Adrenoceptor Agonists and Vatinoxan, a Peripherally Acting Alpha

Department of Equine and Small Animal Medicine Faculty of Veterinary Medicine University of Helsinki Finland

Interaction of alpha2-adrenoceptor agonists and vatinoxan,

a peripherally acting alpha2-adrenoceptor antagonist, in horses

SOILE PAKKANEN

sss

ACADEMIC DISSERTATION

To be presented for public examination, with the permission of the Faculty of Veterinary Medicine, University of Helsinki, in the Paatsama hall, Koetilantie 4, Helsinki, March 22th 2019, at 12 noon.

Helsinki 2019 Supervised by

Professor Outi Vainio, DVM, PhD, DECVPT Docent Marja Raekallio, DVM, PhD Docent Anna Mykkänen, DVM, PhD Faculty of Veterinary Medicine University of Helsinki Finland

Reviewed by

Professor Rikke Buhl, DVM, PhD Faculty of Health and Medical Sciences University of Copenhagen Copenhagen Denmark

Professor Francisco José Teixeira-Neto, DVM, PhD Department of Veterinary Surgery and Anaesthesiology Faculdade de Medicina Veterinária e Zootecnia Universidade Estadual Paulista (UNESP) Botucatu Brazil

Opponent

Professor Görel Nyman, DVM, PhD, DECVAA Faculty of Veterinary Medicine and Animal Science Swedish University of Agricultural Sciences Uppsala Sweden

ISBN 978-951-51-4897-1 (paperback) ISBN 978-951-51-4898-8 (pdf) http://ethesis.helsinki.fi

Layout by Maija Savolainen

Unigrafia Helsinki 2019

2 3 Supervised by

Professor Outi Vainio, DVM, PhD, DECVPT Docent Marja Raekallio, DVM, PhD Docent Anna Mykkänen, DVM, PhD Faculty of Veterinary Medicine University of Helsinki Finland

Reviewed by

Professor Rikke Buhl, DVM, PhD Faculty of Health and Medical Sciences University of Copenhagen Copenhagen Denmark

Opponent

Professor Görel Nyman, DVM, PhD, DECVAA Faculty of Veterinary Medicine and Animal Science Swedish University of Agricultural Sciences Uppsala Sweden To my Family

ISBN 978-951-51-4897-1 (paperback) ISBN 978-951-51-4898-8 (pdf) http://ethesis.helsinki.fi

Layout by Maija Savolainen

Unigrafia Helsinki 2019

2 3 CONTENTS

CONTENTS ...... 4 ABSTRACT ...... 6 LIST OF ORIGINAL PUBLICATIONS ...... 7 ABBREVIATIONS ...... 8 1. INTRODUCTION ...... 9 2. REVIEW OF THE LITERATURE ...... 10 2.1 Alpha-adrenergic receptors ...... 10

2.2 Central α2-adrenoceptors ...... 10

2.3 Peripheral α2-adrenoceptors...... 10 2.4 Central and peripheral imidazoline receptors ...... 11 2.5 Detomidine and romifidine in horses...... 11 2.5.1 Central effects of detomidine and romifidine ...... 11 2.5.1.1 Sedative effect...... 11 2.5.1.2 Antinociceptive effect...... 12 2.5.2 Detomidine and romifidine as preanaesthetics before general anaesthesia...... 13 2.5.3 Pharmacokinetics of detomidine and romifidine...... 13 2.5.4 Cardiovascular and respiratory effects of detomidine and romifidine...... 15 2.5.5 Gastrointestinal effects of detomidine and romifidine...... 16 2.5.6 Effects of detomidine and romifidine on selected neurohormonal variables and changes related to energy metabolism...... 16

2.6. Alphα2-adrenoceptor antagonists in horses ...... 17 2.7 Vatinoxan...... 18 2.7.1 Pharmacokinetics of vatinoxan...... 18 2.7.2 Cardiovascular effects of vatinoxan ...... 19 2.7.3 Metabolic effects of vatinoxan...... 19 2.7.4 Other effects of vatinoxan ...... 19

2.8 Simultaneous use of vatinoxan and α2-adrenoceptor agonists...... 19

2.8.1 Vatinoxan and cardiovascular effects of 2α -adrenoceptor agonists...... 19 2.8.2 Effect of vatinoxan on drug plasma concentrations and sedation...... 20

2.8.3 Influence of vatinoxan on metabolic and other effects of2 α -adrenoceptor agonists...... 21

3. AIMS OF THE STUDY...... 22 4. MATERIALS AND METHODS...... 23 4.1 Animals...... 23 4.2 Instrumentation...... 24 4.3 Treatments...... 24

4 5 4.4 Study designs...... 25 4.5 Clinical assessment of sedation and general anaesthesia...... 25 4.6 Analyses of drug plasma concentrations ...... 26 4.7 Pharmacokinetic calculations...... 26 4.8 Cardiorespiratory monitoring ...... 26 4.9 Assessment of intestinal borborygmi...... 27 4.10 Biochemisrty analyses from serum and plasma ...... 27 4.11 Statistical analyses and power calculations...... 28

5. RESULTS...... 29 5.1 Clinical sedation (I, II)...... 29 5.2 Induction time...... 29 5.3 Plasma drug concentrations...... 30 5.4 Cardiovascular effects...... 33 5.4.1 Cardiovascular effects of detomidine, romifidine and MK-467 in conscious horses (I, II)...... 33 5.4.2 Cardiovascular performance in horses under general anaesthesia (III)...... 34 5.5 Intestinal borborygmi (I,II)...... 36 5.6 Effects of romifidine and MK-467 on selected neurohormonal variables and changes related to energy metabolism (IV)...... 37

6. DISCUSSION ...... 40 6.1 Clinical sedation and plasma drug concentrations...... 40 6.2 Cardiovascular effects...... 41 6.2.1 Cardiovascular effects in conscious horses...... 41 6.2.2 Cardiovascular effects in horses under general anesthesia...... 42 6.3 Effects on intestinal borborygmi...... 42 6.4 Effects on selected neurohormonal variables and changes related to energy metabolism...... 43 6.5 Study limitations and methodological considerations...... 44 6.6 Clinical implications and future prospects...... 46

7. CONCLUSIONS...... 47

ACKNOWLEDGEMENTS...... 48 REFERENCES ...... 50 APPENDIX ...... 61

4 5 ABSTRACT

These studies aimed to investigate the effects of vatinoxan alone and combined with detomidine and romifidine in horses. Detomidine and romifidine are 2α -adrenoceptor agonists that are widely used in equine medicine for their sedative and analgesic properties. Vatinoxan, in turn, is an α2-adrenoceptor antagonist that differs from the commercially available and more extensively studied α2-adrenoceptor antagonists in its poor ability to cross the blood-brain barrier, limiting its actions mainly to peripheral α2-adrenoceptors.

The main objective of these studies was to evaluate the potential of vatinoxan in preventing the adverse effects, mediated by peripheral 2α -adrenoceptors, of detomidine and romifidine, while sparing the desired sedative action in horses. These adverse reactions include cardiovascular effects (decreased heart rate and cardiac output, increased systemic vascular resistance), decreased intestinal motility and changes in plasma glucose concentrations. In addition, plasma drug concentrations were investigated to elucidate the pharmacokinetic agonist-antagonist interactions.

To assess cardiovascular effects of the drugs, heart rate and arterial and central venous pressures were measured in horses treated with α2-adrenoceptor agonists, vatinoxan and a combination of these drugs. In horses under general anaesthesia, premedicated with either detomidine alone or combined with vatinoxan, cardiac index was measured and systemic vascular resistance and tissue oxygen delivery were calculated. Sedation and intestinal borborygmi were scored after treatment with detomidine, romifidine and vatinoxan or a combination of vatinoxan with the α2-adrenoceptor agonists. From venous blood samples, plasma glucose, ACTH and lactate, serum insulin, FFA, cortisol and triglyceride, and whole-blood potassium and sodium were analysed.

When administered in combination with α2-adrenoceptor agonist, vatinoxan attenuated the bradycardia, hypertension and intestinal hypomotility induced by detomidine and romifidine, while the effect on sedation was minor. Vatinoxan alone increased heart rate, but had no effect on mean arterial blood pressure. Combined premedication with vatinoxan and detomidine resulted in severe hypotension during general anaesthesia, and high doses of dobutamine were required for blood pressure support. In addition, combined premedication resulted in decreased systemic vascular resistance and increased cardiac index and tissue oxygen delivery compared with premedication with detomidine alone. These improvements in cardiovascular performance may, in fact, be at least partly related to high doses of dobutamine. Vatinoxan affected disposition of detomidine and romifidine, leading to greater volume of distribution and lower plasma concentration of these

α2-adrenoceptor agonists.

Romifidine induced hyperglycaemia in horses, which vatinoxan alleviated, but vatinoxan alone had no effect on plasma glucose concentration. Serum FFA concentration was significantly higher after vatinoxan alone than after romifidine alone or combined with vatinoxan.

The results indicate that vatinoxan was able to alleviate the cardiovascular and intestinal effects of detomidine and romifidine in standing horses, while no clinically relevant effect occurred on sedation. When vatinoxan was administered before general anaesthesia, marked hypotension followed. Romifidine-induced hyperglycaemia was attenuated by vatinoxan. Vatinoxan increased the volume of distribution and decreased plasma concentrations of intravenously administered detomidine and romifidine. Romifidine had no significant impact on plasma concentration of vatinoxan.

6 7 LIST OF ORIGINAL PUBLICATIONS

This thesis is based on the following publications, which in the text are referred to by their Roman numerals:

I Vainionpää M, Raekallio M, Pakkanen S, Ranta-Panula V, Rinne V, Scheinin M, Vainio O. Plasma

drug concentrations and clinical effects of a peripheral alpha-2-adrenoceptor antagonist, MK-467, in horses sedated with detomidine. Veterinary Anaesthesia and Analgesia, 2013, 40, pp 257-264.

II de Vries A, Pakkanen S, Raekallio M, Ekiri A, Scheinin M, Taylor P, Vainio O. Clinical effects and

pharmacokinetic variables of romifidine and the peripheral 2α -adrenoceptor antagonist MK-467 in horses. Veterinary Anaesthesia and Analgesia, 2016, 43, pp 599-610.

III Pakkanen S, Raekallio M, Mykkänen A, Salla K, de Vries A, Vuorilehto L, Scheinin M, Vainio O.

Detomidine and the combination of detomidine and MK-467, a peripheral alpha-2 adrenoceptor antagonist, as premedication in horses anaesthetized with isoflurane.Veterinary Anaesthesia and Analgesia, 2015, 42, pp 527-536.

IV Pakkanen S, de Vries A, Raekallio M, Mykkänen A, Palviainen M, Sankari S, Vainio O. Changes in energy metabolism, and levels of stress-related hormones and electrolytes in horses after

intravenous administration of romifidine and the peripheral alpha-2 adrenoceptor antagonist vatinoxan. Acta Veterinaria Scandinavica, 2018, 60: 27.

These publications have been reprinted with the permission of their copyright holders. In addition, some unpublished material is presented.

6 7 ABBREVIATIONS

DO2 tissue oxygen delivery

FFA free fatty acid

FiO2 inspired oxygen fraction

H+ hydrogen ion

HR heart rate

IM intramuscular

IV intravenous

K+ potassium

LiDCO lithium dilution technique

MAC minimum alveolar concentration

MAP mean arterial blood pressure

Na+ sodium

NLC nucleus locus coeruleus

PaCO2 arterial partial pressure of carbon dioxide

PaO2 arterial partial pressure of oxygen

PeCO2 end-tidal partial pressure of carbon dioxide

RIA radioimmunoassay

SaO2 arterial oxygen saturation

SD standard deviation

SVR systemic vascular resistance

T½ß elimination half-life

Tmax time to maximum plasma concentration

Vd volume of distribution

8 9 Introduction

1. INTRODUCTION

Alphα2-adrenoceptor agonists, such as detomidine and romifidine, are widely used sedative agents in equine medicine. In addition to their sedative properties, they induce antinociception, which is useful for minor veterinary procedures. These advantageous effects come, however, with some less desired side-effects. These include changes in cardiovascular function manifesting as bradycardia, conduction disturbances, increased peripheral vascular resistance and decreased intestinal motility. The latter may be particularly harmful in horses, a species prone to ileus and colic.

The effects of 2α -adrenoceptor agonists can be prevented or reversed by α2-adrenoceptor antagonists; atipamezole is in clinical use in small animals. In recent years, increasing interest has focused on vatinoxan, an

α2-adrenoceptor antagonist that acts mainly on peripheral α2-adrenoceptors because of its low lipophilicity, resulting in poor ability to cross the blood-brain barrier (Clineschmidt et al. 1988). The peripherally re-

stricted action of vatinoxan creates a possibility to counteract the undesired side-effects of α2-adrenoceptor

agonists, which are mediated mainly via peripheral α2-adrenoceptors, while preserving the sedative and analgesic effects needed for veterinary procedures.

Before our studies, the use of vatinoxan had been studied mostly in a restricted number of species, such as sheep and dogs; only one study on cardiovascular effects of medetomidine and vatinoxan included horses (Bryant et al. 1998). The aim of our studies was to investigate the potential of vatinoxan in counteracting the peripheral effects, with the main focus on cardiovascular function during sedation and general anaesthesia and intestinal hypomotility, of detomidine and romifidine in horses. The effects of combined administration of romifidine and vatinoxan on selected neurohormonal variables and changes related to energy metabolism were also assessed. Lastly, the plasma concentrations of detomidine, romifidine and vatinoxan and the inter-

action between these α2-adrenoceptor agonists and the antagonist affecting each other’s plasma concentrations were studied to gain a clear understanding of the observations made during the trials.

Our hypothesis was that vatinoxan would antagonize the peripheral side-effects of detomidine and romifidine while preserving the sedative effect. In practice, this would be manifested as stable heart rate (HR), arterial blood pressure (ABP) and systemic vascular resistance (SVR), and conserved intestinal motility

when vatinoxan is combined with the α2-adrenoceptor agonist. Furthermore, we hypothesized that vatinoxan would have an effect on disposition of detomidine and romifidine, which would be seen as increased volume

of distribution (Vd) and decreased area under plasma concentration time curve (AUC) of these drugs. In addition, we expected romifidine to induce hyperglycaemia, hypoinsulinaemia and a decrease in serum FFA concentrations, and that these effects would be prevented when romifidine is administered in combination with vatinoxan.

8 9 Review of the literature

2. REVIEW OF THE LITERATURE 2.1 Alpha-adrenergic receptors

Adrenergic receptors and their natural ligands noradrenaline and adrenaline mediate the actions of the sympathetic nervous system, which controls a wide variety of vital functions of the body. The classification of adrenergic receptors to alpha (α) and beta (β) was initially presented by Ahlquist (1948). Further division of

α-adrenergic receptors to α1 and α2 subclasses was made according to their differences in affinity to receptor ligands (Langer 1974). The classic example of α1-adrenoceptor is that it is located at the postsynaptic nerve terminal and ligand binding results in activation of the target cell. The α2-adrenoceptor, in turn, was first discovered at a presynaptic location exerting an inhibitory action of noradrenaline release from the nerve terminal (Berthelsen and Bettinger 1977), although these receptors were later found also on postsynaptic sites (Drew and Whitling 1979; Langer et al. 1980; Madjar et al. 1980). Furthermore, the α1-adrenoceptors were divided into α1A, α1B and α1D (see review in Zhong and Minneman 1999) and α2-adrenoceptors into α2A,

α2B, α2C and α2D subclasses (Bylund et al.1988; Blaxall et al. 1991). Alphα2A receptor subtype was first discovered in human platelets and rabbit spleen (Bylund et al.1988; Michel et al. 1989), α2B in rat lung and kidney

(Latifpour et al. 1982; Bylund et al.1988), α2C in opossum kidney cells (Murphy and Bylund 1988; Blaxall et al. 1991) and α2D in rat submaxillary gland and bovine pineal gland (Michel et al. 1989; Simonneaux et al. 1991). Alphα2A and α2D subtypes are, in fact, very similar in structure and there is a possibility that they are the same receptor discovered in different mammalian species and the fourth 2α -adrenoceptor subtype is actually found in zebrafish (Ruuskanen et al. 2004, see commentary in Bylund 2005).

2.2 Central α2-adrenoceptors

The sedative action of α2-adrenoceptor agonist was discovered when clonidine induced sleep in several animal species by central sympatholytic effect, decreasing monoaminergic transmission and inhibiting the spontaneous firing of noradrenergic neurons in mammalian nucleus locus coeruleus (NLC) (Holman et al. 1971; Svensson et al. 1975). The NLC is the center of noradrenergic pathways in the central nervous system.

It is densely populated with α2-adrenoceptors and is located in the close vicinity of the fourth ventricle at the level of the pons (see review in Foote et al. 1983). In addition to induction of sleep, clonidine is reported to alter behaviour and sexual activity, modify anxiety, learning and memory, cause ataxia and decrease motility in rodents (Nomura et al. 1980; Clark et al. 1984; Delini-Stula 1984; Arnsten and Goldman-Rakic 1985;

Luttinger et al. 1985; Pellow et al. 1985). Several other α2-adrenoceptor agonists, e.g. detomidine and dexmedetomidine, are reported to cause sedation mediated by central α2- adrenoceptors in laboratory animals (Hsu 1981; Virtanen et al. 1985; Doze et al. 1989). The hypnotic action of dexmedetomidine was located at NLC (Boyajian et al. 1987; Correa-Sales et al. 1992).

In addition to sedative and behaviour-modifying action (later this action was specified to be related to

α2A-adrenoceptor), α2-adrenoceptors mediate central hypotension caused by α2-adrenoceptor agonists (Kubo and Misu 1981; MacMillan et al. 1996). Alphα2-adrenoceptors in the central nervous system also alter nociception (Luttinger et al. 1985; Jones and Gebhart 1986), thermoregulation (Myers et al. 1987) and induce mydriasis (Berridge et al. 1983).

2.3 Peripheral α2-adrenoceptors

Alphα2-adrenoceptor activation is known to affect numerous body functions also in peripheral receptor sites.

Soon after their discovery, agonists of α2-adrenoceptors were noted to cause vasoconstriction (Timmermans and van Zwieten 1980). Later, α2A and α2B-adrenoceptors in vascular smooth muscle were specified to medi-

10 11 Review of the literature

ate peripheral vasoconstriction and hypertensive effect, whereas activation of 2Cα -receptor subtype does not result in cardiovascular effects (Link et al. 1996; MacMillan et al. 1996). Hypertension is reported to result in bradycardia and conduction disturbances mediated by baroreceptor reflex (Sarazan et al. 1989).

Another body system that is markedly affected by 2α -adrenergic function is the gastrointestinal tract. Gastro-

intestinal motility and gastric acid secretion are decreased by α2-adrenoceptor agonists, the latter of which might also involve a central mechanism (Ruwart et al. 1980; Del Tacca et al. 1982). Furthermore, activation

of α2-adrenoceptors inhibits lipolysis in human adipose tissue (Lafontan and Berlan 1980) and insulin secretion from pancreatic β-cells in mice, resulting in hyperglycaemia (Nakadate et al. 1980; Angel et al. 1988).

Lastly, α2-adrenoceptors mediate platelet aggregation (Barthel and Marward 1974; Shattil et al. 1981) and antagonize the effects of antidiuretic hormone, leading to increased urine production (Smyth et al. 1985; Nuñez et al. 2004). 2.4 Central and peripheral imidazoline receptors

Central imidazoline receptors were discovered to participate in the central hypotensive effect of clonidine,

an α2-adrenoceptor agonist with imidazoline structure (Bousquet et al. 1984; Ernsberger et al. 1990). After- wards, three separate types of imidazoline receptors have been described according to their affinities to dif-

ferent ligands. These are clonidine-preferring I1 and its ligand imidazoline receptor antisera selected (IRAS)

(Piletz et al 2000), idazoxan-binding I2 and its ligand monoamine oxidase (MAO) (Tesson et al. 1995) and

I3, which shows insulinotropic action independent of α2-adrenergic action (Schultz and Hasselblatt 1989). As

many of the other α2-adrenoceptor agonists possess an imidazoline ring in their structure, their actions might

also be mediated by imidazoline receptors, in addition to α2-adrenoceptors (Göthert and Molderings 1991). 2.5 Detomidine and romifidine in horses

Alphα2-adrenoceptor agonists are widely studied and used in veterinary medicine for sedation, analgesia and premedication before general anaesthesia. The most widely used drugs in this group are medetomidine

and dexmedetomidine, which are also the most α2-specific. Xylazine (IUPAC N-(2,6-dimethylphenyl)-5,6- dihydro-4H-1,3-thiazin-2-amine), detomidine (IUPAC 5-[(2,3-dimethylphenyl)methyl]-1H-imidazole) and romifidine (IUPAC N-(2-bromo-6-fluorophenyl)-4,5-dihydro-1H-imidazol-2-amine), in turn, are commonly

used in equine medicine (see review in England and Clarke 1996). Reported α2:α1 specificities are 260:1 for detomidine (Virtanen et al. 1988) and 340:1 for romifidine (Muir 2009). Detomidine is reported not to have

selectivity to any α2-adrenoceptor subtype in vitro (Schwartz and Clark 1998).

