Electroencephalographic and cardiovascular variables as nociceptive indicators in isoflurane-anaesthetized horses

Transcription

1 Veterinary Anaesthesia and Analgesia, 2005, 32, RESEARCH PAPER Electroencephalographic and cardiovascular variables as nociceptive indicators in isoflurane-anaesthetized horses Henning A Haga DVM, PhD & Nils I Dolvik DVM, PhD Department of Companion Animal Clinical Sciences, The Norwegian School of Veterinary Science, Oslo, Norway Correspondence: Henning A Haga, Department of Companion Animal Clinical Sciences, The Norwegian School of Veterinary Science, PO Box 8146 Dep, N-0033 Oslo, Norway. andreas.haga@veths.no Abstract Objective To evaluate Fourier-transformed electroencephalographic (EEG) variables, mean arterial blood pressure (MAP) and pulse rate as nociceptive indicators in isoflurane-anaesthetized horses. Animals Five standardbred and three Norwegian cold-blooded trotter stallions undergoing castration, aged 2 4 years, mass kg. Materials and methods All horses received intravenous (IV) detomidine (10 lg kg )1 IV) and butorphanol (0.01 mg kg )1 IV). Additional detomidine (4 lg kg )1 IV) was administered in the induction area. Anaesthesia was induced with ketamine (2.5 mg kg )1 IV) and diazepam (40 lg kg )1 IV), and maintained for 30 minutes with isoflurane (end-tidal concentration of 1.4%) vaporized in oxygen. The electroencephalogram, MAP and pulse rate were recorded for 15 minutes, beginning 5 minutes before skin incision. Differences between the mean values of recordings taken before, and during surgery were calculated and tested for significant differences using a two-sided Student s t-test. Results A significant rise in MAP and a fall in pulse rate were found. No significant change was found in any EEG variable. Conclusion/clinical relevance Of the variables evaluated, MAP seems to be the most sensitive and reliable indicator of nociception in isoflurane-anaesthetized horses. Keywords cardiovascular, EEG, horse, isoflurane, nociception. Introduction Surgery performed under general anaesthesia induces nociceptive signals which enter the central nervous system and evoke stress responses which may increase the appreciation of post-operative pain (Thurmon et al. 1996). Consequently, the identification of reliable indicators of nociception during general anaesthesia is a necessary prerequisite to a desirable reduction in nocistimulation. The electroencephalogram (EEG) obtained through skin electrodes is a recording of cerebrocortical electrical activity. When noxious stimuli are applied during general anaesthesia, a phenomenon termed arousal may be observed (Bimar & Bellville 1977). This is an increase in the frequency and a decrease in amplitude of the EEG signal. The arousal reaction has been used as an indicator of anaesthetic quality in horses by recording their EEG while anaesthetized with volatile agents. Using Fourier transformation, a number of variables may be calculated from the raw EEG signal, producing numerical characteristics of the EEG and arousal. A response to nociception has been found in the following Fourier-transformed variables: alpha/delta ratio; beta/delta ratio; theta/delta ratio; median frequency (MED); spectral edge frequency 80% (SEF 128

2 80%) and total power (Ekström et al. 1993; Otto et al. 1993, 1994, 1996, 1998; Short & Ekström 1993; Miller et al. 1995; Murrell et al. 1999, 2000a,b). Nociceptive impulses reaching the CNS may increase sympathetic nervous activity, raising both blood pressure and pulse rate. These cardiovascular variables have traditionally been used as indicators of nociception during anaesthesia. The objective of this study was to evaluate the reliability of the alpha/delta ratio, beta/delta ratio, theta/delta ratio, MED, SEF 95%, total power, mean arterial blood pressure (MAP) and heart rate as nociceptive indicators in isoflurane-anaesthetized horses. Materials and methods Five Standardbred and three Norwegian cold-blooded trotter stallions, which were judged to be healthy on the basis of physical examination, were included in this study. These were clinical cases and so owner consent was obtained in all cases. A clinical examination was performed on the day before surgery, the horse