- Email: [email protected]
A Comparison of Argon, Carbon Dioxide, and Nitrogen in a Broiler Killing System
PROCESSING AND PRODUCTS A Comparison of Argon, Carbon Dioxide, and Nitrogen in a Broiler Killing System G. H. POOLE and D. L. FLETCHER1 Department of Poultry Science, University of Georgia, Athens, Georgia 30602
1995 Poultry Science 74:1218-1223
INTRODUCTION Electrical stunning has received considerable attention in recent years concerning the relative benefits of automated killing and humane slaughter vs stunningrelated carcass damage. Ma and Addis (1973) reported that electrical stunning of turkeys causes an increase in wing and keel breakages. Mohan Raj et al. (1990b) reported that 55% of birds stunned electrically and then mechanically plucked had broken bones. Gregory and Wilkins (1989) found that with an increase in stunning current between 74 to 269 mA per bird there was a corresponding increase in birds with one or more broken bones.
Received for publication August 8, 1994. Accepted for publication February 14, 1995. ^ o whom correspondence should be addressed.
The use of gas killing systems has been proposed as a possible alternative to electrical stunning systems. The potential advantages of a gas killing system would be to reduce the carcass damage associated with electrically stunned birds (Mohan Raj and Gregory, 1990a), as well as the fact that gas stunning could be performed prior to hanging (Mohan Raj and Gregory, 1990b), thus reducing another major source of carcass damage (Albright, 1991). There have been various benefits reported with the use of a gas kill system. Carbon dioxide has been associated with increased blood loss (Kotula et al, 1957), decreased initial breast muscle pH (Mohan Raj et al, 1990b; Flemming et al, 1991) and terminal breast muscle p H (Mohan Raj et al, 1990b), decreased broken bones, and decreases in both superficial and deep muscle hemorrhaging in broiler chickens (Mohan Raj et al, 1990b).
1218
Downloaded from http://ps.oxfordjournals.org/ by guest on May 12, 2015
ABSTRACT Argon, C0 2 , and N 2 gasses were each evaluated in a broiler chicken gas killing system. Birds were killed by individual exposure to one of the three gasses for 2 min in a flow-through system. The gasses were evaluated by determining the time, in seconds, for the following responses: first reaction to the gas, loss of posture, eye closure, initiation of death struggle, and cessation of respiration. Percentage blood loss over a 3-min bleed time was determined by weight loss. Breast muscle pH values were determined at 15 min and 24 h post-mortem on the Pectoralis major muscle. Carbon dioxide resulted in the earliest first reaction, loss of posture, eye closure, and initiation of struggle. Argon and N 2 exhibited a delayed first reaction, a less severe early reaction, but a more severe unconscious death struggle. All birds died in approximately 75 s. Results indicate that the flow-through gas system takes longer to kill broilers than the immersion systems previously reported. Gas killing resulted in lower (P < .05) blood loss. Initial breast muscle pH values were significantly highest for the birds killed with C 0 2 , followed by the control treatment, which was significantly higher than the values for broilers killed with either Ar or N 2 . After 24 h of chilling, there were no differences in broiler breast muscle p H among the four treatments. These results indicated that a flow-through gassing chamber may be a feasible, although slower, method of performing gas killing as compared to an immersion system. (Key words: gas killing, argon, carbon dioxide, nitrogen, breast muscle pH)
GAS KILLING SYSTEM FOR BROILERS
41.3 cm
FIGURE 1. Schematic drawing of flow-through broiler gassing box. Gas (either Ar, C0 2 , or N2) enters from the left side at 100 KPa. Box has internal dimensions of 41.3 x 30.5 x 28.0 cm, for a volume of .035 m3. Baffle (left side of box) is 2.5 cm from inner side of box.
