Effects of opioids and anesthetic drugs on body temperature in cats

Veterinary Anaesthesia and Analgesia, 2010, 37, 35 43 doi:10.1111/j.1467-2995.2009.00508.x RESEARCH PAPER Effects of opioids and anesthetic drugs on body temperature in cats Lysa P Posner*, Alana A Pavuk*, Jennifer L Rokshar*, Jennifer E Carter* & Jay F Levine *Department of Molecular and Biomedical Sciences, North Carolina State College of Veterinary Medicine, Raleigh, NC, USA Department of Population Health and Pathobiology, North Carolina State College of Veterinary Medicine, Raleigh, NC, USA Correspondence: Lysa Posner, Department of Molecular and Biomedical Sciences, North Carolina State College of Veterinary Medicine, 4700 Hillsborough St, Raleigh, NC 27606, USA. E-mail: lysa_posner@ncsu.edu Abstract Objective To determine which class of opioid alone or in conjunction with other anesthetic drugs causes post-anesthetic hyperthermia in cats. Study design Prospective, randomized, crossover study. Animals Eight adult, healthy, cats (four spayed females and four castrated males weighing 3.8 ± 0.6 kg). Methods Each cat was instrumented with a wireless thermistor in the abdominal cavity. Temperature in all phases was recorded every 5 minutes for 5 hours. Population body temperature (PBT) was recorded for 8 days. Baseline body temperature is the final 24 hours of the PBT. All injectable drugs were given intramuscularly. The cats were administered drugs in four phases: 1) hydromorphone (H) 0.05, 0.1, or 0.2 mg kg )1 ; 2) morphine (M) (0.5 mg kg )1 ), buprenorphine (BUP) (0.02 mg kg )1 ), or butorphanol (BUT) (0.2 mg kg )1 ); 3) ketamine (K) (5 mg kg )1 ) or ketamine (5 mg kg )1 ) plus hydromorphone (0.1 mg kg )1 ) (KH); 4) isoflurane in oxygen for 1 hour. Fifteen minutes prior to inhalant anesthetic, cats received either no premed (I), hydromorphone (0.1 mg kg )1 ) (IH), or hydromorphone (0.1 mg kg )1 ) plus ketamine (5 mg kg )1 ) (IHK). Results Mean PBT for all unmedicated cats was 38.9 ± 0.6 C (102.0 ± 1 F). The temperature of cats administered all doses of hydromorphone increased from baseline (p < 0.03) All four opioids (H, M, BUP and BUT) studied increased body temperature compared with baseline (p < 0.005). A significant difference was observed between baseline temperature values and those in treatment KH (p < 0.03). Following recovery from anesthesia, temperature in treatments IH and IHK was different from baseline (p < 0.002). Conclusions and clinical relevance All of the opioids tested, alone or in combination with ketamine or isoflurane, caused an increase in body temperature. The increase seen was mild to moderate (<40.1 C (104.2 F) and self limiting. Keywords buprenorphine, butorphanol, feline, hydromorphone, hyperthermia, isoflurane, ketamine, morphine, opioids. Introduction Severe hyperthermia [>41.7 C (107.0 F)] has been reported in cats following anesthesia and surgery (Niedfeldt & Robertson 2006; Posner et al. 2007). In both of those reports, hydromorphone, a pure mu opioid agonist, was positively associated with post-anesthetic hyperthermia (Niedfeldt & Robertson 2006; Posner et al. 2007). Although 35

mu opioid receptor agonist-evoked hyperthermia has long been documented (Clark & Cumby 1978), when administered alone in cats, morphine and fentanyl (pure mu opioids) cause only mild to moderate increases in body temperature (Wallenstein 1978; Gellasch et al. 2002). Whether hydromorphone has a unique effect on thermoregulation and whether that effect is dose dependent are unclear. Currently, partial mu opioid agonists (e.g. buprenorphine) and kappa opioid agonists (e.g. butorphanol) are commonly used for analgesia in cats. No reports of post-anesthetic hyperthermia have been associated with their use, but there have been reports of cats that were not administered a pure mu opioid agonist also becoming hyperthermic (Posner et al. 2007). Ketamine, a dissociative anesthetic, is commonly used in cats, and has been suggested to exacerbate the hyperthermia associated with hydromorphone administration (Posner et al. 2007). Cats that were administered hydromorphone plus ketamine had the greatest incidence of elevated temperature (100%) as well as the greatest absolute temperatures (41.6 C, 107.0 F) (Posner et al. 2007). Although ketamine has been shown to