Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Assessment of distress associated with carbon dioxide euthanasia in laboratory rats 2006

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
ubc_2006-200599.pdf
ubc_2006-200599.pdf [ 6.53MB ]
Metadata
JSON: 1.0093022.json
JSON-LD: 1.0093022+ld.json
RDF/XML (Pretty): 1.0093022.xml
RDF/JSON: 1.0093022+rdf.json
Turtle: 1.0093022+rdf-turtle.txt
N-Triples: 1.0093022+rdf-ntriples.txt
Citation
1.0093022.ris

Full Text

ASSESSMENT OF DISTRESS ASSOCIATED WITH CARBON DIOXIDE EUTHANASIA IN LABORATORY RATS by LEE ERIN NIEL B.Sc, Simon Fraser University, 2000 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Animal Science) The University of British Columbia July 2006 © Lee Erin Niel, 2006 Abstract C a r b o n d iox ide (CO2) gas is the most w i d e l y used euthanasia agent for laboratory rodents. H o w e v e r , it has the potential to cause both pa in and dyspnea, an unpleasant sensation o f breathless, w h i l e an imals are st i l l consc ious . The a ims o f this dissertation were to determine whether g radua l - f i l l CO2 euthanasia causes distress i n laboratory rats, and to examine potential sources o f distress, i n c l u d i n g pa in , dyspnea and novelty . The first study examined the behavioural responses o f rats dur ing g radua l - f i l l CO2 euthanasia. Rats showed increased exploratory behaviours and escape behaviours dur ing CO2 euthanasia, suggesting that this procedure does cause distress. The second and th i rd studies used approach-avoidance testing to investigate avers ion to CO2 i n rats, by e x a m i n i n g their w i l l ingness to enter a test cage conta in ing CO2 for access to an attractive food reward. Rats were found to avo id CO2 concentrations that are suff ic ient to cause unconsciousness. Spec i f i ca l l y , w h e n tested w i t h static concentrations o f CO2 rats showed avoidance at 1 5 % CO2 and greater, and w h e n tested w i t h gradual ly increasing concentrations o f CO2 at f l o w rates ranging f r o m 3 to 2 7 % per minute , rats showed avoidance at 13 to 1 6 % CO2. Th i s avoidance indicates that rats are at least moderately averse to CO2 concentrations occur r ing dur ing g radua l - f i l l CO2 euthanasia, and that forced exposure l i k e l y causes distress. Concentrat ions o f CO2 that were associated w i t h behav ioura l responses and avers ion were not consistent w i t h prev ious data o n pa in due to CO2. H o w e v e r , s imi la r concentrations have been shown to cause dyspnea i n humans. The final study examined the role o f nove l ty i n rats' responses to CO2, and found that novel ty was not a major source o f distress dur ing g radua l - f i l l CO2 euthanasia. In summary , these studies suggest that g radua l - f i l l CO2 euthanasia causes distress i n rats, and that this distress is l i k e l y due to dyspnea. Further research is necessary to examine the effects o f CO2 on other rodent species such as m i c e , and to ident i fy alternative methods o f euthanasia that cause unconsciousness wi thout distress. Table of Contents Abstract i i Tab le o f Contents i i i L i s t o f Tables v L i s t o f F igures v i L i s t o f Abbrev ia t ions v i i A c k n o w l e d g e m e n t s v i i i C o - A u t h o r s h i p Statement i x C H A P T E R 1: Genera l Introduction 1 1.1 Introduction 1 1.2 C a r b o n D i o x i d e Euthanasia 3 1.3 A n i m a l Distress and C a r b o n D i o x i d e 9 1.4 Assessment o f Distress D u r i n g CO2 Euthanasia 16 1.5 Object ives •. 27 1.6 References 29 C H A P T E R 2 : Behav iou ra l responses o f rats to g radua l - f i l l carbon d iox ide euthanasia and reduced o x y g e n concentrations 41 2.1 Introduct ion 41 2.2 Mater ia l s and M e t h o d s 43 2.3 Resul ts 47 2.4 D i s c u s s i o n 49 2.5 References 61 C H A P T E R 3 : Rats a v o i d exposure to carbon d iox ide and argon. . . . 66 3.1 Introduction 66 3.2 Mater ia l s and M e t h o d s 67 3.3 Resul ts 71 3.4 D i s c u s s i o n 73 3.5 References 81 C H A P T E R 4: E f fect o f f l o w rate on avers ion to g radua l - f i l l carbon d iox ide euthanasia i n rats 85 4.1 Introduction 85 4.2 Mater ia l s and M e t h o d s 87 4.3 Results 90 4.4 D i s c u s s i o n 91 4.5 References 95 C H A P T E R 5: E f fects o f novel ty on rat responses to CO2 exposure 97 5.1 Introduct ion 97 5.2 Mate r ia l s and M e t h o d s 99 5.3 R e s u l t s , 104 iii 5.4 D i s c u s s i o n 106 5.5 References 113 C H A P T E R 6: Genera l D i s c u s s i o n 116 6.1 Distress i n rats dur ing CO2 euthanasia 116 6.2 Sources o f distress 117 6.3 Cr i t ique o f methods 122 6.3 Future direct ions 125 6.4 C o n c l u s i o n s 129 6.5 References 130 iv List of Tables Table 2.1 Descr ipt ions o f rat behaviours recorded dur ing basel ine and dur ing exposure to CO2 or reduced O2 concentrations 56 Table 2.2 D i f fe rence f r o m basel ine for each o f the f ive behav ioura l responses o f rats dur ing air and CO2 exposure (n =' 8 rats). D a t a are presented as medians w i t h 2 5 t h and 7 5 t h percenti les, and statistical compar isons were made w i t h W i l c o x o n S igned R a n k s Test (T; based on N values >0) 57 Table 2.3 D i f fe rence f r o m basel ine for each o f the f ive behav ioura l responses o f rats dur ing air and exposure to reduced O2 concentrations (n = 8 rats). D a t a are presented as medians w i t h 25th and 75th percenti les, and statistical compar isons were made w i t h the W i l c o x o n S igned R a n k s Test (T; based o n N va lues>0) 58 Table 3.1 M e d i a n (with 2 5 t h and 7 5 t h percenti les) number o f reward items eaten dur ing the entire session and eating and d w e l l i n g t imes dur ing the first entry w i t h either air or argon i n the test cage (n = 9 rats) 79 Table 5.1 Descr ip t ions o f rat behaviours recorded dur ing basel ine and dur ing exposure to CO2 or peppermint odour (Exper iment 3) 110 v List of Figures Figure 2.1 A v e r a g e concentrations o f O2 (open markers) and CO2 ( f i l led markers) i n the chamber dur ing the first 600 s o f the f i l l i n g process. Concentrat ions were taken 5 c m (triangles) and 15 c m (squares) f r o m the chamber bot tom 59 Figure 2.2 Figure 3.1 Figure 4.1 Figure 5.1 Figure 5.2 Responses by rats dur ing the baseline per iod and then dur ing exposure to either air ( f i l led squares) or CO2 (open squares) starting at t = 0. M e d i a n values per 1 5 s per iod are s h o w n for a) act iv i ty , b) rears , c) nose to l i d , d) escape behaviours , and e) voca l i zat ions (n = 8 rats) .60 M e d i a n ( ± interquarti le ranges) a) number o f reward i tems eaten, b) eating t ime ( f i l led squares) and d w e l l i n g t ime (open squares) for the first entry, and c) number o f attempted entries into the test cage dur ing sessions w i t h 0, 5 , 10, 15 and 2 0 % C 0 2 (n = 9 rats) .80 Least squares mean ( ± S E M ) a) number o f reward i tems eaten, b) latency to stop eating (open) and to leave the test cage ( f i l led) , and c) CO2 concentrat ion at the t ime w h e n rats stopped eating (open) and left the test cage ( f i l led) dur ing test sessions w i t h CO2 flow rates o f 3 , 7, 14, 20 and 2 7 % o f the test cage v o l u m e per minute (n = 8 rats) .94 Approach -avo idance responses to CO2 for the first three days and for the final day (Day 16) o f exposure (Exper iment 1). M e a n ( ± S E M ) (a) latency to stop eating and leave the test cage, (b) C 0 2 concentrat ion when rats stopped eating and left the test cage, and (c) number o f reward i tems eaten (n - 9 rats) I l l B e h a v i o u r a l responses o f rats to CO2 euthanasia and peppermint odour exposure (Exper iment 3). Least squares mean ( ± S E M ) (a) number of rears, and (b) t ime spent w i t h the nose i n contact w i t h the test cage l i d dur ing basel ine and dur ing exposure to either C 0 2 (n = 16 rats) or air w i t h peppermint odour (n = 16 rats) 112 VI List of Abbreviations A C T H adrenocort icotropic hormone CNS Centra l nervous system c o 2 C a r b o n d iox ide C R H Cort icot rophin - re leas ing hormone CSF Cerebral spinal f l u i d EEG Elect roencephalogram HPA Hypotha lamic -p i tu i ta ry -adrena l axis o 2 O x y g e n SAM ' Sympathet ic -adrenerg ic -medul lary axis SEP Somatosensory -evoked potential USV Ul t rason ic voca l i za t ion vii Acknowledgements I am grateful to my supervisor, Dan Weary, for his endless enthusiasm and encouragement. Dan has allowed me tremendous freedom to pursue research that I am passionate about. He has also given me every opportunity to learn about all aspects of academia, from writing grants to teaching and training students. I am especially thankful for his constant barrage of questions and criticisms, which have been crucial to my growth as an independent and critical researcher. I have thoroughly enjoyed my disagreements with Dan, and I hope that future colleagues are as forthcoming with their opinions. I would like to thank my committee members, David Fraser, Jim Love and Ken Craig for their insight into my thesis area, and for their help with editing. Dave has been a fount of knowledge on both animal welfare and grammar, and I have learned vast amounts from him during my time in the Animal Welfare Program. I am grateful to Jim and Ken for their scepticism, which has made me think carefully about appropriate interpretation and presentation of my data. ' I would also like to thank a number of people who have made contributions to the studies in this thesis. First and foremost, I want to thank Cassandra Tucker, with whom I have spent countless hours discussing research ideas, experimental design and statistical analysis. I would also like to thank Richard Kirkden, for his insights on animal motivation. Gilles Galzi, Sylvia Leung and Jurgen Pehlke have been incredibly helpful with the day to day running of the lab. Murray Hodgson provided technical knowledge on ultrasound, and very generously lent me equipment for the ultrasound recordings in Chapter 2. Sarah Stewart, Marianne Pfaffinger and Gary Lee provided assistance with day to day lab duties and with data collection (Sarah, Chapters 4 and 5; Marianne, Chapters 3 and 5; Gary, Chapter 5). Mitja Sedlbauer, Nicole Fenwick, and Amanda Grout provided assistance with animal care. I am also thankful to the many members of the Animal Welfare Program for their support and friendship. Lunchtime discussions and presentations have been a high point in my time as a graduate student. I am eternally indebted to my family and friends for their unwavering support. A special thank you goes to my husband, Colin, who has helped me to stay sane throughout this process and has done far more than his fair share of the dog-walking these past few months. Finally, completion of this work would not have been possible without the generous financial support that I received through the National Sciences and Engineering Research Council and the U B C Faculty of Land and Food Systems (Elizabeth Roxann Howland Fellowship, Leonard S. Klink Memorial Fellowship, Beryl March Travel Award and Charles River Laboratories Scholarship). viii Co-Authorship Statement Chapter 2: I w a s responsib le for identifying the research area , exper imental des ign , data col lect ion, statistical ana lyses , and writing the manuscr ipt . D a n W e a r y contributed to the exper imental des ign and to the interpretation and presentation of the manuscr ipt . Chapter 3: I w a s responsib le for identifying the research area , exper imental des ign , data col lect ion, statistical ana lyses , and writing the manuscr ipt . D a n W e a r y contributed to the exper imental des ign and to the interpretation and presentation of the manuscr ipt . Chapter 4: I w a s responsib le for a portion of the data col lect ion, and for identifying the research area , exper imental des ign , statistical ana lyses , and writing the manuscr ipt . S a r a h Stewart contributed to the exper imental des ign and data col lect ion. D a n W e a r y contributed to the exper imental des ign and to the interpretation and presentation of the manuscr ipt . Chapter 5: I w a s responsib le for identifying the research area , exper imental des ign , data col lect ion, statistical ana lyses , and writing the manuscr ipt . D a n W e a r y contributed to the exper imental des ign , and to the interpretation and presentation of the manuscr ipt . ix CHAPTER 1: General Introduction 1.1 Introduction In the industr ia l ized countries, an imals are rout inely bred and used i n sc ient i f ic exper imentat ion that serves to advance basic b i o l o g i c a l knowledge , advance h u m a n and veterinary med ic ine , and ensure the safety o f humans , animals and the envi ronment through regulatory testing. Rodents are the most w i d e l y used species i n this research, and i n C a n a d a alone approx imate ly one m i l l i o n m i c e and rats are used i n research each year ( C C A C , 2006) The vast major i ty o f research and breeding an imals are eventual ly k i l l e d , either for exper imental purposes or to reduce surplus stock. D u r i n g an exper iment , an imals might be k i l l e d i n order to al leviate pa in and suffer ing due to an exper imental manipu lat ion or to a l l o w for tissue co l lec t ion and analysis . The term euthanasia is c o m m o n l y used to refer to the k i l l i n g o f research animals . Euthanasia is der ived f r o m Greek for ' g o o d death ' , w h i c h suggests a process w h i c h does not invo l ve pa in or distress ( B l a c k m o r e , 1993). A l t h o u g h i n practice it m a y not be possib le to develop a procedure for the k i l l i n g o f animals that is complete ly devo id o f stress, the goal is c lear ly to m i n i m i z e any pa in and distress associated w i t h the procedure. A number o f different euthanasia methods are current ly approved for k i l l i n g laboratory rodents i n C a n a d a ( C C A C , 1993). Euthanasia guidel ines developed by the Canad ian C o u n c i l on A n i m a l Care are general ly i n l ine w i t h the suggested euthanasia methods o f other western countr ies such as those o f the U S A ( A V M A , 2001) , the U K ( U K H o m e O f f i c e , 1997), countries o f the European U n i o n (C lose et a l . , 1997) and A u s t r a l i a and N e w Zea land ( A N Z C C A R T , 1993). These methods inc lude phys ica l techniques, injectable anaesthetics, and anaesthetic and non-anaesthetic gases. It is recognized that some o f these methods have the potential to result i n both pa in and distress, but the chosen method does not depend sole ly o n a reduct ion o f an imal 1 suffer ing. It also depends o n pragmatic concerns such as the purpose o f the k i l l i n g and the constraints o f t ime , money , and safety to humans, other an imals and the envi ronment . The Canad ian C o u n c i l o n A n i m a l Care ( C C A C ) , the body that governs an imal research i n Canada , stipulates i n their E th ics o f A n i m a l Investigation guidel ines (1989, p . l ) that research an imals "must not be subjected to unnecessary pa in or d istress" and " i f pa in or distress is necessary concomitant to the study, it must be m i n i m i z e d both i n intensity and durat ion" . Therefore, where mul t ip le methods are avai lable for ach iev ing the same a i m s , the method that causes the least pa in and distress should be used. Furthermore, where a pa in fu l or distressful procedure is required to meet the a ims o f the study, efforts must be made to mit igate these effects. A l t h o u g h these guidel ines a l l o w for pragmatic concerns to be addressed, they require that a n i m a l pa in and distress be m i n i m i z e d to the greatest extent poss ib le w i t h i n these conf ines. In order to meet this requirement, we must first determine w h i c h methods cause p a i n and distress, and then determine h o w these states can be m i n i m i z e d . T o assess whether k i l l i n g methods cause pa in and distress w e can examine three l ines o f ev idence: 1) human self - report data to determine what subjective states are potential ly associated w i t h exposure, 2) phys io log ica l responses to observe the method 's potential for act ivat ing nociceptors or neuro-endocr ine stress responses, and 3) behav ioura l data to assess whether the method results i n behaviours indicat ive o f pa in or distress, and to determine the leve l o f avers ion associated w i t h the st imulus. A l t h o u g h it m a y not be poss ib le to complete ly prevent pa in and distress, it should be possib le to ident i fy those procedures that result i n the least pa in and distress, and thus prov ide a basis for recommendat ions that can be used by an imal caregivers. 2 1.2 Carbon Dioxide Euthanasia C a r b o n d iox ide (CO2) is the most w i d e l y used euthanasia agent for laboratory rodents. In CO2 euthanasia o f rodents, the a n i m a l is either p laced into a chamber that is p re - f i l l ed w i t h CO2 at a concentrat ion o f greater than 7 0 % , or the an imal is p laced i n an empty chamber that is then gradual ly f i l l e d w i t h CO2. A l t h o u g h some prev ious w o r k has examined whether CO2 euthanasia causes distress i n rats, the results to date have been inconc lus ive . 1.2.1 M o d e o f a c t i o n A l t h o u g h CO2 is categorized as a non-anaesthetic gas, it does have anaesthetic properties. In the rat, it causes a decrease i n bra in exc i tabi l i ty at concentrations as l o w as 5 % , l ight anaesthesia beg inn ing at 2 5 % , and deeper anaesthesia at approx imate ly 4 0 % (rev iewed by W o o d b u r y et a l . , 1958). H u m a n s show electroencephalogram ( E E G ) depression and are unable to carry out s imple commands w h e n breathing 20% C 0 2 ( M e y e r et a l . , 1966). Death can occur at concentrations o f approx imately 3 0 % CO2 and higher (Danneman et a l . , 1997; Sharp et a l . , 2006) , l i k e l y due to depression o f brain centers responsible for c i rcu lat ion and respirat ion; the exact CO2 concentrat ion necessary for death depends on the durat ion o f exposure (Danneman et a l , 1997). D u r i n g metabo l i sm, the body uses oxygen (O2) and produces CO2 and these two gases are constantly exchanged by the respiratory system. A i r contains approx imate ly 0 . 0 3 % CO2 and 2 0 . 9 % O2 (and 7 8 % nitrogen), but due to metabo l i sm, the tissues conta in higher levels o f CO2 and lower levels o f O2. T h i s results i n a gas concentrat ion gradient between the air and the b lood at the lungs that a l lows for the uptake o f O2 into the b l o o d and release o f CO2 into the air, and between the b l o o d and the tissues i n the rest o f the body that a l l ows for the release o f O2 to the tissues and the uptake o f C 0 2 into the b lood . Because o f this gradient, C 0 2 general ly occurs 3 only at low concentrations (<6%) in the body. Total CO2 content in the body consists of carbamino compounds, CO2, bicarbonate ions (HCO3") and carbonic acid (H2CO3). The last three exist in the following equilibrium: C 0 2 + H 2 0 «-+ H2CO3 H + + HCO3" A s the concentration of CO2 in the air is increased, it builds up in the tissues and shifts this equilibrium to the right, producing hydrogen ions and reducing p H . Although the body has adaptations that increase the removal of C 0 2 at the lungs under normal conditions, the concentrations of CO2 used in euthanasia are far beyond the body's compensatory mechanisms. Furthermore, the blood brain barrier is very permeable to CO2, and the C S F has little buffering capacity. This results in a greater drop in p H in the C S F than in the blood and other tissues (Brodie & Woodbury, 1958), and it is thought that this reduction in p H causes CO2 narcosis and death (e.g. Brodie & Woodbury 1958; Eisele et al., 1967; Martoft et al. 2003; Meyer et al., 1961; Woodbury et al., 1958). Specifically, these changes in C S F p H are thought to reduce neuron excitability. For example, in vitro studies have demonstrated that a reduction in extracellular p H due to exposure 20% CO2 reduces the activity of neurons in the hippocampus via a p H dependent mechanism (e.g. Dulla et al., 2005; Hsu et al., 2000; Lee et al., 1996; Velisek, 1998). The exact mechanisms for these effects on neuron functioning are not currently known, but p H effects are thought to be mediated by alterations in cell ion gradients (e.g. Gifford et al., 1990; Pasternack et al., 1992; Tang et al., 1990; Tombaugh and Somjen, 1996), possibly through reconfiguration of relevant proteins associated with cell ion channels, receptors or enzymes (Somero, 1986). 4 1.2.2 Time-course of CO2 euthanasia in rats A number o f studies have observed the t ime course o f CO2 euthanasia i n rats, and this in fo rmat ion prov ides an ind icat ion o f the durat ion o f exposure and the m a x i m u m concentrations o f CO2 that occur w h i l e the an imal is st i l l consc ious 1 . These results have been obtained for pre- f i l l and g radua l - f i l l CO2 euthanasia o f rats. Authors d i f fer i n their cr i ter ia for loss o f consciousness, but the major i ty o f experiments use onset o f atax ia to indicate the in i t ia l depression o f the central nervous system ( C N S ) , and complete loss o f posture to indicate loss o f consciousness. Some studies e x a m i n i n g species other than rats have used E E G s (Raj and Gregory , 1994; Ra j et a l . , 1997; Ra j et a l . , 1998; Mar to f t et a l . , 2002) , and somatosensory- evoked potentials ( S E P s ) (Raj and Gregory , 1994; Raj et a l . , 1997; Mar to f t et a l . , 2002) to gain a more accurate estimate o f t inie to unconsciousness, and there appears to be a sl ight t ime lag between loss o f posture and changes i n E E G s and S E P s (Coenen et a l . , 2 0 0 0 ; Ra j et a l . , 1992). Hewett et a l . (1993) found that dur ing g radua l - f i l l CO2 exposure at a f l o w rate o f 2 0 % per minute , rats showed i m m o b i l i t y after 72 s and loss o f r ight ing ref lex after 109 s. H o w e v e r , i m m o b i l i t y was on ly def ined as inact iv i ty so the delay between complete loss o f posture and loss o f r ight ing ref lex w o u l d have been shorter. In fact, C o e n e n et a l . (1995) found that complete loss o f posture i n rats was w e l l correlated w i t h onset o f a f u l l y aberrant E E G , and that the delay was less than 10 s dur ing g radua l - f i l l CO2 exposure. F o r the sake o f s i m p l i c i t y , ataxia and loss o f posture w i l l be used to summar ize the duration data for each o f the different CO2 methods, w i t h the assumpt ion that these measures represent the onset o f loss o f consciousness and that complete insensib i l i ty l i k e l y f o l l o w s short ly afterwards. . 1 The term 'conscious' is used here according to what Block (1995) has referred to as "phenomenal consciousness". This refers to awareness and the ability to experience events, but does not imply being able to report on this experience or of being self-conscious. 'Conscious' is used throughout this dissertation to indicate when the animal is awake and therefore is presumably capable of experiencing negative affective states. 5 D u r i n g p r e - f i l l euthanasia i n rats w i t h CO2 concentrations greater than 7 0 % , t ime to loss o f posture general ly ranges from 7 s to 20 s ( B l a c k s h a w et a l . , 1988; K o h l e r et a l . , 1999; Hewett et a l . , 1993; D a n n e m a n et a l . , 1997; C o e n e n et a l . , 1995; Br i t t , 1987), and atax ia is not general ly observed because o f the speed o f col lapse. W i t h a fast f l o w rate (>50%» o f the chamber v o l u m e be ing added per minute) , ataxia occurs at 13 s to 18 s and loss o f posture at 26 s to 48 s (Coenen et a l . , 1995; S m i t h and Harrap , 1997). W i t h a m e d i u m f l o w rate ( 1 5 % to 5 0 % o f the chamber v o l u m e added per minute) ataxia ranges f r o m 42 s to 120 s, and loss o f posture ranges f r o m 90 s to 120 s (Hornett and Haynes , 1984; Hackbar th et a l . , 1999; S m i t h and Harrap , 1997; D a n n e m a n et a l . , 1997). S l o w f i l l rates (<15% o f the chamber v o l u m e added per minute) take m u c h longer, w i t h ataxia occur r ing at 120 s to 180 s and loss o f posture occur r ing at 120 s to 240 s (Hornett and Haynes , 1984). L i t t l e is k n o w n about the concentrations o f CO2 that are required to cause loss o f consciousness dur ing g radua l - f i l l CO2 euthanasia. W i t h a m e d i u m CO2 f l o w rate, rats have been found to lose consciousness at concentrations as l o w as 4 0 % (Smi th and Har rap , 1997). The major i ty o f studies e x a m i n i n g g radua l - f i l l CO2 exposure have not moni tored CO2 concentrat ions, so it is not clear whether f l o w rate has an effect o n the CO2 concentrat ion needed to cause loss o f consciousness i n rats. H o w e v e r , results i n m i c e suggest that a m e d i u m flow rate o f 3 0 % per minute is associated w i t h a lower concentrat ion o f CO2 at loss o f consciousness than a fast flow rate o f 6 0 % per minute (Ambrose et a l . , 2000) . W h i l e the cause o f this effect is u n k n o w n , the longer durat ion o f CO2 exposure w i t h s l o w f i l l rates m a y a l l o w more t ime for p H adjustments to occur i n the b l o o d and C S F . In summary , p r e - f i l l CO2 exposure takes less t ime to cause loss o f consciousness than g radua l - f i l l exposure, but an imals are exposed to higher C 0 2 concentrations dur ing the per iod o f consciousness. W i t h g radua l - f i l l CO2 exposure, faster flow rates cause a shorter durat ion to loss 6 o f consciousness, but m a y be associated w i t h higher CO2 concentrations at the t ime o f loss o f consciousness. 1.2.3 Contexts of CO2 use C a r b o n d iox ide is recommended for euthanasia o f laboratory rodents i n m a n y western countr ies such as C a n a d a ( C C A C , 1993) the U S A ( A V M A , 2000) , the U K ( U K H o m e O f f i c e , 1997), the countries o f the European U n i o n (C lose et a l . , 1997) and A u s t r a l i a and N e w Zea land ( A N Z C C A R T , 1993). The recommended method o f del ivery var ies, but i n general , gradual induct ion is recommended over p r e - f i l l . A l t h o u g h some recommendat ions stipulate a f l o w rate for gradual induct ion , anecdotal reports suggest that few fac i l i t ies moni to r f l o w rate. C a r b o n d iox ide is also used for euthanasia o f in jured w i l d l i f e i n rehabi l i tat ion centres and for stunning or k i l l i n g o f farmed m i n k , p igs , poul t ry , and f i sh (v ia water bath). Recommendat ions for use o f CO2 w i t h these species d i f fer f r o m those appl icable to laboratory animals i n that they often require p r e - f i l l i n g or rap id ly f i l l i n g the chamber rather than gradual f i l l . F o r example , the U K Wel fa re o f A n i m a l s (Slaughter or K i l l i n g ) Regu la t ion (1995) stipulates that for the stunning o f p igs , the chamber must reach a concentrat ion o f at least 7 0 % i n less than 30 seconds, and for the k i l l i n g o f m i n k , the an imal must be p laced into a chamber conta in ing 1 0 0 % C 0 2 ( U K H o m e O f f i c e , 1995). In contrast, the U K C o d e o f Pract ice for k i l l i n g o f exper imental an imals recommends exposure to a gradual increase i n CO2 ( U K H o m e O f f i c e , 1997). H o w e v e r , there has been no research to just i f y the use o f different de l ivery methods for different species. 1.2.4 Advantages of CO2 euthanasia There are a number o f reasons for the widespread use o f CO2 euthanasia o f laboratory an imals . A n e c d o t a l reports suggest that there is a general percept ion w i t h i n the scient i f ic 7 c o m m u n i t y that the effects o f CO2 on exper imental results are k n o w n , and that u n k n o w n effects o n exper imental results m a y be introduced by changing to other procedures. Init ial experiments o n the use o f CO2 for k i l l i n g occurred i n the late 1800's ( rev iewed by H i l l and F l a c k , 1908), and there is therefore a long history o f CO2 use. There have also been a number o f studies conducted to determine the effects o f CO2 o n spec i f ic metabol ites and tissues, i n order to ensure that exper imental results are not affected (e.g. F a w e l l et a l . , 1972; Pecaut et a l . , 2 0 0 0 ; Bergersweeney et a l . , 1994). In situations where a confound ing effect on tissue measures is expected, such as dur ing tox i co log ica l research, other euthanasia methods l ike decapitat ion and cerv ica l d is locat ion are often employed . CO2 also has a number o f advantages for the operator i n compar ison to other methods. Because it can be del ivered i n a chamber, it invo lves l itt le hand l ing o f an imals and can be used to q u i c k l y euthanize large numbers o f animals at a t ime. In compar ison to phys ica l methods and injectable anaesthetics, it requires considerably less t ime. Th i s method also requires l ittle direct interact ion w i t h animals and results in f e w effects that m a y be d isturb ing to the operator. A s a respiratory gas, CO2 is also safer for the operator than other gas euthanasia agents such as carbon m o n o x i d e , ch lo ro fo rm, ether and gas anaesthetics, w h i c h can pose s ignif icant health r isks . CO2 also has cost advantages i n compar ison to other methods. Decapi tat ion and cerv ica l d is locat ion require m i n i m a l equipment, but have considerable costs i n terms o f employee t ime. Other phys ica l methods such as m i c r o w a v e i r radiat ion require the purchase o f expensive equipment. Anaesthet ics tend to be m u c h more expensive than CO2. There are also indirect costs for anaesthetic methods, such as the purchase o f needles (for injectable anaesthetics), and anaesthetic machines and scavenging systems (for gaseous anaesthetics). W i t h respect to the welfare o f an imals CO2 prov ides m a n y benefits i n compar ison to other euthanasia methods. M o s t important ly , CO2 is easy to de l i ver proper ly w i t h l i tt le t raining. 8 P h y s i c a l methods, injectable anaesthetics and inert gases have the potential to be administered incorrect ly and to result i n considerable p a i n and distress for the an ima l . Gases i n general are also advantageous because they require m i n i m a l hand l ing o f the an ima l . Laboratory rodents are not general ly habituated to regular hand l ing and restraint, and these procedures can therefore cause considerable distress (e.g. Sharp et a l . , 2 0 0 2 , 2003) . Furthermore, inject ions into the peri toneal cav i ty can be pa in fu l , and injectable anaesthetics can be i r r i tat ing, poss ib ly s t imulat ing v iscera l nociceptors. A l t h o u g h CO2 has many advantages over other euthanasia methods, there has been ongo ing debate as to whether it causes pa in and distress i n laboratory rodents. 1.3 Animal Distress and Carbon Dioxide 1.3.1 What is animal distress? A n i m a l wel fare is general ly considered to be enhanced by good health, posi t ive affect, and the abi l i ty to per form natural behaviours , and reduced by poor health, negative affect and inadequate outlets for behaviour (Fraser et a l . , 1997). W i t h a procedure such as CO2 euthanasia, the m a i n concern is w i t h the affect ive state o f the an ima l . In descr ib ing everyday usage, The Canad ian O x f o r d D ic t ionary (2000) defines distress as "severe trouble, anxiety , so r row" (p.275), ind icat ing a negative emot ional state. Selye (1975) was the first to def ine distress sc ient i f ica l ly , and he used this term to refer to b io log ica l responses to negative stressors, where stressor refers to an actual or perceived threat to homeostasis . W i t h speci f ic reference to an imal wel fare , the term distress has been def ined i n a number o f different ways ; however , i n general there is agreement that 'distress' refers to a negative state that develops w h e n an organism is unable to adapt to a stressor (e.g. K i t c h e n et a l . , 1987; M o b e r g , 2 0 0 0 ; N R C , 2 0 0 3 ; R o w a n et a l . , 1998). Th i s def in i t ion , therefore, acts as an 9 umbre l la term that encompasses negative affect associated w i t h more speci f ic negative states such as p a i n , d iscomfor t and fear. 1.3.2 Potential mechanisms for CO2 distress There are three m a i n mechanisms by w h i c h CO2 might cause distress i n rats. The rationale for the first two mechanisms that are discussed is largely dependent o n human experiences dur ing CO2 exposure. I f humans report pa in or distress dur ing CO2 exposure, then f o l l o w i n g the log ic o f D a w k i n s (1980) it can be hypothesized that CO2 has the potential for s imi la r sensations i n animals w i t h s imi la r anatomy and phys io logy . A l t h o u g h humans have higher order cogni t ive process ing that can affect the qual i ty o f a sensation, the general percept ion o f a st imulus as appetit ive or aversive is l i k e l y conserved between humans and other m a m m a l s . H o w e v e r , it is important to recognize that sensations m a y also di f fer between different species, s ince each is adapted to a part icular n iche. 1.3.2.1 Pain induced by carbonic acid formation The most w i d e l y discussed m e c h a n i s m for distress dur ing CO2 euthanasia is that CO2 can f o r m carbonic ac id o n nasal mucous membranes, and this st imulates t r igeminal nociceptors and causes pa in (e.g. L e a c h et a l . , 2002a) . Th i s can be extended by cons ider ing that CO2 also has the potential to affect the cornea and conjunct iva o f the eyes, and poss ib ly chemoreceptors i n the lower respiratory tract that are sensit ive to other irritants, a l though this has not been fu l l y examined . A number o f studies have assessed the potential for CO2 to activate nociceptors and evoke p a i n i n humans w h e n appl ied to the nasal mucosa , cornea and conjunct iva . Negat i ve nasal m u c o s a l potentials have been recorded f r o m the nasal septum and used as a non - invas ive method for measurement o f t r igeminal nociceptor act ivat ion. Negat i ve m u c o s a l potentials "are 10 thought to be the result o f summat ing receptor potentials o f chemosensi t ive nociceptors o f the t r igemina l nerve" (Thurauf et a l . , 1993, p.293). These potentials have been recorded i n response to CO2 concentrations over 4 5 % , and have been found to increase w i t h CO2 concentrat ion and appl icat ion durat ion (Thurauf, 1993). H u m a n self - report data indicate that CO2 is detectable at the nasal m u c o s a at concentrations o f on ly 2 0 % , and that it becomes overt ly pa in fu l at concentrations above approx imate ly 5 0 % ( A n t o n et a l . , 1992; Thurauf et a l . , 2002) . H o w e v e r , the concentrat ion that is perceived as pa infu l does vary , w i t h the pa in threshold ranging f r o m 3 2 . 