Current attitudes to, and use of, peri-operative analgesia in dogs and cats by veterinarians in New Zealand

193 New Zealand Veterinary Journal 53(3), 193-202, 2005 Scientific Article Current attitudes to, and use of, peri-operative analgesia in dogs and cats by veterinarians in New Zealand VM Williams *, BDX Lascelles and MC Robson Abstract AIM: To investigate the attitudes of veterinary practitioners in New Zealand to pain and analgesia, and their use of analgesic drugs, in dogs and cats. METHODS: A questionnaire posted to 1,200 practising veterinarians was used to gather information about the use of analgesia in dogs and cats, assessment of pain, attitudes to pain relief, analgesic drugs and procedures used, factors affecting choice of analgesic agent, and veterinary demographics, continuing education and staffing. RESULTS: Three hundred and twenty questionnaires with useable data were returned, a response rate of 28%. Male and female veterinarians were evenly represented. The analgesic agents most commonly used were morphine (opioids) and carprofen (a non-steroidal anti-inflammatory drug; NSAID). Use of peri-operative pain relief ranged from 50% for castration of cats to 91% for fracture repair in dogs. For most procedures, female veterinarians scored pain at a significantly higher level than their male colleagues. Fifty-eight percent of respondents considered their knowledge in the area of assessment and treatment of pain was adequate. CONCLUSIONS: This survey was considered representative of veterinarians working in companion animal practice in New Zealand. Results indicated a relatively high use of peri-operative analgesia, including both pre-emptive and multi-modal analgesia, in cats and dogs, although there was still some disparity between the perception of how painful a procedure was and the consequent use of pain relief. CLINICAL RELEVANCE: The establishment of current attitudes and practices indicates to practising veterinarians how their own use of analgesics compares with that of their colleagues. It also provides information to educators on potential areas of focus, given that 42% of respondents felt their knowledge in the area of assessment and treatment of pain was inadequate. KEY WORDS: Dogs, cats, analgesia, veterinarians, attitudes, New Zealand * School of Natural Sciences, Unitec Institute of Technology, Private Bag 92025, Auckland, New Zealand. Comparative Pain Research Laboratory, College of Veterinary Medicine, North Carolina State University, Raleigh, NC 27606, USA. Veterinary Specialist Group, Unitec Institute of Technology, Private Bag 92025, Auckland, New Zealand. Author for correspondence. Email: vwilliams@unitec.ac.nz Introduction Recent years have seen an increase in research into, and knowledge and interest in, evaluating and treating pain in animals. There has also been a substantial escalation in the knowledge and understanding of the pathophysiological processes involved in pain transmission (Woolf 2000), leading to relatively new theories regarding optimal use of peri-operative analgesia. These include the suggestion that certain analgesic agents might be most effective if administered prior to the onset of pain (pre-emptive analgesia; Coderre et al 1993; Dahl and Kehlet 1993; Woolf and Chong 1993; Lascelles et al 1995, 1997; Kissin 2000; Moiniche et al 2002), and the suggestion that simultaneous use of multiple and different classes of analgesic drugs might produce the best peri-operative analgesia (Kaneko et al 1994; Slingsby and Waterman-Pearson 2001). Although the pattern seems to be changing, recent surveys suggest that the use of analgesic drugs in veterinary practice is still suboptimal. Hansen and Hardie (1993) were the first to report on the incidence of analgesic use. In that survey, performed in a major referral hospital in the United States of America over 13 years ago, only 28% of dogs received one or more doses of an analgesic agent following extubation, and only 19% of dogs received analgesia for more than 8 h post-operatively. That survey also found that only 1/15 cats received one dose of an analgesic agent post-operatively. A Canadian survey (Dohoo and Dohoo 1996ab) revealed that respondents routinely used an analgesic agent in 84% of dogs and 70% of cats undergoing orthopaedic surgery, 13% of dogs and 17% of cats undergoing ovariohysterectomy, and 10% of dogs and 9% of cats undergoing castration. That survey demonstrated that respondents tended to be users of analgesic drugs, or not; 49.5% of respondents reported giving analgesic drugs to 50% of their cases. An Australian survey (Watson et al 1996) performed at a similar time found use of analgesia varied from 94% for patients with acute severe trauma, through to 6% for ovariohysterectomy and 4% for castration of dogs. More recently, a British survey (Capner et al 1999; Lascelles et al 1999) revealed that 97% and 94% of respondents routinely used an analgesic in dogs and cats, respectively, undergoing orthopaedic surgery, 53% and 26% for ovariohysterectomy, and 32% and 16% for castration. In a small South African survey (Joubert 2001), 65% of respondents used an analgesic drug routinely in cats undergoing ovariohysterectomy, but only 31% of respondents used an analgesic drug routinely in dogs undergoing ovariohysterectomy or castration. These surveys suggest that attitudes towards providing peri-op- NSAID VCNZ Non-steroidal anti-inflammatory drug Veterinary Council of New Zealand