was weighed and detomidine (10 lg kg )1 ) (Domosedan; Orion Corporation, Turku, Finland) was administered intravenously (IV) before a catheter was placed in the left jugular vein. The location for the EEG electrodes was shaved and defatted with diethyl ether (eth-oxy-ethane). On the day of surgery, sodium penicillin ( IU IV) (Penicillin A-L; Apothekernes Laboratorium A.S., Oslo, Norway) was administered and after further skin defatting with diethyl ether, self adhesive silver chloride EEG electrodes (Zipprep; Aspect Medical Systems, Natick, MA, USA) were applied. The recording electrodes were placed bilaterally over the frontal bones 1 cm medial to the temporal line and 2 cm caudal to the lateral canthus of the eyes. The reference electrode was placed in the midline rostral to the medial canthus of the eyes and the ground electrode was placed in the atlanto-occipital region on the left side (Fig. 9). This configuration has been described previously in horses (Ekström et al. 1993). The EEG was recorded with an EEG monitor designed for intra-operative monitoring of humans beings (A-1000; Aspect Medical Systems). The data were automatically transferred to, and stored on computer (Dell Latitude Cpi D266XT, Round Rock, TX, USA). The variables calculated from a combination of both recording electrodes were used in the analyses. Before each EEG recording, the impedance was checked and kept below 2000 ohms at 16 Hz. The filters of the EEG monitor were set as follows: high frequency filter, 70 Hz; 50/60 Hz filter, 50 Hz; and low frequency filter, 2.00 Hz. Detomidine (10 lg kg )1 IV) and butorphanol (10 lg kg )1 IV) (Torbugesic; Fort Dodge Animal Health, Kansas City, MO, USA) were administered for pre-anaesthetic medication, and in the induction area additional detomidine (4 lg kg )1 ) was given IV before anaesthesia was induced with ketamine (2.5 mg kg )1 IV) (Narketan; Chassot AG, Berne, Switzerland) and diazepam (40 lg kg )1 IV) (Stesolid; Dumex-Alpharma A/S, Copenhagen, Denmark). After orotracheal intubation, the horse was placed on the operating table in dorsal recumbency and the endotracheal tube connected to a circle breathing system. Anaesthesia was maintained with isoflurane (Forene; Abbott Scandinavia AB, Kista, Sweden) vaporized in oxygen. Gas was continuously drawn from the proximal end of the endotracheal tube to an Figure 1 Changes in mean arterial blood pressure (MAP) for each horse (dotted lines) and the mean of all horses (solid line) with time. Onset of surgery (*). The grey bar marks the period when the tunica vaginalis was clamped. Ó Association of Veterinary Anaesthetists, 2005, 32,

3 anaesthesia monitor (Datex-Engstrom AS/3, Helsinki, Finland) where the concentrations of inspiratory and expiratory isoflurane, oxygen and carbon dioxide were analysed. Mean arterial blood pressure and pulse rate were measured using a catheter placed in the facial artery connected to a pressure transducer (Baxter Pressure Monitoring Kit; Uden, Holland) zeroed at the level of the thoracic inlet. Intermittent positive pressure ventilation was begun and adjusted to ensure end-tidal CO 2 concentration remained between 4.5 and 5.5%. When instrumentation was completed, glycopyrrolate (5 lg kg )1 ) (Robinul; Wyeth Lederle, Hants, England) was administered IV. During anaesthesia, dobutamine (Dobutrex; Eli Lilly & Co., Indianapolis, IN, USA) was infused with a syringe driver (P6000/TIVA; Ivac, San Diego, CA, USA) in order to maintain MAP above 60 mmhg. A polyionic balanced electrolyte solution (Ringer acetate; Fresenius Kabi, Oslo, Norway) was infused. After an end-tidal isoflurane concentration of 1.4% had been maintained for 25 minutes, dobutamine infusion rate, fluid infusion rate and ventilation were held constant to avoid bias in the MAP and pulse rate measurements. Five minutes later, EEG recording began. Mean arterial blood pressure and pulse rate values were recorded every 30 seconds. After 5 minutes of recording, surgery began and recording continued for a further 10 minutes. Ketamine was used to control movement during surgery; data recording was stopped when this occurred. Surgery involved skin incision and isolation of the common vaginal tunic using blunt dissection. The