Pharmacodynamics and plasma concentrations of dexmedetomidine with or without vatinoxan as a constant- rate infusion in horses anaesthetized with isoflurane— A pilot study

The aim was to determine the effects of vatinoxan on dexmedetomidine plasma concentrations and effects on cardiovascular and intestinal tissue pharmacodynamics. In a prospective randomized study, six horses were premedicated intravenously with dexmedetomidine 3.5 µg kg −1 followed by a constant- rate infusion of 7 µg kg −1 h −1 (group DEX) and six horses with dexmedetomidine of the same dose (bolus and constant- rate infusion) combined with vatinoxan 130 µg kg −1 followed by 40 µg kg −1 h −1 (group VAT). Anaesthesia was induced with ketamine and diazepam and maintained with isoflurane. Venous blood samples were withdrawn before and at predefined points in time after drug application. During sedation and anaesthesia, cardiopulmonary variables, gastrointestinal tissue perfusion and oxygenation were recorded. Data were analysed using two- way- ANOVA, unpaired- t- test and Dunnett's- t - test ( p < 0.05). Group VAT had significantly higher oxygen delivery and lower oxygen extraction ratio, venous admixture, alveolar dead space and alveolar- arterial- oxygen difference. Tissue perfusion of buccal mucosa was reduced during anaesthesia in group DEX. Plasma concentrations of dexmedetomidine in group VAT ( n = 6) and group DEX ( n = 5) were comparable between groups. In the present pilot study, co- administration of vatinoxan with dexmedetomidine did not alter plasma concentrations of dexmedetomidine but ameliorated tissue perfusion and global oxygenation variables.

typical alpha-2 adrenoceptor agonist side-effects, although due to its pharmacokinetic profile these are only minor and short-lasting (Bettschart-Wolfensberger et al., 2005;Ranheim et al., 2015). After bolus administration of dexmedetomidine (3.5 µg kg −1 ), gastrointestinal tissue perfusion and oxygenation are decreased (Neudeck et al., 2018), potentially related to alpha-2b adrenoceptor mediated constriction of vascular smooth muscles and consecutive reduction of cardiac output. Physiologically, tissue can counteract the reduced blood flow associated with lower oxygen delivery by increased oxygen extraction, which in turn, leads to maintenance of tissue oxygenation. Nevertheless, this protective mechanism only works to a certain extent until tissue oxygenation becomes dependent on blood supply. During balanced anaesthesia with desflurane and dexmedetomidine (7 µg kg −1 h −1 ), the global cardiovascular function, as well as the gastrointestinal oxygenation, were well-maintained, whereas the gastrointestinal perfusion was minimally reduced (Neudeck et al., 2018). The peripheral alpha-2 adrenoceptor antagonist MK-467 or L-659.066, which was recently named vatinoxan, poorly crosses the blood/brain barrier due to its low lipid solubility (Clineschmidt et al., 1988;Honkavaara et al., 2020). Several of the aforementioned alpha-2 adrenoceptor agonist side-effects, including bradycardia, gastrointestinal hypomotility as well as reduced cardiac output can be ameliorated or avoided by concomitant administration of vatinoxan with alpha-2 adrenoceptor agonists in sheep (Raekallio et al. 2010), dogs (Honkavaara et al., 2011(Honkavaara et al., , 2012 and horses (Pakkanen et al., 2015;Vainionpää et al., 2013;Vries et al., 2016).
A bolus of 250 µg kg −1 and 200 µg kg −1 vatinoxan during general anaesthesia in horses induced severe hypotension leading to impaired gastrointestinal blood flow (Pakkanen et al., 2015;Wittenberg-Voges et al., 2017). However, the gastrointestinal oxygenation was not affected, even though blood flow was markedly reduced. The cardiac index (CI) was well-maintained and probably responsible for the sustained oxygen delivery . Vatinoxan administration of 140 µg kg −1 with medetomidine in horses 10 min prior to anaesthesia also led to hypotensive conditions and increased dobutamine requirements compared to horses only treated with medetomidine (Tapio, Raekallio, Mykkänen, Al-Ramahi, et al., 2019). Therefore, an appropriate dose ratio between the alpha-2 agonist and vatinoxan is necessary to avoid marked hypotension, especially during general anaesthesia.
In cats (Pypendop et al., 2017), dogs (Restitutti et al., 2017) and sheep (Adam et al., 2018) intramuscular bolus co-administration of vatinoxan with an alpha-2 adrenoceptor agonist resulted in an increased absorption rate and early-stage plasma concentration as well as an enhanced induction of sedation, but also an enhanced cessation of sedation. In addition, after an intravenous bolus administration of vatinoxan, plasma concentrations of detomidine in horses  and of dexmedetomidine or medetomidine in dogs (Bennett et al., 2016;Honkavaara et al., 2012) and cats (Pypendop et al., 2016) were decreased. Currently, there are no studies investigating the effect of a constant-rate infusion (CRI) of vatinoxan on plasma concentration of dexmedetomidine in horses.
The aim of the study was to determine the effect of a CRI of both dexmedetomidine and vatinoxan on global and peripheral perfusion and oxygenation parameters. Furthermore, a potential influence of vatinoxan on dexmedetomidine plasma concentration was investigated.