Williams et al New Zealand Veterinary Journal 53(3), 2005 194 erative analgesia may vary geographically, but also suggest that attitudes may be changing over time. Attitudes to the use of analgesia have also been found to vary according to both gender and age (Dohoo and Dohoo 1996b; Capner et al 1999; Lascelles et al 1999), as female veterinarians allocated significantly higher pain scores to their patients than did males, as did more recent graduates compared with graduates of longer standing. The current study was designed to investigate the attitudes of veterinary practitioners in New Zealand to pain and analgesia and their use of analgesic drugs, in dogs and cats. The establishment of current attitudes and practices gives an indication to practising veterinarians of how their own use of analgesia compares with that of their colleagues, and of where that use might be altered or improved. It also highlights areas of focus for both education and research. Prior to the study presented here, published information on the use of peri-operative analgesia in dogs and cats in veterinary practice in New Zealand was lacking. Materials and methods Questionnaires were posted to the 1,200 individual veterinarians identified from the Veterinary Council of New Zealand (VCNZ) register as being involved in clinical practice. Veterinarians specifically identified as dealing only with species other than cats and dogs were excluded. The questionnaire was based on a similar survey conducted in Britain (Capner et al 1999; Lascelles et al 1999), and was designed to be completed within 20 minutes. Data were obtained about the use of analgesia in cats and dogs. The survey was sent out at the end of 2001, and data were collated in 2002. The survey consisted of eight sections: Part I collected demographic information such as the veterinary school attended, qualification and year of graduation, post-graduate qualifications, gender, type of practice, number of veterinarians within the practice, percentage of time spent treating small animals, location of practice (urban/rural/mixed), and number of veterinary nurses with and without qualifications within the practice. Part II asked respondents to identify the analgesic drugs and/or procedures they used from a list that included those available in New Zealand. Part III asked for information influencing the respondent s decision to use the two major groups of analgesic drugs, opioids and NSAIDs. For both groups, respondents were asked to classify as very important, important, not very important or not important at all, the potential for side-effects, cost, availability of information on the drug, and analgesic efficacy. In addition, for opioids, respondents were asked to similarly classify the degree of sedation and the requirement to keep records, and for NSAIDs the lack of sedation and the anti-inflammatory effect of these agents. Respondents were also invited to specify, with no prompts given, potential side-effects for both opioids and NSAIDs. Part IV explored attitudes to pain relief through a series of questions. Respondents were asked to indicate their response on a scale of 1 10, where 1 indicated strong agreement with a statement, 5 indicated a neutral position, and 10 indicated strong disagreement. Results were divided into three groups, where rankings of 1 4 were taken as agreement with the statement, 6 10 as disagreement, and 5 as neutral. Part V related to the use of analgesic drugs in dogs for a range of surgical procedures, including fracture repair, cruciate ligament repair, laparotomy, ovariohysterectomy, castration, dental with major extractions, and lateral ear-canal resection. Information sought for each procedure included: whether analgesia was used routinely, reactively or never; drugs used; whether drugs were given pre-operatively, during surgery, after surgery but prior to extubation or post-extubation; whether additional doses of analgesic agents were routinely, reactively or never given; and whether analgesia was routinely prescribed on discharge of the patient, and if so, the drug and duration of therapy used. Part VI related to the use of analgesic agents in cats for a range of surgical procedures, using major orthopaedic procedure, ruptured diaphragm, laparotomy, ovariohysterectomy, castration, and dental with major extractions, as the surgical examples. The same information was sought as in Part V. Part VII asked respondents to rate on a scale of 1 10 their assessment of the severity of pain that animals undergoing the surgical procedures specified in Parts V and VI would experience during the first 12 h after surgery if no analgesic drugs were given. The sliding scale from 1 (minimal pain) to 10 (maximal pain) was grouped into three categories, comprising 1 3, 4 7, and 8 10. Part VIII was concerned with continuing education and staffing, asking what the most valuable sources of information about recognising and treating pain were, whether respondents considered their knowledge in this area adequate, what forms of continuing education would be most appropriate for increasing their knowledge, whether veterinarians or veterinary nurses were principally responsible for monitoring animals post-operatively, and how veterinarians rated their nurses knowledge of pain assessment and treatment. A reply-paid envelope was sent out with each survey and an announcement was placed in the New Zealand Veterinary Association magazine, VetScript, after the mailing, to encourage a high return rate. The questionnaires were assigned a number on receipt. Statistical analysis All data were recorded anonymously onto SPSS version 11.5 for Windows (SPSS Inc, Chicago IL, USA). If questions were not answered or answered ambiguously they were registered as missing data. If respondents gave a range when a numerical classification was called for, the average of the range (e.g. 2 for 1 3) or the lower number (e.g. 1 for 1 2) was routinely taken. Simple descriptive statistics were produced for all variables in the dataset. Cross tabulation was used to determine combinations of drugs chosen by those respondents who used more than one analgesic drug for a procedure. Depending on the number of groups, differences in scores and proportions were tested using the non-parametric Kruskal Wallis or Mann-Whitney U tests. Results Part I: Demographics Of the 1,200 questionnaires sent out, 53 were returned by veterinarians who considered themselves inappropriately targeted.