In addition to their sedative and antinociceptive effects, 2α -adrenoceptor agonists are reported to cause a variety of other, usually less desirable, effects such as cardiac disturbances, ataxia, mydriasis, sweating, piloerection, penile prolapse and increased frequency of urination in horses (Vainio 1985; Clarke & Taylor 1986; Kamerling et al. 1988a; Hamm et al. 1995).

2.5.1 Central effects of detomidine and romifidine

2.5.1.1 Sedative effect

Detomidine is reported to induce sufficient sedation for veterinary procedures at intravenous (IV) doses of 5-20 µg/kg, although the recommended dosage was initially much greater; 100-300 µg/kg (Vainio 1985; Clarke and Taylor 1986). The sedative effect is dose-dependent, but the dose of 20 µg/kg results in maximal sedation and higher doses seem to increase only the duration of the effect (Vainio 1985; Jöchle and Hamm

10 11 Review of the literature

1986; Kamerling et al. 1988b; Hamm et al. 1995). Romifidine at a dose of 80 µg/kg causes its maximal sedative effect, which is similar to detomidine at 20 µg/kg. The depth and duration of romifidine-induced sedation also increase with the dose, although romifidine has shown a reduced depth of sedation at a high dose of 120 µg/kg (Jöchle and Hamm 1986; England et al. 1992; Hamm et al. 1995; Freeman and England 2000). Romifidine-induced sedation is, in addition, longer lasting than sedation with detomidine (Freeman and England 2000). Horses sedated with α2-adrenoceptor agonist become ataxic, and this effect is reported to be dose-dependent. Romifidine is considered to induce less ataxia than detomidine (Clarke et a. 1991; England et al. 1992; Hamm et al. 1995; Freeman and England 2000).

Detomidine is useful as a sedative also via intramuscular (IM) administration, but the doses needed to achieve a proper sedative effect are higher than with the IV route. Intramuscular administration is particularly useful in horses that are difficult to handle (Clarke and Taylor 1986; Jöchle and Hamm 1986). The time required to achieve a sedative effect with IM dosing is greater (up to 12 minutes) than with IV dosing (30 seconds) (Lowe and Hilfiger 1986). During the last few years detomidine has been registered also for sublingual administration and provides a reliable sedative effect also by this route. The sedative effect of detomidine administered sublingually at a dose of 40 µg/kg is reported to be appropriate for routine veterinary and husbandry procedures (Gardner et al. 2010) and maximal sedation is achieved 60-100 minutes after administration (Dimaio Knych and Stanley 2011; Kaukinen et al. 2011; L’Ami et al. 2013).

Detomidine and romifidine are routinely combined with opioids, the most widely used of which is butorphanol. Combination usage results in better antinociception and sedation, manifested as decreased responsiveness to touch or audiovisual stimuli, which is very useful during veterinary procedures (Clarke and Paton 1988; Clarke et al. 1991).

2.5.1.2 Antinociceptive effect

The mechanism of antinociception caused by α2-adrenoceptor agonists was investigated in dogs; it was found to be the result of their action at the level of the spinal cord (Sabbe et al. 1994). Variability in methods by which the antinociceptive effect is assessed in horses limits the comparison of published studies. Furthermore, the effect of detomidine on the nociceptive threshold depends on the method used or even the anatomical location to which the stimulus is targeted (Kamerling et al. 1988b). In the majority of the reports, detomidine increases the nociceptive threshold for mechanical stimuli in the limbs and trunk (Vainio 1985; Chambers et al. 1993; Moens et al. 2003; Mama et al. 2009), noxious thermal stimulus of the skin (Kamer- ling et al. 1988b), stimulus created by electric current (Jöchle and Hamm 1986; Hamm et al. 1995; Moens et al. 2003; Rohrbach et al. 2009) and distention of viscera in horses (Lowe and Hilfiger 1986; Elfenbein et al. 2009). Published reports indicate that the detomidine dose of 20 µg/kg produces adequate antinociception in most cases relative to doses ranging from 5 to 10 µg/kg; with doses under 20 µg/kg, the antinociceptive effect has been inconsistent. Furthermore, the duration of antinociceptive effect is dose-dependent, ranging from 5 to 320 minutes with detomidine doses of 5-160 µg/kg (Lowe and Hilfiger 1986; Kamerling 1988b; Chambers et al. 1993; Hamm et al. 1995; Mama et al. 2009). The nociceptive threshold of horses suffering from chronic pain, which results in increased sensitivity to painful stimuli, is also increased by detomidine, but less than in healthy horses (Chambers et al. 1993; Owens et al. 1996). Detomidine is reported to be a very effective analgesic in clinical cases of colic, leading to satisfactory or highly satisfactory effect in at least 96% of cases. The horses in which pain control was unsatisfactory had colic due to severe small intestinal strangulation, requiring surgical intervention or euthanasia (Jöchle 1989; Jöchle et al. 1989).

Results regarding the antinociceptive effect of romifidine are conflicting. Romifidine is reported to have no effect on the nociceptive threshold for electrical stimulus at doses of 40-120 µg/kg (Hamm et al. 1995) and for mechanical stimulus at a dose of 80 µg/kg (Moens et al. 2003) or to have a significantly lower antino-

12 13 Review of the literature

ciceptive effect at a dose of 80 µg/kg relative to 20 µg/kg of detomidine measured with electrical stimulus (Moens et al. 2003). In contrast, there are reports of good and relatively long-lasting (55-120 minutes) antinociceptive effect measured with electrical stimulus at a dose of 80 µg/kg, which induced better antinociception than 20 µg/kg of detomidine (Spadavecchia et al. 2005; Rohrbach et al. 2009).

2.5.2 Detomidine and romifidine as preanaesthetics before general anaesthesia

Alphα2-adrenoceptor agonists are commonly used as premedication before induction of general anaesthesia in horses (Clarke 2014). Detomidine is known to decrease the requirement of other anaesthetic agents when it is used as a part of the anaesthetic protocol. In laboratory animals, this is seen as lengthened barbiturate-induced sleeping time (Vainio 1985). A dose-dependent decrease in the minimum alveolar concentration (MAC) of isoflurane was also observed after injection of detomidine in equine species (Short et al. 1986a; Steffey and Pascoe 2002). Detomidine premedication was recognized to lengthen the anaesthesia induction

time compared with xylazine (Saarinen 1986). Cardiovascular effects of2 α -adrenoceptor agonists, such as decrease of HR and CO, might impair the distribution of intravenous injectables (Mama et al. 2009). A greater decrease of HR and CO after intravenous detomidine relative to xylazine (Wagner et al. 1990), causing delayed distribution of the drug to the central nervous system, could explain the differences in induction times between these agents. The quality of induction was reported to be smooth and muscle relaxation good after detomidine premedication (Saarinen 1986; Short et al. 1986a). In addition, romifidine has proven to provide sufficient sedation for smooth induction of anaesthesia with ketamine (Diamond et al. 1993).

Taylor and co-workers (2001) compared detomidine (20 µg/kg) and romifidine (100 µg/kg) as premedi-

cations before general anaesthesia in horses. They concluded that both α2-adrenoceptor agonists perform satisfactorily as premedication, and the difference between the two compounds was that romifidine caused a greater decrease of mean arterial blood pressure (MAP) during anaesthesia induced with ketamine and

maintained with halothane. The reduction of HR after α2-adrenoceptor agonists might result in decreased CO, which is commonly present in horses under general anaesthesia (Clarke 2014).

2.5.3 Pharmacokinetics of detomidine and romifidine

Publications on the pharmacokinetics of detomidine and romifidine in horses are scarce. In addition, the results vary between the studies (Salonen et al. 1988; 1989; Grimsrud et al. 2009; Wojtasiak-Wypart et al.

2012; Cenani et al. 2017). The maximum observed plasma concentration (Cmax) of detomidine is detected a few minutes after a single bolus IV dose, after which the concentration declines rapidly (Salonen et al. 1989; Grimsrud et al. 2009; Mama et al. 2009). The absorption after subcutaneous dosing in rats and IM dosing in

calves is reported to be fast, with Cmax occurring in 10-15 minutes (Salonen 1986). In horses, by contrast, the

Cmax of detomidine after IM administration was found to occur at 77 minutes (Grimsrud et al. 2009). After

sublingual administration, Cmax of detomidine is reached in 36 minutes (Dimaio Knych and Stanley 2011). The possibility that part of the product is either expelled from the mouth or swallowed after sublingual administration might explain the greater individual variation of pharmacokinetic parameters compared with the IV route (Dimaio Knych and Stanley 2011). A summary of the pharmacokinetic parameters of detomidine and romifidine is presented in Table 1.

The time-concentration curve of romifidine is very similar to that of detomidine, with maxC occurring a few minutes after IV administration, and fitted best the two-compartmental model (Wojtasiak-Wypart et al.

2012; Cenani et al. 2017). Compared with detomidine, romifidine has a larger Vd, faster total body clearance

(Cl) and longer elimination half-life (T½ß).

12 13 Review of the literature

Table 1. Summary of basic pharmacokinetic parameters in median (range) or mean (± SD) of detomidine and romifidine in horses.

Drug Route Dose Vd Cl T½ß Reference (µg/kg) (l/kg) (ml/kg/min) (min)

DET * IV 50 6.8 8.1 582 Salonen 1986

DET IV 80 0.74 7.1 71.4 Salonen et al. 1989 (± 0.25) (±1.6) (± 16.2)

DET IM 80 1.56 10.1 106.8 Salonen et al. 1989

DET IV 30 0.48 11.66 24.1 Grimsrud et al. 2009 (0.22–0.69) (10.10–18.37) (12.7–41.3)

DET IM 30 NA NA 51.4 Grimsrud et al. 2009 (35.9–90.1)

ROM IV 80 4.56 32.4 138.2 Wojtasiak- (3.63–5.12)ss (25.5–38.4) (104.6–171.0) Wypart et al. 2012

ROM IV 100 2.03 20 127 Cenani et al. 2017 (1.13-3.37)ss (16-26) (94-141)

* Preliminary study with one horse DET, detomidine; ROM, romifidine;ss at steady state; NA, not reported in study in question.

In earlier studies, elimination of detomidine has been described as biphasic after a dose of 50-80 µg/kg (Salonen 1986; Salonen et al. 1989), but a more recent publication with a smaller dose of detomidine (30 µg/kg) reports first-order kinetics (Grimsrud et al. 2009). The pharmacokinetic profile of detomidine is best-described with a one-compartmental model (Elfenbein et al. 2009; Grimsrud et al. 2009; Hubbel et al.

2009). Detomidine seems to alter its own pharmacokinetics, as Cl is decreased and T½ß increased after larger doses (Salonen et al. 1989; Elfenbein et al. 2009). This phenomenon is reported also in dogs treated with medetomidine and dexmedetomidine and is explained by decreased CO and peripheral vasoconstriction, which change the disposition of the drugs and prolong their elimination (Salonen et al. 1995, Honkavaara et al. 2012).

Pharmacokinetics of IV detomidine is also altered by exercise; Hubbel and co-workers (2009) reported significantly increased ½ßT , Vd and residence time after maximal exercise compared with non-exercised horses. Increased CO, redistribution of blood flow and changes in hydrogen (H+) ion concentration of tissues after exercise were considered to be the most important factors contributing to these changes; increased CO and blood flow to muscle tissue may have resulted in increased Vd, decreased blood flow to liver and kidneys, decreased metabolic rate and increased H+ concentration in tissues via excretion to urine, leading to ion trapping of detomidine to tissues (Hubbel et al. 2009).

Elimination and metabolites of detomidine in urine were first investigated in rats, in which the most important route of elimination of detomidine is biotransformation in the liver (Salonen and Suolinna 1988). Detomidine undergoes hydroxylation of the methyl substituent at the 3-position of the aromatic ring. The

14 15 Review of the literature

metabolite, hydroxydetomidine, is then further oxidized to detomidine carboxylic acid and conjugated to glucuronides. Some other minor elimination pathways were also reported in the rat (Salonen et al. 1988). Later, urine was analysed for metabolites of detomidine after peroral and IV administration in horses, and the metabolic pathway was reported to be similar to that in the rat, although the ratios of the metabolites varied between the species. In the horse, oxidation of hydroxydetomidine was the major pathway, and the rate of glucuronide conjugation was limited (Salonen et al. 1992). In rat urine, the major metabolites present were reported to be glucuronide conjugates (Salonen et al. 1988), in contrast to horses, in which the detomidine carboxylic acid was most abundant metabolite. In horses, the glucuronide conjugates might be excreted mostly in bile (Salonen et al. 1992). According to the report of the European Medicines Agency, romifidine is degraded very quickly in horses after IV administration, with elimination occurring mainly via kidneys and to a lesser extent via faeces (EMA 2013).

2.5.4 Cardiovascular and respiratory effects of detomidine and romifidine

The cardiovascular actions of detomidine were initially studied in laboratory animals. Detomidine was discovered to dose-dependently lower HR and to cause a biphasic effect on blood pressure in rats and cats. After detomidine administration, blood pressure is initially increased, after which a second mildly hypoten-

sive phase follows (Savola 1986). Hypertension caused by α2-adrenoceptor agonists in rats is mediated by

peripheral α2-adrenoceptors on vascular smooth muscle (Savola 1989), whereas hypotensive and bradycardic, predominantly vagally mediated, effects in rats (De Jonge et al. 1982) and cats (Kitagawa and Walland 1982; Savola et al. 1985) were found to be mediated by a central mechanism. The same effect is seen also in horses after administration of detomidine. The increase in SVR (Wagner et al. 1991; Yamashita et al. 2000) and, as a result, hypertension are evident in minutes after IV administration of detomidine (Vainio 1985; Clarke and Taylor 1986; Short et al. 1986a; Clarke and Paton 1988; Sarazan et al. 1989; Raekallio et al. 1991a; Yamashita et al. 2000). Although vasoconstriction is manifested also as reduced coronary arterial flow in horses, no evidence of myocardial ischaemia was detected (Sarazan et al. 1989). Furthermore, signs

of myocardial ischaemia were not detected even after large doses of α2-adrenoceptor agonist dexmedetomidine in pigs (Jalonen et al. 1995). A decreasing effect on ABP is reported after initial hypertension, especially with lower doses (10-20 µg/kg), but the decrease is rather mild (Clarke and Taylor 1986; Clarke and Paton 1988; Sarazan et al. 1989), and it is not detected in all studies (Yamashita et al. 2000). Initial hypertension followed by a decrease in ABP is also detected after romifidine (Clarke et al. 1991).

In horses, the hypertension results in a secondary decrease of HR accompanied by bradyarrhythmia manifested as sino-atrial and atrio-ventricular (AV) blocks (Sarazan et al. 1989), which are detected in the electrocardiogram of horses treated with detomidine and romifidine (Buhl et al. 2007). The effect is mediated by baroreceptor reflex (see review in Fernandez et al. 2015). This negative chronotropism is seen after IV administration of detomidine (Vainio 1985; Clarke and Taylor 1986; Short et al. 1986a; 1986b; Clarke and Paton 1988; Sarazan et al. 1989; Raekallio et al. 1990; 1991a; England et al. 1992; Buhl et al. 2007) and romifidine (Gasthuys et al. 1990; Clarke et al. 1991; England et al. 1992; Diamond et al. 1993; Buhl et al. 2007) and after IM (Wagner et al. 1991; Mama et al. 2009) and sublingual (Dimaio Knych and Stanley 2011; Kaukinen et al. 2011; L’Ami et al. 2013) administration of detomidine. The duration of negative chronotropism is reported to be dose-dependent (Kamerling et al. 1988a), lasting 45-240 minutes after detomidine IV or IM doses of 10-60 µg/kg (Raekallio et al. 1991a; Wagner et al. 1991; England et al. 1992). The decrease of HR is reported also in horses with elevated basal heart rate caused by abdominal pain (Jöchle 1989; Jöchle et al. 1989).

Although some studies suggest that detomidine decreases ventricular contractility in horses (Sarazan et al. 1989; Wagner et al. 1991), the most important changes in cardiac function are decreased HR, alterations

14 15 Review of the literature in blood pressure, increased afterload and, as a result, decreased cardiac output (CO) (Sarazan et al. 1989; Gasthuys et al. 1990; Wagner et al. 1991; Yamashita et al. 2000). Stroke volume is reported either to be unaffected (Wagner et al. 1991) or to decrease (Yamashita et al. 2000) after detomidine. The effects of detomidine and romifidine on cardiac function of horses are reflected also as changes in echocardiographic measurements (Buhl et al. 2007).

In several studies, detomidine and romifidine decreased respiratory rate and arterial partial pressure of oxygen (PaO2) in conscious horses, but in some reports these effects are very short-lived (Reitemeyer et al. 1986; Clarke and Paton 1988; Kamerling et al. 1988a; Gasthuys et al. 1990; Clarke et al. 1991; Wagner et al. 1991; Yamashita et al. 2000; Nyman et al. 2009). There is also a contradictory report with no effect, although in that study the first sample was taken 15 minutes after medication so the effect could have already disappeared (Raekallio et al. 1990). Decreased respiratory rate after detomidine was observed also in horses suffering from abdominal pain (Jöchle 1989; Jöchle et al. 1989), but not in healthy horses after high doses (100-300 µg/kg, Vainio 1985). An indicator of hypoventilation, arterial partial pressure of carbon dioxide

(PaCO2), remained unchanged or slightly elevated after detomidine (Short et al. 1986a; 1986b; Clarke and Paton 1988; Raekallio et al. 1990; Wagner et al. 1991; Yahashita et al. 2000; Nyman et al. 2009), but increased after romifidine administration (Clarke et al. 1991). No reports demonstrating the other possible mechanisms of decreased PaO2, such as ventilation perfusion mismatch, pulmonary shunt, diffusion barrier or low mixed-venous O2, after administration of α2-adrenoceptor agonists to conscious horses can be found.

The magnitude of changes in PaO2 and PaCO2 were not clinically significant in these studies.

2.5.5 Gastrointestinal effects of detomidine and romifidine

Alphα2-adrenergic tone affects the motility and tone of the equine gastrointestinal tract (Lester et al. 1998), and α2-adrenoceptors with binding properties of both α2A and α2B receptors are found in smooth muscle of equine ileum (Re et al. 2001). Interestingly, the binding properties of ileal α2-adrenoceptors were different, indicating differences in distribution of these receptors, in male and female horses (Re et al. 2001). De- tomidine and romifidine are reported to cause a rapid and profound decrease in duodenal and large colon motility in horses (Roger and Ruckebusch 1987; Merrit et al. 1998; Freeman and England 2001; Elfenbein et al. 2009). The studies reporting an inhibitory effect of detomidine on duodenal motility indicate dose-dependency in its duration, lasting 50-60 minutes with the recommended doses of 10-20 µg/kg and up to 180 minutes with a very high dose of 100 µg/kg (Roger and Ruckebusch 1987; Merrit et al. 1998; Elfenbein et al. 2009). The same applies to romifidine, after which the inhibitory effect lasted significantly longer with a high dose of 120 µg/kg than with 80 µg/kg. The duration and magnitude of the effect differed between different parts of the gastrointestinal tract, and after a dose of 120 µg/kg some inhibitory effect on motility of the small intestine and large colon was still detectable after 120 minutes, but not after 180 minutes from drug administration (Freeman and England 2001). The effect was also noted usingin vitro equine duodenal preparations, which proved that the decrease in motility is mediated by a peripheral mechanism (Zullian et al. 2011). Although the exact location of the site of action remains unknown, it might be related to stimulation of α2A receptors located in the enteric nervous system, which may inhibit the release of acetylcholine from the Auerbach plexus (Blandizzi et al, 1991). Detomidine induced relaxation of the large colon accompanied by signs of pain relief in an experimental colic model of horses (Roger and Ruckebusch 1987).

