Use of midlatency auditory-evoked potentials as indicator of unconsciousness in the dog: characterisation of the effects of acepromazine–thiopentone, medetomidine–thiopentone and medetomidine–butorphanol–midazolam combinations

Research in Veterinary Science 1999, 67, 35–39 Article No. rvsc.1998.0273, available online at http://idealibrary.com on

Use of midlatency auditory-evoked potentials as indicator of unconsciousness in the dog: characterisation of the effects of acepromazine–thiopentone, medetomidine–thiopentone and medetomidine–butorphanol–midazolam combinations B. PYPENDOP, L. PONCELET, J. VERSTEGEN Departments of Small Animal Reproduction and Small Animal Surgery, School of Veterinary Medicine, University of Liege, Bd de Colonster, 20, B44, Sart Tilman B4000 Liege, Belgium SUMMARY Middle latency auditory-evoked potentials were measured in sedated and anaestetised dogs to determine their possible usefulness in monitoring of unconsciousness during anaesthesia and to compare the effects of anaesthetic protocols. There were three groups of five dogs: group I received acepromazine; groups 2 and 3 received medetomidine; 30 minutes later, groups 1 and 2 received thiopentone and group 3 received midazolam and butorphanol. Groups 2 and 3 received atipamezole 60 minutes after medetomidine was administered. Auditory-evoked potentials were recorded at time 15, 40 and 75 minutes. Thiopentone administration resulted in a profound modification of the pattern of response, and several peaks were no longer identified. In group 3, the administration of midazolam-butorphanol tended to increase the latency of the different peaks, but lesser than thiopentone did. Middle latency-evoked potentials appeared to be potentially useful in the monitoring of unconsciousness in the dog.

THE goals of general anaesthesia are unconsciousness, analgesia, autonomic reflex stability and muscle relaxation. The different drugs commonly used participate at varying degrees to one or several of these components. Adequate depth of anaesthesia for a particular surgical procedure implies induction of a sufficient level of each of these components, without compromising vital organ functions. However, despite the extensive use of anaesthesia, no definite clinical sign has proven total reliability in assessing depth of anaesthesia or of its components. Equipments or techniques to assess the depth of each component of anaesthesia should be available, because specific drugs can act on one component without affecting the others; for example it is possible to obtain unconsciousness without adequate analgesia. In human medicine, awareness during general anaesthesia is a common complication, despite the increased monitoring of physiologic variables (Heier and Steen 1996a,b). Although awareness would be very difficult to assess in animals, monitoring the depth of central nervous system depression could undoubtly provide valuable information, especially in critical patients. Different techniques have been tested in attempts to find an adequate method of monitoring unconsciousness. Among these, electroencephalogram (EEG) derivations seem, on a physiological point of view, a logical approach. But, due to the difficulty of interpretation of the EEG, computerised spectral indices have been proposed and are probably up to date the only clinically usefull variables. However, they still need to be validated during routine anaesthesia in humans as well as in animals (Thornton 1991).

0034-5288/99/040035 + 5 $18.00/0

Other EEG-derived variables are evoked potentials. Somatosensory, auditory- and visual-evoked potentials have been studied in humans during anaesthesia and surgery to monitor neurological function. A number of studies showed that the middle latency response of the auditory evoked potentials corresponds to the primary non-cognitive cortical processing of auditory impulses. Moreover, it has been reported that middle latency auditory evoked potentials are influenced in a dose dependent fashion by different anaesthetic and hypnotic drugs (i.e. their latency increased while their amplitude decreased). Consequently, in human medicine, it is considered that monitoring of middle latency auditory evoked potentials (MLAEP) could be clinically useful to assess the level of unconsciousness (Thornton et al, 1989; Thornton 1991, Schwender et al 1994, de Beer et al 1996). In veterinary literature, few studies characterising MLAEP are available (Sims and Moore 1984, Johnson et al 1995; Johnson and Taylor 1997). Sims and Moore (1984) reported a response comparable to that in humans in conscious curarised dogs. They also showed an inhibition of the response following the administration of 20 mg kg–1 of thiamylal, intravenously (IV). The aims of the study were to characterise the influence of anaesthesia on midlatency auditory evoked potentials in the dog, using acepromazine–thiopentone, medetomidine– thiopentone and medetomidine–butorphanol–midazolam combinations. We hypothesised that anaesthesia increases the latency and decreases the amplitude of MLAEP deflections, and that the three combinations tested influence the MLAEP in the same fashion.