5 % to 5 5 % depending on the ind iv idua l ( A n t o n et a l . , 1992). D a n n e m a n et a l . (1997) had subjects rate the noxiousness o f CO2 at concentrations ranging f r o m 5 0 % to 1 0 0 % w h e n inhaled v i a the nose, and found that the major i ty o f subjects rated 50%) CO2 as unpleasant or uncomfor tab le , and 1 0 0 % CO2 as pa in fu l . Furthermore, at each concentrat ion some subjects felt they cou ld not take a complete breath, and the words t ing l ing , p r i c k l i n g and burn ing were used to describe the sensation at a l l concentrations. In rats, CO2 has been shown to activate dorsal horn neurons that receive input f r o m t r igeminal nociceptors i n the nasal m u c o s a ( A n t o n et a l . , 1 9 9 1 ; Peppe l and A n t o n , 1993), suggesting that CO2 also has the potential to cause pa in i n rats. Peppe l and A n t o n (1993) found that rat nociceptors respond to CO2 concentrations that are s imi la r to those prev ious ly reported for h u m a n nociceptors . CO2 concentrations b e l o w 3 7 % were subthreshold for the major i ty o f rat neurons, and the C 0 2 noc icept ion threshold ranged f r o m 3 7 % to 5 0 % CO2 for the major i ty o f neurons. Furthermore, the graded response tended to increase i n a l inear fash ion unt i l saturation was reached at approx imately 8 7 % CO2. Another method o f assessing the potential for a substance to e l ic i t p a i n at the upper respiratory tract is through its abi l i ty to e l ic i t changes i n cardiorespiratory rhythms. Inhaled irritants are k n o w n to induce a ref lex apnea and heart rate reduct ion, and these responses are thought to reduce transfer o f ha rmfu l substances into the body ( W i d d i c o m b e , 1986). In rats, 11 1 0 0 % CO2 e l ic i ts this apnea and bradycardia , but CO2 concentrations o f 10, 25 and 5 0 % do not ( Y a v a r i et a l . , 1996). The threshold necessary for e l i c i t ing cardiorespiratory alterations was not determined, but this data suggests that C 0 2 concentrations less than 50% are not i r r i tat ing to the nasal m u c o s a i n rats. The effect o f CO2 o n pa in and act ivat ion o f nociceptors i n the cornea and conjunct iva appears to be s imi la r to its effect at the nasal mucosa . F e n g and S i m p s o n (2003) examined pa in thresholds i n humans i n response to CO2 appl icat ion to the cornea and conjunct iva and found that they occurred at 31 % and 5 4 % , respect ively . Part ic ipants again character ized the sensation at threshold as burn ing or st inging. In another study e x a m i n i n g the effect o f CO2 on the cornea, subjects reported m i l d st inging at a mean CO2 concentrat ion o f 33.5%, and overt pa in at 47.5% (Chen et a l . , 1995). C h e n et a l . also examined the responses o f cat corneal nociceptors to CO2 appl icat ion and found that they responded at a mean threshold o f 4 0 % C 0 2 , w h i c h is s imi la r to the pa in responses observed i n humans. A c t i v a t i o n o f rat corneal nociceptors by CO2 has also been demonstrated (Hi rata et a l . , 1999), a l though the threshold leve l o f C 0 2 needed to e l ic i t nociceptor act ivat ion was not ident i f ied . Th i s evidence suggests that CO2 has the potential to cause pa in i n rats through st imulat ion o f nociceptors i n the nasal m u c o s a and cornea. P a i n is l i k e l y to occur w i t h p r e - f i l l methods i n w h i c h a conscious an imal is exposed to h igh concentrations o f CO2 (>70%). D u r i n g gradual f i l l CO2 euthanasia animals reach unconsciousness at about 40% (Smi th and Harrap , 1997), so there is some potential for l o w levels o f pa in to occur around the t ime o f loss o f consciousness. 1.3.2.2 Non-pain discomfort associated with hypercapnia and hypoxia Distress dur ing CO2 euthanasia cou ld also occur through the effects o f hypercapnia (elevated b l o o d CO2) and h y p o x i a (reduced b l o o d O2), w h i c h have been s h o w n to cause dyspnea 12 i n humans. D y s p n e a is an unpleasant sensation o f breathlessness w h i c h is sometimes accompanied by other negative sensations such as headache, f lush , restlessness, heart pound ing , drowsiness , and d izz iness (e.g. M o o s a v i et a l . , 2003) . W i t h m i l d increases i n inspi red CO2 and decreases i n inspi red O2, increased vent i lat ion results i n a reduct ion or e l iminat ion o f dyspnea (Banzett et a l . , 1996; F o w l e r , 1957; Shea et a l . , 1996), but there are l imi ts to this compensatory m e c h a n i s m such that dyspnea st i l l occurs dur ing spontaneous breathing w i t h moderate hypercapnia and h y p o x i a (Shea et a l . , 1996). N u m e r o u s studies have demonstrated that hypercapnia results i n dyspnea i n humans dur ing spontaneous breathing (e.g. Banzett et a l . , 1996; Shea et a l . , 1996). M o s t studies e x a m i n i n g dyspnea due to hypercapnia have moni tored end- t ida l part ia l pressures (PET) o f CO2 (concentration o f CO2 breathed out) rather than inspi red levels o f CO2, and it is therefore d i f f i cu l t to determine the inspi red concentrations o f CO2 that result i n different levels o f dyspnea. H o w e v e r , some studies have examined inspi red CO2 levels and can prov ide insight into the CO2 concentrations necessary to e l ic i t dyspnea. L i o t t i et a l . (2001) had subjects rate their sensations o f dyspnea o n a 100 point scale dur ing exposure to 8 % CO2 i n O2, and dyspnea levels were rated as either 73 or 55 depending o n whether the gas mix ture was del ivered us ing a facemask or a mouthpiece, respect ively . D r i p p s and C o m r o e (1947) found that approx imately 3 0 % o f study part ic ipants reported dyspnea w h e n exposed to either 7.6 or 1 0 . 4 % CO2 i n O2. In the late 1800's and early 1900's researchers often used themselves as exper imental subjects i n the invest igat ion o f respiratory phys io logy (summar ized by H i l l and F l a c k , 1908). These studies prov ide insight into the concentrations o f CO2 that result i n moderate to severe dyspnea dur ing spontaneous breathing. A c c o r d i n g to H i l l and F l a c k (1908) , G r e e n w o o d found that he was able to breathe a mixture o f 15.3% CO2 and 1 4 . 5 % O2 w i t h marked dyspnea, but that h igher levels o f CO2 resulted i n c losure o f the glottis and prevented inhalat ion. Ha ldane and S m i t h (1892) found that w h e n breathing 1 8 . 6 % CO2 i n air, they deve loped profound dyspnea i n 13 one to two minutes that was accompanied by throbbing i n the head and mental dul lness. Together w i t h the results f r o m modern studies, these results indicate that dur ing spontaneous breathing, dyspnea and other negative sensations can occur w i t h C O 2 concentrations as l o w as 8%, and that severe dyspnea occurs at C O 2 concentrations greater than approx imately 15%. H y p o x i a also appears to have the potential to cause dyspnea i n humans, but to a lesser extent than hypercapnia . M o o s a v i et a l . (2003) found that dur ing constrained breathing, part icipants reported onset o f dyspnea w h e n PETO2 decreased b e l o w approx imate ly 60 Torr (PETO2 levels were 108 Tor r dur ing basel ine w h e n breathing air). H o w e v e r , at the study's ethical PETO2 l i m i t o f 40 Torr (the lowest inspi red O2 tested was 7%) the major i ty o f part icipants rated their dyspnea as less than 4 0 o n a v isua l analogue scale w h i c h ranged f r o m 0 (no dyspnea) to 100 (extreme dyspnea). Th i s indicates that O2 concentrations o f less than 7 % are needed to evoke moderate sensations o f dyspnea i n humans dur ing constrained breathing. The potential for dyspnea due to h y p o x i a alone i n spontaneously breathing humans has not been fu l l y investigated. H y p o x i a - i n d u c e d loss o f consciousness occurs occas iona l l y i n pi lots as a result o f cab in pressur izat ion fai lure (Cab le , 2003) , suggesting that spontaneously breathing humans do not experience dyspnea pr ior to loss o f consciousness w i t h hypox ia . H o w e v e r , h y p o x i a has a synergist ic effect o n venti latory responses to hypercapnia (e.g. N i e l s o n and S m i t h , 1952), and augments sensations o f dyspnea due to hypercapnia i n humans (Banzett et a l . , 1996; M a s u d a et a l , 2001) . These results indicate that h y p o x i a alone is u n l i k e l y to cause sensations o f dyspnea dur ing free breathing, but that it m a y increase dyspnea due to hypercapnia . Thus , i f dyspnea occurs dur ing C 0 2 euthanasia, h y p o x i a m a y be a contr ibut ing factor. T o date, no models have been developed for the assessment o f dyspnea i n conscious an imals , so it is not k n o w n whether hypercapnia and h y p o x i a cause dyspnea i n rats. D y s p n e a i n rats m a y be indicated by increases i n breathing depth and frequency. The term dyspnea is somet imes used i n the veterinary literature to refer d i rect ly to these breathing changes (e.g. 14 Hornett and Haynes , 1984), but 'dyspnea ' actual ly refers to a sensation o f breathlessness rather than the phys i ca l changes i n breathing that sometimes accompany this sensation. W h i l e increases i n breathing (sometimes referred to as gasping or laboured breathing) have been observed i n rats dur ing CO2 euthanasia, it is not clear whether they occur before or after loss o f consciousness, and whether they are more severe w i t h p r e - f i l l or g radua l - f i l l exposure (Coenen et a l . , 1995; Hornett and Haynes , 1984; Iwarsson & Rehbinder , 1993; S m i t h and Harrap , 1997). Furthermore, human m e d i c a l studies suggest that breathing alterations and sensations o f dyspnea do not a lways occur concurrent ly ( L u s h et a l . , 1988), w h i c h br ings into quest ion the usefulness o f this measure for assessing dyspnea i n rats. W h i l e there is no evidence c o n f i r m i n g that dyspnea occurs i n rats, it is reasonable to assume that such a basic response to changes i n air compos i t i on and b l o o d gas levels w o u l d be conserved between species. I f so, both p r e - f i l l (<6% O2, > 7 0 % CO2) and g radua l - f i l l CO2 euthanasia (<12% 0 2 , > 4 0 % CO2) have the potential to cause strong sensations o f dyspnea in rats, m a i n l y due to hypercapnia . 1.3.2.3 Fear due to novelty N o v e l t y has been suggested to induce an approach-avoidance conf l ic t i n rats, result ing f r o m an interact ion between exploratory mot ivat ion and fear ( M o n t g o m e r y , 1955). CO2 cou ld therefore cause distress i n rats by act ing as a nove l st imulus that el ic i ts fear. Odour percept ion occurs as a result o f both o l factory and t r igeminal s t imulat ion (e.g. C a i n and M u r p h y , 1980). W h i l e CO2 is thought to st imulate m a i n l y t r igeminal neurons, humans perceive an odour qual i ty w h e n asked to describe sensations occur r ing w i t h CO2 inhalat ion ( C a i n and M u r p h y , 1980). Rats can detect CO2 at concentrations between 0.04 and 1 . 7 % (Youngentob , 1991), w h i c h is far b e l o w the leve l required to stimulate t r igeminal neurons (Peppel and A n t o n , 1993). The exact qual i ty o f C 0 2 that rats are perce iv ing at these l o w levels is u n k n o w n , but a large port ion o f the 15 rat brain is devoted to odour detection so rats m a y be more sensit ive to the odour qual i ty associated w i t h CO2 than humans. W a l l a c e and R o s e n (2000) demonstrated that exposure o f rats to nove l odours, such as butyr ic ac id (s imi lar to ranc id butter) and i s o a m y l acetate (s imi lar to banana), causes avoidance, reduces g rooming t ime and increases f reez ing t ime, suggesting that n o v e l odours can el ic i t fear i n rats. Thus , CO2 m a y cause distress i n rats because it is act ing as a nove l st imulus that el ic i ts fear. 1.4 Assessment of Distress During C 0 2 Euthanasia In the assessment o f an imals ' subjective states it is not possib le to obtain verbal reports o f their exper ience, and it is therefore necessary to use other measures. There are two types o f measures that can be used to assess whether a procedure such as CO2 euthanasia causes pa in and distress i n an imals : 1) phys io log ica l measures o f stress, and 2) behav ioura l measures o f distress and avers ion. 1.4.1 Physiological assessment of distress The term stress general ly refers to a b io log ica l response to an actual or perceived threat to homeostasis (e.g. M o b e r g , 2000) , and invo lves both behavioural and phys io log ica l responses that act to mainta in homeostasis. P h y s i o l o g i c a l responses, such as act ivat ion o f the sympathet ic - adrenerg ic -medul lary axis ( S A M ) and the hypothalamic -p i tu i tary -adrenal axis ( H P A ) , occur i n response to both phys io log ica l and psycho log ica l stressors and serve to redirect body systems towards cop ing w i t h the stressor. A c t i v a t i o n o f the S A M axis occurs almost immediate ly i n response to a stressor. Th i s system activates the body by releasing epinephrine and norepinephr ine, increasing glucose metabo l i sm, respirat ion, heart rate and b l o o d pressure, and reduc ing funct ions that are not o f immediate necessity ( summar ized by Toates, 1995). 16 A c t i v a t i o n o f the H P A axis results i n release o f cort icotrophin - re leas ing hormone ( C R H ) and other secretagogues f r o m the hypothalamus, that lead to the release o f adrenocort icotropic hormone ( A C T H ) f r o m the pituitary and then Cortisol or cort icosterone ( in the rat) f r o m the adrenal cortex. Th i s system causes a number o f effects i n c l u d i n g m o b i l i z a t i o n o f energy, suppression o f the i m m u n e system, and release o f endorphins to modulate pa in responses (summar ized by Toates, 1995). The presence or absence o f act ivat ion o f the S A M and H P A axes can prov ide in format ion as to whether a st imulus acts as a stressor, but as w i l l be discussed later, their usefulness for assessing distress is l im i ted . Consistent act ivat ion o f the S A M axis i n response to reduced O2 and elevated CO2 has not been observed i n rats. B o r o v s k y et a l . (1998) found that a 30 s exposure to 1 0 0 % CO2 or 1 0 0 % ni t rogen was suff ic ient to increase p l a s m a norepinephrine levels i n rats, and F u k u d a et a l . (1989) found that 1 0 0 % CO2 increases cardiac sympathetic act iv i ty . H o w e v e r , other studies us ing lower CO2 concentrations and higher O2 concentrations have fa i led to observe act ivat ion o f the S A M axis . Hodges et a l . (2002) found that w h e n rats were exposed to either 1 2 % O2 in air or 7 % CO2 i n O2, vent i lat ion increased, but heart rate and b l o o d pressure d i d not. S i m i l a r l y , R a f f and Roar ty (1988) found that b l o o d pressure was not affected by l o w O2 (10%) O2), h igh CO2 ( 4 % or 8 % C 0 2 ) or a combinat ion o f l o w 0 2 and h igh C 0 2 ( 7 % 0 2 and either 4 % or 8 % C 0 2 ) . Three studies have examined S A M variables dur ing CO2 euthanasia i n rats. W i t h p r e - f i l l exposure, a l l studies found that heart rate (Coenen et a l . , 1995; Sharp et a l . , 2 0 0 6 ; S m i t h and Har rap , 1997) and b l o o d pressure (Sharp, 2 0 0 6 ; S m i t h and Harrap , 1997) decreased immediate ly . W i t h g radua l - f i l l exposure, Coenen et a l . (1995) found that heart rate decreased, but S m i t h and Harrap (1997) found that heart rate and b l o o d pressure in i t ia l l y increased before dec l in ing . The immediate decl ine in heart rate w i t h p r e - f i l l euthanasia can probably be expla ined by rap id depression o f the C N S . A l s o , irritant s t i m u l i , i nc lud ing CO2 at h igh concentrations, are k n o w n to cause bradycard ia v i a s t imulat ion o f t r igeminal nociceptors ( Y a v a r i et a l . , 1996). 17 Unfortunately the results for gradual fill are difficult to interpret because neither study used handled controls or acclimatized the animals to the chamber. It is therefore unclear how these variables relate to resting values, and whether activation of the S A M axis occurred as a result of handling. Increased HPA axis activity during moderate hypoxia and hypercapnia has also been examined in rats. Marotta et al. (1976) found that prolonged exposure to low O2 (10% O2 in nitrogen) or high CO2 (10% CO2, 20% O2, in nitrogen) resulted in corticosterone levels almost eight times those observed in control animals. Similarly, Raff and Roarty (1988) measured A C T H after exposure to low 0 2 (10% 0 2 ) , high C 0 2 (4% C 0 2 or 8% C 0 2 ) or a combination of low 0 2 and high C 0 2 (7% 0 2 and either 4% or 8% C0 2 ) , and found that it was elevated with 8% CO2 and with the combination mixtures. Although one study has specifically examined A C T H and corticosterone levels during gradual-fill CO2 euthanasia and found no increase (Hackbarth et al., 1999), the results were likely affected by problems with experimental design. Not only were the sample sizes extremely small for analysis of these variables (N = 4), but decapitation was performed 30 s, 75 s, and 120 s into the procedure. This time frame does not allow for increases in these hormones to occur in the blood stream (Terlouw et al., 1997), thus no effect could be expected. The two previous studies indicate that gradual-fill CO2 euthanasia could result in activation of the HPA axis. There are, however, three major limitations with the use of physiological stress measures in the assessment of distress associated with CO2 euthanasia. Firstly, S A M and HPA activation do not imply that the stressor or the response is negative or positive for the animal. In fact, physiological stress responses are associated with neutral and positive events such as exercise and mating. Thus a stress response does not necessarily mean that an animal's welfare is compromised (Dawkins, 1998). However, by examining whether the animal finds the stressor 18 appetit ive or aversive, it is often possible to determine whether the stressor is perceived i n a negative or posi t ive manner. Secondly , S A M and H P A act ivat ion occur i n response to both phys io log ica l and p s y c h o l o g i c a l stressors, such that the actual effects o f the st imulus o n the subjective state o f the a n i m a l are d i f f i cu l t to determine. In the case o f the H P A ax is , it appears that there are at least two types o f C N S pathways that can e l ic i t release o f C R H . H e r m a n and C u l l i n a n (1997) suggest that there are direct pathways for immediate phys io log ica l threats, and indirect pathways v i a higher bra in c i rcuits for s t imul i that require interpretation w i t h respect to prev ious experience. H P A responses to what the authors label "p rocess i ve" stressors l i k e restraint and novel ty appear to be associated w i t h relay through higher bra in areas such as the forebrain and l i m b i c system nuc le i , whereas responses to " s y s t e m i c " stressors l ike h y p o x i a and ether exposure are more l i k e l y d i rect ly re layed f r o m brainstem nuc le i ( rev iewed by H e r m a n and C u l l i n a n , 1997; H e r m a n et a l . , 1996). Th i s suggests that H P A responses to CO2 euthanasia c o u l d occur wi thout percept ion by higher bra in centres. I f so, they w o u l d be a poor measure o f psycho log ica l distress associated w i t h this procedure. In fact, it has been s h o w n that hypotha lamic vasopress in - conta in ing neurons, w h i c h secrete the C R H secretogogue vasopress in , produce s imi la r increases i n c - F o s ( indicat ing act ivat ion o f the neuron) i n response to CO2 exposure i n both awake and anesthetized rats ( K c et a l . , 2002) . A l t h o u g h these measures can prov ide a gauge o f h o w the st imulus affects the an imal phys io log ica l l y , and thus the potential for caus ing distress, they do not necessar i ly a l l o w one to assess the psycho log ica l state o f the an ima l . F i n a l l y , because CO2 euthanasia occurs re lat ively q u i c k l y , it is not poss ib le to measure H P A responses to the port ion o f the procedure that occurs w h i l e the an imal is st i l l conscious. In p r e - f i l l and gradual CO2 euthanasia, rats become unconsc ious i n approx imate ly 15 s and 90 s respect ively , and in terms o f distress we are concerned on ly w i t h what the an imal experiences dur ing this t ime. Changes i n the S A M occur w i t h i n seconds and can be measured dur ing this 19 procedure, but evidence o f act ivat ion o f the H P A axis does not appear this q u i c k l y and the state o f the an imal at the t ime o f loss o f consciousness cannot be preserved. B y observ ing this system after p ro longed exposure or after the an imal regains consciousness, w e are no longer observ ing what occurred dur ing that l im i ted per iod o f t ime. F o r example , noc icept ive CO2 concentrations might occur after loss o f consciousness, and this might have an effect o n H P A variables that are not associated w i t h the a n i m a l ' s consc ious experience. 1.4.2 Behavioural assessment of distress U n l i k e phys io log ica l measures, behavioural measures can be easi ly moni tored dur ing the course o f CO2 exposure, and the types o f behaviours per formed can prov ide us w i t h in format ion about h o w the an imal perceives the st imulus. D u r i n g the euthanasia process w e can observe whether the a n i m a l exhibi ts behaviours that are ind icat ive o f distress such as escape behaviours , distress voca l i zat ions and behaviours associated w i t h pa in or d iscomfort . In tests o f avers ion w e can observe whether the an imal avoids CO2 exposure, and whether it w i l l pay a cost to avo id exposure. W e can also use other behav ioura l tests to determine whether CO2 exposure el ic i ts fear i n an imals . 1.4.2.1 Behaviour during CO2 exposure D u r i n g exposure to an aversive or pa in fu l gas, an imals might s h o w both general and spec ies -spec i f ic signs o f distress. Genera l signs might inc lude behaviours associated w i t h escape, such as increased act iv i ty , explorat ion, and concentrated act iv i ty at k n o w n exits. In the rat this might be indicated by increased locomotor act iv i ty and rear ing, as w e l l as w a l l - c l i m b i n g , invest igat ion o f chamber surfaces and push ing , scratching and b i t ing at potential exits. A l t h o u g h an increase i n act iv i ty indicates that the an imal is responding to the st imulus , an increase i n actual escape behaviours , such as push ing , scratching and b i t ing at potential exi ts , is important 20 to show that the st imulus is perce ived as negative and is not s i m p l y e l i c i t ing explorat ion. Rats have also been shown to respond to aversive s t imul i by freezing (e.g. Barnett , 1975; D ie lenberg and M c G r e g o r , 2001) , w h i c h differs f r o m s imple inact iv i ty i n terms o f increased musc le tone and a lack o f head movements . A l t h o u g h this seems contradictory to the increase i n act iv i ty descr ibed above, it is w i d e l y recognized that an imals , i n c l u d i n g rats, can respond proact ive ly or react ively w h e n confronted w i t h a stressor ( rev iewed by K o o l h a s et a l . , 1999). Therefore distress cou ld be indicated by either an increase i n act iv i ty and escape behaviour or by f reez ing behaviour . In assessment o f either behavioural response, it is important to compare it to stable basel ine or contro l data to ensure that the response is i n fact due to st imulus exposure, and not to some other factor such as handl ing or exposure to a nove l envi ronment . Distress can also be indicated by the product ion o f voca l i zat ions . V o c a l i z a t i o n s produced dur ing distress might serve a number o f purposes, such as to w a r n conspec i f ics o f danger, to e l ic i t a ca re -g i v ing response or to assist i n regain ing contact w i t h conspeci f ics ( rev iewed by K l u m p and Shalter, 1984, reproduced i n Hauser , 1996). Rats have been shown to produce ul t rasonic vocal i zat ions ( U S V s ) i n a var iety o f contexts, and these ca l ls m a y be an indicator o f affect ive state (Knutson et a l . , 2002) . U S V s i n the 50 k H z and 20 k H z ranges tend to occur in response to appetit ive and aversive s t imul i respectively . F o r example , U S V s i n the 2 0 - 3 0 k H z range have been observed i n response to pa in fu l s t imulat ion ( D i n h et a l . , 1999; Colpaert et a l . , 1987; Jourdan et a l . , 1995, 1998), exposure to predators (B lanchard et a l . , 1991), drug w i thdrawal ( V i v i a n and M i c z e k , 1991), f ight ing w i t h conspec i f ics (Thomas et a l . , 1983), acoust ic startle (Kal twasser , 1990) active avoidance learning ( C u o m o et a l . , 1992) and touching by an unfami l ia r human (B rud zynsk i and O c i e p a , 1992). W h i l e these ca l ls appear to be re lat ively non -spec i f i c and their s igna l l ing context is not fu l l y understood, they do have an associat ion w i t h situations that appear to be negative. 21 The occurrence o f behaviours associated w i t h pa in or w i t h act ivat ion o f a stress response m a y also indicate that CO2 euthanasia causes distress. Pa in - re lated behaviours might inc lude shaking o f the head or p a w i n g at the nose and eyes. U r i n a t i o n and defecat ion are also sometimes used as measures o f distress because they are associated w i t h act ivat ion o f the autonomic nervous system dur ing the stress response ( rev iewed by V ingerhoets et a l . , 1985). A l t h o u g h ur inat ion and defecat ion can be indicators o f distress, their absence does not necessari ly indicate a lack o f distress because they can be affected by recent feeding and e l iminat ion behaviour . A number o f studies have observed the behaviour o f rats dur ing CO2 euthanasia, but often conc lus ions are d r a w n on the basis o f o p i n i o n rather than object ive data. Responses to CO2 have been character ized by some as "apprehens ive" (Hornett and Haynes , 1984), or " a b n o r m a l " , " e x c i t e d " and "ag i ta ted" (Coenen et a l . , 1995), wi thout precise def in i t ions or descript ions o f behav ioura l measures. Hackbar th et a l . (1999) suggest that they saw no signs o f " fea r " i n rats dur ing CO2 euthanasia. H o w e v e r , no data are p rov ided to support this assert ion, even though they describe a number o f behav ioura l var iables that were measured. Other studies have used object ive measures o f behav iour , but the usefulness o f these measures is compl i ca ted by smal l sample sizes or a lack o f appropriate distress cr i ter ia, controls or acc l imat i zat ion pr ior to gas exposure. B l a c k s h a w et a l . (1988) compared rats' responses dur ing exposure to either air or p r e - f i l l CO2, and measured movement , t ime spent stationary and w a l l touches and c l i m b s dur ing the first 10 s o f exposure. D u r i n g CO2 exposure, rats exhib i ted less overa l l act iv i ty , but more w a l l c l i m b i n g . The decrease i n act iv i ty is d i f f i cu l t to interpret because it m a y be due to the an imal f reez ing or the anaesthetic effects o f CO2. However , the increased inc idence o f w a l l c l i m b i n g does suggest increased explorat ion . W a l l c l i m b i n g was even greater dur ing exposure to ether and c h l o r o f o r m , w h i c h are k n o w n irr itants, suggesting that this behaviour was indicat ive o f some leve l o f distress, but also that the response was not 22 m a x i m a l . H o w e v e r , the results o f this study must be interpreted w i t h caut ion because o n l y three an imals per group were observed. S m i t h and Harrap (1997) recorded the behavioural responses o f rats to both r a p i d - f i l l and g radua l - f i l l euthanasia. They observed w a l l c l i m b i n g i n on ly one a n i m a l and d i d not observe any other escape behaviours , but they d i d observe c i r c l i n g o f the chamber by the major i ty o f an imals i n both groups and immediate ur inat ion by a l l rats i n the rapid fill group. The c i r c l i n g behaviour was not quant i f ied so it is not k n o w n whether it d i f fered between groups, and this study d i d not use controls or acc l imat i za t ion , so it is unclear h o w m u c h o f the effect was due to hand l ing and novelty . H o w e v e r , ur inat ion by a l l rats i n one group and none i n the other does suggest distress i n the rap id f i l l group. F i n a l l y , Br i t t (1987) examined the responses o f two strains o f rats dur ing basel ine and dur ing exposure to either p r e - f i l l or g radua l - f i l l CO2 euthanasia. H o w e v e r , interpretation o f this data is d i f f i cu l t because statistical analyses were not per formed and in format ion o n var iab i l i ty was not p rov ided . P r e - f i l l euthanasia was on ly completed w i t h one rat. D u r i n g g radua l - f i l l exposure, Sprague D a w l e y rats showed increases i n act iv i ty and c l i m b i n g , w h i l e L is ter H o o d e d rats showed a decrease i n act iv i ty , no change i n c l i m b i n g and an increase i n backward movements . Increased rear ing and m o v i n g indicate increased exp lorat ion , and the reduct ion i n act iv i ty i n the second strain i s 'd i f f i cu l t to interpret because o f the narcot ic effects o f CO2. The presence o f backward movements , w h i c h are not general ly seen i n rats, m a y indicate that the an imals were t r y ing to remove themselves from the st imulus. B o t h strains also showed increases i n shaking , ur inat ion and defecation. Increased shaking indicates that contrary to prev ious suggestions, g radua l - f i l l exposure m a y result i n pa in pr ior to loss o f consciousness, and increased ur inat ion and defecation suggest act ivat ion o f the autonomic nervous system. H o w e v e r , shak ing was not def ined, m a k i n g the occurrence o f this behaviour d i f f icu l t to 23 \ interpret. A l t h o u g h ultrasonic data were recorded, no vocal i zat ions were detected for either strain. In summary , the major i ty o f research on the behav ioura l responses o f rats to C O 2 exposure has been poor ly designed and the results are d i f f i cu l t to interpret. W h i l e some studies have found no evidence o f distress dur ing C O 2 exposure, others have found behaviours that suggest distress and poss ib l y pa in . W e l l - c o n t r o l l e d research is necessary to c lar i fy these discrepancies. 1.4.2.2 Aversion Testing A v e r s i o n testing can be used to examine whether an an imal w i l l a v o i d a st imulus , as w e l l as the strength o f avers ion to that st imulus. F i rs t ly , we can observe whether an an imal w i l l remove i tse l f f r o m a st imulus i f g i ven the opportunity , and h o w q u i c k l y it does so. Second ly , w e can test the strength o f avers ion to the st imulus by h a v i n g the an imal w o r k to avo id the st imulus , or by c o m p a r i n g the st imulus to another st imulus o f k n o w n value us ing either approach- avoidance testing or avo idance-avo idance testing. In approach-avoidance conf l ic t , an attractive st imulus is paired w i t h an aversive s t imulus , and the an imal must determine whether it is w i l l i n g to accept the aversive s t imulus i n order to gain access to the attractive one. In avo idance - avoidance test ing, the a n i m a l must choose between exposure to one o f two aversive s t imul i . B y c o m p a r i n g the test st imulus to other s t imul i o f k n o w n value, it is poss ib le to rate h o w aversive the test st imulus is (Rushen, 1996). 'A Preference studies have been used to determine whether rats f i n d different concentrations o f C O 2 aversive, and h o w these rank against a rgon - induced h y p o x i a (<2% O 2 ) and gaseous anaesthetics. L e a c h et a l . (2002 a,b) tested rats i n a preference system consist ing o f two chambers connected by a tunnel , where one chamber contained the test gas and the other contained air. D u r i n g contro l sessions w i t h air, rats w i thdrew f r o m the test chamber i n 24 approx imate ly 15 s, whereas dur ing testing w i t h CO2 at concentrations greater than 2 5 . 5 % , rats w i t h d r e w f r o m the test chamber i n approximately . 1 s. Furthermore, dur ing the 180 s test session, the t ime spent i n the CO2 chamber was o n l y 1 to 2 s, w h i l e the t ime spent i n this chamber dur ing contro l sessions was 35 to 50 s. T i m e to exit and t ime spent i n the chamber were s igni f icant ly greater for severe h y p o x i a and anaesthetic gases, suggesting that h y p o x i a and anaesthetic gases are less aversive than CO2. These results indicate that rats are able to q u i c k l y detect CO2 at concentrations greater than 2 5 . 5 % and that they f ind these concentrations aversive. H o w e v e r , because there was no cost to leav ing the chamber , these results do not indicate the strength o f avers ion to CO2. Furthermore, these results on ly show the response to moderate, static concentrations o f CO2, and responses to g radua l - f i l l exposure might di f fer . It is thought that anaesthesia begins to occur at l o w levels o f CO2, and an imals m a y therefore be part ia l ly anesthetized w h e n moderate concentrations o f CO2 occur dur ing g radua l - f i l l exposure. W h i l e the strength o f avers ion to CO2 has not yet been examined i n rats, approach- avoidance testing has been used to examine CO2 avers ion i n other species. Studies w i t h poult ry have found that a large proport ion o f birds w i l l enter a chamber conta in ing CO2 at concentrations greater than 6 0 % i n order to gain access to food or soc ia l contact (Gerr i tzen, et a l . , 2 0 0 0 ; R a j , 1996; Webster & Fletcher , 2004) . Interestingly, Ra j (1996) found that birds that entered a chamber conta in ing 7 5 % CO2 d i d so w h i l e exh ib i t ing symptoms o f pa in and distress such as gasping, head shaking and v o c a l i z i n g . A v e r s i o n to CO2 has also been examined i n p igs , and w h i l e they w i l l general ly tolerate 3 0 % CO2 i n order to ga in access to a food reward , they w i l l not tolerate exposure to 9 0 % CO2, even after a 2 4 - h per iod o f food depr ivat ion (Raj and Gregory , 1995). M i n k have been shown to w o r k to obtain access to nove l objects ( M a s o n et a l . , 2001) , but w i l l avo id a chamber conta in ing a n o v e l object w h e n it contains 1 0 0 % CO2 (Cooper et a l . , 1998). The fact that p igs and m i n k are w i l l i n g to forgo access to desired i tems to avo id h igh concentrations o f CO2 suggests that it is aversive. H o w e v e r , responses to moderate CO2 25 concentrations that are suff ic ient to induce loss o f consciousness have not been examined. S i m i l a r studies are needed to determine whether rats exhib i t s igni f icant avers ion to CO2 exposure. 1.4.2.3 Tests of anxiety' Researchers have used the V o g e l conf l ic t test to determine whether CO2 causes what they have termed 'anx iety ' i n rats. F o r the V o g e l conf l i c t test, rats are g i ven l im i ted access to water, and dur ing spec i f ied per iods they can dr ink but on ly w i t h concomitant exposure to m i l d electr ic shock. It is predicted that w h e n an anx ie ty -p rovok ing st imulus is del ivered pr ior to water access, d r i n k i n g w i l l be reduced: C u c c h e d d u et a l . (1995) observed the effect o f a 10- minute exposure to gradual fill w i t h a 3 5 % CO2 / 6 5 % O2 mixture , and found a 4 0 % reduct ion i n d r ink ing . Th i s result was s imi la r to the effect o f dos ing w i t h an anx iogenic compound . It was also determined that this suppression cou ld be e l iminated by dos ing the rats w i t h anx io ly t ics pr ior to CO2 exposure. A l t h o u g h this study uses an indirect method o f measurement, it does indicate that exposure to an increasing concentrat ion o f CO2 causes anxiety i n rats. 1.4.2.4 Summary of Behavioural Measures: M o s t studies have not found behaviours that are indicat ive o f distress dur ing exposure o f rats to CO2, either us ing the p r e - f i l l or g radua l - f i l l methods. H o w e v e r , the results o f behavioural tests w i t h rats suggest that they are averse to concentrations o f CO2 that are suff ic ient to cause loss o f consciousness. A v e r s i o n to g radua l - f i l l CO2 exposure has not yet been examined , and the strength o f CO2 avers ion i n rats is s t i l l u n k n o w n . 