The effects of two non-steroidal anti-inflammatory drugs on the mobility of laying hens with keel bone fractures

Veterinary Anaesthesia and Analgesia, 2015, 42, 197 204 doi:10.1111/vaa.12175 RESEARCH PAPER The effects of two non-steroidal anti-inflammatory drugs on the mobility of laying hens with keel bone fractures Mohammed AF Nasr*,, Christine J Nicol*, Lindsay Wilkins* & Joanna C Murrell* *School of Veterinary Science, University of Bristol, Langford House, Langford, UK Faculty of Veterinary Medicine, Animal Wealth Development Department, Zagazig University, Sharkia, Egypt Correspondence: Mohammed AF Nasr, Faculty of Veterinary Medicine, Animal Wealth Development Department, Zagazig University, Sharkia, 44519, Egypt. E-mails: mohammed.nasr@bristol.ac.uk; Nasr.maf@gmail.com Abstract Objective Investigate the effects of administration of meloxicam and carprofen on the mobility of hens with and without keel fractures. Study design Within each of two experiments a blinded randomised cross over design whereby birds received either the test drug (carprofen or meloxicam) or saline. Animals Two groups of Lohman Brown hens with and without keel bone fractures. Methods The first group (n = 63) was treated with carprofen 25 mg kg 1 and saline subcutaneously, twice. The second group (n = 40) was treated with meloxicam (5 mg kg 1 ) and saline subcutaneously. The latency of birds to fly down from perches 50, 100 and 150 cm above the ground was measured after each treatment. Data from experiment 1 and 2 were analysed separately; the effects of drug treatment compared with saline on landing time for birds with and without keel bone fractures were evaluated using MLwiN. Results In both experiments latency to fly down from perches was longer in hens with keel fractures and there was a significant interaction between perch height and fracture status. For carprofen, at the 50 cm, 100 cm and 150 cm perch heights, birds with fractures took (mean SD) 2.5 2.9, 6.8 9.7 and 11.5 13.2 seconds respectively to fly down compared with 1.3 0.5, 2.3 1.2 and 4.2 3.1 seconds for birds without fractures. For meloxicam, at the 50 cm, 100 cm and 150 cm perch heights, birds with fractures took 2.9 2.5, 49.8 85.4 and 100.3 123.6 seconds respectively compared with 0.7 0.5, 2.5 7.1 and 3.0 4.6 seconds to fly down for birds without fractures. There was no significant effect of carprofen or meloxicam treatment. Conclusion and clinical relevance These data provide further confirmation that keel fractures reduce the willingness of birds to move from perches. Keywords fracture, hens, keel, mobility, non steroidal anti-inflammatory drug, pain. Introduction Keel bone fractures in laying hens housed in noncaged and furnished caged systems occur frequently, with prevalences of up to 80% reported in some studies (Wilkins et al. 2011). We have recently shown that healed keel bone fractures reduce bird mobility (Nasr et al. 2012a). Mobility can be partially improved (by approximately 20%) by the administration of butorphanol, a kappa opioid agonist (Nasr et al. 2012b), suggesting that keel bone fractures may be painful in hens. However, opioids also have complex effects on mood and reward systems (Bruchas et al. 2010). In mammals, kappa agonists produce dysphoria 197

and a feeling of unpleasantness (Wee & Koob 2010). Although, the effects of kappa agonists on mood, reward and behaviour in birds are unknown, it is possible that butorphanol improved the mobility of the hens with keel bone fractures indirectly through an effect on behaviour mediated by an interaction between butorphanol and dopaminergic pathways, rather than via a direct analgesic effect mediated by kappa opioid receptors located in areas of the central nervous system involved in nociceptive processing and pain. To provide further evidence about whether the impairment of mobility in hens with keel bone fractures can be attributed to pain it is logical to investigate the effects of other classes of analgesic drugs on hen mobility in the same model. Non steroidal anti-inflammatory drugs (NSAIDs) are widely used in mammals for analgesia. The major mechanism of action of NSAIDs is a reduction in the synthesis of prostaglandins; mediators that play a key role in both peripheral and central nociceptive pathways, contributing to upregulated sensory processing and heightened sensitivity to pain following inflammation (Vane et al. 1998; Kelly et al. 2001). The antinociceptive effects of different NSAIDs have been studied in birds with experimentally induced arthritis (Hocking et al. 2005; Cole et al. 2009; Paul-Murphy et al. 2009). The NSAIDs carprofen, flunixin and ketoprofen altered behaviours in chickens with induced arthritis, though effects were discerned only at doses much higher than those recommended for administration to cats and