testis was then held with a towel clamp and the cremaster muscle and funiculus clamped, then ligated using absorbable suture material. The common vaginal tunic and contents were cut with a scalpel, after which the subcutaneous tissue and skin were sutured. The times at skin incision, vaginal tunic clamping and testis removal were noted. To reduce post-operative discomfort, flunixin (1.1 mg kg )1 ) (Finadyne; Schering-Plough, Segre, France) was administered IV before the horse recovered. Epochs of 2 seconds were used in the Fourier transformation analysis. During recording, the monitor automatically detected epochs with artefacts or noise and marked these as low-quality signal epochs. These were discarded before further analysis. Mean values of MAP and pulse rate were recorded. Raw EEG data were transformed and computations made for SEF 95%, MED, total power, alpha/delta ratio, beta/delta ratio and theta/delta ratio for 5 minutes before and 10 minutes during surgery. Mean values for EEG variables, MAP and pulse rate were calculated for the presurgery and surgery time periods. The differences between means were calculated and tested for difference from zero using two-sided Student s t-test and an alpha value of 5%. Commercially available software (Microsoft Excel for Windows 95 version 7.0, Redmond, WA, USA and Jump version 3.25, SAS Institute Inc., Cary, NC, USA) were used for data handling and statistical analysis. Results Four horses moved during surgery, three of these during surgery on the second testis. This did not interfere with recordings. One horse, which moved 3.5 minutes after surgery began, received ketamine; recording from this animal was discontinued. The differences between the mean values of MAP, pulse rate, SEF 95%, MED, total power, alpha/delta ratio, beta/delta ratio and theta/delta ratio recordings before and during surgery are presented in Table 1 as mean and standard deviations. The p-values for differences are also listed. The variables examined are plotted against time in Figs 1 8. The mean time (SD; range) from beginning of surgery until the first spermatic cord was clamped was 2.25 minutes (0.33; ). There were significant differences in MAP (p ¼ ) and HR (p ¼ 0.011) when values before and during surgery were compared. There were no significant differences between pre-surgical and intraoperative values of the electroencephalographic variables examined. Table 1 Mean and standard deviations for differences between mean values recorded before and during surgery for the variables evaluated. p-values refer to a two-sided Student s t-test for differences from zero Variables Mean differences SD p-values MAP (mmhg) Pulse rate (beats minute )1 ) ) SEF (Hz) ) MED (Hz) Total power (db) ) Alpha/delta ratio ) Beta/delta ratio ) Theta/delta ratio Ó Association of Veterinary Anaesthetists, 2005, 32,

4 Figure 2 Changes in mean pulse rate for each horse (dotted lines) and the mean of all horses (solid line) with time. Onset of surgery (*). The grey bar marks the period when the Figure 3 Changes in mean SEF 95% for each horse (dotted lines) and the mean of all horses (solid line) with time. Onset of surgery (*). The grey bar marks the period when the Figure 4 Changes in mean MED for each horse (dotted lines) and the mean of all horses (solid line) with time. Onset of surgery (*) The grey bar marks the period when the Discussion Nociceptive impulses reaching the CNS can induce sympathetic nervous reactions resulting in increased pulse frequency and blood pressure. These cardiovascular variables have traditionally been used as an indicator of nociception in many species. In this study an increase in MAP occurred during castration, which coincided with a fall in pulse frequency. This fall was associated with manipulation of the Ó Association of Veterinary Anaesthetists, 2005, 32,

