Introduction

A relationship between horses and humans has existed for around 12,000 years. There is evidence that, before horse domestication, their first use was as a food source (^{Hintz, 1995}). Currently, meat horse production represents 0.24 % of total world meat production (~740 thousand tons), and has remained without significant changes over the last decades (^{FAOSTAT, 2018}).

The estimated per capita worldwide consumption of equine meat is 0.10 kg, traditionally the main consumers are European countries, although there are those such as Mexico which stands out in horse meat production and export; although, the acceptability of this meat by Mexican consumers is still unknown (^{Gill, 2005}). Even though horse meat has properties that can be considered as a “healthy meat” (^{Lorenzo et al., 2010}), is not popular in most countries as food for humans. It has been corroborated that regular consumption of horse meat significantly reduces serum levels of total cholesterol and low- density lipoproteins, while increasing polyunsaturated fatty acids levels which it’s improving people’s health (^{Del Bó et al., 2013}). Currently, a major concern for consumers is to demand that foods have passed a rigorous classification and evaluation process before being introduced to the market. Even though in Mexico there are well-defined regulations and procedures to classify beef, sheep, pork and poultry meat (taken into consideration age and weight at slaughter, sex, and body conditions score, among others), in the case of horses there is little information in this regard, even though the amount of horses being slaughtered is increasing. Hence, the establishment of regulation protocols becomes more relevant because a large part of the horses slaughtered in Mexico come from outside the country. For this reason, the objective of this study was to determine the effects of sex, age and weight at slaughter, and body condition score on culled quarter horse carcass characteristics. Due to the important proportion of horses from USA that are sacrificed in Mexico, origin (USA and Mexico) was include as factor in the study.

Material and methods

Results and discussion

Overall, the average slaughter weight was 364.6 ± 90 kg. Adult quarter horse can weight from 400-570 kg, the lower weight registered in this study was mainly due by the lower BCS and weight loss during transportation and management. A 74.5 % of slaughter horse were from the United States of America (USA). This large proportion of horses coming from the USA for slaughter in Mexico is due to the closure of horses slaughter plants in the United States by the law “American Horse Slaughter prevention”, which prevents the slaughter of equines for human consumption in that country (112th Congress act 2011-2012). Accordingly, USA horse exports for slaughter increased by 148 and 660 percent to Canada and Mexico, respectively (Forrest, 2017, The Atlantic Journal). Almost 50 % of slaughtered horses were 8 years old or older, and had lower BCS. This can be explained since currently horse meat producers in USA have decreased. As expected, sex proportion was similar (51.5 % female vs 48.5 % males).

Horses from USA had a heavier (P<0.05) SW (389 vs 283 kg), greater leg circumference (4 %) and compactness index (4.4 %), heavier HCW (215 vs 206 kg), and greater carcass yield (59.02 vs. 57.29 %) without difference on REA nor BFT when compared to Mexican horses (Tables 1 and 2). This could be due to the different production systems used in each country; since these are not well defined they cannot be evaluated as these animals fulfill different functions, not as food animals. ^{Franco et al. (2013)} mentioned that the differences in the carcass morphometric measurements are related to the different genotypes, and not by feed, when are raised in similar production systems. Either way, previous reports had denoted that carcass performance in horses is highly variable. For example, ^{Lanza et al. (2009)}, similar to our findings, registered a 59 % carcass yield in Sanfratellano and Haflinger horses. However, carcass yields up to 67 % have been reported for Burguete horses with ages between 16 and 24 months and a SW close to 400 kg. This is in agreement with carcass yield of 65 and 68 % registered for Burguete and Hispano-Breton horses slaughtered at 18 months of age with a SW around of 400 kg (^{Juárez et al., 2009}). In opposite, ^{Franco et al. (2013)} reported carcass yields of 50 and 52 % in Gallego de Monte breed horses and their crosses with Hispano-Breton. Furthermore, a carcass yield as low as 47 % was reported 15 month-old Gallego de Monte horses with a SW below 200 kg. As discussed below factors such as the BCS and SW are important to explain carcass yield in horse.

