± 0,65a	1,43 ± 0,51a	2,00 ± 0,55b
Ash (%)	1,33 ± 0,01a	1,39 ± 0,00a	1,23 ± 0,03b

Data are presented as means ± standard deviations. Mean values (row) with different lowercase letters are significantly different (p < 0.05)

Meat color, pH measurements, and drip loss

Regarding color, significant differences were found between the different degrees of severity of the WB, particularly the L* value of the cranial part increased as the degree of severity progressed, which denotes greater luminosity or whiter breasts. These results are in accordance with the reports of Mudalal et al. (2014) and Dalle Zotte et al. (2017). Probably the compositional changes that occurred in the muscle caused the increase in luminosity. When analyzing the b* value, an increase was observed in the cranial portion as the gravity of the WB advanced (which refers to more yellow areas), otherwise, in the caudal portion b* decreased (Table 2). Petracci et al. (2013) showed that fillets affected by WB are more yellow than those unaffected, probably due to fibrosis that forms in the tissues. Concerning the a* value, no changes were found between different WB degrees.

Table 2.

L*, a* and b* values ​​obtained for each category of the Wooden Breast myopathy score considering the different measurement areas (cranial, middle and caudal)

Score	Cranial			Middle			Caudal
	L*	a*	b*	L*	a*	b*	L*	a*	b*
Normal	63,80 ± 2,97A.a	6,01 ± 1,16 A.b	14,35 ± 2,09A.a	63,37 ± 2,63A.a	5,21 ± 0,72A.ab	14,15 ± 2,25A.a	68,09 ± 2,71A.b	4,25 ± 2,42A.a	14,55 ± 2,04B.a
Moderate	65,29 ± 2,00B.a	7,20 ± 1,19B.b	15,88 ± 2,66B.a	66,50 ± 2,21A.a	5,58 ± 1,30A.a	13,58 ± 4,51A.b	67,44 ± 4,38A.a	5,05 ± 1,81A.a	12,38 ± 2,51A.c
Severe	66,99 ± 2,32C.b	6,59 ± 1,93AB.b	17,04 ± 2,25C.b	64,76 ± 3,82A.a	6,03 ± 1,81A.b	13,66 ± 2,07A.a	65,89 ± 3,47A.ab	4,68 ± 1,54A.a	12,64 ± 1.74A.a

Data are presented as mean ± standard deviations. Mean values with different capital letters are significantly different (p < 0.05) by the Tukey test in the column. Mean values with different lowercase letters are significantly different (p < 0.05) by the Tukey test in the row

Concerning pH, Oliveira et al. (2021) reported that the samples with WB are characterized by having higher pH values. The values obtained for pH at 24 h postmortem were 5.93 ± 0.13 (normal), 6.01 ± 0.26 (moderate) and 6.25 ± 0.03 (severe). Although a trend to increase pH was observed along with the progress of WB severity, there were no significant differences. These values are close to those informed by Kuttappan et al. (2017), who reported a pH of 5.93 at 24 h for normal breasts, and 6.15 for WB breasts.

The loss of water by dripping analysis showed a higher drip loss as the degree of WB severity progressed, both at 24 h and 48 h, in all areas of the analyzed breasts. Furthermore, the greatest loss was found in the caudal area (Table S1). Dalgaard et al. (2018) also reported increased drip loss values in severe breasts, although the area from which they obtained the sample is not established. It is relevant to mention that the pH values higher than 6 are above the isoelectric point of the meat proteins, therefore, a greater water retention capacity would be expected. As this behavior does not occur, some authors attributed the increased release of water to an extensive loss of membrane integrity, and to the presence of a thin layer of fluid viscous material on the wooden chest (Sihvo et al. 2014).

Texture studies

Effect of wooden breast (WB) on instrument texture

Warner–Bratzler shear force

In the Warner–Bratzler maximum shear force test, the samples of the cranial area of normal, moderate, and severe cooked breasts did not show significant differences (Table S2). These results agree with previous studies where no obvious differences were found in the shear force between normal and WB muscles (Cai et al. 2018; Soglia et al. 2017).

Compression test

Raw WB severe fillets exhibited significantly higher compressive force than normal and moderate fillets (35,54 vs. 8,63; 14,65 N, respectively). These results are shown in Table S2. This high increase in the compression force in the instrumental analysis of the breast with myopathy shows a very noticeable hardness in the raw meat, these results coincide with those of Mudalal et al. (2014). Therefore, the weight test that was used for myopathy characterization may be useful in establishing the presence of wood abnormality in raw breast fillets.

Trained panel study

The panelists analyzed the three degrees of severity of WB myopathy in the different areas of each breast. The main changes detected by the sensory study were observed in the cohesiveness parameter, which was lower for the samples classified as "severe", and in the juiciness parameter, which was higher for the severe degree (Table S4). The rest of the studied attributes (hardness, denseness, springiness, fracturability, and adhesiveness) did not present significant differences among the different WB degrees. The higher juiciness perceived by the panelists could be related to an a loss of muscle fibers integrity of severe breasts. Regarding the cohesiveness decrease, Petracci and Cavani (2011) established that an emerging quality problem in poultry is associated to the poor meat cohesion, caused by immature intramuscular connective tissue which is related to the very early slaughter age of modern chicken strains.