decrease body temperature in primates (Lopez et al. 2002), thermoregulation might be altered differently when ketamine is combined with hydromorphone. In both of the recent reports of severe hyperthermia, affected cats had been anesthetized and maintained on gas anesthesia (Niedfeldt & Robertson 2006; Posner et al. 2007). Anesthesia with gas anesthetics produces hypothermia primarily through inhibition of shivering, and heat loss through vasodilation (Sessler 1997). In the affected cats, body temperature at extubation was inversely related to the degree of hyperthermia, such that the coldest cats at extubation reached the highest temperatures during recovery (Posner et al. 2007). Hypothermia associated with gas anesthetics might exacerbate the hyperthermia caused by opioid agonists following anesthesia. The aforementioned reports (one prospective and one retrospective) were from client-owned animals in a clinical setting. This precluded studying drugs individually and necessitated intervention (active heating or cooling) to safeguard patient well-being. In addition, repeated rectal temperature measurement is stressful to most cats and may have interfered with the accuracy of the measurements. Temperature acquired by telemetry is likely a less stressful alternative and may provide more representative temperature measurements. Gastrointestinal and abdominal telemetric temperature acquisition is commonly used in laboratory animals and has been validated for use in humans and rats (Harkin et al. 2002; McKenzie & Osgood 2004). Accordingly we used telemetry to measure temperature variation in a laboratory trial in which we tested the following hypotheses: 1) Hydromorphone would cause a dose-dependent increase in body temperature. 2) That the pure mu opioid agonists would cause a greater increase in body temperature compared with partial agonists or a kappa agonist. 3) That ketamine would exacerbate the hyperthermia seen with hydromorphone administration. 4) That patients anesthetized with isoflurane and administered hydromorphone and ketamine would have the greatest increase in body temperature following anesthesia. Materials and methods Eight young adult (>6 months), purpose-bred domestic short hair cats were enrolled in the study. There were four spayed females and four castrated males. The cats weighed 3.8 ± 0.6 kg. The cats were group housed by gender in a climate controlled room with a 12-hour light cycle. The cats were fed commercial adult cat food and free choice water. Experiments were all conducted in climate-controlled rooms where temperature ranged from 21.1 to 22.2 C (70 72 F). Experiments were conducted with approval of the North Carolina State University Institutional Animal Care and Use Committee. Thermistor placement After a 2-week acclimation period, the cats were fasted for 12 hours and moved to individual cat kennels. Each cat was anesthetized with medetomidine (medetomidine hydrochloride; Pfizer Animal Health, NY, USA) (10 lg kg )1 ), ketamine (ketamine HCL; Fort Dodge Animal Health, IA, USA) (5 mg kg )1 ), and hydromorphone (hydromorphone HCL; Baxter Healthcare Corporation, IL, USA) (0.05 mg kg )1 ) IM. Isoflurane was administered by facemask, the cats tracheas were then intubated and they were maintained on isoflurane in oxygen. Heart rate, respiratory rate, and arterial hemoglobin oxygen saturation and blood pressure were monitored during anesthesia. The abdomen was shaved and aseptically prepared for surgery. An 36 Ó 2010 The Authors. Journal compilation Ó 2010 Association of Veterinary Anaesthetists, 37, 35 43

approximately 5 cm ventral midline incision was made through the skin, subcutaneous tissue and linea alba. A thermistor (3 cm VM-FH temperature transmitter, Vital View; Respironics, OR, USA) was placed in the abdomen (free floating), but was directed dorsally. The incision was closed routinely with 3-0 PDS (Ethicon; Johnson & Johnson, NJ, USA) in the linea and subcutaneous tissue. All cats received 10 ml kg )1 of lactated Ringers solution (Hospira, IL, USA) SC during anesthesia and meloxicam 0.2 mg kg )1 (Boehringer Ingelheim Vetmedica, MO, USA) SC at extubation. Cats were allowed to recover for 2 weeks before data acquisition. Manual palpation of the abdomen of the cats easily identified the thermistor