2.5.6 Effects of detomidine and romifidine on selected neurohormonal variables and changes related to energy metabolism

In horses, blood glucose increases in a dose-dependent fashion after IV detomidine administration. Hyper- glycaemia has also been reported after IM detomidine and after IV romifidine injection (Short et al. 1986a;

16 17 Review of the literature

Gasthyus et al. 1987; Carroll et al. 1997; Luna et al. 2011; Kullmann et al. 2014). Moreover, an increase in blood glucose was noted in horses under general anaesthesia when detomidine was used as a premedication

(Short et al. 1986a). The hyperglycaemic effect of 2α -adrenoceptor agonists is related to their action on pancreatic β-cells, leading to inhibition of insulin secretion and secondary hyperglycaemia (Angel et al. 1988). Administration of detomidine has also resulted in a significant decrease in serum FFA (Carroll et al. 1997). The findings considering the effect of detomidine on serum cortisol concentration are conflicting; no effect (Raekallio et al. 1991b) and decreasing effect (Raekallio et al. 1992; Carroll et al. 1997) are reported in research settings. In addition, detomidine decreased serum cortisol concentration in clinical equine patients; the effect might be explained by high baseline cortisol concentrations of these horses (Raekallio et al. 1992).

2.6 Alphα2-adrenoceptor antagonists in horses

Among the α2-adrenoceptor antagonists, atipamezole (previously known as MPV-1248), yohimbine and

tolazoline are the most extensively studied in horses. They act on both central and peripheral α2-adreno-

ceptors and can be used for partial or full reversal of the sedative effect of 2α -adrenoceptor agonists (Nils- fors and Kvart 1986; Raekallio et al. 1990; Kollias-Baker et al. 1993; Ramseyer et al. 1998; Paddleford

and Harvey 1999; Hubbell and Muir 2006). Administration of α2-adrenoceptor antagonists has resulted in side-effects such as re-sedation, increased responsiveness to external stimuli and even excitation (Ramseyer et al. 1998; Dimaio Knych et al. 2011a; 2011b). In addition to the behavioural effects, administration of yohimbine alone resulted in a transient increase in HR and prevention of AV conduction disturbances present in some non-medicated horses (Dimaio Knych et al. 2011b), although the effect of yohimbine on HR was not detected in another study (Dimaio Knych et al. 2011a). When administered with xylazine or detomi-

dine, α2-adrenoceptor antagonist drugs (atipamezole, yohimbine and tolazoline) partially prevented nega-

tive chronotropism and AV conduction disturbances induced by the α2-adrenoceptor agonists (Nilsfors and Kvart 1986; Raekallio et al. 1990; Kollias-Baker et al. 1993; Ramseyer et al. 1998). In addition, tolazoline increased MAP significantly in xylazine-treated horses (Kollias-Baker et al. 1993). In detomidine-treated horses, in turn, MAP remained unchanged after tolazoline (Carroll et al. 1997) and increased after atipamezole (Nilsfors and Kvart 1986).

Atipamezole and yohimbine are reported to prevent the α2-adrenoceptor agonist-induced decrease in motility of equine duodenal preparations in vitro (Zullian et al. 2001). When administered alone to conscious horses, yohimbine increased the myoelectric activity of several regions of the gastrointestinal tract (Lester et al. 1998), increased gastrointestinal sounds and resulted in a high frequency of defecation (Dimaio Knych et al. 2011a; 2011b), although neither atipamezole nor yohimbine had any effect on motility of duodenal preparations in vitro (Zullian et al. 2001). In addition to the effects on normal gastrointestinal function, yohimbine restored blood flow, contractions of the caecum and colon (Eades and Moore 1993) and gastric emptying compromised by endotoxin infusion in horses (Meisler et al. 2000). Yohimbine also alleviated experimentally induced postoperative ileus in horses (Gerring and Hunt 1986).

Yohimbine is reported to have no effect on blood glucose concentration in horses (Dimaio Knych et al. 2011b), and atipamezole had no effect on detomidine-induced hyperglycaemia (Luna et al. 2011). In contrast, administration of tolazoline after IV detomidine resulted in a significant increase in blood glucose concentration (Carroll et al. 1997). Other detected effects of tolazoline after detomidine treatment were increases in serum cortisol and FFA concentrations (Carroll et al. 1997).

16 17 Review of the literature

2.7 Vatinoxan

Vatinoxan (IUPAC N-[2-[2R,12bS]-2’-oxospiro[1,3,4,6,7,12b-hexahydro-[1]benzofuro[2,3-a]quino- lizine-2,5’-imidazolidine]-1’-yl]ethyl]methanesulfonamide), previously known as L-659,066 and MK-476, is an α2-adrenoceptor antagonist that belongs to spirocyclic-substituted benzofuroquinolizines (Clineschmidt et al. 1988) and contains an imidazoline group in its structure. It has low lipophilicity, resulting in poor pen- etration of the blood-brain barrier, and an ability to selectively block α2-adrenoceptors in peripheral sites

(Clineschmidt et al. 1988). This feature distinguishes vatinoxan from other α2-adrenoceptor antagonists, which act also on central α2-adrenoceptors. An α2: α1 selectivity of 105 is reported for vatinoxan (Clineschmidt et al.1988). In horse and sheep, no effect of vatinoxan was detected on cardiovascular response to methoxamine, an α1-adrenoceptor agonist, suggesting that it has no clinically relevant effect on 1α -adrenoceptors (Bryant et al. 1998).

2.7.1 Pharmacokinetics of vatinoxan

Plasma concentrations and certain pharmacokinetic parameters of vatinoxan were first reported by Honka- vaara and co-workers (2012) in dogs, although they measured the plasma concentration only for 60 minutes, which may cause some inaccuracy in the calculations. The same applies to the results of Bennet and co-workers (2016), which were calculated after a sampling period of 120 min. Reports in cats were based on a much longer follow-up time (8 hours), enabling more reliable calculations of pharmacokinetic parameters (Pypendop et al. 2016; 2017). A summary of the published pharmacokinetic data on vatinoxan is presented in Table 2. Dexmedetomidine did not have a significant effect on the plasma concentrations of vatinoxan in dogs or cats (Honkavaara et al. 2012; Pypendop et al. 2016; 2017).

Table 2. Summary of basic pharmacokinetic parameters of vatinoxan presented as mean ± SD or median (range) depending on how they appear in the original reports.

Species Dose Route Vd Cl T½ß Reference (µg/kg) (l/kg) (ml/kg/min) (min)

Dog 250 IV 0.41 7.8 39 Honkavaara et al. 2012 ± 0.13 ± 3.4 ± 7.6

Cat 300 IV 0.491 3 122 Pypendop et al. 2016 (0.379–0.604) (2–4.5) (99–139)

Dog 250 IV 1.0 10.6 65.6 Bennett et al. 2016 ± 0.9 ± 6.9 ± 20.1

Cat 600 IV 0.558 4.5 97 Pypendop et al. 2017 (0.501-0.605) (3.8-6.3) (69-105)

Cat 600 IM 0.607 5.9 NA Pypendop et al. 2017 (0.388-0.900)* (2.8-8.5)*

* V/F and Cl/F for IM treatment, reported F (%) was 99.5 (52.6-148.7) NA, not applicable or not reported, F, bioavailability

18 19 Review of the literature

2.7.2 Cardiovascular effects of vatinoxan

Vatinoxan, administered as a sole agent or as a premedication before other medications, is reported to have a variety of effects on cardiovascular function in various species. In dogs, vatinoxan alone increased HR, enhanced cardiac performance measured by CO and cardiac index (CI) (Pagel et al. 1998; Enouri et al. 2008; Honkavaara et al. 2011) and decreased SVR (Pagel et al. 1998; Enouri et al. 2008). The latter effect was reported to be dose-dependent (Pagel et al. 1998). In addition, a dose-dependent increase of HR and decrease of MAP was reported in rats (Szemeredi et al. 1989). In cats, only a transient increasing trend of HR was described (Honkavaara et al. 2017). In dogs, MAP, left ventricular pressure and coronary vascular resistance remained unchanged after administration of vatinoxan (Pagel et al. 1998). In methoxamine-pretreated horse and sheep, in turn, vatinoxan had no effect on HR (Bryant et al. 1998). In that study, MAP increased immediately after IV administration in sheep, but no effect on blood pressure was detected in horses. In human volunteers, vatinoxan caused a slight increase in blood pressure during the first hour after oral administration, but HR remained unchanged (Warren et al. 1991). In hypovolaemic rats, HR and central venous pressure (CVP) increased, but MAP remained unchanged after administration of vatinoxan (Tiniakov and Scrogin 2006).

2.7.3 Metabolic effects of vatinoxan

Initially, research of vatinoxan was focused on its effects on plasma glucose and insulin concentrations. Vatinoxan induced a transient increase in plasma glucose and insulin concentrations in fasted mice (Gold- man et al. 1989), but did not affect plasma glucose concentration in human volunteers (Warren et al. 1991; Schafers et al. 1992). Furthermore, vatinoxan enhanced insulin response to IV glucose administration in mice (Goldman et al. 1991), but not in humans (Schafers et al. 1992). Insulin response to exercise in humans was also enhanced by vatinoxan (Schiberras et al. 1994), and the effects in mice and exercised humans were dose-dependent (Goldman et al. 1989; Schiberras et al. 1994). Vatinoxan dose-dependently increased resting level, further enhanced the exercise-induced increase in plasma FFA concentration in humans (Schiberras et al. 1994) and increased the plasma concentration of noradrenaline and its metabolites in rats (Szemeredi et al. 1989).

2.7.4 Other effects of vatinoxan

There are minor side-effects reported in conjunction with administration of vatinoxan.At high doses (ex- ceeding a total of 12 mg), vatinoxan is reported to cause nausea and abdominal discomfort with restless- ness and lightheadedness in humans (Schafers et al. 1992). Some dogs showed excessive salivation almost immediately after administration of vatinoxan (250 µg/kg), but the salivation lasted only a few minutes (Honkavaara et al. 2011). In mice, a high dose (30 mg/kg) of vatinoxan had a hypothermic effect, which was not seen after administration of lower doses (3 and 10 mg/kg) (Durcan et al. 1991). Vatinoxan also caused an antidiuretic effect in rats after oral water-load. The effect was not observed in normally hydrated rats (Jackson et al. 1992).

2.8 Simultaneous use of vatinoxan and α2-adrenoceptor agonists

2.8.1 Vatinoxan and cardiovascular effects of 2α -adrenoceptor agonists

Currently, studies have focused on the co-administration of vatinoxan with α2-adrenoceptor agonists. Orig- inally, intravenous vatinoxan was discovered to counteract the changes in haemodynamic function (arising from increased afterload caused by vasoconstriction, which impairs systolic function and leads to reflex bradycardia and decreased CO) induced by dexmedetomidine in dogs (Pagel et al. 1998). Studies have reported

18 19 Review of the literature the same in dogs treated with (dex)medetomidine (Enouri et al. 2008; Honkavaara et al. 2008; 2011; Rolfe et al. 2012). The interaction of vatinoxan and dexmedetomidine on haemodynamic function was described to be dose-dependent, and a dose ratio of 50:1 (vatinoxan:dexmedetomidine) resulted in the most stable cardiovascular performance in dogs (Honkavaara et al. 2011). Results showing that tissue oxygen delivery

(DO2) was preserved (Honkavaara et al. 2011) and plasma lactate concentrations were lower (Restitutti, 2012) when vatinoxan was administered with dexmedetomidine provide further evidence that global tissue oxygenation is better preserved when vatinoxan is given in conjunction with α2-adrenoceptor agonists.

Haemodynamic function after vatinoxan and α2-adrenoceptor agonists has also been studied in other species; preservation of cardiovascular performance by vatinoxan was reported in dexmedetomidine-treated sheep (Raekallio et al. 2010) and medetomidine-treated sheep and horses pretreated with methoxamine (Bryant et al. 1998). Recently, dose-dependent prevention of negative chronotropism induced by IV dexmedetomidine has been demonstrated also in cats (Honkavaara et al. 2017).

After IM administration, vatinoxan attenuated the cardiovascular effects of IM medetomidine in dogs, and with the doses used in that study the dose-dependency was seen only in duration, not in magnitude of the action of vatinoxan (Restitutti et al. 2017). In another report, vatinoxan attenuated the effects of IM medetomidine in a manner comparable to IV administration (Rolfe et al. 2012).

2.8.2 Effect of vatinoxan on drug plasma concentrations and sedation

The interaction between α2-adrenoceptor agonist and antagonist agents on the pharmacokinetics of each drug was reported in the work of Salonen and colleagues (1995), which showed that atipamezole decreased

AUC and T½ß and increased Cl of medetomidine. The same interaction seems to appear between vatinoxan and α2-adrenoceptor agonists, as vatinoxan markedly changes the early disposition of medetomidine and dexmedetomidine administered intravenously to dogs (Honkavaara et al. 2012; Bennett et al. 2016) and cats (Pypendop et al. 2016), resulting in significantly lower AUC of dexmedetomidine. This phenomenon is explained by enhanced cardiac function and peripheral blood flow due to lower peripheral vascular resistance after combined administration of dexmedetomidine and vatinoxan compared with dexmedetomidine alone.

When the medications are given intravenously, vatinoxan is reported to have no effect or to cause a slight, but not clinically relevant, reduction in the sedation score or in the duration of sedative effect of α2-adrenoceptor agonist in dogs (Honkavaara et al. 2008; Restitutti et al. 2011; Rolfe et al. 2012; Bennett et al. 2016), cats (Honkavaara et al. 2017), humans (Warren et al. 1991) and sheep (Raekallio et al. 2010). The minor effect is most likely related to the lower plasma concentration of the agonists when they are co-administered with vatinoxan (Honkavaara et al. 2012; Bennett et al. 2016). Bennett and co-workers (2016) reported that medetomidine-induced antinociception in dogs was reduced by co-administration of vatinoxan. The reduction was also most likely related to lower plasma concentrations of medetomidine caused by vatinoxan.

After IM administration to dogs (Restitutti et al. 2017) and cats (Pypendop et al. 2017), vatinoxan facilitated absorption of IM medetomidine and dexmedetomidine, resulting in higher Cmax, which was reached more quickly than with treatment with (dex)medetomidine alone. Increased Cl of dexmedetomidine was reported in cats after combined IM administration with vatinoxan (Pypendop et al. 2017). The sedative action of dexmedetomidine in dogs followed the pattern of plasma concentrations; the sedative effect appeared faster, it was shorter in duration and the maximum sedation score was higher when IM medetomidine was combined with vatinoxan (Restitutti et al. 2017).

20 21 Review of the literature

2.8.3 Influence of vatinoxan on metabolic and other effects

of α2-adrenoceptor agonists

Vatinoxan antagonized hyperglycaemia and hypoinsulinaemia induced by clonidine in humans (Warren et al. 1991) and dexmedetomidine in dogs (Restitutti et al. 2012). Of these, the effect in humans was reported to be dose-dependent. Vatinoxan antagonized clonidine-induced decrease in blood noradrenaline concentration in humans (Warren et al. 1991). Vatinoxan had a delayed effect on the decrease of serum FFA concentration caused by dexmedetomidine in dogs (Restitutti et al. 2012). The FFA concentration initially decreased after administration of dexmedetomidine alone and when it was combined with vatinoxan. In the latter group, the FFA concentration started to increase again after 35 minutes, while in the former group the concentration slowly continued to decrease (Restitutti et al. 2012).

Clonidine inhibited colonic propulsion in mice, which was expressed as slower expulsion of a glass bead from the colon. This effect was also antagonized by vatinoxan (Clineschmidt et al. 1988).

20 21 Aims of the study

3. AIMS OF THE STUDY

1) To evaluate the effect of vatinoxan on clinical depth of sedation induced by detomidine and romifidine in horses (Studies I and II).

2) To investigate the effects of vatinoxan on selected cardiovascular variables of conscious horses treated with detomidine (HR and CVP) or romifidine (HR and ABP) (Studies I and II).

3) To investigate the effects of vatinoxan on the haemodynamic function (CO/CI, DO2, ABP and SVR) of horses that received detomidine as premedication before inhalation (isoflurane) anaesthesia (Study III).

4) To characterize the effect of vatinoxan on plasma concentrations of detomidine and romifidine when administered simultaneously (Studies I, II and III).

5) To compare the effects of detomidine and romifidine alone with the effects of their combination with vatinoxan on intestinal motility of horses (Studies I and II).

6) To examine the effects of romifidine, vatinoxan and their combination on selected neurohormonal variables and changes related to energy metabolism in horses (Study IV).

22 23 Materials and methods

4. MATERIALS AND METHODS

4.1 Animals

A total of 16 horses (9 Finnhorses and 7 Standardbreds) were used to conduct a series of four trials. Studies II and IV were performed simultaneously using the same horses. For studies that did not involve general anaesthesia (Studies I, II and IV), some animals were used in more than one trial (Table 3). A total of 6 Finnhorses participated in Study I, 7 Finnhorses in Studies II and IV and 7 Standardbreds in Study III. None of the mares were pregnant, and the horses were considered healthy based on clinical examination, including inspection of mucous membranes, capillary refill time, auscultation of heart, lungs and intestinal borborygmi and measurement of rectal temperature. In addition, for horses in Study III, routine haematology and serum biochemistry analyses were performed before the beginning of the trial. The studies were approved by the National Exper- imental Board of Finland (license numbers ESAVI-2010-06403/Ym-23 (I), ESAVI/1130/04.10.03/2011 (III), ESAVI/7608/04.10.03/2012 (II, IV)).

Table 3. Data on the horses participating in studies I-IV. The experiments were performed over a three-year period; the age of the horses is the age at the time of the first study in which the horse participated and the weight is a mean of all measurements of the same horse.

Horse Race Gender Age Weight Study (years) (kg) 1 FH Mare 19 581 I 2 FH Mare 17 580 I, II, IV 3 FH Mare 16 587 I 4 FH Mare 16 612 I, II, IV 5 FH Mare 12 585 I, II, IV 6 FH Mare 7 560 I, II, IV 7 STB Mare 8 550 III 8 STB Stallion 4 411 III 9 STB Mare 25 450 III 10 STB Gelding 6 520 III 11 STB Mare 6 475 III 12 STB Gelding 4 465 III 13 STB Mare 20 409 III 14 FH Mare 7 536 II, IV 15 FH Mare 19 662 II, IV 16 FH Mare 16 615 II, IV

FH, Finnhorse; STB, Standardbred

Prior to the trials, horses were fed hay and haylage, except for animals enrolled in Study III, which received only hay. All horses had free access to water before and after the trials. Before the induction of general anaesthesia in Study III, food was withheld for 12 hours. The horses in Study III were euthanized at the end

22 23 Materials and methods of the trial to enable tissue samples to be collected for further research. Horses were euthanized by an IV injection of T 61 (T 61 vet; Intervet, the Netherlands) without letting them recover from the second general anaesthesia.

4.2 Instrumentation

All of the catheters were placed after clipping and scrubbing the skin and under infiltration of local anaesthetic. A venous catheter was placed in the jugular vein for administration of drugs, and a central venous catheter was placed in the cranial vena cava via the opposite jugular vein. In Studies I, II and IV, an additional jugular catheter was placed in the jugular vein cranial to the first jugular catheter; this catheter was used for blood sampling. In these studies, the catheter used for administration of treatments was removed immediately after drug injection. In Study III, venous blood samples were drawn from the central venous catheter.

For Studies II and IV, an arterial catheter was placed in the transverse facial artery. In Study III, an arterial catheter was placed in the facial artery under general anaesthesia. All catheters that were left in place for the monitoring period of the trial were sutured to adjacent skin. Arterial catheters were used for arterial blood sampling for blood gas measurements, monitoring of blood pressure and measurements of CO with lithium dilution technique. All catheters were removed at the end of each follow-up period.

4.3. Treatments

In all studies, horses received α2-adrenoceptors agonist and vatinoxan; the drug doses and combinations are summarized in Table 4.

