Stunning before slaughter is performed to reduce fear in the animal, as well as to prevent pain and suffering during exsanguination. According to EFSA [
6], stunning must induce immediate and unequivocal loss of consciousness and sensibility or, if this is not immediate, the induction should not cause fear, pain, or suffering in conscious animals. In addition, unconsciousness must last until brain death is achieved due to bleeding. For the stun method to be acceptable, it should also ensure a satisfactory level of carcass and meat quality. In this study, we examined two gas mixtures for stunning pigs (90C and 20C2O) to compare (1) the effects on aversion and discomfort during the induction to unconsciousness, (2) the capacity to maintain unconsciousness after stunning (stun quality), and (3) the meat and carcass quality.
4.1. Induction to Unconsciousness
In the behaviors analyzed during both gas stunning batches, pigs were not only aversive to both gas mixtures (90C and 20C2O) but also showed behavioral indications of discomfort (pain and distress) in nearly all groups (albeit at varying periods). This was more evident in pigs stunned in 90C compared to 20C2O. Pigs frequently started lifting the head upwards, pointing the snout towards the cage ceiling. They also often attempted to escape from the cage, with the head stretched up and mouth wide open, while struggling to avoid falling over.
Such behavioral signs of discomfort should be considered a welfare concern. According to animal welfare legislation [
22], animal fear, pain, and suffering should be avoided during stunning and killing, highlighting the urgency to seek alternatives to CO
2 stunning/killing [
23].
The first sequence of behaviors shown during gas exposure in both stunning methods were retreat and escape attempts, both considered signs of aversion [
8,
24]. The first appearance of aversive behavior could be due to the inhalation of the gas mixture. However, the displayed aversion in hypercapnic-hypoxia stunning may not be due to acidification of the nasal cavity, as would happen with hypercapnia, as the threshold of CO
2 concentration has been found to be between 20 and 30%, and severe hyperventilation would have been seen [
25], which was not confirmed. It is also possible that the environment in the gondola, including new sounds and smells, could provoke a first aversive reaction. For instance, the sound of the chains against the cogs came suddenly when the gondolas moved through the machine, which could be heard via the digital sound recorders placed in the crates. Additionally, the sound of the gas injection was sudden and intermittent, and sometimes vocalizations from pigs from below gondolas could be heard when pigs were in the upper level. Given the conditions of this study, it was not possible to determine the drivers of escape and retreat attempts, and whether aversion was provoked by inhalation of a small concentration of the gas mixture or by the system environment, or both. In any case, from our results, it can be confirmed that the prevalence of animals showing these signs of aversion were not different between the two gas mixtures, but they appeared earlier in 90C than 20C2O. According to this, it is plausible that the onset of aversive response has to do with the level of CO
2 concentration, being quicker when the CO
2 concentration is higher. These results are in accordance with those from Velarde et al. [
2], Llonch et al. [
16], and Verhoeven et al. [
9], who found that higher concentrations of CO
2 (e.g., from 70% to 90%) trigger a quicker aversive response than mixtures containing lower concentrations of CO
2 (from 30% to 80%).
Gasping is an indicator of breathlessness during inhalation of gas containing CO
2 [
2]. According to our results, pigs exposed to 90C showed a higher prevalence and a quicker onset of gasping compared to 20C2O. Llonch et al. [
17] also found that gasping occurred more frequently and appeared earlier in animals exposed to 90C compared to 20C2O, the difference of which was even higher (87% vs. 19%) compared to results from this study (93% vs. 62%). This effect of CO
2 concentration on gasping was already stated by Gregory [
26], suggesting that the residual medullary activity in the brainstem that induces gasping when the atmospheric concentration becomes hypercapnic increases with higher concentrations of CO
2.
The time to lose posture is considered the first behavioral indicator of the onset of unconsciousness [
8]. In our study, loss of posture was the time when animals showed the inability to stand in an upright position, but it did not necessarily mean that they started to lay, as in numerous occasions pigs struggled to maintain their position. Therefore, we also considered the time in which pigs started to lay when the whole body was resting and the head was on the ground and not flexing upwards. Considering either the onset of loss of posture or laying behavior, pigs exposed to 90C showed a quicker response than 20C2O, suggesting that higher concentrations of CO
2 induce unconsciousness faster, confirming the results of similar studies conducted in experimental conditions [
16,
17]. However, in this occasion, the rapidness to induce the loss of posture (even considering laying behavior) was quicker compared to [
16,
17]. Our hypothesis is that in the present study conducted in commercial conditions, the circular movement of the crates never stopped, which helped improve the gas uniformity within the pit (upper vs. lower depth). In the studies of Llonch et al. [
16,
17], pigs were stunned with a dip–lift system with a single cage moving up and down, which may not have contributed as much to uniformity within the pit. Conversely, in the commercial paternoster system, the gas concentration was more homogeneous within the pit, which makes it that pigs probably met the highest CO
2 and the lowest O
2 concentration in a shorter time after starting the exposure compared to the dip–lift system, facilitating a faster effect in the paternoster system compared to the dip–lift.