dogs (Hocking et al. 2005; Leece et al. 2005). The behavioural effects of carprofen and meloxicam (also an NSAID) have been investigated in parrots with experimentally induced arthritis in one limb (Cole et al. 2009; Paul-Murphy et al. 2009). Carprofen (3 mg kg 1 ), did not change motor activity or weight bearing on the affected limb compared to parrots treated with saline (control). In comparison, meloxicam (1 mg kg 1 ), increased weight bearing on the affected limb compared to parrots treated with saline, indicative of an antinociceptive effect in this model. Very recently, in a fracture model in pigeons, Desmarchelier et al. (2012) reported that meloxicam did not show any effect at low dose (0.5 mg kg 1 ), but a higher dose (2 mg kg 1 ) improved the weight bearing and lowered the pain score. The effects of both carprofen (25 mg kg 1 ) and meloxicam (5 mg kg 1 ) on the gait characteristics of lame broiler chickens were investigated using kinematic analysis (Caplen et al. 2013a). At the doses studied, both drugs increased walking velocity, indicating improved walking ability, accompanied by changes in gait characteristics at a slow walking speed that may have been indicative of improved walking stability. The effect of carprofen on the agility of lame broiler chickens has also been investigated (McGeown et al. 1999; Danbury et al. 2000). Carprofen (1 mg kg 1 ) decreased the time taken by lame broiler chickens to complete an obstacle course compared to untreated lame birds, whereas no effect of carprofen on the agility of non-lame birds was found. Collectively these studies demonstrate that some NSAIDs modulate the behaviours of birds with both experimentally induced and naturally occurring lameness, although the efficacy and dose of NSAID necessary to produce a behavioural effect varied markedly between studies. Given the well described mechanism of action of NSAIDs in other spontaneous diseases associated with lameness (for example osteoarthritis in dogs), it is very likely that the improved mobility in birds with lameness following NSAID administration can be attributed to the anti-inflammatory effect of this class of drugs. However underlying mechanisms following NSAID administration have not yet been robustly investigated in birds. The aim of this study was to test the prediction that carprofen and meloxicam would improve the mobility of hens with keel fractures. We used a mobility test that had previously discriminated between injured and uninjured birds. When hens were placed on perches of different heights, and provided with a food reward placed on the ground, individuals with healed keel bone fractures took longer to leave the perch than individuals with no fractures (Nasr et al. 2012a). The latency to leave the perch was (somewhat) reduced when hens with keel fractures were given an appropriate dose of an opioid drug (Nasr et al. 2012b). If hens with fractures are unwilling to leave a perch to fly down to the ground because they are in pain, then we suggest that NSAID drugs should similarly reduce the latency to fly down. We therefore compared the time taken for hens with keel bone fractures to leave perches of different heights following NSAID and saline administration. We also predicted that NSAID administration would have no effect on latency in hens without keel fractures. 198 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 42, 197 204

Materials and methods Birds and management Lohman Brown laying hens housed in a free-range system with access to perches obtained from a commercial farm were used in this experiment. Birds were selected by palpation of the keel (Wilkins et al. 2004) to detect birds with healed keel bone fractures and birds with no keel bone fractures. Two groups of hens were studied sequentially; the first group of hens (n = 63, 35 week old hens) was treated with carprofen. The second group of hens (n = 40, 38 weeks old) was treated with meloxicam. Hens were transported to the University of Bristol, School of Veterinary Sciences and on arrival, were individually identified by a leg tag and weighed. Birds were subsequently weighed on alternative days during the experimental period. Each group of hens was housed in one of two floor pens (3 9 3.5 m), bedded with wood shavings, and provided with a metal single tier nest box unit. The hens were provided with ad libitum layers mash from two suspended poultry feeders and water from two suspended poultry drinkers. The lighting programme was 14L: 10D and temperature inside the pen ranged between 19 and 21 C. Each group of hens was allowed to acclimatise for at least 3 days before experiments started. The experiment was carried out under Home Office Project Licence 30/2685. Landing test Training was carried out by catching the hens and carrying them to the test room individually