Reflexes

Reflexes are automatic, stereotyped movements that are produced as the direct result of a stimulus and are mediated by the central nervous system (Carlson, Reference Carlson2007). The presence of central reflexes are indicators of consciousness that are linked to functioning of the brain stem or spinal cord. Brain stem reflexes are regulated by 12 pairs of cranial nerves that enter and exit the brain and are not under cortical control. Two cranial nerves (I and II) enter from the forebrain and the other nerves (III to XII) enter and exit from the brain stem (Carlson, Reference Carlson2007; Rubin and Safdieh, Reference Rubin and Safdieh2007). Brain stem reflexes that are used to assess unconsciousness after stunning in livestock are cornea or blinking, palpebral, pupillary light and threat reflex. The cornea reflex causes involuntary blinking of the eyelids in response to stimulation of the cornea and is in general the last reflex to be lost in anaesthetised animals (Dugdale, Reference Dugdale2010). The palpebral reflex also results in blinking as a response to touching the medial canthus of the eye and disappears earlier than the cornea reflex in anaesthetised animals. Both the cornea and palpebral reflex require a functional afferent cranial nerve V (trigeminal) and efferent cranial nerve VII (facial) and the relevant eye muscles to function adequately (Adams and Sheridan, Reference Adams and Sheridan2008). The pupillary light reflex is tested by letting light fall on the eye and observing whether the pupil adapts to it. The reflex is controlled by cranial nerves II (optic) and III (oculomotor) and is not considered a reliable reflex during exsanguination, as exsanguination interferes with the blood supply to the retina (Blackman et al., Reference Blackman, Cheetham and Blackmore1986). When testing the threat reflex, an object (finger or pencil) suddenly approaches the eye and a conscious animal will close its eye or withdraw the head. This reflex requires a functional efferent cranial nerve VII (facial) and integration of the motor cortex, but is not often applied, as it requires the eye to be open. Focused eye movement, not a reflex, is considered a definite sign of consciousness, as it needs cortical activity for perception and controlled motor activity from the eyeball muscles (Grillner et al., Reference Grillner, Wallén, Saitoh, Kozlov and Robertson2008; Vogel et al., Reference Vogel, Badtram, Claus, Grandin, Turpin, Weyker and Voogd2011). It is pointed out that positive eye reflexes alone do not necessarily indicate consciousness, as positive brain stem reflexes might occur on the basis of residual brain stem activity and do not distinguish clearly between consciousness and unconsciousness (Anil, Reference Anil1991). This especially holds true for animals that are electrically stunned, which was documented as early as 80 years ago (Roos and Koopmans, Reference Roos and Koopmans1936; Blackmore and Delany, Reference Blackmore and Delany1988; von Holleben et al., Reference von Holleben, von Wenzlawowicz, Gregory, Anil, Velarde, Rodríguez, Cenci Goga, Catanese and Lambooij2010). In both sheep and calves, brain stem reflexes were present long after electrical stunning, even though the EEG was suppressed or iso-electric (Anil, Reference Anil1991; Anil and McKinstry, Reference Anil and McKinstry1991). On the other hand, eye reflexes may be inhibited after electrical stunning, whereas the cerebral cortex still functions and the animal may be conscious (Blackmore and Delany, Reference Blackmore and Delany1988). There is no literature available on the frequency of such incidences and its risk for animal welfare is therefore difficult to estimate. After effective captive bolt stunning, however, no eye reflexes should be present, because of the brain trauma produced (Finnie, Reference Finnie1995; Gregory and Shaw, Reference Gregory and Shaw2000). Thus, cranial nerve reflexes can be good indicators for impaired midbrain or brain stem activity, but only work reliably in one way: when absent, it is very likely that the animal is unconscious, but when they are present, the animal is not necessarily conscious. Spinal reflexes include stretch and flexor reflexes. The stretch reflex, a monosynaptic reflex, is the most basic reflex and plays an important role in control of posture. It does not involve the brain and is therefore not used to assess unconsciousness (Carlson, Reference Carlson2007). The flexor reflex, a polysynaptic reflex, involves activation of nociceptors and is used to assess unconsciousness (Anil, Reference Anil1991; Erasmus et al., Reference Erasmus, Turner and Widowski2010). An example of a flexor reflex is the pain withdrawal reflex, which is elicited by applying a painful stimulus to the animal, such as a nose or ear prick. In a survey on expert opinion, the pain withdrawal reflex was ranked high, and thus valued highly, as an indicator to assess unconsciousness after all types of stunning (Gerritzen and Hindle, Reference Gerritzen and Hindle2009). The pedal reflex is elicited by, for instance, pinching the skin between the toes of an animal. This reflex is often used for assessment of depth of anaesthesia in laboratory animals, such as rodents and rabbits, but is only occasionally applied in livestock after stunning, as all spinal reflexes are difficult to assess when animals exhibit convulsions or body movements (Tidswell et al., Reference Tidswell, Blackmore and Newhook1987). This especially holds true for animals that are physically stunned, for example, captive bolt stunning, when there is lack of inhibition from the brain and spinal reflexes may occur more vigorously (Blackmore and Delany, Reference Blackmore and Delany1988). Again, electrically stunned animals may exhibit this reflex long after losing consciousness and the reflex may occur more vigorously when the animal is handled (Blackmore and Newhook, Reference Blackmore and Newhook1982). The righting reflex refers to any reflex that tends to bring the body into its normal upright position. It is often assessed when animals are removed from the stunning box or are hung to the bleeding rail and is also referred to the head righting reflex. This reflex is also difficult to assess when animals exhibit convulsions or involuntary body movements (Blackmore and Newhook, Reference Blackmore and Newhook1982; Anil, Reference Anil1991). Table 1 shows an overview of the different brain stem and spinal reflexes used to assess unconsciousness after stunning.