is pumped through a chamber in which the bird has already been placed. Such a system could be more applicable for commercial conditions requiring a continuous killing system. A potential disadvantage of a flow-through system would be the time lag between first introduction of the animal to the gas and the induction of anoxia. Nitrogen is a possible alternative to Ar, as it is less expensive, more readily available, and should also produce anoxia. However, because N2 is not as dense as air, it is less suitable for use in an immersion system. The purpose of this project was to evaluate and compare the individual use of Ar, CO2, and N2 in a flow-through gas killing system for broilers. MATERIALS AND METHODS Gas Killing System The gas killing system was comprised of a plexiglass box (approximate internal dimensions of 41.3 x 30.5 x 28.0 cm with a nominal volume of .035 m 3 ) with 1.9-cm diameter inlet and exhaust ports on opposite sides of the box (Figure 1). Within the box, on the gas inlet side was a baffle, 28.0 x 25.4 cm, mounted 2.5 cm from the inside of the box. This plexiglass baffle was designed to prevent the direct impinge-
Downloaded from http://ps.oxfordjournals.org/ by guest on May 12, 2015
The use of Ar has also been associated with an early and rapid breast muscle pH decline, which led to a reported increase in breast meat tenderness at 2 h postmortem over electrically stunned broilers (Mohan Raj et al, 1991b). Gregory (1992) also reported that with Ar killing there is some struggling but no stress because the struggle occurs after the bird is unconscious. Mohan Raj et al. (1990a) reported that there were no initial adverse reactions to placing the birds into an Ar environment (i.e., a low-oxygen environment), as is usually evident with placing birds into a CO2 environment. They also reported that with C0 2 stunning, birds lost consciousness between 21 and 30 s, but did not start struggling for 45 s, indicating that the C0 2 produced anaesthesia prior to struggle. The authors also found that there was no correlation between eye closure and loss of somatosensory evoked potentials (SEP), but because eye closure occurred shortly after loss of SEP, the observation of eye closure time could be used to determine when the bird lost consciousness. However, in Ar-induced anoxia, the closing of the eyes occurs before the loss of SEP (Mohan Raj et al, 1991a), which would mean that this visual criterion might not be a good indicator of unconsciousness. In the same study, the researchers noted that the suppression of the electroencephalogram and the loss of SEP occurred prior to convulsions; therefore the birds were unconscious when the convulsions started. All previous work cited was accomplished using two different types of immersion gassing chambers using either CO2 or Ar gas, both of which are denser than air. The first was an immersion system, in which the gas concentration was established in the chamber, and the bird or birds were lowered into it (Mohan Raj and Gregory, 1990b). The other method that was used consisted of a reservoir box, filled with the appropriate concentration of gases, located on top of the stunning chamber. After the bird was placed in the stunning chamber, the gas mixture from the reservoir was released into the stunning section (Mohan Raj et al, 1990a). An alternate system could be a flowthrough chamber system, in which the gas
1219
1220
POOLE AND FLETCHER
Experiment 1 In each of three replicate trials consisting of 24 broiler chickens, live birds were obtained from the holding area of a commercial processing plant, and held for 4 d with free access to feed and water. The feed and water were removed from the birds 8 h prior to slaughter. The birds were divided equally between the three treatments of Ar, N 2 , and C 0 2 gas killing. The birds were timed, in seconds, for the first visual appearance of the following responses: time of first reaction to the gas, loss of posture, eye closure, onset of struggle, and cessation of respiration.
Experiment 2 Birds were obtained at 1 d of age from a commercial hatchery and were grown to market age under simulated commercial conditions. Four replicate trials were conducted using 100 birds in Trial 1 and 60 birds each in Trials 2 to 4. The birds were removed from feed 8 h before processing. The birds were divided equally between four killing treatments using Ar, N2, C 0 2 , and a control treatment of no stunning. The birds were killed, weighed, bled for 3 min in killing cones, and reweighed to determine percentage blood loss by weight difference. Bleeding followed a standard unilateral neck cut to sever both the carotid artery and
the jugular vein (within 20 s of death for gas-killed birds). The carcasses were then sampled for initial breast muscle pH at approximately 15 min post-mortem, chilled in a static ice slush for 24 h and then resampled for terminal breast muscle pH. Determination of p H was done using the iodoacetate method described by Jeacocke (1977).
Statistical Analysis In Experiment 1, the model was a 3 x 3 x 8 arrangement of replication by treatment by number of birds for a total n of 72 birds. The data were analyzed using analysis of variance in the General Linear Models procedure of SAS®, and the means separated using the Duncan's Multiple Range option (SAS Institute, 1988). Main effects of replication, treatment, and replication by treatment were tested by residual error. Because there was no significant replication by treatment interaction, the data were presented by treatment with significance determined using the residual mean square error. In Experiment 2, treatment, replication, and treatment by replication interaction were tested by residual error. Because the treatment by replication interaction was significant for blood loss, that treatment effect was tested by the interaction mean square error. For both 15-min a n d 24-h p H data, there were no interactions and the treatment effects were determined by residual error. For all analyses, significance was determined at P < .05.
RESULTS AND DISCUSSION Experiment 1 Broilers reacted very quickly to the C 0 2 flow, as evidenced by a first reaction time of 7 s (Table 1). Initial reaction times were significantly slower for either the Ar or N 2 gasses at 24 and 22 s, respectively. Carbon dioxide resulted in loss of posture and eye closure in 20 and 28 s, respectively. Nitrogen resulted in loss of posture and eye closure significantly slower than C 0 2 , at 25 and 34 s, respectively, but quicker than Ar at 30 and 40 s, respectively. There were no differences in the time to struggle between the Ar and N 2 gasses (36 and 33 s,
Downloaded from http://ps.oxfordjournals.org/ by guest on May 12, 2015
ment of the gas on the bird and to aid in mixing the incoming gas. Birds were first placed into a wire mesh cage constructed to the same internal dimensions of the gas box. This wire cage was designed to reduce possible damage to both the bird and the gas box during the death struggle. Preliminary trials were conducted to determine a suitable average gas flow condition for the three gasses, Ar, C 0 2 , and N 2 . It was found that a gas inlet pressure of 100 KPa through the inlet orifice took approximately 35 to 40 s to bring the oxygen concentration in the chamber below the 2% required to produce anoxia as reported by Mohan Raj et al. (1990b). In all subsequent experiments, individual birds were gas killed by applying the pure test gas at 100 KPa constant flow for 2 min.
