Measuring methods for fish welfare during slaughter based on electrical impedance, EEG, ECG and blood parameters

Abstract

Improved animal welfare during industrial slaughtering of fish is the aim of the scientific work presented in this thesis. The thesis is based on four publications that cover different stages of an automated industrial slaughtering line for fish. The publications are presented in a similar order to those on a slaughtering line. The results from paper I are relevant for all types of pre-chilling of fish before slaughtering and reveal the physiological effects of live chilling in Atlantic salmon (Salmo salar). Chilling of fish is commonly used in the industry, both during transportation and processing of the fish in the slaughtering house. The publication is based on two experiments where the first experiment included fish (mean weight 840 g) acclimatized to a water temperature of either 16, 8, or 4°C and which were directly transferred horizontally or vertically (9 combinations) to temperatures of 16, 8, 4, or 0°C using a dip net. In the second experiment, fish (mean weight 916 g) acclimatized to 16°C were exposed to four temperature-drop regimes (no physical handling): 16–4°C (over 5 h), 16–4°C (over 1 h), 16–0°C (over 5 h), and 16–0°C (over 1 h). Physical transfers in the first trial, i.e., temperature drops, resulted in immediate (1 h) increases in blood lactate concentrations at all three temperatures, but levels were significantly reduced and close to pretransfer levels after 6 h. Horizontal transfers, i.e. 16–16°C, 8–8°C, and 4–4°C, resulted in similar increases and were not significantly different from the groups exposed to temperature drops. The most severe vertical transfer (16-0°C) resulted in a swift loss of equilibrium and eventually death. In experiment No.2, temperature drops from 16 to 4°C and from 16 to 0°C over a period of one or 5 h, without physically handling the fish, resulted in no significant increases in any of the measured parameters 1 h post-transfer, except in the 16–0°C (1 h) group. The latter experienced a significant increase in blood sodium, glucose, lactate and cortisol levels compared to all other groups. The results suggest that salmon are capable of tolerating relatively steep temperature drops without any significant negative effects on blood stress parameters and that physical stress from gentle handling overrides the effect of thermal insults, which is important for the slaughtering procedure. The overall objective of the study in paper II was to find the optimal configurations for industrial percussive and electrical stunning by evaluating the methods under laboratory conditions. In an automated slaughtering line electrical and percussive stunning are common methods used to ensure unconsciousness, which is critical for fish welfare, before bleed out. The work described in this publication defines the settings, especially voltage and air pressure, needed for efficiently rendering the fish unconscious and also to verify the effect of the stunning machines. Evidence of unconsciousness and insensibility of Atlantic salmon was provided on the electroencephalogram (EEG) by the appearance of slow waves and spikes, followed by a strong depression in electrical activity. This phenomenon was observed in 17 salmon after percussive stunning using an air pressure of 8.1 to 10 bars, whilst 8 fish were considered conscious at pressures below 8.1 bars, although some were seemingly unconscious in behavior. Consequences were a haemorrhage in the brain cavity in 15 out of 17 fish, broken upper or lower jaws in 9 fish and eye burst in 8 fish. A general epileptiform insult (unconscious and insensible) was obtained by delivering a voltage, consisting of a direct current (DC) coupled with 100 Hz alternating current (AC) with a peak value of ::112 volt (V), head to body, for approximately 0.5 s. The total duration of the insult was 62±44 s (mean±SD; n=25) which was followed by minimal brain activity in 19 fish. The heart rate was 20±7 beats/min prior to stunning. After stunning, the electrocardiogram (ECG) revealed fibrillation for 22±15 s and became irregular and showed extrasystolae (ventrical contraction) afterwards. Exposing the salmon for 5 s with electricity followed by a gill cut resulted in 1 out of 3 fish temporarily recovering after 3 min. Haemorrhages were not observed in the fillets. Average current for head to body electrical dry stunning was 668 milliampere (mA) root mean square (RMS) with an average stunning voltage of 107.9 Vrms. Electrical head to body stunning can be recommended when using coupled AC and DC current of 668 mArms and ::107 Vrms. The salmon can be stunned in approximately 0.5 s. However, the experiment concluded that a correct bleeding procedure should be developed. For percussive stunning it was concluded that if sufficient force is used the fish will be rendered unconscious and insensible, however this resulted in damage to the carcass, whereas a combined AC and DC signal is recommended for dry electrical head to body stunning. The objective of paper III was to verify the optimal AC frequency range to be used during industrial electro stunning, i.e. electronarcosis, of Atlantic salmon by investigating the electrical impedance spectra of the combined fish and electro stunning device entity. This is an important task since the frequency of the electrical signals is crucial to the electro stunner’s effect. The electrical impedance and associated phase shift was measured in the frequency range 40 Hz to 1.0 MHz for individual fish (n = 11) placed in a regular electrical stunner. The results of the experiment showed that the average overall impedance of the combined fish and electrical stunning device increases with frequency from 40 to 60 Hz before leveling out in the range from 60 to 800 Hz. Thereafter the impedance decreases to a negligible value at 1 MHz. Measurements on impedance and phase angle showed that the highest average electrical impedance appeared at 100 Hz. Furthermore, there were individual peak impedance variations between 70 and 100 Hz. In all fish measured, the impedance at 900 Hz was observed to be lower than that at adjacent frequencies. Due to average measured impedance values and the expected influence of the alpha dispersions on the cell surface, as reported in previous research, it was concluded that the optimal AC frequency range for electro stunning of the Atlantic salmon brain is 70 to 100 Hz. The aim of paper IV was to understand the importance of electrical signal frequency spectrum on stunning, recovery and inflicting injuries. Hemorrhaging in the filet, caused by broken backbones, has been a quality problem for the industry when electro stunning is used to render the fish unconsciousness. The paper also shows the effect of chilling during bleed out. In this article Atlantic salmon were exposed for 5 seconds to either 217 Vrms, 50 Hz, AC or 107 Vrms coupled AC+DC at 200Hz, with and without a high frequency spectrum. Post stun the fish were placed back into water, either at ambient seawater temperature (10.4 °C) or cold water (-1.3 °C), to investigate recovery or mortality. The results showed that a high frequency spectrum, but low amplitude prevented the muscles from contracting and causing spinal injuries and hemorrhaging, for all individuals. Injury rates of 14 and 18% was observed when using electrical signals containing only low frequencies of 200 Hz AC+DC and 50 Hz, AC. The high frequency spectrum also reduced the stimulation of the brain as fish recovered faster with no mortality. Adding a cold shock post stunning delayed or prevented recovery of all groups within the time span required to kill the fish by exsanguination. Papers III and IV will potentially have relevance for other disciplines, such as medicine, where electroshock and electronarcosis are used.