26 1.5 Objectives A l t h o u g h p r e - f i l l CO2 euthanasia o f rats is re lat ively fast, h u m a n self - report data and data o n nociceptor s t imulat ion suggest that it has a h igh potential for pa in i n rats. It also has a h i g h potential for n o n - p a i n distress due to dyspnea. The potential for distress dur ing g radua l - f i l l CO2 euthanasia is less clear because loss o f consciousness occurs at approx imately 4 0 % CO2 ( S m i t h and Har rap , 1997). In general , 4 0 % CO2 is not suff ic ient to cause pa in i n humans, or to st imulate nociceptors i n rat nasal mucosa . H o w e v e r , it is suff ic ient to cause dyspnea i n humans, and it is poss ib le that rats also experience this sensation dur ing exposure to increased levels o f CO2. The major i ty o f studies e x a m i n i n g responses to CO2 exposure have not observed clear behavioural signs o f distress i n rats, but the studies to date have been poor ly executed. B e h a v i o u r a l testing has determined that CO2 exposure causes avers ion i n rats and other species, but the strength o f rats' avers ion to CO2 and the effects o f g radua l - f i l l exposure have not yet been determined. F r o m the in format ion currently avai lable it is not poss ib le to determine conc lus ive ly whether either method o f CO2 euthanasia causes distress in rats. H o w e v e r , p r e - f i l l CO2 exposure has a h igh potential for caus ing pa in , so I dec ided to focus m y dissertation research on gradual - fill CO2 euthanasia. The two m a i n objectives o f m y thesis were: 1) to determine whether g radua l - f i l l CO2 euthanasia causes distress in laboratory rats, by e x a m i n i n g behavioural responses dur ing euthanasia, and avers ion dur ing approach-avo idance testing, and 2) to determine whether p a i n , dyspnea and novel ty are l i k e l y sources o f distress dur ing g radua l - f i l l CO2 euthanasia. In m y first study (Chapter 2) I per formed a detai led analysis o f behavioural responses o f rats dur ing g radua l - f i l l CO2 euthanasia i n order to determine whether they show behavioural signs o f distress. I also examined rats' responses to a reduct ion i n O2 concentrations to 27 determine whether h y p o x i a p layed a role i n their responses to CO2. In m y second study (Chapter 3), I used approach-avoidance testing to determine whether rats f i n d CO2 more aversive than a valuable food reward , and to determine w h i c h concentrations o f CO2 rats f i n d aversive w h e n tested w i t h either static or gradual ly increasing CO2. I also examined rat avers ion to argon - induced h y p o x i a , w h i c h has been suggested as an alternative gas euthanasia method. In m y third study (Chapter 4) , I expanded m y invest igat ion o f rat avers ion to gradual ly increasing concentrations o f CO2 by determining whether C 0 2 flow rate affects rat avers ion to CO2. W i t h i n the first three studies I also examined the CO2 concentrations that resulted i n behavioural signs o f distress and avers ion, and compared these values w i t h those f r o m previous studies on the potent ial for CO2 to cause pa in and dyspnea. In do ing this I was able to assess whether pa in and dyspnea were l i k e l y causes o f this, distress and aversion. In m y final study (Chapter 5), I investigated whether novel ty was a cause o f distress and avers ion dur ing g radua l - f i l l CO2 exposure. 1) 28 1.6 References 2000 . The Canad ian O x f o r d D ic t ionary , B isset , A . (ed). D o n M i l l s : O x f o r d Un ivers i t y Press. A m b r o s e , N . , W a d h a m , J . , M o r t o n , D. , 2000 . Ref inement i n Euthanasia . In: B a l l s , M . , v a n Ze l le r , A . M . , Ha ider , M . E . (Eds) , Progress i n the R e d u c t i o n , Ref inement and Replacement o f A n i m a l Exper imentat ion , E lsev ie r Sc ience, A m s t e r d a m , p p . 1 1 5 9 - 1 1 6 9 . A m e r i c a n Veter inary M e d i c a l A s s o c i a t i o n , 2 0 0 1 . 2 0 0 0 Report o f the A V M A Pane l on Euthanasia . Journal o f the Veter inary M e d i c a l A s s o c i a t i o n 2 1 8 , 6 6 9 - 6 9 6 . A n t o n , F., Euchner , I., Handwerker , H .O . , 1992. Psychophys i ca l examinat ion o f pa in induced by def ined CO2 pulses appl ied to the nasal mucosa . P a i n 4 9 , 5 3 - 6 0 . A n t o n , F., Peppe l , P. , Euchner , I., Handwerker , H .O . , 1991. Cont ro l led nox ious chemica l s t imulat ion : responses o f rat t r igeminal brainstem neurones to CO2 pulses appl ied to the nasal mucosa . N e u r o s c i . Lett . 123, 2 0 8 - 2 1 1 . Aus t ra l ian and N e w Zea land C o u n c i l for the Care o f A n i m a l s i n Research and Teach ing , 1993. Euthanas ia o f A n i m a l s U s e d for Sc ient i f i c Purposes, A N Z C C A R T , G l e n O s m o n d . Banzett , R . B . , L a n s i n g , R . W . , E v a n s , K . C . , Shea, S . A . , 1996. St imulus - response characteristics o f C02 - induced air hunger i n no rmal subjects. Resp . P h y s i o l . 103, 1 9 - 3 1 . Barnett , S . A . 1975. The Rat : A Study i n Behav io r . C h i c a g o : The Un ive rs i t y o f C h i c a g o Press. Bergersweeney , J . , Berger , U .V . , Sharma, M . , P a u l , C . A . , 1994. Ef fects o f carbon d i o x i d e - induced anaesthesia on cho l inerg ic parameters i n rat -brain. L a b . A n i m . S c i . 4 4 , 3 6 9 - 3 7 1 . B l a c k m o r e , D . K . , 1993. Euthanasia ; not a lways eu. Aus t . Vet . J . 70 , 4 0 9 - 4 1 3 . B l a c k s h a w , J . K . , F e n w i c k , D . C . , Beatt ie , A . W . , A l l a n , D . J . , 1988. The behaviour o f ch ickens , m i c e and rats dur ing euthanasia w i t h ch lo ro fo rm, carbon d iox ide and ether. L a b . A n i m . 2 2 , 6 7 - 7 5 . B l a n c h a r d , R . J . , B l a n c h a r d , D . C . , A g u l l a n a , R., W e i s s , S . M . , 1991. T w e n t y - t w o k H z a larm cries 29 to presentation o f a predator, by laboratory rats l i v i n g i n v i s ib le bur row systems. P h y s i o l . Behav . 5 0 , 9 6 7 - 9 7 2 . B l o c k , N . , 1995. O n a confus ion about a funct ion o f consciousness. B e h a v . B r a i n S c i . 18, 2 2 7 - 2 8 7 . B o r o v s k y , V . , H e r m a n , M . , D u n p h y , G . , C a p l e a , A . , E l y , D . , 1998. CO2 asphyx ia increases p l a s m a norepinephrine i n rats v i a sympathetic nerves. A m . J . o f P h y s i o l . 2 7 4 , R 1 9 - R 2 2 . Br i t t , D . P. , 1987. The humaneness o f carbon d iox ide as an agent o f euthanasia for laboratory rodents. In: Euthanas ia o f U n w a n t e d , Injured or D iseased A n i m a l s or for Educat iona l or Sc ient i f i c Purposes, Un ivers i t ies Federat ion for A n i m a l We l fa re , Potters B a r , pp. 1 9 - 3 1 . B r o d i e , D . A . , W o o d b u r y , D . M . , 1958. A c i d - b a s e changes i n bra in and b l o o d o f rats exposed to h i g h concentrations o f carbon d iox ide . A m . J . P h y s i o l . 192, 9 1 - 9 4 . B r u d z y n s k i , S . M . , O c i e p a , D. , 1992. U l t rason ic voca l i za t ion o f laboratory rats i n response to hand l ing and touch. P h y s i o l . Behav . 5 2 , 6 5 5 - 6 6 0 . C a b l e , G . G . , 2 0 0 3 . In - f l ight h y p o x i a incidents i n mi l i ta ry aircraft : causes and impl icat ions for t ra in ing. A v i a t . Space E n v i r o n . M e d . 74 , 169 -172 . C a i n , W . S . , M u r p h y , C . L . , 1980. Interaction between chemorecept ive modal i t ies o f odour and i r r i tat ion. Nature 2 8 4 , 255 - 257 . Canad ian C o u n c i l o n A n i m a l Care , 1989. E th ics o f A n i m a l Investigation (1989). A v a i l a b l e at: ht tp ://www.ccac.ca/en/CCAC Programs/Guidel ines P o l i c i e s / P O L I C I E S / E T H I C S . H T M . A c c e s s e d M a y 2 0 0 6 . Canad ian C o u n c i l on A n i m a l Care , 1993. G u i d e to the Care and U s e o f Exper imenta l A n i m a l s , V o l u m e 1, 2 n d E d i t i o n , eds E . D . Ol fert , B . M . C ross and A . A . M c W i l l i a m . Ottawa, C C A C . Canad ian C o u n c i l o n A n i m a l Care , 2006 . C C A C Survey o f A n i m a l U s e - 2004. A v a i l a b l e at: http://www.ccac.ca/en/Publications/New Facts Figures/analysis/analysis index .htm. A c c e s s e d M a y 2006 . 30 Chen, X . , Gallar, J . , Pozo, M . A . , Baeza, M . , Belmonte, C , 1995. CO2 stimulation of the cornea: a comparison between human sensation and nerve activity in polymodal nociceptive afferents of the cat. Eur. J. Neurosci. 7, 1154-1163. Close, B. , Banister, K . , Baumans, V . , Bernoth, E . , Bromage, N . , Bunyan, J. , Erhardt, W. , Flecknell, P., Gregory, N . , Hackbarth, H . , Morton, D . , Warwick, C , 1997. European Commission Working Party Report: Recommendations for euthanasia of experimental animals, Part I. Lab. Anim. 30, 293-316. Coenen, A . M . , Drinkenburg, W . H . , Hoenderken, R., van Luijtelaar, G . L . , 1995. Carbon dioxide euthanasia in rats: oxygen supplementation minimizes signs of agitation and asphyxia. Lab. Anim. 29, 262-268. Coenen, A . , Smit, A . , Zhonghua, L . , van Luijtelaar, G . , 2000. Gas mixtures for anaesthesia and euthanasia in broiler chickens. World Poultry Sci. J. 56, 225-234. Colpaert, F . C . , 1987. Evidence that adjuvant arthritis in the rat is associated with chronic pain. Pain 28, 201-222. Cooper, J . , Mason, G . , Raj, M . , 1998. Determination of the aversion of farmed mink (Mustela vison) to carbon dioxide. Vet. Rec. 143, 359-61. Cuccheddu, T., Floris, S., Serra, M . , Porceddu, M . L . , Sanna, E . , Biggio, G . , 1995. Proconflict effect of carbon dioxide inhalation in rats. Life Sci. 56, 321-324. Cuomo, V . , Cagiano, R., De Salvia, M . A . , Mazzoccoli, M . , Persichella, M . , Renna, G . , 1992. Ultrasonic vocalization as an indicator of emotional state during active avoidance learning in rats. Life Sci. 50, 1049-1055. Danneman, P.J., Stein, S., Walshaw, S.O., 1997. Humane and practical implications of using carbon dioxide mixed with oxygen for anaesthesia or euthanasia of rats. Lab. Anim. Sci. 47, 376-85. Dawkins, M.S . , 1980. Animal Suffering: The Science of Animal Welfare. Chapman and Hall, 31 L o n d o n . D a w k i n s , M . S . 1998. E v o l u t i o n and an imal welfare. Q. R e v . B i o l . 7 3 : 3 0 5 - 328 . D ie lenberg , R . A . , Car r i ve , P. , M c G r e g o r , I.S., 2 0 0 1 . The cardiovascular and behavioural response to cat odor i n rats: uncondi t ioned and condi t ioned effects. B r a i n Res . 897, 2 2 8 - 2 3 7 . D i n h , H . K . , L a r k i n , A . , G a t l i n , L. , P iepmeier , E. Jr., 1999. Rat ul t rasound m o d e l for measur ing pa in result ing f r o m intramuscular ly injected ant imicrobia ls . P D A J . P h a r m . S c i . Tech . 5 3 , 4 0 - 4 3 . D r i p p s , R . D . , C o m r o e , J . H . , 1947. The respiratory and c i rculatory response o f normal m a n to inhalat ion o f 7.6 and 10.4 per cent C O v w i t h a compar i son o f the m a x i m a l vent i lat ion produced by severe muscular exercise, inhalat ion o f CO2 and m a x i m a l voluntary hypervent i lat ion. A m . J . P h y s i o l . 149, 4 3 - 5 1 . D u l l a , C . G . , D o b e l i s , P. , Pearson, T., F rengue l l i , B . G . , Staley, K . J . , M a s i n o , S . A . 2 0 0 5 . A d e n o s i n e and A T P l i n k Pco2 to cort ical exc i tabi l i ty v i a p H . N e u r o n 4 8 , 1011-1023 E i se le , J . H . , Eger , E . L , M u a l l e m , M . , 1967. N a r c o t i c properties o f carbon d iox ide i n the dog. Anesthes io logy 2 8 , 8 5 6 - 8 6 4 . F a w e l l , J . K . , T h o m s o m , C . , C o o k e , L., 1972. Respi ratory artefact produced by carbon d iox ide and pentobarbitone s o d i u m euthanasia i n rats. L a b . A n i m . 6, 3 2 1 - 3 2 6 . F e n g , Y . , S i m p s o n , T. L. , 2 0 0 3 . N o c i c e p t i v e sensation and sensit iv i ty evoked f r o m h u m a n cornea and conjunct iva st imulated by CO2. Invest. Ophth . V i s . S c i . 4 4 , 5 2 9 - 5 3 2 . F o w l e r , W . S . , 1954. B r e a k i n g point o f breath -ho ld ing . J . A p p l . P h y s i o l . 6, 5 3 9 - 5 4 5 . Fraser, D . , W e a r y , D . M . , Pajor , E . A . , M i l l i g a n , B . N . , 1997. A sc ient i f ic concept ion o f an imal wel fare that reflects ethical concerns. A n i m . Wel fa re 6, 1 8 7 - 2 0 5 . F u k u d a , Y . , Satao, A , S u z u k i , A , T r z e b s k i , A . , 1989. A u t o n o m i c nerve and cardiovascular response to chang ing b lood oxygen and carbon d iox ide levels i n the rat. J . A u t o n . N e r v . Syst! 2 8 , 6 1 - 7 4 . . 32 Gerr i t zen , M . A . , L a m b o o i j , E. , H i l l e b r a n d , S . J . W . , Lanhaar , J . A . C . , Pieterse, C , 2000 . Behav io ra l responses o f broi lers to different gaseous atmospheres. Pou l t ry S c i . 7 9 , 9 2 8 - 9 3 3 . G i f f o r d , R . G . , M o n y e r , H . , Chr is t ine , C . W . , C h o i , D . W . A c i d o s i s reduces N M D A receptor act ivat ion, glutamate neurotox ic i ty , and oxygen -g lucose depr ivat ion neuronal in jury i n cort ical cultures. B r a i n R e s . 506 , 3 3 9 - 3 4 2 . Hackbar th , H . , K u p p e r s , N . , Bohnet , W . , 2000 . Euthanasia o f rats w i t h carbon d i o x i d e — a n i m a l wel fare aspects. L a b . A n i m . 34 , 9 1 - 9 6 . Ha ldane , J . , S m i t h , F., 1892. J . Patho l . Bacter io l . 168. C i t e d by H i l l and F l a c k , 1908. Hauser , M . D . 1996. The E v o l u t i o n o f C o m m u n i c a t i o n . M I T Press, C a m b r i d g e , Massachusetts . H e r m a n , J .P . & C u l l i n a n , W . E . , 1997. Neuroc i rcu i t r y o f stress: central contro l o f the hypothalamo-p i tu i tary -adrenocort ica l axis . Trends N e u r o s c i . 2 0 , 7 8 - 8 4 . H e r m a n , J . P. , Prewit t , C . M . F., C u l l i n a n , W . E., 1996. N e u r o n a l c i rcui t regulat ion o f the hypothalamo-p i tu i tary -adrenocort ica l stress ax is . Cr i t . R e v . N e u r o b i o l . 10, 3 7 1 - 3 9 4 . Hewett , T . A . , K o v a c s , M . S . , A r t w o h l , J .E . , Bennett , B.T . , 1993. A compar ison o f euthanasia methods i n rats, us ing carbon d iox ide i n p r e - f i l l e d and fixed flow rate f i l l e d chambers. L a b . A n i m . S c i . 4 3 , 5 7 9 - 5 8 2 . H i l l , L. , F l a c k , M . , 1908. The effect o f excess o f carbon d iox ide and o f want o f oxygen upon the respirat ion and the c i rcu lat ion . J . P h y s i o l . 3 7 , 7 7 - 1 1 1 . H i ra ta , H . , H u , J . W . , Bereiter , D . A . , 1999. Responses o f medul la ry dorsal horn neurons to corneal s t imulat ion by CO2 pulses i n the rat. J . N e u r o p h y s i o l . 82 , 2092 - 2 1 0 7 . Hodges , M . R . , Forster, H . V . , Papanek, P . E . , D w i n e l l , M . R . & H o g a n , G . E . , 2 0 0 2 . Vent i la tory phenotypes a m o n g four strains o f adult rats. J . A p p l . P h y s i o l . 9 3 , 9 7 4 - 9 8 3 . Hornett , T . D . , Haynes , A . R . , 1984. C o m p a r i s o n o f carbon dioxide/air mixture and nitrogen/air mixture for the euthanasia o f rodents. D e s i g n o f a system for inhalat ion euthanasia. A n i m a l Techno logy 3 5 , 9 3 - 9 9 . 33 H s u , K . , L i a n g , Y . , H u a n g , C , 2 0 0 0 . Inf luence o f an extracel lualr ac idos is on excitatory synaptic t ransmiss ion and long - te rm potentiat ion i n the C A I reg ion o f rat h ippocampal s l ices. J . N e u r o s c i . Res . 6 2 , 4 0 3 - 4 1 5 . Iwarsson, K . , Rehb inder , C , 1993. A study o f different euthanasia techniques i n guinea p igs , rats, and m i c e . A n i m a l response and postmortem f indings . Scan . J . L a b . A n i m . S c i . 2 0 , 1 9 1 - 2 0 5 . Jourdan, D. , A r d i d , D . , Chapuy , E. , Escha l ie r , A . , L e Bars , D . , 1995. A u d i b l e and ultrasonic voca l i za t ion e l ic i ted by single electr ical noc icept ive s t imu l i to the ta i l i n the rat. P a i n 6 3 , 2 3 7 - 2 4 9 . Jourdan, D. , A r d i d , D . , Chapuy , E., L e Bars , D . , Eschal ie r , A . , 1998. E f fect o f analgesics o n audible and ultrasonic pa in - induced voca l i za t ion i n the rat. L i f e S c i . 6 3 , 1 7 6 1 - 1 7 6 8 . Ka l twasser , M . T . , 1990. Star t le - induc ing acoustic s t imul i evoke ul t rasonic voca l i za t ion i n the rat. P h y s i o l . Behav . 4 8 , 1 3 - 1 7 . K c , P . , H a x h i u , M . A . , T routh , C O . , B a l a n , K . V . , A n d e r s o n , W . A . , M a c k , S . O . , 2 0 0 2 . C 0 2 - induced c - F o s express ion i n hypothalamic vasopressin conta in ing neurons. Resp . P h y s i o l . 1 2 9 , 2 8 9 - 2 9 6 . K i t c h e n , H . , A r o n s o n , A . L . , B i t t l e , J . L . , M c P h e r s o n , C . W . , M o r t o n , D . B . , Pakes , S . P . , R o l l i n , B . E . , R o w a n , A . N . , Sechzer , J . A . , V a n d e r l i p , J .E . , W i l l , J . A . , C l a r k , A . S . , G l o y d , J .S . , 1987. Pane l report o n the C o l l o q u i u m o n recogni t ion and a l lev iat ion o f a n i m a l pa in and distress. J . A m . Ve t . M e d . A s s o c . 191, 1 1 8 6 - 1 1 9 1 . K l u m p , G . M . , Shalter, M . D . , 1984. A c o u s t i c behaviour o f b i rds and m a m m a l s i n the predator context: I. Factors af fect ing the structure o f a larm signals. II. The funct ional s igni f icance and evo lu t ion o f a la rm signals. Z . T ie rpsycho l . 6 6 , 1 8 9 - 2 2 6 . C i t e d i n Hauser , M . D . 1996 The E v o l u t i o n o f C o m m u n i c a t i o n , M I T Press, C a m b r i d g e , p p . 4 1 3 - 4 1 9 . K n u t s o n , B . , Burgdor f , J . , Panksepp, J . , 2 0 0 2 . U l t rason ic voca l i zat ions as indices o f affect ive 34 states i n rats. P s y c h o l . B u l l . 128, 9 6 1 - 9 7 7 . K o h l e r , I., M e i e r , R., Busato , A . , N e i g e r - A e s c h b a c h e r , G . , Schatzmann, U . , 1999. Is carbon d iox ide (CO2) a useful short act ing anaesthetic for sma l l laboratory an imals? L a b . A n i m . 3 3 , 1 5 5 - 1 6 1 . K o o l h a s , J . M . , K o r t e , S . M . , D e Boer , S . F . , v a n der Veg t , B . J . , van Reenen , C . G . , Hopster , H . , de Jong , L C , R u i s , M . A . W . , B l o k h u i s , H.J . , 1999. C o p i n g styles i n an imals : current status i n behavior and stress -physio logy. N e u r o s c i . B iobehav . R. 2 3 , 9 2 5 - 9 3 5 . L e a c h , M . C , B o w e l l , V . A . , A l l a n , T .F . , M o r t o n , D . B . , 2 0 0 2 a . A v e r s i o n to gaseous euthanasia agents i n rats and m i c e . Comparat i ve M e d . 5 2 , 2 4 9 - 2 5 7 . L e a c h , M . C , B o w e l l , V . A . , A l l a n , T .F . , M o r t o n , D . B . , 2002b . Degrees o f avers ion shown by rats and m i c e to different concentrations o f inhalat ional anaesthetics. Ve t . R e c . 150, SOS- S I S . L e e , J . , Ta i ra , T., P ih la ja , P. , R a n s o m , B . R . , K a i l a , K . , 1996. Ef fects o f CO2 on excitatory t ransmiss ion apparently caused by changes i n intracel lular p H i n the rat h ippocampa l s l ice . B r a i n Res . 7 0 6 , 2 1 0 - 2 1 6 . L i o t t i , M . , B rannan , S . , E g a n , G . , Shade, R., M a d d e n , L. , A b p l a n a l p , B . , R o b i l l a r d , R., Lancaster , J . , Zamar r ipa , F .E . , F o x , P.T. , Denton , D. , 2 0 0 1 . B r a i n responses associated w i t h consciousness o f breathlessness (air hunger). P roc . N a t . A c a d . S c i . 98 , 2 0 3 5 - 2 0 4 0 . L u s h , M . T . , Janson -B je rk l ie , S . , Car r ie r i , V . K . , L o v e j o y , N . , 1988. D y s p n e a i n the vent i lator - assisted patient. Heart L u n g , 17, 5 2 8 - 5 3 5 . Maro t ta , S . F . , S i th ichoke , N . , G a r c y , A . M . , Y u , M . , 1976. Adrenocor t i ca l responses o f rats to acute h y p o x i c and hypercapnic stresses after treatment w i t h aminerg ic agents. Neuroendocr ino logy 2 0 , 182 -192 . Mar to f t , L. , L o m h o l t , L. , K o l t h o f f , C , Rodr igues , B . E . , Jensen, E . W . , Jorgensen, P .F . , Pedersen, H . D . , F o r s l i d , A . , 2 0 0 2 . Ef fects o f CO2 anaesthesia o n central nervous system act iv i ty i n 35 swine. L a b . A n i m . 3 6 , 115 -126 . Mar to f t , L. , S todki lde - Jorgensen, H . , F o r s l i d , A . , Pedersen, H . D . & Jorgensens, P .F . , 2 0 0 3 . CO2 induced acute respiratory acidosis and intracel lular p H : a 3 I P N M R study i n swine. L a b . A n i m . 3 7 , 2 4 1 - 2 4 8 . M a s o n , G . J . , Cooper , J . , C la rebrough , C , 2001 Frustrations o f fu r - fa rmed m i n k . Nature 4 1 0 , 3 5 - 36. M a s u d a , A . , O h y a b u , Y . , K o b a y a s h i , T., Y o s h i n o , C , Sakakibara , Y . , K o m a t s u , T. , H o n d a , Y . , 2 0 0 1 . L a c k o f positive' interact ion between CO2 and h y p o x i c s t imulat ion for Pco2 - V A S response slope i n humans. Resp . P h y s i o l . 126, 1 7 3 - 1 8 1 . M e y e r , J .S . , G o t o h , F., T a z a k i , Y . , 1961. CO2 narcosis : an exper imental study. N e u r o l o g y 11, 5 2 4 - 5 3 7 . M e y e r , J .S . , G o t o h , F., T o m i t a , M . , 1966. A c u t e respiratory academia. Cor re lat ion o f jugu lar b l o o d c o m p o s i t i o n and electroencephalogram dur ing CO2 narcosis . N e u r o l o g y 16, 4 6 3 - 4 7 4 . M o b e r g , G . P . , 2 0 0 0 . B i o l o g i c a l responses to stress: impl i cat ions for a n i m a l welfare. In: M o b e r g , G . P . , M e n c h , J . A . (eds), B i o l o g y o f A n i m a l Stress, C A B I P u b l i s h i n g , N e w Y o r k , pp. 1-21 M o n t g o m e r y , K . C . , 1955. The relat ion between fear induced by n o v e l s t imulat ion and exploratory behavior . J . C o m p . P h y s i o l . P s y c h o l . 4 8 , 2 5 4 - 2 6 0 . V M o o s a v i , S . H . , Go lestan ian , E. , B i n k s , A . P . , L a n s i n g , R . W . , B r o w n , R., Banzett , R . B . , 2 0 0 3 . H y p o x i c and hypercapnic dr ives to breathe generate equivalent levels o f air hunger i n humans. J . A p p l . P h y s i o l . 94 , 141 -154 . N a t i o n a l Research C o u n c i l , 2 0 0 3 . Gu ide l ines for the Care and U s e o f M a m m a l s i n Neurosc ience and B e h a v i o r a l Research . The N a t i o n a l A c a d e m i e s Press, W a s h i n g t o n , D . C . , p. 16. N i e l s o n , M . , S m i t h , H . , 1952. Studies o n the regulat ion o f respirat ion i n acute hypox ia . A c t a . P h y s i o l . Scand . 2 4 , 2 9 3 - 3 1 3 . Pasternak, M . , Bount ra , C , V o i p i o , J . , K a i l a , K . , 1992. Inf luence o f extracel lular and 36 intracel lular p H o n G A B A - g a t e d ch lor ide conductance i n c ray f ish musc le f ibers. Neurosc ience 4 7 , 9 2 1 - 9 2 9 . Pecaut, M . J . , S m i t h , A . L . , Jones, T . A . , G r id ley , D . S . , 2000 . M o d i f i c a t i o n o f i m m u n o l o g i c and hemato log ic var iables by method o f CO2 euthanasia. Comparat i ve M e d . 50 , 5 9 5 - 6 0 2 . P e p p e l , P. , A n t o n , F., 1993. Responses o f rat medul la ry dorsal horn neurons f o l l o w i n g intranasal nox ious chemica l s t imulat ion : effects o f st imulus intensity, durat ion, and interst imulus interval . J . N e u r o p h y s i o l . 70 , 2 2 6 0 - 2 2 7 5 . Raf f , H . , Roar ty , T . P . , 1988. R e n i n , A C T H , and aldosterone dur ing acute hypercapnia and h y p o x i a i n conscious rats. A m . J . P h y s i o l . 2 5 4 , R 4 3 1 - R 4 3 5 . R a j , A . B . , 1996. A v e r s i v e reactions o f turkeys to argon, carbon d iox ide and a mixture o f carbon d iox ide and argon. Ve t . R e c . 138, 5 9 2 - 5 9 3 . R a j , A . B . , Johnson, S . P . , W o t t o n , S . B . , M c l n s t r y , J . L . , 1997. We l fa re impl icat ions o f gas stunning p igs : 3 . The t ime to loss o f Somatosensory evoked potentials and spontaneous e lectrocort icogram o f p igs dur ing exposure to gases. Ve t . J . 153, 3 2 9 - 3 3 9 . R a j , A . B . M . , W o t t o n , S . B . , Gregory , N . G . , 1992. Changes i n the somatosensory evoked potential and spontaneous electroencephalogram o f hens dur ing stunning w i t h a carbon d iox ide and argon mixture . Br i t . Ve t . J . 4 8 , 147 -156 . R a j , A . B . , W o t t o n , S . B . , M c K i n s t r y , J . L . , H i l l e b r a n d , S . J . , Pieterse, C , 1998. Changes i n the somatosensory evoked potentials and spontaneous electroencephalogram o f broi ler chickens dur ing exposure to gas mixtures . Br i t . Poul t ry S c i . 39 , 6 8 6 - 6 9 5 . R a j , A . B . M . , Gregory , N . G . , 1995. Wel fare impl icat ions o f the gas stunning o f pigs 1. Determinat ion o f avers ion to the in i t ia l inhalat ion o f carbon d iox ide or argon. A n i m . Wel fa re 4 , 2 7 3 - 2 8 0 . R a j , M . , Gregory , N . G . , 1994. A n evaluat ion o f humane gas stunning methods for turkeys. V e t . R e c . 135, 2 2 2 - 2 2 3 . 37 R o w a n , A . N . , Stephens, M L . , D o l i n s , F., G l e a s o n , A . , D o n l e y , L. , 1998. A n i m a l wel fare perspectives o n pa in and distress management i n research and testing. In: Proceedings for P a i n Management and H u m a n e Endpoints , a workshop o f The Johns H o p k i n s Center for A l ternat ives to A n i m a l Test ing , he ld N o v e m b e r 2 - 3 , 1998, W a s h i n g t o n , D C . A v a i l a b l e at: http://altweb.ihsph.edu/meetings/pain/rowan.htm accessed M a y 2006 . Rushen , J . , 1996. U s i n g avers ion learning techniques to assess the menta l state, suffer ing and welfare o f f a r m animals . J . A n i m . S c i . 74 , 1 9 9 0 - 1 9 9 5 . Se lye , H . 1975. C o n f u s i o n and controversy i n the stress f i e ld . J . H u m . Stress 1, 3 7 - 4 4 . Sharp, J . , A z a r , T. , L a w s o n , D. , 2006 . C o m p a r i s o n o f carbon d iox ide , argon, and nitrogen for induc ing unconsciousness.or euthanasia o f rats. J . A m . A s s o c . L a b . A n i m . S c i . 4 5 , 2 1 - 2 5 . Sharp, J . , Z a m m i t , T., A z a r , T., L a w s o n , D. , 2 0 0 2 . St ress - l ike responses to c o m m o n procedures i n rats housed alone or w i t h other rats. C o n t e m p . Top . L a b . A n i m . 4 1 , 8 - 1 4 . Sharp, J . , Z a m m i t , T., A z a r , T., L a w s o n , D. , 2 0 0 3 . St ress - l ike responses to c o m m o n procedures i n i n d i v i d u a l l y and group-housed female rats. C o n t e m p . Top . L a b . A n i m . 4 2 , 9 - 1 8 . Shea, S . A . , H a t t y , H .R . , Banzett , R . B . , 1996. Se l f - cont ro l o f leve l o f mechan ica l vent i lat ion to m i n i m i z e C O 2 induced air hunger. Resp . P h y s i o l . 1 1 3 - 1 2 5 . S m i t h , W . , Harrap , S . B . , 1997. Behav ioura l and cardiovascular responses o f rats to euthanasia us ing carbon d i o x i d e gas. L a b . A n i m . 3 1 , 3 3 7 - 3 4 6 . Somero , G . N . , 1986. Protons, osmolytes , and fitness o f internal m i l i e u for protein funct ion. A m . J . P h y s i o l . 2 5 1 . R 1 9 7 - R 2 1 3 . Tang , C M . , D ichter , M . , M o r a d , M . , 1990. M o d u l a t i o n o f the N-methy l -D -aspartate channel by extracel lular H + . P roc . N a t l . A c a d . S c i . 87, 6 4 4 5 - 6 4 4 9 . T e r l o u w , E . M . C , Schouten, G . P . , L a d e w i g , J . , 1997. P h y s i o l o g y . In: A p p l e b y , M . C , Hughes , B . O . (eds), A n i m a l We l fa re , C A B I P u b l i s h i n g , W a l l i n g f o r d , pp . 1 4 3 - 1 5 8 . T h o m a s , D . A . , Takahash i , L . K . , B a r f i e l d , R . J . , 1983. A n a l y s i s o f ultrasonic vocal i zat ions 38 emitted by intruders dur ing aggressive encounters a m o n g rats (Rattus norvegicus) . J . C o m p . P s y c h o l . 9 7 , 2 0 1 - 2 0 6 . Thurauf , N . , Gunther , M . , P a u l i , E. , K o b a l , G . , 2 0 0 2 . Sensi t iv i ty o f the negative mucosa l potential to the t r igeminal target st imulus CO2. B r a i n R e s . 9 4 2 , 2 7 - 8 6 . Thurauf , N . , H u m m e l , T. , Ket tenmann , B . , K o b a l , G . , 1993. N o c i c e p t i v e and ref lex ive responses recorded f r o m the human nasal mucosa . B r a i n Res . 6 2 9 , 2 9 3 - 2 9 9 . Toates, F., 1997. Stress: Conceptua l and B i o l o g i c a l Perspect ives. Chichester , John W i l e y and Sons. T o m b a u g h , G . C . , S o m j e n , G . C . , 1996. Ef fects o f extracel lular p H on vol tage-gated N a + , K + , and C a + currents i n isolated rat C A I neurons. J . P h y s i o l . 4 9 3 , 7 1 9 - 7 3 2 . U n i t e d K i n g d o m H o m e O f f i c e , 1995. The Wel fa re o f A n i m a l s (Slaughter or K i l l i n g ) Regulat ions 1995, H e r Majesty 's Stationery O f f i c e , N o r w i c h . U n i t e d K i n g d o m H o m e O f f i c e , 1997. The H u m a n e K i l l i n g o f A n i m a l under Schedule 1 to the A n i m a l s (Sc ient i f ic Procedures) A c t 1986 C o d e o f Pract ice , H e r Majesty 's Stationery O f f i c e , N o r w i c h . V e l i s e k , L. , 1998. Ex t race l lu lar acidosis and h igh levels o f carbon d iox ide suppress synaptic t ransmiss ion and prevent the induct ion o f long - term potentiat ion in the C A I region o f rat h i p p o c a m p a l s l ices. H i p p o c a m p u s 8, 2 4 - 3 2 . V ingerhoets , J . J . M . , 1985. The role o f the parasympathetic d i v i s i o n o f the autonomic nervous system i n stress and emotions. Int. J . Psychosomat ics 3 2 , 2 8 - 3 3 . V i v i a n , J . A . , M i c z e k , K . A . , 1991. U l t rasounds dur ing morphine wi thdrawal i n rats. Psychopharmaco logy 104, 1 8 7 - 1 9 3 . W a l l a c e , K . J . , R o s e n , J . B . , 2000 . Predator odor as an uncondi t ioned fear st imulus in rats: e l ic i tat ion o f f reez ing by t r imethy l th iazol ine , a component o f fox feces. Behav . N e u r o s c i . 1 1 4 , 9 1 2 - 9 2 2 . 39 Webster , A . B . , F letcher , D . L . , 2004. Assessment o f the avers ion o f hens to different gas atmospheres us ing an approach-avoidance test. A p p l . A n i m . Behav . S c i . 88 , 2 7 5 - 2 8 7 . W i d d i c o m b e , J . G . , 1986 Ref lexes f r o m the upper respiratory tract. In: H a n d b o o k o f P h y s i o l o g y , The Respiratory Sys tem, N . S . Chern iak , J . G . W i d d i c o m b e (Eds.) , A m e r i c a n P h y s i o l o g i c a l Soc ienty , Bethesda, 3 6 3 - 3 9 4 . W o o d b u r y , D . M . , R o l l i n s , L .T. , Gardner , M . D . , H i r s c h i , W . L . , H o g a n , J .R . , R a l l i s o n , M . L . , Tanner, G . S . , B r o d i e , D . A . , 1958. Ef fects o f carbon d iox ide o n bra in exc i tabi l i ty and electrolytes. A m . J . P h y s i o l . 192, 7 9 - 9 0 . Y a v a r i , P. , M c C u l l o c h , P .F . , Panneton, W . M . , 1996. T r i gemina l l y -med ia ted alteration o f cardiorespiratory rhythms dur ing nasal appl icat ion o f carbon d iox ide i n the rat. J . A u t o n . N e r v . Syst. 6 1 , 1 9 5 - 2 0 0 . Youngentob , S . L . , H o r n u n g , D . E . , M o z e l l , M . M . , 1991. Determinat ion o f carbon d iox ide detect ion thresholds i n trained rats. P h y s i o l . .Behav . 4 9 , 2 1 - 2 6 . 40 CHAPTER 2: Behavioural responses of rats to gradual-fill carbon dioxide euthanasia and reduced oxygen concentrations 2.1 Introduction M e t h o d s c o m m o n l y used for euthanasia o f sma l l laboratory rodents inc lude phys ica l S techniques, injectable anaesthetics, and exposure to anaesthetic and non-anaesthetic gases. Exposure to CO2 is one o f the most w i d e l y recommended euthanasia methods for rats ( A N Z C C A R T , 1993; A V M A , 2 0 0 0 ; Canad ian C o u n c i l o n A n i m a l Care , 1993; C l o s e et a l . , 1997; U K H o m e O f f i c e , 1997). Rats are either p laced into a chamber p r e - f i l l e d w i t h gas or the gas is gradual ly int roduced into an a i r - f i l l e d chamber , and this results i n narcosis due to the properties o f CO2 f o l l o w e d by death. B o t h methods are re lat ively easy to per fo rm, inexpensive , safe for laboratory workers , and invo lve l itt le handl ing and restraint o f an imals . Ideal ly , a euthanasia method should also result i n a qu ick death w i t h m i n i m a l pa in and distress before loss o f consciousness, but it is not clear whether CO2 euthanasia meets these last cr i ter ia. CO2 forms carbonic ac id when it comes into contact w i t h moisture. It begins to stimulate nociceptors i n rat nasal m u c o s a at CO2 concentrations above 2 5 % , and the threshold for the major i ty o f nociceptors is between 3 7 % and 5 0 % CO2 ( A n t o n et a l . , 1 9 9 1 ; Peppel & A n t o n , 1993). In humans , CO2 is detectable and begins to become pa in fu l at the cornea, conjunct iva and the nasal m u c o s a at concentrations between 30%) and 5 4 % ( A n t o n et a l . , 1992; C h e n et a l , 1995; F e n g & S i m p s o n , 2003) . H i g h CO2 concentrations can also cause dyspnea, or shortness o f breath, w h i c h inc ludes the sensations o f both air hunger and increased breathing effort ( Lans ing et a l . , 2000) . A t l o w levels o f CO2 dyspnea can be overcome by vent i latory adjustments (Shea et a l , 1996), but spontaneously breathing humans report this sensation at CO2 concentrations o f on ly 8%> (Dr ipps & C o m r o e , 1947; L io t to et a l . , 2001) and severity increases w i t h increasing 2 A version of this chapter has been published. Niel, L. , Weary, D . M . , 2006. Behavioural responses of rats to gradual-fill carbon dioxide euthanasia and reduced oxygen concentrations. Appl. Anim. Behav. Sci. (in press). 41 CO2 concentrations (Banzett et a l . , 1996). A d d i t i o n o f CO2 to a chamber causes a reduct ion i n O2 levels w i t h displacement o f air , w h i c h m a y also cause dyspnea. The O2 concentrat ion i n ambient air is 2 0 . 9 % , and humans report dyspnea at O2 levels o f less than 8 % w h e n compensatory breathing is constrained ( M o o s a v i et a l . , 2003) . Th i s sensation was al lev iated w i t h spontaneous breathing, but O2 concentrations less than 7 % were not examined . D u r i n g pre- f i l l CO2 euthanasia rats are exposed to CO2 concentrations above 7 0 % , so it seems l i k e l y that rats experience both pa in and dyspnea us ing this method. H o w e v e r , dur ing g radua l - f i l l CO2 euthanasia rats t yp ica l l y lose consciousness at CO2 concentrations b e l o w 4 0 % (Smi th & Harrap , 1997) and so m a y avo id some o f these negative sensations. I f CO2 does cause pa in or dyspnea i n rats, 'd istress' behaviours w o u l d be expected dur ing exposure. These c o u l d inc lude behaviours associated w i t h pa in such as head -shak ing and rubbing the nose and eyes, behaviours associated w i t h gas avoidance such as increased exp lorat ion and escape attempts, and general distress behaviours such as increases i n part icular voca l i zat ions . B e h a v i o u r a l studies on rats dur ing CO2 euthanasia have general ly examined responses to different methods o f CO2 de l i very , wi thout compar i son to a contro l session. In these studies, some authors have reported a lack o f distress behaviours dur ing p r e - f i l l (Smi th and Har rap , 1997) and g radua l - f i l l C 0 2 exposure (Hackbar th et a l . , 2 0 0 0 ; Hornett & Haynes , 1984; S m i t h & Har rap , 1997). Other studies have reported 'ag i tat ion and asphyx iat ion ' dur ing both p r e - f i l l and g radua l - f i l l exposure (Coenen et a l . , 1995), and ' m i l d to moderate stress' dur ing p r e - f i l l exposure ( Iwarsson & Rehbinder , 1993). H o w e v e r , these authors prov ide f e w detai led behav ioura l descr ipt ions or data to support their conc lus ions . Three studies have taken object ive behav ioura l measures dur ing both CO2 and air exposure. B l a c k s h a w et a l . (1988) found that p r e - f i l l CO2 exposure caused a decrease i n act iv i ty . L e a c h et a l . (2002) examined behavioural responses and avers ion o f rats to static CO2 concentrations greater than 2 5 . 5 % , and found that rats avo ided CO2 exposure, and that it caused increased face -wash ing . Br i t t (1987) 42 examined responses to g radua l - f i l l exposure and found that changes i n act iv i ty and w a l l - c l i m b i n g var ied w i t h strain, but that shaking a lways increased. I f dyspnea does occur dur ing CO2 euthanasia, some o f this effect m a y be due to reduced O2 levels . N o studies to date have measured the responses d f rats to O2 reduct ion at the levels that occur dur ing g radua l - f i l l euthanasia. H o w e v e r , studies us ing subjective assessments o f distress have c l a i m e d that O2 supplementat ion decreases responses o f rats dur ing p r e - f i l l ( Iwarsson & Rehbinder , 1993) and g radua l - f i l l (Coenen et a l . , 1995) CO2 exposure. The a ims o f the current study were to determine whether rats show behav ioura l signs o f distress dur ing g radua l - f i l l CO2 euthanasia and dur ing an equivalent reduct ion i n O2 level caused by d isp lac ing air w i t h argon. W e predicted that distress dur ing exposure w o u l d be accompanied by increased explorat ion , escape attempts and voca l i za t ion , and that pa in at the m u c o s a l membranes and cornea w o u l d be accompanied by head shaking and face wash ing . 2.2 Materials and Methods 2.2.1 Subjects Sixteen 4 0 0 - 5 0 0 g, mature, male Sprague D a w l e y rats were obtained as surplus stock (i.e. an imals already slated for euthanasia) f r o m the U B C Rodent B r e e d i n g Un i t . A n i m a l s were group-housed at 21°C under a 12 :12-hr l ight -dark cyc le , and g iven ad l i b i t u m access to food (Lab D ie t 5 0 0 1 , P M I N u t r i t i o n International, Indiana, U S A ) and tap water. A l l testing was conducted dur ing the l ight port ion o f the l ight -dark cyc le . 