during the course of this experiment. For 4 weeks (before and after thermistor implantation) cats intermittently were acclimated to standard stainless steel individual kennels for 4 6 hour time periods by placing them in the cages with a blanket, food, water, litter pan, and toys. Data acquisition During times of data acquisition, cats were placed in individual stainless steel cages (24 00 W 22 00 H 24 00 D) with a towel and a litter pan, and a receiver (TR-3000 Telemetry Receiver, Vital View; Respironics) for the thermistor was placed in the grate below the floor of the kennel. The receiver was connected to a data acquisition system (Vital View Software and Hardware Card; Respironics). Each thermistor was purchased with calibration data that indicated 0.1 C accuracy within the 10 45 C range. For all phases of the study, on the day of the experiment, cats were placed in individual kennels approximately 2 hours prior to drug injection to assure data acquisition was functioning and to allow the cats to become accustomed to their surroundings. The data acquisition system was programmed to record thermistor temperature every 5 minutes for 24 hours. Body temperatures Population body temperature (PBT) was taken from data recorded every 5 minutes for 12,485 minutes (8 days) in alternating 24 hour periods. Cats had routine access to food, water, litter pans and normal human interaction during those time periods. All data were compiled to determine the mean temperature and a standard deviation (SD) of this population of cats. Baseline body temperature (BBT) was taken from the final 24 hours of the PBT (above). Drug administration phases For each phase, drug administration was randomly assigned such that each cat received all doses and all treatments. All drugs were given intramuscularly (IM) in alternating thigh muscles with a 23- gauge needle. Due to limitations in receivers, four cats were studied per day and there was a 1 week washout period between experiments. Phase 1 Cats were administered hydromorphone 0.05 mg kg )1 (HL), 0.1 mg kg )1 (H), or 0.2 mg kg )1 (HH) IM. Phase 2 Cats were administered morphine (morphine sulfate; Baxter Healthcare Corporation) (M) (0.5 mg kg )1 ), buprenorphine (buprenorphine hydrochloride; Hospira) (BUP) (0.02 mg kg )1 ), or butorphanol (butorphanol tartrate; Phoenix Pharmaceutical, St. Joseph, MO 64507) (BUT) (0.2 mg kg )1 ) IM. Phase 3 Cats were administered ketamine (K) (5 mg kg )1 )or ketamine (5 mg kg )1 ) plus hydromorphone (0.1 mg kg )1 ) (KH) IM. Phase 4 Cats were anesthetized for 1 hour with isoflurane (Abbot Animal Health, IL, USA) in oxygen; initially administered by face-mask and then by orotracheal intubation with an appropriately sized cuffed endotracheal tube. Fifteen minutes prior to inhalant anesthetic, cats received either no premed (I), hydromorphone (0.1 mg kg )1 ) IM (IH), or received hydromorphone (0.1 mg kg )1 ) plus ketamine (5 mg kg )1 ) IM (IHK). Following injection, cats were kept in cages with the thermistor receiver until they were administered isoflurane. Following intubation cats were placed directly on the thermistor receiver. During anesthesia, cats were continuously monitored for heart rate, respiratory rate, arterial hemoglobin oxygen saturation, and end-tidal carbon dioxide tension (PE CO 2 ); oscillometric blood pressure was evaluated every Ó 2010 The Authors. Journal compilation Ó 2010 Association of Veterinary Anaesthetists, 37, 35 43 37

3 minutes (Cardell 9402; Sharn Veterinary Inc., FL, USA). Following intubation, each cat had a 22- gauge catheter placed in a cephalic vein and received lactated Ringer s solution at 10 ml kg )1 hour )1. All cats breathed spontaneously. Depth of anesthesia was adjusted to keep cats lightly anesthetized and to maintain PE CO 2 <55 mmhg (7.3 kpa) and MAP >60 mmhg. A towel was placed between the cat and receiver. Cats were neither actively warmed nor cooled. After 1 hour of isoflurane anesthesia, the isoflurane was discontinued and cats continued breathing 100% oxygen until extubated. Cats were then returned to their individual kennel and data acquisition was continued. Data analysis Graphic visualization of the data indicated that the first 5 hours following drug administration was the primary area of interest (the only time period where