Duration of muscular excitation was calculated as the time between when the first pig started to move uncontrollably until cessation of muscle excitation in the last pig of the crate. In other studies, the duration of muscular excitation is measured individually, but this was not possible in our study, as video recording did not allow the individual identification of pigs throughout the exposure. Still, results are in accordance with other studies [
17,
18] where muscular excitation lasted longer in pigs exposed to 20C2O compared to 90C. Raj [
14] states that the acidification of the central nervous system, when inhalation of high concentrations of CO
2, inhibits muscular excitation during induction. Therefore, the higher the concentration of CO
2, the shorter the duration of muscular excitation.
There has been some debate on whether muscular excitation occurs as a voluntary response to the gas mixture, and is therefore indicative of discomfort, or if it is a period of involuntary movements provoked by a lack of modulation of the neuronal structures that regulates motor activity [
27]. According to a report issued by EFSA [
6], muscular excitation during CO
2 exposure might be an aversive response, with implicit consciousness. In studies assessing the brain activity during exposure from 80% to 95% CO
2 [
9,
28] and N
2 and CO
2 gas mixtures [
18], evidence suggests that the start of muscle excitation occurs before significant changes in brain function appear, which could indicate that pigs were conscious. Furthermore, some of these studies [
9,
18] show evidence that most of the muscular excitation period occurs while brain activity is still high, indicating that the pig is still likely to be conscious.
4.2. Stun Quality
All (100%) pigs stunned with 90C were appropriately stunned after exposure and remained unconsciousness until brain death after sticking. According to instructions from the manufacturer (Butina), the minimum exposure time should be not less than 3 min, and the CO2 concentrations should be no less than 80% at the first stop and more than 90% at the bottom of the pit. As CO2 concentration and time of exposure were always higher than indicated by the manufacturer, it is not surprising that all pigs exposed to 90C were appropriately stunned.
The stun quality was satisfactory (no pigs recovered consciousness) despite the fact that the stun-to-stick interval was often longer than one minute, which is the maximum time recommended by EFSA [
6]. This may be explained because most of the pigs were probably dead when they were released from the stunning unit. This supports the findings from Llonch et al. [
18], who stated that the majority of pigs were dead after a 3 min exposure to 90C in a dip-lift system. The death of pigs after exposure to gas stunning safeguards animal welfare, as the stun-to-stick interval becomes irrelevant. For example, Atkinson et al. [
19] found that a 90% concentration of CO
2 stunning in six Swedish slaughter plants consistently provided adequate stunning (and probably death) despite stun-to-stick times longer than EFSA recommendations.
Conversely, 20C2O could not achieve the stun quality of 90C stunning, with 7.4% of pigs showing some level of inadequate stunning, which is also over EFSA’s recommendations of no more than 5% of pigs having signs of inadequate stunning [
6]. It is worth noting though that only five of the 796 pigs assessed had a stun level of four, indicating a high risk of poor animal welfare. There are several reasons that could explain this lower stun quality of 20C2O. The most notorious reason is the concentration of oxygen inside the pit. According to our results, a presence of oxygen above 2% in atmospheric air was associated with a higher percentage of signs of recovery before sticking. The facilities used in our study were designed to be used with high concentrations of CO
2. However, they may not necessarily be suitable to contain a modified atmosphere containing 80% N
2 and 20% CO
2, (with less than 2% O
2). Gas mixtures containing high concentrations of N
2 are difficult to contain because its lower molecular weight makes it difficult to contain in a pit [
15]. In our study, there was a need for a constant supply of 20C2O, suggesting that the gas mixture was continuously escaping from the pit. Therefore, gas stunning facilities should be adapted to improve containment of N
2 and CO
2 gas mixtures where the O
2 concentration can also be kept low. Our results suggest that if O
2 levels can be kept under 2%, the likeliness of recovery after a 5.5 min exposure to 20C2O is dramatically reduced. The stunning quality should be further investigated with maintained low (<2%) oxygen levels before hypercapnic-hypoxia can be considered as a real alternative to hypercapnia
The time of exposure to the gas mixture has also an effect on the stun quality. Llonch et al. [
18] recommended that for a good stunning quality using 20C2O, pigs need to be exposed to the gas for at least five minutes. The previous statement was based on pigs stunned with an individual dip–lift system without stops in an 8m
3 pit volume. In the present study, similar to many other commercial abattoirs, the stunning unit encompassed six cages each loaded with a nominal group of 3 to 4 pigs, rotating in a 63 m
3 pit volume and stopping at different points (and gas concentrations) inside the pit. A proof of that is the difference in O
2 concentration between the first and second stop (2.2 and 5.6 m depth) shown in this study. Although we already extended the exposure time (by 30 s) from what was recommended by Llonch et al. [
17], we still found some pigs with inadequate stunning. In order to reduce the likeliness of recovery, a further extension could be implemented, seeking for complete breathing cessation and death of pigs, as occurred in 90C stunning. In this sense, Raj [
14] recommended an exposure time of at least seven minutes in pigs exposed to a mixture of inert gas (argon) and CO
2. This extension may be controversial if implemented in commercial abattoirs, as it may decrease the speed of the processing line and limit the production capacity of the plant. However, this can be compensated by increasing the capacity of the gondola system, which can be achieved by increasing the capacity of the gondola (more pigs per cage) or adding more cages.
