Efficacy of intra-muscular analgesics for acute pain in sheep C GRANT*, RN UPTON+ and TR KUCHELS

SUMMARY: The analgesic action of intramuscularly injected buprenorphine, metha- done, flunixin meglumine and xylazine was examined in sheep, using algesimetry based on a leg withdrawal response to an electrical stimulus. No analgesic response was detected for buprenorphine, methadone or flunixin meglumine. Only the a2-adrenocep- tor agonist, xylazine, produced an analgesic response. The current required to elicit a response increased by 170% (4.5 f 0.43 mA to 12.23 f 1.14 mA; mean f SE) after a dose of 0.05 mgkg xylazine; by 180% (4.73 f 0.3 mA to 13.28 f 2.35 mA) after 0.1 mgkg; and by 510% (4.52 f 0.29 mA to 27.63 f 3.89 mA) after 0.2 mgkg. Intramuscular xylazine appears to be an effective analgesic agent for acute pain in the sheep and further investigation into ideal administration regimens and dosage may provide more detailed information on relationships between dose, analgesic and sedative effects. The findings also suggest that some common analgesic agents, and opioids in particular, may be ineffective for the management of acute pain in sheep and that any analgesic should be administered only on the basis of its proven efficacy in that species. Aust Vet J 73: 129 - 132

Introduction Pain in animals is a topical and emotive subject. Whereas most

people would acknowledge the need for alleviation of pain in a suffering animal, the routine use of analgesics have been generally limited to animals of significant economic or social importance and to research animals, where a legislative obligation exists to exhort the application of modern principles of analgesia. Likewise, studies examining the clinical use of these analgesics is mainly in the dog, cat or horse. The impetus for this study came from a desire to examine methods of effective control of post-operative pain for chronically catheterised sheep (Huang et a1 1992). The ideal analgesic in this situation would cause a minimum of sedation and recumbency, be fast acting and easily administered. The aim of this study was to examine the efficacy of a number of commonly used intramuscular analgesics in the form they are presented to the practicing veterinar- ian for use in sheep. Although there are studies involving the use of analgesics in sheep, extrapolation of the results of most of them beyond the conditions of the experiment is of dubious value.

We encountered some potential problems when choosing an anal- gesic to use in the sheep: Analgesics are often administered on the basis of data derived from studies in humans or other species, even though considerable interspecies differences in drug disposition are well known (Davis 1983). In a number of studies the sheep has served as a pharmacokinetic/dynamic model for investigation into the action of certain analgesics, but their application is directed towards under- standing basic mechanisms (Gray et a1 1994). The general route of administration of these compounds in veterinary practice is intra- muscular (IM) because of its ease, but most previous studies have used intravenous or even intrathecal administration (Nolan el a1 1987; Waterman et a1 1988). In past studies many of the drugs were given at doses to produce maximum effect, where analgesia could not be distinguished clearly from anaesthesia or profound sedation (Kamerling et a1 1991).

This study was designed to compare in the sheep the analgesic action of a number of intramuscularly injected compounds, using a leg with- drawal response to an electrical stimulus as a measure of analgesia.

* Department of Anaesthesia and Intensive Care, University of Adelaide, Adelaide, South Australia 5005

t Department of Anaesthesia and Intensive Care, Royal Adelaide Hospital, Adelaide, South Australia, 5000

$ Institute of Medical and Veterinary Science, Frome Road, Adelaide, South Australia 5000

Materials and Methods The drugs investigated were the opioids, buprenorphine* (0.005

mgkg) and methadoneq (0.6 mgkg), the a2-adrenoceptor agonist, xylazine# (0.05, 0.1 and 0.2 mgkg), and flunixin meglumine" (2 m a g ) , an anti-prostaglandin agent with an analgesic component. Saline placebo (1 mL, 0.9%) was included as a control group. Acetylpromazinett was also included as a sedative non-analgesic control (0.15 mgkg) to examine if there were any changes in the response to electrical stimulus, due to sedation alone. Each drug- dosage group consisted of 6 sheep.

All drugs were administered by IM injection via 26G needle into the upper thigh area. The maximum injected volume was 4.1 mLand the minimum volume was 0.1 mL. Injection volumes above 1 mL were administered using multiple injection sites.