| Study design
Prospective, randomized and final experimental study with the observers being aware of the treatment administered. Twelve horses were allocated to two groups by drawing lots: -group DEX (n = 6) | 3 NEUDECK Et al.
In the morning of each experiment and after subcutaneous infiltration of lidocaine (Lidocainhydrochlorid 2%, Bela-Pharm GmbH & Co. KG, Germany), a 12G catheter (Intraflon 2, 12G-80 mm, VYGON GmbH & CO.KG, France) was aseptically placed in the left jugular vein. Additionally, a port system (Exacta,8.5 Fr; Argon Medical Devices Inc., IL, USA) for a pulmonary catheter (Balloon wedge pressure catheter, 7 Fr, 160 cm; Arrow International Inc., NC, USA) was placed aseptically in the right jugular vein. The pulmonary catheter was then inserted into a pulmonary artery under ultrasound guidance and control of location-specific pressure curves.
During anaesthesia, an arterial catheter (Venocan™ PLUS IV Catheter 20G.33 mm, Kruuse A/S, Denemark) was placed in the facial artery to measure the arterial blood pressure and the blood gases. The cardiac output measurements were performed by the lithium dilution technique as described (Linton et al., 2000). The arterial and pulmonary catheters were connected to a fluid-filled, low compliance extension line of a pressure transducer (Argon Safedraw Transducer; Argon Medical Devices Inc., IL, USA), which was placed at the level of the scapulohumeral joint and zeroed to atmospheric pressure.

| Anaesthesia
The vatinoxan powder was weighed individually for each horse and was then dissolved in sterile saline to reach a concentration of 10 mg ml −1 . Group VAT received an intravenous bolus of dexmedetomidine (Dexdomitor ® , 0.5 mg ml −1 , Orion Pharma, Finnland) and vatinoxan (Vetcare, Finland). Simultaneously, a CRI of dexmedetomidine and of vatinoxan was started by using a syringe driver (Braun Perfusor ® Compact pump, BBraun, Germany). In group DEX, a bolus of dexmedetomidine was administered by concurrent start of a CRI of dexmedetomidine, which was also administered by using a syringe driver. In all horses, anaesthesia was induced with 2.2 mg kg −1 ketamine IV (Narketan 100 mg ml −1 ; Vetoquinol GmbH, Germany) and 50 µg kg −1 diazepam IV (Ziapam ® 5 mg ml −1 , Ecuphar, Germany). During the induction and throughout the procedure none of the CRIs used in group DEX or VAT was discon- administered at 5 ml kg −1 h −1 . In case of the decrease of mean arterial blood pressure below 60 mmHg, the infusion rate was increased to 10 ml −1 kg −1 h −1 and a CRI of dobutamine (Dobutamin-ratiopharm ® 250 mg Trockensubstanz, Ratiopharm, Germany) starting at 0.33 µg kg −1 min −1 was initiated. However, in case of a dobutamine requirement, CRI was withheld 5 min before each measurement.