© 1999 Harcourt Publishers Ltd

B. Pypendop, L. Poncelet, J. Verstegen

36

MATERIALS AND METHODS All experiments were carried out according to the Belgian regulations for animal research and experimentation and to the NIH guide for the care and use of laboratory animals. Dogs Fifteen healthy young adult Beagles, (age range 8 months to 1 year), weighing 11 to 15 kg (mean ± SD = 13.5 ± 1.7), were used. Dogs were housed individually and allowed to exercise outdoors 2 hours a day. They were fed a commercial diet (Hill’s Science Plan Canine Maintenance) once daily with ad libitum access to water. Food and water were withheld for 12 hours prior to the experiments. Before the study, an examination of the ear canal was performed, to exclude any gross pathology or abnormality. Treatment Dogs were randomly allotted to one of three groups (n=5). In group 1, acepromazine, 0.05 mg kg–1, was administered intravenously at T0 min. In groups 2 and 3, medetomidine, 1 mg m2–1 of body surface area (BSA), was administered intramuscularly (IM) at T0 min. Thiopentone, at a dose allowing intubation and supressing the oculopalpebral reflex, was administered IV in groups 1 and 2 at T30 min. In group 3, butorphanol (0.1 mg kg–1) and midazolam (1 mg kg–1) were administered IV at T30 min. Finally, atipamezole, 2.5 mg m2–1 BSA, was administered IM at T60 min. in groups 2 and 3. Auditory evoked potentials recordings Measurements were carried out in a quiet, non soundproofed room, in sedated or anaesthetised animals as preliminary studies indicated that recording MLAEP in awake dogs was not practical. After the T0 administration, dogs were placed in sternal recumbency. Subdermal electrodes were placed at the vertex, just rostral to the base of the left ear and connected to the non-inverting input of the pre-amplifler, and in the dorsal cervical region, about 20 cm caudal to the occipital protuberance, connected to the inverting input. The interelectrode impedance was less than 5 kΩ. Acoustic rarefaction clicks, at 90 dB hearing level of 100 µs duration were delivered monaurally to the left ear via an insert headphone at a stimulus rate of 6.1 sec–1. The signal was amplified. A bandpass from 5 Hz to 3 kHz was used, with automatic artifact rejection. The first 120 milliseconds (msec) after each click was recorded, and 1000 responses were averaged and stored (Pathfinder II, Nicolet). Each recording was repeated twice. Amplitudes and latencies of identified positive and negative peaks were measured off-line. Evoked potentials were recorded at T15 (sedated animal), T40 (anaesthetised animal) and T75 min (recovery). Statistical analysis Data are presented as mean values ± SD. Each peak was analysed separately. Repeated measures ANOVA for time effect within each group was used. For between group comparisons at each time point, one-way ANOVA was used. When

a significant effect was detected, the groups were compared two-by-two using a Bonferonni test. Moreover, in order to include the fact that some peaks disappeared during anaesthesia, and that their latency can be considered to be infinite, (latency)–1 was calculated for each peak (resulting in 0 value when a peak was absent), and the statistical analysis was repeated on the data generated. Differences were considered significant when P<0.05 (Lentner 1982).

RESULTS In groups 1 and 2, dogs were administered 15.5±2.09 and 4.4±1.34 mg kg–1 thiopentone IV respectively, to achieve the desired stage of anaesthesia. MLAEP recordings in sedated, conscious dogs consisted in a succession of positive (P) and negative (N) peaks identified as N0, P0, Na, Pa, Nb, and Pb, by analogy with the peaks identified in the human (Fig 1). N0 and Pb were only identified in some recordings and were therefore excluded from statistical analysis, while the other peaks were identified in each of them. The amplitude of the peaks appeared widely variable between individuals, while latencies were more constant (Tables 1 and 2). Neither amplitude nor latency of any peak was significantly different among groups at T15 minutes (sedated dogs) (Tables 1 and 2). Thiopentone administration resulted in a profound modification of the pattern of response. Na, Pa and Nb were no longer identified following thiopentone administration in the groups 1 and 2 (Tables 1 and 2, Fig 2). P0 was significantly delayed in the medetomidine-thiopentone group (group 2), and was no longer identified in the acepromazinethiopentone group (group 1). The analysis performed on (latencies)–1 showed the same differences. In group 3, the administration of midazolam-butorphanol increased the latency of the different peaks. However, due to wide variation, this was not significant. There was also no significant difference with group 1 and 2 (thiopentone administration). However, the analysis performed on (latencies)–1 allowed detection of a significant difference in Na and Nb with group 1 and 2. Effect on amplitude was not detected (Tables 1 and 2, Fig 3). Measurements realised during recovery were similar to those obtained before the induction of anaesthesia, in sedated dogs (Tables 1 and 2).