data appeared to deviate from baseline measurement) and subsequent data analysis focused on this time period after administration. Summary statistics for the BBT ( C) of each cat and body temperature after each drug or drug combination from the different phases of the study were estimated using JMP Version 7.0, 2008 (SAS Institute, NC, USA). During phase one a paired t-test using the JMP Matched Pairs Platform was used to compare the mean body temperature of the eight cats for the 5 hours following administration of each dose with each other dose (0.05 versus 0.1, 0.05 versus 0.2, 0.1 versus 0.2). Missing data were autoconverted to m which was not counted by the statistics program. Due to the multiple comparisons for the three doses of hydromorphone a Bonferroni correction was used to adjust the p value for multiple tests, a p value of <0.016 was considered significant. Additional statistical analysis focused on comparing the temperature-related response of each drug or drug combination with baseline values. The difference between mean baseline temperature values over 5 hours obtained during phase 1, phase 2, and phase 3 was compared with the mean temperature values (the same 5 hours period from the control recordings) after administration of each drug using a paired t-test (JMP Version 7, 2008). A p value of <0.05 was considered statistically significant. An initial decrease, then increase in body temperature back toward baseline was apparent during the first 100 minutes after administration of isoflurane and isoflurane drug combinations. The increase back toward baseline corresponded approximately with the recovery from anesthesia (Fig. 4). Accordingly, the data from phase 4 obtained for isoflurane and isoflurane drug combinations were compared for the first 100 minutes post-administration and then for minutes 105 300, to accurately reflect the physiologic changes associated with the approximate recovery from anesthesia. Moderate and severe hyperthermia are described as body temperatures exceeding 2 SD and 3 SD, respectively. Results Mean (±SD) PBT for all cats (for 8 days) was 38.9 ± 0.6 C (102.0 ± 1 F). The BBT for all cats at each time point are shown in Figs 1 4. The highest recorded individual temperature for a cat in each treatment is presented in Table 1. Phase 1 All doses of hydromorphone produced temperatures that were different from baseline (p < 0.03), but were not different from each other (p > 0.08) (Fig. 1). Data generated by treatment H are also shown in Figs 2 4. Phase 2 Mean data for treatments B, M, BUT, and BUP are shown in Fig. 2. All temperatures were significantly increased from baseline (p < 0.005). Phase 3 A significant difference was observed between temperature during baseline and treatment KH (p < 0.03) (Fig. 3). Ketamine produced a mild increase in body temperature for about 1 hour after administration, but body temperature was not significantly different from baseline (p > 0.88). At 5mgkg )1 IM, cats administered ketamine alone or with hydromorphone were sedated but not fully anesthetized. Phase 4 During the first 100 minutes of phase 4, a significant difference was observed between baseline body temperature values and body temperature in treatments I, IH, and IHK (p < 0.02) (Fig. 4). During the 38 Ó 2010 The Authors. Journal compilation Ó 2010 Association of Veterinary Anaesthetists, 37, 35 43

Figure 1 Mean body temperatures following administration of hydromorphone at 0.05, 0.1, or 0.2 mg kg )1. Mean baseline temperature included for comparison. The straight line is the mean temperature for this population of cats and the dotted line is 1 SD from the mean. Treatments are all significantly different from baseline. Figure 2 Mean body temperatures following administration of hydromorphone (0.1 mg kg )1 ), morphine (0.5 mg kg )1 ), buprenorphine (0.02 mg kg )1 ), or butorphanol (0.2 mg kg )1 ). Mean baseline temperature included for comparison. The straight line is the mean temperature for this population of cats and the dotted line is 1 SD from the mean. All treatments are significantly different from baseline. next 200 minutes of phase 4, there was a significant difference between baseline body temperature and treatments IH and IHK (p < 0.002) (Fig. 4). Discussion Administration of both mu (full