Table 4. Doses and combinations of detomidine (DET), romifidine (ROM) and vatinoxan (V) given to horses during Studies I-IV.

Dose of DET/ROM Dose of vatinoxan Study α -adrenoceptor agonist Treatment 2 (µg/kg) (µg/kg)

DET I DET 10 250 DET+V

DET III DET 20 200 DET+V

ROM II, IV ROM 80 200 ROM+V V

24 25 Materials and methods

Study III was performed under general anaesthesia, and horses underwent arthroscopy for the simultaneously conducted orthopaedic study. Before maintenance of inhalation anaesthesia with isoflurane (Isoba; Schering-Plough Limited, UK) vaporized in oxygen, all drugs were administered intravenously according to the following protocol:

- Benzylpenicillin sodium 40 000 IU/kg (Geepenil 24.0 g; Orion Pharma, Finland)

o at least 10 minutes before any other medication - Butorphanol 20 µg/kg (Butordol, Intervet, the Netherlands)

o at T-15 (15 minutes before induction of anaesthesia) - Detomidine 20 µg/kg (Equisedan, Vetcare, Finland), either alone or with 200 µg/kg Vatinoxan (Merck & Co., Inc., PA, USA)

o at T-10 (10 minutes before induction of anaesthesia) - Ketamine 2.2 mg/kg (Ketaminol; Intervet, the Netherlands) and midazolam 0.06 mg/kg (Midazolam Hameln; Hameln Pharmaceuticals, Germany)

o at T0 (induction of general anaesthesia) The time from administration of induction agents to recumbency was recorded. Under general anaesthesia, the lungs of horses were ventilated mechanically. Target end-tidal isoflurane concentration was 1.3% to maintain an appropriate level of anaesthesia. If MAP decreased below 70 mmHg, dobutamine 0.25 µg/kg/ minute (Dobuject; Primex Pharmaceuticals, Finland) was infused, and the rate was doubled if needed every 5 minutes until 70 mmHg was reached. The total amount of dobutamine (mg/kg) required for maintaining MAP > 70 mmHg was recorded.

4.4 Study designs

Study I had a randomized, crossover design with a minimum of 14 days between treatments. The investi- gator assessing the sedation and intestinal scores was unaware of the treatment. The same applies to Stud- ies II and IV, except that the washout period between treatments was a minimum of 6 days. Study III had a crossover design with a washout period of 14 days, starting alternately with DET or DET + V. Investiga- tors monitoring cardiovascular status of the horses and the entire staff in StudyIII were not blinded to the treatments. The personnel and laboratories analyzing the samples of serum and plasma biochemistry for Study IV were unaware of the treatments.

4.5 Clinical assessment of sedation and general anaesthesia (I, II, III)

The sedative effect of the drugs was scored in StudiesI and II. Attitude, posture, height of head, eyelid aper- ture and movement of ears were scored following a scoring system described by Rohrbach and co-workers (2009) (see Appendix 1). In Study III, sufficiency of sedation was evaluated according to posture of head and responses to external stimuli, but a pre-defined scoring system was not used. Clinical depth of general anaesthesia in Study III was assessed by observing the position and movements of the eye and the magnitude of the blinking reflex to touch of the eyelids and muscle relaxation of the horse.

24 25 Materials and methods

4.6 Analyses of drug plasma concentrations (I, II, III)

Venous blood samples for drug plasma concentration analyses were collected at pre-defined time-points from the jugular catheter (Studies I and II) or the central venous catheter (Study III) to EDTA tubes. Plasma was separated by centrifugation and stored at -20°C until analysed. The concentrations of drugs in plasma were analysed with high-performance liquid chromatography-tandem mass spectrometry (API 4000, AB Sciex, MA, USA). Detomidine, romifidine, butorphanol and vatinoxan were analysed after solid-phase extraction with Sep-Pak tC18 96-well extraction plates (Waters Co., MA, USA). Concentrations of midazolam were determined after liquid-liquid extraction with di-isopropylether containing 2% isopropanol and ketamine after protein precipitation with methanol. Internal standards for the analyses were diphenylim- idazole (Aldrich, Sigma-Aldrich Co., UK) for detomidine, RS-79948 (Tocris Biosciences, R&D Systems, MN, USA) for vatinoxan, buprenorphine (Schering-Plough Europe, Belgium) for butorphanol, nitrazepam (Roche, Switzerland) for midazolam and dexmedetomidine (Tocris) for ketamine. Concentration of romifidine was determined without an internal standard. More detailed descriptions of the analysis methods are available in the original manuscripts (I, II, III).

4.7 Pharmacokinetic calculations (I, II, III)

Pharmacokinetic variables were calculated from the plasma detomidine, romifidine and vatinoxan concentration time data with the WinNonlin Professional software package, version 5.3 (Pharsight Corporation, CA, USA), using non-compartmental methods (I, II). In Study III, only mean plasma concentrations and AUC of the drugs were compared. Area under time concentration curves was calculated with the trapezoidal method.

4.8 Cardiorespiratory monitoring (I, II, III)

Heart rate and blood pressures, specifically HR and CVP in Study I, and HR and ABP in Study II, were recorded with a multi-parameter monitor (S⁄5 Compact Critical Care Monitor, Datex-Ohmeda, UK). An anaesthesia monitor (Tafonius, Hallowell EMC, Pittsfield, MA, USA) was used in Study III for HR and ABP monitoring. When monitors were not connected or HR was too low for measurement by the monitor, HR was counted from auscultation of the left thorax. A zero reference point for blood pressure measurements was set at the level of the shoulder in standing horses (Studies I and II) and at the level of Manubrium sterni in horses in dorsal recumbency (Study III). In Study I, CVP was measured as changes from baseline of each individual horse to exclude the possible effect of variation in catheter tip placement.

Cardiac output was measured by the lithium indicator dilution method (LiDCO Plus Haemodynamic Moni- tor, LiDCO Ltd., UK) as described for horses by Hallowell and Corley (2005). A dose of 2.25 mmol of LiCl was used for each measurement. Standard values of 10 g/l haemoglobin and 140 mmol/l of sodium were used and later corrected with values obtained from simultaneously collected arterial blood samples (III). Corrections are needed for accurate measurements of lithium concentration. Haemoglobin concentration is used to calculate the fraction of blood (red cells) that will not contain lithium. Sodium concentration, in turn, is needed for calculations because the lithium sensor measures both lithium and sodium, and the amount of sodium needs to be excluded to obtain the lithium concentration.

In Studies II and III, a base apex electrocardiogram (ECG) was recorded continuously (S ⁄ 5 Compact Crit- ical Care Monitor) during experiments, and heart rhythm was assessed retrospectively from the recordings.

26 27 Materials and methods

Respiratory rate was recorded by counting thorax movements (II). In Study III, end-tidal partial pressure of

carbon dioxide (PeCO2) of horses was monitored with Tafonius and inspired oxygen fraction (FiO2) with S⁄5 Compact Critical Care Monitor. Samples for blood gas measurements in Study III were analysed with ABL 800 basic (Radiometer Medical ApS, Denmark) and in Study II with IDEXX VetStat (ME, USA). Calculated haemodynamic variables are presented in Table 5.

Table 5. Equations used for calculations of cardiovascular parameters.

Variable Equation

CI (l/kg/min) CO (l/min)/BWT (kg)

SVR (dynes*sec*cm-5) 80*[MAP (mmHg)-CVP (mmHg)]/CO

Ratio of arterial oxygen tension to inspired oxy- PaO2 (mmHg) /FiO2 gen

DO2 (l/min) [CaO2 (ml/l)*CO (l/min)]/ 1000

CaO2 = (Hb (g/l)*1.39*SaO2)+ (PaO2(kPa)*0.225)

BWT, Body weight; CaO2, arterial oxygen content; SaO2, arterial oxygen saturation

4.9 Assessment of intestinal borborygmi (I, II)

Intestinal sounds in Studies I and II were scored as described by Mama et al. (2009). Left and right dorsal and ventral flank areas were auscultated and number of borborygmi counted over 30 seconds from each quadrant; a score of 0.5 was used when uncoordinated rumbling or gaseous sounds were present. A single borborygmi score was achieved by summation of the scores from each quadrant.

4.10 Biochemistry analyses from serum and plasma (IV)

In Study IV, selected biochemistry analyses were run from serum and plasma samples. Glucose was analysed with the photometric glucose hexokinase 2-reagent method (Konelab™ Glucose HK, Thermo Fish- er Scientific Ltd., Vantaa, Finland). The enzymatic colorimetric method was used for the determination of FFAs (NEFA-C, Wako Chemicals GmbH, Neuss, Germany; KONE Pro Selective Chemistry Analyzer, Thermo Fisher Scientific Ltd.). Adrenocorticotropic hormone (ACTH) was analysed with a solid-phase, two-site sequential chemiluminescent immunometric assay (Immulite 2000, Siemens Healthcare Diagnos- tics Products GmbH, Marburg, Germany). Serum cortisol concentration was analysed by radioimmunoassay (RIA, Spectria cortisol RIA kit, Orion Diagnostica Ltd., Espoo, Finland). Blood potassium (K+) and sodium (Na+) concentrations were analysed by a blood gas analyzer (IDEXX VetStat; ME, USA).

Serum insulin was initially analysed with a solid-phase, enzyme-labelled chemiluminescent immunometric assay (Immulite 2000) because the RIA kits for insulin were not available at the time of the trial. The plasma and serum samples of horses were stored at -20°C after the initial analyses and four years later, when the RIA kits were commercially available again, insulin analyses of the plasma samples were repeated with the RIA method (Hu Ins Spec RIA 250TKit I-125<185 KBqs, EMD Millipore, MA, USA).

26 27 Materials and methods

4.11 Statistical analyses and power calculations

Power calculation was performed for Study I in which the number of the animals included (n = 6) was con- firmed to demonstrate an 80% power and 95% level of confidence for HR 10 beats/minute (25 versus 35) with standard deviation (SD) of six and for area under sedation time curve (AUCsed) 60 (300 versus 360) SD of 40, respectively.

The normality assumptions of the data in Studies I, II and IV were evaluated with Shapiro-Wilk tests. In Study III, normality was tested with the Kolmogorov–Smirnov test.

Repeated two-way analysis of variance was used for comparisons of HR and CVP (I); HR, CI, MAP, SVR, lactate and blood gas variables (III); glucose, FFA, ACTH, triglycerides, cortisol, lactate, K+ and Na+ (IV) between the treatments of the study in question. In Study III, a logarithmic transformation was applied to normalize the distribution of HR readings, and the model was fitted for the transformed response. Further- more, in Study II, linear mixed modelling for repeated measures was performed to test whether combined administration of vatinoxan and romifidine attenuated effects of romifidine. For each clinical variable, fixed effects were analysed to assess mean differences among the three treatments over the entire measurement period. Within each treatment, mean differences were analysed between each time interval and baseline in the same horse.

A post hoc t-test with Bonferroni correction was used to compare HR and CVP between treatments in Study I. In Studies II and IV, all significant values of the above-mentioned variables were adjusted by Bonferroni correction for multiple tests.

The medians of sedation and borborygmi scores (I, II) were tested for difference relative to baseline within treatments with Friedman’s two-way analysis of variance for dependent samples with Bonferroni correction, and the Kruskall-Wallis test was used for comparisons between treatments. Areas under the time curves for sedation and borborygmi scores between the treatments were analysed with Friedman’s two-way analysis of variance for dependent samples; Wilcoxon test for dependent samples was used for pairwise comparisons. Friedman’s two-way analysis of variance for dependent samples was used also for comparisons of peak sedation score between treatments.

Pairwise comparisons between treatments for pharmacokinetic parameters (I, II), drug concentrations in venous blood, induction times and amount of dobutamine administered (III) were performed with the paired Student’s t-test.

For serum insulin concentrations from analysis by chemiluminescent immunometric assay in Study IV, non-parametric test had to be used because the concentrations were below the detection limit in many samples. Friedman’s two-way analysis of variance for related samples with Bonferroni correction was used to compare concentrations to baseline within each treatment, and the Kruskall-Wallis test was used for comparisons of concentrations between treatments. The same statistical methods were used for additional insulin results from analysis by RIA.

For all comparisons, significance was set at p < 0.05.

28 29 Results

5. RESULTS 5.1 Clinical sedation (I, II)

Clinical sedation of horses was assessed in Studies I and II; high sedation score was achieved after both detomidine and romifidine administration. Sedation scores and effects of vatinoxan on sedation are summarized in Figure 1. No significant difference was detected in peak median sedation scores between detomidine and romifidine alone and combined with vatinoxan. Area under sedation time curve was, however, signifi-

cantly smaller after DET+V than after DET alone (I). No difference was detected in AUCsed with ROM and ROM+V (II).

Figure 1. Median sedation scores, as real gained scores without corrections, of horses after IV treatment with detomidine (10 µg/kg; DET), detomidine combined with vatinoxan (10 µg/kg + 250 µg/kg; DET+V), romifidine (80 µg/kg; ROM), vatinoxan (200 µg/kg; V) and romifidine combined with vatinoxan (80 µg/kg + 200 µg/kg; ROM+V). No significant difference was detected in peak median sedation scores between detomidine and romifidine alone and combined with vatinoxan. Area under sedation time curve was, however,

significantly smaller after DET+V than after DET alone. No difference was detected in AUCsed with ROM and ROM+V. Error bars indicate range.

5.2 Induction time (III)

Induction time, defined as time between the beginning of injection (midazolam and ketamine) and reach- ing lateral recumbency, was significantly faster after DET+V (59.2 ± 4.4 seconds) than after DET alone (89.6 ± 12.5 seconds).

28 29 Results

5.3 Plasma drug concentrations (I, II, III)

Plasma drug concentrations were analysed in Studies I, II and III. In all of these studies, DET+V and ROM+V resulted in lower plasma concentrations of these agents than either DET or ROM alone. The plasma concentration of DET and ROM alone and combined with vatinoxan are presented in Figure 2a and 2b.

Figure 2. Mean (± SD) plasma concentrations of (a) detomidine (DET, 10 µg/kg IV) and (b) romifidine (ROM, 80 µg/kg IV) in a semilogarithmic scale when administered with and without vatinoxan. Error bars indicate standard deviation. a)

30 31 Results

The plasma drug concentrations of Study III are presented in Table 6. Ketamine plasma concentration of one horse is not included in the data because the horse needed an additional dose of ketamine 0.2 mg/kg at T10 due to head and limb movements. Also, in this setting, vatinoxan decreased the plasma detomidine concentration. Plasma butorphanol concentrations were also lower after combined pretreatment with DET+V than after DET alone.

Table 6. Plasma drug concentrations (mean ± SD) and area under plasma concentration time curves (AUC) of seven horses premedicated at T-10 with detomidine (DET, 20 µg/kg IV) alone or combined with vatinoxan (DET+V, 200 µg/kg IV) during isoflurane-maintained anaesthesia. General anaesthesia was induced with midazolam and ketamine at T0, which are therefore not analysed at T-1. Ketamine plasma concentration of one horse is not included because of an extra dose of ketamine administered at T10. * Significant difference (p < 0.05) between DET and DET+V (III).

Drug Treatment T-1 T20 T40 T60 AUC (min ng/ml)

Vatinoxan (ng/ml) DET+V 205 135 101 84 7773 ± 29.7 ± 29.7 ± 11.7 ± 4.87 ± 806 Detomidine (ng/ml) DET 13.1* 6.39* 4.43* 3.41* 400* ± 2.37 ± 1.17 ± 0.78 ± 0.59 ± 53

DET+V 10.1* 4.68* 2.96* 2.46* 286* ± 2.03 ± 1.29 ± 0.47 ± 0.31 ± 40 Butorphanol (ng/ml) DET 37.8* 10.1 8.16 7.64 843* ± 6.91 ± 2.60 ± 2.08 ± 3.36 ± 171 DET+V 28.8* 7.77 7.01 6.16 663* ± 10.1 ± 3.01 ± 2.43 ± 2.16 ± 212 Midazolam (ng/ml) DET 42.7 32.7 32.9 756 ± 9.56 ± 9.84 ± 12.9 ± 218 DET+V 47.8 35.7 32.5 803 ± 10.0 ± 9.99 ± 10.2 ± 189 Ketamine (ng/ml) DET 768 480 350 11178 ± 162 ± 91 ± 58 ± 2160 DET+V 682 407 326 10080 ± 197 ± 99 ± 82 ± 2593

30 31 Results

Pharmacokinetic parameters, which are presented in Table 7, were calculated in Studies I and II. In Study II, the plasma concentration of vatinoxan in one horse at T240 was excluded because of an exceptionally high reading, possibly the result of contamination or a technical error. When detomidine and romifidine were administered in combination with vatinoxan, AUC of detomidine or romifidine decreased, and Vd and Cl, in turn, increased relative to administration of detomidine or romifidine alone. Furthermore, elimination half-life of romifidine increased after ROM+V. After DET and DET+V, the ½ßT of detomidine did not differ significantly between treatments.

Table 7. Pharmacokinetic parameters (mean ± SD) of IV detomidine (10 µg/kg; DET), romifidine (80 µg/ kg; ROM), administered with and without vatinoxan, and vatinoxan (200 and 250 µg/kg; V200, V250) in horses. * Significant difference (p < 0.05) between DET or ROM alone relative to administration combined with vatinoxan.

T½ß AUCinf Vd Cl (minutes) (hours ng/ml) (l/kg) (ml/min/kg)

DET 37.1 ± 4.1 567 ± 207* 1.0 ± 0.3* 19.4 ± 5.9* DET (with V250) 42.6 ± 6.5 289 ± 97* 2.4 ± 1.0* 38.2 ± 13.5* ROM 75.6 ± 3.0* 62.9 ± 7.0* 2.3 ± 0.2* 21.4± 2.4* ROM (with V200) 89.1 ± 12.8* 49.6 ± 4.8* 3.5 ± 0.5* 27.1 ± 2.7* V200 141 ± 28.6 569.3 ± 94.2 1.2 ± 0.1 6.0 ± 1.0 V200 (with ROM) 170 ± 82.7 650.1 ± 204.2 1.2 ± 0.2 5.5 ± 1.6 V250 (with DET) 138 ± 17 655.7 ± 93.2 1.3 ± 0.1 6.5 ± 0.9

AUCinf, AUC extrapolated to infinity

32 33 Results

5.4 Cardiovascular effects 5.4.1 Cardiovascular effects of detomidine, romifidine and vatinoxan in conscious horses (I, II)

Both detomidine and romifidine induced cardiovascular effects immediately after administration. Heart rate (Fig. 3) decreased significantly within minutes. Heart rate was significantly lower than at baseline after DET, ROM and ROM+V. After DET+V, there was no significant difference in HR compared with baseline. After V, HR increased significantly relative to baseline. Heart blocks were detected in all horses after both DET and ROM. No heart blocks were detected after DET+V. In contrast, second-degree AV-blocks were detected in horses after ROM+V, but they disappeared within three minutes. After DET, HR was lower than after DET+V. Furthermore, HR was significantly higher after V than after ROM or ROM+V; there was no significant difference in HR between ROM and ROM+V.

Figure 3. Mean HR of horses after IV treatment with detomidine (10 µg/kg; DET), detomidine combined with vatinoxan (10 µg/kg + 250 µg/kg; DET+V), romifidine (80 µg/kg; ROM), vatinoxan (200 µg/kg; V) and romifidine combined with vatinoxan (80 µg/kg + 200 µg/kg; ROM+V). Mean HR was significantly lower after DET than after DET+V, and after ROM and ROM+V than after V. There was no significant difference between ROM and ROM+V. After DET, ROM and ROM+V, HR decreased, and after V, HR increased significantly. Error bars indicate standard deviation.

32 33 Results

Romifidine increased ABP until 50 minutes after administration of the drug, and MAP was significantly higher after ROM than after ROM+V (Fig. 4). After ROM+V, MAP decreased for 40 minutes. Hypoten- sion (MAP < 60 mmHg) was not detected during the 90-minute cardiovascular monitoring. No differences were detected in MAP after V.

Figure 4. Mean MAP of horses after IV treatment with romifidine (80 µg/kg; ROM), vatinoxan (200 µg/kg; V) and romifidine combined with vatinoxan (80 µg/kg + 200 µg/kg; ROM+V). Mean arterial blood pressure increased significantly after ROM and decreased after ROM+V; it was significantly higher after ROM than after ROM+V. Error bars indicate standard deviation.