2.2.2 Experimental Apparatus The euthanasia chamber was a 20 L po lypropy lene cage 20.5 c m h i g h , 45 .5 c m long and 24 c m w i d e at the top (Lab Products Inc.), f i tted w i t h a P lex ig las l i d . The l i d had a gas inlet 43 centered at one end, two air outlets pos i t ioned at the opposite end, and a gas sampl ing tube inserted at the center o f the chamber to a depth o f h a l f the chamber height. The air outlets were covered w i t h m e s h to prevent the rats f r o m pushing their noses outside the chamber. The back and sides o f the chamber were covered w i t h b lack paper so that the an imals cou ld not see the person conduct ing the experiment. A r g o n , an inert gas, was used to displace air i n the C^ - reduct ion treatment group. C a r b o n d iox ide and argon were del ivered to the chamber f r o m compressed gas cy l inders (Praxair , R i c h m o n d , B . C . ) , w h i l e r o o m air was del ivered v i a an air compressor situated i n an adjo in ing r o o m . The treatment gases were passed through a copper c o i l i n a r o o m temperature water bath to regulate the temperature o f the gas before it entered the chamber. P re l iminary tests indicated that the chamber temperature d i d not drop dur ing the f i l l i n g process. F l o w rates o f the gases were measured by a var iable area f lowmeter ( M o d e l V S B - 6 6 - B V , D w y e r Instruments, Inc., M i c h i g a n ) , and measured f l o w rates for CO2 and argon were adjusted for density by the correct ion factors 0 .812 and 0.852 respect ively . Gas concentrations i n the chamber were moni to red dur ing the exper iment v i a a gas sampl ing tube us ing a M o c o n L F 7 0 0 D O2 analyzer. It was assumed that any decrease i n O2 was direct ly related to a decrease i n air and a cor responding increase i n the treatment gas. Therefore the f o l l o w i n g f o r m u l a was used to calculate the concentrat ion o f CO2 at speci f ic t ime points (t = x) dur ing the f i l l i n g process: [ C 0 2 ( , = *>] = 1 0 0 - ( 1 0 0 * ( [0 2 ( t = X ) ] / [ 0 2 ( , = „)])). 2.2 .3 A p p a r a t u s T e s t i n g Befo re starting the a n i m a l exper iments, gas concentrat ions were measured i n different areas o f the empty chamber dur ing the CO2 filling process. The chamber was d i v ided into twelve sectors by part i t ioning the chamber into three segments i n the x -p lane (length), two segments i n the y -p lane (width) and two segments i n the z -p lane (height). The gas sampl ing tube 44 was p laced i n the center o f the sector to be tested, and CO2 was added at a rate o f 3.5 L/min. The O2 concentrat ion was recorded every 5 s for 10 m i n . E a c h sector was tested three t imes. The sectors were tested i n random order to account for m i n o r f luctuations i n env i ronmental parameters. 2 .2.4 E x p e r i m e n t a l P r o c e d u r e A n i m a l s were randomly al located to the CO2 or O2 reduct ion (using argon) treatment groups (n = 8 for both). F o r both groups, an imals were first tested w i t h air exposure and then w i t h the treatment gas on the f o l l o w i n g day. O n both testing days, each a n i m a l was ind i v idua l l y p laced into the euthanasia-chamber for a 1 5 - m i n per iod o f acc l imat i za t ion dur ing w h i c h air was added to the chamber at a rate o f 3.5 L/min. The length o f the acc l imat i za t ion per iod was based o n p re l iminary observations s h o w i n g h o w l o n g it took an imals to cease exp lorat ion and become inact ive after entry into the testing apparatus. A f t e r acc l imat i za t ion , air f l o w ceased and either air , CO2 or argon f l o w was started at a rate o f 3.5 L/min. Th i s rate corresponded to 1 7 . 2 5 % o f the chamber v o l u m e be ing added per minute. A l t h o u g h air f l o w ceased dur ing treatment gas exposure, the air compressor remained on throughout the exper iment i n order to control for noise effects. C02 - t reated animals remained i n the chamber and were moni tored unt i l death but argon-treated an imals were removed f r o m the chamber at the end o f the 105-s observat ion per iod . P re l iminary observations showed that C02 - t reated an imals ceased a l l purposeful movement w i t h i n this per iod , so any relevant effects o f O2 depr ivat ion w o u l d be present dur ing this t ime. The leve l o f O2 reduct ion result ing f r o m argon addi t ion was not suff ic ient to cause unconsciousness or death, and was used on ly to s imulate reduct ions i n O2 levels that occur w h e n a chamber is filled w i t h CO2. 45 2.2 .5 B e h a v i o u r a l A n a l y s i s The euthanasia chamber and O2 meter readout were v ideo recorded dur ing the exper imental procedure. E a c h a n i m a l was scored cont inuously dur ing the last 105 s o f the acc l imat i zat ion per iod (baseline) and the first 105 s after gas flow began (exposure) for pre - def ined behaviours thought to relate to pa in and distress (Table 2.1). The t ime unt i l complete recumbency and cessation o f breathing was also recorded. R e c u m b e n c y was def ined as a loss o f posture and musc le tone. The t ime unt i l onset o f ataxia was not recorded as it c o u l d not be accurately assessed i n a l l an imals . Sound data were co l lected w i t h a ^ " c o n d e n s e r mic rophone (Brue l and K jaer , Type 4135) , connected to a preampl i f ie r (Brue l and K jaer , T y p e 2619) and a measur ing ampl i f ie r (B rue l and K jaer , T y p e 2636) . The s ignal was recorded direct ly to a h igh -capac i ty hard d isk at a rate o f 2 5 0 k H z us ing a 330 k H z P C I - D A S 1 2 0 0 / J R data acquis i t ion card (Computerboards, Inc.) and C B D i s k 1.4 software (Engineer ing D e s i g n , B e l m o n t , M A ) . The microphone end was suspended 0.5 c m into the euthanasia chamber through one o f the air outlets. Sound was recorded dur ing the last 105 s o f the acc l imat i za t ion per iod (baseline) and the first 105 s after gas flow began (exposure). Sound analysis was w i t h S I G N A L 4.0 (Engineer ing D e s i g n , B e l m o n t , M A ) . C a l l s were ident i f ied by their f o r m and as be ing dist inct f r o m ambient noise. Sounds o f less than 5 m s durat ion were d i f f i cu l t to d is t inguish f r o m ambient noise and were discarded f r o m the analysis . Suspected vocal i zat ions were p layed back i n a frequency range audible to humans by s l o w i n g the recordings by a factor o f 0.05 to 0 . 1 . W h i s t l e - l i k e sounds were accepted as vocal i zat ions w h i l e c l i c k s and other mechan ica l sounds were d iscarded. Rats produce w h i s t l e - l i k e U S V s by push ing air through a 1 to 2 m m hole fo rmed v i a the tight oppos i t ion o f the two v o c a l cords (Sanders et a l . , 2001) . It is poss ib le that rats use another p roduct ion m e c h a n i s m to produce c l i c k s , but i n this study it was not poss ib le to adequately d is t inguish v o c a l c l i c k s f r o m those result ing f r o m movement i n the cage. 46 V o c a l i z a t i o n s were subjected to spectrographic analysis to determine ca l l durat ion, peak frequency, m a x i m u m frequency and m i n i m u m frequency. 2.2.6 Statistical Analysis The behavioural data were n o n - n o r m a l w i t h unequal var iances and cou ld not be corrected through the use o f transformations, so non-parametr ic statistics were used for analysis . H e a d - s h a k i n g was not observed i n any animals and face -wash ing was observed on ly i n two rats dur ing basel ine and one rat dur ing CO2 exposure, so these behaviours were not inc luded i n the analys is . CO2 and reduced 0 2 exposures were not conducted concurrent ly , therefore direct .comparisons between the treatments were not per formed. The W i l c o x o n S igned R a n k s Test was used to compare the change i n behaviour f r o m basel ine dur ing air exposure w i t h the change f r o m basel ine dur ing either CO2 or reduced O2 exposure. A l l behavioural data are presented as medians w i t h 2 5 % and 7 5 % interquarti le ranges. V o c a l i z a t i o n parameters and t ime to recumbency and death are presented as means ± standard deviat ions. 2.3 Results CO2 concentrations i n the empty chamber rose asymptot ica l l y dur ing the f i l l i n g process, reaching 8 7 % after 600 s (F ig . 2.1). Concentrat ions o f CO2 tended to be greater at the bot tom o f the chamber than the top, w i t h a peak dif ferent ial o f 7%> (bottom at 9 . 4 % , top at 2%) after 20 s. A f t e r this t ime , the concentrations began to converge. There were no concentrat ion differences i n the y -p lane dur ing f i l l i n g , and the x -p lane CO2 concentrat ion di f ferent ial between the posi t ions closest and furthest f r o m the gas inlet was less than 2 % throughout the filling process. D u r i n g the actual exper iment gas samples were taken at a depth o f h a l f the chamber height, and 47 CO2 and argon concentrations consistently fe l l w i t h i n the range o f values found dur ing p re l iminary testing. D u r i n g the basel ine per iod rats were general ly inact ive, but a l l behavioural measures increased after CO2 f l o w began (F ig . 2.2). R e a r i n g and touch ing o f the nose to the l i d started to increase w i t h i n the first 15 s. A c t i v i t y and rear ing peaked between 15 s and 60 s, and touching o f the nose to the l i d , escape behaviours and vocal i zat ions peaked between 60 s and 90 s after CO2 began. O2 and CO2 concentrations reached approx imate ly 2 0 % and 5 % after 15 s, 1 7 % and 20%) after 60 s, and 15%> and 28%) after 90 s. Comple te recumbency occurred o n average 106 ± 12 s after f l o w in i t ia t ion , at O2 and CO2 concentrations o f approx imate ly 14%> and 3 3 % . Rats d i d not stop breathing unt i l 443 ± 14 s into the procedure, at 0 2 and CO2 concentrations o f approx imate ly 4 % and 8 0 % . W h e n compared to air exposure, rats exposed to CO2 were more act ive, and showed s igni f icant increases i n the frequency o f rear ing, escape behaviours , voca l i zat ions and i n the t ime spent w i t h the nose contact ing the chamber l i d (Table 2.2). H o w e v e r , there was considerable var iat ion i n behavioural response a m o n g an imals , as ref lected by the large interquarti le ranges for several behaviours. F o r example , dur ing the CO2 exposure per iod one rat per formed 34 escape behaviours , w h i l e two others per formed none. V o c a l i z a t i o n s dur ing CO2 exposure consisted o f pure tones and cal ls w i t h frequency modu la t ion , and these d i d not fa l l into obv ious categories. C a l l s var ied i n length and frequency, ranging f r o m 5 to 150 m s and 8.6 to 102.1 k H z . O n average the ca l l durat ion was 33 ± 28 m s , the average f requency range was 22 ± 19 k H z and the peak f requency was 44 ± 20 k H z . D u e to the s m a l l number o f voca l i zat ions produced dur ing air and argon exposure, ca l l parameters were not compared between treatments. Rats that were exposed to reduced O2 concentrations us ing argon exhib i ted on ly a smal l increase i n the t ime spent w i t h the nose contact ing the chamber l i d and no increases i n any other 48 var iables i n compar i son w i t h air exposure (Table 2.3). Rats d i d not show any signs o f ataxia or recumbency dur ing the 105-s observat ional per iod for this treatment. 2.4 Discussion In contrast to several prev ious studies o n g radua l - f i l l CO2 euthanasia (Hackbarth et a l . , 2 0 0 0 ; Hornett & Haynes , 1984; S m i t h & Har rap , 1997), w e found that this procedure does cause behavioural signs o f distress i n rats. N o t on ly d i d the rats exhib i t general signs o f explorat ion such as increased l o c o m o t i o n , rear ing, and touch ing the nose to the l i d , they also showed escape behaviours and vocal i zat ions . Th i s behavioural response began w i t h i n the first 15 s after the start o f gas f l o w , demonstrat ing that rats respond to even l o w (approx. 5%) concentrations o f CO2. W e d i d not observe increases i n head-shaking or face -wash ing dur ing exposure, suggesting that animals d i d not experience pa in dur ing the t ime w h e n they were capable o f mount ing a behav ioura l response. H o w e v e r , it is also possible that the measured behaviours were not appropriate indicators o f upper respiratory pa in i n rats. O u r results are consistent w i t h the subjective assessments o f 'ag i tat ion ' dur ing gradual - f i l l reported by C o e n e n et a l . (1995) , and w i t h the increase i n act iv i ty reported by Br i t t (1987) for Sprague D a w l e y rats. Interestingly, Br i t t reported a decrease i n act iv i ty for L is ter H o o d e d rats i n the same study, ind icat ing that strain m a y be an important factor i n response differences. Other studies have reported f e w behavioural changes i n Wis ta r ( B l a c k s h a w et a l . , 1988; Hornett & Haynes , 1984) and F - 3 4 4 rats (Hackbarth et a l . , 2000) dur ing CO2 exposure. In our study, w e found considerable var iat ion i n response among ind iv idua ls , w i t h some animals d isp lay ing numerous escape attempts and others showing little response dur ing the procedure. It is unclear whether this var iat ion indicates a dif ference i n the leve l o f distress result ing f r o m the procedure, 49 or a di f ference i n h o w animals respond to distress. A lack o f behav ioura l response does not necessari ly indicate that the rats perceive the procedure as innocuous. W h i l e L e a c h et a l . (2002) have demonstrated that rats w i l l avo id static CO2 concentrations o f 2 5 . 5 % and greater, this is the first study to show increased escape attempts by rats dur ing CO2 euthanasia. The design o f prev ious euthanasia experiments m a y have discouraged expression o f escape behaviours. The rats i n our study had access to the chamber l i d and t ime to explore it thoroughly before CO2 exposure. Rats i n other studies m a y not have had access to the chamber l i d and i n most other experiments rats were unfami l ia r w i t h the chamber at the t ime o f exposure, w h i c h m a y have inhib i ted escape attempts. O n l y one other study has attempted to measure voca l i zat ions dur ing CO2 exposure, and no cal ls were detected (Britt , 1987). The author d i d not prov ide details on the sound co l lec t ion apparatus, so the sensit iv i ty and frequency range o f the equipment is u n k n o w n . W e found that voca l i zat ions were present at l o w levels dur ing basel ine and argon exposure, but increased dur ing CO2 exposure. The major i ty o f studies o n rat U S V s have focused on what have been descr ibed as the 22 k H z (ranging f r o m approx imately 2 0 - 3 0 k H z and 300 - 3000 ms) and 50 k H z (ranging f r o m approx imate ly 3 0 - 7 0 k H z and < 80 ms) cal ls ( rev iewed i n K n u t s o n et a l . , 2002) . C a l l s i n the current study appear to be consistent w i t h those that have been descr ibed as 50 k H z ca l ls , and no 22 k H z cal ls were observed. C a l l s that have been grouped under the 50 k H z label i n prev ious studies have actual ly var ied considerably i n length, f requency and shape, and f e w studies have p rov ided detai led ca l l descr ipt ions. Th i s makes it d i f f icu l t to determine whether these cal ls are consistent across studies, and whether they are indicat ive o f s imi la r states. A l t h o u g h the 50 k H z cal ls have been observed dur ing pos i t ive , contexts such as ant ic ipat ion o f reward (Burgdor f et a l . , 2 0 0 0 ; K n u t s o n et a l . , 1998; K n u t s o n et a l . , 1999), they have also been observed dur ing intermale aggression (Sales, 1972; Thomas et a l . , 1983) and exposure to anesthetized conspeci f ics (B lanchard et a l . , 1993). These occurrences dur ing 50 potentially negative contexts suggest that the calls may also be associated with distress. However, the CC^-exposed animals were more active and spent more time near the lid in the vicinity of the microphone, and this may have improved our ability to detect calls during this condition. The increase in calls could also have been a by-product of increases in breathing frequency and depth that occur during hypercapnia, but the effects of breathing changes on U S V production has not previously been examined. The densities of the gases used in the current study were higher than air and may also have affected U S V characteristics. Roberts (1975) found that a reduction in gas density results in an increase in fundamental frequency and a decrease in amplitude of USVs , and it is therefore likely that an increase in gas density would have the opposite effect. However, this effect is due to changes in the speed of sound in gases with different densities, and the effect of helium is much greater than the effects of either CO2 or argOn. The speed of sound in air, CO2 and argon at 2 1 ° C is approximately 343, 269 and 320 m/s respectively, whereas the speed of sound in helium is approximately 1000 m/s. Furthermore, the gases in the current experiment were mixed with air in relatively low concentrations at the time when rats were responding. While this may have had a minor effect on U S V characteristics, it is unlikely that density alone was responsible for the increased number of calls detected during CO2 exposure. Britt's (1987) conclusion that rats experience distress during CO2 exposure was based partly on the occurrence of shaking, but this behaviour was not seen in our study. Britt (1987) did not state the flow rates used, but higher flow rates tend to result in a higher concentration of CO2 in the chamber at the time of loss of consciousness (Ambrose et al., 2000) and increase the likelihood that rats experience pain. Previous studies have also found increases in urination and defection during exposure to C 0 2 (Britt, 1987; Smith & Harrap, 1997). We did not record urination and defecation as they often occurred when the animal was initially placed in the 51 chamber , and the l i k e l i h o o d o f future events is strongly related to the t ime since these last occurred! The potential for animals to experience distress dur ing euthanasia is l im i ted to the per iod o f consciousness, and i n this exper iment w e have assumed that complete loss o f consciousness occurred w h e n the an imals became f u l l y recumbent. R e c u m b e n c y i n rats dur ing CO2 exposure is associated w i t h a loss o f pedal and corneal ref lexes (Hornett and H a y n e s , 1984) and w i t h a drop i n heart rate and onset o f an aberrant E E G (Coenen et a l . , 1995). H o w e v e r , some researchers have found a delay between col lapse and loss o f ref lexes (Danneman et a l . , 1997; Hewett et a l . , 1993), so it is poss ib le that the depth o f unconsciousness varies at the t ime o f recumbency . W e found that dur ing g radua l - f i l l CO2 exposure, rats became recumbent after an average o f 106 s, and this f i n d i n g is consistent w i t h other studies us ing s imi la r f l o w rates (Hewett et a l . , 1993; Hornett and Haynes , 1984; D a n n e m a n et a l . , 1997). A n assessment o f ataxia w o u l d have prov ided an ind icat ion o f the t ime that loss o f consciousness began to occur , but w e were unable to accurately detect ataxia i n the current study because a loss o f musc le coord inat ion can on ly be detected w h e n part icular postures and movements are in i t ia l l y present. F o r example , w h e n a rat is c rouched and stationary a reduct ion i n musc le coord inat ion m a y not be obv ious . In the current exper iment, some rats exh ib i ted recumbency wi thout obv ious atax ia beforehand. T w o possib le reasons for the rats' behavioural response to CO2 are pa in f r o m carbonic ac id fo rmat ion at the m u c o s a and cornea, and dyspnea f r o m hypercapnia and hypox ia . W e found that the O2 and CO2 concentrations i n the chamber at the t ime o f recumbency were approx imate ly 1 4 % and 3 3 % . H o w e v e r , the O2 and CO2 concentrations were approx imately 20%o and 5 % w h e n an imals started to respond, and 1 5 % and 2 8 % w h e n a l l behaviours had peaked and were dec l in ing . P h y s i o l o g i c a l data suggest that the CO2 threshold for the major i ty o f nociceptors i n rat nasal m u c o s a is between 37 and 5 0 % (Peppel & A n t o n , 1993). H u m a n self - report data indicate that CO2 is pa in fu l i f appl ied to the nasal m u c o s a at concentrations above 5 2 , • 4 7 % , a l though this va lue ranges f r o m 32.5% to 5 5 % depending o n the ind i v idua l ( A n t o n et a l . , 1992). F o r the cornea, humans report st inging at 33% and overt pa in at 4 7 % CO2 ( C hen et a l , 1995). Receptors found i n the larynx , trachea, and bronchi are also sensit ive to inhaled irritants ( rev iewed i n W i d d i c o m b e , 2001) , but responses to CO2 have not been fu l l y investigated for these receptors. H o w e v e r , D a n n e m a n et a l . (1997) had h u m a n subjects inhale CO2 into the lower a i rways and found that on ly 7 o f the 40 subjects reported that 5 0 % CO2 was overt ly p a i n f u l , suggesting that pa in i n the lower a i rways is not occur r ing at a marked ly lower CO2 concentrat ion than has been demonstrated for the nasal mucosa . A l t h o u g h it is poss ib le that rats i n this study exper ienced some pa in before los ing consciousness, the CO2 concentrat ion dur ing the per iod o f m a x i m a l response was m u c h lower than the probable p a i n threshold, and the rats d i d not exhib i t increases i n head-shaking and face -wash ing . H e a d - s h a k i n g and face -wash ing have been observed i n rats dur ing exposure to moderate to h igh CO2 concentrations (Britt , 1987; L e a c h et a l . , 2002) , and face -wash ing has also been observed i n rats dur ing exposure to i r r i tat ing compounds such as ch lo ro fo rm ( B l a c k s a w et a l . , 1988). Hence , pa in is u n l i k e l y to be the cause o f the rats' responses i n the current study. In contrast, the concentrations o f CO2 i n the chamber dur ing the per iod o f m a x i m a l response have been found to cause sensations o f dyspnea i n spontaneously breathing humans (Dr ipps & C o m r o e , 1947; L i o t t i et a l . , 2001) , and m a y cause s imi la r effect i n rodents. In humans , hypercapnia is also associated w i t h other negative phys i ca l symptoms such as headache, f lush , restlessness, heart pound ing , drowsiness and d izz iness ( M o o s a v i et a l . , 2003) . It therefore seems l i k e l y that the distress response to g radua l - f i l l CO2 exposure is due to dyspnea and other symptoms o f hypercapnia rather than pa in . Furthermore, D r i p p s and C o m r o e (1947) found that humans vary i n their response to hypercapnia , a f i n d i n g consistent w i t h the var iabi l i ty i n the rats' responses to CO2 exposure in the current study. Prev ious studies have assessed breathing i n rats dur ing CO2 exposure and found that it causes changes descr ibed as " laboured 53 breath ing" ( Iwarsson & Rehbinder , 1993), "gasp ing and a s p h y x i a " (Coenen et a l . , 1995) and "gasp ing or laboured breath ing" (Smi th & Harrap , 1997) pr ior to loss o f consciousness, and these changes might be indicat ive o f the sensation o f dyspnea. In the current study we d i d not assess changes i n breathing because we cou ld not make an accurate assessment f r o m v ideo . Ideal ly breathing c o u l d be assessed object ive ly by measur ing its f requency and depth, but this was beyond the scope o f the current study. H o w e v e r , h u m a n m e d i c a l studies suggest that laboured breathing is poor l y correlated w i t h the sensation o f dyspnea ( L u s h et a l . , 1988), so it is not clear that breathing measures are useful as an indicator o f dyspnea dur ing CO2 exposure. O u r results indicate that O2 reduct ion alone causes on ly m i n i m a l distress i n rats over the range o f O2 concentrations that we examined . A l t h o u g h rats exhib i ted a sl ight increase i n touch ing the nose to the l i d dur ing argon treatment, the increase i n this behaviour was m u c h less than that seen i n C02 - exposed an imals and there were no other behav ioura l changes. Th i s result is consistent w i t h h u m a n self - report data suggesting that O2 levels i n the range seen dur ing this exper iment do not cause dyspnea dur ing spontaneous breathing ( M o o s a v i et a l . , 2003) . H o w e v e r , hypercapnia and h y p o x i a are k n o w n to have synergist ic effects on venti latory responses ( N i e l s o n & S m i t h , 1952), such that the response to increased C 0 2 m a y be potentiated by a leve l o f O2 reduct ion that has no effect o n its o w n . M a s u d a et a l . (2001) found that increasing levels o f h y p o x i a augment the effect o f hypercapnia o n dyspnea scores i n humans. A s s u m i n g that these results are appl icable to rats, it w o u l d appear that h y p o x i a dur ing CO2 euthanasia m a y increase dyspnea, and this is supported by results s h o w i n g that O2 supplementat ion reduces the adverse effects o f CO2 exposure on rats (Coenen et a l . , 1995). It is important to note that these results have no bear ing o n distress associated w i t h argon w h e n used for euthanasia. Rats can surv ive for greater than 20 minutes w h e n exposed to O2 concentrations o f 4 . 9 % i n argon ( A l t l a n d et a l . , 1968), and our O2 concentrations were on ly reduced to 1 4 % . F o r t ime ly unconsciousness and death o f pigs and poultry w i t h argon, concentrations o f 9 0 % or 54 greater have been used to reduce 0 2 levels below 2% (e.g., Raj, 1999; Raj et al., 1998), and gradual fill exposure has not been investigated. Our study indicates that gradual-fill C 0 2 euthanasia causes distress in rats, and the concentrations involved suggest that this distress is due to dyspnea rather than pain. The lack of consistency between experiments suggests that further research is necessary to examine whether there are strain differences in sensitivity or responsiveness to C 0 2 . Further research is also necessary to determine the extent of the distress caused by C 0 2 exposure, and to determine whether other gas euthanasia agents cause less distress. The variability in behavioural responses to C 0 2 suggests that tests of motivation might be a better approach. For example, approach- avoidance testing could be used to examine whether rats will forgo an attractive reward of known value to avoid exposure to C 0 2 and other gas euthanasia agents. 55 Table 2.1. Descriptions of rat behaviours recorded during baseline and during exposure to CO2 or reduced O2 concentrations. Behaviour Description Activity Movement that results in the back feet crossing a line that divides the length of the chamber in half (event). Rear Raising of the upper body while standing on the two back feet. Includes wall climbing. Climbing on the air sampling tube while chewing it and rearing during grooming were excluded (event). Nose to lid Time spent with the nose in contact with the chamber lid (state). Escape behaviours: Scratch at lid A rapid movement of the front paw from the lid through at least a 90° downward angle (event). Push at lid A push at the chamber lid using the nose or front paw evidenced by body and lid movement (event). Head shake Rapid rotation of the head about the axis (event). Face washing Placement of one or both paws to the nose. Performance during grooming was excluded (event). 56 Table 2.2. Difference from baseline for each of the five behavioural responses of rats during air and C O 2 exposure (n = 8 rats). Data are presented as medians with 25 t h and 75 t h percentiles, and statistical comparisons were made with Wilcoxon Signed Ranks Test (T; based on N values >0). Air C 0 2 Behaviour Med (25 m , 75 t h) Med (25 t h , 75 t h) " T ( N ) P-value Activity (no.) 0.0 (-0.5, 0.5) 3.5 (2.0, 4.5) 0(8) <0.005 Rears (no.) 0.0 (-2.5, 1.0) 10.5 (8.0,13.0) 0(8) <0.005 Nose to lid (s) 0.0 (-15,6.5) 23.0 (9.0,25.5). 3(7) <0.05 Escape behaviours (no.) 0.0 (0.0, 0.0) 4.0(0.5,10.5) 0(7) <0.05 Vocalizations (no.) -0.5 (-1.0, 0.0) 6.0 (-0.5,13.0) 1.5 (7) <0.05 57 Table 2.3. D i f fe rence f r o m basel ine for each o f the f ive behav ioura l responses o f rats dur ing air and exposure to reduced O2 concentrations (n = 8 rats). D a t a are presented as medians w i t h 25th and 75th percenti les, and statistical compar isons were made w i t h the W i l c o x o n S igned R a n k s Test (T; based o n N values >0). A i r R e d u c e d O2 B e h a v i o u r Med ( 2 5 t h , 7 5 t h ) Med ( 2 5 t h , 7 5 t h ) T ^ / > V a , U e A c t i v i t y (no.) 0.0 ( - 0 . 5 , 0 . 5 ) 0 .0(0 .0 ,1 .5) 5(8) NS Rears (no.) -0.5 ( - 2 . 5 , 0 . 0 ) 3.0(0.0,6.0) 4 (7 ) NS N o s e to l i d (s) 0.0 ( - 3 . 5 , 0 . 0 ) 2.5(1.0,8.0) 1.5(8) <0.05 Escape behaviours (no.) 0 . 0 ( 0 . 0 , 0 . 0 ) 0 .0 (0 .0 ,0 .0 ) V o c a l i z a t i o n s (no.) 0.0 ( - 0 . 5 , 0 . 0 ) 0 .0 (0 .0 ,0 .5 ) 5 (7 ) NS N S signif ies P > 0.05 58 c o ro i_ "c CD O c o o V) CD o 100 90 - 80 - 70 60 50 40 30 20 10 0 -A- 5cm O2 - • - 15cm 0 2 - • - 5cm C0 2 -•- 15cm C0 2 60 120 180 240 300 360 Time (s) 420 480 Figure 2.1. Average concentrations of 0 2 (open markers) and CO2 (filled markers) in the chamber during the first 600 s of the filling process. Concentrations were taken 5 cm (triangles) and 15 cm (squares) from the chamber bottom. j 59 a) Activity b) Rears Time (s) Time (s) c) Nose to lid d) Escape behaviours to o <o o <o o o) AS. cp y co <o o <o o <o o <o <- co V C O ^ 0 5 o i? £> in P "o o <o o o> f x C O V CO * -I / I I I I <o o <o o <o O <p » - CO V CO K O ) o Time (s) Time (s) e) Vocalizations 1.2 -I 1 - "O 0 *k_ 0.8 - cu Q . 0.6 - ^ » O 0.4 - z 0.2 - 0 - CO to o <o o Co Time (s) Figure 2.2. Responses by rats during the baseline period and then during exposure to either air (filled squares) or CO2 (open squares) starting at t = 0. Median values per 15 s period are shown for a) activity, b) rears , c) nose to lid, d) escape behaviours, and e) vocalizations (n = 8 rats). 60 2.5 References A l t l a n d , P . D . , B r u b a c h , H .F . , Parker , M . G . 1968. Ef fects o f inert gases o n tolerance o f rats to h y p o x i a . J . A p p l . P h y s i o l . 2 4 , 7 7 8 - 7 8 1 . A m e r i c a n Veter inary M e d i c a l A s s o c i a t i o n , 2 0 0 1 . 2 0 0 0 Report o f the A V M A Pane l on Euthanasia . J . A m . Ve t , M e d . A s s o c . 2 1 8 , 6 6 9 - 6 9 6 . A m b r o s e , N . , W a d h a m , J . , M o r t o n , D. , 2000 . Ref inement i n Euthanasia . In: B a l l s , M . , van Ze l le r , A . M . , Ha ider , M . E . (Eds) , Progress i n the Reduc t ion , Ref inement and Replacement o f A n i m a l Exper imentat ion , E l sev ie r Sc ience , A m s t e r d a m , p p . 1 1 5 9 - 1 1 6 9 . A n t o n , F., Euchner , I., Handwerker , H .O . , 1992. Psychophys i ca l examinat ion o f pa in induced by def ined CO2 pulses appl ied to the nasal mucosa . P a i n 4 9 , 5 3 - 6 0 . A n t o n , F., P e p p e l , P. , Euchner , I., Handwerker , H .O . , 1991. Cont ro l l ed nox ious chemica l s t imulat ion : responses o f rat t r igeminal brainstem neurones to CO2 pulses appl ied to the nasal mucosa . N e u r o s c i . Lett . 123, 2 0 8 - 2 1 1 . Aus t ra l ian and N e w Zea land C o u n c i l for the Care o f A n i m a l s i n Research and Teach ing , 1993. Euthanas ia o f A n i m a l s U s e d for Sc ient i f i c Purposes G l e n O s m o n d , A N Z C C A R T . Banzett , R . B . , L a n s i n g , R . W . , E v a n s , K . C . , Shea, S . A . 1996. St imulus - response characteristics o f C 0 2 - i n d u c e d air hunger i n normal subjects. Resp . P h y s i o l . 103, 1 9 - 3 1 . B l a c k s h a w , J . K . , F e n w i c k , D . C . , Beatt ie , A . W . , A l l a n , D . J . , 1988. The behaviour o f ch ickens , m i c e and rats dur ing euthanasia w i t h ch lo ro fo rm, carbon d iox ide and ether. L a b . A n i m . 2 2 , 6 7 - 7 5 . ' B l a n c h a r d , R . J . , Y u d k o , E . B . , B l a n c h a r d , D . C . , T a u k u l i s , H . K . , 1993. H igh - f requency (35 -70 k H z ) ultrasonic vocal i zat ions i n rats confronted w i t h anesthetized conspec i f ics : effects o f gepirone, ethanol , and d iazepam. P h a r m . B i o c h e m . Behav . 4 4 , 3 1 3 - 3 1 9 . Br i t t , D . P. , 1987. The humaneness o f carbon d iox ide as an agent o f euthanasia for laboratory 61 rodents. In: Euthanasia o f U n w a n t e d , Injured or D iseased A n i m a l s or for Educat ional or Sc ient i f i c Purposes, pp . 1 9 - 3 1 . Potter 's B a r : Un ivers i t ies Federat ion for A n i m a l Wel fa re . Burgdor f , J . , K n u t s o n , B . , Panksepp, J . 2000 . A n t i c i p a t i o n o f reward ing electr ical bra in s t imulat ion evokes ultrasonic vocal i zat ions i n rats. Behav . N e u r o s c i . 114, 3 2 0 - 3 2 7 . Canad ian C o u n c i l on A n i m a l Care , 1993. G u i d e to the Care and U s e o f Exper imenta l A n i m a l s , V o l u m e 1, 2 n d E d i t i o n , Ol fert , E . D . , C ross , B . M . , M c W i l l i a m , A . A (Eds) . Ot tawa, C C A C . C h e n , X . , Ga l la r , J . , P o z o , M . A . , B a e z a , M . , B e l m o n t e , C , 1995. C 0 2 s t imulat ion o f the cornea: a compar i son between h u m a n sensation and nerve act iv i ty i n p o l y m o d a l nocicept ive afferents o f the cat. Eur . J . N e u r o s c i . 7, 1 1 5 4 - 1 1 6 3 . C l o s e , B . , Banister , K . , B a u m a n s , V . , Bernoth , E., B r o m a g e , N . , B u n y a n , J . , Erhardt , W . , F l e c k n e l l , P. , G regory , N . , Hackbar th , H . , M o r t o n , D. , W a r w i c k , C , 1997. European C o m m i s s i o n W o r k i n g Party Report : Recommendat ions for euthanasia o f exper imental an imals , Part I. L a b . A n i m . 30 , 2 9 3 - 3 1 6 . C o e n e n , A . M . , D r inkenburg , W . H . , Hoenderken , R., van Lui j te laar , G . L . , 1995. C a r b o n d iox ide euthanasia i n rats: oxygen supplementat ion m i n i m i z e s signs o f agitat ion and asphyx ia . L a b . A n i m . 2 9 , 2 6 2 - 2 6 8 . D a n n e m a n , P . J . , Ste in , S . , W a l s h a w , S . O . , 1997. H u m a n e and pract ical impl icat ions o f us ing carbon d iox ide m i x e d w i t h oxygen for anesthesia or euthanasia o f rats. L a b . A n i m . S c i . 4 7 , 3 7 6 - 8 5 . D r i p p s , R . D . , C o m r o e , J . H . , 1947. The respiratory and c i rculatory response o f normal m a n to inhalat ion o f 7.6 and 10.4 per cent CO2 w i t h a compar i son o f the m a x i m a l vent i lat ion produced by severe muscular exercise, inhalat ion o f CO2 and m a x i m a l voluntary hypervent i lat ion. A m . J . P h y s i o l . 149, 4 3 - 5 1 . F e n g , Y . , S i m p s o n , T. L. , 2 0 0 3 . N o c i c e p t i v e sensation and sensit iv i ty evoked f r o m human cornea and conjunct iva st imulated by CO2. Invest. Ophth . V i s . S c i . 4 4 , 5 2 9 - 5 3 2 . 62 Hackbar th , H . , K u p p e r s , N . , Bohnet , W . , 2000 . Euthanasia o f rats w i t h carbon d i o x i d e — a n i m a l wel fare aspects. L a b . A n i m . 34 , 9 1 - 9 6 . Hewett , T . A . , K o v a c s , M . S . , A r t w o h l , J .E . , Bennett , B.T . , 1993. A compar ison o f euthanasia methods i n rats, us ing carbon d iox ide i n p re - f i l l ed and f i x e d f l o w rate f i l l e d chambers. L a b . A n i m . S c i . 4 3 , 5 7 9 - 5 8 2 . Hornett , T . D . , Haynes , A . R . , 1984. C o m p a r i s o n o f carbon dioxide/air mixture and nitrogen/air mixture for the euthanasia o f rodents. D e s i g n o f a system for inhalat ion euthanasia. A n i m a l Techno logy 3 5 , 9 3 - 9 9 . Iwarsson , K . , Rehbinder , C , 1993. A study o f different euthanasia techniques i n guinea p igs , rats, and m i c e . A n i m a l response and postmortem f indings . Scan . J . L a b . A n i m . S c i . 2 0 , 1 9 1 - 2 0 5 . K n u t s o n , B . , Burgdor f , J . , Panksepp, J . 1998. A n t i c i p a t i o n o f p lay el ic i ts h igh - f requency ultrasonic vocal i zat ions i n young rats. J . C o m p . P s y c h o l . 112, 6 5 - 7 3 . K n u t s o n , B . , Burgdor f , J . , Panksepp, J . 1999. H igh - f requency u l t rasonic vocal i zat ions index condi t ioned pharmaco log ica l reward i n rats. P h y s i o l . B e h a v . 66 , 6 3 9 - 6 4 3 . K n u t s o n , B . , Burgdor f , J . , Panksepp, J . , 2 0 0 2 . U l t rason ic voca l i zat ions as indices o f affective states i n rats. P s y c h o l . B u l l . 1 2 8 , 9 6 1 - 9 7 7 . L a n s i n g , R . W . , Im , B . S . H . , T h w i n g , J.I., L e g e d z a , A . T . R . , Banzett , R . B . 2000 . The percept ion o f respiratory w o r k and effort can be independent o f the percept ion o f air hunger. A m . J . Resp i r . Cr i t . Care M e d . 162, 1690 -1696 . L e a c h , M . C , B o w e l l , V . A . , A l l a n , T .F . , M o r t o n , D . B . , 2 0 0 2 . A v e r s i o n to gaseous euthanasia agents i n rats and m i c e . Comparat i ve M e d . 