Algesimetry The algesimetry method used has been reported and validated

previously and is based on a leg withdrawal response to a subcutan- eous electrical stimulus (Ludbrook el a1 1995). It has been shown that repeated measurements are stable for a period of at least 60 min and that analgesic profiles can be determined for sedativdhypnotic and analgesic drugs by measuring the increase above baseline values of the current required to induce a leg withdrawal response .

Adult Merino sheep of both sexes (weight 43.5 to 68.5 kg) were used throughout the study and were conditioned to accept housing in metabolic crates. For each experiment the sheep were placed in a sling within their crate, which prevented them from lying down but otherwise appeared comfortable. Two 26G needles, which acted as electrodes, were then inserted subcutaneously in the lower hind leg. To measure analgesia a modified peripheral nerve stimulatortt was used to deliver pulsed DC current to the needles, increasing stepwise between 0 and a maximum of 50 mA. The endpoint was taken as a deliberate withdrawal of the leg, at which point the current was stopped by an observer and the maximum level of current recorded.

4 Temgesic', Reckitt and Colman, Hull, UK P Methone injection', Parnell Laboratories, Silverwater, NSW U Xylazil-SO', Troy Laboratories, Smithfield, NSW ** Finadyne', Heriot AgVet, Rowville, Vic t t ACP lo", Delta Veterinary Laboratories, Mt. Kuring-gai, NSW $$ Digistim3', Neuro Technologies, Houston, TX, USA

Australian Veterinary Journal Vol. 73, No. 4, April 1996 129

lo! 0 0 10 20 30 40

Time (rnin)

Figure 1. The analgesic response (end-point current to stimulate leg withdrawal) and plasma concentrations of methadone after adminstra- tion of 1 m@g of methadone in one sheep. This graph represents the typical pharmacokinetic profile of methadone after intramuscular administration in sheep.

30 t 25

20

15

10

5

0 . 0 5 mg/kg x y l a z i n e

I

30

0.1 mg/kg xylaz ine

L 20

0.2 rng/kg xy laz ine

5

I I I I I I

0 10 20 30 40 50 60

Time (mi.)

Figure 2. The mean * SE analgesic response of sheep (n = 6) after intramuscular administration of xylazine at doses of 0.05,O.l and 0.2 msn(9.

Study sequences occurred every 30 s, the ramp time for each reading being about 5 s. When the pre-treatment response to the stimulus became stable, generally within 2 to 10 min, baseline recordings were made for 5 min (that is, 10 study sequences), after which the selected dose of drug was injected IM into the contralateral leg. Readings continued to be taken every 30 s for a period of 40 to 60 min.

Surgery Under general anaesthesia some sheep were prepared for venous

blood sampling by placing a 7F catheter in the jugular vein. The catheter was fastened to the strap muscles of the neck using a plastic plate and exteriorised. Sheep were then recovered and the catheters continuously flushed with heparinised saline. All sheep were allowed to recover for at least 5 days before any studies were commenced. For the methadone studies 2 mL blood samples were taken at 2.5,5, 10, 15, 20, 30 and 60 min after the dose, and were placed in heparinised plastic tubes, then frozen and stored at -5°C. The samples were assayed by gas chromatography using a technique described by Gourlay el a1 (1982).

Statistical Analysis A one-way analysis of variance (ANOVA) was performed on the

end-point current measurement series for each drug-dosage regimen to determine whether any values within the group were statistically different. Any data set that exhibited statistical significance was then subjected to a paired 1-test to identify which points reached statistical significance when compared with baseline values. A value of P < 0.05 was taken as statistically significant.

Results There was no change in the end-point current from baseline values

in response to administration of 1 mL of 0.9% saline. There was no significant change from baseline values in the

end-point current required to induce leg withdrawal for the drugs buprenorphine, methadone or flunixin meglumine, indicating these drugs produced no analgesic response to the electrical current stimulus.

All sheep in the methadone group that had blood samples taken (n=3) showed consistent increases in plasma methadone concentra- tions, with peak concentration occurring between 15 and 20 min. Typical data are shown in Figure 1.

Acetylpromazine produced no visible signs of sedation and no change in response to the electrical stimulus, but in some cases, particularly at the higher doses, the sheep became agitated and reacted violently to the electrical stimulus. Reliable readings could not be made under such circumstances.

