Assessment of stunning effectiveness in donkeys using electroencephalogram

Donkeys and other equids are currently being slaughtered all around the world for human consumption and their by-products. To ensure they are in an irreversible state of unconsciousness prior to bleeding, they are usually stunned with a penetrating captive bolt (PCB) device. In commercial settings, the effectiveness of stunning is usually assessed by observing behavioural signs and brainstem reflexes. However, electroencephalography (EEG) provides a more direct indicator of functional brain activity, making it the most accurate tool for assessing stunning success. Literature on factors affecting PCB stunning is available for the more common livestock species, while little information on donkeys. In addition, there are no guidelines for the ideal shot angle and position for donkeys, and anecdotally, the position suggested for horses is usually used (HSA, 2016). With these premises, this study aimed to assess the stunning effectiveness of donkeys shot with pneumatic PCB using EEG analysis and to determine if there are differences between shot deviation from ideal position and presence of normal-like EEG after stunning. The study was conducted in a commercial abattoir in Mexico on 13 donkeys. Subdermal electrodes were placed and EEG recordings were taken before, during and 30 seconds after stunning. After stunning, heads were removed and shot deviation from ideal position (for horses), both rostro-caudal and lateral, was measured. EEG recordings were analysed and classified in 5 categories: normal-like EEG, movement artifacts, transitional EEG, high amplitude-low frequency EEG and isoelectric EEG (Gibson et al., 2009). Associations between the presence of normal-like EEG and shot deviation were tested with Mann-Whitney test. Six donkeys (46.2 %) had periods of normal-like EEG after stunning, for a maximum of 9 sec. During this phase, there was an increase in alpha, beta and theta activity. Moreover, four donkeys (30.8%) did not reach isoelectric EEG during the 30 seconds of recording. For shot position, most of the shots were deviated caudally and to the left. There was a significant association (p=0.02) between rostral-caudal deviation and ineffectively stunned donkeys (i.e. the ones that had periods of normal-like EEG). This study revealed a high number of ineffective stunning and a number of animals that did not reach isoelectric EEG, namely a reliable sign of irreversible unconsciousness, before the end of the recordings. These results highlight welfare concerns related to the delayed or incomplete loss of consciousness in these donkeys. In conclusion, it is important to ensure exsanguination occurs without delay whilst assessing stunning effectiveness in each individual. Furthermore, the association between conscious brain activity and shot deviation confirmed the need for species-specific guidelines for shot angle and position in donkeys.