Table 1 Reflexes used to assess unconsciousness in livestock after stunning

¹ Presence and absence of reflexes are presented as follows:+=present, −=absent, (+)=may be present, (−)=may be absent.

Behavioural indicators

Loss of posture, the inability of the animal to remain in an initial standing or sitting position, is considered a valuable indicator as it is often the first sign to be lost after successful stunning and indicates that the cerebral cortex is no longer able to control posture (Raj et al., Reference Raj, Wotton and Gregory1992; Raj and Gregory, Reference Raj and Gregory1996; Llonch et al., Reference Llonch, Rodríguez, Jospin, Dalmau, Manteca and Velarde2013). Both mechanical and electrical stunning should lead to immediate collapse (AVMA, 2013). Nystagmus, involuntary rapid horizontal eye flickering, is caused by damage to the vestibular, labyrinthine or central nervous system and was more present in cattle that had a shallow depth of concussion following captive bolt. It was observed in only 3% of 1608 cattle, but was associated with a greater chance of rhythmic breathing. Its presence could add strength to the conclusion that the depth of concussion has been shallow (Gregory et al., Reference Gregory, Lee and Widdicombe2007). In a study by Bourquet et al. (Reference Bourguet, Deiss, Tannugi and Terlouw2011), nystagmus was observed in one out of 95 captive bolt shot cattle. This animal was reshot, and this supported the study by Gregory et al. (Reference Gregory, Lee and Widdicombe2007), which indicated that when nystagmus was observed, there was a one in three chance that the quality of the stun was insufficient. Nystagmus may occur as a result of electrical stunning (Grandin, Reference Grandin2002), but in CO₂-stunned pigs, nystagmus was not observed once (Atkinson et al., Reference Atkinson, Velarde, Llonch and Algers2012). It is stated that under no circumstances should a stunned animal vocalise, as vocalisation after stunning indicates consciousness and probably distress and pain (Grandin and Smith, Reference Grandin and Smith2004; Gouveia et al., Reference Gouveia, Ferreira, da Costa, Vaz-Pires and da Costa2009). A large network of brain regions is involved in the production of vocalisations, including the frontal lobe and primary motor cortex and vocalisations are considered a conscious response (Carlson, Reference Carlson2007). The involuntary passage of air along the vocal cords, however, may cause sounds that can be mistaken for vocalisations. Absence of vocalisations on the other hand, is certainly no guarantee for absence of pain or distress, as the occurrence of vocalisations also depends on the species. A sheep often does not vocalise when injured, where a pig will scream loudly (Broom, Reference Broom2001; EFSA, 2004). Grandin (Reference Grandin2002) believes an animal to be unconscious when it shows a limp head and protruding tongue. The tongue is controlled by nerve XII (hypoglossal) and when relaxed this may indicate loss of cranial nerve function. A study by Gregory et al. (Reference Gregory, Lee and Widdicombe2007) showed that a protruding tongue was not associated with depth of concussion after captive bolt stunning, but was proposed as indicator following exsanguination, when 40% of the cattle had a protruding tongue while hanging on the bleeding rail. Similarly, relaxation