1221
GAS KILLING SYSTEM FOR BROILERS
TABLE 1. Time (± SEM), to first reaction to the gas (FIRST), loss of posture (LOP), eye closure (EYE), onset of struggle (STRUG), and death (DEATH), for broilers killed with argon, carbon dioxide, and nitrogen gasses in a flow-through gassing system (n = 24) Gas
FIRST
LOP
EYE
STRUG
(s) Argon 24 ± 1.1» 30 ± 1.4" 40 ± 2.2» 36 ± Carbon dioxide 7 ± .4b 20 ± .8C 28 ± l.C 27 ± Nitrogen 22 ± .8» 25 ± 1.4b 34 ± i.gb 33 ± '"Values within a column with no common superscript differ significantly
1.3» 79 ± 1.6 1.1>> 75 ± 1.5 .9a 75 ± 1 - 7 (P < .05).
of action is primarily displacement of oxygen and induction of anoxia. Differences between the two gasses could primarily be due to their relative densities (Ar = 1.784 g/L and N 2 = 1.251 g/L at 0 C), affecting the rate of displacement in both the kill chamber and within the birds' respiratory systems. Mohan Raj et al. (1990b) reported the observation that broilers killed with Ar exhibited more wing flapping than those killed with C0 2 . Similar results were observed in this study. Although C0 2 concentrations were greater in this study, the birds that were killed with Ar and N 2 exhibited a more severe death struggle with violent wing flapping and muscular contractions. Experiment 2 The nonstunned control treatment had significantly higher blood loss of 3.7%, as compared to the other treatments (Table 2). Blood losses for the birds killed with Ar, CO2, and N 2 averaged 3.2% and were not different from each other. Blood loss for the control birds is in agreement with previously published values (Newell and Shaffner, 1950; Kotula and Helbacka, 1966; Harris and Carter, 1977). The blood losses for birds killed with Ar and C0 2 (3.3 and 3.2%, respectively) were comparable to those reported by Mohan Raj and Gregory (1991). The initial breast muscle pH values were significantly higher for the birds killed with C02 (6.55), followed by the control birds (6.49), and were lowest for the birds killed with Ar and N^ which were not different from each other (6.40 and 6.44, respectively). These results are similar in pattern to those reported by Mohan Raj et al. (1990b), who reported that the pH at 20 min postmortem was 5.93 for broilers killed with Ar,
Downloaded from http://ps.oxfordjournals.org/ by guest on May 12, 2015
respectively), but both were significantly longer than for C0 2 at 27 s. The times to death, as determined by cessation of respiration, ranged between 75 and 79 s for the three gasses and were not different. The time to death of 75 s for C0 2 in the flow-through box was similar to the 76 s reported by Mohan Raj et al. (1992) using an immersion system. However, it should be noted that the order of events prior to death was somewhat different in the two studies. Mohan Raj et al. (1992) reported that loss of posture preceded eye closure by about 8 s, followed by the onset of the "clonic phase" (vigorous struggle) 4 s later. Using the flowthrough gas exchange in this study, loss of posture and onset of struggle preceded eye closure by approximately 8 and 1 s, respectively. As with C0 2 , the use of Ar in a flowthrough system showed similar responses to those reported by Mohan Raj et al. (1991a) using an immersion system. In the immersion system, there was a similar order of behavioral reactions for the two gases, with the struggle following eye closure. However, for the flow-through system, the order was reversed, with eye closure following the onset of struggle. Mohan Raj et al. (1991a) reported that death due to immersion in Ar took approximately 62 s as compared with 79 s for the flow-through design used in this study. This difference in the times can be attributed to the delay of the flow-through system to reduce the oxygen concentration in the chamber to a level suitable to induce anoxia in the bird. Nitrogen and Ar resulted in similar times for the first reaction to the gas, onset of struggle and death. However, N2 resulted in a significantly quicker loss of posture and eye closure than the Ar. These results are not unexpected because both Ar and N 2 are physiologically neutral gasses whose mode
DEATH
1222
POOLE AND FLETCHER
TABLE 2. Blood loss, initial breast muscle pH (15 min), and terminal breast muscle pH (24 h) (± SEM) for broilers killed with argon, carbon dioxide, and nitrogen gasses in a flow-through system, or killed without stunning (Control), (n = 70) PH Gas
Blood loss
15 min
24 h
Argon Carbon dioxide Nitrogen Control
(%) 3.3 ± 3.2 ± 3.2 ± 3.7 ±
6.40 6.55 6.44 6.49
5.69 5.73 5.70 5.72
.lb .l b .l b .1*
± ± ± ±
.02^ .02" .02c .02b
± ± ± ±
.02 .01 .02 .02
a_c
Values within a column with no common superscript differ significantly (P < .05).
ACKNOWLEDGMENTS This research was supported in part by state and Hatch funds allocated to the Georgia Agricultural Experiment Station.