5 2 , 2 4 9 - 2 5 7 . L i o t t i , M . , B rannan , S . , E g a n , G . , Shade, R., M a d d e n , L. , A b p l a n a l p , B . , R o b i l l a r d , R., Lancaster , J . , Zamar r ipa , F .E . , F o x , P.T. , Denton , D. , 2 0 0 1 . B r a i n responses associated w i t h consciousness o f breathlessness (air hunger) . P roc . Nat . A c a d . S c i . 9 8 , 2 0 3 5 - 2 0 4 0 . 63 L u s h , M . T . , Janson -B je rk l ie , S . , Car r ie r i , V . K . , L o v e j o y , N . 1988. D y s p n e a i n the venti lator - assisted patient. Heart and L u n g , 17, 5 2 8 - 5 3 5 . M a s u d a , A . , O h y a b u , Y . , K o b a y a s h i , T. , Y o s h i n o , C , Sakakibara , Y . , K o m a t s u , T. , H o n d a , Y . , 2 0 0 1 . L a c k o f posi t ive interaction between CO2 and h y p o x i c s t imulat ion for Pco2 - V A S response slope i n humans. Resp . P h y s i o l . 126, 1 7 3 - 1 8 1 . M o o s a v i , S . H . , Go lestan ian , E . , B i n k s , A . P . , L a n s i n g , R . W . , B r o w n , R., Banzett , R . B . , 2 0 0 3 . H y p o x i c and hypercapnic drives to breathe generate equivalent levels o f air hunger i n humans. J . A p p l . P h y s i o l . 94 , 141 -154 . N i e l s o n , M . , S m i t h , H . , 1952. Studies o n the regulat ion o f respirat ion i n acute hypox ia . A c t a . P h y s i o l . Scand. 24 , 2 9 3 - 3 1 3 . P e p p e l , P. , A n t o n , F., 1993. Responses o f rat medul lary dorsal horn neurons f o l l o w i n g intranasal nox ious c h e m i c a l s t imulat ion : effects o f st imulus intensity, durat ion, and interst imulus interval . J . N e u r o p h y s i o l . 70 , 2 2 6 0 - 2 2 7 5 . R a j , A . B . M . , 1999. B e h a v i o u r o f p igs exposed to mixtures o f gases and the t ime required to stun and k i l l them: welfare impl i ca t ions . Ve t . R e c o r d 144, 1 6 5 - 1 6 8 . R a j , A . B . M . , W o t t o n , S . B . , M c K i n s t r y , J . L . , H i l l e b r a n d , S . J . W . , Pieterse, C , 1998. Changes in the somatosensory evoked potentials and spontaneous electroencephalogram o f broi ler ch ickens dur ing exposure to gas mixtures . Br i t . Pou l t ry S c i . 39 , 6 8 6 - 6 9 5 . Roberts , L . H . , 1975. The rodent ultrasound product ion m e c h a n i s m . U l t rason ics 13, 8 3 - 8 8 . Sales, G . D . 1972. U l t rasound and aggressive behaviour i n rats and other smal l m a m m a l s . A n i m . Behav . 2 0 , 8 8 - 1 0 0 . Sanders, I., W e i s z , D . J . , Y a n g , B . Y . , F u n g , K . , A m i r a l i , A . , 2 0 0 1 . The m e c h a n i s m o f ultrasonic voca l i za t ion i n the rat. Soc . N e u r o s c i . Abs t r . , V o l . 2 7 , P r o g r a m N o . 88 .19 . Shea, S . A . , Har ty , H . R . , Banzett , R . B . , 1996. Se l f - cont ro l o f mechan ica l vent i la t ion to m i n i m i z e CO2 induced air hunger. Resp . P h y s i o l . 103, 1 1 3 - 1 2 5 . 64 S m i t h , W . , Harrap , S B . , 1997. Behav ioura l and cardiovascular responses o f rats to euthanasia us ing carbon d iox ide gas. L a b . A n i m . 3 1 , 3 3 7 - 3 4 6 . T h o m a s , D . A . , Takahash i , L . K . , B a r f i e l d , R . J . 1983. A n a l y s i s o f ultrasonic vocal i zat ions emitted by intruders dur ing aggressive encounters among rats (Rattus norvegicus). J . C o m p . P s y c h o l . 9 7 , 2 0 1 - 2 0 6 . U n i t e d K i n g d o m H o m e O f f i c e , 1997. The H u m a n e K i l l i n g o f A n i m a l s under Schedule 1 to the A n i m a l s (Sc ient i f ic Procedures) A c t 1986 C o d e o f Pract ice. N o r w i c h , He r Majesty 's Stationery O f f i ce . W i d d i c o m b e , J . , 2 0 0 1 . A i r w a y receptors. Resp . P h y s i o l . 125, 3 - 1 5 . 65 CHAPTER 3 : Rats avoid exposure to carbon dioxide and argon3 3.1 Introduction Small laboratory rodents are euthanized using a number of methods, including physical techniques, injectable anaesthetics, and exposure to volatile anaesthetics and other gases. One of the most common methods is exposure to CO2. Animals are exposed to either a gradually increasing concentration of CO2 or a pre-filled chamber, and this causes unconsciousness followed by death. Ideally, a euthanasia method should result in a quick death with minimal pain and distress. In humans CO2 is known to cause dyspnea, a sensation of breathlessness, at concentrations of 8% (Dripps & Comroe, 1947; Liotti et al., 2001). At CO2 concentrations ranging from 30% to 54%, humans also experience pain at the cornea (Chen et al., 1995; Feng & Simpson, 2003), conjunctiva (Feng & Simpson, 2003) and nasal mucosa (Anton et al., 1992; ,Thurauf et al., 2002). In rats, the threshold for the majority of nociceptors in the nasal mucosa is between 37% and 50%> C 0 2 (Anton et al., 1991; Peppel & Anton, 1993). Some studies have reported no behavioural evidence of distress in rats during pre-fill (Blackshaw et al., 1988; Smith & Harrap, 1997) and gradual-fill C 0 2 exposure (Hackbarth et al., 2000; Hornett & Haynes, 1984; Smith & Harrap, 1997). However, others have suggested that CO2 causes behavioural signs of distress (Britt, 1987; Coenen et al., 1995; Iwarsson & Rehbinder, 1993; Chapter 2). This variability between experiments suggests that simply monitoring behavioural responses during exposure may be inadequate as a method for assessing the rat's perception of C 0 2 . 3 A version of this chapter has been accepted for publication. Niel, L. , Weary, D . M . , 2006. Rats avoid exposure to carbon dioxide and argon. Appl. Anim. Behav. Sci. (accepted). 66 A v e r s i o n to CO2 exposure has also been examined us ing preference testing. L e a c h et a l . (2002) found that rats m o v e d to an air f i l l ed chamber w h e n exposed to moderate CO2 concentrations. H o w e v e r , on ly static concentrations between 2 5 . 5 % and 50.8%) were tested. R a t s ' responses to lower concentrations and to g radua l - f i l l exposure have not been examined , and to date no studies have addressed the strength o f avers ion to CO2. Another f o r m o f preference testing, the approach-avoidance test, has been used to examine avers ion to CO2 i n m i n k (Cooper et a l . , 1998), p igs (Raj & Gregory , 1995) and poult ry ( R a j , 1 9 9 6 ; Ger r i t zen et a l . , 2 0 0 0 ; Webster & Fletcher , 2004) . D u r i n g this procedure, entry into the chamber is vo luntary and animals are mot ivated to enter and remain for a reward. I f an imals a v o i d the chamber , even w h e n it contains something that they are trained and mot ivated to obtain such as a food reward, this indicates that they find the condi t ions o f the test cage aversive. Approach -avo idance testing has not prev ious ly been used to examine avers ion to CO2 i n rats. The a i m o f this study was to use approach-avoidance testing to characterize rats' avers ion to static and g radua l - f i l l CO2 exposure. W e also examined whether rats exhibi t avers ion to 9 0 % argon i n air, w h i c h causes death by reduc ing O2 levels to 2% and has been proposed as an alternative method o f gas euthanasia for rats ( Leach et a l . , 2002) . 3.2 Materials and Methods 3.2 .1 S u b j e c t s a n d H o u s i n g The subjects were 10 male Wis ta r rats, 4 0 0 to 500g , obtained f r o m the U B C A n i m a l Care Centre Rodent B reed ing U n i t as surplus supply stock and destined for euthanasia. A n i m a l rooms were kept at 21 ± 1 °C under a 12:12-hr l ight -dark cyc le , and rats were g iven ad l i b i t u m 67 access to food (Lab D ie t 5 0 0 1 , P M I N u t r i t i o n International, R i c h m o n d , U S A ) and tap water. A l l testing was conducted dur ing the l ight por t ion o f the l ight -dark cyc le . Rats were s ing ly housed i n the testing apparatus, cons is t ing o f two transparent cages connected by an opaque tunnel made o f b lack , r ibbed, P V C tubing w i t h a diameter o f 10 c m and s loped so that one cage was 27 c m higher than the other. The ' h o m e ' cage measured 48 x 38 x 20 c m , and contained food , water, bedding , an opaque nestbox and a N y l a b o n e dog chew. The secondary cage measured 45 x 24 x 20 c m and contained bedding . The bot tom cage was a lways used for testing because the test gases were denser than air. D u r i n g pre l iminary testing we determined that the test gases were restricted to the bot tom cage and the lower port ion o f the tunnel . T o determine the effect o f cage fami l ia r i t y , h a l f o f the rats were tested i n the home cage and the other h a l f were tested i n the secondary cage. Th i s was accompl i shed by pos i t ion ing the home cage on the bot tom for h a l f o f the rats, and the secondary cage o n the bot tom for the other h a l f o f the rats. H o w e v e r , many rats spent a port ion o f their t ime i n the secondary cage, so were fami l ia r w i t h both cages. 3 .2 .2 T e s t i n g P r o c e d u r e D u r i n g exper imental test ing, each an imal and its testing apparatus were transferred i n d i v i d u a l l y to a test r o o m . A t this t ime, the nest box was removed and the wi re l i d on the test cage was replaced w i t h a p lex ig las l i d that featured two air outlets pos i t ioned at the end closest to the tunnel , a gas inlet at the far end o f the test cage, and a gas sampl ing tube inserted at the center o f the test cage. The air outlets were covered w i t h mesh to prevent the rats f r o m pushing their noses outside the chamber. The experimenter was concealed beh ind a b l i n d dur ing testing. The testing apparatus and 0 2 meter readout were v ideo recorded dur ing testing. A i r , C O 2 and argon were del ivered to the test cage f r o m compressed gas cy l inders (Praxair , R i c h m o n d , B .C . ) . The treatment gases were passed through a copper c o i l i n a r o o m 68 temperature water bath to regulate gas temperature before entering the test cage. F l o w rates o f gases were measured us ing a var iable area f l o w meter ( D w y e r Instruments V S B - 6 6 - B V ) , and measured f l o w rates for CO2 were adjusted for density us ing a correct ion factor o f 0 .812. O2 concentrations i n the test cage were moni tored dur ing the exper iment us ing a M o c o n L F 7 0 0 D 0 2 analyzer , and were used to calculate CO2 concentrations at speci f ic t ime points (t = x ) w i t h the fo rmula : C 0 2 ( 1 = x ) = 100 - (100 * ( [0 2 { , = x ) ] / [0 2 (, = <,)])). Test ing o f the apparatus was completed to ensure that CO2 concentrations were u n i f o r m throughout the test cage dur ing g radua l - f i l l CO2 addi t ion. D u r i n g CO2 addi t ion at a rate o f 1 7 % o f the test cage v o l u m e per minute , gas concentrations were moni tored at a depth o f h a l f the test cage height at f i ve different sites. There were no obv ious trends for lower CO2 concentrations at the end o f the chamber closest to the tunnel , and CO2 concentrations at the different sites i n the chamber var ied by less than 3 % . Befo re beg inn ing the experiment, rats were trained to per form the approach-avoidance task. Rats were trained for 10 days w i t h air on ly and this was f o l l o w e d by 9 days o f t ra in ing where air and different concentrations o f CO2 were alternated to fami l ia r i ze the animals w i t h gas exposure and remove any effects o f novelty . F o r this f ina l stage o f t ra in ing a l l rats were exposed to air , g radua l - f i l l CO2 at 1 7 % o f the test cage v o l u m e per minute and static CO2 concentrations o f 5 , 10, 15 and 2 0 % . A n i m a l s were not exposed to argon pr ior to testing because we were concerned that this w o u l d affect performance i n general . In p re l iminary test ing one rat that was exposed to argon appeared to have d i f f i cu l ty determining w h i c h cage contained air and refused to run the task the f o l l o w i n g day. Furthermore, argon was not expected to evoke a novel ty response because it is an odourless and non- i r r i tat ing gas. D u r i n g both t ra in ing and exper imental sessions, rats were first l o c k e d into the top cage for 5 m i n to a l l o w t ime for addi t ion o f treatment gases and food rewards to the test cage. F o l l o w i n g lock remova l they were able to enter the lower test cage for a food reward o f 20 69 H o n e y N u t C h e e r i o s ™ (General M i l l s , Inc., M innesota ) . F o r static exposure, the test cage was p r e - f i l l e d w i t h either CO2 or argon. F o r gradual f i l l exposure, CO2 flow into the test cage was init iated w h e n the rat started eating the food reward. The session ended 300 s after l ock remova l , and animals were a l l o w e d to exit and re-enter the test cage dur ing this per iod . A t the end o f the session, the remain ing reward i tems were removed and the rat was returned to the h o l d i n g r o o m . O n the final day o f t ra in ing, a l l rats were run w i t h air i n the test cage and per formed the task correct ly , consuming at least 19 reward i tems each dur ing the first entry into the test cage. The exper iment consisted o f three test per iods. The a i m o f Part 1 was to determine w h i c h static concentrations o f CO2 rats find aversive by filling the test cage w i t h air or static CO2 at concentrations o f 5, 10, 15 and 2 0 % . E a c h rat was tested once w i t h each cond i t ion over a 5 - d per iod . Treatment order was al located accord ing to a L a t i n square and counterbalanced accord ing to home cage pos i t ion ing (high vs. low) . W e recorded the total number o f reward i tems eaten over the entire test session as w e l l as the eating and d w e l l i n g t imes for the first entry (the m a x i m u m exposure t ime tolerated), and predicted that these var iables w o u l d decl ine w i t h increasing CO2 concentrat ion. W e also recorded attempted entries into the test cage as a measure o f mot i va t ion to enter the test cage w i t h each treatment. F i n a l l y , we predicted that i f a treatment was aversive, it w o u l d increase the t ime taken to enter the test cage o n the f o l l o w i n g day. Therefore, we recorded the latency to enter the tube leading to the test cage f o l l o w i n g lock remova l . The a i m o f Part 2 was to determine what concentrat ion o f CO2 rats find aversive dur ing g radua l - f i l l exposure, i n order to compare it to the static fill data f r o m Part 1. Rats were g iven a single exposure to g radua l - f i l l CO2 at a rate o f 17%) o f the test cage v o l u m e per minute. W e were spec i f i ca l l y interested i n the gas concentrations w h e n the rats stopped eating and left the test cage so the O2 concentrat ion was the on ly var iable recorded. Th i s test took place the day after the end o f testing for Part 1. 70 The a i m o f Part 3 was to evaluate rats' responses to a static concentrat ion (90%) o f argon gas i n air ( 2 % O2), and compare this w i t h air exposure. Th i s test took place the day after testing w i t h g radua l - f i l l C O 2 exposure. A s descr ibed above, some adverse reactions to argon were observed dur ing p i lo t testing, and due to the potential for carryover effects f o l l o w i n g argon exposure, rats were exposed to air o n the first day and argon on the second day. Va r iab les recorded were ident ical to Part 1 except the treatments were not counterbalanced, so the latency to enter the tube leading to the test cage was not recorded. 3.2.3 Statistical Analysis One an imal d i d not learn the task and was removed f r o m the experiment. The data for Part 1 and Part 3 were n o n - n o r m a l w i t h unequal var iances and cou ld not be corrected through the use o f transformations, so non-parametr ic statistics were used for analysis . A n in i t ia l eva luat ion o f the data indicated there were no differences for any var iables between rats tested i n the home cage versus the secondary cage, therefore the data were p o o l e d for further analysis . F o r Part 1, the F r i e d m a n ' s test was used to compare dif ferences i n dependant variables across static C O 2 concentrations o f 0 , 5 , 10, 15 and 2 0 % . F o r Part 2 , descr ipt ive statistics are presented only . F o r Part 3 , the W i l c o x o n S igned R a n k s test was used to compare dif ferences i n dependent var iables for air and argon exposure. 3.3 Results 3.3.1 Part 1 - Exposure to static concentrations of CO2 A l l 9 rats entered the test cage at 0 , 5 , 10 and 15%. C 0 2 . A t 2 0 % . C 0 2 one rat refused to enter the test cage. A t 0, 5 and 1 0 % C O 2 rats ate a l l o f the reward i tems prov ided , but the number eaten dec l ined w i t h 15 and 2 0 % C O 2 (F ig . 3 .1a ; P < 0 .005) . A t 1 5 % one an imal refused 71 to eat. A t 2 0 % CO2 on ly two animals ate, and each consumed o n l y one or two reward i tems. C o n s i d e r i n g on ly the first entry into the test cage, the eating and d w e l l i n g t imes were reduced at the highest C 0 2 concentrations (for both : P < 0 .005) . F r o m 0 % to 1 0 % C 0 2 these variables dec l ined on ly s l ight ly (F ig . 3.1b) , but there was increased var iab i l i t y i n response at 10%) C 0 2 . A t 15%> there was a steep drop i n eating and d w e l l i n g t imes (median 32 s and 46 s, respect ively) , and at 2 0 % C 0 2 these values had dropped again (median 2 s and 5 s, ; respect ively) . The total number o f t imes that the rats attempted to enter the test cage dur ing the entire session di f fered across the f ive C 0 2 concentrations (F ig . 3 .1c ; P < 0 .005) . A t 0 and 5%> C 0 2 rats entered the test cage on ly once and remained for the major i ty o f the session, but at higher concentrations rats showed an increasing numbers o f entries. The t ime taken to enter the tube leading to the test cage was not affected by the treatment on the prev ious day (P > 0.1). O n average rats took 2.3 ± 2 s to enter the tube. 3.3.2 Part 2 - Gradual-fill C 0 2 exposure D u r i n g g radua l - f i l l C 0 2 exposure, rats stopped eating and left the test cage at C 0 2 concentrations o f 17.3 ± 2.1% (mean ± standard deviat ion) and 18.4 ± 2.0% respect ively . 3.3.3 Part 3 - Argon exposure D u r i n g argon exposure, three rats refused to enter the test cage, no rats ate, and the m e d i a n d w e l l i n g t ime was 3 s (Table 3.1). The m e d i a n number o f entries dur ing argon exposure was two . In contrast, dur ing air exposure rats ate a l l o f the reward i tems and spent almost the entire session i n the test cage. The number o f reward items eaten, and the eating and d w e l l i n g t imes were a l l s ign i f icant ly greater dur ing the session w i t h air (P < 0 .005) . 72 3.4 Discussion The ult imate a i m o f gas euthanasia is to del iver gases i n such a w a y as to render the a n i m a l unconsc ious wi thout caus ing distress. In the current study rats tolerated extended exposure to 5 % and 1 0 % CO2, but this was not suff ic ient to cause unconsciousness. The rats were u n w i l l i n g to tolerate extended exposure to 1 5 % and 2 0 % CO2, and concentrations greater than 30%o are necessary to cause loss o f consciousness (Chap in & Edgar , 1963; Chapter 2) . Noc icepto rs i n rat nasal m u c o s a beg in to respond to CO2 concentrations o f approx imately 2 5 % ( A n t o n et a l . , 1 9 9 1 ; Peppel & A n t o n , 1993), so pa in is un l ike l y to be the cause o f CO2 avers ion at the two highest concentrations that we tested. D y s p n e a due to hypercapnia more l i k e l y accounts for our results. Some humans report dyspnea at CO2 levels o f on ly 8 % (Dr ipps & C o m r o e , 1947; L i o t t i et a l . , 2001) , and this sensation increases i n severity w i t h higher CO2 concentrat ions. D y s p n e a can also be accompanied by other negative p h y s i c a l symptoms such as headache, f lush , restlessness, heart pound ing , drowsiness and d izz iness ( M o o s a v i et a l . , 2003) . H o w e v e r , humans vary i n their tolerance to hypercapnia (Dr ipps & C o m r o e , 1947). S i m i l a r di f ferences i n tolerance i n rats might exp la in the var iab i l i t y i n eating and d w e l l i n g t imes observed at 1 0 % CO2 i n the current study. The var iab i l i ty at 1 0 % CO2 and the steep drop i n eating and d w e l l i n g t imes at 15%> CO2 suggests that the onset o f severe dyspnea in rats m a y occur w i t h 1 0 % to 1 5 % C 0 2 . It has been suggested that g radua l - f i l l CO2 exposure results i n a s l o w onset o f unconsciousness without distress, but i n the current study rats left the test cage w h e n the CO2 concentrat ion reached o n average 1 8 . 4 % . Th i s CO2 concentrat ion is consistent w i t h rats' avoidance o f static CO2 concentrations, suggesting that g radua l - f i l l exposure does not cause a gradual loss o f consciousness without causing avers ion. In other research we have found that rats do not become recumbent unt i l a CO2 concentrat ion o f approx imate ly 3 0 % is reached, about 73 105 s after gas flow is in i t iated at a flow rate o f 1 7 % per minute (Chapter 2) . The depth o f unconsciousness at the onset o f recumbency is u n k n o w n since some studies report a short delay before a loss o f ref lexes occurs (Danneman et a l . , 1997; Hewett et a l . , 1993). Th i s suggests that some awareness o f aversive CO2 concentrations might persist for a short per iod after onset o f recumbency . In the present study, a concentrat ion o f 1 8 % was reached after 60 s, suggesting that dur ing g radua l - f i l l CO2 exposure, rats are exposed to aversive CO2 concentrations for at least 45 s before los ing consciousness. Increased flow rates w o u l d expose rats to higher CO2 concentrat ions, but l i ke l y for a shorter durat ion before unconsciousness. The net effects on the rats o f this conf l i c t between intensity and durat ion are u n k n o w n . F o r static CO2 exposure, the number o f reward i tems eaten, and eating and d w e l l i n g t imes show s imi la r trends, but eating and d w e l l i n g t imes appear to be more sensit ive measures, at least at lower concentrations o f CO2. W h i l e eating and d w e l l i n g t imes indicate the m a x i m u m t ime rats are w i l l i n g to tolerate exposure, the number o f reward i tems eaten prov ides a measure o f the rats' act iv i ty i n the test cage over the entire testing procedure and is affected by re-entries. Furthermore, rats can consume the same amount i n a shorter exposure per iod by increasing eating speed. A l t h o u g h rats ate the same amount w i t h air and 5%> CO2, at 5%> CO2 the median t ime spent eating was s l ight ly reduced without an increase i n entries, suggesting that they s i m p l y ate more q u i c k l y . The responses o f rats i n the current study m a y not be indicat ive o f h o w rats w o u l d respond i f exposed to CO2 for the first t ime. In order to per fo rm w i t h i n - a n i m a l compar isons it was necessary to remove the effects o f novel ty by f a m i l i a r i z i n g an imals w i t h a l l C 0 2 concentrations and del ivery methods pr ior to testing. Rats tended to be less tolerant o f static CO2 dur ing t ra in ing , suggesting that novel ty made exposure more aversive. Thus , by r e m o v i n g the effects o f novel ty w e have obtained a better ind icat ion o f avers ion to the properties o f the gas itself. A n i m a l s were not trained w i t h argon, but it is an inert gas w i t h no perceptible odour, so a 74 novel ty response to odour w o u l d not be expected. Increases i n inspi red CO2 and reductions i n inspi red O2 both cause an in i t ia l increase i n vent i lat ion rate and depth ( L u m b , 2000) , w h i c h c o u l d also contribute to novelty . H o w e v e r , because hypercapnia and h y p o x i a have s imi la r effects o n vent i lat ion , t ra in ing w i t h CO2 w o u l d l i k e l y also have been effect ive for argon exposure. O u r results demonstrate that approach-avoidance testing can prov ide a sensit ive and object ive method for e x a m i n i n g rats' avers ion to gas euthanasia agents. The major i ty o f prev ious studies e x a m i n i n g distress associated w i t h CO2 have recorded behavioural responses dur ing p r e - f i l l or g radua l - f i l l euthanasia, and the results have been var iable . Some studies have reported no behav ioura l evidence o f distress dur ing C 0 2 exposure (Hornett & Haynes , 1984; S m i t h & Har rap , 1997; Hackbar th et a l . , 2000) . H o w e v e r , others have conducted subjective assessments o f distress ( Iwarsson & Rehbinder , 1993; C o e n e n et a l . , 1995) and taken object ive behav ioura l measures (Britt , 1987; Chapter 2) and conc luded that indicat ions o f distress were present. S o m e var iab i l i ty among experiments m a y be due to differences i n the w a y behavioural responses were interpreted. F o r example , Br i t t (1987) reported that Sprague D a w l e y rats exh ib i ted increased act iv i ty and L is te r H o o d e d rats exh ib i ted decreased act iv i ty , yet both responses were interpreted as indicat ive o f ' d i scomfor t ' . The current methodology prov ides a measure o f avers ion that is less open to subjective interpretation. Preference testing has prev ious ly been used to determine whether rats w i l l avo id exposure to 2 5 . 5 , 34.9 and 5 0 . 8 % CO2 ( Leach et a l . , 2002) . Th i s prev ious study reported consistent avoidance at a l l three CO2 concentrations, but imposed no cost to leav ing the chamber , m a k i n g it d i f f i cu l t to assess the strength o f avers ion to CO2. W i t h the current approach-avoidance des ign, rats had to forgo a food reward to avo id CO2 exposure. W e can therefore conc lude that the rats' mot ivat ion to avo id CO2 at concentrations above 15%> is stronger than their mot ivat ion to obtain a palatable food reward w h e n fed ad l i b i tum. A l t h o u g h 75 the strength o f rats' mot i va t ion for sweet foods w h e n fed ad l i b i t u m has not spec i f i ca l l y been investigated i n the current study, there is evidence to suggest that it ranges f r o m moderate to h igh . In the present study rats were qu ick to enter the cage f o l l o w i n g l o c k r e m o v a l , and consumed the entire food reward dur ing a l l sessions w i t h air. Furthermore, prev ious studies have s h o w n that mot i va t ion for sucrose w h e n fed ad l i b i t u m is as m u c h as 5 0 - 7 5 % o f the mot i va t ion for sucrose w h e n food depr ived. M c G r e g o r et a l . (1999) trained rats to l i c k a sipper tube for access to 8 . 6 % sucrose, and dur ing a 45 minute test session w i t h a progressive ratio schedule o f reward it was found that ad l i b i t u m fed rats l i c k e d approx imate ly 1500 t imes, wh i le food depr ived rats l i c k e d approx imately 2000 t imes. The m a x i m u m number o f l i cks for a s ingle reward was approx imately 55 for ad l i b i t u m rats and 70 for food -depr i ved rats. In another study rats were trained to bar press for access to sucrose, and it was found that ad l i b i t u m fed rats bar pressed approx imate ly 5 5 % as m u c h as food -depr ived rats d i d for access to 4 % and 16%> sucrose (Co l l i e r and B o l l e s , 1968). These results suggest that rats i n this exper iment were w e l l mot ivated by the food reward, and thus indicates that the rats' mot i va t ion to avo id exposure to CO2 concentrations o f 1 5 % and greater was at least as moderate. The strength o f rats' avers ion to CO2 cou ld be more accurately assessed i n future studies by increasing hunger levels to ensure that mot ivat ion for the food reward is h igh . Approach -avo idance testing has also been used to examine avers ion to CO2 i n other species. The major i ty o f these studies have examined moderate to h igh CO2 concentrations because gas stunning regulations for fa rm animals general ly require p r e - f i l l exposure (e.g., European U n i o n , 1993). Cooper et a l . (1998) found that m i n k w i l l avo id a chamber conta in ing more than 8 0 % CO2, even w h e n it contains a nove l object that they are mot ivated to obtain. P igs have been found to a v o i d a food reward w h e n it is paired w i t h 90% CO2, even f o l l o w i n g food depr ivat ion , but w i l l tolerate moderate durations o f exposure to 30%> CO2 (Raj & Gregory , 1995). P rev ious studies w i t h poultry have found that turkeys and ch ickens w i l l enter CO2 76 concentrations greater than 60% for a reward of food or social contact, and w i l l lose consciousness before they are able to exit the chamber (Raj, 1996; Gerritzen et al., 2000; Webster & Fletcher, 2004). This suggests that poultry have a greater tolerance for C 0 2 exposure, but it is not clear whether this interspecific variability demonstrates a difference in gas perception or in motivation to obtain the reward. The birds were found to exhibit behavioural and physiological signs of distress during exposure such as hyperventilation, coughing and head shaking, suggesting that the gas was likely detectable and unpleasant. CO2 euthanasia is widely used because it is easy to perform, inexpensive, safe for laboratory workers, and involves little handling and restraint for animals. Another gas euthanasia agent which meets these criteria is argon gas, which causes unconsciousness and death by O2 displacement. In pigs and poultry, unconsciousness and death occur at argon concentrations greater than 90%>, which lowers O2 concentrations below 2% (e.g., Raj, 1999; Raj & Tserveni-Gousi, 2000). During approach-avoidance testing, pigs, turkeys and chickens have been found to enter and remain in lethal concentrations of argon for a food reward (Raj & Gregory, 1995; Raj, 1996; Webster & Fletcher, 2004). During preference testing, rats and mice have been found to tolerate argon exposure for longer than CO2 exposure, but to exit before loss of consciousness for both gases (Leach et al., 2002). This suggests some level of aversion with both CO2 and argon. In the current study we found that when the test cage contained 90% argon, some rats refused to enter and others exited immediately. This indicates that rats can detect argon-induced hypoxia and that they find lethal argon concentrations aversive. Argon is an odourless and non-irritating gas, so this aversion is not likely to be due to the properties of argon. However, low levels of inspired O2 results in hypoxia, and this could cause a sensation of dyspnea. The O2 concentration in ambient air is 20.9%, and Moosavi et al. (2003) found that humans report dyspnea at less than 8% O2 when breathing is constrained. This sensation was alleviated with spontaneous breathing, but O2 concentrations less than 1% were not examined. 77 H o w e v e r , h u m a n pi lots have been found to lose consciousness after cab in depressurizat ion without apparent efforts at correct ion, ind icat ing that humans m a y not experience dyspnea w i t h h y p o x i a levels that are suff ic ient to cause loss o f consciousness (Cable , 2003) . The design o f the current study does not a l l o w for direct compar isons between responses to CO2 and argon, but it appears that rats are not w i l l i n g to tolerate exposure to either gas for suff ic ient periods to cause unconsciousness. In humans, there is a delay to onset o f dyspnea after a change i n inspi red levels o f CO2 or O2. Th i s delay is due to the t ime necessary for changes i n b l o o d CO2 and O2 levels to reach the per ipheral and central chemoreceptors, w h i c h is about 5 to 15 s i n humans ( rev iewed by C u n n i n g h a m et a l . , 1986), and for h y p o x i a and hypercapnia to reach levels that are suff ic ient to evoke dyspnea. Banzett (1996) calculated the ha l f - t ime for development o f a stable level o f dyspnea i n humans to be approx imately 32 s. In the current study, the m e d i a n latency to leave w i t h 2 0 % CO2 and w i t h 9 0 % argon exposure was on ly 5 s and 3 s, respect ively , and this response was m u c h qu icker than the t ime taken for dyspnea to develop i n humans. H o w e v e r , the c i rculatory delay is on ly 2 s i n rats (Lagneaux, 1986), and the dynamics o f dyspnea i n rats is u n k n o w n , so dyspnea cannot be ru led out as a potential source o f avers ion dur ing both CO2 and argon exposure i n the current study. O u r results suggest that p r e - f i l l and g radua l - f i l l CO2 exposure, and 90%> argon exposure, cause avers ion i n rats. They also indicate that approach-avoidance testing is a sensitive and object ive method for assessing avers ion to gas euthanasia methods i n rats. Further w o r k is needed to assess h o w avers ion to CO2 compares w i t h other gas and non-gas methods o f euthanasia, so that the most humane methods o f euthanasia can be implemented . 78 Table 3.1. M e d i a n (with 2 5 t h and 7 5 t h percenti les) number o f reward i tems eaten dur ing the entire session and eating and d w e l l i n g t imes dur ing the first entry w i t h either air or argon i n the test cage (n = 9 rats). „ ~ W i l c o x o n S igned T y p e 0 f g a S R a n k s test A i r A r g o n R e w a r d i tems eaten (no.) 20 (20,20) 0 (0 ,0 ) 22.5 <0.005 Eat ing t ime (s) 261 ( 2 5 2 , 2 8 1 ) 0 (0 ,0 ) 22.5 <0.005 D w e l l i n g t ime (s) 296(296,297) 3 ( 3 , 6 ) 22.5 <0.005 79 250 -i 200 - CD 150 - E h- 100 - 50 - </> 4 g> -*—» iS 3 H E 2 < 1 —I 1 1 — 5 10 15 C a r b o n dioxide concentrat ion (%) 20 F i g u r e 3 .1. Median (± interquartile ranges) a) number of reward items eaten, b) eating time (filled squares) and dwelling time (open squares) for the first entry, and c) number of attempted entries into the test cage during sessions with 0, 5, 10, 15 and 20% CO2 (n = 9 rats). 80 3.5 References A n t o n , F., Euchner , I., Handwerker , H .O . , 1992. Psychophys i ca l examinat ion o f pa in induced by def ined CO2 pulses appl ied to the nasal mucosa . P a i n 4 9 , 5 3 - 6 0 . A n t o n , F., P e p p e l , P. , Euchner , I., Handwerker , H .O . , 1991. Cont ro l l ed nox ious chemica l s t imulat ion : responses o f rat t r igeminal brainstem neurones to CO2 pulses appl ied to the nasal mucosa . N e u r o s c i . Lett . 123, 2 0 8 - 2 1 1 . Banzett , R . B . , 1996. D y n a m i c response characteristics o f C C ^ - i n d u c e s air hunger. Resp . P h y s i o l . 105, 4 7 - 5 5 . , B l a c k s h a w , J . K . , F e n w i c k , D . C . , Beatt ie , A . W . , A l l a n , D . J . , 1988. The behaviour o f ch ickens , m i c e and rats dur ing euthanasia w i t h c h l o r o f o r m , carbon d iox ide and ether. L a b . A n i m . 2 2 , 6 7 - 7 5 . Br i t t , D . P. , 1987. The humaneness o f carbon d iox ide as an agent o f euthanasia for laboratory rodents. In: Euthanasia o f U n w a n t e d , Injured or D iseased A n i m a l s or for Educat iona l or Sc ient i f i c Purposes, pp. 19-3.1. Potter 's B a r : Un ivers i t ies Federat ion for A n i m a l Wel fa re . C a b l e , G . G . , 2 0 0 3 . In - f l ight h y p o x i a incidents i n mi l i ta ry aircraft : causes and impl icat ions for t ra in ing. A v i a t . Space E n v i r o n . M e d . 74 , 169 -172 . C h a p i n , J . L . , Edgar , J . L . R . 1963. C o o l i n g o f rats i n carbon d iox ide . A m . J . P h y s i o l . 2 0 4 , 7 2 3 - 726. C h e n , X . , Ga l la r , J . , P o z o , M . A . , B a e z a , M . , B e l m o n t e , C , 1995. C 0 2 s t imulat ion o f the cornea: a compar i son between h u m a n sensation and nerve act iv i ty i n p o l y m o d a l nocicept ive afferents o f the cat. Eur . J . N e u r o s c i . 7, 1 1 5 4 - 1 1 6 3 . C o e n e n , A . M . , D r inkenburg , W . H . , Hoenderken , R., van Lui j te laar , G . L . , 1995. Carbon d iox ide euthanasia i n rats: oxygen supplementat ion m i n i m i z e s signs o f agitat ion and asphyx ia . L a b . A n i m . 2 9 , 2 6 2 - 2 6 8 . 81 C o l l i e r , G . , B o l l e s , R., 1968. Hunger , thirst, and their interaction as determinants o f sucrose consumpt ion . J . C o m p . P h y s i o l . P s y c h . 66 , 6 3 3 - 6 4 1 . Cooper , J . , M a s o n , G . , R a j , M . , 1998. Determinat ion o f the avers ion o f fa rmed m i n k (Mustela vison) to carbon d iox ide . V e t . R e c . 143, 3 5 9 - 3 6 1 . C u n n i n g h a m , D . J .C . , R o b b i n s , P. A . , W o l f f , C . B . 1986. Integration o f respiratory response to changes i n a lveolar part ia l pressures o f C 0 2 and O2 and i n arterial p H . In: Chern iak , N . S . , W i d d i c o m b e , J . G . (eds), H a n d b o o k o f P h y s i o l o g y , Sect ion 3 : The Respiratory Sys tem, V o l u m e II: Con t ro l o f Breath ing , Part 2 , A m e r i c a n P h y s i o l o g i c a l Soc iety , Wash ington , D . C . , p p . 