of the jaw may be taken into account, but can be observed in conscious animals (Gregory et al., Reference Gregory, Spence, Mason, Tinarwo and Heasman2009). Both jaw relaxation and tongue protruding are not used as single indicators to assess unconsciousness, but can support other indicators of unconsciousness (Grandin, Reference Grandin2002; von Holleben et al., Reference von Holleben, von Wenzlawowicz, Gregory, Anil, Velarde, Rodríguez, Cenci Goga, Catanese and Lambooij2010). Beside the important role regarding consciousness, the brain stem also houses the regulatory centres for respiratory and circulatory systems. Rhythmic breathing movements after stunning indicate that the corticospinal, ventral and lateral columns of the spinal cord are still intact and may thus indicate consciousness (Mitchell and Berger, Reference Mitchell and Berger1975). The presence of rhythmic breathing after stunning is generally accepted to indicate that an animal may not be fully unconscious and is thought to be one of the first signs of recovery after CO₂ and electrical stunning (Gerritzen and Hindle, Reference Gerritzen and Hindle2009; Anastasov and Wotton, Reference Anastasov and Wotton2012). In captive bolt stunned cattle, rhythmic breathing immediately disappears after an effective shot because of axonal injuries to the brainstem (Finnie et al., Reference Finnie, Blumbergs, Manavis, Summersides and Davies2000). The occurrence of convulsions, observed as uncontrolled movements of the body, indicates effective stunning in electrical or mechanical stunned animals, but also occur in unconscious animals that are gas stunned (Adams and Sheridan, Reference Adams and Sheridan2008; Marzin et al., Reference Marzin, Collobert, Jaunet and Marrec2008; von Holleben et al., Reference von Holleben, von Wenzlawowicz, Gregory, Anil, Velarde, Rodríguez, Cenci Goga, Catanese and Lambooij2010). These convulsions are thought to be incompatible with consciousness due to the absence of higher motor control (Lambooij, Reference Lambooij2004). They can, however, sometimes be mistaken for rhythmic breathing, as they can occur as almost rhythmic body movements (Wotton and Sparrey, Reference Wotton and Sparrey2002). Gagging refers to low-frequency inhalations with the neck positioned towards the front legs and occasional emission of sounds similar to snoring and is considered an indicator of deep unconsciousness (Rodríguez et al., Reference Rodríguez, Dalmau, Ruiz-de-la-Torre, Manteca, Jensen, Rodríguez, Litvan and Velarde2008). Gasping is seen when an animal takes deep breaths through an open mouth and is considered an indicator of onset of breathlessness during CO₂ stunning, which continues long after loss of consciousness even when brain activity is no longer recorded, but may also occur after electrical stunning (Blackmore and Petersen, Reference Blackmore and Petersen1981; Newhook and Blackmore, Reference Newhook and Blackmore1982; Grandin, Reference Grandin2013). Interpretation of all individual indicators mentioned above can be doubtful unless supported by other information (Blackmore, Reference Blackmore1984; Gerritzen and Hindle, Reference Gerritzen and Hindle2009; Anastasov and Wotton, Reference Anastasov and Wotton2012). Table 2 shows an overview of the different behavioural indicators used to assess unconsciousness after stunning.