Xylazine was the only drug examined that showed significant analgesic effects (Figure 2). At a dose of 0.05 mgkg there was an increase in the end-point current value at leg withdrawal, from an average baseline value of 4.5 f 0.43 mA to a maximum level of 12.23 * 1.14 mA, or 170% above baseline, occurring at 21 min. Statistically significant increases above baseline (P c 0.05) were observed for all values from 15 min to 60 min.

A dose of 0.1 mgkg of xylazine produced a faster increase to the maximum end-point current value at leg withdrawal (from an aver- age baseline value of 4.73 * 0.3 mA to a maximum level of 13.28 * 2.35 mA or 180% above baseline occurring at 13 min). Statistically significant increases above baseline (P c 0.05) were observed for all values from 3 min to 60 min.

The dose of 0.2 mgkg produced marked analgesic response in all sheep, with a peak effect of 27.63 f 3.89 mA occurring at 14 min, or 510.3% above a baseline of 4.52 * 0.29 mA. Statistically signifi- cant increases above baseline (P c 0.05) were observed for all values from 2 min to 60 min. In 3 sheep there was a significant degree of sedation as shown by postural changes such as head drooping and reduced alertness.

130 Australian Veierinary Journal Vol. 13 , No. 4, April 1996

Discussion The lack of change in response to the electrical stimulus within the

saline placebo control group indicates that any increases in current required to elicit a response in the drug groups is due to pharma- cological effects rather than experimental method. The lack of change in response to the electrical stimulus also shows there is no significant acclimatisation or sensitisation to the stimulus over time.

Acetylpromazine was not useful as a control agent to correct for sedation effects. The variable response, which included excitation, had not previously been well described in the sheep, but is an acknowledged clinical extrapyramidal effect in horses (Hall and Clarke 1983).

The failure of the opioids, buprenorphine and methadone, to produce any significant analgesic effect in the sheep is not particu- larly surprising. There have been anecdotal reports that opioids have atypical effects in ruminants (Davis 1983; Livingston et a1 1991). What is surprising is the continued use of these agents in sheep. Their widespread and successful use in producing analgesia in many species has caused them to be seen by some clinicians as a universal panacea for all situations. However, there are also a number of other possible explanations for the failure of opioids to produce analgesic response in this experimental series.

Stimulus type -The nature of the stimulus itself could be respon- sible for the lack of analgesic response, as another opioid, morphine. has been shown to be effective in treating slow C fibre pain sensations but less effective for fast A-delta pain (Cooper et a1 1982). The peripheral and acute nature of the electrical stimulus may be unsuit- able for the study of opioids although it has been shown that mor- phine attenuates the nociceptive response to electrocutaneous stimuli in primates (Vierck et a1 1983).

With respect to pain types, it has been shown in sheep that the antinoceptive activity of intravenous buprenorphine is evident in response to a thermal stimulus but not mechanical pressure (Nolan eta1 1987). This has been reasoned to be due to activation ofdifferent afferents by different stimuli. However, in a comparison of pain types we would expect A-delta pain produced by thermal and electrical stimulus to be more similar, and quite distinct from C mechanore- ceptor pain, which would respond to pressure stimulus.

Administration - As most of the previous work using opioids in sheep has been based on intravenous rather than intramuscular administration, the lack of analgesic response may well be due to the different routes of administration, which can cause changes in the pharmacological profile of these drugs, thus altering their absorption, bioavailability and distribution (Baggot 1992). The proposition that the lack of an analgesic response could be explained by the pharma- cokinetics of the drugs as administered was tested by a study of venous blood samples. This revealed a profile of methadone concen- trations that suggest that a sufficient plasma concentration was available to produce anti-nociception to the electrical stimulus (Fig- ure I). With studies in humans (Shir et a1 1990) post-operative analgesia was maintained with about half the methadone plasma concentrations we recorded in the sheep.

Increased doses of methadone and buprenorphine were given in order to determine whether sufficient dosage was given to prevent nociception to the electrical stimulus. Buprenorphine was given at varying doses from 0.005 to 0.2 mgkg, whilst methadone was given at doses from 0.2 to 1 m a g , with no change in the response to electrical stimulus.

Some of the conflicting findings regarding the action of opioids in sheep may also be due to methodological aberrations within some studies.