195 New Zealand Veterinary Journal 53(3), 2005 Williams et al Table 1. Demographic data of respondents to a survey of attitudes to, and use of, peri-operative analgesia in dogs and cats by veterinarians in New Zealand (Survey Part I). Gender Demographic Male Female Total Percentage 166 a (52%) 153 (48%) 319 b Year of graduation Pre-1960 2 0 2 <1% 1961 1970 22 2 24 8% 1971 1980 54 17 71 22% 1981 1985 27 19 46 14% 1986 1990 26 27 53 17% 1991 1995 20 37 57 18% 1996 2000 14 51 65 20% Total 165 a 153 318 b Veterinary school attended New Zealand (Massey) 136 121 257 81% Australia 14 5 19 6% South Africa 7 1 8 3% Great Britain 7 15 22 7% Other European 2 8 10 3% North American 0 3 3 1% Post-graduate qualifications None 131 123 254 80% MACVSc 19 14 33 10% Other 16 16 32 10% Practice location Urban 78 66 144 45% Rural 37 56 93 29% Urban/rural 51 31 82 26% Time spent treating small animals <10% 15 9 24 8% 11 30% 22 19 41 13% 31 60% 20 19 39 12% 61 90% 25 21 46 14% >90% 83 85 168 53% a,b Discrepancies in numbers are attributable to occasional missing data MACVSc = Membership of the Australian College of Veterinary Scientists Of the remaining 1,147, there were 320 questionnaires which contained useable data, a response rate of 28%. Discrepancies in total numbers reported relate to occasional missing data for some variables. Details of demographic data are summarised in Table 1. Male and female veterinarians were evenly represented. A total of 221/318 (69%) respondents had graduated in the previous 20 years, and two veterinarians had graduated prior to 1960. The majority of respondents were graduates of Massey University (New Zealand), the remainder coming from Australia, South Africa, Great Britain, Europe and North America. While the most common postgraduate qualification was Membership of the Australian College of Veterinary Scientists, other qualifications mentioned included Doctor of Philosophy, Fellowship of the Australian College of Veterinary Scientists, Diplomate of various American veterinary post-graduate colleges, Master of Veterinary Science, and postgraduate diploma. The highest number of respondents came from practices classed as urban. Of 318 respondents, 253 (80%) spent >30% of their clinical time treating small animals, and 168/318 (53%) spent >90% of their time on small animals. A comparison of the survey population with the equivalent population taken from VCNZ statistics of veterinarians holding current practising certificates as at May 2004 is detailed in Table 2, which shows the years since graduation of survey respondents was broadly representative of registered veterinarians in New Zealand at that time. Part II: Analgesic drugs and analgesic procedures used on a regular basis Details of analgesic drugs and procedures used on a regular basis are summarised in Table 3. The opioid used by the greatest number of respondents was morphine, while the NSAID used by most respondents was carprofen. Two thirds of respondents used local anaesthetic agents for procedures including skin blocks, nerve blocks, intra-articular blocks and epidurals. Part III: Choice of analgesic agent The importance placed on factors influencing the choice to use opioid or NSAID analgesia in cats and dogs in the peri-operative period is summarised in Table 4. Analgesic efficacy was considered important or very important by most respondents for both opioids and NSAIDs. The most common side-effect specified Table 2. Comparison of the demographic data of survey respondents with the equivalent graduating population taken from Veterinary Council of New Zealand (VCNZ) statistics of veterinarians holding current practising certificates as at May 2004. Current survey VCNZ Year of graduation Male Female Total % Male Female Total % Pre-1960 2 2 <1% 36 36 2% 1961 1970 22 2 24 8% 169 4 173 10% 1971 1980 54 17 71 22% 376 87 463 26% 1981 1985 27 19 46 14% 178 106 284 16% 1986 1990 26 27 53 17% 140 152 292 17% 1991 1995 20 37 57 18% 107 118 225 13% 1996 2000 14 51 65 20% 105 176 281 16% Total 165 153 318 1,111 643 1,754

Williams et al New Zealand Veterinary Journal 53(3), 2005 196 Table 3. Percentages of survey respondents (n=320) who used named analgesic drugs or procedures on a regular basis (Survey Part II). Drug or procedure % Respondents who used the drug or procedure Opioids Buprenorphine 60 Butorphanol 23 Morphine 66 Pethidine 36 Codeine 10 Fentanyl 4 NSAIDs Carprofen 83 Ketoprofen 62 Meloxicam 50 Flunixin meglumine 12 Tolfenamic acid 25 Other Local anaesthetic agent 64 Ketamine 38 Tricyclic antidepressants 4 Procedures Skin block 70 Nerve block 23 Intra-articular block 20 Epidural (local anaesthetic) 19 Epidural (morphine) 4 Acupuncture 7 Fentanyl patch 4 Homeopathy 5 NSAIDs = Non-steroidal anti-inflammatory drugs for opioids was respiratory depression (26% respondents), while cardiovascular effects (cited as bradycardia, low blood pressure, cardiac irregularities and cardiovascular depression) and gastrointestinal problems, including nausea and vomiting, were raised by 15% of respondents. For NSAIDs, the potential for side-effects was rated highly, and the most common specified (% respondents) were renal compromise (43%), followed by gastrointestinal problems, including ulceration and bleeding (32%), clotting problems (16%) and hepatic compromise (8%). Part IV: Attitudes to pain relief Attitudes of respondents towards statements concerning pain and analgesia are detailed in Table 5. Most respondents agreed that animals should be given the same considerations for pain relief as humans (84%), that the rate of recovery from surgery was improved by the use of analgesic drugs (79%), and that prevention of pain was better than pain relief (83%). Of 316 respondents, 35 (11%) agreed that they only gave pain relief to those animals whose owners were willing to pay for it. Parts V and VI: Use of analgesic drugs in dogs and cats The percentage of respondents routinely using peri-operative analgesic drugs for common surgical procedures in dogs and cats are shown in Table 6. Both routine and multi-modal (>1 analgesic drug) use of analgesics was highest for fracture repair and lowest for castrations. The