Table 2 Behavioural indicators used to assess unconsciousness in livestock after stunning

¹ Presence and absence of reflexes are presented as follows:+=present; −=absent; (+)=may be present; (−)=may be absent.

Brain activity (EEG)

When monitoring brain functioning, activity can be presented in an EEG, which displays electrical activity derived from electrodes attached to various locations on the surface of the head. The EEG is considered the most objective method for assessing unconsciousness and reflects the sum of underlying electrical activity of populations of neurones supported by glia cells (Murrell and Johnson, Reference Murrell and Johnson2006). There are four different types of wave patterns in the EEG that can be distinguished based on their respective frequencies and that are related to the state of consciousness: δ (0 to 4 Hz), θ (4 to 8 Hz), α (8 to 12 Hz) and β (>12 Hz) waves. Both δ and θ (slow wave) activity is related to sleep or reduced consciousness. α activity is prominent in subjects that are conscious, but mentally inactive (closing eyes and relaxation) and β waves are associated with active movements and increased alertness (Kooi et al., Reference Kooi, Tucker and Marshall1978; Niedermeyer et al., Reference Niedermeyer, Schomer and Da Silva2011). Depending on the method of stunning, the EEG shows a characteristic pattern of change when animals lose consciousness. Generally, an increase in low frequency activity is accompanied by an increase in amplitude. When neurons depolarise at the same time or frequency, they fire in a synchronised fashion creating slow high amplitude waves as seen in unconscious states suggesting a depression of the reticular formation (Lopes da Silva, Reference Lopes da Silva1982). Consciousness on the other hand is characterised by high frequency (α and β), low amplitude waves (Seth et al., Reference Seth, Baars and Edelman2005). When looking more specifically at EEG wave patterns, the EEG can be broken down in different time segments, better known as epochs. These epochs can be analysed for frequency (Hz), amplitude (µV) and power (µV²), together representing the amount of activity in the brain. Four stages of EEG can be distinguished during the process of stunning and slaughter and are related to the level of consciousness, namely: active, transitional, unconscious and iso-electric (flat) EEG. In the first (active) stage, normal awake activity is recorded with high frequency, low amplitude waves, indicating the animal is conscious. In the second (transitional) stage, the amplitude of the EEG increases together with a decrease in frequency. When these changes become more profound, the animal is considered unconscious. When loss of consciousness progresses, the EEG turns iso-electric and brain activity is no longer recorded (Gibson et al., Reference Gibson, Johnson, Stafford, Mitchinsont and Mellor2007; McKeegan et al., Reference McKeegan, McIntryre, Demmers, Lowe, Wathes, van den Broek, Coenen and Gentle2007). The exact moment when unconsciousness sets in, based on the EEG, is difficult to determine as changes are often gradual. The iso-electric EEG, however, is never compatible with consciousness. There is no consistency in the literature regarding the number of stages used in the assessment of unconsciousness. Other research may only differentiate between the stages conscious and unconscious or contrary, use additional stages besides the four mentioned above.

Derivatives of the EEG

Another way of analysing raw EEG data, next to visual appraisal of the EEG, would be to compute a Fast Fourier Transformation (FFT). The output thereof represents the frequency composition of the signal, or alternatively formulated, how much power is presented in the different frequency bands. The principle is similar to defining the EEG in different EEG types that consist of slow or fast waves with high or low amplitudes (Davidson, Reference Davidson2006). Further (automatic) calculations of the FFT can lead to EEG derivatives presenting a single value or percentage that is easier to standardise.