and partial agonists) and kappa opioid agonists produced mild to moderate increases in body temperature in cats in this study. This is in contrast to previous reports that did not show an association between buprenorphine and post-anesthetic hyperthermia in cats (Niedfeldt & Robertson 2006) and is the first report linking butorphanol to post-anesthetic hyperthermia. Interestingly, maximum individual body temperatures differed by less than 0.8 C for cats in each treatment and the maximum temperature recorded for any cat in this study [40.7 C (105.3 F)] did not approach the severe hyperthermia [>41.7 C (107.0 F)] reported previously (Niedfeldt & Robertson 2006; Posner et al. 2007). Only one, clinically relevant, dose of morphine, buprenorphine or butorphanol was chosen to be evaluated. It is possible that lower or higher doses would produce different results. After administration of the three doses of hydromorphone tested, the maximum temperature for any cat was 40.5 C (104.9 F) and the maximum temperature by treatments only differed by 0.5 C. Based upon these findings, administering lower Ó 2010 The Authors. Journal compilation Ó 2010 Association of Veterinary Anaesthetists, 37, 35 43 39

Figure 3 Mean body temperature following administration of ketamine (5 mg kg )1 ), ketamine (5 mg kg )1 ) plus hydromorphone (0.1 mg kg )1 ). Baseline and hydromorphone values from Phase 1 are included for comparison. The straight line is the mean temperature for this population of cats and the dotted line is 1 SD from the mean. The KH treatment produced temperatures different from baseline. Figure 4 Mean body temperature during and following administration of isoflurane, isoflurane plus hydromorphone (0.1 mg kg )1 ), and isoflurane plus hydromorphone (0.1 mg kg )1 ) plus ketamine (5 mg kg )1 ). Baseline values are included for comparison. The straight line is the mean temperature for this population of cats and the dotted line is 1 SD from the mean. During the first 100 minutes, treatments I, IH, and IHK were different from baseline. During the next 200 minutes, treatments IH, and IHK were different from baseline. doses of hydromorphone (e.g. 0.05 mg kg )1 ) does not prevent hyperthermia. However, inspection of Fig. 1 shows that body temperature in the HH treatment was the greatest and had not returned to the reference interval within 5 hours. It is possible that increasing doses might produce a greater increase in body temperature in addition to a more prolonged effect. A previous study showed a greater incidence and greater absolute temperature in cats administered 40 Ó 2010 The Authors. Journal compilation Ó 2010 Association of Veterinary Anaesthetists, 37, 35 43

Table 1 Maximum individual body temperature for any cat within a treatment within the 5 hour experimental period Treatment Highest individual temperature Cat Baseline 39.8 C (103.6 F) Topsail HL 40.0 C (104.0 F) Cayman/Fiji H 40.5 C (104.9 F) Topsail HH 40.2 C (104.4 F) Bali H 40.5 C (104.9 F) Topsail M 40.1 C (104.2 F) Fiji BUT 40.3 C (104.5 F) Fiji BUP 40.2 C (104.4 F) Cayman K 39.9 C (103.9 F) Fiji HK 40.3 C (104.5 F) Fiji I 39.9 C (103.9 F) Bali IH 40.5 C (104.9 F) Bali IHK 40.7 C (105.3 F) Curacao both ketamine and hydromorphone (Posner et al. 2007). Ketamine is a phencyclidine dissociative anesthetic. When ketamine was administered alone to cats at sedative doses (10 mg kg )1 ) it occasionally increased body temperature, whereas when administered at anesthetic doses (20 mg kg )1 ), it decreased body temperature in cats by 1.6 C (2.9 F) (Beck & Coppock 1971). Although the cats in this study were sedated at 5 mg kg )1 (IM) they were not completely anesthetized, and a mild increase in body temperature was observed during the first hour after ketamine administration. It is possible that the increased motor activity associated with recovery from ketamine in conjunction with increased muscle tone may increase body temperature following anesthesia. In this study ketamine did not exacerbate the hyperthermia of hydromorphone as average temperatures for cats administered hydromorphone or hydromorphone and ketamine were quite similar, and none of the cats in either treatment had a temperature exceeding 40.3 C (104.5 F). In previous reports of post-anesthetic hyperthermia, all of the cats