5 Figure 5 Changes in mean total power for each horse (dotted lines) and the mean of all horses (solid line) with time. Onset of surgery (*). The grey bar marks the period when the Figure 6 Changes in mean alpha/ delta ratio for each horse (dotted lines) and the mean of all horses (solid line) with time. Onset of surgery (*). The grey bar marks the period when the tunica vaginalis was clamped. Figure 7 Changes in mean beta/delta ratio each horse (dotted lines) and the mean of all horses (solid line) with time. Onset of surgery (*). The grey bar marks the period when the testis and supports that literature which identifies genital surgery as a cause of increased vagal tone and intraoperative bradycardia (Smith 1992). Furthermore, it is our conviction that manipulation of the testes in lightly anaesthetized horses induces bradycardia. This may be related specifically to testicular surgery or represent a general effect of surgical stimulation under inadequate analgesia. In the current study, glycopyrrolate was given to prevent problems caused by increased vagal tone during 132 Ó Association of Veterinary Anaesthetists, 2005, 32,

6 Figure 8 Changes in mean theta/ delta ratio each horse (dotted lines) and the mean of all horses (solid line) with time. Onset of surgery (*). The grey bar marks the period when the Figure 9 The EEG electrode configuration used was two recording electrodes medial to the temporal lines, a reference electrode in the midline rostral to the eyes and a ground electrode in the left atlantooccipital region. testicular manipulation. However, stimulation during castration seems to have partially overridden the chronotropic effect of glycopyrrolate. Despite decreasing pulse rate, MAP increased, probably caused by sympathetic nervous stimulation increasing systemic vascular resistance. Nociceptive stimuli reaching the CNS are likely to alter the functional state of the cerebral cortex, increasing the frequency and decreasing the amplitude of the EEG, that is causing arousal (Bimar & Bellville 1977). The opposite effect a decrease in frequency and an increase in amplitude has also been observed in response to nociception (Bimar & Bellville 1977). Other studies have found changes in alpha/delta ratio, beta/delta ratio, theta/delta ratio, MED, SEF 80% and total power in response to nociceptive stimulation in horses anaesthetized with halothane or isoflurane (Ekström et al. 1993; Otto et al. 1994, 1996; Murrell et al. 1999, 2000a,b). The Fourier transformations chosen in the current study were based on these previous results. We used SEF 95% rather than SEF 80% as the EEG monitor used in the current study could not give SEF 80%. No significant change was found in any of the EEG variables evaluated. There is conflicting information in the literature concerning nociception-induced EEG arousal in horses anaesthetized with volatile agents. During halothane anaesthesia, some studies found changes in beta/delta ratio, MED and total power (Ekström et al. 1993; Short & Ekström 1993; Murrell et al. 1999, 2000a,b). However, others failed to record any EEG responses (Miller et al. 1995). During isoflurane anaesthesia, alpha/delta ratio, beta/delta Ó Association of Veterinary Anaesthetists, 2005, 32,

7 ratio, theta/delta ratio, MED and SEF 80% changed in response to nociceptive stimulation (Otto et al. 1994, 1996) while no EEG changes were recorded in other studies (Ekström et al. 1993; Short & Ekström 1993; Otto et al. 1998). There is electroencephalographic evidence that the CNS is more responsive during halothane than isoflurane anaesthesia (Ekström et al. 1993). In one study, the dose response curves for CNS depression at equal MAC multiples of halothane and isoflurane were different (Johnson & Taylor 1998). The results of the current study support the view that EEG responsiveness to noxious stimulation is suppressed during isoflurane anaesthesia in horses. A significant increase in MAP occurred in response to surgery. In several studies of halothane-anaesthetized horses, an increase in MAP occurred in response to nociception (Wagner et al. 1992, 1995, 1996; Taylor et al. 1998). In one EEG study in halothane-anaesthetized horses, no significant difference was found in the EEG variables evaluated (Miller et al. 1995) but MAP was higher during, and lower after surgery when compared with a control group. This indicates that MAP is a more sensitive indicator of nociception than the EEG in halothane-anaesthetized horses. However, it is difficult to draw firm conclusions from this study because dobutamine was administered when needed and the halothane