According to our results, no association was found between stun-to-stick interval and the stun quality under the conditions of this trial. However, it is also true that if pigs are not killed by the exposure to the gas, the likeliness of regaining conscious increases with longer periods between the end of stunning and exsanguination [
6]. In both stunning methods, the stun-to-stick time for the last pig in the group exceeded the Swedish recommendation of 60 s (average was 1 min 21 s for 90C and 1 min 23 s for 20C2O). As 38 pigs from 20C2O showed gagging after exiting the stunning unit, the combination of pigs coming out of the stunning unit alive with extended stun to stick times, and the physical action of gagging (deep breaths) would have facilitated O
2 inhalation and the associated recovery. Evidence that consciousness was regained after the stun-to-stick interval was that corneal reflex appeared just before sticking and not when exiting the stunning unit. As a corrective measure, some authors [
14,
18] have stated that in order to guarantee unconsciousness of pigs until brain death when exposed to N
2/CO
2 gas mixtures, additional killing methods are required, such as electrically mediated cardiac arrest.
4.3. Meat and Carcass Quality
The results obtained in this study are similar to research obtained in experimental trials, where conditions and study design were controlled. Llonch et al. [
17] found a lower pH45 in carcasses from pigs stunned with 20C2O compared to 90C (6.66 ± 0.05 vs. 6.26 ± 0.06;
p < 0.001). In that study, differences in pH45 were clearly due to the effect of a different stunning gas mixture. It was suggested that the longer muscular excitation period during the exposure to the gas mixture increased abruptly the energy demands of the muscular tissue, which is associated with the anaerobic metabolism of glucose into lactic acid. The accumulation of lactic acid into the muscle results in a more pronounced decrease of pH compared to pigs with lower energy requirements closely before slaughter [
29,
30]. The extended muscular excitation period of pigs stunned with 20C2O compared to 90C shown in our study confirms that the significant pH reduction would also be due to the greater energy expenditure, likely due to a longer muscular activity before killing.
The decline of the muscular pH after slaughter is determinant for pork quality. The acidification of the muscular milieu provokes protein denaturalization, which leads the transformation of muscle into meat [
29]. However, if the denaturalization process occurs too quickly, the capacity to hold water from proteins decreases, leading to an exudative meat [
31]. PSE meat can be determined using different criteria. Two criteria that are frequently used are pH45 (lower than 6) and EC (higher than 6 µS at 24 h after slaughter) [
21]. According to these criteria, 11 of the 223 carcasses (4.9%) from pigs stunned with 20C2O showed PSE meat, whereas no carcasses from 90C did so. Although the differences in percentage of PSE carcasses are significant, the number of affected carcasses is much lower than those reported by Llonch et al. [
17] in experimental conditions (13% of carcasses with PSE). However, as the genetic background (and many other affecting variables) of both studies were different, it is difficult to compare results. PSE affecting nearly 5% of the carcasses from pigs stunned with 20C2O is still noteworthy, and the use of alternatives to 90C, as such as 20C2O, in commercial conditions should be promoted as long as product quality insults are addressed.
The presence of ecchymosis was used to evaluate carcass quality [
21]. In the experiment from Llonch et al. [
17], one out of every four carcasses from pigs stunned with hypercapnic-hypoxia showed ecchymosis in the hams. According to our results, no carcass from 90C stunned pigs showed ecchymosis, and it was insignificant in 20C2O, as only one out of the 223 carcasses had ecchymosis. These results indicate that carcass quality is not affected when nitrogen and CO
2 gas mixture stunning is performed under commercial slaughter. However, these results should be taken with care because the hams were not assessed completely, being that two thirds of the ham were covered by skin and could not be assessed. Thus, further studies would be needed to confirm these results.