At the beginning of the monitoring period, DET induced a minor increase in CVP, which was attenuated by vatinoxan, leading to a significant difference between DET and DET+V. CVP was significantly reduced after both treatments.

5.4.2 Cardiovascular performance in horses under general anaesthesia (III)

Heart rate was higher in horses after IV premedication with DET+V relative to DET at T30, T40 and T45 (Fig. 5). Cardiac index was significantly higher at T40 after premedication with DET+V and SVR after DET at all measuring points. Tissue oxygen delivery was significantly higher after DET+V than DET at T40 and T60. Blood lactate concentration showed no significant differences compared with baseline or between treatments (Table 8).

34 35 Results

Figure 5. Mean ± HR of seven horses under general anaesthesia pretreated with detomidine (20 µg/kg; DET IV) alone or combined with vatinoxan (250 µg/kg; DET+V IV). Error bars indicate standard deviation. * Significant difference between treatments (p < 0.05).

Table 8. Mean ± SD CI, SVR, DO2 and blood lactate concentration of seven horses under general anaesthesia pretreated with detomidine (20 µg/kg; DET IV) alone or combined with vatinoxan (250 µg/kg; DET+V IV). * Significant difference (p < 0.05) between DET and DET+V.

Treatment T20 T40 T60

CI DET 0.048 ± 0.020 0.043 ± 0.009* 0.041 ± 0.007

(l/kg/min) DET+V 0.051 ± 0.008 0.065 ± 0.011* 0.054 ± 0.012

SVR DET 247 ± 72* 352 ± 111* 353 ± 85*

(dynes*sec*cm-5) DET+V 134 ± 19* 206 ± 61* 272 ± 89*

DO2 DET 4544 ± 2572 4295 ± 911 4061 ± 1305

(ml/min) DET+V 4661 ± 747 7301 ± 1582* 6441 ± 3466*

Lactate DET 0.84±0.19 0.90±0.18 0.91±0.21

(mmol/l) DET+V 1.07±0.37 1.16±0.31 1.13±0.24

34 35 Results

Mean arterial blood pressures of the horses in Study III are presented in Figure 6. At the beginning of the monitoring period, MAP was in the target range (> 70 mmHg) in horses pretreated with DET. In contrast, after DET+V, MAP readings were noticeably lower. Afterwards, MAP decreased in DET-pretreated horses and dobutamine was infused after both treatments to keep MAP over 70 mmHg. The horses needed a significantly larger total amount of dobutamine for blood pressure support after premedication with DET+V (32.4 ± 15.6 mg) than after DET (10.4 ± 10.2 mg).

Figure 6. Mean arterial blood pressure of seven horses under general anaesthesia pretreated with detomidine (20 µg/kg; DET IV) alone or combined with vatinoxan (250 µg/kg; DET+V IV). Error bars indicate standard deviation. * Significant difference (p < 0.05) between treatments (III).

5.5 Intestinal borborygmi (I, II)

Borborygmi scores are presented in Table 9. Both DET and ROM decreased intestinal motility, significantly from T10 to T50 after DET and from T30 to T120 after ROM. After DET+V or ROM+V, there were no significant differences in borborygmi scores compared with baseline.Vatinoxan had no effect on borborygmi scores, but three horses had signs of abdominal discomfort, such as kicking towards the abdomen, after administration of vatinoxan. The median AUC for borborygmi was significantly smaller after DET than after DET+V, and after ROM than after ROM+ V or V; the latter two were not significantly different from each other.

36 37 Results

Table 9. Median (range) intestinal borborygmi scores of the horses after IV detomidine 10 µg/kg alone (DET) and combined with vatinoxan 250 µg/kg (DET+V), romifidine 80 µg/kg alone (ROM) and combined with vatinoxan 200 µg/kg (ROM+V), and vatinoxan 200 µg/kg alone (V). Borborygmi score was significantly lower (p < 0.05) than baseline from T10 to T50 after DET and from T30 to T120 after ROM. abcSignificant difference between treatments with the same superscript.

Time (min) DET DET+V ROM ROM+ V V

Baseline 4.75 (2.5-7.5) 5.5 (4-9) 5.5 (3.5-7.5) 4.5 (3-7) 6 (4-9.5)

10 0 (0-2) 5.25 (1.5-7) 1.5 (0.5-4) 3.5 (1-7) 7 (3-9)

20 0.25 (0-1) 5 (2-8) NA NA NA

30 0.25 (0-1) 4 (2-7) 0 (0-2.5) 2.5 (0-7) 6.5 (4-8.5)

40 0.25 (0-1) 4.75 (2.5-6) NA NA NA

50 0.5 (0-1) 4.5 (2.5-7) NA NA NA

60 1 (0-2) 4 (2-6.5) 0 (0-5.5) 4 (1-7.5) 6 (3-8)

90 2 (1-3) 5 (3-9) 0.5 (0-2.5) 2 (0.5-7.5) 6 (4-7.5)

120 NA NA 0 (0-2) 2.5 (1-8) 6.5 (4.5-8)

180 NA NA 1 (1-7.5) 5.5 (3-7.5) 5.5 (3-8)

240 NA NA 2 (2-5.5) 4.5 (4-8.5) 6 (3.5-8)

AUC 425a 86a 500b,c 968b 1428c

NA = Not applicable, borborygmi was not scored at the time-point in question.

5.6 Effects of romifidine and vatinoxan on selected neurohormonal variables and changes related to energy metabolism (IV)

The effects of romifidine, vatinoxan and their combination on concentrations of plasma glucose and ACTH, serum insulin, FFA, cortisol and blood K+ are presented in Table 8. Romifidine induced hyperglycaemia in horses. Plasma glucose concentration increased throughout the monitoring period after ROM; the increase was significantly different relative to baseline at T60 and T120. After ROM+V, in turn, the glucose plasma concentration increased (significantly different from baseline at T30, T60 and T120), but to a lesser extent than after ROM. No effect on glucose plasma concentration was detected after vatinoxan.

Serum insulin concentration (results from chemiluminescent immunometric assay) decreased significantly after V; the decrease was significant at T30 and T120. There were no significant differences after ROM or ROM+V within treatments, and no significant differences were detected between the three treatments. The RIA results (Fig. 7) showed a decrease in plasma insulin concentrations after all treatments, and the decrease was statistically significant after ROM and ROM+V. Plasma insulin concentration was significantly lower after ROM than after V.

36 37 Results

Figure 7. Box-and-whisker plots of plasma insulin concentration of seven horses, measured by RIA. The horses received romifidine (ROM, 80 μg/kg IV), ROM with vatinoxan (ROM+V, 80 μg/kg + 200 μg/kg IV) and vatinoxan (V, 200 μg/kg IV) at T0. Each box represents the interquartile range, the central horizontal line is the median value and the whiskers represent the range. Plasma insulin concentrations were significantly lower (p < 0.05) than baseline at T30 and T60 after ROM and at T15 and T30 after ROM+V. *Significant difference (p < 0.05) between ROM and V.

Serum FFA concentration increased after all treatments, and it was the highest after V. The differences between treatments were significant between V and ROM and between V and ROM+V. The difference in FFA concentrations was significant compared with baseline after all treatments: at T30 after ROM, at T15 and T30 after ROM+V and at T30 and T60 after V.

The plasma ACTH concentration showed an increasing tendency after all treatments, but the increase was statistically different from baseline only after V at T15 and T30. No significant differences emerged in plasma ACTH concentrations between treatments. In addition, serum cortisol concentration increased after all treatments. Serum cortisol was significantly increased from baseline atT15 after all treatments and at T30 after V. Differences in serum cortisol concentrations between the treatments were not detected.

Blood K+ tended to decrease after all treatments, and it was significantly lower than baseline atT15 after ROM, at T15 and T60 after ROM+V and at T60 after V. Blood K+ was significantly higher after V than after ROM+V. Plasma triglyceride, lactate or blood Na+ concentrations were stable throughout the monitoring period, and therefore, differences relative to baseline within treatments or in comparisons between treatments were not detected (data not shown).

38 39 Results

Table 10. Median (highest concentration) serum insulin (results from chemiluminescent immunometric assay) and mean ± SD plasma glucose and ACTH, serum FFA, cortisol and blood K+ concentrations. The lowest insulin concentration after all treatments at all time-points was below the detection limit (<2 µIU/ ml). The horses received romifidine (ROM, 80 μg/kg IV), ROM with vatinoxan (ROM+V, 80 μg/kg + 200 μg/kg IV) and vatinoxan (V, 200 μg/kg IV) at T0. a,b, Significant difference between treatments with the same superscript at the time-point in question.

Analyte Treatment Baseline T15 T30 T60 T120 Glucose ROM 5.10 5.33 6.69b 8.87a 9.33a ± 0.48 ± 0.57 ± 0.82 ± 1.43 ± 1.43 (mmol/l) ROM+V 4.74 5.11a 5.79a 5.77a 5.73a ± 0.19 ± 0.33 ± 0.51 ± 0.55 ± 0.69 V 4.94 4.69a 4.33ab 4.37a 4.86a ± 0.34 ± 0.34 ± 0.22 ± 0.26 ± 0.40 Insulin ROM 6.0 <2 <2 <2 <2 (83) (49) (45) (32) (42) (µIU/ml) ROM+V 4.8 <2 <2 <2 <2 (58) (45) (62) (39) (59) V 7.8 3.2 <2 4.3 <2 (80) (74) (47) (58) (43) FFA ROM 0.10 0.26 0.28 0.27b 0.24 ± 0.01 ± 0.15 ± 0.17 ± 0.16 ± 0.20 (mmol/l) ROM+V 0.05 0.29 0.27a 0.24a 0.23a ± 0.01 ± 0.19 ± 0.14 ± 0.10 ± 0.09 V 0.09 0.50 0.55a 0.64ab 0.40a ± 0.07 ± 0.32 ± 0.31 ± 0.27 ± 0.11 ACTH ROM 16.8 32.3 25.0 18.7 44.7 ± 3.83 ± 22.4 ± 12.5 ± 7.66 ± 62.4 (pg/ml) ROM+V 19.4 28.5 21.6 21.4 19.1 ± 11.1 ± 12.4 ± 8.39 ± 13.2 ± 11.1 V 17.3 32.2 29.0 23.0 16.1 ± 6.53 ± 6.66 ± 5.71 ± 5.68 ± 5.68 Cortisol ROM 59.0 120 106 86.6 113 ± 13.9 ± 34.7 ± 37.6 ± 35.4 ± 51.7 (nmol/ml) ROM+V 51.9 107 84.1 76.5 66.3 ± 16.1 ± 24.7 ± 23.1 ± 29.0 ± 26.5 V 61.2 131 120 102 79.4 ± 14.7 ± 34.2 ± 30.4 ± 23.5 ± 32.7

Analyte Treatment Baseline T5 T15 T30 T60 K+ ROM 4.30 4.16 4.09 4.16 4.17 ± 0.26 ± 0.28 ± 0.28 ± 0.30 ± 0.27 (mmol/l) ROM+V 4.20 4.11 3.87a 3.91 3.84 ± 0.29 ± 0.40 ± 0.28 ± 0.25 ± 0.27 V 4.44 4.39 4.30 a 4.21 3.94 ± 0.36 ± 0.28 ± 0.31 ± 0.36 ± 0.31

38 39 Discussion

6. DISCUSSION

In conscious horses, vatinoxan showed some beneficial effects when it was administered in combination with α2-adrenoceptor agonists. Vatinoxan alleviated peripherally mediated effects of detomidine and romifidine, which include cardiovascular changes (such as decreased HR and increased ABP), hyperglycaemia and decreased intestinal motility. Prevention of intestinal hypomotility could improve patient safety by preventing post-operative ileus and colic in horses. At the same time, the sedative effect of detomidine and romifidine was preserved. As a premedication, vatinoxan decreased induction time, but induced marked hypotension during general anaesthesia at the dose tested. Drug plasma concentration results indicate that

α2-adrenoceptor agonists induce changes in drug disposition via altered cardiovascular function. Vatinoxan was able to alleviate these changes.

6.1 Clinical sedation and plasma drug concentrations

Studies I, II and III describe the effect of vatinoxan on clinical sedation after IV administration of α2-adrenoceptor agonists detomidine and romifidine in horses. Administration of either 10 µg/kg detomidine or 80 µg/kg romifidine induced a rapid-onset sedative effect in horses, and the magnitude of the effect seemed to be similar after both drugs. This finding challenges a previous report in which doses of 20 µg/kg detomidine and 80 µg/kg romifidine appeared similar in potency (England et al. 1992). Both Studies I and II indicate that vatinoxan does not prevent the sedative effect of detomidine and romifidine, but the efficacy is slightly diminished. The difference was detectable in AUCsed in Study I, but not in the peak sedative effect, indicating that vatinoxan affected the duration rather than the magnitude of sedation. This finding is in agreement with earlier investigations of IV administration of vatinoxan and dexmedetomidine in dogs (Honkavaara et al. 2008; Restitutti et al. 2011; Rolfe et al. 2012) and cats (Honkavaara et al. 2017), which report only a minor decrease caused by vatinoxan on α2-adrenoceptor agonist-induced sedation, and the sedative effect was considered bioequivalent between administration of α2-adrenoceptor agonists alone and combined with vatinoxan. Bennet and co-workers (2016), instead, reported a shorter duration of sedative action and a markedly reduced antinociceptive action of IV medetomidine when it was co-administered with vatinoxan to dogs, but this is likely due to the small dose of medetomidine in their study.

In Study III, when vatinoxan was used as a part of the premedication protocol before induction of general anaesthesia, investigators observed a slightly lighter sedation compared with premedication with detomidine alone, in agreement with the results of Studies I and II. This is, however, only an observation because the level of sedation was not scored for statistical comparison between the treatments, and the observation may be biased because the investigators monitoring sedation and induction were not blinded to the treatments of the horses. Despite this, the inductions were smooth on all occasions. The induction time was significantly reduced when vatinoxan was administered as a part of the premedication protocol (III).

The effects of vatinoxan on clinical sedation and anaesthesia induction time are explained by the effects of detomidine, romifidine and vatinoxan on cardiovascular function and plasma concentrations of the drugs administered simultaneously. Basically, vatinoxan alleviated vasoconstriction and bradycardia caused by detomidine and romifidine. This resulted in increased Vd and lower plasma concentrations of detomidine and romifidine via enhanced circulation and organ perfusion. In practice, this was manifested as a slightly lighter clinical sedation when vatinoxan was a part of the medication protocol. The effect of 2α -adrenoceptor agonists and antagonists on drug plasma concentrations is reported also in dogs; atipamezole administered after medetomidine decreased the AUC and increased Cl of medetomidine (Salonen et al. 1995). Furthermore, vatinoxan decreased AUC and increased Vd and Cl of simultaneously administered medetomidine (Bennett et al. 2016) and dexmedetomidine (Honkavaara et al. 2012) in dogs. These observations indicate that also

40 41 Discussion

the anaesthetic-sparing effect of the 2α -adrenoceptor agonists could be due to the altered cardiovascular sta-

tus and perfusion of the peripheral tissues, leading to decreased Vd and increased drug plasma concentrations

of the drugs administered in conjunction with α2-adrenoceptor agonists. This conclusion is in agreement with the previous finding in which dexmedetomidine dose-dependently modified its own pharmacokinetics via altered cardiovascular performance (Dutta et al. 2000). The slower circulation after administration of

α2-adrenoceptor agonists results in slower distribution of other drugs administered intravenously, manifested

as delayed anaesthesia induction time in Study III. The slower circulation also explains the finding that maxC of detomidine, when administered alone, seemed to be reached a few minutes after IV administration, whe-

reas with vatinoxan the Cmax of detomidine was detected already in the first blood sample (1 minute) after

administration. A similar delayed detection of Cmax of detomidine in horses has previously been reported after IV administration (Mama et al. 2009).

Romifidine did not alter the plasma concentrations of vatinoxan, which is explained by stable cardiovascular performance when vatinoxan was a part of the medication protocol. This result supports further the link between drug disposition and cardiovascular function.

6.2 Cardiovascular effects

6.2.1 Cardiovascular effects in conscious horses

In Study II, romifidine transiently increased MAP; this phenomenon is previously well-described after administration of detomidine and romifidine in horses (Sarazan et al. 1989; Clarke et al. 1991). Increased MAP results from vasoconstriction manifesting as increased SVR (Wagner et al. 1991; Pimenta et al. 2011), although detailed haemodynamic monitoring was not performed in Study II. When vatinoxan was administered with romifidine, the hypertensive effect was prevented and MAP decreased below the baseline va-

lue. Vatinoxan seemed to antagonize the vasoconstrictive effect of romifidine on2 α -adrenoceptors in blood vessels. Further reasons for decreased MAP could be romifidine-induced central sympatholysis and related bradycardia, which were only partially prevented in this study.

Detomidine and romifidine induced negative chronotropic effects of the same magnitude (Studies I and II).

Negative chronotropism after administration of α2-adrenoceptor agonists is caused by baroreceptor reflex to increased ABP (Sarazan et al. 1989; Gasthyus et al. 1990). Negative chronotropism can be considered an appropriate physiological response to hypertension caused by increased SVR because the hypertensive effect is intensified and prolonged when bradycardia is antagonized by anticholinergics (Pimenta et al. 2011).

Vatinoxan was able to prevent the negative chronotropic effect caused by detomidine and alleviated the same effect when administered in combination with romifidine (StudiesI and II). Alleviation of the negative chronotropic effect coincided with the prevention of the hypertensive effect of romifidine, which provides further evidence that negative chronotropism was due to baroreceptor reflex caused by increased ABP. In Study I, detomidine was administered at a moderately low dose of 10 µg/kg relative to romifidine 80 µg/ kg in Study II, as a detomidine dose of 20 µg/kg can be considered equipotent to romifidine 80 µg/kg (Eng-

land et al.1992). At the receptor level, the predominance of either α2-adrenoceptor agonistic or antagonistic actions depends on the concentration and receptor affinity of these competing substances in the synaptic cleft. This is manifested as dose-dependency of the interaction between agonists and antagonists, previously reported in several studies (Jöchle and Hamm 1986; Kameing et al. 1988b; Hamm et al. 1995). Therefore, the comparatively smaller dose of detomidine may explain the difference of the antagonistic action of detomidine and romifidine-induced bradycardia by vatinoxan noted in our studies. This explanation is supported by previous findings showing that detomidine might cause more intense negative chronotropism than equi- 40 41 Discussion

potent doses of other α2-adrenoceptor agonists (Wagner et al. 1991; Yamashita et al. 2000). Thus, providing that doses of these α2-adrenoceptor agonists were equipotent, the negative chronotropic effect would have been more intense after detomidine than after romifidine. By contrast, Raekallio and co-workers (1991a) administered 10 µg/kg and 20 µg/kg detomidine to horses and the magnitude of the decrease in HR was similar in their report. In conclusion, differences in the antagonism of the negative chronotropic effects of detomidine and romifidine by vatinoxan could have been related to the use of different doses of the α2-adrenoceptor agonists evaluated.

In Study I, administration of detomidine resulted in a slight increase in CVP, which was prevented by simultaneously administering vatinoxan. We assume that the increase in CVP in our study was also a result of vasoconstriction. The complete antagonism of bradycardia by vatinoxan in Study I suggested that peripheral vasoconstriction was probably present after detomidine alone, but not when it was administered with vatinoxan because the initial bradycardia results from increased blood pressure via baroreceptor reflex (Sarazan et al. 1989).

6.2.2 Cardiovascular effects in horses under general anaesthesia

When vatinoxan was administered as a part of the premedication protocol before general anaesthesia, the most remarkable effect was severe hypotension (III). A slight decrease of blood pressure by vatinoxan, administered in combination with α2-adrenoceptor agonist, was also seen in standing horses (II), and it has been reported before in dogs (Honkavaara et al. 2011). In addition, atipamezole has been described to cause a transient decrease in blood pressure in detomidine-treated horses (Raekallio et al. 1990). This suggests that α2-adrenoceptor antagonists, in general, decrease arterial blood pressure via vasodilatation, and that α2-adrenoceptor agonist premedication supports blood pressure during general anaesthesia. For example, dexmedetomidine-infusion increased ABP in horses under isoflurane anaesthesia (Risberg et al.