were administered inhalant anesthetics (Niedfeldt & Robertson 2006; Posner et al. 2007). Although all cats had a significant decrease in body temperature during isoflurane anesthesia, inspection of Graph 4 shows that following recovery from inhalant anesthesia there were minimal differences between baseline and treatment I. When hydromorphone or hydromorphone and ketamine were administered before isoflurane the post-anesthesia body temperatures were similar to those of cats administered hydromorphone with or without ketamine without the isoflurane (Figs 3 & 4). It is therefore unlikely that isoflurane itself contributes to hyperthermia. Hyperthermia has been suggested to be most severe in cats administered hydromorphone, ketamine, and isoflurane (Niedfeldt & Robertson 2006; Posner et al. 2007). In this study, temperature in cats in the IHK treatment was not statistically different from the IH treatment; however, inspection of Fig. 4 shows that mean body temperatures in the cats in the IHK treatment were generally higher (0.6 C) than those in the IH treatment. Furthermore, the highest individually recorded body temperature was from a cat in the IHK treatment. Although there was not a significant difference, the difference of 0.6 C might be clinically relevant and that individuals might be more prone to hyperthermia suggests that the combination be evaluated further. Thermoregulation is a complex interaction between thermal sensing, central processing, and behavioral and physiologic responses. The pre-optic region of the hypothalamus is the primary center for thermoregulatory control. The thermoregulatory center is composed of two distinct areas: the heatloss and the heat-promoting centers. Peripheral (in the skin and mucous membranes) and central thermoreceptors (sense changes in blood temperature) sense changes in body temperature and the thermal information is transmitted up the spinal cord to the hypothalamus which coordinates the information. The hypothalamus then compares these values with the threshold temperatures that trigger thermoregulatory responses. If the body temperature has exceeded the threshold temperatures (either above or below) a series of behavioral and physiologic responses will be triggered. Responses to hyperthermia include behavioral responses (decreased muscle activity, increasing surface area, extension of limbs) and physiologic responses (vasodilation, panting, decreased metabolism). Pyrogens and cytokines can cause the hypothalamus to raise the thermoregulatory set point by the productions of prostaglandin E 2 (PGE 2 ), and interleukin-1 and -6 (IL-1 and IL-6). Opioids alter thermoregulation by resetting the threshold point controlled by the hypothalamus (Cox et al. 1976). Therefore, it is possible that opioids affect the cat s ability to limit a thermogenic response. Furthermore, post-anesthetic hyperthermia has been inversely associated with body tem- Ó 2010 The Authors. Journal compilation Ó 2010 Association of Veterinary Anaesthetists, 37, 35 43 41

perature at extubation, such that the colder the cats were at extubation the warmer they were during the postoperative period (Posner et al. 2007). This exaggerated rebound from hypothermia might be enabled by the resetting action of opioids upon the hypothalamic thermoregulatory threshold. Some of the cats in the Posner et al. study were severely hypothermic at extubation [<34.4 C (94 F)], likely due to surgery time, surgical incisions, the use of flush solutions, and different warming strategies. In the current study, body temperature at extubation for all treatments was 36.4 C (97.5 F). Thus, a possible reason that no severe hyperthermia was seen in this study is because the cats were not as cold at extubation and the same level of rebound response, as hypothesized by Posner et al., did not occur. Another factor in support of opioids as a major factor in feline hyperthermia is that the authors anecdotally have noted that naloxone will quickly lower the body temperature of hyperthermic cats following anesthesia. The authors have treated cats with body temperatures exceeding 41.1 C (106.0 F) following anesthesia with naloxone at 0.01 mg kg )1 IM or SC and have seen body temperature return to normal range in <30 minutes. In addition to general