concentration delivered during surgery was not equivalent in the two groups. In halothane-anaesthetized horses, constant infusions of dobutamine exert an increasing pressor effect with time (Young et al. 1998). This could have affected the results in the current study by contributing to the rise in blood pressure during surgery. The dobutamine infusion rate was constant for 5 minutes before recording baseline values for blood pressure and as Fig. 1 shows, baseline MAP values were stable compared with the increase observed once surgery began. On this basis, we believe that surgery itself, and not the increasing pressor effect of dobutamine with time, was the cause of the rising MAP observed at the time of surgery in the current study. Glycopyrrolate, 2.5 lg kg )1 IV, increases heart rate in xylazine-sedated horses for 40 minutes after injection (Singh et al. 1997). It has been stated that in horses during volatile agent anaesthesia, the effect of glycopyrrolate at 5 10 lg kg )1 IV wanes after 1 2 hours, when a second dose may be required (Taylor & Clarke 1999). In the current study, a glycopyrrolate dose of 5 lg kg )1 IV was administered to all horses, and we cannot exclude the possibility that its effects may have waned in some of the horses during castration. This may explain the decreased pulse rate during castration, and indicate that our results with regard to pulse rate as a nociceptive indicator should be interpreted with caution. However, while vagolysis may influence MAP, we believe that the observed rise in MAP was a valid response to nociception. The results of the current study agree with others in this regard (Wagner et al. 1992, 1995, 1996; Taylor et al. 1998), and is probably caused by an increase in systemic vascular resistance resulting from an increase in sympathetic nervous vasomotor tone. Electroencephalographic responses to surgical stimulation are more likely to be encountered in halothane, compared with isoflurane-anaesthetized horses (Ekström et al. 1993). In this study, the EEG response was not accompanied by an increase in systemic arterial pressure or heart rate. This may be explained by the fact that the halothane concentration was not held constant; indeed, the concentration delivered was partially based on cardiovascular variables. Furthermore, dobutamine was infused as needed. In another study, EEG Fourier transformation variables were evaluated in isoflurane-anaesthetized horses undergoing soft tissue surgery, orthopaedic surgery or anaesthesia alone (Otto et al. 1996). In this study, a significant difference was found in EEG variables between those horses undergoing surgery and the control group, but no significant difference was observed in haemodynamic variables, indicating that EEG was a more sensitive indicator of nociception. However, dobutamine was also infused as needed in this study. The potential of EEG and cardiovascular variables to indicate nociception was evaluated in isofluraneanaesthetized pigs (Haga et al. 2001). Nociceptive stimulation caused an increase in MAP, but significant changes in the EEG were not encountered. In another study using isoflurane anaesthetized-goats with isolated brain circulation, it was determined that the brain had a limited role in suppressing cardiovascular responses to noxious stimuli during isoflurane anaesthesia (Antognini & Berg 1995). This may explain the increase in MAP after nociceptive stimulation without observable changes in the Fourier-transformed EEG signal. A rise in MAP in response to nociception indicates that nociceptive stimuli have reached the brain stem. An EEG response indicates a functional change in the cerebral cortex. While knowledge of 134 Ó Association of Veterinary Anaesthetists, 2005, 32,