Derivatives of the EEG include: the total power (P _tot), which is the area underneath the frequency spectrum curve, the medium frequency (F ₅₀), which is the frequency below which 50% of the total power is located and the spectral edge frequency (F ₉₅), which is the frequency below which 95% of the power is located. These readout parameters are considered quantitative tools to describe changes in EEG activity (Murrell and Johnson, Reference Murrell and Johnson2006). An initial increase in P _tot may represent a loss of functional cerebrocortical activity when amplitudes of EEG waves increase because of synchronised firing of neurons. But as the level of unconsciousness deepens, a decrease in all three derivatives is seen (Bager et al., Reference Bager, Braggins, Devine, Graafhuis, Mellor, Tavener and Upsdell1992; Martoft et al., Reference Martoft, Jensen, Rodriguez, Jorgensen, Forslid and Pedersen2001). In electrically stunned livestock, an increase in power of all frequency bands is first observed in the first 5 to 15 s post-stun because of initial epileptiform activity (Velarde et al., Reference Velarde, Ruiz-de-la-Torre, Roselló, Fàbrega, Diestre and Manteca2002; Beyssen et al., Reference Beyssen, Babile and Fernandez2004). Automatic FFT is applied during human surgeries and on a smaller scale during animal surgeries, where the raw EEG and its FFT are used to assess anaesthetic depth. Established anaesthesia monitors are used to assess depth of anaesthesia, but they differ in the algorithm used to analyse the EEG (Bruhn et al., Reference Bruhn, Myles, Sneyd and Struys2006). To the authors’ knowledge, only one of such monitors, namely the Index of Consciousness or IoC, has been used in a study concerning stunning in animals. During gas stunning of pigs, the raw EEG was recorded and based on that data a dimensionless variable (IoC) was calculated (Llonch et al., Reference Llonch, Rodríguez, Jospin, Dalmau, Manteca and Velarde2013). This variable ranges from 100 (awake) to 0 (iso-electric) and decreases with increasing anaesthetic and sedative depth. Values between 40 and 60 are suggested to represent an adequate hypnotic effect of the subject under general anaesthesia (Grover and Bharti, Reference Grover and Bharti2008). In the study by Llonch et al. (Reference Llonch, Rodríguez, Jospin, Dalmau, Manteca and Velarde2013), time to loss of posture occurred almost 20 s earlier then the accompanying decrease in IoC. A delay in IoC reading, compared with loss of balance, was also seen in pigs anaesthetised with propofol, but with a delay of only 7 s (Llonch et al., Reference Llonch, Andaluz, Rodríguez, Dalmau, Jensen, Manteca and Velarde2011). Muscular excitations that occur during CO₂ stunning probably affected the IoC calculation, as movement artefacts are known to influence EEG data and calculations made in anaesthesia monitors (Teplan, Reference Teplan2002). This is one of the reasons offline calculation is used to more adequately compare and correlate brain activity data with behavioural indicators. Though many studies have looked at behavioural indicators or the EEG separately, only a few have studied correlations between these different read-out parameters for assessing unconsciousness. In a study by Benson et al. (Reference Benson, Alphin, Rankin, Caputo, Kinney and Johnson2012), loss of posture was correlated to the α/δ ratio extracted from the EEG, in an effort to find a more objective and alternative method (as opposed to loss of posture) to assess loss of consciousness in broilers. A correlation and no difference was found between time to unconsciousness as observed by the two methods, supporting the use of α/δ ratio as method to assess unconsciousness. The study shows that such correlations can provide additional, more objective data to support the use of behavioural indicators as a measure of unconsciousness and provide details when certain behaviours may be present or absent in an animal that loses consciousness.