frequency with which analgesic drugs were prescribed pre-emptively, either prior to or during anaesthesia, is shown in Table 7. The drugs used for routine peri-operative analgesia by more than 5% of the respondents who used analgesics routinely, cross-tabulated against surgical procedure (Table 8), included: morphine, buprenorphine, pethidine, carprofen, meloxicam, ketofen, tolfenamic acid, local anaesthetic agents (dogs only), and nonspecified NSAIDs. The drugs used by fewer than 5% of respondents included butorphanol and flunixin in dogs and cats, and fentanyl, codeine and local anaesthetic agents in cats. Local anaes- Table 4. Importance placed by survey respondents on various factors in the decision whether to use an opioid or non-steroidal anti-inflammatory drug (NSAID) in dogs and cats (Survey Part III). % Respondents regarding factor as: Unimportant/not very important Important/very important Factor Dogs Cats Dogs Cats Opioids Potential for side-effects 34 26 66 74 Cost 86 88 14 12 Sedation 32 35 68 65 Available information 14 13 86 87 Need for record-keeping 50 51 50 49 Analgesic efficacy 1 1 99 99 NSAIDs Potential for side-effects 7 7 93 93 Cost 60 70 40 30 Lack of sedation 69 70 31 30 Available information 8 7 92 93 Anti-inflammatory effect 12 12 88 88 Analgesic efficacy 2 2 98 98

197 New Zealand Veterinary Journal 53(3), 2005 Williams et al Table 5. Attitudes of survey respondents (n=315) towards statements concerning pain and analgesia in dogs and cats (Survey Part IV). Attitude % Agree % Neutral % Disagree A degree of post-operative pain is good because it keeps the animal inactive 16 8 77 a Animal should be given the same considerations for pain relief as humans 84 6 11 a I only give pain relief to those animals whose owners are willing to pay for it 11 7 82 The potential side-effects of using analgesia outweigh the benefits 10 4 86 The rate of recovery from surgery is improved by the use of analgesia 79 9 11 a Pain prevention is better than pain relief 83 6 11 Using pain relief prolongs recovery from anaesthesia 20 20 60 I am constrained from using the degree of pain relief I feel is ideal by practice policy 11 8 81 I feel constrained from using the degree of pain relief I feel is ideal by concerns over owner budget 25 11 64 It is important to accurately calculate dose-rates of analgesic drugs 90 2 8 I use standard analgesic doses on estimated bodyweights 48 9 44 a a Discrepancies in totals attributable to rounding of percentages Table 6. Percentage of survey respondents who used analgesia routinely, and percentage of routine users who used multi-modal analgesia, for common surgical procedures in dogs and cats (Survey Parts V and VI). Use of drugs Routine >1 drug >2 drugs Procedure Animal (% respondents) (% routine users) (% routine users) Fracture repair Dog 98 72 14 Cat 97 61 11 Cruciate repair Dog 97 70 22 Cat Diaphragmatic hernia repair Dog Cat 90 50 7 Laparotomy Dog 76 48 3 Cat 70 37 8 Ovariohysterectomy Dog 76 39 1 Cat 64 22 1 Castration Dog 65 27 1 Cat 50 14 1 Dental Dog 81 48 7 Cat 75 43 4 Lateral ear-canal resection Dog 93 66 7 Cat Table 7. Percentage of survey respondents (n=320) who administered at least one analgesic drug either prior to or during surgery in dogs and cats (Survey Parts V and VI). Procedure Dog Cat Fracture repair 91 77 Cruciate repair 87 Diaphragmatic hernia repair 76 Laparotomy 77 69 Ovariohysterectomy 74 60 Castration 65 50 Dental 74 67 Lateral ear-canal resection 81 thesia was used by 4% of respondents for cats undergoing dental procedures. The combinations of drugs used for peri-operative analgesia, cross-tabulated against surgical procedure, are reported in Table 9. The most common drug combination used in dogs was morphine/carprofen (29% of respondents who used two drugs), followed by morphine and an NSAID (12%). For surgical procedures in cats for which two drugs were used, the most common combination used for fractures, castrations and dental procedures was morphine/ketoprofen; for diaphragmatic hernia it was buprenorphine/carprofen; for ovariohysterectomy it was morphine/ carprofen; and for laparotomy it was morphine/buprenorphine. The percentage of respondents providing pain relief after surgery to dogs and cats at discharge, for common surgical procedures, is shown in Table 10. The drugs prescribed at discharge by more than 5% of the respondents who discharged animals with drugs, cross-tabulated against surgical procedure (Table 11), included: carprofen, meloxicam, ketoprofen, codeine and unspecified NSAIDs in both dogs and cats, and tolfenamic acid in cats. The percentage of respondents who prescribed drugs at discharge for more than 5 days supply is also shown in Table 11. Drugs prescribed by fewer than 5% of respondents were: buprenorphine (dogs and cats), and tolfenamic acid and fentanyl (dogs only).

Williams et al New Zealand Veterinary Journal 53(3), 2005 198 Table 8. Percentage of survey respondents who used analgesia routinely, who used various drugs for various surgical procedures in dogs and cats. Results are only shown for combinations used by 5% of qualifying respondents (Survey Parts V and VI). Procedure Animal Morph Carp Bup Melox Peth Keto Tolf A LA Unspec NSAID Drug Fracture repair Dog 62 39 28 15 7 5 5 14 Cat 45 24 36 5 7 27 13 7 Cruciate repair Dog 59 40 26 14 6 5 5 16 14 Cat Diaphragmatic hernia repair Dog Cat 35 23 36 8 21 12 6 Laparotomy Dog 47 23 20 10 10 Cat 36 20 28 5 15 9 5 Ovariohysterectomy Dog 47 19 13 13 5 9 Cat 30 16 18 11 6 Castration Dog 40 14 9 11 6 Cat 45 24 36 5 7 27 13 7 Dental Dog 46 27 16 10 6 6 12 Cat 35 21 23 23 13 6 Lateral ear-canal resection Dog 56 38 22 15 10 6 5 14 Cat Morph = morphine; Carp = carprofen; Bup = buprenorphine; Melox = meloxicam; Peth = pethidine; Keto = ketofen; Tolf A = tolfenamic acid; LA = local anaesthetic agent; Unspec NSAID = unspecified non-steroidal anti-inflammatory drug Table 9. Percentage of those survey respondents who used two or three analgesic drugs for surgical procedures in dogs and cats (where n 10 cases) who used particular drug combinations (Survey Parts V and VI). Two drugs: opioid plus Three drugs: opioid plus NSAID Opioid Total