8 the CNS level to which nociceptive stimuli have reached may be of some value as an indicator of general nociception in isoflurane-anaesthetized horses, MAP seems to be more sensitive than any of the EEG Fourier transformation variables evaluated in the current study. Acknowledgements This study was financially supported by the Norwegian Research Council. We acknowledge Henning Moerch for technical assistance and Professor Aage Tverdal for statistical advice. References Antognini JF, Berg K (1995) Cardiovascular responses to noxious stimuli during isoflurane anesthesia are minimally affected by anesthetic action in the brain. Anesth Analg 81, Bimar J, Bellville JW (1977) Arousal reactions during anesthesia in man. Anesthesiology 47, Ekström PM, Short CE, Geimer TR (1993) Electroencephalography of detomidine-ketamine-halothane and detomidine-ketamine-isoflurane anesthetized horses during orthopedic surgery. A comparison. Vet Surg 22, Haga HA, Tevik A, Moerch H (2001) Electroencephalographic and cardiovascular indicators of nociception during isoflurane anaesthesia in pigs. Vet Anaesth Analg 28, Johnson CB, Taylor PM (1998) Comparison of the effects of halothane, isoflurane and methoxyflurane on the electroencephalogram of the horse. Br J Anaesth 81, Miller SM, Short CE, Ekström PM (1995) Quantitative electroencephalographic evaluation to determine the quality of analgesia during anesthesia of horses for arthroscopic surgery. Am J Vet Res 56, Murrell JC, Johnson CB, Waterman-Pearson AE (1999) Spontaneous EEG changes in the equine surgical patient. J Vet Anaesth 26, 42 (Abstract). Murrell JC, Johnson CB, Waterman-Pearson AE et al. (2000a) Changes in the equine EEG during surgery: The effect of an intravenous infusion of thiopentone. Veterinary anaesthesia and analgesia 27, 113 (Abstract). Murrell JC, White KL, Johnson CB et al. (2000b) Spontaneous EEG Changes in the Equine Surgical Patient: The Effect of an Intravenous Infusion of Lidocaine. Proceedings of the 7th World Congress of Veterinary Anaesthesia, pp Otto K, Short CE, Voigt S (1993) Identification of noxious stimuli and analgesic drug effects using computer assisted EEG analysis during anaesthesia in horses a preliminary report. J Vet Anaesth 20, (Abstract). Otto K, Voight S, Deegen E et al. (1994) Methoden der Narkosetiefenbestimmung beim Pferd-II. Elektroenzephalographie, Pferdeheilkd, pp Otto KA, Voight S, Piepenbrock S et al. (1996) Differences in quantitated electroencephalographic variables during surgical stimulation of horses anesthetized with isoflurane. Vet Surg 25, Otto KA, Voight S, Piepenbrock S et al. (1998) Effects of low dose ketamine on haemodynamic and electroencephalographic variables during surgery in isoflurane-anaesthetized horses. J Vet Anaesth 25, Short CE, Ekström PM (1993) Compressed Spectral Analysis of the EEG as an Indicator of Anesthetic Quality During Anesthesia for Orthopedic Surgery in the Horse. Proceedings of the Thirty-Eight Annual Convention of the American Association of Equine Practitioners, Orlando, Florida, pp Singh S, Young SS, McDonell WN et al. (1997) Modification of cardiopulmonary and intestinal motility effects of xylazine and glycopyrrolate in horses. Can J Vet Res 61, Smith M (1992) Komplikationer i forbindelse med anæstesi. In: Komplikationer i forbindelse med anæstesi Klinisk anæstesi og analgesi af vore husdyr (2nd edn). DSR forlag, København, pp Taylor PM, Clarke KW (1999) Anaesthetic problems. In: Handbook of Equine Anaesthesia (1st edn). W.B. Saunders, London, pp Taylor PM, Kirby JJ, Shrimpton DJ et al. (1998) Cardiovascular effects of surgical castration during anaesthesia maintained with halothane or infusion of detomidine, ketamine and guaifenesin in ponies. Equine Vet J 30, Thurmon JC, Tranquilli WJ, Benson GJ (eds) (1996) Perioperative pain and distress. In: Veterinary Anaesthesia (3rd edn). Williams & Wilkins, Baltimore, MD, pp Wagner AE, Dunlop CI, Heath RB et al. (1992) Hemodynamic function during neurectomy in halothaneanesthetized horses with or without constant dose detomidine infusion. Vet Surg 21, Wagner AE, Dunlop CI, Wertz EM et al. (1995) Hemodynamic responses of horses to anesthesia and surgery, before and after administration of a low dose of endotoxin. Vet Surg 24, Wagner AE, Dunlop CI, Wertz EM et al. (1996) Evaluation of five common induction protocols by comparison of hemodynamic responses to surgical manipulation in halothane-anesthetized horses. J Am Vet Med Assoc 208, Young LE, Blissitt KJ, Clutton RE et al. (1998) Temporal effects of an infusion of dobutamine hydrochloride in horses anesthetized with halothane. Am J Vet Res 59, Received 10 April 2001; accepted 1 March Ó Association of Veterinary Anaesthetists, 2005, 32,