The apparent stoicism or limited expression of indicators of pain in sheep can lend itself to the observer regarding the animal as being able to ‘cope’ with pain or painful procedures without the need for analgesics or to perhaps appear in a subjective assessment that a given analgesic is being effective.

Another possible methodological problem may be error in defining endpoints during algesimetric testing. We (Ludbrook et a1 1995). as have others (Nolan et a1 1985; Livingston et a1 1991). observed the agitation in sheep administered high doses of intravenous opioids. These effects included increased spontaneous head movement, in- tense chewing and increased locomotor activity. These are contrary to the response seen in humans when administered opioids, and may be kappa receptor mediated. They are not related to analgesia, and changes in leg lift or ear flick response (as used in thermal testing) cannot be reliably used as an index of analgesia when both these responses are so significantly altered. A discrete and purposeful ear flick is very difficult to determine in an agitated and confused animal that frequently moves its head. Likewise, a voluntary leg lift on a weight-bearing limb by a sheep in a similar state may also be difficult to distinguish from the general increased movement of the sheep. These dysphoric effects were not observed after the intramuscular doses of methadone and buprenorphine in the present study, which may be due to differences between intramuscular and intravenous administration.

Nunixin meglumine is a potent anti-inflammatory agent and ap- pears to have some analgesic properties, particularly in dogs. The proportion of its analgesic action that can be ascribed to its anti-in- flammatory mechanism remains to be determined, but in the sheep i t is best considered an anti-inflammatory agent applicable to par- ticular post-operative circumstances. This drug had no measurable effect on the response of sheep to electrically induced pain in this study.

Xylazine produced significant dose-dependent changes in the response to the electrical stimulus. These were determined to be ‘real’ effects and not changes induced by inhibited locomotor activ- ity. This was shown during maximal drug effect where the current threshold required to produce leg lift was at a level that would be equatable to significant levels of pain (as tested on one of the authors), yet an innocuous non-painful stimulus such as touching the leg would elicit a leg-lifting response. So it would appear that the transmission of sensory signals can remain unaffected by these doses of xylazine and that any increase in the current required to produce a leg lifting response is not due to a depression of central reflexes. Thus, xylazine at sub-anaesthetic doses would appear to effectively alter the sheep’s awareness of pain but not its ability to respond to the pain.

The range of response from the three doses demonstrates that varying levels and combinations of anaesthesidsedatiodanalgesia can be achieved by careful dose selection. These can range from light analgesia with no sedation to profoundly sedated, anaesthetic type. states. The only major difference between the 0.05 and 0.1 mgkg analgesic response was a more rapid onset of effect of the higher dose. This may indicate that with increased dosage a plateau of analgesic effect is reached beyond which increased pain relief comes at the price of greater sedative effects, as seen for the 0.2 mgkg dose.

From this experimental paradigm and from clinical evidence the aZadrenoceptor agonists would seem to hold the key to successful acute pain management in the sheep. Indeed, the range of new a2-adrenoceptor agonists may be even more successful in producing analgesia. This study highlights the need for further investigation into new agents and ideal administration regimens for satisfactory analgesic effect and duration while minimising unwanted anaesthetic and sedative effects such as respiratory depression or sternal recum- bency.

This study also confirms that many standard analp -ics, which may exhibit substantial analgesic properties in other species, can be ineffective in the sheep. The choice of analgesic for controlling pain in the sheep, and indeed for any species, should be based on evidence of its efficacy particular to that species.

Ausiralian Veterinary Journal Vol. 73, No. 4, April 1996 131

Acknowledgments We wish to thank Mr Brian Lewis for coordinating access to IMVS

Animal Research facilities and assistance provided, and Ms Loren Matthews for technical assistance. Dr Ted Mayther is thanked for statistical advice, and Mr Colin McLean, Flinders Medical Centre, for the methadone sample assays.