in a single box; hens could hear but could not see other birds. Food was withdrawn 2 hours before morning training and testing time and 1 hour before afternoon training. All the hens were initially trained by placing them on a low perch (20 cm height) and allowing them to see a mealworm. Hens were considered trained when they jumped from this low perch in <10 seconds and ate the food reward. One hen in experiment 2 was excluded because she could not be trained. Carprofen experiment Trained hens were allocated randomly to receive treatment with either carprofen (Rimadyl 50 mg ml 1 solution, Pfizer Animal Health, UK) 25 mg kg 1 injected subcutaneously in the dorsal neck or an equivalent volume of saline in a cross over design. All hens were treated with both carprofen and saline twice, with 48 hours allowed to elapse between each treatment. Birds that received carprofen as the first treatment, received saline as the second treatment, while birds that received saline first were subsequently treated with carprofen, and this sequence was repeated. One hour after carprofen or saline injection food was withdrawn from the hens; 3 hours after injection the landing test started. Each landing test comprised three trials with the perch set at 50 cm, 100 cm and 150 cm from the ground respectively. The time taken for birds to leave the perch and fly to the ground was recorded (using a digital handheld sport stopwatch with a 1/100 second chronograph) by an observer standing inside the room where the experiment was carried out, the observer was unaware to the treatment (drug or saline) and the fracture status. At the end of the experiment birds were killed and the keel bone was then dissected to confirm the presence of an old or recent keel bone fracture (for details see Sandilands et al. 2009) or absence of a keel bone fracture and hens were grouped into two categories (no fracture and fracture group based on this dissection). Meloxicam experiment Using the same experimental design as for the carprofen experiment, hens were randomly allocated to receive treatment with either meloxicam (Metacam 5 mg ml 1 solution, Boehringer Ingelheim, UK) 5 mg kg 1 injected subcutaneously in the dorsal neck or an equivalent volume of saline. All hens were treated with both meloxicam and saline twice, with 48 hours allowed to elapse between each treatment. The landing test was carried out 3 hours after injection, with food withdrawn for 2 hours before the start of the landing test. The landing test was carried out as described for the carprofen experiment. At the end of the experiment birds were killed and the keel bone dissected to ascertain fracture status. The doses of carprofen and meloxicam used in this study were based on work carried out by our research group in broiler birds and were sufficient to prevent the development of thermal primary hyperalgesia in broiler chickens with experimentally induced arthritis (Caplen et al. 2013b). Carprofen and meloxicam at a dose of 25 mg kg 1 and 5mgkg 1 subcutaneously, respectively, altered 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 42, 197 204 199

gait parameters in lame broilers when compared with saline injection (Caplen et al. 2013a). The testing time was based on a pharmacokinetic study of carprofen in broiler chickens, which revealed that the peak plasma level of carprofen used with different doses (15, 25 and 35 mg kg 1 ) occurred 3 6 hours after injection and concentrations were negligible 24 hours after injection (Hothersall et al. 2012). In addition, a study of meloxicam in five bird species reported an elimination half life of meloxicam in chicken of 3.21 hours (Baert & De Backer 2003). Statistical analyses The mean latency to land from each perch height was calculated for each drug and for saline for each bird prior to statistical analysis. The statistical analysis was conducted using MLwiN 2.2 (Rasbash et al. 2012) as this allowed non-independence in the dataset to be accounted for. A base (random effects logistic regression) model acknowledged the hierarchical structure of the dataset where repeated observations (six per hen) occurred within clusters (three perch heights for each drug or saline treatment) and where hen identity may also have played a role (63 hens for the carprofen experiment and 40 hens for the meloxicam experiment). This model allowed each level of the hierarchy to be treated as a random, rather than a fixed, effect. The statistical model contained fixed terms for slope and intercept based on the mean latency to land across all observations for the slope of the correlation between the variables. The random parts of the model (error terms) denoted random variation in latencies to land partitioned into components between observations, clusters and individuals. Subsequently, the influences