2016). This interplay of α2-adrenoceptor agonistic and antagonistic action is an indication of tonic α2-adrenoceptor-mediated control of blood pressure (Savola 1989). However, the blood pressure support is a result of vasoconstriction and could in theory lead to lower DO2, which was detected in Study III. In Study III, the hypotensive effect was probably amplified by a combination of vasodilatory effects of isoflurane and vatinoxan. The MAP of horses premedicated with vatinoxan was, indeed, so low that it had to be considered a significant risk for patient safety III( ).

The effect of detomidine and vatinoxan on HR was similar in standing horses (I) and horses under general anaesthesia (III), although the difference in HR between treatments was smaller in the anaesthetized horses. Hoisting the horse to the operating table and time required for instrumentation prevented cardiovascular monitoring during the first 15 minutes after induction, thus not allowing early cardiovascular changes to be detected (III).

6.3 Effects on intestinal borborygmi

Detomidine and romifidine caused a marked decrease in intestinal borborygmi, which is indicative of intestinal hypomotility (Studies I and II). This side-effect of α2-adrenoceptor agonists is well described in the literature (Merritt et al. 1998; Freeman and England 2001; Mama et al. 2009). This phenomenon is of par- ticular importance in horses because the species is prone to ileus and colic. In clinical practice, the problem is exaggerated after standing operations because the sedatives are used as repeated boluses or infusion, and furthermore, in conjunction with colic surgery, after which ileus is often a major concern and an important cause of mortality (Cohen et al. 2004). Simultaneous administration of vatinoxan with detomidine and romifidine significantly reduced the intestinal hypomotility.

42 43 Discussion

Administration of vatinoxan alone resulted in watery faeces and behaviour indicative of abdominal discomfort in some of the horses (Study II). This finding is in agreement with a previous study, where yohimbine increased motility of the gastrointestinal tract in horses (Lester et al. 1998). In humans, vatinoxan caused nausea and abdominal discomfort (Schafers et al. 1992), and in dogs transient salivation, which could be indicative of nausea (Honkavaara et al. 2011). These effects are most likely a result

of change in the α2-adrenenergic tone of the gastrointestinal tract (Lester et al. 1998), which is mediat-

ed by α2-adrenoceptors in the smooth muscle layer. The effect could be related to 2Aα -receptors located in the enteric nervous system, which may inhibit the release of acetylcholine from the Auerbach plexus

when activated (Blandizzi et al. 1991). In horses, alphα2-adrenoceptors have been detected at least

in the ileum, where they have binding properties of both α2A and α2B-adrenoceptors (Re et al. 2001).

6.4 Effects on selected neurohormonal variables and changes related to energy metabolism

Romifidine-induced hyperglycaemia has been reported earlier in horses (Kullmann et al. 2014), and it was demonstrated also in Study IV. The plasma glucose concentration increased throughout the monitoring period, which means that the peak concentration could possibly have emerged even after 120 minutes, which was the last time-point of measurements. Vatinoxan alleviated the hyperglycaemic effect of romifidine and the incomplete effect was most likely related to a low dose ratio of vatinoxan (200 µg/kg) to romifidine (80

µg/kg). In theory, there could also be other explanations for incomplete antagonism, such as α1-adrenergic action of romifidine, which could affect plasma glucose concentration via glucose metabolism in the liver. Vatinoxan alone had no effect on plasma glucose concentration (IV), which is in agreement with previous reports in healthy human volunteers (Warren et al. 1991; Schafers et al. 1992).

In Study IV, the RIA kits for analysis of serum insulin concentrations were not commercially available at the time of the trials, and the chemiluminescent immunometric assay method (Immulite) was chosen instead. As Immulite is designed for analyses of clinical samples, its lower detection limit of insulin, 2 µIU/ml, was obviously too high for research purposes, as many of our samples had insulin concentrations below that, leading to loss of potentially important information about the serum insulin concentrations of the horses. Serum insulin concentration, measured with Immulite, had a decreasing trend after all treatments, but the level of significance was reached only after V. Decreased insulin concentrations in plasma after administra-

tion of α2-adrenoceptor agonists is a well-known phenomenon and is a result of α2-adrenoceptor activation in pancreatic β-cells, which, in turn, lead to decreased insulin secretion. The effect was firstly discovered in mice (Angel et al. 1988) and following hyperglycaemia. Because of the limitations of our insulin analyses, this association in horses remains to be investigated elsewhere. Also, the effect of vatinoxan alone on serum insulin concentrations warrants further research in horses. The decrease detected in the present study was minor and individual variation in insulin concentrations was very high, necessitating further investigations before reliable conclusions can be drawn.

Contrary to an old report indicating that around 75% of insulin is lost during prolonged storage at -20°C (Feldman and Chapman 1973), many of the insulin concentrations were, surprisingly, higher in the results yielded by RIA than by chemiluminescent immunometric assay (IV). The reference standards of the RIA kit gave the expected values, indicating that the kit itself and the analysis procedure were successful. Samples were measured in duplicate and there was a good correspondence within the results of duplicate samples. In summary, we were unable to add any reliable information about the insulin concentrations of the horses in Study IV, but further investigation by using RIA as an analysis method for adequate samples might give new

information about the effects of vatinoxan alone and combined with α2-adrenoceptor agonists on plasma insulin concentrations in horses.

42 43 Discussion

All of the treatments in Study IV resulted in an increase in serum FFA concentrations; the increase was of the same magnitude after ROM and ROM+V, indicating a mechanism other than the peripheral α2-adrenoceptor-mediated mechanism. After V, the increase in serum FFA concentration was significantly greater than after ROM or ROM+V. In previous studies, α2-adrenoceptor agonists have been reported to decrease the serum FFA concentration in horses (Carroll et al. 1997) and dogs (Ambrisko and Hikasa 2002; 2003), whereas vatinoxan increased serum FFA concentration and lipolytic response to exercise in humans (Schiberras et al. 1994). In contrast, serum FFA in goats remained unchanged after medetomidine (Carroll et al. 2005). The- refore, the increase in serum FFA concentration after all treatments in Study IV was most likely a manifes- tation of stress from instrumentation and restraining non-sedated horses in stocks for the monitoring period, which the administration of romifidine alleviated. This is supported by an observation that at least some of the horses seemed to be stressed in the stocks when romifidine was not part of the medication protocol. Furt- hermore, catheter placement is reported to be a probable reason for increased serum FFA in goats (Carroll et al. 1998). In another study, the increase in baseline FFA concentration of goats was prevented when baseline blood samples were taken two hours after placement of catheters, when the stress from catheterization had already ceased, resulting in unchanged FFA concentration after medetomidine administration (Carroll et al. 2005). The effect of vatinoxan and dexmedetomidine on plasma FFA concentration was previously studied in dogs (Restitutti et al. 2012). Administration of dexmedetomidine alone resulted in a decrease in plasma FFA concentration, as expected. When vatinoxan was co-administered with dexmedetomidine, the effect was biphasic; at first, the FFA concentration decreased similarly to treatment with dexmedetomidine alone, but after 35 minutes, the concentration started to increase in combined administration while it decreased further after dexmedetomidine (Restitutti et al. 2012). The exact mechanism for the biphasic response remained unexplained, but it was speculated to be a result of the balance between the positive action of catecholamines and the negative action of insulin on hormone-sensitive lipase, which is the main regulator of lipolysis (see review in Langin et al. 1996). The balance between the actions of catecholamines and insulin might be the mechanism behind the changes in serum FFA concentrations in Study IV, although this requires further investigation.

Serum cortisol increased quickly after all treatments, and ACTH showed a similar trend, but the level of significance was reached only after V, most likely because of high individual variation in plasma ACTH concentrations and the small number of horses used in the study. The concentrations of cortisol and ACTH returned near baseline levels within the monitoring period, which could be indicative of stress response related to instrumentation because the blood samples for baseline measurements were drawn in the stall before instrumentation. One horse became very agitated in the stocks unless it was heavily sedated; the same horse had exceptionally high ACTH concentrations in some of the blood samples, increasing the SD of the results in such a small population. In previous equine studies, detomidine has been reported to have no effect (Raekallio et al. 1991b; Carroll et al. 1997) or to decrease (Raekallio et al 1992) serum cortisol concentration depending on its baseline. In contrast, medetomidine increased serum cortisol concentration in cattle and sheep, which might have resulted from stress induced by hypoxaemia, which is typical for ruminants (Ranheim et al. 2000). In Study IV, fairly heavy instrumentation was needed because of the concomitant cardiovascular study, probably causing significant stress to the horses. To differentiate between the effect of stress and the direct effect of the drugs studied, further investigations are needed.

6.5 Study limitations and methodological considerations

These studies have several limitations. Firstly, the number of horses was low in all studies. A larger study population would have added statistical power to the results, but financial constraints, and the licenses for animal testing, restricted the number of horses used in the studies. However, power calculations and the

44 45 Discussion

results of Study I indicated that the number of horses was sufficient to detect statistical differences between the treatments.

Secondly, the use of control groups would have allowed more reliable conclusions to be drawn. The addition of treatment with vatinoxan alone to Study I and non-treated controls to Studies I, II and IV would have added information about vatinoxan administered as a sole agent to horses and the possible effect of stress caused by instrumentation and restraining of horses in stocks for the monitoring period, respectively. Without a non-treated control, differentiation of the effects of vatinoxan from the stress-related responses is challenging. This is especially true in Study IV, in which the baseline blood samples for biochemistry analyses were drawn in the stall before the horses were taken to the examination room, predisposing them to stress from instrumentation. Placement of catheters in facial arteries of non-sedated horses required immobility, which had to be ensured by use of a twitch. In addition, some of the horses mildly resisted their placement in stocks, which could have resulted in stress response. The limited availability of vatinoxan and economic restrictions prevented the use of additional treatment groups in the studies. The use of a larger study population would have increased statistical power.

All of the observers in the studies were not blinded to the treatments used, which could bias the results. True blinding of the observers was hindered by some practical issues since the cardiovascular variables differed considerably between the treatments and the cardiovascular variables seen on the screen of the multiparameter monitors revealed the treatment administered. In studies where blinding was included, this problem was addressed by preventing access of blinded observers to the screen of multiparameter monitors.

Cardiovascular monitoring was limited to HR (Studies I and II), ABP (Study II) and CVP (Study I) in conscious horses. In an ideal setting, a complete haemodynamic evaluation (including CO) would have been performed. Measurement of CO would allow determination of whether vatinoxan alleviated the negative chronotropic effect of detomidine and romifidine via prevention of vasoconstrictive effect and increased

SVR. This effect is reported in other species treated with 2α -adrenoceptor agonists combined with vatinoxan

(Pagel et al. 1998). In addition, measurement of CO would be needed to assess whether DO2 improves in

horses when α2-adrenoceptor agonists are combined with vatinoxan.

Central venous pressure was measured in Study I. Central venous pressure would provide more information if MAP and CO were measured simultaneously, as SVR could be calculated if all of these variables were measured. Central venous pressure alone is not a reliable indicator of cardiovascular function. The measurement of CVP in standing horses includes many sources of error, such as variation in head posture (Norton et al. 2011), which might change slightly during the trials as the level of sedation changes. We tried to minimize this by supporting the heads of the horses with ropes. Another important source of error is that the placement of the catheter tip was assessed only by the shape of the CVP curve in the monitor. During the trial there was some variation in the shape of the curve in addition to high variation in CVP readings; thus, the possibility exists that the catheter tip was moving during the monitoring period.

In horses under general anaesthesia (Study III), the evaluation of haemodynamic function was confounded by dobutamine infusion. The larger dose of dobutamine needed for blood pressure support in vatinoxan-treated

horses could have markedly influenced the CO and DO2 values, hindering accurate evaluation of the effect of vatinoxan on these variables. The use of dobutamine was, however, necessary to avoid possible compli- cations of hypotension during general anaesthesia.

A variety of methods are available for CO measurements in horses. The lithium dilution technique is considered accurate for CO measurements in anaesthetized horses (Linton et al. 2000). In addition, the availability of equipment and a minimally invasive technique supported the use of the LiDCO technique for CO

44 45 Discussion measurements in Study III. Compared with the bolus thermodilution technique, the LiDCO technique is reported to overestimate CO (Ambrisko et al. 2012). The most likely reason for overestimation is that some drugs, such as detomidine and ketamine, which were also used in Study III, are reported to interact with the LiDCO sensor, possibly biasing the results of CO measurements (Ambrisko et al. 2013). We assumed that the potential bias in Study III would be similar in magnitude between the treatments of the study, as the doses of the drugs, excluding dobutamine and vatinoxan, were the same in both treatments. Dobutamine infusion did not cause a discrepancy in CO results between LiDCO and thermodilution methods in horses, whereas a significant bias was recorded in conjunction with xylazine infusion (Hopster et al. 2015), so it is justifiable to assume that dobutamine did not significantly affect CO results in Study III. Another contributing factor with unknown impact on the results is the additive effect of different drugs, which has been reported with xylazine and ketamine in vitro (Ambrisko et al. 2013), but further investigation is needed to estimate the effects of various drug combinations on LiDCO results.

6.6 Clinical implications and future prospects

The studies indicate that the combination of vatinoxan and an α2-adrenoceptor agonist could be beneficial compared with administration of an α2-adrenoceptor agonist alone for sedation in conscious horses. The first advantage is the prevention of negative chronotropic effect and peripheral vasoconstriction, leading to preserved cardiac function and oxygen delivery to peripheral tissues. Complete haemodynamic evaluation

(including CO) of conscious horses medicated with vatinoxan combined with α2-adrenoceptor agonist is needed to confirm the effects of vatinoxan on SVR and 2DO .

Another favourable feature of combined administration is maintenance of intestinal motility during sedation with an α2-adrenoceptor agonist. This effect may be particularly important when 2α -adrenoceptor agonists are used as infusions during standing operations that may last for several hours. The standing surgical and dental operations are becoming more common as the risk and cost of general anesthesia can be avoi- ded with the standing techniques. In practice, abdominal pain is a fairly common side-effect after long-lasting sedation with α2-adrenoceptor agonists. It would be beneficial to investigate the effects of vatinoxan combined with α2-adrenoceptor agonists administered as constant rate infusion for standing procedures in horses.

On the other hand, as a part of premedication before general anesthesia, vatinoxan caused severe hypotension (MAP <50 mmHg), which must be resolved before the drug is potentially usable for this indication in horses. However, the possible improvement in peripheral perfusion might be beneficial in horses under general anaesthesia, as equine species are prone to muscular damage due to high body weight and compromised peripheral perfusion. Determination of the optimal dose of vatinoxan in both conscious horses and as a premedication before general anaesthesia requires further investigation. Also, the effects of vatinoxan in compromised patients, such as horses suffering from colic or conditions affecting cardiovascular and other important organ systems, await elucidation.

In Study IV, the metabolic changes induced by romifidine and vatinoxan were investigated only in clinically healthy animals. These effects need to be evaluated in horses with underlying diseases related to the endocrine system (such as pituitary pars intermedia dysfunction and equine metabolic syndrome) and criti- cally ill patients (such as horses with colic and neonatal foals). This is especially important because these effects seem to last longer than the sedative effect of romifidine.A possible target for future research is the evaluation of these effects in horses under stressful situations.

46 47 Conclusions

7. CONCLUSIONS

1) Vatinoxan had a minor but statistically significant effect on the clinical depth of sedation by

detomidine or romifidine; the AUCsed was diminished after detomidine, but the peak sedation score did not differ between the treatments I,( II).

2) The negative chronotropism induced by detomidine and romifidine and the increasedABP induced by romifidine in horses were alleviated when vatinoxan was administered simultaneously with these drugs. Administration of vatinoxan combined with detomidine also resulted in lower CVP compared with detomidine alone (I, II).

3) Due to marked hypotension, vatinoxan is not useful as a premedication before general anaesthesia at the dose tested. Otherwise cardiovascular function seemed to improve with vatinox

an (decreased SVR and increased DO2); however, these effects could be at least in part due to the large doses of dobutamine necessary for blood pressure support under these circumstances (III).

4) Vatinoxan affected the disposition of detomidine and romifidine. This manifested as increased

Vd and T½ß and reduced AUCinf of detomidine and romifidine when these were administered in combination with vatinoxan (I, II, III).

5) The reduction of intestinal motility induced by detomidine and romifidine was alleviated when

these α2-adrenoceptor agonists were administered simultaneously with vatinoxan (I, II).

6) Romifidine induced hyperglycaemia in horses, which was alleviated by vatinoxan. Serum FFA concentration increased after administration of vatinoxan, which, in turn, was alleviated by romifidine (IV).

46 47 Acknowledgements

ACKNOWLEDGEMENTS

The studies were carried out at the Department of Equine and Small Animal Medicine, Faculty of Veterinary Medicine, University of Helsinki and at MTT Agricultural Research Centre in Ypäjä, Finland.

I gratefully acknowledge the following foundations and companies for generous support: Finnish Veterinary Foundation (The Mercedes Zachariassen Foundation), Finnish Foundation of Veterinary Research, Kymin Osakeyhtiön 100 vuotissäätiö, Helsinki University, Suomen Hippos ry (Erkki Rajakoski Foundation) and Vetcare Oy for financial support; Merck, Sharpe & Dohme, Boehringer Ingelheim and IDEXX for kindly providing drugs and laboratory equipment for the studies.

My deepest appreciation and gratitude are owed to my supervisors Professor Outi Vainio, Docent Marja Rae- kallio and Docent Anna Mykkänen. Your vast knowledge, patience and encouragement were instrumental to my completing this project. I was privileged to work under your supervision. Anna, your positive attitude and amazing skill to view every setback as minor gave me the confidence to overcome the hard times during these years.

The reviewers of this thesis, Professor Rikke Buhl and Associate Professor Francisco José Teixeira-Neto, are thanked for valuable and constructive comments that greatly improved this manuscript, and Professor Görel Nyman, for accepting the role as my opponent during the defence of this thesis. I also thank Carol Ann Pelli for editing the language of the thesis.

My gratitude goes also to my co-researchers: Annemarie (Marieke) deVries, DVM, PhD, Dipl. ECVAA, and Mari Vainionpää, DVM, PhD, thank you for your constructive comments and tireless ambition. Marieke, I so admire your practical skills in clinical trials. Kati Salla, DVM, PhD, your contribution and routine with the LiDCO procedure were invaluable during the anaesthesia trial. Many thanks to Tytti Niemelä, DVM, for smooth co-operation in planning and executing the simultaneous trials with the horses in Study III. I also thank Professor Satu Sankari, Kaj-Roger Hurme and Mari Palviainen, PhD, for technical assistance and important contributions in laboratory work, Lauri Vuorilehto, MSc, Valtteri Rinne, MSc, Ville Ranta-Panula, MSc, and Professor Mika Scheinin for drug plasma concentration analysis and Marianna Myllymäki for taking care of the practical issues with horses in Ypäjä. Special thanks to Ninja Karikoski, DVM, PhD, for help and support with every problem encountered; it meant the world to me.

Last, but not least, I am indebted to my family and friends for supporting me during this work. Lotta and Maija, thank you for sorting out every practical task I needed help with.

48 49 $FNQRZOHGJHPHQWV

The support and help of my mom and dad gave me the time needed to write this thesis. I hope I have made you proud. Tiina, your friendship and support throughout the years have been priceless.

My beloved Petri, your love, endless understanding and taking responsibility for the kids and housework during these years made this work possible. My precious Senni and Martti, I love you to the moon and back!

Löytty, February 2019

49 References

REFERENCES

Ahlquist R. A Study of the adrenotropic receptors. American Journal of Physiology, 1948, 153, pp 586-599.

Ambrisko T, Hikasa Y. Neurohormonal and metabolic effects of medetomidine compared with xylazine in beagle dogs. Canadian Journal of Veterinary Research, 2002, 66, pp 42-49.

Ambrisko T, Hikasa Y. The antagonistic effects of atipamezole and yohimbine on stress-related neurohormonal and metabolic responses induced by medetomidine in dogs. Canadian Journal of Veterinary Research, 2003, 67, 64-67.

Ambrisko T, Coppens P, Kabes, R, Moens Y. Lithium dilution, pulse power analysis, and continuous thermodilution cardiac output measurements compared with bolus thermodilution in anaesthetized ponies. British journal of anaesthesia, 2012, 109, pp 864-869.

Ambrisko T, Kabes R, Moens Y. Influence of drugs on the response characteristics of the LiDCO sensor: anin vitro study. British journal of anaesthesia, 2013, 110, pp 305–310.