Appreciation is extended to Tony Reed of Airco Inc., Gainesville, GA 30501, for his technical and material support. The authors also wish to thank Reginald Smith and Petri Papinaho for their laboratory assistance.
REFERENCES Albright, G. W., 1991. Identifying field and plantcaused downgrades. Proceedings of the 26th National Meeting on Poultry Health and Condemnations, October 17-18, Ocean City, MD. Flemming, B. K., G. W. Froning, M. M. Beck, and A. A. Sosnicki, 1991. The effect of carbon dioxide as a preslaughter stunning method for turkeys. Poultry Sci. 70:2201-2206. Gregory, N. G., 1992. Stunning of broilers. Pages 345-349 in: Proceedings of XIX World's Poultry Congress. Vol. 2. Amsterdam, The Netherlands. Gregory, N. G., and L. J. Wilkins, 1989. Effect of stunning current on carcase quality in chickens. Vet. Rec. 124:530-532. Harris, C. E., and T. A. Carter, 1977. Broiler blood losses with manual and mechanical killers. Poultry Sci. 56:1827-1831. Jeacocke, R. E., 1977. Continuous measurements of the pH of beef muscle in intact beef carcasses. J. Food Technol. 12:375-386. Kotula, A. W., E. E. Drewniak, and L. L. Davis, 1957. Effect of carbon dioxide immobilization on the bleeding of chickens. Poultry Sci. 36:585-589. Kotula, A. W., and N. V. Helbacka, 1966. Blood volume of live chickens and influence of slaughter technique on blood loss. Poultry Sci. 45:684-688. Ma, R.T.-I., and P. B. Addis, 1973. The association of struggle during exsanguination to glycolysis, protein solubility and shear in turkey pectoralis muscle. J. Food Sci. 38:995-997. Mohan Raj, A. B., and N. G. Gregory, 1990a. Effect of rate of induction of carbon dioxide anaesthesia on the time of onset of unconsciousness and convulsions. Res. Vet. Sci. 49:360-363. Mohan Raj, A. B., and N. G. Gregory, 1990b. Investigation into the batch stunning/killing of
Downloaded from http://ps.oxfordjournals.org/ by guest on May 12, 2015
as compared with 6.34 for broilers killed with C 0 2 . However, the relatively low initial pH reported by Mohan Raj et al. (1990b) for Ar and the large apparent difference with C 0 2 were not seen in this study. Terminal breast muscle pH values (24 h) averaged 5.71 and were not affected by the killing treatment. In contrast, Mohan Raj et al. (1990b) reported that after 24 h, birds killed with Ar exhibited significantly higher pH values than birds killed with C 0 2 (5.81 and 5.77, respectively). The differences in early post-mortem muscle pH values between C 0 2 and Ar are consistent with differences in the severity of death struggle. Increased struggle at time of death has been documented to result in accelerated rates of early pH decline (Ma and Addis, 1973). It is difficult to determine whether the differences noted between this study and those reported by Mohan Raj et al. (1990b) are due to the application method of the gas killing (immersion vs flow-through system) or are if they are due to other differences inherent in the production, handling, and slaughter of broilers between the United Kingdom and the United States. Although these results do not necessarily support the overall use of gas killing systems for broilers, they do indicate that a flow-through system may be a feasible alternative to immersion systems. Also, because Ar and N 2 both function by the induction of anoxia, N 2 may be a suitable alternative to Ar for application in a continuous system.
GAS KILLING SYSTEM FOR BROILERS chickens using carbon dioxide or argon induced hypoxia. Res. Vet. Sci. 49:364-366. Mohan Raj, A. B., and N. G. Gregory, 1991. Efficiency of bleeding of broilers after gaseous or electrical stunning. Vet. Rec. 128:127-128. Mohan Raj, A. B., N. G. Gregory, and S. B. Wotton, 1990a. Effect of carbon dioxide stunning on somatosensory evoked potentials in hens. Res. Vet. Sci. 49:355-359. Mohan Raj, A. B., N. G. Gregory, and S. B. Wotton, 1991a. Changes in the somatosensory evoked potentials and spontaneous electroencephalogram of hens during stunning in argon-induced anoxia. Br. Vet. J. 147:322-330. Mohan Raj, A. B., T. C. Grey, A. R. Audsely, and N. G. Gregory, 1990b. Effect of electrical and
1223
gaseous stunning on the carcase and meat quality of broilers. Br. Poult. Sci. 31:725-733. Mohan Raj, A. B., T. C. Grey, and N. G. Gregory, 1991b. Effect of early filleting on the texture of breast muscle of broilers stunned with argoninduced anoxia. Br. Poult. Sci. 32:319-325. Mohan Raj, A. B., S. B. Wotton, and N. G. Gregory, 1992. Changes in the somatosensory evoked potentials and spontaneous electroencephalogram of hens during stunning with a carbon dioxide and argon mixture. Br. Vet. J. 148: 147-156. Newell, G. W., and C. S. Shaffner, 1950. Blood loss by chickens during killing. Poultry Sci. 29:271-275. SAS Institute, 1988. SAS/STAT® User's Guide. Release 6.03 Edition. SAS Institute Inc., Cary, NC.