thermoregulation, the hypothalamus modulates fever in response to PGE 2 which is triggered by endogenous or exogenous pyrogens. Phagocytic cells produce cytokines (e.g. IL-1, IL-6) and tumor necrosis factor-a which are released into general circulation to the circumventricular organs of the brain and activate the arachidonic acid pathway. The arachidonic acid pathway is mediated by the enzymes phospholipase A2 (PLA2), cyclooxygenase-2 (COX-2), and PGE 2 synthase which mediates the synthesis and release of PGE 2 (Ganong 2001). PGE 2 is the ultimate mediator of the febrile response (Blatteis 2007). Tissue trauma can trigger the febrile response (Biddle 2006). In human patients undergoing orthopedic or urologic surgery, approximately 12% developed post-surgical fevers unrelated to infection (Saavedra et al. 2008). Noninfectious post-surgical pyrexia is generally considered a part of the acute phase inflammatory response which is mediated by cytokines; including IL-1 and IL-6. In human cardiac patients, IL-6 was positively correlated with postoperative pyrexia (up to 24 hours) in patients without infection (Mitchell et al. 2007). A possible reason that the cats in this study did not show severe hyperthermia may be that they did not undergo surgery and have pyrogenic cytokines released. However, in humans, elevation of IL-6 (and pyrexia) following surgery is from 4 to 24 hours post-surgery (Mitchell et al. 2007). The resolution of the hyperthermia in the cats in this study and other studies within 5 hours argues against inflammatory cytokines as the primary cause of the hyperthermia in cats. Furthermore, nonsteroidal anti-inflammatory drugs do not attenuate the hyperthermia seen in cats (Niedfeldt & Robertson 2006). This experimental study removed many of the other factors that might contribute to hyperthermia in cats. The cats in this study were tested in a controlled environment compared with clinical reports. In addition to not having an invasive procedure, they were not exposed to: environmental temperature fluctuations, IV fluid administration, antibiotic administration, medications for other disease processes, inconsistent post-anesthetic heating, stress from unfamiliar animals and people, stress from being away from the home setting, and repeated rectal temperature assessment. It is possible that other factors in addition to drug therapy contribute to post-anesthetic hyperthermia. The wireless thermistors provided a robust, nonstressful, way of recording body temperature in cats. This rich data set also facilitated easy graphic visualization of the data and demonstration of the consistent extended duration of the moderate level of hyperthermia that was observed. Data acquisition at times was compromised when cats climbed the bars of the cage and the thermistor was out of range of the receiver. This occurred infrequently and over the 24 hour test period 90% of data points were recorded. There were outlier values in our baseline measurements that we hypothesize may have been due to the cat s abdomen being in contact with the metal base of the cage but as this could not be confirmed the data were included in the data set. These outlier values may explain some of the downward spikes in our baseline data. Telemetric temperature measurement has been successfully validated in rats and humans (Harkin et al. 2002; McKenzie & Osgood 2004). Although this model has not been validated for cats, the 8 day body temperature average and range was comparable to values reported in cats (Merck 2002). Although the large number of observations recorded using the thermistors heightened the likelihood that minor differences in body temperature would be statistically significant, that difference was not always biologically relevant. 42 Ó 2010 The Authors. Journal compilation Ó 2010 Association of Veterinary Anaesthetists, 37, 35 43