Opioid/ NSAID/ Opioid/ Total Procedure Animal (%) (%) respondents NSAID (%) local (%) local (%) respondents Fracture repair Dog 88 10 172 83 17 41 Cat 82 14 149 78 13 6 32 Cruciate repair Dog 85 8 140 34 53 13 64 Cat Diaphragmatic hernia repair Dog Cat 79 18 116 78 6 18 Laparotomy Dog 83 15 136 8 Cat 76 20 99 70 10 10 Ovariohysterectomy Dog 90 10 115 3 Cat 89 10 62 3 Castration Dog 86 10 79 2 Cat 92 5 39 2 Dental Dog 80 11 122 25 70 5 20 Cat 82 14 115 23 62 15 13 Lateral ear-canal resection Dog 89 9 156 84 11 5 19 Cat NSAID = non-steroidal anti-inflammatory drug Part VII: Assessment of pain Perceptions of the severity of pain, quantified using subjective pain scores, that animals were likely to experience during the first 12 h after a specified surgical procedure if no analgesic drugs were given are reported in Table 12. For both dogs and cats, fracture repair was rated most painful and castration least painful. For every surgical procedure except lateral ear-canal resection in dogs, women scored pain at a significantly higher level than men (p<0.05; Mann-Whitney U test), while for every operation except cruciate ligament repair and lateral ear-canal resection in dogs, and diaphragmatic hernia repair in cats, respondents who graduated from 1986 onwards scored pain at a significantly higher level than those graduating prior to 1986 (p<0.05; Mann-Whitney U test). No significant differences (Kruskal Wallis test) in pain scores were evident between veterinarians working in urban, rural or mixed practice locations. Part VIII: Continuing education and staffing The value placed on sources of information that contributed to a respondent s knowledge about recognising and treating pain in animals is summarised in Table 13. Of the 302 respondents who

199 New Zealand Veterinary Journal 53(3), 2005 Williams et al Table 10. Percentage of survey respondents who provided pain relief at time of discharge for dogs and cats, following various surgical procedures (Survey Parts V and VI). Procedure Dog Cat Fracture repair 79 68 Cruciate repair 78 Diaphragmatic hernia repair 40 Laparotomy 24 17 Ovariohysterectomy 11 3 Castration 6 1 Dental 42 34 Lateral ear-canal resection 61 answered this question, 174 (58%) considered their knowledge in the area of pain assessment and treatment was adequate. For those who considered their knowledge inadequate, the value placed on different sources of information for gaining further knowledge on recognising and treating pain in animals is summarised in Table 14. Post-operative monitoring was noted as principally the responsibility of veterinarians in 124/312 (40%) cases; of the veterinary nurses in 166/312 (53%); and of both in 22/312 (7%). Of 310 respondents, 186 (60%) considered their veterinary nursing staff had adequate knowledge about pain and its treatment, while 160/291 (55%) thought the nursing staff themselves felt their knowledge was adequate. Discussion The 28% response rate of our survey was higher than the 10% recorded for a similar study in Australia (Watson et al 1996) and lower than the 50% recorded by Capner et al (1999) and Lascelles et al (1999) in the United Kingdom (UK), but the latter survey represented only 10.7% of practising veterinarians in the UK. Our survey was estimated to represent 18% of practising veterinarians in New Zealand, although this is a conservative estimate since it is based on numbers holding current practising certificates as at May 2004, a total (1,754 veterinarians) which is likely to be higher than 2 years previously. Also, in contrast to other studies where responses were sought from a sample of the veterinary population, the relatively small total number of veterinarians in New Zealand meant that the survey population comprised all those veterinarians, nationally, who were possibly involved in small animal work. Thus, our survey is considered representative of veterinarians in small animal practice in New Zealand, although bias introduced by the voluntary nature of completion of questionnaires must be considered. A comparison of the survey population with the equivalent graduating population taken from VCNZ statistics of veterinarians holding current practising certificates as at May 2004 (Table 2) showed that percentages of respondents stratified by year of graduation were comparable, while the differences in gender ratios indicated a higher proportion of female graduates working in small animal practice than male graduates. Opioid drugs appeared to be widely available and frequently used in small animal practices. Morphine and buprenorphine were the opioid drugs most frequently used in our survey, in contrast with the British survey (Capner et al 1999; Lascelles et al 1999) where buprenorphine and butorphanol were the most frequently used, and the Canadian survey (Dohoo and Dohoo 1996ab) where butorphanol, followed by pethidine (meperidine), were most frequently used. The range of factors influencing the use of opioids in the peri-operative period was similar between the British study and our survey, although sedation was considered to be slightly more important in the current study. As in the British survey, the potential for side-effects was deemed important, and more respondents considered it important in cats than in dogs. Interestingly, 7% and 10% of veterinarians who provided single-agent pain relief in patients undergoing fracture repair and lateral earcanal resection, respectively, used pethidine. Pethidine is a very Table 11. Percentage of those survey respondents who discharged animals with drugs following specific surgical procedures, who used specific drugs or who dispensed drugs for use over >5 days. Results are only shown for drugs used by 5% of qualifying respondents (Survey Parts V and VI). Procedure Animal Carp Melox Keto Tolf A Codeine Unspec NSAID >5 days Drug Fracture repair Dog 64 11 7 15 52 Cat 10 7 58 16 7 20 Cruciate repair Dog 65 11 6 15 53 Cat Diaphragmatic hernia repair Dog Cat 9 6 60 15 7 14 Laparotomy Dog 54 11 8 6 17 14 Cat 8 55 18 10 6 Ovariohysterectomy Dog 61 8 6 19 6 Cat 10 40 10 40 Castration Dog 44 17 11 22 6 Cat 33 67 Dental Dog 58 8 13 17 9 Cat 7 67 14 5 3 Lateral ear-canal resection Dog 64 9 10 14 37 Cat Carp = carprofen; Melox = meloxicam; Keto = ketofen; Tolf A = tolfenamic acid; Unspec NSAID = unspecified non-steroidal anti-inflammatory drug