of the independent categorical variables fracture status and treatment were examined. The influence of perch height was included as a continuous variable allowing all two way interactions and also the three-way interaction between treatment, fracture status and perch height to be investigated. The influence of perch height as both a linear and quadratic continuous variable was considered, but as this did not affect the significance of any results, the models including a linear term for perch height are reported. All models were also run using untransformed data for latency to land, and log transformed data, but this did not affect the significance of any results and the models including the untransformed data are reported here. Pearson correlations analysed the relationship between the body weight and latency to land from different perch heights. Values were considered significant at p 0.05. The data were summarised using the mean and standard deviation. Results Influence of hen bodyweight In the carprofen experiment, there was no significant difference in the bodyweight of the hens with (n = 43; (mean SD) 1.74 0.13 kg) and without (n = 20; 1.74 0.18 kg) keel fractures. Also, no bivariate correlations were detected between bodyweight and latency to land from any perch height when fractured and non-fractured birds were considered together, or when birds of different fracture status were examined separately. In the meloxicam experiment, there was again no significant difference in the bodyweight of hens with (n = 23; 1.77 0.14 kg) and without (n = 16; 1.71 0.16 kg) keel fractures. No bivariate correlations between bodyweight and latency to land from any perch height were detected when fractured and non-fractured birds were considered together. When birds of different fracture status were examined separately, there were no significant correlations between body weight and latency to land for birds with keel fractures at any perch height. However, within the non-fractured group (n = 16), there was a positive correlation between bird weight and latency to land from the 50 cm (r = 0.67, p = 0.01) and the 100 cm (r = 0.63, p = 0.01) perch, though not from the 150 cm perch (r = 0.07, p = 0.80). Carprofen Dissection of the keel bones of birds at the end of the experiment confirmed that 43 birds had a healed keel bone fracture and 20 hens had no keel bone fracture. No birds had evidence of new keel bone fractures. The statistical model is shown in supporting information in the online version of this article (Fig. S1). There was a significant interaction between perch height and fracture status ( 0.055/ 0.020 = z = 0.035; p < 0.006). At the 50 cm, 100 cm and 150 cm perch heights, birds with fractures took 2.5 2.9, 6.8 9.7 and 11.5 13.2 seconds respectively to fly down 200 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 42, 197 204

compared with 1.3 0.5, 2.3 1.2, and 4.2 3.1 seconds for birds without fractures. There was no significant effect of carprofen treatment ( 1.28/1.74 = z = 0.76; p = 0.46) (Fig. 1). Meloxicam Dissection of the keel bones of birds at the end of the experiment confirmed that 23 hens had a healed keel bone fracture and 16 hens had no keel bone fracture. No birds had evidence of new keel bone fractures. The statistical model is shown in Fig S2, supporting information online only. There was a significant interaction between perch height and fracture status ( 0.849/0.231 = z = 3.67; p < 0.0002). At the 50 cm, 100 cm and 150 cm perch heights, birds with fractures took 2.9 2.5, 49.8 85.4 and 100.3 123.6 seconds respectively compared with 0.7 0.5, 2.5 7.1 and 3.0 4.6 seconds to fly down for birds without fractures. There was no significant effect of meloxicam treatment ( 8.079/ 22.596 = z = 0.36; p = 0.72) (Fig 2). It is noteworthy that the birds with keel bone fractures in the meloxicam experiment had markedly longer latencies to land from 100 to 150 cm perch heights to the ground, irrespective of treatment (saline or NSAID) compared to fractured birds in the carprofen experiment. Discussion Birds with keel bone fractures in both the carprofen and the meloxicam experiment took longer to fly down from perches to the ground than birds without keel bone fractures, a finding that has been reported previously (Nasr et al. 2012a,b). There was also a significant interaction between the perch height and Figure 1 Latency of hens to fly down to the ground from different perch heights in the group of birds treated with carprofen. Mean SEM latency to fly down (seconds) from three different perch heights (50, 100 and 150 cm) in hens with (n = 43) and without (n = 20) keel bone fractures following treatment with carprofen (25 mg kg 1 ) and saline injected subcutaneously in the dorsal neck. Latency to fly from the perch to the ground (seconds) 14 12 10 8 6 4 2 0 No fracture & carprofen No fracture & saline Fracture & carprofen Fracture & saline 50 100 150 Perch heights (cm) Latency to fly from the perch to the ground (seconds) 120 100 80 60 40 20 0 No fracture & meloxicam No fracture & saline Fracture & meloxicam Fracture & saline 50 100 150 Perch heights (cm) Figure 2 Latency of hens to fly down to the ground from different perch heights in the group of birds treated with meloxicam. Mean SEM latency to fly down (seconds) from three different perch heights (50, 100 and 150 cm) in hens with (n = 23) and without (n = 16) keel bone fractures following treatment with meloxicam (5 mg kg 1 ) and saline injected subcutaneously in the dorsal neck. 