Severe hyperthermia, seen in some cats, might be an idiosyncratic reaction from an individual or from a particular (genetic, geographic) group, as some cats have been shown to have markedly different responses to opioids (Johnson et al. 2007) as well as other drugs, such as nitrous oxide (Pypendop et al. 2003). In the two previous reports, cats studied were likely from the same geographic area (able to get to a teaching hospital) and in the present study, it is likely that some of the cats were related. Interestingly, when the cat with the highest individual temperature was identified for each treatment, five of the eight cats were represented. The increased body temperature in all of the experimental treatments within the present study was self limiting, and 8 of the 12 treatments had returned to normal within 5 hours. There was no apparent morbidity or mortality associated with the hyperthermia seen in the experimental subjects. The term hyperthermia has been used unequally in previous studies. Niedfeldt considered cats hyperthermic when body temperature exceeded 40 C (104 F) (Niedfeldt & Robertson 2006), whereas Posner labeled cats hyperthermic when body temperature exceeded 1 SD [39.2 C (102.5 F)] (Posner et al. 2007) from the normal body temperature published in the Merck Manual (Merck 2002). In this study, hyperthermia was defined based on the normal body temperature for this population of cats. Cats having a body temperature exceeding the mean ± 2 SD or 3 SD were considered to be moderately and severely hyperthermic, respectively. The thermistor-based measurements in this controlled trial indicate that hydromorphone, morphine, butorphanol, and buprenorphine all cause an increase in body temperature in cats. All doses of hydromorphone increased body temperature equivalently. Administration of ketamine or isoflurane in addition to hydromorphone does not produce a clinically relevant increase in body temperature compared with that of administration of hydromorphone alone. Acknowledgements Dr Pavuk and Ms Rokshar were supported by the Merck-Merial Student Research Scholars Program. References Beck C, Coppock R (1971) Evaluation of VETALAR (ketamine HCL). Vet Med 66, 993 996. Biddle C (2006) The neurobiology of the human febrile response. AANA J 74, 145 150. Blatteis CM (2007) The onset of fever: new insights into its mechanism. Prog Brain Res 162, 3 14. Clark WG, Cumby HR (1978) Hyperthermic responses to central and peripheral injections of morphine sulphate in the cat. Br J Pharmacol 63, 65 71. Cox B, Ary M, Chesarek W et al. (1976) Morphine hyperthermia in the rat: an action on the central thermostats. Eur J Pharmacol 36, 33 39. Ganong W (2001) Review of Medical Physiology. Lange Medical Books, New York. Gellasch KL, Kruse-Elliott KT, Osmond CS et al. (2002) Comparison of transdermal administration of fentanyl versus intramuscular administration of butorphanol for analgesia after onychectomy in cats. J Am Vet Med Assoc 220, 1020 1024. Harkin A, O Donnell J, Kelly J (2002) A study of VitalView for behavioural and physiological monitoring in laboratory rats. Physiol Behav 77, 65 77. Johnson JA, Robertson SA, Pypendop BH (2007) Antinociceptive effects of butorphanol, buprenorphine, or both, administered intramuscularly in cats. Am J Vet Res 68, 699 703. Lopez KR, Gibbs PH, Reed DS (2002) A comparison of body temperature changes due to the administration of ketamine-acepromazine and teletamine-zolazepam anesthetic in cynomolgus macaques. Contemp Topics 41, 47 50. McKenzie J, Osgood D (2004) Validation of a new telemetric core temperature monitor. J Therm Biol 29, 605 611. Merck (2002) Merck Veterinary Manual. Merck & Co, Rahway, NJ. Mitchell JD, Grocott HP, Phillips-Bute B et al. (2007) Cytokine secretion after cardiac surgery and its relationship to postoperative fever. Cytokine 38, 37 42. Niedfeldt RL, Robertson SA (2006) Postanesthetic hyperthermia in cats: a retrospective comparison between hydromorphone and buprenorphine. Vet Anaesth Analg 33, 381 389. Posner LP, Gleed RD, Erb HN et al. (2007) Post-anesthetic hyperthermia in cats. Vet Anaesth Analg 34, 40 47. Pypendop BH, Ilkiw JE, Imai A et al. (2003) Hemodynamic effects of nitrous oxide in isoflurane-anesthetized cats. Am J Vet Res 64, 273 278. Saavedra F, Myburg C, Lanfranconi MB et al. (2008) Postoperative fever in orthopedic and urologic surgery. Medicina (B Aires) 68, 6 12. Sessler DI (1997) Mild perioperative hypothermia. N Engl J Med 336, 1730 1737. Wallenstein M (1978) Temperature response to morphine in paralyzed cats. Eur J Pharmacol 49, 331 333. Received 20 May 2008; accepted 15 March 2009. Ó 2010 The Authors. Journal compilation Ó 2010 Association of Veterinary Anaesthetists, 37, 35 43 43