4 7 5 - 5 2 8 D a n n e m a n , P . J . , Ste in , S . , W a l s h a w , S . O . , 1997. H u m a n e and pract ical impl i cat ions o f us ing carbon d iox ide m i x e d w i t h oxygen for anaesthesia or euthanasia o f rats. L a b . A n i m . S c i . 4 7 , 3 7 6 - 8 5 . D r i p p s , R . D . , C o m r o e , J . H . , 1947. The respiratory and c i rculatory response o f no rmal m a n to inhalat ion o f 7.6 and 10.4 per cent CO2 w i t h a compar i son o f the m a x i m a l vent i lat ion produced by severe muscu lar exercise, inhalat ion o f CO2 and m a x i m a l voluntary hypervent i lat ion. A m . J . P h y s i o l . 149, 4 3 - 5 1 . European U n i o n 1993. C o u n c i l D i rec t i ve 93/119/EC o f 22 December 1993 o n the protect ion o f an imals at the t ime o f slaughter or k i l l i n g . O f f i c i a l Journal L 3 4 0 , 31/12/1993, 0 0 2 1 - 0 0 3 4 . F e n g , Y . , S i m p s o n , T . L., 2 0 0 3 . N o c i c e p t i v e sensation and sensit iv i ty evoked f r o m human cornea and conjunct iva st imulated by CO2. Invest. Ophth . V i s . S c i . 4 4 , 5 2 9 - 5 3 2 . Ger r i t zen , M . A . , L a m b o o i j , E. , H i l l e b r a n d , S . J . W . , Lanhaar , J . A . C . , Pieterse, C , 2000 . B e h a v i o r a l responses o f broi lers to different gaseous atmospheres. Poul t . S c i . 79 , 9 2 8 - 9 3 3 . Hackbar th , H . , K u p p e r s , N . , Bohnet , W . , 2000 . Euthanas ia o f rats w i t h carbon d i o x i d e — a n i m a l wel fare aspects. L a b . A n i m . 34 , 9 1 - 9 6 . Hornett , T . D . , Haynes , A . R . , 1984. C o m p a r i s o n o f carbon dioxide/air mix ture and nitrogen/air 82 mixture for the euthanasia o f rodents. D e s i g n o f a system for inhalat ion euthanasia. A n i m a l Techno logy 3 5 , 9 3 - 9 9 . Hewett , T . A . , K o v a c s , M . S . , A r t w o h l , J .E . , Bennett , B.T . , 1993. A compar ison o f euthanasia methods i n rats, us ing carbon d iox ide i n p re - f i l l ed and f i x e d f l o w rate f i l l e d chambers. L a b . A n i m . S c i . 4 3 , 5 7 9 - 5 8 2 . Iwarsson , K . , Rehb inder , C , 1993. A study o f different euthanasia techniques i n guinea p igs , rats, and m i c e . A n i m a l response and postmortem f ind ings . Scan . J . L a b . A n i m . S c i . 2 0 , 1 9 1 - 2 0 5 . Lagneaux , D. , 1986. Vent i la tory responses o f the rat to m i l d hypercapni s t imulat ion before and after a lmitr ine bismesylate. Resp . P h y s i o l . 6 5 , 3 7 9 - 3 8 8 . L e a c h , M . C , B o w e l l , V . A . , A l l a n , T .F . , M o r t o n , D . B . , 2 0 0 2 . A v e r s i o n to gaseous euthanasia agents i n rats and m i c e . Comparat i ve M e d . 52 , 2 4 9 - 2 5 7 . L i o t t i , M . , B rannan , S . , E g a n , G . , Shade, R., M a d d e n , L., A b p l a n a l p , B . , R o b i l l a r d , R. , Lancaster , J . , Zamar r ipa , F .E . , F o x , P.T. , Denton , D. , 2 0 0 1 . B r a i n responses associated w i t h consciousness o f breathlessness (air hunger) . P r o c . Nat . A c a d . S c i . 9 8 , 2 0 3 5 - 2 0 4 0 . L u m b , A . B . , 2 0 0 0 . N u n n ' s A p p l i e d Respi ratory Phys io logy . But te rwor th -He inemann , W o b u r n , M A , pp. 8 2 - 1 0 6 . M c G r e g o r , I.S., Saharov, T., Hunt , G . E . , Topp le , A . N . , 1999. B e e r consumpt ion in rats: the inf luence o f ethanol content, food depr ivat ion, and cocaine. A l c o h o l 17, 4 7 - 5 6 . M o o s a v i , S . H . , Go lestan ian , E., B i n k s , A . P . , L a n s i n g , R . W . , B r o w n , R., Banzett , R . B . , 2 0 0 3 . H y p o x i c and hypercapnic drives to breathe generate equivalent levels o f air hunger i n humans. J . A p p l . P h y s i o l . 94 , 141 -154 . P e p p e l , P. , A n t o n , F., 1993. Responses o f rat medul la ry dorsal horn neurons f o l l o w i n g intranasal nox ious c h e m i c a l s t imulat ion : effects o f st imulus intensity, durat ion, and interst imulus interval . J . N e u r o p h y s i o l . 7 0 , 2 2 6 0 - 2 2 7 5 . 83 R a j , A . B . M . , 1996. A v e r s i v e reactions o f turkeys to argon, carbon d iox ide and a mixture o f carbon d iox ide and argon. Ve t . R e c . 138, 5 9 2 - 5 9 3 . R a j , A . B . M . , 1999. B e h a v i o u r o f pigs exposed to mixtures o f gases and the t ime required to stun and k i l l them: welfare impl icat ions . Ve t . R e c . 144, 1 6 5 - 1 6 8 . R a j , A . B . M . , G regory , N . G . , 1995. Wel fa re impl icat ions o f the gas stunning o f p igs 1. Determinat ion o f avers ion to in i t ia l inhalat ion o f carbon d iox ide or argon. A n i m . Wel fa re 4 , 2 7 3 - 2 8 0 . R a j , A . B . M . , T s e r v e n i - G o u s i , A . , 2000 . Stunning methods for poult ry . W o r l d Pou l t ry S c i . J . 56 , 2 9 1 - 3 0 4 . S m i t h , W . , Harrap , S . B . , 1997. Behav iou ra l and cardiovascular responses o f rats to euthanasia us ing carbon d iox ide gas. L a b . A n i m . 3 1 , 3 3 7 - 3 4 6 . Thurauf , N . , Gunther , M . , P a u l i , E. , K o b a l , G . , 2 0 0 2 . Sensi t iv i ty o f the negative m u c o s a l potent ial to the t r igeminal target st imulus CO2. B r a i n R e s . 9 4 2 , 2 7 - 8 6 . Webster , A . B . , F letcher , D . L . , 2 0 0 4 . Assessment o f the avers ion o f hens to different gas atmospheres us ing an approach-avoidance test. A p p l . A n i m . Behav . S c i . 88 , 2 7 5 - 2 8 7 . 84 CHAPTER 4: Effect of flow rate on aversion to gradual-fill carbon dioxide euthanasia in rats4 4.1 Introduction Laboratory rats are c o m m o n l y euthanized us ing CO2, w h i c h induces unconsciousness f o l l o w e d by death. A n i m a l s are either p laced into a chamber that has been p re - f i l l ed w i t h CO2, or the chamber conta in ing the an imals is gradual ly f i l l e d w i t h CO2 unt i l death is conf i rmed . H o w e v e r , the term 'g radua l ' encompasses a large range o f f l o w rates, and opt imal f l o w rates for m i n i m i z i n g distress have not yet been ident i f ied. Ideal ly , a method o f euthanasia should induce death q u i c k l y wi thout caus ing pa in or distress. CO2 concentrations o f greater than 8 % are k n o w n to cause dyspnea, w h i c h is an unpleasant sensation o f breathlessness, i n humans (Dr ipps & C o m r o e , 1947; L io t to et a l . , 2001) , and m a y cause s imi la r sensations i n an imals . CO2 also forms carbonic ac id o n the mucous membranes and is k n o w n to cause pa in i n humans at concentrations greater than 30 to 5 0 % ( A n t o n et a l . , 1992; C h e n et a l . , 1995; F e n g & S i m p s o n , 2 0 0 3 ; Thurauf et a l . , 2002) . Noc icepto rs i n the nasal m u c o s a and cornea o f rats are also st imulated by C 0 2 (Peppel and A n t o n , 1993; H i ra ta et a l . , 1999), suggesting CO2 l i k e l y also causes pa in in rats. D u r i n g gradual - f i l l CO2 euthanasia, faster f l o w rates cause loss o f consciousness and death more q u i c k l y (Hornett & Haynes , 1984; C o e n e n et a l . , 1995). H o w e v e r , w i t h faster f l o w rates animals lose consciousness at higher CO2 concentrations ( A m b r o s e et a l . , 2000) , l i k e l y because loss o f consciousness dur ing CO2 exposure is dependent o n p H changes i n the cerebral spinal f l u i d and s l o w f i l l rates a l l o w more t ime for these p H changes to occur . Th i s exposure to higher CO2 concentrations before loss o f consciousness m a y increase the severity o f any pa in and dyspnea that occur. Thus , the potential for distress m a y vary w i t h f l o w rate; h igh f l o w rates w i l l expose 4 A version of this chapter has been accepted for publication. Niel, L. , Stewart, S.A., Weary, D . M . 2006. Effect of flow rate on aversion to gradual-fill carbon dioxide euthanasia in rats. Appl. Anim. Behav. Sci. (accepted). 85 animals to gas concentrations w i t h a h igh potential for caus ing distress for a short per iod , w h i l e l o w f l o w rates w i l l cause pro longed exposure to gas concentrations w i t h a l o w to moderate potential for caus ing distress. It has been suggested anecdotal ly that a s l o w l y increasing CO2 concentrat ion a l lows for gradual onset o f unconsciousness, wi thout the an imal exper ienc ing aversive CO2 concentrations. T h e interaction between f l o w rate and m a x i m u m CO2 concentrat ion before unconsciousness indicates that this m a y be p laus ib le , such that a range o f s l o w f l o w rates m a y a l l o w for euthanasia without causing distress. Rats ' reactions to g radua l - f i l l CO2 euthanasia have been assessed us ing both behavioural responses to exposure and preference testing. W h i l e some studies have reported a lack o f behavioural response to g radua l - f i l l CO2 euthanasia (Hackbarth et a l . , 2 0 0 0 ; Hornett & Haynes , 1984; S m i t h & Harrap , 1997), others have reported behavioural responses that suggest distress (Britt , 1987; C o e n e n et a l . , 1995; Chapter 2) . T w o studies have spec i f i ca l l y examined the effect o f f l o w rate dur ing g radua l - f i l l CO2 exposure i n rats. Hornett and Haynes (1984) d i d not find an effect o f f l o w rate on behavioural responses, but C o e n e n et a l . (1995) found that a rate o f 1 2 5 % o f the chamber v o l u m e per minute caused greater gasping than 14%> o f the chamber v o l u m e per minute . Th i s increase i n gasping suggests a potential for dyspnea w i t h the faster f l o w rate. In Chapter 3 w e found that rats show avers ion to g radua l - f i l l CO2 exposure at a rate o f 17%> per minute , but the effects o f f l o w rate were not examined. The a i m o f this study was to use approach-avoidance test ing to determine whether avers ion to g radua l - f i l l euthanasia varies w i t h flow rate. 86 4.2 Materials and Methods 4.2 .1 Sub jec ts a n d H o u s i n g The subjects were eight 8 -month o l d , male Wis ta r rats destined for euthanasia as surplus stock f r o m the U B C Rodent B reed ing Un i t . A n i m a l rooms were kept at 21 ± 1 °C under a 1 2 : 1 2 - hr l ight -dark cyc le , and rats were g iven ad l i b i t u m access to food ( Lab D ie t 5 0 0 1 , P M I N u t r i t i o n International, R i c h m o n d , U S A ) and tap water. A l l testing was conducted dur ing the l ight port ion o f the l ight -dark cyc le . Rats were s ing ly housed i n the testing apparatus, cons is t ing o f two transparent cages connected by a s loped, opaque tunnel . The top or ' h o m e ' cage measured 48 x 38 x 20 c m , and contained f o o d , water, bedding , an opaque nestbox and a N y l a b o n e d o g chew. The bottom or 'test' cage measured 45 x 24 x 20 c m and contained bedding . The home cage was 27 c m higher than the test cage, and the connect ing tunnel was made o f b l a c k , r ibbed, P V C tubing w i t h a diameter o f 10 c m . 4.2.2 T e s t i n g P r o c e d u r e D u r i n g exper imental test ing, each an imal and its test ing apparatus were transferred ind i v idua l l y to a separate r o o m . A t this t ime, the wi re l i d o n the test cage was replaced w i t h a p lex ig las l i d f itted w i t h a gas inlet i n the center, two air outlets (1.5 c m i n diameter) pos i t ioned at the end closest to the tunnel , and a gas sampl ing tube inserted at the far end o f the test cage. The air outlets were covered w i t h mesh to prevent the an imals f r o m push ing their noses outside the test cage. A part i t ion was p laced behind the exper imental set-up to conceal the experimenter dur ing testing. Because the test cage opened direct ly into the tunnel , the total v o l u m e (24 L ) was calculated to inc lude the v o l u m e o f the test cage p lus the v o l u m e o f the por t ion o f the tunnel that was at the same height as the test cage. P re l iminary testing was conducted to examine var iabi l i ty 87 i n CO2 concentrations i n the test cage dur ing the f i l l i n g process. CO2 concentrations were moni tored at a depth o f ha l f the test cage height at 5 different sites w i t h f l o w rates ranging f r o m 5 % to 2 8 % o f the test cage v o l u m e per minute , and were found to vary by less than 3 ^ 5 % w i t h any g iven f l o w rate. Furthermore, there was no obv ious trend for lower CO2 concentrations at end o f the test cage closest to the tunnel . Because CO2 concentrations tend to be greater near the bot tom o f the chamber than at the top (Britt , 1987; Chapter 2) , measurements dur ing the exper iment were taken 10 c m above the site o f reward del ivery . A i r and CO2 were del ivered to the test cage f r o m compressed gas cy l inders (Praxair , R i c h m o n d , B .C . ) . The treatment gases were passed through a copper c o i l i n a r o o m temperature water bath to regulate the temperature o f the gas before it entered the test cage. F l o w rates o f gases were measured w i t h a var iable area f lowmeter ( M o d e l V S B - 6 6 - B V , D w y e r Instruments, Inc.) , and measured CO2 f l o w rates were adjusted for density us ing a correct ion factor o f 0 .812. Rats were trained to enter the lower cage for a food reward , and had prev ious ly been tested w i t h exposure to static and gradual ly increasing concentrations o f CO2 for a separate experiment. D u r i n g exper imental sessions, rats were first l o c k e d i n the top cage for 2 m i n . A f t e r the l o c k was removed the rats were able to enter the lower test cage for a food reward o f 20 H o n e y N u t C h e e r i o s ™ (General M i l l s , Inc., Minnesota ) . A s soon as the rat entered the test cage and started eating the C h e e r i o s ™ , either air or CO2 flow was init iated at a pre -determined rate. Rats c o u l d remain i n the test cage for a m a x i m u m o f 300 s f r o m the t ime that gas flow began, after w h i c h the test session was ended. I f the rat entered the home cage dur ing this per iod the test session was stopped. A t the end o f the session, the remain ing reward i tems were removed and counted, and the rat was returned to the ho ld ing r o o m . The test ing apparatus and O2 meter readout were v ideo recorded dur ing testing. W e also recorded the total number o f reward i tems eaten over the entire test session as w e l l as the 88 latency to stop eating and the latency to leave the test cage after gas f l o w had begun. G a s concentrations i n the test cage were moni tored dur ing the exper iment v i a the gas sampl ing tube us ing a M o c o n L F 7 0 0 D O2 analyzer. The O2 concentrat ion was recorded and used to calculate C 0 2 concentrations at each o f these t imes (t = x ) w i t h the fo rmula : C02(t= X) = 100 - (100 * ([O2 (t = x ) ] / [O 2 ( , = 0)])). Rats were tested i n two replicates o f eight test sessions, and i n each replicate rats were tested o n f ive days w i t h C 0 2 and o n three control days w i t h air. F o r both repl icates, rats were tested w i t h f i ve different C 0 2 f l o w rates: 3 , 7, 14, 2 0 , and 2 7 % o f the test cage v o l u m e per minute . In the first repl icate, a f l o w rate o f 2 1 % per minute was used for a l l three test sessions w i t h air. In the second repl icate, f l o w rates o f 4 , 17 and 3 3 % per minute were used for the three test sessions w i t h air. Treatment order for CO2 and air was balanced across rats and days by an 8 x 8 L a t i n square. O n l y f l o w rates less than 30%> per minute were examined because the results o f A m b r o s e et a l . (2000) suggest that faster f l o w rates result i n potent ial ly pa in fu l CO2 concentrations before an imals lose consciousness. 4.2.3 Statistical Analysis D a t a were averaged w i t h i n rat and CO2 f l o w rate for the two repl icates, result ing i n 40 observations for the analys is o f CO2 f l o w rate (8 rats and 5 f l o w rates). F o r the analysis o f air f l o w rate, the three f l o w rates f r o m the second replicate were examined , result ing i n 24 observations (8 rats and 3 f l o w rates). Dependent var iables were analyzed us ing a m i x e d m o d e l ( S A S v9.1) that inc luded rat (7 d.f.) as a random effect, and tested for l inear and quadratic effects o f f l o w rate (1 d.f. for each) against an error term w i t h 30 d.f. for the test o f CO2 f l o w rate and 14 d.f. for the test o f air f l o w rate. Latency to leave the test cage was not tested i n the air f l o w analysis because a l l an imals remained i n the test cage for the entire testing per iod . 89 4.3 Results D u r i n g test sessions w i t h air, rats ate on average ( ± S E ) 19.3 ± 0.3 reward i tems out o f 2 0 , and f in ished eating 2 7 0 ± 6 s after entering the test cage. A l l rats remained i n the test cage for the entire 300 s testing per iod for a l l air sessions. Changes i n air f l o w rate d i d not affect the latency to stop eating ( l inear: F\,u - 1.74, P > 0 . 1 ; quadratic : F i j 4 = 0 .19 , P > 0.1) or the number o f reward i tems eaten (l inear: F i j 4 = 0 .29 , P > 0 . 1 ; quadratic : F i j 4 = 0 .07, P > 0.1). In contrast, the number o f reward i tems eaten, the latency to stop eat ing, and the latency to leave the test cage decreased w i t h increasing CO2 f l o w rates (F ig . 4.1 a, b). B o t h the l inear and curv i l inear effects were s igni f icant for the number o f reward i tems eaten (l inear: Fi^o = 6 7 . 2 1 , P < 0 . 0 0 1 ; quadratic: F i > 3 0 = 5 .02 , P < 0.05) , the latency to stop eating ( l inear: F i , 3 0 = 128.36, P < 0 . 0 0 1 ; quadratic: F U o = 10 .91 , P < 0 .01) , and the latency to leave the test cage ( l inear: F i > 3 0 = 171.24, P < 0 . 0 0 1 ; quadratic: Fh30 = 11-84, P < 0.01). The rats d i d not remain i n the test cage for long enough to lose consciousness at any o f the f l o w rates, but there was a curv i l inear relat ionship between CO2 f l o w rate and the CO2 concentrat ion at the t ime rats stopped eating (F i > 3 o = 5.65 , P < 0.05) and left the test cage (Fi , 3 o = 9 .02 , P < 0.01). Rats stopped eating and left the test cage at lower CO2 concentrations w i t h the lowest and highest f l o w rates (F ig . 4.1 c). Rats left the test cage at the highest CO2 concentrat ion ( 1 5 . 9 % CO2) w h e n tested at the 14%> per minute f l o w rate. H o w e v e r , the CO2 concentration w h e n rats left the test cage var ied considerably across rats. W h e n averaged w i t h i n rat across a l l days for a l l f l o w rates, the average CO2 concentrat ion w h e n rats left the test cage ranged f r o m 1 1 . 1 % to 18.6%). The m a x i m u m and m i n i m u m CO2 concentrations tolerated before rats left the test cage ranged f r o m 4 . 8 % to 25.3%>. V a r i a b i l i t y was also observed for ind i v idua l rats; for example , one rat left the test cage at 4 . 8 % CO2 on one day and at 21.5%) CO2 on a different day. 90 4.4 Discussion It has been suggested anecdotal ly that s l o w CO2 f i l l rates can result i n loss o f consciousness i n rats before aversive CO2 concentrations occur . H o w e v e r , rats i n the current study left the test cage before l os ing consciousness for a l l test sessions w i t h CO2. Th i s result demonstrates that rats are averse to g radua l - f i l l CO2 exposure w i t h flow rates ranging f r o m 3 % to 2 7 % o f the test cage v o l u m e per minute. F l o w rate had no effect o n any variables dur ing test sessions w i t h air , ind icat ing that it was CO2 exposure that resulted i n avers ion rather than sound or air currents associated w i t h changes i n gas flow dynamics . In the current study, the latency to leave the test cage decreased w i t h increasing flow rate, such that o n average rats left the test cage w h e n CO2 concentrations were between 1 3 . 0 % and 1 5 . 9 % . In a prev ious study e x a m i n i n g avers ion to CO2, rats showed avers ion to static CO2 at concentrations o f 1 5 % , and to a gradual ly increasing concentrat ion o f CO2 at approx imately 18%> (Chapter 3). These results indicate that there is a threshold CO2 concentrat ion that rats find aversive, and that it is re lat ively consistent regardless o f flow rate. Th i s concentrat ion o f CO2 is u n l i k e l y to cause pa in i n rats. The major i ty o f receptors i n the nasal m u c o s a respond to CO2 concentrations between 37 and 5 0 % CO2 ( A n t o n et a l . , 1991; Peppe l & A n t o n , 1993). Furthermore, pa in fu l s t imulat ion o f the nasal m u c o s a is k n o w n to e l ic i t apnea and bradycardia , and this response is not observed i n rats at CO2 concentrations ranging f r o m 10%> to 5 0 % ( Y a v a r i et a l . , 1996). H o w e v e r , CO2 concentrations as l o w as 8 % have been associated w i t h a sensation o f dyspnea, or breathlessness, i n humans (Dr ipps & C o m r o e , 1947; L i o t t i et a l . , 2001) , and this sensation m a y occur i n rats. W e therefore suggest that dyspnea is more l i k e l y to be the cause o f avers ion i n the current study. The CO2 concentrat ion at leav ing t ime var ied w i t h flow rate, w i t h rats tolerat ing s l ight ly higher CO2 concentrations at intermediate flow rates. A t l o w flow rates, rats are l i k e l y leav ing at 91 lower CO2 concentrations because the extended per iod o f exposure to lower concentrations reduces their overa l l tolerance for CO2. Sensations o f dyspnea due to hypercapnia are l i k e l y mediated by central and per ipheral chemoreceptors ( A m e r i c a n Thorac ic Soc iety , 1999), w h i c h are sensit ive to reductions i n the p H o f b l o o d and cerebral spinal f l u i d . Ex tended exposure w o u l d a l l o w greater t ime for these adjustments, such that the m a x i m u m tolerance is reached at a lower concentrat ion. R e d u c e d tolerance at h igh f l o w rates indicates that CO2 detect ion mechanisms might be sensit ive to not on ly absolute CO2 concentrat ion, but also to the rate at w h i c h CO2 is increasing. O n l y f l o w rates less than 3 0 % per minute were examined i n the current study because the results o f A m b r o s e et a l . (2000) suggest that h igh f l o w rates result i n CO2 concentrations that are suff ic ient to cause pa in before unconsciousness i n m i c e . A m b r o s e et a l . (2000) found that a f l o w rate o f 6 0 % o f the chamber v o l u m e per minute resulted i n CO2 concentrations above 50%) i n the 10 s or so before m i c e lost consciousness, w h i l e at 30% per minute m i c e became unconsc ious at CO2 concentrations under 50%>. Thus , a l though faster f l o w rates result i n a shorter durat ion o f exposure before loss o f consciousness, s lower f l o w rates m a y prevent exposure to CO2 levels that are suff ic ient to cause pa in . W h i l e f l o w rates greater than 30%) per minute were not examined , the CO2 concentrat ion at w h i c h rats left the test cage var ied i n a parabol ic manner , suggesting that rats w o u l d leave at lower concentrations w i t h - f l o w rates above those tested i n the current study. The approach-avoidance test used i n the current study indicates that the m a x i m u m CO2 concentrat ion tolerated varies w i t h f l o w rate, but this study prov ides l i tt le in format ion on the leve l o f distress that rats w o u l d have exper ienced had they been forced to remain i n the test cage unt i l death. Th i s distress w o u l d be dependent both on the durat ion o f the per iod between onset o f avers ion and unconsciousness, and the strength o f avers ion to the CO2 concentrations occur r ing dur ing this per iod . Prev ious studies have attempted to examine behavioural responses 92 dur ing the entire euthanasia procedure as a means o f assessing distress. W h i l e some studies have found evidence to suggest a distress response (Britt , 1987; C o e n e n et a l . , 1995; Chapter 2) , others have reported no effect (Hackbarth et a l . , 2 0 0 0 ; Hornett & Haynes , 1984; S m i t h & Har rap , 1997). S o m e o f these studies have spec i f ica l l y compared behav ioura l responses dur ing g radua l - f i l l CO2 exposure at different flow rates. Hornett and Haynes (1984) examined flow rates ranging f r o m 6 to 4 0 % per minute w i t h rats and w h i l e adverse reactions were not reported for any flow rate, a rate o f 1 9 . 5 % per minute was recommended based o n a subjective assessment o f the procedure and t ime to unconsciousness and death. T i m e to unconsciousness was approx imate ly 4 m i n at 6 % per minute , but was reduced to approx imately 2 m i n for flow rates o f 13 to 40%» per minute . C o e n e n et a l . (1995) compared flow rates o f 1 4 % and 1 2 5 % per minute and found that both treatments resulted a s imi la r per iod o f 'exc i tat ion and agi tat ion ' , but that gasping was s l ight ly h igher i n the fast fill group, suggesting increased dyspnea. H o w e v e r , the t ime to recumbency , aberrant E E G and abnormal E C G were s ign i f icant ly longer in the 1 4 % per minute group. In conc lus ion , rats show avoidance dur ing exposure to g radua l - f i l l CO2 exposure w i t h flow rates ranging f r o m 3 % to 2 7 % per minute. The CO2 concentrat ion at the t ime rats left the test cage var ied w i t h flow rate, ind icat ing that a flow rate o f 1 4 % per minute is opt imal in terms o f in i t ia l avers ion. H o w e v e r , forced exposure to CO2 beyond this in i t ia l avers ion is l i k e l y to result i n distress w i t h a l l flow rates, therefore further research is needed to develop better methods o f euthanasia for laboratory rats. 93 a) b) c) 8 -. ci c te n 6 - CO CD m s 4 - CD ~o 2 - CD $ CD 0 - 100 -i 80 - „—s (/} >. 60 - o c a> 40 - ro _ i 20 - 0 - 20 - i ^ 16 - CD "O 12 -X o T3 C 8 - o ro 4 - O 0 - 10 15 20 Flow rate (%/min) 10 15 20 Flow rate (%/min) 10 15 20 Flow rate (%/min) 25 30 25 30 25 30 F i g u r e 4 . 1 . Least squares mean ( ± S E M ) a) number o f reward i tems eaten, b) latency to stop eating (open) and to leave the test cage ( f i l led) , and c) CO2 concentrat ion at the t ime w h e n rats stopped eating (open) and left the test cage ( f i l led) dur ing test sessions w i t h CO2 f l o w rates o f 3 , 7, 14, 20 and 2 7 % o f the test cage v o l u m e per minute (n = 8 rats). 94 4.5 References A m b r o s e , N . , W a d h a m , J . , M o r t o n , D. , 2000 . Ref inement i n Euthanasia . In: B a l l s , M . , van Ze l le r , A . M . , Ha ider , M . E . (Eds) , Progress i n the R e d u c t i o n , Ref inement and Replacement o f A n i m a l Exper imentat ion , E l sev ie r Sc ience , A m s t e r d a m , p p . 1 1 5 9 - 1 1 6 9 . A m e r i c a n Thorac ic Society . 1999. Dyspnea : mechan isms , assessment, and management; a consensus statement. A m . J . Resp i r . Cr i t . Care M e d . 159, 3 2 1 - 3 4 0 A n t o n , F., Euchner , I., Handwerker , H .O . , 1992. P s y c h o p h y s i c a l examinat ion o f pa in induced by def ined CO2 pulses appl ied to the nasal mucosa . P a i n 4 9 , 5 3 - 6 0 . A n t o n , F., P e p p e l , P. , Euchner , I., Handwerker , H .O . , 1991. Cont ro l led nox ious chemica l s t imulat ion : responses o f rat t r igeminal brainstem neurones to CO2 pulses appl ied to the nasal mucosa . N e u r o s c i . Lett . 123, 2 0 8 - 2 1 1 . Br i t t , D . P . , 1987. The humaneness o f carbon d iox ide as an agent o f euthanasia for laboratory rodents. In: Euthanasia o f U n w a n t e d , Injured or D iseased A n i m a l s or for Educat iona l or Sc ient i f i c Purposes, pp. 1 9 - 3 1 . Potter 's B a r : Un ivers i t ies Federat ion for A n i m a l Wel fa re . C h e n , X . , Ga l la r , J . , P o z o , M . A . , B a e z a , M . , B e l m o n t e , C , 1995. CO2 s t imulat ion o f the cornea: a compar i son between human sensation and nerve act iv i ty i n p o l y m o d a l nocicept ive afferents o f the cat. Eur . J . N e u r o s c i . 7, 1 1 5 4 - 1 1 6 3 . C o e n e n , A . M . , D r inkenburg , W . H . , Hoenderken , R. , van Lui j te laar , G . L . , 1995. C a r b o n d iox ide euthanasia i n rats: oxygen supplementat ion m i n i m i z e s signs o f agitat ion and asphyx ia . L a b . A n i m . 2 9 , 2 6 2 - 2 6 8 . D r i p p s , R . D . , C o m r o e , J . H . , 1947. The respiratory and c i rculatory response o f normal m a n to inhalat ion o f 7.6 and 10.4 per cent CO2 w i t h a compar i son o f the m a x i m a l vent i lat ion produced by severe muscular exercise, inhalat ion o f CO2 and m a x i m a l voluntary hypervent i lat ion. A m . J . P h y s i o l . 149, 4 3 - 5 1 . 95 F e n g , Y . , S i m p s o n , T . L. , 2 0 0 3 . N o c i c e p t i v e sensation and sensit iv i ty evoked f r o m human cornea and conjunct iva st imulated by CO2. Invest. Ophth . V i s . S c i . 4 4 , 5 2 9 - 5 3 2 . Hackbar th , H . , K u p p e r s , N . , Bohnet , W . , 2000 . Euthanasia o f rats w i t h carbon d i o x i d e — a n i m a l wel fare aspects. L a b . A n i m . 34 , 9 1 - 9 6 . H i ra ta , H . , H u , J . W . , Bereiter , D . A . , 1999. Responses o f medul la ry dorsal horn neurons to corneal s t imulat ion by CO2 pulses i n the rat. J . N e u r o p h y s i o l . 8 2 , 2092 - 2107 . Hornett , T . D . , Haynes , A . R . , 1984. C o m p a r i s o n o f carbon dioxide/air mixture and nitrogen/air mixture for the euthanasia o f rodents. D e s i g n o f a system for inhalat ion euthanasia. A n i m a l Techno logy 3 5 , 9 3 - 9 9 . L i o t t i , M . , B rannan , S . , E g a n , G . , Shade, R., M a d d e n , L. , A b p l a n a l p , B . , R o b i l l a r d , R. , Lancaster , J . , Zamar r ipa , F .E . , F o x , P.T. , Denton , D. , 2 0 0 1 . B r a i n responses associated w i t h consciousness o f breathlessness (air hunger). P roc . Nat . A c a d . S c i . 98 , 2 0 3 5 - 2 0 4 0 . P e p p e l , P. , A n t o n , F., 1993. Responses o f rat medul la ry dorsal horn neurons f o l l o w i n g intranasal nox ious c h e m i c a l s t imulat ion: effects o f st imulus intensity, durat ion, and interst imulus interval . J . N e u r o p h y s i o l . 7 0 , 2 2 6 0 - 2 2 7 5 . S m i t h , W . , Harrap , S . B . , 1997. Behav ioura l and cardiovascular responses o f rats to euthanasia us ing carbon d iox ide gas. L a b . A n i m . 3 1 , 3 3 7 - 3 4 6 . Thurauf , N . , Gunther , M . , P a u l i , E. , K o b a l , G . , 2 0 0 2 . Sensi t iv i ty o f the negative mucosa l potential to the t r igeminal target st imulus CO2. B r a i n Res . 9 4 2 , 2 7 - 8 6 . Y a v a r i , P. , M c C u l l o c h , P .F . , Panneton, W . M . , 1996. T r i g e m i n a l l y - m e d i a t e d alteration o f cardiorespiratory rhythms dur ing nasal appl icat ion o f carbon d iox ide i n the rat. J . A u t o n . N e r v . Syst. 6 1 , 195 -200 . 96 CHAPTER 5: Effects of novelty on rat responses to C 0 2 exposure5 5.1 Introduction C a r b o n d iox ide is w i d e l y used for k i l l i n g laboratory rodents, but recent ev idence suggests that exposure to CO2 m a y cause distress and avers ion before loss o f consciousness i n rats. Exposure o f rats to a gradual ly increasing concentrat ion o f CO2 has been shown to result i n ^ escape behaviours , such as push ing and scratching at the chamber l i d , and increased explorat ion (Chapter 2) . Furthermore, L e a c h et a l . (2002) found that rats a v o i d CO2 concentrations o f 2 5 . 5 % and greater, and , as descr ibed i n Chapters 3 and 4, rats w i l l forgo a palatable food reward i n order to a v o i d CO2 concentrations o f 15%> and greater. B o t h dyspnea (an unpleasant sensation o f breathlessness) and pa in have been suggested as potential causes o f distress and avers ion dur ing CO2 euthanasia. CO2 concentrations o f greater than 30%> are necessary to cause loss o f posture i n rats (Chapter 2 ; S m i t h and Harrap, 1997), and loss o f posture indicates the approximate onset o f unconsciousness (e.g. Coenen et a l . , 1995). CO2 forms carbonic ac id w h e n it comes into contact w i t h moisture at the mucosa l membranes , and starts to cause pa in in humans at concentrations o f 30 to 5 0 % ( A n t o n et a l . , 1992; C h e n et a l . , 1995; F e n g & S i m p s o n , 2 0 0 3 ; Thurauf et a l . , 2002) . Rats have nociceptors in the nasal m u c o s a that respond to CO2 at s imi lar concentrations ( A n t o n et a l . , 1 9 9 1 ; Peppel and A n t o n , 1993), and so rats m a y experience pa in at moderate CO2 concentrations. CO2 also causes dyspnea i n humans at concentrations o f on ly 8%> (Dr ipps & C o m r o e , 1947; L io t t i et a l . , 2001) . In prev ious studies on CO2 euthanasia, rats have s h o w n behavioural signs o f distress and avers ion dur ing exposure to relat ively l o w concentrations o f CO2 (Chapters 2 , 3 and 4). These 5 A version of this chapter has been submitted for publication. Niel, L., Weary, D.M. 2006. Effects of novelty on rat responses to C 0 2 exposure. Appl. Anim. Behav. Sci. (submitted). 97 responses occurred at CO2 concentrations that were lower than noc icept ive thresholds, therefore dyspnea is a more l i k e l y cause o f distress and avers ion than pa in . H o w e v e r , another potential source o f distress dur ing CO2 exposure is novel ty . N o v e l t y has been suggested to induce an approach-avoidance conf l i c t i n rats, result ing f r o m an interact ion between exploratory mot i va t ion and fear ( M o n t g o m e r y , 1955). W a l l a c e and R o s e n (2000) demonstrated that exposure o f rats to nove l odours, such as butyr ic ac id (s imi lar to ranc id butter) and i s o a m y l acetate (s imi lar to banana), causes avoidance, reduces g rooming t ime and increases f reez ing t ime , and these responses suggest that nove l odours can el ic i t fear. Odour percept ion occurs as a result o f s t imulat ion o f both o l factory and t r igemina l neurons, w i t h the latter contr ibut ing m a i n l y to pungency ( C a i n and M u r p h y , 1980). CO2 is thought to stimulate m a i n l y t r igeminal neurons i n the nasal mucosa , but because odour percept ion occurs as a result o f both o l factory and t r igemina l input, humans perceive an odour qual i ty w h e n asked to describe sensations occur r ing w i t h CO2 inhalat ion (Ca in and M u r p h y , 1980). Rats can detect CO2 at concentrations between 0.04 and 1 . 7 % (Youngentob , 1991), w h i c h is b e l o w the levels required to st imulate t r igeminal neurons i n rat nasal m u c o s a (Peppel and A n t o n , 1993). The exact qual i ty o f CO2 that rats are responding to is u n k n o w n , but they have a sensit ive ol factory system and m a y have an enhanced ab i l i t y to detect the odour qual i ty o f CO2. In order to test whether distress and avers ion associated w i t h CO2 exposure is due i n part to novel ty , we used two different exper imental approaches. F o r the first set o f exper iments, we used an approach-avoidance test to examine gas avers ion i n rats, by pa i r ing gas exposure w i t h a food reward. In Exper iment 1 we examined rat avers ion to g radua l - f i l l CO2 exposure. Rats were tested w i t h repeated CO2 exposure to document whether their responses w o u l d show habituat ion. In Exper iment 2 we used the same methodology to examine h o w a nove l odour affects rat per formance o n the approach-avoidance test. W e used peppermint as the nove l odour , and this st imulus was not expected to produce pa in or respiratory s t imulat ion . For both 98 experiments, we compared rat responses to the first exposure and subsequent exposures to determine whether each condition was aversive and whether novelty was a source of aversion, using the following logic. We reasoned that: 1. a lack of aversion would be indicated by similar eating and dwelling times with air and the treatment gas, 2. aversion due to novelty would result in animals eating less and leaving earlier on the initial exposure, but that this reaction would decrease on subsequent exposures, and 3. aversion without an effect of novelty would result in the animals eating less and leaving earlier on all exposures. For the second experimental approach, we examined the behavioural responses of rats during exposure to a gradually increasing concentration of either CO2 or peppermint odour (Experiment 3). We predicted that behaviours that were due to novelty would be present with both treatments, but that those occurring as a result of other factors, such as pain and dyspnea, would occur only in the CO2 treatment group. 5.2 Materials and Methods 5.2.1 S u b j e c t s , H o u s i n g , a n d E q u i p m e n t Rats were obtained as surplus stock (i.e. animals already slated for euthanasia) from the U B C Rodent Breeding Unit, and housed at 21°C under a 12:12-hr light-dark cycle with ad libitum access to food (Lab Diet 5001, PMI Nutrition International, Indiana, USA) and tap water. A l l testing was conducted during the light portion of the light-dark cycle. CO2 and air were delivered from compressed gas cylinders (Praxair, Richmond, B.C.). For some treatments, air was scented with peppermint odour by routing the air flow through a 266 mL chamber containing three cotton balls soaked in 2 mL of peppermint extract (Canada Safeway Ltd., Calgary). The treatment gases were passed through a copper coil in a room temperature water bath to regulate the temperature of the gas before it entered the chamber. 