DISCUSSION AND CONCLUSIONS Because immobility is warranted when recording evoked potentials, baseline measurement were not recorded in nonsedated dogs but in conscious sedated animals. As sedative drugs depress the central nervous system, an effect on MLAEP cannot be excluded. However, the only alternative to sedation is the administration of neuromuscular blocking agents in conscious animals, which is ethically not acceptable. It has been reported that in normal conscious dogs, the MLAEP pattern was similar to that observed in human (Sims and Moore 1984). The first 120 ms following each stimulation were studied because most human studies showed that

MLAEP and unconsciousness in the dog Table 1:

37

Latencies of the different peaks in the three treatment groups. Mean (SD)

Group

Time

P0

(ms)

Na

(ms)

Pa

(ms)

Nb

(ms)

15 40 75 15

10.8(2.1) absent* 12.0(0.2) 10.0(0.7)

16.2(4.1) absent* 17.4(2.5) 14.3(2.0)

23.5(4.5) absent* 23.8(5.3) 21.1(3.6)

41.9(10.0) absent* 40.8(12.1) 57.1(20.4)

2 MDT-THIO

40 75 15

15.1(2.7)* 10.8(1.1) 13.3(3.3)

absent* 16.6(2.0) 17.9(5.2)

absent* 26.3(3.6) 28.2(13.3)

absent* 47.5(1.7) 52.6(2.8)

3 MDT-BUT-MDZ

40 75

51.7(83.0) 13.8(2.6)

88.4(101.9)$ 29.9(22.7)

98.3(94.2) 31.2(13.4)

100.4(63.0)$ 62.1(15.9)

1 ACP-THIO

ACP-THIO: acepromazine-thiopentone group (group 1); MDT-THIO: medetomidine-thiopentone group (group 2); MDT-BUT-MDZ: medetomidine-butorphanol-midazolam group (group 3). *significantly different from the respective measurement at time 15 minutes (P<0.05). $significantly different from acepromazine-thiopentone and medetomidine thiopentone at time 40 minutes (analysis on latencies–1)(P<0.05)

Table 2:

Amplitudes of the different peaks in the three treatment groups. Mean (SD)

Group

Time

P0

(µV)

Na

(µV)

Pa

(µV)

Nb

(µV)

15 40 75 15

1.3(0.6) absent* 2.0(1.4) 0.8(0.5)

–0.9(0.6) absent* 1.0(3.0) –0.1(0.5)

2.1(0.7) absent* 3.0(1.7) 1.4(0.2)

–3.7(1.9) absent* –2.9(0.8) –3.9(1.8)

2 MDT-THIO

40 75 15

1.5(1.1) 0.7(0.6) 0.8(0.8)

absent* –1.2(0.6) 0.5(1.8)

absent* 1.8(0.9) 2.3(0.9)

absent* –2.8(1.1) –2.9(1.1)

3 MDT-BUT-MDZ

40 75

0.7(1.0) 0.7(1.0)

0.5(0.7) 0.5(0.7)

1.9(2.1) 1.7(1.1)

–2.0(1.6) –2.5(1.0)

1 ACP-THIO

See above for key.

Pb

P0

Pa

Na

Nb

FIG 1: Example of MLAEP in a dog after acepromazine administration (two repetitions without time interval). Vertical division = 2.5 µV, horizontal division = 5 ms. Only the 50 first ms are shown. This recording shows that MLAEP consisted in a succession of positive and negative deflections labelled P0, Na, Pa, Nb and Pb, by analogy with the peaks identified in human being.

FIG 2: Example of MLAEP in a dog after acepromazine and thiopentone administration (four repetitions at a 5 minutes interval). Vertical division = 2.5 µV, horizontal division = 5 ms. Only the 50 first ms are shown. Compared to the recording after acepromazine administration, MLAEP are inhibited and can not be identified.