Downloaded from http://ps.oxfordjournals.org/ by guest on May 12, 2015
1995 Poultry Science 74:1218-1223
INTRODUCTION Electrical stunning has received considerable attention in recent years concerning the relative benefits of automated killing and humane slaughter vs stunningrelated carcass damage. Ma and Addis (1973) reported that electrical stunning of turkeys causes an increase in wing and keel breakages. Mohan Raj et al. (1990b) reported that 55% of birds stunned electrically and then mechanically plucked had broken bones. Gregory and Wilkins (1989) found that with an increase in stunning current between 74 to 269 mA per bird there was a corresponding increase in birds with one or more broken bones.
Received for publication August 8, 1994. Accepted for publication February 14, 1995. ^ o whom correspondence should be addressed.
The use of gas killing systems has been proposed as a possible alternative to electrical stunning systems. The potential advantages of a gas killing system would be to reduce the carcass damage associated with electrically stunned birds (Mohan Raj and Gregory, 1990a), as well as the fact that gas stunning could be performed prior to hanging (Mohan Raj and Gregory, 1990b), thus reducing another major source of carcass damage (Albright, 1991). There have been various benefits reported with the use of a gas kill system. Carbon dioxide has been associated with increased blood loss (Kotula et al, 1957), decreased initial breast muscle pH (Mohan Raj et al, 1990b; Flemming et al, 1991) and terminal breast muscle p H (Mohan Raj et al, 1990b), decreased broken bones, and decreases in both superficial and deep muscle hemorrhaging in broiler chickens (Mohan Raj et al, 1990b).
1218
Downloaded from http://ps.oxfordjournals.org/ by guest on May 12, 2015
ABSTRACT Argon, C0 2 , and N 2 gasses were each evaluated in a broiler chicken gas killing system. Birds were killed by individual exposure to one of the three gasses for 2 min in a flow-through system. The gasses were evaluated by determining the time, in seconds, for the following responses: first reaction to the gas, loss of posture, eye closure, initiation of death struggle, and cessation of respiration. Percentage blood loss over a 3-min bleed time was determined by weight loss. Breast muscle pH values were determined at 15 min and 24 h post-mortem on the Pectoralis major muscle. Carbon dioxide resulted in the earliest first reaction, loss of posture, eye closure, and initiation of struggle. Argon and N 2 exhibited a delayed first reaction, a less severe early reaction, but a more severe unconscious death struggle. All birds died in approximately 75 s. Results indicate that the flow-through gas system takes longer to kill broilers than the immersion systems previously reported. Gas killing resulted in lower (P < .05) blood loss. Initial breast muscle pH values were significantly highest for the birds killed with C 0 2 , followed by the control treatment, which was significantly higher than the values for broilers killed with either Ar or N 2 . After 24 h of chilling, there were no differences in broiler breast muscle p H among the four treatments. These results indicated that a flow-through gassing chamber may be a feasible, although slower, method of performing gas killing as compared to an immersion system. (Key words: gas killing, argon, carbon dioxide, nitrogen, breast muscle pH)
GAS KILLING SYSTEM FOR BROILERS
41.3 cm
FIGURE 1. Schematic drawing of flow-through broiler gassing box. Gas (either Ar, C0 2 , or N2) enters from the left side at 100 KPa. Box has internal dimensions of 41.3 x 30.5 x 28.0 cm, for a volume of .035 m3. Baffle (left side of box) is 2.5 cm from inner side of box.