Scanning electron microscopy combined with image analysis techniques

Figure 3 shows SEM images of the cranial area of the raw and cooked normal and severe breast surfaces. Image analysis using the GLMC technique showed significant differences for the CON and ASM parameters when comparing the two degrees of WB severity. In the case of raw samples, the severe WB degree presented higher contrast values compared to those of normal breasts, suggesting that they have higher hardness, these results are in agreement with the compression force test (Table S2). With respect to ASM parameter, severe samples presented lower values than normal ones, suggesting that the images were less homogeneous. In cooked samples, both severe and normal samples showed higher CON values than those observed in raw breasts, so that the hardness would increase to the cooking process, being higher for normal samples. Soglia et al. (2017) suggested that in cooked wooden breast the texture changes are related to the denaturation and solubilization of the thermally labile collagen, which occurs at temperatures between 53 and 63 °C.

Fig. 3.

Scanning electron micrographs of chicken breasts obtained from the cranial surface at 1000 × magnification (Coding: A: Raw normal sample. B: Raw severe sample. C: Cooked normal sample. D: Cooked severe sample). The table shows the parameters of the concurrency matrix

Principal component analysis

A principal component analysis was performed (Fig. 4). Principal Component 1 (CP1) represented 85.8% of the total variation between samples. CP1 was mainly explained by the attributes: HOM, COR, and ASM, on the right side, and these parameters could be used to differentiate normal raw samples from severe raw samples. The normal raw samples (R0) presented higher HOM and ASM, which denotes smoother and more homogeneous images. Towards the left side, the ENT and CON parameters could be used to differentiate the cooked samples, and the values were higher for the normal degree (C0). Soglia et al. (2017) stated that the differences in texture between normal breasts and those affected by WB were detected mainly in raw meat and not in cooked meat, so this technique would allow a new alternative to differentiate between severe and normal breasts, both raw as well as cooked, using in each case the parameters that differ in this graph.

Fig. 4.

Principal Component Analysis (PCA). CP1 and CP2 were obtained from the image texture analysis (Coding: ASM = energy COR = correlation, ENT = entropy, CON = contrast, HOM = homogeneity) of breasts samples from cranial area (C0: normal cooked, C2: severe cooked, R0: normal raw, R2: severe raw)

Conclusion

In this study, it was possible to perform for the first time a characterization of wooden breast myopathy (WB) in broilers from Argentina. This work indicated that the prevalence of WB myopathy in broilers during the summer was close to 50% (considering moderate and severe degrees of severity). The severity of the myopathy increased with the broiler’s weight and was predominant in male birds. Additionally, composition changes occurred, being the most relevant the decrease in protein content (23.84% normal, compared to 21.00% severe) and increased drip loss of water in affected breasts, which can generate losses at the industrial purposes.

Regarding sensory studies, no great differences were detected at this level, however, the breasts turned out to be less cohesive and juicier, which could be related to a loss in the structure of the meat matrix.

Texture studies showed that the hardness analysis performed on cooked meat using both instrumental (Warner–Bratzler shear force), and sensory (trained panel) methods, did not show relevant differences. However, when analyzing raw breasts, both methods showed a hardness increase in severe WB samples. Also, the images analysis showed a very different behavior between cooked and raw meat.

Image analysis in conjunction with GLCM proved to be a promising technique for characterizing and differentiating WB breasts from normal ones. These results comprise a starting point to find the possible application of these techniques within the poultry industry, and the next step to simplify the methodology would be to employ digital camera images.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviation

WB

Wooden Breast

Authors' contribution

TAL: Writing—Original Draft; Methodology; Validation; Formal analysis; Investigation. Project administration. PAG: responsible for the industry. Participation in the classification criteria. LT: participation in physicochemical analysis. FP: specialist in image analysis and matrix application for texture analysis. CS: Writing—Review and Editing; Conceptualization. Supervision. NS: Funding acquisition; Resources; Visualization; Writing—Review & Editing; Supervision. Project administration.

Funding

This study was funded by Agencia Nacional de Promoción Científica y Tecnológica (PICT 2018/01054) and Universidad Nacional de Entre Ríos (PID Novel 2021/9118).

Availability of data and material

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Code availability

Not applicable.

Declarations

Conflict of interest

The authors declare that they have no competing interests. As corresponding author, hereby I declare that the coauthors have accepted to be represented by myself. All the authors certify that they have NO affiliations with or involvement in any organization or entity with any financial interest, or non-financial interest in the subject matter or materials discussed in this manuscript.

Ethical approval

Not applicable. Broilers were slaughtered according to a commercial slaughter process. No experimental treatments or scientific procedures were conducted on living animals.

Consent to participate

All participants sensory panel were informed of the purpose of the research, and the voluntary nature of participation, and read the free and informed consent form to participate in the study. All information gathered about during this study was treated with full confidentiality.

Consent for publication

Not applicable.