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 42, 197 204 201

fracture status for latency to land that was replicated in both the carprofen and meloxicam experiment. A correlation between perch height and latency to land has been reported previously in birds with keel bone fractures (Nasr et al. 2012a). Collectively these new data demonstrate the sensitivity of the landing test in discriminating between birds with and without keel bone fractures and support the utility of the test as a biomarker of mobility in laying hens. On post mortem examination at the end of the experiment the birds with keel bone fractures that were studied had healed (rather than recent) keel bone fractures that were associated with callus formation at the fracture site. This reflects the detection method (keel bone palpation) used to select birds with fractures on farm (Wilkins et al. 2004), which is largely reliant on detecting new bone formation at the site of the healed keel bone fracture. In contrast, detection of new fractures by palpation is very challenging (Richards et al. 2011). Although it might be predicted that new bone fractures, that are likely unstable, are more painful than healed keel bone fractures, there are two compelling pieces of evidence to support the contention that healed keel bone fractures are also painful. The first is the finding that the opioid analgesic drug butorphanol decreases latency to fly down from perches in birds with healed keel bone fractures (Nasr et al. 2012b). Secondly, in a Conditioned Place Preference paradigm, birds with and without healed keel bone fractures were trained to associate the colour and position of an environment with the effects of either butorphanol and saline. In a subsequent choice experiment, birds with healed fractures preferred the environment where they had experienced butorphanol, in contrast to birds without keel bone fractures that exhibited no preference (Nasr et al. 2013). That the conditioned place preference was only shown by the fractured birds suggests that it was the analgesic properties of butorphanol that were rewarding and that healed keel bone fractures are a source of chronic pain. The lack of effect of meloxicam and carprofen in the present study was unexpected because the same test doses improved mobility in broiler chickens with spontaneous lameness (Caplen et al. 2013a) and reduced thermal hyperalgesia in an experimental model of arthropathy (Caplen et al. 2013b). Furthermore, 30 mg kg 1 carprofen improved weight bearing in the microcrystalline sodium urate model of articular pain chickens, although these birds were much younger than the laying hens studied in the present experiment (Hocking et al. 2005). Significantly lower doses of carprofen and meloxicam have improved weight bearing in experimental models of arthritis in other bird species (Cole et al. 2009; Paul- Murphy et al. 2009; Desmarchelier et al. 2012). Non steroidal anti-inflammatory drugs are antihyperalgesic; they do not elevate the normal pain threshold in animal models, but will normalize the exaggerated pain behaviour (hyperalgesia) that is observed after tissue injury or inflammation (Burian & Geisslinger 2005). Bone fractures are associated with inflammation in the time period shortly following tissue injury and fracture occurrence and NSAIDs are widely used to provide analgesia following bone fracture repair in man and companion animals. However, based on keel bone dissection and gross appearance of the fracture site, the keel bone fractures studied in the present model were healed and therefore may not have been associated with ongoing inflammation. This is supported by data from our previous study that showed the temperatures of keel bones in birds with healed fractures were lower than the temperatures of keel bones in normal birds as measured by thermal imaging (Purohit 2006; Nasr et al. 2012a). However peripheral inflammation also causes induction of COX enzyme and prostaglandin production in the spinal cord, causing prolonged changes in spinal nociceptive processing that might be modulated by NSAID administration (Telleria-Diaz et al. 2010). Therefore the absence of peripheral inflammation as detected by thermal imaging does not exclude the possibility that NSAIDs will exert a central antihyperalgesic effect to modulate pain related behaviour. The time of carrying out the landing test relative to the administration of carprofen and meloxicam was based on pharmacokinetic data for the two drugs (Baert & De Backer 2003; Hothersall et al. 2012). However each NSAID was administered only once before the first landing test and then 48 hours was allowed to elapse before a second test dose was administered and the landing test repeated. There was no significant difference in latency to fly down from perches between the first and second landing test. However chronic pain is associated with significant long term changes in sensory processing in the peripheral and central nervous system and it is possible a more sustained course of treatment with the test NSAID would have modulated sensory processing sufficiently to alter the willingness of birds with keel bone fractures to fly down from perches to the ground. Continuous as opposed to 202 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 42, 197 204

intermittent treatment with the NSAID celecoxib was associated with improved pain management in people with osteoarthritis (Luyten et al. 2007; Strand et al. 2011), although the underlying mechanisms were not investigated in these studies. Similarly, in dogs with osteoarthritis treated for 1 month with the NSAID cimicoxib, mobility scores continued to improve between the first and second week of treatment (EMEA Scientific Discussion for Trocoxil). One explanation for the discrepancy between the apparently good efficacy of a single dose of an NSAID to improve mobility in experimental models of arthritis and the present study could be the longevity of the underlying disease process and therefore chronicity of changes in sensory processing. Similarly, broiler birds with spontaneous lameness are very juvenile at the time of testing (approximately 30 days old) (Caplen et al. 2013a), therefore the underlying disease process is of relatively short duration compared with the chronicity of keel bone fractures in laying hens. Somewhat unexpectedly, in the carprofen experiment, birds with keel bone fractures were much quicker at performing the landing test from the 100 and 150 cm perch heights than birds with keel bone fractures in the meloxicam experiment, irrespective of whether they received saline or NSAID treatment. This indicates that birds with fractures in the meloxicam group were extremely hesitant to fly down from the higher perch heights, following both saline and meloxicam, whereas times to land were comparable between experiments from the 50 cm perch height. Data from hens in previous studies utilising landing time as a biomarker for mobility have generated latencies to land that are also longer than the birds treated with carprofen in the present experiment (Nasr et al. 2012a,b). The cause of the variability in latency that we have identified between different experimental groups of birds is difficult to identify but may relate to bird age, behaviour, activity or management within the commercial housing system prior to enrolment in the study, even though all the hens used these studies were from free-range housing systems. It is possible that the very quick latencies to land in the birds that received carprofen reduced the sensitivity of the assay in being able to determine an analgesic effect of carprofen in this group of birds, which may be a limitation of this model. Treating the same birds with both meloxicam and carprofen would have provided a more accurate comparison between the effects of both NSAIDs on bird mobility. Conclusions Data generated by this study provide further confirmation that keel fractures reduce the willingness of birds to move from perches (Nasr et al. 2012a,b), however treatment with the NSAIDs carprofen and meloxicam did not modulate mobility in this model. Acknowledgements The authors would like to thank Professor Toby Knowles for his help with statistical analysis of the data. The BBSRC (grant G000921) and Egyptian Government (The Egyptian Educational and Cultural Bureau) funded the study. References Baert K, De Backer P (2003) Comparative pharmacokinetics of three non-steroidal anti-inflammatory drugs in five bird species. Comp Biochem Physiol C Toxicol Pharmacol 134, 25 33. Bruchas MR, Land DD, Chavkin C (2010) The dynorphin/ kappa opioid system as a modulator of stress induced and pro-addictive behaviours. Brain Res 1314(Suppl.), 44 45. Burian M, Geisslinger G (2005) COX-dependent mechanisms involved in the antinociceptive action of NSAIDs at central and peripheral sites. Pharmacol Ther 107, 139 154. Caplen G, Baker L, Hothersall L