is pumped through a chamber in which the bird has already been placed. Such a system could be more applicable for commercial conditions requiring a continuous killing system. A potential disadvantage of a flow-through system would be the time lag between first introduction of the animal to the gas and the induction of anoxia. Nitrogen is a possible alternative to Ar, as it is less expensive, more readily available, and should also produce anoxia. However, because N2 is not as dense as air, it is less suitable for use in an immersion system. The purpose of this project was to evaluate and compare the individual use of Ar, CO2, and N2 in a flow-through gas killing system for broilers. MATERIALS AND METHODS Gas Killing System The gas killing system was comprised of a plexiglass box (approximate internal dimensions of 41.3 x 30.5 x 28.0 cm with a nominal volume of .035 m 3 ) with 1.9-cm diameter inlet and exhaust ports on opposite sides of the box (Figure 1). Within the box, on the gas inlet side was a baffle, 28.0 x 25.4 cm, mounted 2.5 cm from the inside of the box. This plexiglass baffle was designed to prevent the direct impinge-
Downloaded from http://ps.oxfordjournals.org/ by guest on May 12, 2015
The use of Ar has also been associated with an early and rapid breast muscle pH decline, which led to a reported increase in breast meat tenderness at 2 h postmortem over electrically stunned broilers (Mohan Raj et al, 1991b). Gregory (1992) also reported that with Ar killing there is some struggling but no stress because the struggle occurs after the bird is unconscious. Mohan Raj et al. (1990a) reported that there were no initial adverse reactions to placing the birds into an Ar environment (i.e., a low-oxygen environment), as is usually evident with placing birds into a CO2 environment. They also reported that with C0 2 stunning, birds lost consciousness between 21 and 30 s, but did not start struggling for 45 s, indicating that the C0 2 produced anaesthesia prior to struggle. The authors also found that there was no correlation between eye closure and loss of somatosensory evoked potentials (SEP), but because eye closure occurred shortly after loss of SEP, the observation of eye closure time could be used to determine when the bird lost consciousness. However, in Ar-induced anoxia, the closing of the eyes occurs before the loss of SEP (Mohan Raj et al, 1991a), which would mean that this visual criterion might not be a good indicator of unconsciousness. In the same study, the researchers noted that the suppression of the electroencephalogram and the loss of SEP occurred prior to convulsions; therefore the birds were unconscious when the convulsions started. All previous work cited was accomplished using two different types of immersion gassing chambers using either CO2 or Ar gas, both of which are denser than air. The first was an immersion system, in which the gas concentration was established in the chamber, and the bird or birds were lowered into it (Mohan Raj and Gregory, 1990b). The other method that was used consisted of a reservoir box, filled with the appropriate concentration of gases, located on top of the stunning chamber. After the bird was placed in the stunning chamber, the gas mixture from the reservoir was released into the stunning section (Mohan Raj et al, 1990a). An alternate system could be a flowthrough chamber system, in which the gas
1219
1220
POOLE AND FLETCHER
Experiment 1 In each of three replicate trials consisting of 24 broiler chickens, live birds were obtained from the holding area of a commercial processing plant, and held for 4 d with free access to feed and water. The feed and water were removed from the birds 8 h prior to slaughter. The birds were divided equally between the three treatments of Ar, N 2 , and C 0 2 gas killing. The birds were timed, in seconds, for the first visual appearance of the following responses: time of first reaction to the gas, loss of posture, eye closure, onset of struggle, and cessation of respiration.
Experiment 2 Birds were obtained at 1 d of age from a commercial hatchery and were grown to market age under simulated commercial conditions. Four replicate trials were conducted using 100 birds in Trial 1 and 60 birds each in Trials 2 to 4. The birds were removed from feed 8 h before processing. The birds were divided equally between four killing treatments using Ar, N2, C 0 2 , and a control treatment of no stunning. The birds were killed, weighed, bled for 3 min in killing cones, and reweighed to determine percentage blood loss by weight difference. Bleeding followed a standard unilateral neck cut to sever both the carotid artery and
the jugular vein (within 20 s of death for gas-killed birds). The carcasses were then sampled for initial breast muscle pH at approximately 15 min post-mortem, chilled in a static ice slush for 24 h and then resampled for terminal breast muscle pH. Determination of p H was done using the iodoacetate method described by Jeacocke (1977).
Statistical Analysis In Experiment 1, the model was a 3 x 3 x 8 arrangement of replication by treatment by number of birds for a total n of 72 birds. The data were analyzed using analysis of variance in the General Linear Models procedure of SAS®, and the means separated using the Duncan's Multiple Range option (SAS Institute, 1988). Main effects of replication, treatment, and replication by treatment were tested by residual error. Because there was no significant replication by treatment interaction, the data were presented by treatment with significance determined using the residual mean square error. In Experiment 2, treatment, replication, and treatment by replication interaction were tested by residual error. Because the treatment by replication interaction was significant for blood loss, that treatment effect was tested by the interaction mean square error. For both 15-min a n d 24-h p H data, there were no interactions and the treatment effects were determined by residual error. For all analyses, significance was determined at P < .05.
RESULTS AND DISCUSSION Experiment 1 Broilers reacted very quickly to the C 0 2 flow, as evidenced by a first reaction time of 7 s (Table 1). Initial reaction times were significantly slower for either the Ar or N 2 gasses at 24 and 22 s, respectively. Carbon dioxide resulted in loss of posture and eye closure in 20 and 28 s, respectively. Nitrogen resulted in loss of posture and eye closure significantly slower than C 0 2 , at 25 and 34 s, respectively, but quicker than Ar at 30 and 40 s, respectively. There were no differences in the time to struggle between the Ar and N 2 gasses (36 and 33 s,
Downloaded from http://ps.oxfordjournals.org/ by guest on May 12, 2015
ment of the gas on the bird and to aid in mixing the incoming gas. Birds were first placed into a wire mesh cage constructed to the same internal dimensions of the gas box. This wire cage was designed to reduce possible damage to both the bird and the gas box during the death struggle. Preliminary trials were conducted to determine a suitable average gas flow condition for the three gasses, Ar, C 0 2 , and N 2 . It was found that a gas inlet pressure of 100 KPa through the inlet orifice took approximately 35 to 40 s to bring the oxygen concentration in the chamber below the 2% required to produce anoxia as reported by Mohan Raj et al. (1990b). In all subsequent experiments, individual birds were gas killed by applying the pure test gas at 100 KPa constant flow for 2 min.