Williams et al New Zealand Veterinary Journal 53(3), 2005 200 Table 12. Percentage of respondents who assigned pain scores on a scale of 1 to 10 (minimal to maximal), indicating the degree of pain expected during the first 12 h after a variety of surgical procedures if no analgesic drugs were provided to patients (Survey Part VII). Pain score Procedure Animal 1 3 4 7 8 10 Fracture repair Dog 18 82 Cat 21 79 Cruciate repair Dog 26 74 Cat Diaphragmatic hernia repair Dog Cat 40 60 Laparotomy Dog 26 74 Cat 4 73 23 Ovariohysterectomy Dog 5 73 22 Cat 12 74 15 Castration Dog 22 72 6 Cat 40 55 5 Dental Dog 5 62 33 Cat 6 62 33 Lateral ear-canal resection Dog 1 43 56 Cat Table 13. Percentages of survey respondents (n=302) who placed most and least value on six sources of information (including Other ; data not shown) that contributed to their knowledge of recognising and treating pain in animals (Survey Part VIII). Discrepancies are due to respondents giving equal value to more than one source of information and/or not ranking all sources. Source of information Most valuable Least valuable Under-graduate training 17 28 Journals 17 1 Continuing education lectures 35 5 Experience gained in practice 35 4 Commercial literature 8 3 Table 14. Percentages of those respondents who believed their knowledge of pain was inadequate (n=128), who placed most and least value on various sources of information (including Other ; data not shown) for gaining further knowledge on recognising and treating pain in animals. Discrepancies are due to respondents giving equal value to more than one source of information and/or not ranking all sources. Source of information Most valuable Least valuable Review articles in journals 31 1 Research articles in journals 8 10 National lectures/seminars 22 7 Regional lectures/seminars 52 2 Video/audiocassettes 11 19 Internet 2 42 short-acting opioid, only providing analgesia for 1 2 h (Waterman and Kalthum 1989; Lascelles et al 1997). Pethidine was also the opioid most commonly used by veterinarians in South Africa for treating post-operative pain (Joubert 2001). As in the British survey (Capner et al 1999; Lascelles et al 1999), the NSAIDs appeared to be slightly more frequently used than the opioids for peri-operative pain relief, and the potential for side-effects was cited more frequently as a concern for NSAIDs than for opioids, probably explaining the greater importance placed on available information for NSAIDs compared with opioids. It is interesting to note that cardiac irregularities and cardiovascular depression were raised voluntarily by respondents as side effects of opioids, which are, in fact, known for their cardiovascular stability (bradycardia excepted). This highlights an area for continuing education. As with the opioids, cost appeared not to be a limiting factor in the decision to use NSAIDs, findings similar to the British survey. The NSAIDs used by respondents were very similar to those reported in the British survey, but very different from those reported in the Australian survey (Watson et al 1996), where flunixin meglumine and dipyrone were the NSAIDs most commonly used in dogs, and ketoprofen in cats. This difference between the Australian survey, the current survey and the British survey is undoubtedly due to the changing availability of NSAIDs approved for veterinary use, with the introduction of carprofen, meloxicam and ketoprofen since the Australian survey was performed. Similarly, in South Africa, the NSAIDs most commonly used for peri-operative pain were flunixin, phenylbutazone and aspirin, at a time when carprofen and meloxicam were not registered by the South African Medicines and Drug-Related Authority (Joubert 2001). In contrast to this study, the British (Capner et al 1999; Lascelles et al 1999), Canadian (Dohoo and Dohoo 1996ab) and Australian (Watson et al 1996) surveys did not evaluate the use of analgesic practices such as skin blocks, nerve blocks, joint blocks, epidurals, acupuncture and transdermal patches. With the exception of skin blocks, respondents used these procedures relatively infrequently. This could either be because these procedures were not appropriate for the cases treated by the responding veterinarians, or that the responding veterinarians did not feel adequately trained in such techniques. A total of 174/302 (58%) respondents indicated they believed their knowledge in the area of assessment and treatment of pain was adequate. However, questions about specific aspects of treatment of pain were not asked. Respondents did indicate they used some drugs and procedures that were not included on the questionnaire, and these included methadone, intra-articular morphine, magnetic collars, and tender loving care. A comparison of responses to the questions about attitudes to pain and analgesia between this and the British survey (Capner et al 1999; Lascelles et al 1999) clearly showed that more respondents in New Zealand believed recovery from anaesthesia and surgery was improved by the provision of analgesia (80% vs 67%), and fewer respondents believed that a degree of post-operative pain was useful in keeping the animal inactive (16% vs 30%). It must be remembered, though, that such comparisons should be taken in the temporal context. The British survey was performed in 1996, and over the last few years there has been substantial gains in the knowledge and understanding of peri-operative pain in dogs and cats, and significant post-graduate education to convey these advances. A present-day survey of British veterinarians may well reveal a very different response regarding their attitudes towards pain and analgesia. Veterinarians from our survey expressed a greater degree of concern over an owner s willingness to pay for analgesic medication than was found in previous surveys. It should be noted that because a 10-point scale was used in Part IV, an element of bias may have been introduced by using 5 as the neutral position.