et al. (2013a) Lame broiler chickens respond to non-steroidal anti-inflammatory drugs with objective changes in gait function: a controlled clinical trial. The Vet J 196, 477 482. Caplen G, Baker L, Hothersall B et al. (2013b) Thermal nociception as a measure of non-steroidal antiinflammatory drug (NSAID) effectiveness in broiler chickens with articular pain. The Vet J 198, 616 619. Cole GA, Paul-Murphy J, Krugner-Higby L et al. (2009) Analgesic effects of intramuscular administration of meloxicam in Hispaniolan parrots (Amazona ventralis) with experimentally induced arthritis. Am J Vet Res 70, 1471 1476. Danbury TC, Weeks CA, Chambers JP et al. (2000) Self selection of the analgesic drug carprofen by lame broiler chickens. Vet Rec 146, 307 311. Desmarchelier M, Troncy E, Beauchamp G et al. (2012) Analgesic effects of meloxicam administration on postoperative orthopedic pain in domestic pigeons (Columba livia). AJVR 73, 361 367. EMEA Scientific Discussion for Trocoxil. http://www. ema.europa.eu/docs/en_gb/document_library/epar Scientific_Discussion/veterinary/000132/WC500069 277.pdf 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 42, 197 204 203

Hocking PM, Robertson GW, Gentle MJ (2005) Effects of non-steroidal anti-inflammatory drugs on pain-related behaviour in a model of articular pain in the domestic fowl. Res Vet Sci 78, 69 75. Hothersall B, Caplen G, Nolan A et al. (2012) Pharmacokinetics of carprofen in broiler chickens (abstract). Proceedings of the Association of Veterinary Anaesthetists, Davos, Switzerland, Spring (March) 2012. Kelly DJ, Ahmad M, Brull SJ (2001) Preemptive analgesia I: physiological pathways and pharmacological modalities. Can J Anaesth 48, 1000 1010. Leece EA, Brearley JC, Harding EF (2005) Comparison of carprofen and meloxicam for 72 hours following ovariohysterectomy in dogs. Vet Anaesth Analg 32, 184 192. Luyten FP, Geusens P, Malaise M et al. (2007) A prospective randomised multicentre study comparing continuous and intermittent treatment with celecoxib in patients with osteoarthritis of the knee or hip. Ann Rheum Dis 66, 99 106. McGeown D, Danbury TC, Waterman-Pearson AE et al. (1999) Effect of carprofen on lameness in broiler chickens. Vet Rec 144, 668 671. Nasr MAF, Murrell J, Wilkins LJ et al. (2012a) The effect of keel fractures on egg production parameters, mobility and behaviour in individual laying hens. Anim Welf 21, 127 135. Nasr MAF, Nicol CJ, Murrell JC (2012b) Do laying hens with keel fractures experience pain? PLoS ONE 7(8): e42420. doi:10.1371/journal.pone.0042420. Nasr MAF, Browner WJ, Caplen G et al. (2013) Positive affective state induced by opioid analgesia in laying hens with bone fractures. Appl Anim Behav Sci 147, 127 131. Paul-Murphy J, Sladky KK, Krugner-Higby LA et al. (2009) Analgesic effects of carprofen and liposome-encapsulated butorphanol tartrate in Hispaniolan parrots (Amazona ventralis) with experimentally induced arthritis. Am J Vet Res 70, 1201 1210. Purohit RC (2006) Use of infrared imaging in veterinary medicine. In: Biomedical Engineering Handbook (3rd edn). Bronzino J.D. (ed). Taylor and Francis, London, UK. Rasbash J, Steele F, Browne SN et al. (2012). A User s Guide to MLWiN, v2.26. Centre for Multilevel Modelling, University of Bristol. Richards GJ, Nasr MA, Brown SN et al. (2011) Use of radiography to identify keel bone fractures in laying hens and assess healing in live birds. Vet Rec 169, 279 284. Sandilands V, Moinard C, Sparks NHC (2009) Providing laying hens with perches: fulfilling behavioural needs but causing injury? Br Poult Sci 50, 395 406. Strand V, Simon LS, Dougados M et al. (2011) Treatment of osteoarthritis with continuous versus intermittent celecoxib. J Rheumatol 38, 12. Telleria-Diaz A, Schmidt M, Kreusch S et al. (2010) Spinal antinociceptive effects of cyclooxygenase inhibition during inflammation: involvement of prostaglandinsand endocannabinoids. Pain 148, 26 35. Vane JR, Bakhle YS, Botting RM (1998) Cyclooxygenases 1 and 2. Annu Rev Pharmacol 38, 97 120. Wee S, Koob GF (2010) The role of the dynorphin- j opioid system in the reinforcing effects of drugs of abuse. Psychopharmacologia 210(Suppl.), 121 135. Wilkins LJ, Brown SN, Zimmerman PH et al. (2004) Investigation of palpation as a method for determining the prevalence of keel and furculum damage in laying hens. Vet Rec 155, 547 549. Wilkins LJ, McKinstry JL, Avery NC et al. (2011) Influence of housing system and design on bone strength and keel bone fractures in laying hens. Vet Rec 169, 414. Received 3 April 2013; accepted 21 February 2014. Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. Statistical model for the carprofen experiment. Figure S2. Statistical model for the meloxicam experiment. 204 2014 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesia and Analgesia, 42, 197 204