1221
GAS KILLING SYSTEM FOR BROILERS
TABLE 1. Time (± SEM), to first reaction to the gas (FIRST), loss of posture (LOP), eye closure (EYE), onset of struggle (STRUG), and death (DEATH), for broilers killed with argon, carbon dioxide, and nitrogen gasses in a flow-through gassing system (n = 24) Gas
FIRST
LOP
EYE
STRUG
(s) Argon 24 ± 1.1» 30 ± 1.4" 40 ± 2.2» 36 ± Carbon dioxide 7 ± .4b 20 ± .8C 28 ± l.C 27 ± Nitrogen 22 ± .8» 25 ± 1.4b 34 ± i.gb 33 ± '"Values within a column with no common superscript differ significantly
1.3» 79 ± 1.6 1.1>> 75 ± 1.5 .9a 75 ± 1 - 7 (P < .05).
of action is primarily displacement of oxygen and induction of anoxia. Differences between the two gasses could primarily be due to their relative densities (Ar = 1.784 g/L and N 2 = 1.251 g/L at 0 C), affecting the rate of displacement in both the kill chamber and within the birds' respiratory systems. Mohan Raj et al. (1990b) reported the observation that broilers killed with Ar exhibited more wing flapping than those killed with C0 2 . Similar results were observed in this study. Although C0 2 concentrations were greater in this study, the birds that were killed with Ar and N 2 exhibited a more severe death struggle with violent wing flapping and muscular contractions. Experiment 2 The nonstunned control treatment had significantly higher blood loss of 3.7%, as compared to the other treatments (Table 2). Blood losses for the birds killed with Ar, CO2, and N 2 averaged 3.2% and were not different from each other. Blood loss for the control birds is in agreement with previously published values (Newell and Shaffner, 1950; Kotula and Helbacka, 1966; Harris and Carter, 1977). The blood losses for birds killed with Ar and C0 2 (3.3 and 3.2%, respectively) were comparable to those reported by Mohan Raj and Gregory (1991). The initial breast muscle pH values were significantly higher for the birds killed with C02 (6.55), followed by the control birds (6.49), and were lowest for the birds killed with Ar and N^ which were not different from each other (6.40 and 6.44, respectively). These results are similar in pattern to those reported by Mohan Raj et al. (1990b), who reported that the pH at 20 min postmortem was 5.93 for broilers killed with Ar,
Downloaded from http://ps.oxfordjournals.org/ by guest on May 12, 2015
respectively), but both were significantly longer than for C0 2 at 27 s. The times to death, as determined by cessation of respiration, ranged between 75 and 79 s for the three gasses and were not different. The time to death of 75 s for C0 2 in the flow-through box was similar to the 76 s reported by Mohan Raj et al. (1992) using an immersion system. However, it should be noted that the order of events prior to death was somewhat different in the two studies. Mohan Raj et al. (1992) reported that loss of posture preceded eye closure by about 8 s, followed by the onset of the "clonic phase" (vigorous struggle) 4 s later. Using the flowthrough gas exchange in this study, loss of posture and onset of struggle preceded eye closure by approximately 8 and 1 s, respectively. As with C0 2 , the use of Ar in a flowthrough system showed similar responses to those reported by Mohan Raj et al. (1991a) using an immersion system. In the immersion system, there was a similar order of behavioral reactions for the two gases, with the struggle following eye closure. However, for the flow-through system, the order was reversed, with eye closure following the onset of struggle. Mohan Raj et al. (1991a) reported that death due to immersion in Ar took approximately 62 s as compared with 79 s for the flow-through design used in this study. This difference in the times can be attributed to the delay of the flow-through system to reduce the oxygen concentration in the chamber to a level suitable to induce anoxia in the bird. Nitrogen and Ar resulted in similar times for the first reaction to the gas, onset of struggle and death. However, N2 resulted in a significantly quicker loss of posture and eye closure than the Ar. These results are not unexpected because both Ar and N 2 are physiologically neutral gasses whose mode
DEATH
1222
POOLE AND FLETCHER
TABLE 2. Blood loss, initial breast muscle pH (15 min), and terminal breast muscle pH (24 h) (± SEM) for broilers killed with argon, carbon dioxide, and nitrogen gasses in a flow-through system, or killed without stunning (Control), (n = 70) PH Gas
Blood loss
15 min
24 h
Argon Carbon dioxide Nitrogen Control
(%) 3.3 ± 3.2 ± 3.2 ± 3.7 ±
6.40 6.55 6.44 6.49
5.69 5.73 5.70 5.72
.lb .l b .l b .1*
± ± ± ±
.02^ .02" .02c .02b
± ± ± ±
.02 .01 .02 .02
a_c
Values within a column with no common superscript differ significantly (P < .05).
ACKNOWLEDGMENTS This research was supported in part by state and Hatch funds allocated to the Georgia Agricultural Experiment Station.
Appreciation is extended to Tony Reed of Airco Inc., Gainesville, GA 30501, for his technical and material support. The authors also wish to thank Reginald Smith and Petri Papinaho for their laboratory assistance.