201 New Zealand Veterinary Journal 53(3), 2005 Williams et al Some disparities in attitude were apparent. Almost all the responding veterinarians indicated that fracture repair, cruciate repair, diaphragmatic hernia repair and laparotomy were associated with pain scores of 4 or more (out of 10), but not all routinely used analgesia in these cases. This may be due either to pain not being detected in assessments of individuals undergoing these surgeries, or to a degree of pain being tolerated before intervention with analgesic drugs occurred. Indeed, as mentioned above, 49/315 (16%) respondents agreed with the statement A degree of post-operative pain is good because it keeps the animal inactive. Also interesting is the finding that although the pain score distributions for cruciate surgery and laparotomy in dogs were very similar, 294/303 (97%) respondents used analgesia routinely for cruciate surgery, but only 236/310 (76%) did so for laparotomy. So, although veterinarians appeared to consider certain procedures were associated with pain, in daily practice it may be for procedures where pain was more obvious, such as a leg not being used following cruciate repair, that analgesia was provided more readily. Similar disparities in attitude between considering a surgical procedure painful and actually administering analgesia to patients undergoing these same procedures were also reported in the British (Capner et al 1999; Lascelles et al 1999) and Canadian (Dohoo and Dohoo 1996ab) surveys. This is important, because it demonstrates that this trend has remained in the mind of veterinarians despite the passage of time, changes in teaching curricula, continuing education and increased availability of analgesic agents. Disparities were also noted between species in the present survey. For instance, dogs and cats undergoing ovariohysterectomy received similar pain score distributions, but the percentage of respondents using analgesia routinely in these patients was lower for cats than for dogs (64% vs 76%). In the British survey (Capner et al 1999; Lascelles et al 1999), the pain scores were also the same and, again, cats were routinely treated less frequently. However, in the British survey only 26% and 53% of cats and dogs, respectively, undergoing ovariohysterectomy were routinely treated with analgesic drugs. In the Canadian survey (Dohoo and Dohoo 1996ab), only 13% of dogs and 17% of cats undergoing ovariohysterectomy were routinely provided with analgesia, while the Australian study (Watson et al 1996) found only a 6% rate of use of analgesia for ovariohysterectomy. In the Canadian survey, the proportion of cats undergoing ovariohysterectomy that received analgesia was biased upwards by a single practitioner who always used analgesic drugs. Overall, veterinarians in New Zealand appeared to be the most frequent users of analgesia when surveys from Australia, Canada and the UK were compared. Considering the similarities in teaching programmes between the British, Australian and New Zealand veterinary schools, and the fact that all these schools are accredited by one body (The Royal College of Veterinary Surgeons), it would seem unlikely that veterinarians in New Zealand were more frequent users of analgesia than their Commonwealth counterparts, had they all been evaluated at the same time. What these findings do show is an apparent increase in the usage of analgesia over the last 10 years by veterinarians who undergo similar training programmes. The present survey is the most recent, and attitudes and practices are probably changing quickly. However, differences between countries may exist, as indicated by comparison of rates of use of analgesia for ovariohysterectomy between concurrent studies in Australia (Watson et al 1996) and Canada (Dohoo and Dohoo 1996ab), as described above. The pain scores assigned to particular procedures varied between veterinarians graduated more or less than 16 years, similar to the British (Capner et al 1999; Lascelles et al 1999) and Canadian (Dohoo and Dohoo 1996ab) surveys. The date of 1985 was taken as the cut-off point between recent and less recent graduates because it was in the mid-1980s that instruction on peri-operative analgesia began to be stressed in the veterinary curriculum at Massey University (K Stafford 1, pers. comm.). More recent graduates assigned higher pain scores to surgical procedures, particularly intra-abdominal and neutering procedures, in a manner similar to the results found in the British survey. As in other surveys, female veterinarians appeared to assign higher pain scores to surgical procedures. Interestingly, there were no significant differences noted in pain scores assigned to surgical procedures between veterinarians working in urban, rural or mixed practice. This was not examined in previous surveys, but dispelled the common perception that veterinarians in rural or mixed practice were less sensitive to the recognition of pain in small animals. Compared to the most recent survey (Capner et al 1999; Lascelles et al 1999), responding veterinarians in New Zealand appeared to be using analgesic drugs more frequently pre-operatively, and at least 50% of respondents used at least one analgesic drug pre- or intra-operatively for all procedures in cats, and at least 65% for all procedures in dogs. Again, these findings may indicate a shift in practice over time. The clinical relevance of pre-emptive analgesia is still hotly debated (Moiniche et al 2002), but the controlled studies that have been performed on animals have all found a clinical benefit, albeit short-lived in the studies on the use of NSAIDs (Lascelles et al 1997, 1998; Welsh et al 1997). Veterinarians in New Zealand also seemed to be receptive to the use of combinations of drugs that act at different levels of the pain pathway (multi-modal analgesia), as an aid to optimising peri-operative analgesia (Kaneko et al 1994; Slingsby and Waterman-Pearson 2001). While fracture repair had the highest rate of multi-modal analgesia, even for castrations in dogs and cats, 27% and 14% of respondents, respectively, used more than one analgesic drug. The most common combination of drugs used was an opioid and an NSAID; only for laparotomy in the cat was a combination of two opioids used more frequently. No controlled trials have been reported that evaluate how long analgesia is required post-operatively, or on the quantity of analgesic medication that should be dispensed on discharge of patients. Our survey was the first to seek information about the dispensing of analgesic medication following common surgical procedures. The results probably indicate the length of time veterinarians in New Zealand believed analgesic medication should have been provided post-operatively, as it is likely the majority of practices were sending patients home soon after surgery. There was large variation in the percentage of respondents discharging patients with pain-relief medication when various surgical procedures were compared, e.g. 79% vs 24% following fracture repair vs laparotomy in dogs. This may well be appropriate, and may be in line with the expected duration of pain following these procedures, as no published research could be found on this aspect of pain in animals. It is, however, an area where good clinical research is required. 1 Professor KJ Stafford, Institute of Veterinary, Animal and Biomedical Sciences, Massey University, Palmerston North, New Zealand