References Baggot JD (1992) Clin Pharmacokinet 22:254 Cooper BY, Vierck CJ and Franzen 0 (1982) Soc Neurosci Abstr 8510 Davis LE (1983) In Animal Pain:Perception and Alleviation, edited by

Kitchell RI, Erickson HH, American Physiological Society, Bethesda. p 161

Gourlay GK. Wilson PR and Glynn CJ (1982) Anesthesiology 57:458 Gray EC, Ludbrook GL.Upton RN and Grant C (1994) Proc Snc Clin Exp

Hall LW and Clarke K (1983) In Veterinary Anaesthesia, 8th edn, Bailliere P harmacol Toxicol I :47

Tindall, London. p 53

Huang YF. Upton RN, Rutten AJ and Runciman WB (1992) Er J Anaesth

Kamerling S, Keoween M, Bagwell C and Jochle W (1991) Acta Vet S c a d

Livingston A, Acevedo MEG, Kyles A and Waterman A (1991) Acta Vet

Ludbrook GL, Grant C, Upton RN and Penhall C (1995) JPharmacol Toxicol

Nolan AM, Livingston A and Waterman A (1985) Er J Pharmacol Suppl

Nolan A. Livingston A and Waterman AE (1 987) Er J P harmacol92:527 Shir Y, Eimerl D, Magora F. Damm D, Schulte-Monting J and Chrubasik J

(1990) Br J Anaesth 65204 Vierck CJ, Cooper BY and Cohen RH (1983) In Animal Pain:Perception and

Alleviation, edited by Kitchell RI, Erickson HH, American Physiological Society, Bethesda, p 161

Waterman A, Livingston A and Bouchenafa 0 (1988) Neuropharmacology 27:2 13

(Accepted for publication 24 October 1995)

69:368

Supp 87: 161

Scand Supp 87: 170

Methods 33:17

86:462P

Adverse drug reactions: Report of the Australian Veterinary Association Adverse Drug Reaction Subcommittee, 1994

JE MADDJSON Convenor, AVA Adverse Drug Reaction Subcommittee,

Department of Pharmacology, The University of Sydney, New South Wales 2006

SUMMARY Seventy-seven reports of suspected adverse drug reactions (ADRs) were received by the Adverse Drug Reaction Subcommittee (ADRSc) of the Australian Veteri- nary Association from April 1993 to December 1994 inclusive.

The number of reports receivedhumber of animals involved per species were: dogs (32/44), cats (18/31), horses (17/48), and cattle (10/21). Of these, 49 (64%) were classified as definite ADRs and 9 (12%) as probable ADRs. In 11 (14%) reports an ADR could not be substantiated or there was insufficient information available to make a decision. Eight reports were not classified because the manufacturer and the ADRSc disagreed as to the appropriate classification.

Sixteen reports involved apparent hypersensitivity reactions, which resulted in death on 6 occasions. Six reports were associated with ’off label‘ use and 1 report with use of an expired product. Of the definite, probable and unclassified reports of suspect ADRs, the most frequent types of drugs involved were antimicrobial drugs (13 reports), an- thelmintics (131, insecticides (13, vaccines (lo), nonsteroidal anti-inflammatory prepara- tions (5), chondroprotective agents (4), anaesthetidsedative agents (4) and vitamin preparations (2). Single reports concerning definite, probable or unclassified ADRs to a vasodilator, a corticosteroid, a local anaesthetic and a disinfectant were received. Aust Vet J 73: 132 - 136

Introduction This is the third annual report of the Adverse Drug Reaction

Subcommittee (ADRSc) of the Australian Veterinary Association. Previous reports (Maddison 1992, 1994) have discussed the report- ing of suspected adverse drug reactions (ADRs) and the reader is referred to these reports for further information.

In general, ADRs refer to abnormalities that occur when a drug is administered at an appropriate dose for the purpose intended. How- ever, in veterinary medicine there are conditions for which drugs are used outside the manufacturer’s recommendations (‘off-label’ use, see Maddison 1992). It is appropriate that data be collated also on suspected ADRs to drugs used ‘off label’ because such use is a relatively common occurrence, particularly in companion animal practice.

For the purposes of this annual report, the reports of suspected ADRs received by the ADRSc have been classified into 4 groups:

1. Definite ADR: The ADRSc and manufacturer agree that there is no rational explanation other than an ADR for the animal’s clinical signs.

2. Probable ADR: There is insufficient evidence to confirm an ADR but an ADR is more likely than other possibilities and/or cannot be excluded as a possibility.

3. Non-ADR: Definitely not an ADR (other more likely explana- tion) or not enough information to allow classification.

4. Unclassified ADR: This is a new category for this report. Reports were unclassified if the manufacturer and the ADRSc dis- agreed in their assessment of the report.

132 Australian Veterinary Journal Vol. 73, No. 4, April 1996