REFERENCES Albright, G. W., 1991. Identifying field and plantcaused downgrades. Proceedings of the 26th National Meeting on Poultry Health and Condemnations, October 17-18, Ocean City, MD. Flemming, B. K., G. W. Froning, M. M. Beck, and A. A. Sosnicki, 1991. The effect of carbon dioxide as a preslaughter stunning method for turkeys. Poultry Sci. 70:2201-2206. Gregory, N. G., 1992. Stunning of broilers. Pages 345-349 in: Proceedings of XIX World's Poultry Congress. Vol. 2. Amsterdam, The Netherlands. Gregory, N. G., and L. J. Wilkins, 1989. Effect of stunning current on carcase quality in chickens. Vet. Rec. 124:530-532. Harris, C. E., and T. A. Carter, 1977. Broiler blood losses with manual and mechanical killers. Poultry Sci. 56:1827-1831. Jeacocke, R. E., 1977. Continuous measurements of the pH of beef muscle in intact beef carcasses. J. Food Technol. 12:375-386. Kotula, A. W., E. E. Drewniak, and L. L. Davis, 1957. Effect of carbon dioxide immobilization on the bleeding of chickens. Poultry Sci. 36:585-589. Kotula, A. W., and N. V. Helbacka, 1966. Blood volume of live chickens and influence of slaughter technique on blood loss. Poultry Sci. 45:684-688. Ma, R.T.-I., and P. B. Addis, 1973. The association of struggle during exsanguination to glycolysis, protein solubility and shear in turkey pectoralis muscle. J. Food Sci. 38:995-997. Mohan Raj, A. B., and N. G. Gregory, 1990a. Effect of rate of induction of carbon dioxide anaesthesia on the time of onset of unconsciousness and convulsions. Res. Vet. Sci. 49:360-363. Mohan Raj, A. B., and N. G. Gregory, 1990b. Investigation into the batch stunning/killing of
Downloaded from http://ps.oxfordjournals.org/ by guest on May 12, 2015
as compared with 6.34 for broilers killed with C 0 2 . However, the relatively low initial pH reported by Mohan Raj et al. (1990b) for Ar and the large apparent difference with C 0 2 were not seen in this study. Terminal breast muscle pH values (24 h) averaged 5.71 and were not affected by the killing treatment. In contrast, Mohan Raj et al. (1990b) reported that after 24 h, birds killed with Ar exhibited significantly higher pH values than birds killed with C 0 2 (5.81 and 5.77, respectively). The differences in early post-mortem muscle pH values between C 0 2 and Ar are consistent with differences in the severity of death struggle. Increased struggle at time of death has been documented to result in accelerated rates of early pH decline (Ma and Addis, 1973). It is difficult to determine whether the differences noted between this study and those reported by Mohan Raj et al. (1990b) are due to the application method of the gas killing (immersion vs flow-through system) or are if they are due to other differences inherent in the production, handling, and slaughter of broilers between the United Kingdom and the United States. Although these results do not necessarily support the overall use of gas killing systems for broilers, they do indicate that a flow-through system may be a feasible alternative to immersion systems. Also, because Ar and N 2 both function by the induction of anoxia, N 2 may be a suitable alternative to Ar for application in a continuous system.
GAS KILLING SYSTEM FOR BROILERS chickens using carbon dioxide or argon induced hypoxia. Res. Vet. Sci. 49:364-366. Mohan Raj, A. B., and N. G. Gregory, 1991. Efficiency of bleeding of broilers after gaseous or electrical stunning. Vet. Rec. 128:127-128. Mohan Raj, A. B., N. G. Gregory, and S. B. Wotton, 1990a. Effect of carbon dioxide stunning on somatosensory evoked potentials in hens. Res. Vet. Sci. 49:355-359. Mohan Raj, A. B., N. G. Gregory, and S. B. Wotton, 1991a. Changes in the somatosensory evoked potentials and spontaneous electroencephalogram of hens during stunning in argon-induced anoxia. Br. Vet. J. 147:322-330. Mohan Raj, A. B., T. C. Grey, A. R. Audsely, and N. G. Gregory, 1990b. Effect of electrical and
1223
gaseous stunning on the carcase and meat quality of broilers. Br. Poult. Sci. 31:725-733. Mohan Raj, A. B., T. C. Grey, and N. G. Gregory, 1991b. Effect of early filleting on the texture of breast muscle of broilers stunned with argoninduced anoxia. Br. Poult. Sci. 32:319-325. Mohan Raj, A. B., S. B. Wotton, and N. G. Gregory, 1992. Changes in the somatosensory evoked potentials and spontaneous electroencephalogram of hens during stunning with a carbon dioxide and argon mixture. Br. Vet. J. 148: 147-156. Newell, G. W., and C. S. Shaffner, 1950. Blood loss by chickens during killing. Poultry Sci. 29:271-275. SAS Institute, 1988. SAS/STAT® User's Guide. Release 6.03 Edition. SAS Institute Inc., Cary, NC.
Downloaded from http://ps.oxfordjournals.org/ by guest on May 12, 2015