| Cardiovascular and respiratory variables parameters
Heart rate and pulmonary artery pressure variables (systolic pulmonary artery pressure (sPAP), diastolic pulmonary artery pressure (dPAP), and mean pulmonary artery pressure (mPAP)) were obtained before and after sedation. During the anaesthesia, the following cardiovascular and respiratory variables were measured by a multiparameter anaesthesia monitor (GE Datex-Ohmeda S/5 Compact Anästhesie Monitor, Duisburg, Germany): Heart rate (HR), respiratory rate (fR), end-tidal carbon dioxide tension (E T CO 2 ), inspired (F I ISO) and end-tidal (E T ISO) isoflurane concentrations, systolic, diastolic and mean arterial blood pressure (SAP, DAP, MAP, respectively), sPAP, dPAP and mPAP. Before each experiment a 2-point calibration of the monitor was performed according to the expected concentrations (CO 2 5%, O 2 55%, N 2 O 33%, Desflurane 2%; Quick CalTM Calibrationgas, GE Healthcare Finland OY, Finland). Mixed-venous blood gas samples (2 ml) were taken before as well as five minutes after sedation and simultaneously with arterial blood gas samples (2 ml) and cardiac output measurements during anaesthesia. Blood gas samples were withdrawn in heparinised syringes and were analysed by AVL (AVL995, AVL Medizintechnik, Germany) for oxygen partial pressure (PaO 2 and PvO 2 , respectively), carbon dioxide partial pressure (PaCO 2 and PvCO 2 , respectively), pH (pHa and pHv, respectively), haemoglobin and electrolytes.

| Tissue perfusion and oxygenation
Micro-light guide spectrophotometry (O2C, LEA Medizintechnik, Germany) was used to assess the oral and rectal microperfusion in arbitrary units (AU) by laser-Doppler flowmetry and the oxygenation in percentage (%) by white light spectrophotometry as described before (Krug, 2006). On the buccal mucosa and rectal mucosa a flat surface probe (LF-2 probe) was placed with a light penetration depth of 2.5 mm.

| Blood sampling and drug analysis
Venous blood samples (10 ml) were withdrawn from the port system placed in the right jugular vein and transmitted into EDTA tubes.
The plasma was frozen by −80°C until further analysis of the plasma concentrations.
The concentration of dexmedetomidine (reference standard: racemic medetomidine, TRC, Toronto, Ontario, Canada) was determined with a high-performance liquid chromatography/mass spectrometry after a solid-phase extraction (Sep-Pak ® tC18 96-well extraction plates, Waters Co., Milford, MA, USA). After separation with a Gemini ® C18 column (2 × 150 mm, 5 µm, Phenomenex, Torrance, CA, USA) with a gradient flow system (0.1% formic acid in water and methanol), a quantitative detection was performed in multi-reaction monitoring mode with a triple quadrupole mass spectrometer (AB Sciex 4000 QTrap) connected to a high-performance

| Study protocol
The study was conducted every Wednesday for 12 consecutive weeks with one horse per week. The days leading up to the experiments, horses were familiarized with the stock by being fed in them without manipulation. Every Wednesday morning (7 am), horses were brought into a stock, where they were instrumented with the pulmonary artery catheter. Baseline values (B), as well as sedation values (S), were obtained for tissue perfusion (blood flow), oxygenation (sO 2 ) variables, and cardiovascular variables. T0 indicates the application of dexmedetomidine and the start of CRI of dexmedetomidine in group DEX and the application of dexmedetomidine and vatinoxan and the start of a CRI of dexmedetomidine and vatinoxan in group VAT. Thereafter, additional measurements were performed during the anaesthesia. According to the protocol of the concomitant study, further measurements were performed after 60 min of induction (A60), 80 min (A80), 200 min (A200) and 230 min (A230).
One set of measurements consisted of recordings of cardiovascular variables starting with the determination of cardiac output by simultaneous blood gas sampling and was then followed by the tissue perfusion and oxygenation measurements. The latter was always performed by the same person and in the same order, beginning with rectal, then buccal measurements.
The concurrent study consisted of three main episodes including (1) an equilibration and instrumentation time for 90 min, (2) small intestinal ischaemia for 120 min and (3) reperfusion of the small intestine for 30 min.
Horses were euthanized with 60 mg kg −1 pentobarbital (Release ® 500 mg ml −1 , WDT, Germany) at the end of the study.