Williams et al New Zealand Veterinary Journal 53(3), 2005 202 The medications dispensed were almost always NSAIDs. Codeine was the non-nsaid most frequently used, particularly following abdominal surgery, and especially in cats. No research could be located on the efficacy of oral opioids used post-operatively in dogs or cats, although many people now suggest their use. This survey revealed that 53% of respondents considered nurses were responsible for monitoring animals post-operatively, which compares with 71% in the British survey (Capner et al 1999; Lascelles et al 1999). In another Canadian survey, 88% of animal health technicians (nurses) said they were responsible for monitoring animals post-operatively (Dohoo and Dohoo 1998). Thus, in our survey a high proportion of post-operative evaluation was performed by veterinary nurses in New Zealand, and this suggests continuing education on the management of pain should be targeted at nurses as much as at veterinarians. However, the survey did not ask who was responsible for administering analgesia. If nurses were monitoring patients, but not given responsibility for initiating the administration of analgesic drugs, then animals may be less efficiently medicated. More information is needed on the relationship between veterinarians and veterinary nurses with regard to post-operative analgesic medication. In the only study available, 89% of animal health technicians surveyed in Canada recommended post-operative analgesic therapy to a veterinarian when they believed it was warranted, and 90% reported animals received additional analgesia, based upon their recommendations (Dohoo and Dohoo 1998). In summary, the results of this survey indicated a relatively high use of peri-operative analgesia, including both pre-emptive and multi-modal analgesia, in cats and dogs in New Zealand, although there was still some disparity between the perception of how painful a procedure was and the subsequent use of pain relief for that procedure. Areas noted for further study include the appropriate length of post-operative management of pain, the issues associated with prescription and dispensing of analgesic agents upon discharge, and the relationship between veterinarians and veterinary nurses with regard to post-operative analgesic medication. Acknowledgements We are grateful to the VCNZ for the supply of mailing lists and statistics, and to Rob McNeill of Auckland University for help with statistics. The corresponding author acknowledges the financial support of the Unitec Institute of Technology. References Capner CA, Lascelles BD, Waterman-Pearson AE. Current British veterinary attitudes to perioperative analgesia for dogs. Veterinary Record 145, 95 9, 1999 Coderre TJ, Katz J, Vaccarino AL, Melzack R. Contribution of central neuroplasticity to pathological pain: review of clinical and experimental evidence. Pain 52, 259 85, 1993 Dahl JB, Kehlet H. The value of pre-emptive analgesia in the treatment of postoperative pain. British Journal of Anaesthesia 70, 434 9, 1993 Dohoo SE, Dohoo IR. Postoperative use of analgesics in dogs and cats by Canadian veterinarians. Canadian Veterinary Journal 37, 546 51, 1996 a Dohoo SE, Dohoo IR. Factors influencing the postoperative use of analgesics in dogs and cats by Canadian veterinarians. Canadian Veterinary Journal 37, 552 6, 1996 b Dohoo SE, Dohoo IR. Attitudes and concerns of Canadian animal health technologists toward postoperative pain management in dogs and cats. Canadian Veterinary Journal 39, 491 6, 1998 Hansen B, Hardie E. Prescription and use of analgesics in dogs and cats in a veterinary teaching hospital: 258 cases (1983 1989). Journal of the American Veterinary Medical Association 202, 1485 94, 1993 Joubert KE. The use of analgesic drugs by South African Veterinarians. Journal of the South African Veterinary Association 72, 57 60, 2001 Kaneko M, Saito Y, Kirihara Y, Collins JG, Kosaka Y. Synergistic antinociceptive interaction after epidural co-administration of morphine and lidocaine in rats. Anesthesiology 80, 137 50, 1994 Kissin I. Preemptive analgesia. Anesthesiology 93, 1138 43, 2000 Lascelles BD, Waterman AE, Cripps PJ, Livingston A, Henderson G. Central sensitization as a result of surgical pain: investigation of the pre-emptive value of pethidine for ovariohysterectomy in the rat. Pain 62, 201 12, 1995 Lascelles BD, Cripps PJ, Jones A, Waterman AE. Post-operative central hypersensitivity and pain: the pre-emptive value of pethidine for ovariohysterectomy. Pain 73, 461 71, 1997 Lascelles BD, Cripps PJ, Jones A, Waterman-Pearson AE. Efficacy and kinetics of carprofen, administered preoperatively or postoperatively, for the prevention of pain in dogs undergoing ovariohysterectomy. Veterinary Surgery 27, 568 82, 1998 Lascelles BD, Capner C, Waterman-Pearson A. Current British veterinary attitudes to perioperative analgesia for cats and small mammals. Veterinary Record 145, 601 4, 1999 Moiniche S, Kehlet H, Dahl JB. A qualitative and quantitative systematic review of preemptive analgesia for postoperative pain relief: the role of timing of analgesia. Anesthesiology 96, 725 41, 2002 Slingsby LS, Waterman-Pearson AE. Analgesic effects in dogs of carprofen and pethidine together compared with the effects of either drug alone. Veterinary Record 148, 441 4, 2001 Waterman AE, Kalthum W. Pharmacokinetics of intramuscularly administered pethidine in dogs and the influence of anaesthesia and surgery. Veterinary Record 124, 293 6, 1989 Watson AD, Nicholson A, Church DB, Pearson MR. Use of anti-inflammatory and analgesic drugs in dogs and cats. Australian Veterinary Journal 74, 203 10, 1996 Welsh EM, Nolan AM, Reid J. Beneficial effects of administering carprofen before surgery in dogs. Veterinary Record 141, 251 3, 1997 Woolf CJ. Pain. Neurobiology of Disease 7, 504 10, 2000 Woolf CJ, Chong MS. Preemptive analgesia treating postoperative pain by preventing the establishment of central sensitization. Anesthesia and Analgesia 77, 362 79, 1993 Submitted 28 June 2004 Accepted for publication 03 November 2004