MLAEP are situated in the 10 to 120 ms window (Thornton et al 1989, Thornton 1991, Schwender et al 1994, de Beer et al 1996), and because a previous study in the dog described MLAEP in the 6 to 60 ms range (Sims and Moore 1984). Rarefaction clicks were used because most human studies using MLAEP as monitoring of anaesthesia depth use rarefaction clicks, and preliminary studies showed no difference between condensation and rarefaction clicks, or an alternance of both types. However, Sims and Moore (1984) alternated rarefaction and condensation clicks. Finally, these clicks were applied monaurally because preliminary studies did not show a better response when using binaural stimulation. In this study, peaks similar to those reported in humans could be identified. However, the first negative (N0) and the

latest positive (Pb) deflections were observed in only some of the baseline recordings. This could be due to an effect of sedation, because of their inconstant presence in our baseline recordings, they were excluded from statistical analysis. The two sedative drugs tested (acepromazine and medetomidine) have different modes of action. Phenothiazines produce sedation by depressing the brain stem and connections to the cerebral cortex, probably partially by antagonising dopamine excitatory receptors (Gleed 1987). Medetomidine induces sedation mainly by agonism on the alpha-2 adrenoceptors located in the locus coeruleus (Stenberg 1989). However, the MLAEP recorded after administration of either drug were not significantly different. If sedation has an effect on these evoked potentials, it seems similar with both drugs.

38

B. Pypendop, L. Poncelet, J. Verstegen

FIG 3: Example of MLAEP in a dog after medetomidine-butorphanol-midazolam administration (four repetitions without time interval). Vertical division = 2.5 µV, horizontal division = 5 ms. Only the 50 first ms are shown. This recording shows that this combination partially inhibited MLAEP

Thiopentone was used to observe the effects of unconsciousness on MLAEP. This drug was chosen because ultrashort barbiturates are widely used to induce anaesthesia, produce reliable hypnosis, and have a duration of action compatible with the study design (Thurmon et al 1996 a). The dose of thiopentone necessary to obtain sufficient anaesthesia was significantly different in the two groups receiving this drug. This is due to the higher anaesthetic potentiating effect of medetomidine, compared to acepromazine (Thurmon et al 1996 b). In the two groups receiving thiopentone, latencies of different peaks significantly increased, and most of them could even not be identified anymore. This modification of MLAEP is compatible with an effect of hypnosis and has been reported in the dog following administration of thiamylal, another ultrashort barbiturate (Sims and Moore 1984). A total suppression of P0 was observed in group 1, receiving acepromazine and thiopentone, but not in group 2, receiving medetomidine and thiopentone. This was probably related to the higher dose of thiopentone (15.5 vs 4.4 mg kg–1) required in the first group, inducing a more profound depression of the central nervous system. In previous studies, we reported the effects of a medetomidine-butorphanol-midazolam combination in the dog (Verstegen and Petcho, 1993; Pypendop et al., 1996). Clinically, anaesthesia was obtained (disappearance of oculo-palpebral reflexes, intubation possible, unresponsiveness), although none of the drugs used is known to produce unconsciousness in the dog, even at high dosages (Thurmon et al 1996b). The medetomidine-midazolam-butorphanol combination increased all latencies, but to a lesser extent than did thiopentone, which abolished totally most of the peaks. This leads to a significant difference when the disappearance of the peaks is analysed. The different peaks had an intermediate latency situated in the range between those observed during the sedation stage and those measured in animals receiving thiopentone. However, this was not significantly different from the sedation stage. If we state that MLAEP monitor the level of unconsciousness (i.e. hypnosis), the medetomidine–midazolam–butorphanol combination cannot be considered as inducing an hypnosis comparable to thiopentone. In human medicine, it has been reported that profound analgesia (induced by potent opioids for example) depressed only partially MLAEP (Schwender et al 1993, 1995). The

absence of reaction to surgical stimulation previously reported with the study combination (Verstegen and Petcho, 1993) and an incomplete inhibition of MLAEP compared to true hypnosis could thus represent a similar situation. The present study showed the possibility to obtain MLAEP recordings in sedated and anaesthetised dogs. However, because of the variability in the measurements, especially the amplitudes, further studies are needed to confirm these preliminary results and to establish reference values using different sedative and/or anaesthetic drugs. Nevertheless, the technique appeared interesting for the monitoring of the changes in cerebral activity induced by hypnosis. In conclusion, the results of this study are compatible with the statement that MLAEP could be useful in the monitoring of unconsciousness in the dog. The medetomidinemidazolam-butorphanol combination is unable to depress MLAEP to the same extent than thiopentone, maybe because this combination induces profound neuroleptanalgesia rather than true hypnosis.

AKNOWLEDGEMENTS The authors thank Dr Annick Hamaide, Department of Small Animal Surgery, for her comments and editorial assistance. They thank Hill’s Benelux for the supply of dog food, Pfizer Animal Health Belgium for the supply of medetomidine and atipamezole and Willows Francis Veterinary, for the supply of butorphanol.