99 F l o w rates o f the gases were measured by a var iable area f lowmeter ( M o d e l V S B - 6 6 - B V , D w y e r Instruments, Inc., M i c h i g a n ) , and measured CO2 f l o w rates were adjusted for density w i t h a correct ion factor o f 0 .812. G a s concentrations i n the chamber were moni tored dur ing the exper iment v i a a gas sampl ing tube us ing a M o c o n L F 7 0 0 D O2 analyzer , and the f o l l o w i n g f o r m u l a was used to calculate the concentrat ion o f CO2 at speci f ic t ime points (t = x ) dur ing the f i l l i n g process: [ C 0 2 ( , = s ) ] = 1 0 0 - ( 1 0 0 * ( [0 2 ( t = x)]/ [O2 (,-<»])). 5.2.2 Experiments 1 & 2: Approach-avoidance Testing W e used an approach-avoidance test to examine rats' avo idance o f a gradual addi t ion o f either CO2 (Exper iment 1) or air w i t h peppermint odour (Exper iment 2) . The peppermint was not expected to cause p a i n or respiratory s t imulat ion. Rats were s ing ly housed in the testing apparatus cons is t ing o f two transparent cages connected by an opaque tunnel made o f b lack , r ibbed, P V C tub ing w i t h a diameter o f 10 c m and s loped so that one cage was 27 c m higher than the other. The ' h o m e ' cage measured 48 x 38 x 20 c m , and contained f o o d , water, bedding , an opaque nestbox and a N y l a b o n e dog chew. The secondary cage measured 45 x 24 x 20 c m and contained bedding . The lower cage was a lways used for testing because the test gases were denser than air. D u r i n g p re l iminary testing o f the exper imental apparatus we determined that the test gases remained i n the lower cage and the lower por t ion o f the tunnel . F o r Exper iment 1, h a l f o f the rats were tested i n the home cage and the other h a l f were tested i n the secondary cage for the purposes o f a separate experiment (Chapter 3). F o r Exper iment 2 , a l l rats were tested i n the secondary cage. D u r i n g exper imental test ing, each an imal and its testing apparatus were transferred i n d i v i d u a l l y to a test r o o m . A t this t ime , the nest box was removed f r o m the home cage and the wi re l i d on the test cage was replaced w i t h a p lex ig las l i d that featured two air outlets posi t ioned 100 at the end closest to the tunnel , a gas inlet at the far end o f the cage, and a gas sampl ing tube inserted at the center to a depth o f 10 c m above the cage f loor . The air outlets were covered w i t h mesh to prevent the rats f r o m push ing their noses outside the test cage. The experimenter was concealed behind a b l i n d dur ing testing. The testing apparatus and O2 meter readout were v ideo recorded dur ing testing. D u r i n g both t ra in ing and exper imental sessions, rats were first l o c k e d into the upper cage for 2 m i n . The l o c k was then removed and they were able to enter the lower test cage for a food reward o f 20 H o n e y N u t C h e e r i o s ™ (General M i l l s , Inc., M innesota ) . W h e n the rat started eating the food reward , gas f l o w (air, air w i t h peppermint or CO2) into the test cage was started at a rate o f 1 7 % o f the cage v o l u m e per minute. The session ended 300 s after l ock remova l . The animals were a l l owed to exit and re-enter the test cage throughout the test per iod . A t the end o f the session, the remain ing reward i tems were removed and the rat was returned to the h o l d i n g r o o m . F o r Exper iment 1 the subjects were nine 5 -month o l d , male Wis ta r rats. In i t ia l ly they were trained for 9 days to per form the approach-avoidance task w i t h air f l o w i n g into the test cage. Th i s t ra in ing was f o l l o w e d by 17 days o f testing. The rats were first tested over five consecut ive days, w i t h air on day 1, g radua l - f i l l CO2 on days 2 , 3 , and 4, and air again o n day 5. F r o m days 6 to 15, the rats were tested two t imes w i t h air and w i t h static CO2 concentrations o f 5, 10, 15 and 2 0 % for a separate exper iment (Chapter 2) . These data are not inc luded i n the current study, but served to prov ide the rats w i t h further experience o f CO2 exposure. Rats were then tested again w i t h g radua l - f i l l CO2 and air on days 16 and 17, respect ively . F o r Exper iment 2 the subjects were seven 13 -month o l d , male Wis ta r rats. Be fo re this exper iment , they had performed i n another approach-avoidance exper iment, and had considerable experience w i t h air or CO2 f l o w i n g into the test cage. H o w e v e r they had not p rev ious ly been tested w i t h a n o v e l odour such as peppermint . The rats were tested over five 101 consecut ive days, w i t h air o n day 1, air w i t h peppermint odour o n days 2 , 3 , and 4 , and air again o n day 5. F o r Exper iments 1 and 2, we recorded the total number o f reward i tems eaten dur ing each test session, the latency to stop eating and the latency to leave the test cage after gas f l o w started. F o r Exper iment 1 we also recorded the 0 2 concentrations w h e n the rat stopped eating and w h e n it left the test cage, and used these values to calculate CO2 concentrations as descr ibed above. 5 .2 .3 B e h a v i o u r a l R e s p o n s e s : E x p e r i m e n t 3 In this exper iment we examined the behavioural responses o f rats dur ing the gradual addi t ion o f either CO2 or air w i t h peppermint odour. The exposure chamber was a 20 L po lypropy lene cage measur ing 20.5 x 45 .5 x 24 c m (Lab Products Inc.), f i tted w i t h a P lex ig las l i d . The l i d had a gas inlet centered at the end, two air outlets pos i t ioned at the opposite end, and a gas sampl ing tube inserted at the center o f the chamber to a depth o f h a l f the chamber height. The air outlets were covered w i t h mesh to prevent the rats f r o m push ing their noses outside the chamber. The back and sides o f the chamber were covered w i t h b lack paper so that the animals c o u l d not see the person conduct ing the experiment. The subjects were th i r ty - two 4 to 6 - m o n t h - o l d , male Wis ta r rats. A n i m a l s were randomly a l located to the CO2 or air w i t h peppermint odour treatment groups (n = 16 for both). The rats were ind i v idua l l y p laced into the euthanasia chamber for a 2 7 - m i n per iod o f acc l imat i zat ion , dur ing w h i c h air was added to the chamber at a rate o f 1 7 % o f the chamber v o l u m e per minute. A f t e r acc l imat i za t ion , air f l o w was stopped and either CO2 or air w i t h peppermint odour was started at a rate o f 17%> o f the chamber v o l u m e per minute . C O v t r e a t e d an imals remained i n the chamber and were moni tored unt i l death, but animals treated w i t h peppermint odour were 102 removed f r o m the chamber at the end o f the 135-s observat ion per iod . P re l iminary observations showed that C02-treated animals ceased a l l purposeful movement w i t h i n this per iod , so any relevant effects o f peppermint exposure w o u l d be present du r ing this t ime. The euthanasia chamber and O2 meter readout were v ideo recorded dur ing the exper imental procedure. E a c h a n i m a l was scored cont inuously dur ing the last 135 s o f the acc l imat i zat ion per iod (baseline) and for 135 s after gas f l o w began (exposure) for pre -def ined behaviours thought to relate to distress (Table 1). In a prev ious study these behaviours were found to increase dur ing g radua l - f i l l CO2 exposure (Chapter 2) . 5 .2.4 S t a t i s t i c a l A n a l y s e s 5.2.4.1 Experiment 1 -Approach-avoidance Testing with CO2 Dependent variables were analyzed for the first three days o f CO2 exposure w i t h a m i x e d m o d e l ( S A S v9 .1) w h i c h inc luded rat as a random effect (8 d.f), and tested for a l inear effect o f order (1 d.f.) against an error term w i t h 17 d.f. F o r those var iables where no statistical differences were found across the first three days o f CO2 exposure (al l except number o f reward i tems eaten), data were averaged w i t h i n rat and the average response was then compared to the response o n the final day o f C 0 2 testing (Day 16) w i t h a m i x e d m o d e l w h i c h inc luded rat as a random effect (8 d.f.) and examined the effect o f order (1 d.f.) against an error term w i t h 8 d.f. N u m b e r o f reward i tems eaten was compared between the third and final CO2 test sessions w i t h a s imi la r m i x e d m o d e l . 5.2.4.2 Experiment 2 - Approach-avoidance Testing with Peppermint Odour O n l y 1 o f the 7 rats ate fewer than 20 reward i tems dur ing testing w i t h air or peppermint odour, so statistical analyses were not performed w i t h this var iable . The remain ing dependent var iables were compared across the three days o f testing w i t h peppermint odour w i t h a m i x e d 103 m o d e l w h i c h inc luded rat as a random effect (6 d.f) and tested for a l inear effect o f order (1 d.f.) against an error term w i t h 13 d.f. Rats showed s imi lar responses to peppermint odour exposure over the three days o f testing, so data were averaged w i t h i n rat for the two days o f testing w i t h air and for the three days o f testing w i t h peppermint odour. Dependent variables were then compared w i t h a m i x e d m o d e l w h i c h inc luded rat as a random effect (6 d.f) and examined the effect o f gas treatment (1 d.f.) against an error term w i t h 6 d.f. 5.2.4.3 Experiment 3 - Behavioural Responses The number o f t imes the rat reared and the t ime spent w i t h the nose i n contact w i t h the chamber l i d were analyzed w i t h a m i x e d m o d e l w h i c h inc luded rat as a random effect (30 d.f.) and examined the effect o f per iod (1 d . f ) , gas treatment ( l .d . f ) and the interaction between per iod and gas (1 d.f.) against an error term w i t h 30 d.f. Escape behaviours and act iv i ty (recorded as side changes) were not observed i n a l l an imals , so the number o f animals w h i c h showed increases i n these behaviors dur ing exposure was compared between gas treatments by a G-test w i t h a W i l l i a m ' s correct ion (described i n S o k a l and R o l f , 1995). 5.3 Results 5.3.1 Experiment 1 - Approach-avoidance testing with CO2 D u r i n g approach-avoidance testing on the three contro l days w i t h air (Days 1, 5 and 17), rats consumed a l l 20 reward i tems, and, o n average (mean ± S .E . ) , they stopped eating and left the test cage after 266 ± 4 s and 288 ± 2 s, respect ively . In contrast, o n the four CO2 test days (Days 2 , 3 , 4 and 16), rats consumed an average o f o n l y 2.7 ± 0.2 reward i tems, and their latencies to stop eating and leave the test cage dropped to 30 ± 2 s and 40 ± 2 s, respect ively . 104 A c r o s s the first three days o f testing w i t h CO2, there was no change i n the latencies to stop eating or leave the test cage, or i n the CO2 concentrat ion at w h i c h rats stopped eating and left the test cage (F ig . 5.1 a, b; P > 0.1 for al l ) . H o w e v e r , o n the final day o f testing rats showed a 5 2 % increase i n latency to stop eating (Fi>8 = 18.99, P < 0 .005) and a 25%> increase i n latency to leave the test cage (F\$ = 7 .75, P < 0.05), result ing i n higher CO2 concentrations at these t ime points (stop eating: F ) ; 8 = 3 1 . 9 , P < 0 . 0 0 1 ; leave test cage: Fh% = 15.2 , P < 0.01). The number o f reward i tems eaten showed a l inear increase over the first three days o f testing w i t h CO2 ( F ig . 5.1 c; F\t\i = 5 .45, P < 0.05). The number o f reward items eaten also increased f r o m the th i rd to the final test session (F\# = 2 5 . 5 4 , P < 0 .001) . 5.3.2 Experiment 2 - Approach-avoidance testing with peppermint odour R a t s ' performance o n the approach-avoidance task was s imi la r o n the two control days w i t h air (Days 1 and 5), and on the three days o f testing w i t h peppermint odour (Days 2 , 3 and 4). D u r i n g testing w i t h both air and peppermint odour, s ix o f the seven rats consumed a l l 20 food reward i tems o n each test day. The seventh rat ate for the entire test per iod each day, but on ly consumed 17 to 20 of . the reward i tems due to a s l o w eating rate. In compar ison to air exposure, peppermint exposure d i d not affect the latency for rats to stop eating (232 vs. 235 ± 14 s; Fh6 = 0 .66 , P > 0.1) or leave the test cage (274 vs. 283 ± 6 s; F\fi = 2 .06 , P > 0.1). R a t s ' performance was also consistent across the three days o f testing w i t h peppermint odour. T h e y consumed a s imi la r number o f reward i tems, and there was no difference i n their latency to stop eating ( F U 3 = 0 . 3 3 , P > 0.1) or to leave the test cage ( F U 3 = 0 .44, P > 0.1). 5.3.3 Experiment 3 - Behavioural responses D u r i n g the 135-s basel ine per iod , rats f r o m both the CO2 and the peppermint odour treatment groups reared about four t imes and spent about 12 to 14 s w i t h the nose touch ing the 105 chamber l i d (F ig . 5.2). A f t e r in i t iat ion o f either CO2 or peppermint odour, rats showed increases i n both rearing ( F i , 3 o = 14 .13 , P < 0 .001) , and t ime spent w i t h the nose touching the chamber l i d (^ i ,30 = 8 . 1 1 , P < 0.01). W h i l e the increase i n response for these behaviours was numer ica l l y larger w i t h CO2 than w i t h peppermint odour, we found neither an effect o f gas treatment nor an interact ion between per iod and gas treatment ( P > 0.1). D u r i n g exposure to either CO2 or peppermint odour , less than h a l f o f the rats showed increases i n act iv i ty i n compar ison to basel ine. Furthermore, the number o f rats that showed an increase d i d not d i f fer between the CO2 (6 o f 16 rats) and peppermint odour (7 o f 16 rats) treatment groups ( G = 0 .12 , P > 0.1). D u r i n g the basel ine per iod , escape behaviours were o n l y per formed by one rat. T h i s rat was f r o m the peppermint odour treatment group, and it d i d not per fo rm escape behaviours dur ing exposure to peppermint odour. Increases i n escape behaviours were observed i n 1 o f 16 rats dur ing peppermint exposure, but were observed i n 10 o f 16 rats dur ing CO2 exposure (G = 11.89, P < 0 .001) . The number o f escape behaviours per formed by rats dur ing CO2 exposure ranged f r o m 1 to 2 1 . 5.4 Discussion Prev ious studies have found that w h e n rats are exposed to nove l s t imul i such as nove l envi ronments ( D u b o v i c k y et a l . , 1999; M o n t g o m e r y , 1955) and objects (Zangrossi and F i l e , 1994), they show signs o f habituat ion by the second exposure. In contrast, the rats i n Exper iment 1 d i d not show any reduct ion i n their avoidance o f g radua l - f i l l CO2 exposure over the first three days o f approach-avoidance testing. They ate faster o n the second and th i rd days o f testing, but showed s imi la r latencies to stop eating and leave the test cage. Because the response to CO2 d i f fered f r o m that observed w i t h other s imp le , nove l s t i m u l i , it is un l ike l y that 106 the rats' response was due to novelty . Th is conc lus ion is supported by the results o f Exper iment 2 , i n w h i c h nove l peppermint odour had no effect o n rats' performance dur ing approach- avoidance testing. Hence , exposure to a nove l odour is not suff ic ient to deter rats f r o m this type o f task. Furthermore, by the final day o f approach-avoidance testing w i t h CO2, rats were st i l l s h o w i n g consistent avers ion to CO2 concentrations b e l o w those needed to cause unconsciousness. Th i s result indicates that rats do not habituate to g radua l - f i l l CO2 exposure, and suggests that avers ion to CO2 is due m a i n l y to factors other than novelty . Rats i n the current study in i t ia l l y left the test cage w h e n CO2 concentrations reached 1 4 % , but this increased to 18%) o n the final day o f testing. In prev ious approach-avoidance studies o n CO2 avers ion i n rats, the rats were tested for avers ion to CO2 after be ing fami l ia r i zed w i t h CO2 exposure (Chapters 3 and 4). The current results suggest that this l i k e l y resulted i n a modest underest imation o f the CO2 concentrations that rats find aversive o n in i t ia l exposure. One potential explanat ion for this increased tolerance for CO2 w i t h repeated exposure is that the rats learned to tolerate the unpleasant sensations associated w i t h CO2 exposure, or that they learned strategies, such as breatholding, that a l l o w e d them to remain i n the test cage for longer. A l te rnat i ve l y , the rats m a y have developed an increased p h y s i o l o g i c a l tolerance for CO2. Prev ious studies have found that chronic exposure to elevated CO2 can result i n a reduced vent i latory response to hypercapnia through acid -base adjustments ( L a i et a l . , 1981) or alterations i n chemoreceptor act iv i ty ( M i t c h e l l and Johnson, 2003) . C h r o n i c exposure to l o w levels o f CO2 (<3%) has also been shown to increase the level o f hypercapnia needed to cause dyspnea dur ing acute CO2 exposure i n humans ( B l o c h - S a l i s b u r y et a l . , 1996). H o w e v e r , in compar ison to other studies in w h i c h this increased tolerance has been demonstrated, rats in the current study were exposed to CO2 for on ly short periods ( less than 5 m i n per day) , whereas changes i n dyspnea tolerance i n humans developed on ly after mul t ip le days o f chronic CO2 exposure ( B l o c h - S a l i s b u r y et a l . , 1996). / 107 O n l y one study has looked at the effect o f short, da i l y CO2 exposures o n venti latory responses to an acute CO2 chal lenge. Waters and T i n w o r t h (2001) examined changes i n the vent i latory responses o f piglets to a chal lenge w i t h an inspi red gas mixture o f 9 % 0 2 and 6 % CO2 after seven days o f acute, c y c l i c exposure to this same gas mixture . C y c l i c exposure consisted o f one 4 8 - m i n session per day i n w h i c h fresh air and exposure to the gas mixture were alternated every 4 m i n . F o l l o w i n g seven days o f c y c l i c exposure, the piglets showed a smal ler increase i n breathing rate and v o l u m e i n response to the chal lenge i n compar ison to controls. H o w e v e r , it is not k n o w n whether the re lat ively short per iods o f CO2 exposure that occurred dur ing the current study were suff ic ient to cause such changes, and whether these changes w o u l d have altered the rats' avers ion to CO2 exposure. B e h a v i o u r a l changes o f rats dur ing forced exposure to an unfami l ia r s t imulus , such as peppermint odour, might be due to novel ty , w h i c h w o u l d decrease w i t h repeated exposure, or due to intr ins ic avers ion to the st imulus , w h i c h w o u l d remain constant w i t h repeated exposure. W h i l e humans do not general ly f i n d peppermint odour int r ins ica l ly unpleasant, no previous studies have examined avers ion to peppermint odour i n rats. In E x p e r i m e n t 2 we found that rats showed no s ign o f avers ion to peppermint odour at the concentrations used i n the current study, even o n the first exposure. Th i s result suggests that rats' avers ion to this st imulus was less than their mot i va t ion to obtain the reward i tems. N o t on ly d i d rats consume a s imi la r number o f reward i tems i n a s imi la r amount o f t ime w i t h both air and peppermint odour, but dur ing peppermint odour exposure they remained i n the test cage for an average o f 48 s after they stopped eating. The rats' w i l l ingness to remain i n the test cage even after the reward was f in ished suggests that rats do not f ind peppermint odour aversive, and that any behavioural responses observed dur ing forced exposure were not due to avers ion. In a prev ious study we found that exposure to a gradual ly increasing concentrat ion o f CO2 caused an increase i n both exploratory and escape behaviours i n rats (Chapter 2) . One o f 108 the a ims o f the current study was to determine whether these responses were due to novel ty or other aversive properties o f CO2. In Exper iment 3 , exposure to either CO2 or a nove l odour in gradual ly increasing concentrations resulted i n increased act iv i ty , rear ing, and t ime spent w i t h the nose touch ing the chamber l i d . H o w e v e r , more rats per formed escape behaviours dur ing CO2 exposure than dur ing exposure to peppermint odour. Th i s indicates that exploratory behaviours dur ing g radua l - f i l l CO2 exposure cou ld be due to novel ty , but that escape behaviours are m a i n l y due to other properties o f CO2. Exp loratory behaviours are d i f f i cu l t to interpret i n terms o f an imal distress, but escape behaviours presumably indicate that the an imal w o u l d exit the exposure chamber i f g iven the opportunity. These results therefore suggest that distress dur ing g radua l - f i l l CO2 exposure is not due to novelty . H o w e v e r , one a n i m a l d i d per form escape behaviours i n response to peppermint odour exposure, so exposure to a nove l st imulus m a y also contr ibute to distress dur ing CO2 exposure. Prev ious studies have found that f reez ing i n response to a n o v e l odour depends o n the odour that is used and its intensity (Wa l lace and R o s e n , 2000) , so w e cannot rule out the poss ib i l i t y that another odour st imulus might e l ic i t a s imi la r behavioural response to that seen dur ing CO2 exposure. O f the three potential causes o f distress and avers ion dur ing CO2 exposure - novel ty , p a i n and dyspnea - the current study indicates that nove l ty l i k e l y contr ibutes to the responses o f rats dur ing g radua l - f i l l CO2 exposure, but it is not the m a i n cause o f distress and aversion. M o r e o v e r , prev ious studies have found that behaviora l responses and avers ion to a gradual ly increasing concentrat ion o f CO2 occur at concentrations that are b e l o w the threshold for pa in at the eyes and the nasal m u c o s a (Chapters 2 , 3 and 4). Therefore, the most l i k e l y cause o f distress and avers ion dur ing g radua l - f i l l CO2 exposure is dyspnea, w h i c h occurs i n humans at CO2 concentrations o f o n l y 8 % (Dr ipps & C o m r o e , 1947; L i o t t i et a l . , 2001) . Further research is needed to determine whether dyspnea does occur i n rats i n response to CO2 exposure, and i f so whether it is poss ib le to mit igate sensations o f dyspnea dur ing CO2 euthanasia. 109 Table 5;1. Descr ipt ions o f rat behaviours recorded dur ing basel ine and dur ing exposure to CO2 or peppermint odour (Exper iment 3). B e h a v i o u r Descr ip t ion A c t i v i t y M o v e m e n t that results i n the back feet cross ing a l ine that d iv ides the length o f the chamber i n h a l f (event). Rear R a i s i n g the upper body w h i l e standing on the back feet. Includes w a l l c l i m b i n g . C l i m b i n g o n the air sampl ing tube w h i l e c h e w i n g it and rear ing dur ing g r o o m i n g were exc luded (event). N o s e to l i d T i m e spent w i t h the nose i n contact w i t h the chamber l i d (state). Escape behaviours : Scratch at l i d A rapid movement o f the front p a w f r o m the l i d through at least a 90° d o w n w a r d angle (event). P u s h at l i d A push at the chamber l i d us ing the nose or front p a w ev idenced by body and l i d movement (event). 110 a) 50 40 30 20 10 0 20 n 15 M 10 - 2 5 • Leave • Stop eating final • Leave • Stop eating final 2 3 Exposure number final Figure 5.1. A p p r o a c h - a v o i d a n c e responses to CO2 for the first three days and for the f ina l day (Day 16) o f exposure (Exper iment 1). M e a n ( ± S E M ) (a) latency to stop eating and leave the test cage, (b) CO2 concentrat ion w h e n rats stopped eating and left the test cage, and (c) number o f reward i tems eaten (n = 9 rats). I l l F i g u r e 5 .2 . Behav iou ra l responses o f rats to CO2 euthanasia and peppermint odour exposure (Exper iment 3). Least squares mean ( ± S E M ) (a) number o f rears, and (b) t ime spent w i t h the nose i n contact w i t h the test cage l i d dur ing basel ine and dur ing exposure to either CO2 (n = 16 rats) or air w i t h peppermint odour (n = 16 rats). 112 5.5 References A n t o n , F., Euchner , I., Handwerker , H .O . , 1992. Psychophys i ca l examinat ion o f pa in induced by def ined CO2 pulses appl ied to the nasal mucosa . P a i n 4 9 , 5 3 - 6 0 . A n t o n , F., P e p p e l , P. , Euchner , I., Handwerker , H .O . , 1991. Cont ro l l ed nox ious chemica l s t imulat ion : responses o f rat t r igeminal brainstem neurones to CO2 pulses appl ied to the nasal mucosa . N e u r o s c i . Lett . 123, 2 0 8 - 2 1 1 . B l o c h - S a l i s b u r y , E. , Shea, S . A . , B r o w n , R., E v a n s , K . , Banzett , R . B . , 1996. A i r hunger induced by acute increase i n Pco2 adapts to chronic e levat ion o f Pco2 i n venti lated humans. J . A p p l . P h y s i o l . 8 1 , 9 4 9 - 9 5 6 . C a i n , W . S . , M u r p h y , C L . 1980. Interaction between chemorecept ive modal i t ies o f odour and i rr i tat ion. Nature . 2 8 4 , 255 - 257 . C h e n , X . , Ga l la r , J . , P o z o , M . A . , B a e z a , M . , B e l m o n t e , C , 1995. CO2 s t imulat ion o f the cornea: a compar ison between human sensation and nerve act iv i ty i n p o l y m o d a l nocicept ive afferents o f the cat. Eur . J . N e u r o s c i . 7, 1 1 5 4 - 1 1 6 3 . C o e n e n , A . M . , D r inkenburg , W . H . , Hoenderken , R. , v a n Lui j te laar , G . L . , 1995. Carbon d iox ide euthanasia i n rats: oxygen supplementat ion m i n i m i z e s signs o f agitat ion and asphyx ia . L a b . A n i m . 2 9 , 2 6 2 - 2 6 8 . D r i p p s , R . D . , C o m r o e , J . H . , 1947. The respiratory and c i rculatory response o f normal m a n to inhalat ion o f 7.6 and 10.4 per cent CO2 w i t h a compar i son o f the m a x i m a l vent i lat ion produced by severe muscular exercise, inhalat ion o f CO2 and m a x i m a l voluntary hypervent i lat ion. A m . J . P h y s i o l . 149, 4 3 - 5 1 . D u b o v i c k y , M . , Sku l tetyova , I., Jezova , D. , 1999. Neotnata l stress alters habituat ion o f exploratory behaviour i n adult male but not female rats. P h a r m a c o l . B i o c h e m . B e . 64 , 6 8 1 - 686. 113 F e n g , Y . , S i m p s o m , T. L. , 2 0 0 3 . N o c i c e p t i v e sensation and sensit iv i ty evoked from human cornea and conjunct iva st imulated by CO2. Invest. Ophth . V i s . S c i . 4 4 , 5 2 9 - 5 3 2 . L a i , Y . L . , L a m m , W . J . E , H i ldebrandt , J . , 1981. Vent i la t ion dur ing pro longed hypercapnia i n the rat. J . A p p l . P h y s i o l . 5 1 , 7 8 - 8 3 . L e a c h , M . C . , B o w e l l , V . A . , A l l a n , T . F . , M o r t o n , D . B . , 2 0 0 2 . A v e r s i o n to gaseous euthanasia agents i n rats and mice . Comparat i ve M e d . 5 2 , 2 4 9 - 2 5 7 . L i o t t i , M . , B rannan , S . , E g a n , G . , Shade, R., M a d d e n , L. , A b p l a n a l p , B . , R o b i l l a r d , R. , Lancaster , J . , Zamar r ipa , F .E . , F o x , P.T. , Denton , D. , 2 0 0 1 . B r a i n responses associated w i t h consciousness o f breathlessness (air hunger). Proc . N a t . A c a d . S c i . 9 8 , 2 0 3 5 - 2 0 4 0 . M i t c h e l l , G . S . , Johnson, S . M . , 2 0 0 3 . Neurop last ic i t y i n respiratory motor contro l . J . A p p l . P h y s i o l . 94 , 3 5 8 - 3 7 4 . M o n t g o m e r y , K . C . 1955. The relat ion between fear induced by nove l s t imulat ion and exploratory behavior . J . C o m p . P h y s i o l . P s y c h o l . 4 8 , 2 5 4 - 2 6 0 . P e p p e l , P. , A n t o n , F., 1993. Responses o f rat medul lary dorsal horn neurons f o l l o w i n g intranasal n o x i o u s c h e m i c a l s t imulat ion : effects o f s t imulus intensity, durat ion, and interst imulus interval . J . N e u r o p h y s i o l . 70 , 2 2 6 0 - 2 2 7 5 . S o k a l , R . R . , R o h l f , F .J . , 1995. B i o m e t r y : the pr inc ip les and practice o f statistics i n b io log ica l research, F reeman, N e w Y o r k , p p . 7 2 8 - 7 3 2 . S m i t h , W . , Harrap , S . B . , 1997. Behav iou ra l and cardiovascular responses o f rats to euthanasia us ing carbon d iox ide gas. L a b . A n i m . 3 1 , 3 3 7 - 3 4 6 . Thurauf , N . , Gunther , M . , P a u l i , E. , K o b a l , G . , 2002. Sensi t iv i ty o f the negative mucosa l potential to the t r igeminal target st imulus CO2. B r a i n Res . 9 4 2 , 2 7 - 8 6 . W a l l a c e , K . J . , R o s e n , J . B . 2000. Predator odor as an uncondi t ioned fear st imulus i n rats: e l ic i tat ion o f f reez ing by t r imethy l th iazo l ine , a component o f fox feces. Behav . N e u r o s c i . 1 1 4 , 9 1 2 - 9 2 2 . 114 Waters , K . A . , T i n w o r t h , K . D . , 2 0 0 1 . Depress ion o f vent i latory responses after da i l y c y c l i c hypercapnic h y p o x i a i n piglets. J . A p p l . P h y s i o l . 9 0 , 1 0 6 5 - 1 0 7 3 . Youngentob , S . L . , H o r n u n g , D . E . , M o z e l l , M . M . , 1991. Determinat ion o f carbon d iox ide detection thresholds i n trained rats. P h y s i o l . .Behav . 4 9 , 2 1 - 2 6 . Zangross i , H . , F i l e , S . E . , 1992. Behav io ra l consequences i n a n i m a l tests o f anxiety and exp lorat ion o f exposure to cat odor. B r a i n Res . B u l l . 2 9 , 3 8 1 - 3 8 8 . 115 CHAPTER 6: General Discussion The term euthanasia refers to a good death, w h i c h imp l ies a lack o f distress. H o w e v e r , due to the potential for CO2 to cause pa in and dyspnea, there has been considerable debate as to whether CO2 euthanasia can produce death wi thout distress. The two m a i n objectives o f m y thesis were: 1) to determine whether g radua l - f i l l CO2 euthanasia causes distress i n laboratory rats by e x a m i n i n g behavioural responses dur ing euthanasia, and avers ion dur ing approach- avoidance testing, and 2) to determine whether p a i n , dyspnea and nove l ty are l i k e l y sources o f distress dur ing this procedure. 6.1 Distress in rats during C 0 2 euthanasia A s discussed i n Chapter 1, the conc lus ions o f prev ious research o n distress in rats dur ing CO2 euthanasia have been h igh ly var iable . Some studies have found behavioural responses to CO2 exposure that m a y be indicat ive o f distress (Br i t t , 1986; C o e n e n et a l . , 1995; Iwarsson and Rehbinder , 1993), but others have not noted these effects ( B l a c k s h a w et a l . , 1988; Hackbar th et a l . , 2 0 0 0 ; Hornett and Haynes , 1984; S m i t h and Harrap , 1997). In a more recent study, L e a c h et a l . (2002) found that CO2 causes avoidance i n rats, suggesting that CO2 exposure is aversive. The results o f m y thesis b u i l d o n this prev ious research, and prov ide further evidence that g radua l - f i l l CO2 exposure does cause distress i n laboratory rats. Chapters 2 and 5 demonstrate that two w i d e l y used strains o f laboratory rats, Wis tars and Sprague -Dawleys , exhib i t behaviours that are indicat ive o f distress dur ing g radua l - f i l l CO2 euthanasia. D u r i n g CO2 exposure, both strains showed evidence o f increased explorat ion, i n c l u d i n g increased act iv i ty , rear ing, and t ime spent w i t h the nose touch ing the chamber l i d . M o r e important ly , rats i n both studies showed scratching and push ing at the chamber l i d , behaviours that suggest the rats were t ry ing to escape f r o m the chamber. Increased exploratory 116 behaviour has also been observed i n some prev ious studies (e.g. Br i t t , 1986), and w h i l e these behaviours indicate increased arousal they do not necessari ly suggest distress. H o w e v e r , escape behaviours strongly suggest that the animals w o u l d avo id CO2 exposure i f g iven the opportunity , and that forced CO2 exposure causes some leve l o f distress. Chapters 3 , 4 , and 5 demonstrate that rats w i l l avo id CO2 concentrations necessary to cause unconsciousness, even w h e n this requires that they g ive up a valuable food reward. In Chapter 3 , I found that rats tolerated extended exposure to static CO2 concentrations o f 5 and 1 0 % , but that the latency to leave the test cage dropped dramat ica l ly at 1 5 % CO2. Rats left the test cage at about this same concentrat ion w h e n exposed to a g radua l - f i l l procedure. In Chapter 4 , I found that the f l o w rate used dur ing the g radua l - f i l l procedure had on ly a smal l effect o n avo idance ; rats left the test cage at CO2 concentrations between 13 and 1 6 % w i t h f l o w rates ranging f r o m 3 to 2 7 % o f the test cage v o l u m e per minute . In Chapter 5, I found that rats are averse to CO2 regardless o f habituat ion, al though tolerance does increase s l ight ly w i t h repeated exposure. Th i s increased tolerance suggests that the aversive concentrations reported i n Chapters 3 and 4 w o u l d l i k e l y have been about 4 % lower on in i t ia l contact. Together, these results indicate that rats are averse to CO2 concentrations greater than approx imately 15%> regardless o f h o w CO2 is del ivered. F r o m these results, I conc lude that forced exposure to higher concentrations l i k e l y results i n distress. 6.2 Sources of distress In Chapter 1 ,1 ident i f ied three potential sources o f distress dur ing CO2 euthanasia: pa in , dyspnea and novelty . The occurrence o f pa in and dyspnea were not assessed direct ly i n the current thesis. H o w e v e r , I determined the CO2 concentrations that e l ic i ted behavioural responses and avers ion i n Chapters 2 , 3 , and 4, and compared these data w i t h prev ious research on 117 nociceptor act ivat ion i n rats, and pa in and dyspnea i n humans. The potential for CO2 to e l ic i t distress as a result o f novel ty was examined direct ly i n Chapter 5. 6.2.1 Pain A s discussed i n Chapter 1, CO2 is k n o w n to cause pa in at the nasal m u c o s a and cornea i n humans at concentrations o f 30 to 5 0 % , and this concentrat ion range is s imi la r to that required to st imulate nociceptors in rat nasal m u c o s a ( A n t o n et a l . , 1992; C h e n et a l . , 1995; D a n n e m a n et a l . , 1997; F e n g and S i m p s o n , 2 0 0 3 ; Peppe l and A n t o n , 1993; T h u r a u f et a l . , 2002) . In Chapter 2 , I found that rats started to show behavioural responses to g radua l - f i l l CO2 euthanasia at CO2 concentrations o f on ly 5 % CO2, and this response had peaked and was dec l in ing at CO2 concentrations o f o n l y 28%). S i m i l a r l y , i n Chapters 3 and 4 I found that rats avo ided CO2 concentrations greater than approx imate ly 1 5 % , and the m a x i m u m CO2 concentrat ion tolerated by a s ingle rat dur ing g radua l - f i l l exposure was on ly 2 5 % . These results suggest that pain does not account for the behav ioura l responses o f rats dur ing g radua l - f i l l CO2 euthanasia or for their avers ion to CO2. Furthermore, i n Chapter 2 I d i d not observe an increase i n pain-related behaviours , such as head -shak ing and face -wash ing , dur ing CO2 exposure, suggesting that the rats d i d not experience pa in w h i l e they were able to mount a behav ioura l response. H o w e v e r , it is poss ib le that rats experience pa in around the t ime o f loss o f posture. 