Angel I, Bidet S, Langer S. Pharmacological characterization of the hyperglycemia induced by alpha-2 adrenoceptor agonists. The Journal of pharmacology and experimental therapeutics, 1988, 246, pp 1098-1103.

Arnsten A, Goldman-Rakic P. Alpha 2-adrenergic mechanisms in prefrontal cortex associated with cognitive decline in aged nonhuman primates. Science, 1985, 230, pp 1273-1276.

Barthel W, Markwardt F. Aggregation of blood platelets by biogenic amines and its inhibition by antiadrenergic and antiserotoninergic agents. Biochemical Pharmacology, 1974, 23, pp 37-45.

Bennett R, Salla K, Raekallio M, Hänninen L, Rinne V, Scheinin M, Vainio O. Effects of MK-467 on the antinociceptive and sedative actions and pharmacokinetics of medetomidine in dogs. Journal of Veterinary Pharmacology and Therapeutics, 2016, 39, pp 336-343.

Berridge T, Gadie B, Roach A, Tulloch I. Alpha 2-Adrenoceptor agonists induced mydriasis in the rat by an action within the central nervous system. British journal of pharmacology, 1983, 78, pp 507-515.

Berthelsen S, Pettinger W. A functional basis for classification of α-adrenergic receptors Life Sciences, 1977, 21, pp 595-606.

Blandizzi C, Doda M, Tarkovacs G, Del Tacca M, Vizi ES. Functional evidence that acetylcholine release from

Auerbach’s plexus of guinea-pig ileum is modulated by alpha 2A-adrenoceptor subtype. European Journal of Pharmacology, 1991, 205, pp 311-313.

Blaxall H, Murphy T, Baker J, Ray C, Bylund D. Characterization of the alpha-2C adrenergic receptor subtype in the opossum kidney and in the OK cell line. The Journal of pharmacology and experimental therapeutics, 1991, 259, pp 323-329.

Bousquet P, Feldman J, Schwartz J. Central cardiovascular effects of alpha adrenergic drugs: differences between catecholamines and imidazolines. The Journal of pharmacology and experimental therapeutics, 1984, 230, pp 232-236.

Boyajian C, Loughlin S, Leslie F. Anatomical evidence for alpha-2 adrenoceptor heterogeneity: differential autoradiographic distributions of 3Hrauwolscine and 3Hidazoxan in rat brain. The Journal of pharmacology and experimental therapeutics, 1987, 241, pp 1079-1091.

50 51 References

Bryant C, Thompson J, Clarke K. Characterisation of the cardiovascular pharmacology of medetomidine in the horse and sheep. Research in Veterinary Science, 1998, 65, pp 149-154.

Buhl R, Ersbøll A, Larsen N, Eriksen L, Koch J. The effects of detomidine, romifidine or acepromazine on echocardiographic measurements and cardiac function in normal horses. Veterinary Anaesthesia and Analgesia, 2007, 34, pp 1-8.

Bylund D, Ray-Prenger C, Murphy T. Alpha-2A and alpha-2B adrenergic receptor subtypes: antagonist binding in tissues and cell lines containing only one subtype. The Journal of pharmacology and experimental therapeutics, 1988, 245, pp 600-607.

Bylund D. Alpha-2 adrenoceptor subtypes: are more better? British journal of pharmacology, 2005, 144, pp 159-160.

Carroll G, Matthews N, Hartsfield S, Slater M, Champney T, Erickson S. The effect of detomidine and its antagonism with tolazoline on stress-related hormones, metabolites, physiologic responses, and behavior in awake ponies. Veterinary Surgery, 1997, 26, pp 69-77.

Carroll G, Hartsfield S, Champney T, Slater M, Newman J. Stress-related hormonal and metabolic responses to restraint, with and without butorphanol administration, in pre-conditioned goats. Laboratory animal science, 1998, 48, pp 387-390.

Carroll G, Hartsfield S, Champney T, Geller S, Martinez E, Haley E. Effect of medetomidine and its antagonism with atipamezole on stress-related hormones, metabolites, physiologic responses, sedation, and mechanical threshold in goats. Veterinary Anaesthesia and Analgesia, 2005, 32, pp 147-157.

Cenani A, Brosnan R, Madigan S, Knych H, Madigan J. Pharmacokinetics and pharmacodynamics of intravenous romifidine and propranolol administered alone or in combination for equine sedation. Veterinary Anaesthesia and Analgesia, 2017, 44, pp 86-97.

Chambers J, Livingston A, Waterman A, Goodship A. Analgesic effect of detomidine in thoroughbred horses with chronic tendon injury. Research in veterinary science, 1993, 54, pp 53-56.

Clark J, Smith E, Davidson J. Enhancement of Sexual Motivation in Male Rats by Yohimbine. Science, 1984, 225, pp 847-849.

Clarke K, Taylor P. Detomidine: a new sedative for horses. Equine Veterinary Journal, 1986, 18, 366-370.

Clarke K, Paton B. Combined use of detomidine with opiates in the horse. Equine Veterinary Journal, 1988, 20, 331-334.

Clarke K, England G, Goossens L. Sedative and cardiovascular effects of romifidine, alone and in combination with butorphanol, in the horse. Journal of Veterinary Anaesthesia, 1991, 18, pp 25-29.

Clarke K. Anaesthesia of the horse. In: Clarke K, Trim C, Hall L, editors. Veterinary anaesthesia. St Louis: Saunders Elsevier; 2014, pp 245-311.

Clineschmidt B, Pettibone D, Lotti V, Hucker H, Sweeney B, Reiss D, Lis E, Huff J, Vacca J. A peripherally acting alpha-2 adrenoceptor antagonist: L-659,066. Journal of pharmacology and experimental therapeutics, 1988, 245, pp 32-40.

Cohen N, Lester G, Sanchez L, Merritt A, Roussel A. Evaluation of risk factors associated with development of postoperative ileus in horses. Journal of the american veterinary medical association, 2004, 225, pp 1070-1078.

50 51 References

Correa-Sales C, Rabin B, Maze M. A hypnotic response to dexmedetomidine, an alpha-2 agonist, is mediated in the locus coeruleus in rats. Anesthesiology, 1992, 76, pp 948-952.

De Jonge A, Timmermans P, Van Zwieten P. Quantitative aspects of alpha adrenergic effects induced by clonidine-like imidazolidines. II. Central and peripheral bradycardic activities. The Journal of pharmacology and experimental therapeutics, 1982, 222, pp 712-719.

Delini-Stula A. Simple behavioral measures of central alpha-adrenoceptor blocking activity of drugs. Polish journal of pharmacology and pharmacy, 1984, 36, pp 513-521.

Del Tacca M, Soldani G, Bernardini C, Martinotti E, Impicciatore M. Pharmacological studies on the mechanisms underlying the inhibitory and excitatory effects of clonidine on gastric acid secretion. European journal of pharmacology,1982, 81, pp 255-261.

Diamond M, Young L, Bartram D, Gregg A, Clutton R, Long K, Jones R. Clinical evaluation of romifidine/ ketamine/halothane anaesthesia in horses. Veterinary record, 1993, 132, pp 572-575.

Dimaio Knych H, Stanley S. Pharmacokinetics and pharmacodynamics of detomidine following sublingual administration to horses. American journal of veterinary research, 2011, 72, pp 1378-1385.

Dimaio Knych H, Steffey E, Deuel J, Shepard R, Stanley S. Pharmacokinetics of yohimbine following intravenous administration to horses. Journal of veterinary pharmacology and therapeutics, 2011a, 34, pp 58-63.

Dimaio Knych H, Steffey E, Stanley S.Pharmacokinetics and pharmacodynamics of three intravenous doses of yohimbine in the horse. Journal of veterinary pharmacology and therapeutics, 2011b, 34, pp 359-366.

Doze V, Chen B, Li Z, Maze M. Pharmacologic characterization of the receptor mediating the hypnotic action of dexmedetomidine. Acta Veterinaria Scandinavica. Supplementum,1989, 85, pp 61-64.

Drew G, Whiting S. Evidence for two distinct types of postsynaptic alpha-adrenoceptor in vascular smooth muscle in vivo. British Journal of Pharmacology, 1979, 67, pp 207-215.

Durcan M, Wozniak K, Linnoila M. Modulation of the hypothermic and hyperglycaemic effects of 8-OH-DPAT by α2-adrenoceptor antagonists. British Journal of Pharmacology, 1991, 102, pp 222–226.

Dutta S, Lal R, Karol M, Cohen T, Ebert T. Influence of cardiac output on dexmedetomidine pharmacokinetics. Journal of Pharmaceutical Sciences, 2000, 89, pp 519-527.

Eades S, Moore J. Blockade of endotoxin-induced cecal hypoperfusion and ileus with an alpha 2 antagonist in horses. American journal of veterinary research, 1993, 54, pp 586-590.

Elfenbein J, Sanchez L, Robertson S, Cole C, Sams R. Effect of detomidine on visceral and somatic nociception and duodenal motility in conscious adult horses. Veterinary Anaesthesia and Analgesia, 2009, 36, pp 162-172.

EMA. Committee for veterinary medicinal products. Romifidine. Summary report, 2013, available at: http://www.ema.europa.eu/docs/en_GB/document_library/Maximum_Residue_Limits_-_Report/2009/11/ WC500015833.pdf

England G, Clarke K, Goossens L. A comparison of the sedative effects of three alpha2 -adrenoceptor agonists (romifidine, detomidine and xylazine) in the horse.Journal of Veterinary Pharmacology and Therapeutics, 1992, 15, pp 194-201.

England G, Clarke K. Alpha 2 adrenoceptor agonists in the horse--a review. British Veterinary Journal, 1996, 152, pp 641-657.

52 53 References

Enouri S, Kerr C, McDonell W, O’Sullivan M, Neto F. Effects of a peripheral alphα2 adrenergic-receptor antagonist on the hemodynamic changes induced by medetomidine administration in conscious dogs. American Journal of Veterinary Research, 2008, 69, pp 728-736.

Ernsberger P, Giuliano R, Willette R, Reis D. Role of imidazole receptors in the vasodepressor response to clonidine analogs in the rostral ventrolateral medulla. The Journal of pharmacology and experimental therapeutics, 1990, 253, pp 408-418.

Feldman J, Chapman B. Radioimmunoassay of insulin in serum and plasma. Clinical chemistry, 1973, 19, pp 1250-1254.

Fernandez G, Lee J, Liu L, Gassler J. The Baroreflex in Hypertension.Current Hypertension Reports, 2015, 17, pp 1-8.

Foote S, Bloom F, Aston-Jones G. Nucleus locus ceruleus: new evidence of anatomical and physiological specificity. Physiological reviews, 1983, 63, pp 844-914.

Freeman S, England G. Investigation of romifidine and detomidine for the clinical sedation of horses.Veterinary Record, 2000, 147, pp 507-511.

Freeman S, England G. Effect of romifidine on gastrointestinal motility, assessed by transrectal ultrasonography. Equine Veterinary Journal, 2001, 33, pp 570-576.

Gardner R, White G, Ramsey D, Boucher J, Kilgore W, Huhtinen M. Efficacy of sublingual administration of detomidine gel for sedation of horses undergoing veterinary and husbandry procedures under field conditions. Journal of the American Veterinary Medical Association, 2010, 237, pp 1459-1464.

Gasthuys F, Terpstra P, van den Hende C, De Moor A. Hyperglycaemia and diuresis during sedation with detomidine in the horse. Journal of Veterinary Medicine, 1987, 34, pp 641-648.

Gasthuys F, Parmentier D, Goossens L, De Moor A. A preliminary study on the effects of atropine sulphate on bradycardia and heart blocks during romifidine sedation in the horse.Veterinary Research Communications, 1990, 14, pp 489-502.

Gerring E, Hunt J. Pathophysiology of equine postoperative ileus: Effect of adrenergic blockade, parasympathetic stimulation and metoclopramide in an experimental model. Equine veterinary journal, 1986, 18, pp 249–255.

Goldman M, Pettibone D, Reagan J, Clineschmidt B, Baldwin J, Huff J. Blockade of peripheral α2- adrenoceptors by L-659,066 enhances glucose tolerance and insulin release in mice. Drug Development Research, 1989, 17, pp 141-151.

Grimsrud K, Mama K, Thomasy S, Stanley S. Pharmacokinetics of detomidine and its metabolites following intravenous and intramuscular administration in horses. Equine Veterinary Journal, 2009, 41, pp 361-365.

Göthert M, Molderings G. Involvement of presynaptic imidazoline receptors in the alpha 2-adrenoceptor- independent inhibition of noradrenaline release by imidazoline derivatives. Naunyn-Schmiedeberg’s Archives of Pharmacology, 1991, 343, pp 271-282.

Hamm D, Turchi P, Jöchle W. Sedative and analgesic effects of detomidine and romifidine in horses.Veterinary Record, 1995, 136, pp 324-327.

Holman R, Shillito E, Vogt M. Sleep produced by clonidine (2-(2,6-dichlorophenylamino)-2-imidazoline hydrochloride. British Journal of Pharmacology, 1971, 43, pp 685-695.

52 53 References

Honkavaara J, Raekallio M, Kuusela E, Hyvärinen E, Vainio O. The effects of L-659,066, a peripheral

alphα2-adrenoceptor antagonist, on dexmedetomidine-induced sedation and bradycardia in dogs. Veterinary Anaesthesia and Analgesia, 2008, 35, pp 409-413.

Honkavaara J, Restitutti F, Raekallio M, Kuusela E, Vainio O. The effects of increasing doses of MK-

467, a peripheral alpha(2)-adrenergic receptor antagonist, on the cardiopulmonary effects of intravenous dexmedetomidine in conscious dogs. Journal of Veterinary Pharmacology and Therapeutics, 2011, 34, pp 332-337.

Honkavaara J, Restitutti F, Raekallio M, Salla K, Kuusela E, Ranta-Panula V, Rinne V, Vainio O, Scheinin

M. Influence of MK-467, a peripherally acting 2α -adrenoceptor antagonist on the disposition of intravenous dexmedetomidine in dogs. Drug Metabolism and Disposition, 2012, 40, pp 445-449.

Honkavaara J, Pypendop B, Turunen H, Ilkiw J. The effect of MK-467, a peripheral 2α -adrenoceptor antagonist, on dexmedetomidine-induced sedation and bradycardia after intravenous administration in conscious cats. Veterinary Anaesthesia and Analgesia, 2017, 44, pp 42-51.

Hopster K, Ambrisko T, Stahl J, Schramel J, Kästner S. Influence of xylazine on the function of the LiDCO sensor in isoflurane anaesthetized horses.Veterinary Anaesthesia and Analgesia, 2015, 42, pp 142-149.

Hsu W. Xylazine-induced depression and its antagonism by alpha adrenergic blocking agents. The Journal of pharmacology and experimental therapeutics, 1981, 21, pp 188-192.

Hubbell J, Muir W. Antagonism of detomidine sedation in the horse using intravenous tolazoline or atipamezole. Equine Veterinary Journal, 2006, 38, pp 238-241.

Hubbell J, Sams R, Schmall L, Robertson J, Hinchcliff D, Muir W. Pharmacokinetics of detomidine administered to horses at rest and after maximal exercise. Equine Veterinary Journal, 2009, 41, pp 419-422.

Jackson H, Griffin I, Birkett S, Nutt D. The effects of idazoxan and other2 α -adrenoceptor antagonists on urine output in the rat. British Journal of Pharmacology, 1992, 106, pp 443–446.

Jalonen J, Halkola L, Kuttila K, Perttilä J, Rajalin A, Savunen T, Scheinin M, Valtonen M. Effects of dexmedetomidine on coronary hemodynamics and myocardial oxygen balance. Journal of Cardiothoracic and Vascular Anesthesia, 1995, 9, pp 519-524.

Jones S, Gebhart, G. Characterization of coeruleospinal inhibition of the nociceptive tail-flick reflex in the rat: Mediation by spinal α 2-adrenoceptors. Brain Research, 1986, 364, pp 315-330.

Jöchle W, Hamm D. Sedation and analgesia with Domosedan (detomidine hydrochloride) in horses: dose response studies on efficacy and its duration.Acta Veterinaria Scandinavica. Supplementum, 1986, 82, pp 69-84.

Jöchle, W. Field trial evaluation of detomidine as a sedative and analgesic in horses with colic. Equine Veterinary Journal, 1989, 7, pp 117–120.

Jöchle W, Moore J, Brown J, Baker G, Lowe J, Fubini, S, Reeves M, Watkins J, White N. Comparison of detomidine, butorphanol, flunixin meglumine and xylazine in clinical cases of equine colic. Equine Veterinary Journal, 1989, 7, pp 111–116.

Kamerling S, Cravens W, Bagwell C. Dose-related effects of detomidine on autonomic responses in the horse. Journal of autonomic pharmacology, 1988a, 8, pp 241-249.

Kamerling S, Cravens W, Bagwell C. Objective assessment of detomidine-induced analgesia and sedation in the horse. European Journal of Pharmacology, 1988b, 151, pp 1-8.

54 55 References

Kaukinen H, Aspegren J, Hyyppa S, Tamm L, Salonen J. Bioavailability of detomidine administered sublingually to horses as an oromucosal gel. Journal of Veterinary Pharmacology and Therapeutics, 2011, 34, pp 76-81.

Kitagawa H, Walland A. Augmentation of vagal reflex bradycardia by central 2α adrenoceptors in the cat. Naunyn-Schmiedeberg’s Archives of Pharmacology, 1982, 321, pp 44-47.

Kollias-Baker C, Court M, Williams L. Influence of yohimbine and tolazoline on the cardiovascular, respiratory, and sedative effects of xylazine in the horse. Journal of Veterinary Pharmacology and Therapeutics, 1993, 16, pp 350-358.

Kubo T, Misu Y. Pharmacological characterisation of the α-adrenoceptors responsible for a decrease of blood pressure in the nucleus tractus solitarii of the rat. Naunyn-Schmiedeberg’s Archives of Pharmacology, 1981, 317, pp 120-125.

Kullmann A, Sanz M, Fosgate G, Saulez M, Page P, Rioja E. Effects of xylazine, romifidine, or detomidine on hematology, biochemistry, and splenic thickness in healthy horses. Canadian veterinary journal, 2014, 55, pp 334-340.

Lafontan M, Berlan M. Evidence for the alpha 2 nature of the alpha-adrenergic receptor inhibiting lipolysis in human fat cells. European Journal of Pharmacology, 1980, 66, 87-93.

L’Ami J, Vermunt L, van Loon J, Sloet van Oldruitenborgh-Oosterbaan M. Sublingual administration of detomidine in horses: sedative effect, analgesia and detection time.Veterinary Journal, 2013, 196, pp 253-259.

Langer S. Presynaptic regulation of catecholamine release. Biochemical Pharmacology, 1974, 23, pp 1793- 1800.

Langer S, Massingham R, Shepperson N. Presence of postsynaptic alpha 2-adrenoreceptors of predominantly extrasynaptic location in the vascular smooth muscle of the dog hind limb. Clinical Science Supplementum, 1980, 59, pp 225-228.

Langin D, Holm C, Lafontan M. Adipocyte hormone-sensitive lipase: a major regulator of lipid metabolism. The Proceedings of the Nutrition Society, 1996, 55, pp 93-109.

Latifpour J, Jones S, Bylund D. Characterization of 3Hyohimbine binding to putative alpha-2 adrenergic receptors in neonatal rat lung. The Journal of pharmacology and experimental therapeutics, 1982, 223, pp 606-611.

Lester G, Merritt A, Neuwirth L, Vetro-Widenhouse T, Steible C, Rice B. Effect of alpha2 -adrenergic, cholinergic, and nonsteroidal anti-inflammatory drugs on myoelectric activity of ileum, cecum, and right ventral colon and on cecal emptying of radiolabeled markers in clinically normal ponies. American journal of veterinary research, 1998, 59, pp 320-327.

Link R, Desai K, Hein L, Stevens M, Chruscinski A, Bernstein D, Barsh G, Kobilka B. Cardiovascular

Regulation in Mice Lacking α2-Adrenergic Receptor Subtypes b and c. Science, 1996, 273, pp 803-805.

Linton R, Young L, Marlin D, Blissitt K, Brearley J, Jonas M, O’Brien T, Linton N, Band D, Hollingworth C, Jones R. Cardiac output measured by lithium dilution, thermodilution, and transesophageal Doppler echocardiography in anesthetized horses. American journal of veterinary research, 2000, 61, pp 731-737.