| Statistical analysis
Data analysis and graph creation were performed using GraphPad software (GraphPad Prism 8, GraphPad Software, USA). The concurrent study determined the number of subjects used in this study.
However, based on data on oxygen delivery from Pakkanen et al., simultaneously to the tissue perfusion assessment were analysed (A60; A80; A200; A230). The reason for that was to reduce the influence of inotropic support on cardiovascular and perfusion parameters which was suspended for the measurement period. Visual assessment of qq-plots and the Shapiro-Wilk test were performed to confirm the normal distribution of model residuals on dependent variables. All parametric data are presented as means ± SD and non-parametric as median with min and max. The statistical analysis of global and tissue parameters was undertaken in terms of a twofactorial variance analysis for repeated measurements. In case of TA B L E 1 Cardiorespiratory data, presented as mean and standard deviation, from twelve horses sedated with either dexmedetomidine 3.5 µg kg −1 followed by a CRI of 7 µg kg −1 h −1 (group DEX) or dexmedetomidine 3.5 µg kg −1 intravenously (IV) followed by a CRI of 7 µg kg −1 h −1 and vatinoxan 130 µg kg −1 IV followed by 40 µg kg −1 h −1 (group VAT) under general anaesthesia with isoflurane

| Cardiovascular and respiratory variables
Due to technical problems, six out of 48 cardiac output (CO) measurements failed. Data for the cardiovascular and respiratory variables and pMAP are depicted in Table 1 and Figure 4, respectively.

| Tissue perfusion and oxygenation
One horse in group DEX did not tolerate the baseline tissue perfusion and the oxygenation measurements, which precluded their further analysis due to the relative nature of the values. However, the cardiovascular and the sedation parameters were obtained. The tissue oxygenation was equally maintained in both groups (Figure 1).

Tissue perfusion significantly differed between baseline and points
in time within the groups and is depicted in Figure 2.

| Plasma concentrations
Due to implausible high plasma concentrations in one horse in group DEX, only five horses of this group were included for analysis. One other horse in this group also had implausible high plasma concentrations after 1 and 3 min. Therefore, these measurements were excluded for further analysis. There was no significant difference in the AUC plasma concentration of dexmedetomidine between both groups ( Table 2). The dexmedetomidine/vatinoxan plasma concentration ratio is depicted in Figure 3.

| Anaesthesia
The period of anaesthesia was similar for all horses. The endexpiratory isoflurane concentration is depicted in Table 1. All horses in group VAT and two horses in group DEX needed inotropic support. The average dobutamine dose was 0.51 ± 0.24 µg kg −1 min −1 (n = 2) over a period of 120 ± 42 min in group DEX and 0.44 ± 0.26 µg kg −1 min −1 (n = 6) over a period of 140 ± 27 min in group VAT (p = 0.14). The cessation of dobutamine infusion before the measurement of cardiovascular and peripheral perfusion parameters led to a decrease in blood pressure resulting in a slight hypotension in both groups (Table 1).