6.2.2 Dyspnea A s discussed in Chapter 1, spontaneously breathing humans report dyspnea w i t h CO2 concentrations o f o n l y 8 % (Dr ipps and C o m r o e , 1947; L i o t t i et a l . , 2001) , and severe dyspnea has been reported w i t h CO2 concentrations o f 1 5 % to 20%) CO2 ( rev iewed by H i l l and F l a c k , 1908). W h i l e conc lus ive evidence o f dyspnea i n rats is not avai lable , the CO2 concentrations that e l ic i t dyspnea i n humans are consistent w i t h the concentrations that e l ic i ted a behavioural 118 responses and avers ion in , laboratory rats, suggesting that dyspnea is a l i k e l y cause o f distress dur ing CO2 exposure. H u m a n s general ly show a delay between a change i n inspi red CO2 and the onset o f dyspnea, but very l i tt le delay was observed i n rats. D u r i n g approach-avoidance testing w i t h 2 0 % CO2, one rat refused to enter the test cage, and the latency to leave the test cage for the remain ing rats ranged f r o m 2 to 16 s. I f w e assume a m a x i m u m o f 2 s o f exposure i n the tunnel leading to the test cage, this indicates that avoidance occurred after on ly 2 s o f exposure for the rat that d i d not enter the test cage, and 18 s o f exposure for the rat w i t h the longest latency to leave the test cage. Part o f the delay before onset o f dyspnea is due to the t ime taken for inspi red C 0 2 to increase b l o o d CO2 levels , and stimulate peripheral and central chemoreceptors i n the carot id bodies and the m e d u l l a , respect ively . F o r humans, the delay between a change i n inspired C 0 2 and the onset o f increased vent i lat ion is 5 to 1 5 s depending on whether peripheral or central chemoreceptors are d r i v i n g the response ( rev iewed by C u n n i n g h a m et a l . , 1986). W h i l e responses to hypercapnia tend to occur m a i n l y v i a the central chemoreceptors, for w h i c h there is a longer delay , the per ipheral chemoreceptors also contribute. B l o o d CO2 concentrations cont inue to increase towards inspi red leve ls , resul t ing i n increased vent i lat ion and dyspnea i f hypercapnia is suff ic ient . Banzett (1996) calculated that a step change i n inspi red CO2 against a h y p o x i c background results i n a logar i thmic increase i n dyspneic sensations i n humans, w i t h a ha l f - t ime o f approx imately 32 s for development o f a stable dyspnea rat ing. The t ime to onset o f dyspnea was not reported, but the ha l f - t ime prov ides an ind icat ion o f h o w l o n g it took for moderate levels o f dyspnea to develop under the condi t ions that were used. S i m i l a r l y , Haldane and S m i t h ( rev iewed by H i l l and F l a c k , 1908) found that there was a delay o f 1 to 2 m i n unt i l severe dyspnea set i n dur ing inhalat ion o f 18.6%> CO2. H o w e v e r , rats m a y dif fer f r o m humans i n the t ime taken to develop dyspnea. Lagneaux (1986) found that rats s h o w on ly a 2 s lag i n their 119 ventilatory response to inspiration of 1.5% CO2, suggesting a much shorter circulatory delay in rats. Furthermore, the increase in ventilation occurred much more quickly with 1.5% CO2 than with 0.5 or 1% CO2, indicating that the time for development of ventilation, and likely dyspnea, is dependent on CO2 concentration. These results suggest that the distress and aversion observed in this thesis could have been due to dyspnea. 6.2.3 Novelty The results of Chapter 5 illustrate the role of novelty on rat responses to gas exposure. While novelty results in increased exploratory behaviours and may contribute to distress due to fear, it does not account for performance of escape behaviours or avoidance of CO2. These results suggest that novelty is not a major source of distress during gradual-fill CO2 euthanasia. 6.2.4 Alternative hypotheses One alternative explanation for behavioural signs of distress and aversion in rats during CO2 exposure is that they could detect the onset bf unconsciousness, and that this sensation of diminished consciousness either evoked an innate escape response or was perceived as unpleasant. However, pigs, poultry and humans have been shown to lose consciousness during exposure to low O2 concentrations without demonstrating obvious attempts at avoidance (Cable, 2003; Raj & Gregory, 1995; Raj, 1996; Webster & Fletcher, 2004), suggesting that diminished consciousness on its own does not evoke escape responses or cause unpleasant sensations in other species. Furthermore, recent results from our research group suggest that avoidance of CO2 during gradual-fill exposure is not well correlated with ataxia, an initial indicator of diminished consciousness (Kirkden, unpublished data). Leach et al. (2002a) also found that rats avoided 25.5%) CO2 after only 1.1s, even though ataxia took 30 s to occur at this concentration ofCOi. 120 A v o i d a n c e o f CO2 dur ing approach-avoidance testing c o u l d also have been inf luenced by the effects o f CO2 o n taste. The approach-avoidance task that I used is dependent o n rats be ing h igh ly mot ivated to obtain a food reward , and alterations i n taste cou ld affect this mot ivat ion . CO2 forms carbonic ac id w h e n c o m b i n e d w i t h water, as w o u l d occur at the oral mucosa , and acids are k n o w n to stimulate taste buds to produce a sour taste ( D e S i m o n e et a l . , 2001) . CO2 is w i d e l y descr ibed to have a sour taste, but no in fo rmat ion is avai lable on the concentrat ion o f CO2 that w o u l d be necessary to evoke this taste i n rats. H o w e v e r , rats appeared h igh l y mot ivated to eat right unt i l they left the test cage, and w o u l d often grab one or two reward i tems to take w i t h them. Furthermore, this hypothesis cannot account for the escape behaviours that rats per formed dur ing g radua l - f i l l CO2 euthanasia. 6.2.5 Conclusions on sources of distress and aversion The most l i k e l y source o f distress and avers ion dur ing g radua l - f i l l CO2 exposure appears to be dyspnea. W h i l e there is some quest ion as to whether this sensation c o u l d develop w i t h i n the t ime it took for rats to exhibi t avers ion to CO2, the other hypotheses presented do not fu l l y account for the responses that were observed. In order to determine more conc lus ive ly whether CO2 causes dyspnea i n rats, further studies are needed. In humans , dyspnea due to exper imenta l l y - induced hypercapnia and due to disease has been s h o w n to be rel ieved b y inhalat ion o f aerosol ized furosemide ( M i n o w a et a l . , 2 0 0 2 ; N i s h i n o et a l . , 2 0 0 0 ; O n g et a l . , 2 0 0 4 ; S h i m o y a m a and S h i m o y a m a , 2002) . Furosemide is label led for use as a diuret ic , and the m e c h a n i s m for its effects o n dyspnea is not fu l l y understood. It has been found to increase the act iv i ty o f pu lmonary stretch receptors (Sudo et a l . , 2000) , w h i c h m a y feed back into systems responsible for the generation o f dyspnea. Th i s effect o n pu lmonary stretch receptors is speci f ic to aerosol de l ivery and does not occur after systemic del ivery , suggesting that this is a loca l i zed effect at the leve l o f the respiratory ep i the l ium. A n exper iment to test the effects o f furosemide 121 o n tolerance to CO2 w o u l d prov ide more direct evidence o f the role o f dyspnea i n rat avers ion to C 0 2 . 6.3 Critique of methods T o examine distress associated w i t h CO2 euthanasia i n rats I used two different methods: 1) behavioural assessment o f rat responses dur ing CO2 euthanasia, and 2) approach- avoidance testing w i t h static and gradual ly increasing concentrations o f CO2. W h i l e previous studies have assessed the behaviour o f rats dur ing CO2 euthanasia and found var iable results, I i m p r o v e d upon these studies by deve lop ing a w e l l def ined l ist o f p a i n and distress behaviours , us ing an adequate sample size for statistics, acc l imat i z ing the rats before gas exposure, and c o m p a r i n g behavioural responses to both baseline levels and responses dur ing air exposure. Preference test ing had also been used to examine rats' avoidance o f CO2 ( Leach et a l . , 2002a) , but I i m p r o v e d u p o n this methodology by e x a m i n i n g the strength o f rats' mot i vat ion to a v o i d CO2 us ing approach-avoidance testing. W h i l e this method had been used to examine gas avers ion i n other species (e.g. Cooper et a l . , 1 9 9 8 ; Ger r i t zen et a l . , 2 0 0 0 ; R a j , 1996; Ra j & Gregory , 1995; Webster & Fletcher , 2004) , it had not p rev ious ly been developed for use w i t h rats. B o t h o f the assessment methods that I used prov ided useful and complementary in format ion about the potential for CO2 to e l ic i t distress. Rats per formed escape behaviours dur ing CO2 euthanasia, and left the test cage before loss o f consciousness dur ing approach- avoidance testing. B o t h o f these responses suggest that g radua l - f i l l CO2 exposure was unpleasant and that forced exposure l i k e l y causes distress. H o w e v e r , rats' behavioural responses dur ing CO2 euthanasia were not compared to a k n o w n aversive st imulus , so it is d i f f icu l t to determine the leve l o f distress that is indicated by the level o f responses that I observed. In fact, 122 one rat showed escape behaviours dur ing exposure to peppermint odour , a st imulus that was found to be non-avers ive dur ing approach-avoidance testing. Th i s f i n d i n g suggests that escape behaviours m a y not a lways indicate a h i g h leve l o f distress. A p p r o a c h - a v o i d a n c e test ing prov ides a better ind icat ion o f the severity o f distress caused by CO2 exposure, because mot i va t ion to a v o i d CO2 is compared against mot i va t ion to obtain a food reward. A s discussed i n Chapter 3 , rats' mot i va t ion to obtain sweet foods when fed ad l i b i t u m is moderate to h i g h ; therefore, approach-avoidance test ing indicates that their mot ivat ion to avo id CO2 is at least moderate. A n even better ind icat ion o f strength o f avers ion to CO2 can be gained by f o o d - depr i v ing rats for different per iods o f t ime pr ior to testing to ensure that their mot ivat ion for the food reward is h igh . U s i n g this procedure, K i r k d e n et a l . (2005) found that even after food depr ivat ion for up to 24 h, rats showed m u c h the same avers ion to g radua l - f i l l CO2 exposure. T h i s result suggests that rats find this exposure h igh ly aversive. Approach -avo idance testing appears to be a more sensit ive measure o f avers ion to CO2 i n rats than s imple behavioural responses. A l t h o u g h rats showed m a n y behavioural changes dur ing CO2 euthanasia, the major i ty o f the behaviours observed are d i f f i cu l t to interpret. O n l y escape behaviours prov ide a clear ind icat ion o f avers ion. Behav iou ra l responses to CO2 euthanasia were also h igh l y var iable between rats, w i t h some animals s h o w i n g l itt le or no response. W h i l e this c o u l d be interpreted to indicate that some rats do not find this procedure distressing, rats a lways avoided CO2 concentrations o f approx imately 1 5 % and higher dur ing approach-avoidance testing. Th i s indicates that CO2 was l i k e l y aversive to a l l o f the rats dur ing CO2 euthanasia, but that the behavioural responses that I measured are var iable and l i k e l y poor measures o f distress dur ing forced gas exposure. In part icular , it is interesting that escape behaviours were not observed i n a l l an imals , because the results o f m y thesis and that o f K i r k d e n et a l . (2005) suggest that rats are h igh ly mot ivated to avo id CO2. It is possible that some rats d i d not perceive a potential to exit the exposure chamber and therefore d i d not try to 123 escape. W h i l e there is the poss ib i l i t y that rats were per fo rming relevant behaviours that I d i d not detect, the fact that some rats remained complete ly mot ionless throughout the exposure suggests this not to be the case. H o w e v e r , this raises the poss ib i l i t y that i n d i v i d u a l rats were cop ing w i t h the forced exposure i n different ways . Indiv iduals can respond either proact ive ly or react ively w h e n confronted w i t h a stressor ( rev iewed by K o o l h a s et a l . , 1999). Hence , another explanat ion for the var iab i l i t y i n m y results is that rats were cop ing i n different manners , w i t h some s h o w i n g f reez ing and others s h o w i n g escape behaviours . In experiments where I examined the behavioural responses o f rats dur ing CO2 euthanasia, an imals were tested i n a nove l environment. D u r i n g exposure to a predator odour, rats have been found to show behaviours associated w i t h fear w h e n tested i n a nove l env i ronment , but not w h e n tested i n a fami l ia r env i ronment ( M o r r o w et a l . , 2002) . Th i s f i n d i n g suggests that fear i n rats is enhanced by a nove l envi ronment . It is poss ib le that I m a y have obtained different results i f an imals were tested i n a fami l ia r envi ronment . H o w e v e r , rats were tested i n a fami l ia r envi ronment dur ing approach-avoidance testing and st i l l avo ided CO2 concentrations greater than 1 5 % , ind icat ing that the overa l l conc lus ions o f the thesis are v a l i d . The m a i n l imi ta t ion o f approach-avoidance testing is that it prov ides in format ion on ly about the rats' in i t ia l percept ion o f CO2, and does not address any effect that might occur w i t h cont inued exposure unt i l loss o f consciousness. F r o m the t ime-course o f CO2 euthanasia w i t h a m e d i u m f l o w rate, it appears that there is a per iod o f at least 45 s between the onset o f avers ion and loss o f consciousness. Distress dur ing the entire procedure c o u l d be assessed by compar ing g radua l - f i l l CO2 exposure unt i l loss o f consciousness w i t h a k n o w n aversive st imulus. One w a y to examine this w o u l d be to compare the w i l l ingness o f an imals to re-enter the exposure chamber after exposure to each o f the s t imu l i . F o r example , Jongman et a l . (2000) used this method to compare p i g s ' avers ion to CO2 and to an electr ic shock del ivered from a p rod , and found that p igs w o u l d more readi ly re-enter the exposure chamber after exposure to 9 0 % CO2 124 than after exposure to electr ic shock. Another method o f compar ing two aversive s t imul i is to use avo idance -avo idance testing, where animals must choose between two aversive s t imu l i . The st imulus that is chosen is assumed to be the less aversive o f the two . F o r example , R u s h e n (1986) used a Y - m a z e to compare avers ion to different hand l ing techniques i n sheep. Another l imi ta t ion o f the exper imental design that I used for approach-avoidance testing was that I had to habituate the rats to CO2 exposure pr ior to testing. Th i s resulted i n rats tolerat ing s l ight ly h igher CO2 concentrations than w o u l d be observed o n an in i t ia l exposure, as w o u l d occur dur ing euthanasia. H o w e v e r , it was necessary to habituate the rats to avo id a drift i n response to C 0 2 throughout the experiment. M o r e o v e r , I was able to estimate the magnitude o f this effect i n Chapter 5. 6.3 Future directions 6.3.1 Other rodent species The results o f m y thesis suggest that g radua l - f i l l C 0 2 euthanasia causes distress in rats, but we do not k n o w whether this is also true for other rodent species such as mice . Prev ious studies on distress dur ing CO2 euthanasia i n m i c e have been inconc lus i ve (Ambrose et a l . , 2 0 0 0 ; B l a c k s h a w et a l , 1988; Br i t t , 1986; Iwarsson and Rehbinder , 1993). L i k e the rat studies, these studies suffer f r o m prob lems w i t h exper imental des ign, i n c l u d i n g a lack o f appropriate sample sizes, control groups and acc l imat i zat ion before gas exposure. L e a c h et a l . (2002) demonstrated that m i c e w i l l avo id CO2 dur ing a preference test, but G o d b e y (personal communicat ion ) found that m i c e that are prov ided w i t h a food reward dur ing g radua l - f i l l CO2 euthanasia w i l l lose consciousness w h i l e eating. Th i s latter result suggests that m i c e might be less averse to CO2 than rats, but w e l l contro l led studies are needed to determine i f this is the case. 125 6.3.2 Alternative euthanasia methods G r a d u a l - f i l l CO2 euthanasia appears to cause distress i n rats, suggesting that this method should be replaced w i t h other methods that are more humane. H o w e v e r , f e w studies have examined whether the other methods that are currently avai lable cause distress in rats, and i f so, whether they are better or worse than CO2 euthanasia. One an imal wel fare benefit o f CO2 euthanasia is that it invo lves m i n i m a l hand l ing and restraint o f the animals . A s i d e f r o m CO2 euthanasia, the two major classes o f gas euthanasia agents are inhalant anaesthetics and inert gases, such as argon and nit rogen, w h i c h are used to displace O2 and cause severe h y p o x i a . Inhalant anaesthetics are not k n o w n to be pa in fu l or to cause dyspnea, but do have a pungent odour that rats m a y find unpleasant. U s i n g s imple preference testing, L e a c h et a l . (2002b) demonstrated that rats w i l l avo id exposure to inhalant anaesthetics before los ing consciousness, but further research is needed to examine the strength o f this avers ion. A s discussed i n Chapter 1, h y p o x i a can cause dyspnea i n humans , but does not appear to do so dur ing spontaneous breathing (Cable , 2 0 0 3 ; M o o s a v i et a l . , 2003) . Furthermore, p igs and poult ry w i l l enter a chamber conta in ing 9 0 % argon i n air ( 2 % O2) to access food and conspec i f ics , and w i l l remain for long enough to lose consciousness (Raj & Gregory , 1995; R a j , 1996; Webster & Fletcher , 2004) . Th i s suggests that it might be possib le to k i l l rats w i t h h y p o x i a wi thout caus ing distress. H o w e v e r , i n contrast to these results w i t h other species, Hornett and Haynes (1984) examined euthanasia by h y p o x i a w i t h a gradual ly increasing concentrat ion o f ni t rogen and found that it caused " p a n i c " (p.99) i n rats. Furthermore, L e a c h et a l . (2002a) found that rats avo id 90%> argon i n air before l os ing consciousness i n a s imple preference test. In Chapter 3 , 1 expanded on this w o r k be e x a m i n i n g rats' avers ion to 9 0 % argon i n air us ing approach-avoidance testing, and found that they w o u l d forgo a palatable food reward i n order to a v o i d argon exposure. In fact, the m e d i a n latency to leave w i t h argon was 126 on ly 3 s, w h i c h was s imi la r to that seen for 2 0 % CO2. A s an inert gas, argon is not thought to have a percept ible s m e l l , so the most l i k e l y explanat ion for this avoidance is dyspnea. It is not clear w h y rats behave dif ferent ly f r o m other species, but it is possible that they are more sensit ive to h y p o x i a as a result o f b u r r o w - d w e l l i n g adaptations. Enhanced sensit iv i ty to h y p o x i a and hypercapnia w o u l d be useful i n fossor ia l species for avoidance o f gas irregularit ies that can occur underground. W h i l e Rattus norvegicus is not fossor ia l , this trait m a y be conserved a m o n g rodents. V e n t i l a t i o n i n fossor ia l species is actual ly k n o w n to be less responsive to h y p o x i a and hypercapnia than i n non - fossor ia l species, l i k e l y i n order to tolerate the elevated CO2 (1 to 9 . 5 % ) and reduced 0 2 (6 to 2 0 % ) concentrations that occur i n c losed burrows ( rev iewed by Tenney and B o g g s , 1986). H o w e v e r , the causes and dynamics o f dyspnea are st i l l poor l y understood, and it is poss ib le that these species exper ience a sharp r ise i n dyspneic sensations at levels o f h y p o x i a and hypercapnia that are dangerous to surv iva l . C a r b o n m o n o x i d e gas is another potential alternative euthanasia agent for laboratory rodents. C a r b o n m o n o x i d e is an odourless and non- i r r i tat ing gas that causes death by direct tox ic effects o n ce l ls , and by compet ing for O2 b ind ing sites o n haemog lob in and prevent ing suff ic ient de l ivery o f O2 to body tissues ( rev iewed by K a o and Nanagas , 2005) . It has been suggested that carbon m o n o x i d e po ison ing causes death wi thout distress (C lose et a l . , 1996). H o w e v e r , humans suffer ing f r o m carbon m o n o x i d e po ison ing report symptoms such as headache, nausea, chest pa in , dyspnea, and elevated heart rate and breathing rate, w h i c h suggests that it does have the potential to cause distress ( rev iewed by K a o and Nanagas , 2005) . C a r b o n m o n o x i d e has not been examined as a euthanasia agent for rodents. However , C h a l i f o u x and Da l la i re (1993) found that carbon m o n o x i d e caused voca l i zat ions and signs o f agitation i n some dogs before loss o f consciousness. One major p r o b l e m w i t h the use o f carbon m o n o x i d e i n the laboratory is that it is an odourless, non- i r r i tat ing gas that is potent ia l ly dangerous to human 127 health, and its use w o u l d therefore require precautions to ensure human safety. Further research e x a m i n i n g the effects o f carbon m o n o x i d e o n rodents is necessary. C o m m o n l y used non-gas euthanasia methods for laboratory rats inc lude the phys ica l methods, such as decapitat ion and cerv ica l d is locat ion , and injectable anaesthetics. I f done proper ly , the phys i ca l methods are qu ick , and the m a i n a n i m a l wel fare concern is w i t h handl ing and restraint. H o w e v e r , the C C A C ranks phys ica l methods as on ly condi t iona l l y acceptable because considerable a n i m a l pa in can occur i f these methods are per formed improper ly . Inject ion o f anaesthetics i n rats also requires restraint and some pa in . It appears that a l l o f the c o m m o n l y used euthanasia methods for laboratory rats invo l ve factors that might cause distress before loss o f consciousness. W h i l e m y thesis suggests that CO2 exposure also causes distress I cannot say whether this distress is more o r less severe than that w h i c h occurs w i t h these other procedures. O n l y one study to date has compared avers ion to CO2 i n rats against other gas euthanasia methods. L e a c h et a l . (2002a, b) found that rats w o u l d remain i n inhalant anaesthetics and argon for longer than they w o u l d remain i n C 0 2 , but they avo ided a l l o f the agents before loss o f consciousness. Th i s result indicates that a l l o f the agents are aversive o n in i t ia l exposure, and that further research is necessary to determine w h i c h agent causes unconsciousness w i t h the least leve l o f distress. One next step w o u l d be to compare rats' avers ion to these different gas euthanasia agents us ing the avers ion testing methods descr ibed above i n 6 .3 . Rats w o u l d be exposed to each agent unt i l unconsciousness and then removed. The i r w i l l ingness to re-enter compartments associated w i t h each agent w o u l d then be compared so that the agents can be ranked accord ing to the leve l o f avers ion that they cause. These methods c o u l d also be used to examine whether rats f ind gas euthanasia agents or hand l ing and in ject ion more aversive. The agent or method that caused the least avers ion w o u l d be assumed to cause the least distress before unconsciousness. H o w e v e r , w i t h this methodology it is 128 assumed that rats remember the exposure; therefore, it is important to first ensure that none o f the agents affects memory . 6.4 Conclusions O v e r the past 10 years, there has been increasing concern about the use o f C 0 2 as a euthanasia agent, for laboratory rodents. The H u m a n e Society o f the U n i t e d States has ca l led for a ban on the use o f CO2 for euthanasia o f conscious rodents based o n a n i m a l welfare concerns (Conlee et a l . , 2005) , and recent developments have seen regulatory agencies, such as the European F o o d Safety Author i t y , suggest that CO2 should not be used as a sole euthanasia agent for consc ious animals ( E F S A , 2005) . T h e results o f m y thesis suggest that g radua l - f i l l CO2 euthanasia causes distress i n laboratory rats. I have demonstrated that rats per form escape behaviours dur ing g radua l - f i l l CO2 euthanasia, and that they are at least moderately averse to CO2 concentrations b e l o w those necessary to cause loss o f consciousness, regardless o f the del ivery method. These results suggest that laboratory rats shou ld not be euthanized w i t h CO2. H o w e v e r , prev ious research indicates that rats also show avers ion to other euthanasia agents, so it is not clear that any o f the agents currently avai lable can induce unconsciousness without distress. Further research is needed to determine w h i c h euthanasia agents cause the least amount o f distress i n rats and other rodent species, and to ident i fy alternative methods that can cause death wi thout distress. . 129 6.5 References A m b r o s e , N . , W a d h a m , J . , M o r t o n , D . , 2000 . Ref inement i n Euthanasia . In: B a l l s , M . , van Ze l le r , A . M . , Ha ider , M . E . (Eds) , Progress i n the R e d u c t i o n , Ref inement and Replacement o f A n i m a l Exper imentat ion , E l sev ie r Sc ience, A m s t e r d a m , pp. 1 1 5 9 - 1 1 6 9 . A n t o n , F., Euchner , I., Handwerker , H .O . , 1992. P s y c h o p h y s i c a l examinat ion o f pa in induced by def ined CO2 pulses appl ied to the nasal mucosa . P a i n 4 9 , 5 3 - 6 0 . Banzett , R . B . , 1996. D y n a m i c response characteristics o f CO^ - induces air hunger. Resp . P h y s i o l . 105, 4 7 - 5 5 . B l a c k s h a w , J . K . , F e n w i c k , D . C . , Beatt ie , A . W . , A l l a n , D . J . , 1988. The behaviour o f ch ickens , m i c e and rats dur ing euthanasia w i t h c h l o r o f o r m , carbon d iox ide and ether. L a b . A n i m . 2 2 , 6 7 - 7 5 . Br i t t , D . P. , 1996. The humaneness o f carbon d iox ide as an agent o f euthanasia for laboratory rodents. In: Euthanasia o f U n w a n t e d , Injured or D iseased A n i m a l s or for Educat iona l or Sc ient i f i c Purposes, Un ivers i t ies Federat ion for A n i m a l We l fa re , Potters B a r , pp. 1 9 - 3 1 . C a b l e , G . C , 2 0 0 3 . In - f l ight h y p o x i a incident i n mi l i ta ry aircraft : causes and impl i ca t ions for t ra in ing. A v i a t . Space E n v i r o n . M e d . 74 , 169 -172 . C h a l i f o u x , A . , Da l la i re , A . , 1993. P h y s i o l o g i c and behavioral evaluat ion o f CO2 euthanasia o f adult dogs. A m . J . Ve t . R e s . 44 , 2 4 1 2 - 2 4 1 7 . C h e n , X . , Ga l la r , J . , P o z o , M . A . , B a e z a , M . , B e l m o n t e , C , 1995. CO2 s t imulat ion o f the cornea: a compar ison between human sensation and nerve act iv i ty i n p o l y m o d a l nocicept ive afferents o f the cat. Eur . J . N e u r o s c i . 7, 1 1 5 4 - 1 1 6 3 . C l o s e , B . , Banister , K . , B a u m a n s , V . , Bernoth , E. , B r o m a g e , N . , B u n y a n , J . , Erhardt , W . , F l e c k n e l l , P. , Gregory , N . , Hackbar th , H . , M o r t o n , D. , W a r w i c k , C , 1996. Recommendat ions for euthanasia o f exper imental an imals : Part 1. L a b . A n i m . 3 0 , 2 9 3 - 3 1 6 . 130 C o e n e n , A . M . , D r inkenburg , W . H . , Hoenderken , R., van Lui j te laar , G . L . , 1995. C a r b o n d iox ide euthanasia i n rats: oxygen supplementat ion m i n i m i z e s signs o f agitat ion and asphyx ia . L a b . A n i m . 2 9 , 2 6 2 - 2 6 8 . C o n l e e , K . M . , Stephens, M . L . , R o w a n , A . N . , K i n g , L . A . , 2 0 0 5 . C a r b o n d iox ide for euthanasia: concerns regarding p a i n and distress, w i t h specia l reference to m i c e and rats. L a b . A n i m . 3 9 , 1 3 7 - 1 6 1 . Cooper , J . , M a s o n , G . , R a j , M . , 1998. Determinat ion o f the avers ion o f fa rmed m i n k (Muste la v ison) to carbon d iox ide . Vet . R e c . 143, 3 5 9 - 6 1 . C u n n i n g h a m , D . J .C . , R o b b i n s , P. A . , W o l f f , C . B . 1986. Integration o f respiratory response to changes i n a lveolar part ia l pressures o f CO2 and O2 and i n arterial p H . In: Chern iak , N . S . , W i d d i c o m b e , J . G . (eds), H a n d b o o k o f P h y s i o l o g y , Sect ion 3 : The Respiratory Sys tem, V o l u m e II: Con t ro l o f Breath ing , Part 2 , A m e r i c a n P h y s i o l o g i c a l Soc iety , Wash ing ton , D . C . , p p . 4 7 5 - 5 2 8 D a n n e m a n , P . J . , Ste in , S . , W a l s h a w , S . O . , 1997. H u m a n e and pract ical impl i cat ions o f us ing carbon d iox ide m i x e d w i t h oxygen for anaesthesia or euthanasia o f rats. L a b . A n i m . S c i . 4 7 , 3 7 6 - 8 5 . D e S i m o n e , J . A . , L y a l l , V . , H e c k , G . . , F e l d m a n , G . M . 2001 A c i d detect ion by taste receptor cel ls . Resp . P h y s i o l . 129, 2 3 1 - 2 4 5 . D r i p p s , R . D . , C o m r o e , J . H . , 1947. The respiratory and c i rcu latory response o f normal m a n to inhalat ion o f 7.6 and 10.4 per cent CO2 w i t h a compar i son o f the m a x i m a l vent i lat ion produced by severe muscular exercise, inhalat ion o f CO2 and m a x i m a l voluntary hypervent i lat ion. A m . J . P h y s i o l . 149, 4 3 - 5 1 . E F S A , 2 0 0 4 . O p i n i o n o f the Sc ient i f ic Panel o n A n i m a l Hea l th and Wel fa re on a quest ion related to " A s p e c t s o f the b io logy and welfare o f an imals used for exper imental and other sc ient i f ic purposes" - A d o p t e d by the A H A W Panel o n 14th N o v e m b e r 2005 (Quest ion N ° 131 E F S A - Q - 2 0 0 4 - 1 0 5 ) . The E F S A Journal 2 9 2 , 1-46 F e n g , Y . , S i m p s o n , T . L., 2 0 0 3 . N o c i c e p t i v e sensation and sensit iv i ty evoked f r o m human cornea and conjunct iva st imulated by CO2. Invest. Ophth . V i s . S c i . 4 4 , 5 2 9 - 5 3 2 . Ger r i t zen , M . A . , L a m b o o i j , E. , H i l l e b r a n d , S . J . W . , Lanhaar , J . A . C . , Pieterse, C , 2000 . B e h a v i o r a l responses o f broi lers to different gaseous atmospheres. Poul t ry S c i . 7 9 , 9 2 8 - 9 3 3 . H a c k b a r t h , H . , K u p p e r s , N . , Bohnet , W . , 2 0 0 0 . Euthanas ia o f rats w i t h carbon d i o x i d e — a n i m a l wel fare aspects. L a b . A n i m . 3 4 , 9 1 - 9 6 . Hewett , T . A . , K o v a c s , M . S . , A r t w o h l , J .E . , Bennett , B.T . , 1993. A compar i son o f euthanasia methods i n rats, us ing carbon d iox ide i n p re - f i l l ed and f i x e d f l o w rate filled chambers. L a b . A n i m . S c i . 4 3 , 5 7 9 - 5 8 2 . H i l l , L. , F l a c k , M . , 1908. The effect o f excess o f carbon d iox ide and o f want o f oxygen upon the t respirat ion and the c i rcu lat ion . J . P h y s i o l . 3 7 , 7 7 - 1 1 1 . Hornett , T . D . , H a y n e s , A . R . , 1984. C o m p a r i s o n o f carbon dioxide/air mix ture and nitrogen/air mixture for the euthanasia o f rodents. D e s i g n o f a system for inhalat ion euthanasia. A n i m a l Techno logy 3 5 , 9 3 - 9 9 . Iwarsson , K . , Rehb inder , C , 1993. A study o f different euthanasia techniques i n guinea p igs , rats, and m i c e . A n i m a l response and postmortem f indings . Scan . J . L a b . A n i m . S c i . 2 0 , 1 9 1 - 2 0 5 . Jongman, E . C . , Barnett , J . L . , H e m s w o r t h , P H . , 2 0 0 0 . The aversivenss o f carbon d iox ide stunning i n p igs and a compar i son o f the CO2 stunner crate vs. the V - rest ra iner . A p p l . A n i m . Behav . S c i . 67 , 6 7 - 7 6 . K a o , L . W . , Nanagas , K . A . , 2 0 0 5 . C a r b o n m o n o x i d e po ison ing . M e d . C l i n . N . A m . 89, 1 1 6 1 - 1194. 132 K i r d e n , R., N i e l . , L. , W e a r y , D . M . , 2 0 0 5 . H o w aversive is gradual f i l l carbon d iox ide euthanasia for rats? 2005 C A L A S / A C S A L S y m p o s i u m Proceedings , V a n c o u v e r , B . C . , Canada , p. 31. K o o l h a s , J . M . , K o r t e , S . M . , D e Boer , S . F . , v a n der Veg t , B . J . , van Reenen , C . G . , Hopster , H . , de Jong , I.C., R u i s , M . A . W . , B l o k h u i s , H.J . , 1999. C o p i n g styles i n an imals : current status i n behavior and stress -physio logy. N e u r o s c i . B iobehav . R. 2 3 , 9 2 5 - 9 3 5 . Lagneaux , D. , 1986. Vent i la tory responses o f the rat to m i l d hypercapni s t imulat ion before and after a lmit r ine bismesylate. Resp . P h y s i o l . 6 5 , 3 7 9 - 3 8 8 . L e a c h , M . C . , B o w e l l , V . A . , A l l a n , T . A . , M o r t o n , D . B . 2002a . A v e r s i o n to gaseous euthanasia agents i n rats and mice . Comparat i ve M e d . 5 2 , 2 4 9 - 2 5 7 . L e a c h , M . C . , B o w e l l , V . A . , A l l a n , T . F . & M o r t o n , D . B . 2002b. Degrees o f avers ion shown by rats and m i c e to different concentrations o f inhalat ional anaesthetics. Ve t . R e c . 150, SOS- S I S . L i o t t i , M . , B rannan , S . , E g a n , G . , Shade, R , M a d d e n , L. , A b p l a n a l p , B . , R o b i l l a r d , R., Lancaster , J . , Zamar r ipa , F .E . , F o x , P.T. , Denton , D. , 2 0 0 1 . B r a i n responses associated w i t h consciousness o f breathlessness (air hunger). P roc . Nat . A c a d . S c i . 9 8 , 2 0 3 5 - 2 0 4 0 . M o n t g o m e r y , K . C , 1955. The relat ion between fear induced by nove l s t imulat ion and exploratory behavior . J . C o m p . P h y s i o l . P s y c h o l . 4 8 , 2 5 4 - 2 6 0 . ' M i n o w a , Y . , Ide, T. , N i s h i n o , T. , 2 0 0 2 . Ef fects o f furosemide on C 0 2 venti latory responsiveness i n humans. P u l m . Pharmaco l : Ther. 15, 3 6 3 - 3 6 8 . M o o s a v i , S . H . , Go lestan ian , E., B i n k s , A . P . , L a n s i n g , R . W . , B r o w n , R., Banzett , R . B . , 2 0 0 3 . H y p o x i c and hypercapnic dr ives to breathe generate equivalent levels o f air hunger i n humans. J . A p p l . P h y s i o l . 9 4 , 141 -154 . M o r r o w , B . A . , E l s w o r t h , J . D . , R o t h , R . H . , 2 0 0 2 . F e a r - l i k e b i o c h e m i c a l and behavioural responses i n rats to the predator odor, T M T , are dependent o n the exposure environment. 133 Synapse 4 6 , 1 1 - 1 8 . N i s h i n o , T. , Ide, T. , Sudo , T., Sato, J . , 2000 . Inhaled furosemide greatly al leciates the sensation o f exper imental ly induced dyspnea. A m . J . Resp i r . Cr i t . Care M e d . 161, 1 9 6 3 - 1 9 6 7 . O n g , K . , K o r , A . , C h o n g , W . , Earnest, A . , W a n g , Y . , 2004. E f fects o f inhaled furosemide o n exert ional dyspnea i n chronic obstructive pu lmonary disease. A m . J . Resp i r . Cr i t . Care M e d . 169, 1 0 2 8 - 1 0 3 3 . P e p p e l , P. , A n t o n , F., 1993. Responses o f rat medul la ry dorsal horn neurons f o l l o w i n g intranasal nox ious chemica l s t imulat ion : effects o f st imulus intensity, durat ion, and interst imulus interval . J . N e u r o p h y s i o l . 70 , 2 2 6 0 - 2 2 7 5 . R a j , A . B . M . , 1996. A v e r s i v e reactions o f turkeys to argon, carbon d iox ide and a mixture o f carbon d iox ide and argon. Ve t . R e c . 138, 5 9 2 - 5 9 3 . R a j , A . B . M . , G regory , N . G . , 1995. Wel fa re impl icat ions o f the gas stunning o f pigs 1. Determinat ion o f avers ion to in i t ia l inhalat ion o f carbon d iox ide or argon. A n i m . Wel fa re 4, 2 7 3 - 2 8 0 . R u s h e n , J . , 1986. A v e r s i o n o f sheep for hand l ing treatments: pa i red -cho ice studies. A p p l . A n i m . Behav . S c i . 16, 3 6 3 - 3 7 0 . S h i m o y a m a , N . , S h i m o y a m a , M . , 2 0 0 2 . N e b u l i z e d furosemide as a n o v e l treatment for dyspnea in terminal cancer patients. J . P a i n S y m p t o m M a n a g . 2 3 , 7 3 - 7 6 . S m i t h , W . , Harrap , S . B . , 1997. Behav ioura l and cardiovascular responses o f rats to euthanasia us ing carbon d iox ide gas. L a b . A n i m . 3 1 , 3 3 7 - 3 4 6 . Sudo , T. , H a y a s h i , F., N i s h i n o , T . 2 0 0 0 . Responses o f t racheobronchial receptors to inhaled furosemide i n anesthetized rats. A m . J . Resp i r . Cr i t . Care M e d . 162, 9 7 1 - 9 7 5 Tenney , S . M . , B o g g s , D .F . , 1986. Comparat i ve m a m m a l i a n respiratory contro l . In: Chern iak , N . S . , W i d d i c o m b e , J . G . (eds), H a n d b o o k o f P h y s i o l o g y , Sect ion 3 : The Respiratory Sys tem, V o l u m e II: Con t ro l o f Breath ing , Part 2 , A m e r i c a n P h y s i o l o g i c a l Soc iety , Wash ington , D . C . , 134 p p . 4 7 5 - 5 2 8 Thurauf , N . , Gunther , M . , P a u l i , E. , K o b a l , G . , 2 0 0 2 . Sensi t iv i ty o f the negative mucosa l potential to the t r igeminal target s t imulus CO2. B r a i n Res . 9 4 2 , 2 7 - 8 6 . Webster , A . B . , F letcher , D . L . , 2004. Assessment o f the avers ion o f hens to different gas atmospheres us ing an approach-avoidance test. A p p l . A n i m . Behav . S c i . 88 , 2 7 5 - 2 8 7 . 135

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 4 1
Japan 3 0
France 2 0
Canada 2 0
China 1 12
City Views Downloads
Tokyo 3 0
Ashburn 3 0
Unknown 2 1
Vancouver 2 0
Lewes 1 0
Shanghai 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items