Sex did not affect any characteristic evaluated (Table 3). This finding coinciding with those reported by ^{Sarriés and Beriain (2005)} and ^{Lorenzo et al. (2013b)}. These researchers reported that there is no effect of sex, or an interaction with any other factor for carcass characteristics in horses. It is important to note that in our study all males were castrated, which would further minimize any possible difference attributed to gender

The effects of age at slaughter (AS) on carcass characteristics are shown in Table 4. As expected, AS significantly affected (P<0.05) carcass characteristics. Increasing AS, linearly increased (P<0.05) morphometric measures, compact index, and carcass weight. While cooling loss, REA, and BFT were shown a quadratic effect (P<0.05) being maximal values to horses for categories within from >3<15 years. The differences on carcass characteristics by age can be explained by different level of physiological development and by mature size depending of the breed, at greater age, greater morphometric and carcass measures expressed as absolute measures are expected (^{Sarriés and Beriain, 2005}; ^{Znamirowska, 2005}). In our study, only carcass yield (expressed as %) was not affected by AS. In opposite, ^{Sarriés and Beriain (2005)} attributed improvements in carcass yields to age when evaluated carcass yield, when slaughter occurred at 16 and 24 months. It is important to indicate that 16 and 24 months of age is a very early growth period (low maturity level), and could impact carcass yield. It would be necessary to evaluate shorter age ranges and longer periods of growth (i.e. up to 72 months) in order to determine the maximal carcass yield point according to the horses age. In the case of the REA, it does occur in this manner, due to the fact that this show greater values for interim ages, decreasing in horses older than 15 years. ^{Schauer (2015)} reported that muscle characteristics in horses increase with age until reaching a peak (this is presented in all mammals species). Likewise, subcutaneous fat (BFT) seems to behave in the same manner decreasing in older animals.

The effects of BCS on carcass characteristics are shown in Table 5. As BCS increased, leg length linearly decreased, but increasing leg circumference enough to raise the compactness index value in horses. Similar than feedlot cattle (^{Apple et al., 1999}; ^{Minchin et al., 2009}), as BCS increased, heavier carcass weight, greater REA, and a lower CL were registered. REA are related to rate of weight gain, and rate weigh gain is dependent of energy intake (which is correlated with fat deposition (^{Cunha, 1980}). Thus, it is expected that horse that had better feeding system, greater gain rate and more body fat (and greater BCS), had greater REA values. It is well known that BCS were developed to assess the level of body fat proportion (based on subcutaneous fat index). Therefore, BCS is highly correlated with BFT (^{Henneke et al., 1983}; ^{Gentry et al., 2004}).

Body condition score values in this study are lower than those previously reported (^{Schauer, 2015}; ^{Martin-Giménez et al., 2016}). As mentioned above, differences between studies can be attributed to differences in production systems (background of feeding and weight at slaughter) and by horse breed. The lower CL in horse with high BCS is explained by lower dehydration by greater fat cover (BFT) in carcass.

The most influential factor in carcass characteristics was SW, since it had a significant effect (P<0.05) on every characteristic evaluated (Table 6). As SW increased, carcass morphometric measurements (length, chest depth, leg length, leg circumference, and compactness index) also increased. Carcass weight (HCW and CCW) show the same behavior, with more pronounced increases (P<0.05). Same behavior appeared in the case of BFT, however, anatomical variations in fat deposition, which could confuse fatness estimates, could be different amongst different horse breeds (^{Martin-Giménez et al., 2016}). Therefore, more ultrasonography studies on different anatomical areas are necessary, to determine the areas with the greatest tendency for fat deposition of each breed of horses (^{Quaresma et al., 2013}). In the case of CL, it was inversely proportional to SW, that is, the heavier the horse, the lower CL value. As mentioned previously, CL is inversely related to carcass cover fat (BFT). The values of cooling loos registered in this study are consistent with those reported by ^{Franco et al. (2013)} and ^{Lorenzo et al. (2013a)}, which range between 2 and 3%. But not with ^{Sarriés and Beriain (2005)}, who reported cooling loos values as large as 13%. Generally, REA size is proportional with body size and degree of body fatness, therefore it is expected that at greater SW greater REA. Although, it is necessary to consider the fact that heavy animals are not necessarily fat and light animals are not necessarily thin either, which would directly affect the REA. Even still, ultrasound measurements for equine muscles are reliable and repeatable (^{Abe et al., 2012}). In feedlot cattle, in which there are countless studies related to factors that affect REA, the production system and nutritional status affect body size and degree of fatness same affecting REA size (^{Yokoo et al., 2008}).

Coefficient determination (R2) between factors and variables registered are shown in Table 7. Origin, sex and age explained very little (R2≤0.20) its effects on carcass characteristics, while SW mainly explained the changes in linear measurements (R2≥0.40) and carcass weight (R2=0.55), while BCS mainly explained carcass yield characteristics (R2=0.27) and REA (R2=0.42). Since these variables are not only affected by the factors assessed (origin, sex, AS, BCS and SW), low R2 values (R2<0.50) can be expected despite the fact that the effect is significant (P≤0.05). Unlike the factors of origin, sex and age of the horses, factors such as SW and BCS are strongly related, thus, represent greater R2values.

Conclusion

Origin, sex and age, per se, explained very little its effect on the carcass characteristics. SW mainly explained changes in linear measures and carcass weight, while BCS mainly explained carcass yield and REA.