Evoked responses

The EEG recording is also used to assess unconsciousness by way of generating evoked responses. Evoked responses are responses in the EEG following external stimuli (visual, somatosensory or auditory), generated in specific areas of the cerebral cortex, mid brain and brainstem (Schneider and Sebel, Reference Schneider and Sebel1997; Grover and Bharti, Reference Grover and Bharti2008). Evoked responses are frequently used as additional indicators to assess unconsciousness next to behavioural indicators, and have been applied in sheep, cattle, poultry and pigs. No correlations, however, have been calculated for the presence or absence of evoked potentials and presence or absence of behavioural indicators. Though, similar to EEG derivatives, evoked potentials may in this way provide additional support for the use of certain behavioural indicators. As for now, evoked responses are only used in experimental set-ups. Rapid changes in consciousness are difficult to observe with evoked potentials, as repeated stimulation and averaging of data (EEG) is needed to see these changes (Beyssen et al., Reference Beyssen, Babile and Fernandez2004). Differences in time to loss of consciousness based on the loss of spontaneous EEG or evoked responses have been observed in multiple studies. In hens stunned with different gas mixtures, evoked responses were observed to disappear ∼15 s after the EEG became suppressed, but almost 30 s before the occurrence of an iso-electric EEG (Raj et al., Reference Raj, Gregory and Wotton1991 and Reference Raj, Wotton and Gregory1992). In poultry slaughtered by nine different methods, all without prior stunning, spontaneous brain activity was lost after 23 to 233 s, where visual evoked potentials were lost after 90 to 349 s (Gregory and Wotton, Reference Gregory and Wotton1986). The loss of somatosensory evoked potentials was also recorded before an iso-electric EEG, but after a suppressed EEG in gas-stunned turkeys (Raj and Gregory, Reference Raj and Gregory1993). The presence of an evoked response implies that the afferent pathways to the higher brain centres are intact, but not necessarily that the animal is aware of the stimulus (Raj et al., Reference Raj, Gregory and Wotton1991). Visual evoked potentials have been observed in, for instance, anaesthetised animals (Gregory and Wotton, Reference Gregory and Wotton1986; Gregory, and Wotton, Reference Gregory and Wotton1989). Conversely, the absence of evoked potentials may not always guarantee unconsciousness (Anil et al., Reference Anil, Raj and McKinstry2000). Gregory and Wotton (Reference Gregory and Wotton1990) looked at the effects of multiple electrical stunning currents on spontaneous physical activity and evoked responses and found that the loss of somatosensory evoked potentials indicated a deeper level of unconsciousness than absence of neck tension. All these studies show that the use of different methods to assess unconsciousness may lead to different findings regarding the time to loss of consciousness. The use of absence of evoked responses or iso-electric EEG, may provide more conservative times to loss of consciousness compared with loss of spontaneous EEG. The indicators based on brain activity that can be used to asses unconsciousness after stunning are presented in Table 3.

Table 3 Indicators based on brain activity as presented in an electroencephalogram (EEG) used to assess unconsciousness in livestock after stunning

Difficulties in the use of EEG

Though the EEG may be considered most objective when assessing unconsciousness, there are some disadvantages to its use. First, there is no golden standard for the way in which the division of stages of consciousness is described and this also limits the use of brain function monitors in differentiating between consciousness and unconsciousness, especially during transitional stages (Alkire et al., Reference Alkire, Hudetz and Tononi2008). Second, it is difficult to compare EEG values between species and individuals, because of animal variation caused by electrode placement, skull thickness and differences between equipment. Third, the EEG can be influenced by artefacts that are animal related (eye or muscle movements) or technical related (cable movements, impedance fluctuation or 50/60 Hz interference) (Teplan, Reference Teplan2002). Experimental controlled situations provide a significantly better environment to limit these artefact sources than slaughter plants. These artefacts, however, limit possibilities for EEG application as an evaluation method in slaughter plants at this stage.