Investigation of welfare impacts of gaseous methods for on-farm euthanasia of suckling piglets : a thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Physiology at Massey University, Manawatu, New Zealand

Blunt trauma (BT) to the head is the most common method used for on-farm euthanasia of pre-weaned piglets. When performed correctly, loss of consciousness is immediate, but the potential for delivery of sub-lethal blows, along with aesthetic unacceptability to many operators, has lead to the need for alternative methods to be developed. One recommended alternative is exposure to 100% CO2. Although gas euthanasia is potentially more reliable and less disturbing to perform than BT, there are concerns that CO2 may induce breathlessness and pain before loss of consciousness, thus negatively affecting piglet welfare. This research aimed to evaluate the welfare impact of alternative gases, relative to CO2, for piglet euthanasia. A small pilot study was conducted to select appropriate gases for further evaluation. This identified 100% argon (Ar) and a mixture of 40% CO2-60% argon (CO2-Ar) as possible alternatives to 100% CO2 (CO2) for piglet euthanasia. The relative welfare impacts of CO2, Ar and CO2-Ar were evaluated in two studies. These studies aimed to identify the interval following gas exposure in which the animal may be conscious, and to identify evidence of welfare compromise within this interval. Identifying the period of possible consciousness is important in evaluating welfare impact, as this defines the time period in which the animal is capable of perceiving potential negative experiences associated with euthanasia. In the first study, conscious animals were exposed to the test-gases in a purpose-built chamber. Behavioural and physiological data including escape attempts, vocalisation, loss of coordination, loss of posture, respiratory effort, convulsions, gasping, and respiratory arrest were recorded until death. Loss of posture has previously been used to infer the onset of unconsciousness, whilst escape attempts, vocalisation and laboured breathing are associated with the experience of pain, aversion and distress in animals. Piglet behaviour was examined for evidence of negative experience prior to the onset of unconsciousness. In the second study, EEG and ECG data were recorded from anaesthetised, immobilised pigs during exposure to the same test gases used in the first study. Changes in the amplitude of the raw EEG can provide information on the level of consciousness. Changes in the EEG power spectrum, derived from mathematical transformation of the raw EEG, can provide evidence of noxious stimulation in anaesthetised mammals. EEG recorded during exposure to test gases was analysed to determine the likely latency to loss of consciousness with each gas, and to determine whether nociceptive processing occurred. Changes in heart rate, derived from the ECG, are frequently used as indicators of acute stress in mammals. ECG recorded during gas exposure was examined for indications of physiological stress responses. Behavioural data suggested that the latency to onset of unconsciousness did not differ between gases. However, the changes in the amplitude of the EEG suggested that loss of consciousness may occur sooner with CO2 than with Ar or CO2- Ar. Behavioural data indicated that piglets found CO2 exposure more aversive or unpleasant than exposure to either Ar or CO2-Ar. However, CO2-Ar induced greater respiratory stimulation than Ar alone, suggesting that Ar caused the least negative welfare impact of the 3 gases. ECG data showed that heart rate increased prior to likely loss of consciousness in piglets exposed to CO2 and CO2-Ar but not Ar alone, suggesting that Ar exposure does not induce a physiological stress response. Contrary to expectations, there was no evidence of nociception in piglets exposed to either 40 or 100% CO2, although this may have been influenced by the method used. Together, these data suggest that whilst CO2 induces more rapid loss of consciousness than Ar, it also results in significantly greater welfare impact prior to loss of consciousness. The addition of CO2 to Ar may provide some welfare advantage over CO2 alone, but not over Ar alone. From a welfare perspective, Ar is preferable to either CO2 or CO2-Ar for piglet euthanasia