Lowe J, Hilfiger J. Analgesic and sedative effects of detomidine compared to xylazine in a colic model using i.v. and i.m. routes of administration. Acta Veterinaria Scandinavica. Supplementum, 1986, 82, pp 85-95.

54 55 References

Luna S, Taylor P, Carregaro A. Atipamezole antagonism of an ACTH stimulation test in ponies sedated with detomidine. Journal of Veterinary Pharmacology and Therapeutics, 2011, 34, pp 508-511.

Luttinger D, Ferrari R, Perrone M, Haubrich D. Pharmacological analysis of alpha-2 adrenergic mechanisms in nociception and ataxia. The Journal of Pharmacology and Experimental Therapeutics, 1985, 232, 883-889.

Madjar H, Docherty J, Starke K. An Examination of Pre- and Postsynaptic α-Adrenoceptors in the Autoperfused Rabbit Hindlimb. Journal of Cardiovascular Pharmacology, 1980, 2, pp 619-62

Mama K, Grimsrud K, Snell T, Stanley S. Plasma concentrations, behavioural and physiological effects following intravenous and intramuscular detomidine in horses. Equine Veterinary Journal, 2009, 41, pp 772-777.

MacMillan L, Hein L, Smith M, Piascik M, Limbird L. Central Hypotensive Effects of the 2α -Adrenergic Receptor Subtype. Science, 1996, 273, pp 801-803.

Meisler S, Doherty T, Andrews F, Osborne D, Frazier D. Yohimbine ameliorates the effects of endotoxin on gastric emptying of the liquid marker acetaminophen in horses. Canadian journal of veterinary research, 2000, 64, pp 208-211.

Merritt A, Burrow J, Hartless C. Effect of xylazine, detomidine, and a combination of xylazine and butorphanol on equine duodenal motility. American journal of veterinary research, 1998, 59, pp 619-623.

Michel A, Loury D, Whiting R. Differences between the 2α -adrenoceptor in rat submaxillary gland and the α2A- and α2B-adrenoceptor subtypes. British Journal of Pharmacology, 1989, 98, pp 890-897.

Moens Y, Lanz F, Doherr M, Schatzmann U. A comparison of the antinociceptive effects of xylazine, detomidine and romifidine on experimental pain in horses.Veterinary Anaesthesia and Analgesia, 2003, 30, pp 183–190.

Muir W. Anxiolytics, nonopioid sedative-analgesics, and opioid analgesics. In: Muir W, Hubbel J, editors. Equine anesthesia. St Louis: Saunders Elsevier; 2009, pp 185-209.

Murphy T, Bylund D. Characterization of alpha-2 adrenergic receptors in the OK cell, an opossum kidney cell line. The Journal of pharmacology and experimental therapeutics, 1988, 244, pp 571-578.

Myers R, Beleslin D, Rezvani A. Hypothermia: role of alpha 1- and alpha 2-noradrenergic receptors in the hypothalamus of the cat. Pharmacology, biochemistry, and behavior, 1987, 26, 373-379.

Nakadate T, Nakati T, Muraki T, Kato R. Regulation of plasma insulin level by α2-adrenergic receptors. European Journal of Pharmacology, 1980, 65, pp 421–424.

Nilsfors L, Kvart C. Preliminary report on the cardiorespiratory effects of the antagonist to detomidine, MPV- 1248. Acta Veterinaria Scandinavica. Supplementum, 1986, 82, pp 121-129.

Nomura Y, Oki K, Segawa T. Pharmacological characterization of central α-adrenoceptors which mediate clonidine-induced locomotor hypoactivity in the developing rat. Naunyn-Schmiedeberg’s Archives of Pharmacology, 1980, 311, pp 41-44.

Norton J, Nolen-Walston R, Undenwood C, Boston R, Slack J, Dallap B. Repeatability, reproducibility, and effect of head position on central venous pressure measurement in standing adult horses.Journal of veterinary internal medicine, 2011, 25, pp 575-578.

Nuñez E, Steffey E, Ocampo L, Rodriguez A, Garcia A. Effects of alphα2-adrenergic receptor agonists on urine production in horses deprived of food and water. American Journal of Veterinary Research, 2004, 65, 1342-1346.

56 57 References

Nyman G, Marntell S, Edner A, Funkquist P, Morgan K, Hedenstierna G. Effect of sedation with detomidine and butorphanol on pulmonary gas exchange in the horse. Acta Veterinaria Scandinavica, 2009, 51, pp.22

Owens J, Kamerling S, Stanton S, Keowen M. Evaluation of detomidine-induced analgesia in horses with chronic hoof pain. The Journal of pharmacology and experimental therapeutics, 1996 278, pp 179– 184.

Paddleford R, Harvey R. Alpha 2 agonists and antagonists. Veterinary Clinics of North America: Small Animal Practice, 1999, 29, pp 737-745.

Pagel P, Proctor L, Devcic A, Hettrick D, Kersten J, Tessmer J, Farber N, Schmeling W, Warltier D. A novel

alpha 2-adrenoceptor antagonist attenuates the early, but preserves the late cardiovascular effects of intravenous dexmedetomidine in conscious dogs. Journal of Cardiothoracic and Vascular Anesthesia, 1998, 12, pp 429-434.

Pellow S, Chopin P, File S. Are the anxiogenic effects of yohimbine mediated by its action at benzodiazepine receptors? Neuroscience Letters, 1985, 55, pp 5-9.

Piletz J, Ivanov T, Sharp J, Ernsberger P, Chang C, Pickard R, Gold G, Roth B, Zhu H, Jones J, Baldwin J, Reis D. Imidazoline receptor antisera-selected (IRAS) cDNA: cloning and characterization. DNA and Cell Biology, 2000, 19, pp 319–329.

Pimenta E, Teixeira Neto F, Sá P, Pignaton W, Garofalo N. Comparative study between atropine and hyoscine- N-butylbromide for reversal of detomidine induced bradycardia in horses. Equine Veterinary Journal, 2011, 43, 332-340.

Pypendop B, Honkavaara J, Ilkiw J. Pharmacokinetics of dexmedetomidine, MK-467, and their combination following intravenous administration in male cats. Journal of Veterinary Pharmacology and Therapeutics, 2016, 39, pp 460-468.

Pypendop B, Honkavaara J, Ilkiw J. Pharmacokinetics of dexmedetomidine, MK-467 and their combination following intramuscular administration in male cats. Veterinary Anaesthesia and Analgesia, 2017, 44, pp 823-831.

Raekallio M, Vainio O, Karjalainen J. The influence of atipamezole on the cardiovascular effects of detomidine in horses. Journal of the Association of Veterinary Anaesthetists of Great Britain and Ireland, 1990, 17, pp 50-53.

Raekallio M, Vainio O, Karjalainen J. The influence of detomidine and epinephrine on heart rate, arterial blood pressure, and cardiac arrhythmia in horses. Veterinary surgery, 1991a, 20, pp 468-473.

Raekallio M, Vainio O, Scheinin M. Detomidine reduces the plasma catecholamine, but not cortisol concentration in horses. Journal of Veterinary Medicine Series A, 1991b, pp 153-165.

Raekallio M, Leino A, Vainio O, Scheinin M. Sympatho-adrenal activity and the clinical sedative effect of detomidine in horses. Equine veterinary journal. Supplementum, 1992, 11, pp 66-68.

Raekallio M, Honkavaara J, Vainio O. The effects of L-659,066, a peripheral 2α -adrenoceptor antagonist, and verapamil on the cardiovascular influences of dexmedetomidine in conscious sheep. Journal of Veterinary Pharmacology and Therapeutics, 2010, 33, pp 434-438.

Ramseyer B, Schmucker N, Schatzmann U, Busato A, Moens Y. Antagonism of detomidine sedation with atipamezole in horses. Journal of Veterinary Anaesthesia, 1998, 25, pp 47–51.

Ranheim B, Horsberg T, Søli N, Ryeng K, Arnemo J. The effects of medetomidine and its reversal with atipamezole on plasma glucose, cortisol and noradrenaline in cattle and sheep. Journal of Veterinary Pharmacology and Therapeutics, 2000, 23, pp 379-387.

56 57 References

Re G, Badino P, Odore R, Galaverna D, Girardi C. Characterization of alpha-adrenoceptor subtypes in smooth muscle of equine ileum. American journal of veterinary research, 2001, 62, pp 1370-1374. Reitemeyer H, Klein H, Deegen E. The effect of sedatives on lung function in horses. Acta Veterinaria Scandinavica. Supplementum, 1986, 82, pp 111-120.

Restitutti F, Honkavaara J, Raekallio M, Kuusela E, Vainio O. Effects of different doses of L-659’066 on the bispectral index and clinical sedation in dogs treated with dexmedetomidine. Veterinary Anaesthesia and Analgesia, 2011, 38, pp 415-422.

Restitutti F, Raekallio M, Vainionpää M, Kuusela E, Vainio O. Plasma glucose, insulin, free fatty acids, lactate

and cortisol concentrations in dexmedetomidine-sedated dogs with or without MK-467: a peripheral α-2 adrenoceptor antagonist. Veterinary Journal, 2012, 193, 481-485.

Restitutti F, Kaartinen M, Raekallio M, Wejberg O, Mikkola E, Del Castillo J, Scheinin M, Vainio O. Plasma concentration and cardiovascular effects of intramuscular medetomidine combined with three doses of the

peripheral alphα2-antagonist MK-467 in dogs. Veterinary Anaesthesia and Analgesia, 2017, 44, pp 417-426.

Risberg Å, Ranheim B, Krontveit R, Lervik A, Haga H. The cardiovascular status of isoflurane-anaesthetized horses with and without dexmedetomidine constant rate infusion evaluated at equivalent depths of anaesthesia. Veterinary Anaesthesia and Analgesia, 2016, 43, pp 412-423.

Rohrbach H, Korpivaara T, Schatzmann U, Spadavecchia C. Comparison of the effects of the alpha-2 agonists detomidine, romifidine and xylazine on nociceptive withdrawal reflex and temporal summation in horses. Veterinary Anaesthesia and Analgesia, 2009, 36, pp 384-395.

Roger T, Ruckebusch Y. Colonic alpha 2-adrenoceptor-mediated responses in the pony. Journal of Veterinary Pharmacology and Therapeutics, 1987, 10, pp 310-318.

Rolfe N, Kerr C, McDonell W. Cardiopulmonary and sedative effects of the peripheral 2α -adrenoceptor antagonist MK 0467 administered intravenously or intramuscularly concurrently with medetomidine in dogs. American Journal of Veterinary Research, 2012, 73, pp 587-594.

Ruuskanen J, Xhaard H, Marjamäki A, Salaneck E, Salminen T, Yan Y, Postlethwait J, Johnson M, Larhammar

D, Scheinin M. Identification of duplicated fourth alphα2-adrenergic receptor subtype by cloning and mapping of five receptor genes in zebrafish.Molecular biology and evolution, 2004, 21, pp 14-28.

Ruwart M, Klepper M, Rush B. Clonidine delays small intestinal transit in the rat. Journal of Pharmacology and Experimental Therapeutics, 1980, 212, pp 487-490.

Saarinen H. Preanesthetic use of detomidine in horses--some clinical observations. Acta Veterinaria Scandinavica. Supplementum, 1986, 82, pp 157-165.

Sabbe M, Penning J. Ozaki G, Yaksh T. Spinal and systemic action of the alpha 2 receptor agonist dexmedetomidine in dogs. Antinociception and carbon dioxide response. Anesthesiology, 1994, 80, pp 1057-1072.

Salonen J. Pharmacokinetics of detomidine. Acta Veterinaria Scandinavica. Supplementum, 1986, 82, pp 59-66.

Salonen J, Suolinna E. Metabolism of detomidine in the rat. I. Comparison of 3 H-labelled metabolites formed in vitro and in vivo. European Journal of Drug Metabolism and Pharmacokinetics, 1988, 13, pp 53-58.

Salonen J, Vuorilehto L, Eloranta M, Karjalainen A. Metabolism of detomidine in the rat. II. Characterisation of metabolites in urine. European Journal of Drug Metabolism and Pharmacokinetics, 1988, 13, pp 59-65.

Salonen J, Vähä-Vahe T, Vainio O, Vakkuri O. Single-dose pharmacokinetics of detomidine in the horse and cow. Journal of Veterinary Pharmacology and Therapeutics, 1989, 12, pp 65-72.

58 59 References

Salonen J, Vuorilehto L, Gilbert M, Maylin G. Identification of detomidine carboxylic acid as the major urinary metabolite of detomidine in the horse. European Journal of Drug Metabolism and Pharmacokinetics, 1992, 17, pp 13-20.

Salonen S, Vuorilehto L, Vainio O, Anttila M. Atipamezole increases medetomidine clearance in the dog: an agonist-antagonist interaction. Journal of Veterinary Pharmacology and Therapeutics, 1995, 18, pp 328-332.

Sarazan R, Starke W, Krause G, Garner H. Cardiovascular effects of detomidine, a new alpha2 -adrenoceptor agonist, in the conscious pony. Journal of Veterinary Pharmacology and Therapeutics, 1989, 12, pp 378- 388.

Savola J, Ruskoaho H, Puurunen J, Kärki N. Cardiovascular action of detomidine, a sedative and analgesic imidazole derivative with alpha-agonistic properties. European journal of pharmacology, 1985, 26, 118, pp 69-76.

Savola J. Cardiovascular actions of detomidine. Acta Veterinaria Scandinavica. Supplementum, 1986, 82, pp 47-57.

Savola J. Cardiovascular actions of medetomidine and their reversal by atipamezole. Acta Veterinaria Scandinavica. Supplementum, 1989, 85, pp 39-47.

Schafers R, Elliott H, Howie C, Reid J. A preliminary, clinical pharmacological assessment of L-659,066, a

novel α2-adrenoceptor antagonist. British Journal of Clinical Pharmacology, 1992, 34, pp 521-526.

Schwartz D, Clark T. Affinity of detomidine, medetomidine and xylazine for alpha-2 adrenergic receptor subtypes. Journal of Veterinary Pharmacology and Therapeutics, 1998, 21, pp 107-111.

Sciberras D, Reed J, Elliott C, Blain P, Goldberg M. The effects of a peripherally selective alpha2 -adrenoceptor antagonist, MK-467, on the metabolic and cardiovascular response to exercise in healthy man. British Journal of Clinical Pharmacology, 1994, 37, pp 39-44.

Schulz A, Hasselblatt A. An insulin-releasing property of imidazoline derivatives is not limited to compounds that block α-adrenoceptors. Naunyn-Schmiedeberg’s Archives of Pharmacology, 1989, 340, pp 321-327.

Shattil S, Mcdonough M, Turnbull J, Insel P. Characterization of alpha-adrenergic receptors in human platelets using [3H]clonidine. Molecular pharmacology, 1981, 19, pp 179-183.

Short C, Matthews N, Harvey R, Tyner C. Cardiovascular and pulmonary function studies of a new sedative/ analgetic (detomidine/Domosedan) for use alone in horses or as a preanesthetic. Acta Veterinaria Scandinavica. Supplementum, 1986a, 82, pp 139-155.

Short C, Stauffer J, Goldberg G, Vainio O. The use of atropine to control heart rate responses during detomidine sedation in horses. Acta veterinaria Scandinavica, 1986b, 27, pp 548-559.

Simonneaux V, Ebadi M, Bylund D. Identification and characterization of alpha2D -adrenergic receptors in bovine pineal gland. Molecular pharmacology, 1991, 40, pp 235-241.

Smyth D, Umemura S, Pettinger W. Alpha 2-adrenoceptor antagonism of vasopressin-induced changes in sodium excretion. The American journal of physiology, 1985, 248, pp F767-772.

Spadavecchia C, Arendt-Nielsen L, Andersen O, Spadavecchia L, Schatzmann U. Effect of romifidine on the nociceptive withdrawal reflex and temporal summation in conscious horses.American journal of veterinary research, 2005, 66, pp 1992-1998.

Steffey E, Pascoe P. Detomidine reduces isoflurane anesthetic requirement (MAC) in horses.Veterinary Anaesthesia and Analgesia, 2002, 29, pp 223-227.

58 59 References

Svensson T, Bunney B, Aghajanian G. Inhibition of both noradrenergic and serotonergic neurons in brain by the α-adrenergic agonist clonidine. Brain Research, 1975, 92, pp 291-306.

Szemeredi K, Stull R, Kopin I, Goldstein D. Effects of a peripherally acting alpha2 -adrenoceptor antagonist (L-659,066) on hemodynamics and plasma levels of catechols in conscious rats. European journal of pharmacology, 1989, 170, pp 53-59.

Taylor P, Bennett R, Brearley J, Luna S, Johnson C. Comparison of detomidine and romifidine as premedicants before ketamine and halothane anesthesia in horses undergoing elective surgery. American journal of veterinary research, 2001, 62, pp 359-363.

Tesson F, Limon-Boulez I, Urban P, Puype M, Vande- Kerckhove J, Coupry I, Pompon D, Parini A. Localization of I2-imidazoline binding sites on monoamine oxidases. The Journal of Biological Chemistry, 1995, 270, pp 9856–9861.

Timmermans P, van Zwieten P. Vasoconstriction mediated by postsynaptic alpha 2-adrenoceptor stimulation. Naunyn Schmiedebergs Archives of Pharmacology, 1980, 313, pp 17-20.

Tiniakov R, Scrogin K. The serotonin 5-hydroxytryptaphan1A receptor agonist, (+)8- hydroxy-2-(di-n- propylamino)-tetralin, stimulates sympathetic- dependent increases in venous tone during hypovolemic shock. The Journal of Pharmacology and Experimental Therapeutics, 2006, 319, pp 776-782.

Vainio O. Detomidine, a new sedative and analgesic drug for veterinary use: pharmacological and clinical studies in laboratory animals, horses and cattle. Academic dissertation, University of Helsinki, 1985.

Virtanen R, Ruskoaho H, Nyman L. Pharmacological evidence for the involvement of alpha-2 adrenoceptors in the sedative effect of detomidine, a novel sedative analgesic.Journal of Veterinary Pharmacology and Therapeutics, 1985, 8, pp 30-37.

Virtanen R, Savola J, Saano V, Nyman L. Characterization of the selectivity, specificity and potency of

medetomidine as an α 2-adrenoceptor agonist. European Journal of Pharmacology, 1988, 150, pp 9-14.

Wagner A, Muir W, Hinchcliff K. Cardiovascular effects of xylazine and detomidine in horses.American journal of veterinary research, 1991, 52, pp 651-657.

Warren J, Dollery C, Sciberras D, Goldberg M. Assessment of MK-467, a peripheral alpha 2-adrenergic receptor antagonist, with intravenous clonidine. Clinical Pharmacology & Therapeutics, 1991, 50, pp 71-77.

Wojtasiak-Wypart M, Soma L, Rudy J, Uboh C, Boston R, Driessen B. Pharmacokinetic profile and pharmacodynamic effects of romifidine hydrochloride in the horse.Journal of Veterinary Pharmacology and Therapeutics, 2012, 35, pp 478-488.

Zhong H, Minneman K. α1-Adrenoceptor subtypes. European Journal of Pharmacology, 1999, 375, pp 261- 276.

Zullian C, Menozzi A, Pozzoli C, Poli E, Bertini S. Effects of α2-adrenergic drugs on small intestinal motility in the horse: an in vitro study. Veterinary Journal, 2011, 187, pp 342-346.

Yamashita K, Tsubakishita S, Futaoka S, Ueda I, Hamaguchi H, Seno T, Katoh S, Izumisawa Y, Kotani T, Muir W. Cardiovascular effects of medetomidine, detomidine and xylazine in horses.Journal of Veterinary Medical Science, 2000, 10, pp 1025-1032.

60 61 Appendix

APPENDIX 1.

Sedation scoring according to Rohrbach et al. (2009). Maximum sedation score is 10.

1. Attitude Nervous 0 Calm 1 Apathetic 2 Stuporous 3

2. Standing ability Stands well 0 Leans slightly on stocks 1 Leans strongly on stocks 2 Difficulty standing 3

3. Head Moving 0 Quiet 1 Hanging in halter 2

4. Eyes Normal 0 Slightly closed 1

5. Ears Moving 0 Not moving 1

60 61