REFERENCES DE BEER, N.A.M., VAN HOOFF, J.C., BRUNIA, C.H.M., CLUITMANS, P.J.M, KORSTEN, H.H.M. AND BENEKEN, J.E.W. (1996) Midlatency auditory evoked potentials as indicators of perceptual processing during general anaesthesia. British Journal of Anaesthesia 77, 617–624. GLEED R.D. (1987) Tranquillizers and sedatives. In: Short C.E., ed. Principles and practice of veterinary anesthesia. Baltimore, USA: Williams & Wilkins, 16–27. HELER, T. AND STEEN, P.A. (1996b) Awareness in anaesthesia: incidence, consequences and prevention. Acta Anaesthesiologica Scandinavica 40, 1073–1086. HEIER, T. AND STEEN, P.A. (1996a) Assessment of anaesthesia depth. Acta Anaesthesiologica Scandinavica 40, 1087–1100 JOHNSON, C.B. AND TAYLOR, P.M. (1997) Effects of alfentanil on the equine electroencephalogram during anaesthesia with halothane in oxygen. Research in Veterinary Science 62, 159–163. JOHNSON, C.B., WASHBOURNE, J.R. AND TAYLOR, P.M. (1995) Origin of the equine middle latency auditory evoked potential. Journal of Veterinary Anaesthesia 22, 15–18. LENTNER, G. (1982) Statistical methods. In: Lentner, G., ed. Geigy scientific tables. Introduction to statistics, statistical tables and mathematical formulae. 8th ed. Basel, Switzerland: Ciba Geigy, 199–203. PYPENDOP, B., SERTEYN, D. AND VERSTEGEN, J. (1996) Hemodynamic effects of medetomidine-midazolam-buprenorphine combinations and reversibility by atipamezole in dogs. American Journal of Veterinary Research 57, 724–730. SCHWENDER, D., KLASING, S., MADLER, C., PÖPPEL, E. AND PETER, K. (1994) Midlatency auditory evoked potentials and purposeful movements after thiopentone bolus injection. Anaesthesia 49, 99–104. SCHWENDER, D., RIMKUS, T., HAESSLER, R., KLASING, S., PÖPPEL, E. AND PETER, K. (1993) Effects of increasing doses of alfentanil, fentanyl and morphine on mid-latency auditory evoked potentials. British Journal of Anaesthesia 71, 622–628. SCHWENDER, D., WENINGER, E., DAUNDERER, M., KLASING, S., PÖPPEL, E. AND PETER, K. (1995) Anesthesia with Increasing Doses of Sufentanil and Midlatency Auditory Evoked Potentials in Humans. Anesthesia and Analgesia 80, 499–505. SIMS, M.H. AND MOORE, R.E. (1984) Auditory-evoked response in the clinically normal dog: middle latency components. American Journal of Veterinary Research 45, 2028–2033. STENBERG, D. (1989) Physiological role of alpha-2 adrenoceptors in the regulation of vigilance and pain: effect of medetomidine. Acta Veterinaria Scandinavica 85, 21–28.

MLAEP and unconsciousness in the dog THORNTON, C., BARROWCLIFFE, M.P., KONIECZKO, K.M., VENTHAM, P., DORÉ, C.J., NEWTON, D.E.F. AND JONES, J.G. (1989) The auditory evoked response as an indicator of awareness. British Journal of Anaesthesia 63, 113–115. THORNTON, C. (1991) Evoked potentials in anaesthesia. European Journal of Anaesthesiology 8, 89–107. THURMON, J.C., TRANQUILLI, W.J. AND BENSON, G.J. (1996a) Injectable Anesthetics. In: Thurmon JC, Tranquilli WJ, Benson GJ, eds. Lumb and Jones’ Veterinary Anesthesia. 3rd ed. Baltimore, USA: Williams & Wilkins, 210–240.

39

THURMON, J.C., TRANQUILLI, W.J. AND BENSON, G.J. (1996b) Preanesthetics and anesthetic adjuncts. In: Thurmon JC, Tranquilli WJ, Benson GJ, eds. Lumb and Jones’ Veterinary Anesthesia. 3rd ed. Baltimore, USA: Williams & Wilkins, 183–209. VERSTEGEN, J. AND PETCHO, A. (1993) Medetomidine-butorphanol-midazolam for anaesthesia in dogs and its reversal by atipamezole. Veterinary Record 132, 353–357. Accepted January 28, 1999