| DISCUSS ION
This is the first study investigating the effect of the co-administration of vatinoxan CRI and dexmedetomidine CRI on global and peripheral hypothesis. The addition of vatinoxan did not alter the plasma concentrations of dexmedetomidine in horses in the current study but increased the requirements for inotropic support to sustain MAP.
The marked hypotension induced by the addition of vatinoxan during isoflurane anaesthesia is related to the alpha-2 adrenoceptorblocking effect and the additional vasodilatory effect of isoflurane (Grosenbaugh & Muir, 1998). Group VAT showed minimally higher CI without reaching significancy and significantly lower SVR values compared to group DEX. In the literature, higher CI is reported when vatinoxan was added to an alpha-2 adrenoceptor agonist, however, an increased dose of dobutamine CRI was used due to more pronounced hypotension in vatinoxan treated horses. As dobutamine CRI was not discontinued before measurements of CI, it is nearly impossible to distinguish between the effects of vatinoxan and dobutamine, respectively (Pakkanen et al., 2015;Tapio, Raekallio, Mykkänen, Al-Ramahi, et al., 2019). Nevertheless, in sedated horses co-administration of vatinoxan to medetomidine significantly increased CI and decreased SVR ten minutes after application . The MAP in our study was similarly sustained in both groups due to the fact that MAP in group VAT was maintained in physiological limits by starting a dobutamine CRI. As dobutamine increases the gastrointestinal perfusion, as well as the cardiac output in a dose-dependent fashion (Dancker et al., 2018), dobutamine was ceased five minutes before measuring the cardiac output and the peripheral perfusion to prevent marked influence. Inotropic support was initiated in case of hypotension due to protocol requirements of the concomitant surgical study. Other potential reasons for the exacerbating hypotension might be seen in the nature of dorsal recumbency (Tapio, Raekallio, Mykkänen, Al-Ramahi, et al., 2019) and IPPV (Araos et al., 2020), since both could have led to a reduced venous return and cardiac output, which again can induce hypotension. Another possible explanation for the hypotension in this study is an endotoxaemia induced by imitating a small intestinal strangulation according to the concurrent study protocol as endotoxins can lead to vasodilation and vasoplegia (Burrows, 1970) as well as reduced cardiac contractility (Brady et al., 1992). The potential endotoxaemia might have influenced the blood pressure especially at the end of the study where reperfusion was allowed. However, this does not explain why a hypotension was present straight from the beginning and during equilibration time.
The slight increase in CI and decrease in SVR also influenced oxygenation variables, illustrated by a moderately higher arterial oxygen tension, significantly lower alveolar dead space and venous admixture in group VAT. The oxygen delivery index was significantly lower and the ERO 2 was significantly higher in group DEX leading to similar VO 2 I compared to values obtained from group VAT. The increase in the ERO 2 in group DEX is most likely responsible for the sustained tissue oxygenation at the buccal and rectal mucosa. This protective mechanism was well-maintained, even though blood flow This observation is in accordance with studies concluding that the bolus administration of alpha-2 adrenoceptor agonists reduce the tissue blood flow by increasing SVR and decreasing CO resulting in a decreased DO 2 I (Edner et al., 2002;Neudeck et al., 2018). By alleviating these effects with the addition of vatinoxan, the tissue blood flow can be promoted. In dogs, the concomitant administration of vatinoxan and dexmedetomidine prevented the dexmedetomidine induced reduction in intestinal blood flow measured by contrast-enhanced ultrasonography (Restitutti et al., 2013). This result is in contrast with a study in horses where vatinoxan decreased the gastrointestinal tissue blood flow 5 min after the bolus administration during anaesthesia, although these horses were markedly hypotensive (MAP 32-43 mmHg). In this study, the MAP was most likely beneath a critical perfusion pressure (MAP <50 mmHg and CI <40 ml kg −1 min −1 ) which is associated with a deteriorated intestinal tissue perfusion . In medetomidine sedated sheep the atipamezole administration failed to restore distributional changes in organ blood flow enhanced by medetomidine (Talke et al., 2000). In this study, the perfusion pressure was well maintained even though slight hypotension occurred. The effect of F I G U R E 3 Plasma concentrations of dexmedetomidine (a), vatinoxan (b) and vatinoxan/dexmedetomidine plasma concentration ratio (c) from eleven horses (DEX n = 5; VAT n = 6) sedated with either dexmedetomidine 3.5 µg kg −1 intravenously (IV) followed by a CRI of 7 µg kg −1 h −1 (group DEX) or dexmedetomidine 3.5 µg kg −1 IV followed by a CRI of 7 µg kg −1 h −1 and vatinoxan 130 µg kg −1 IV followed by 40 µg kg −1 h −1 (group VAT) and under general anaesthesia with isoflurane. Hashes indicate a significant difference (p < 0.05) to baseline values in group VAT The mean pulmonary artery pressure was equally increased after sedation and decreased during anaesthesia between both groups ( Figure 4). Dexmedetomidine, like other alpha-2 adrenoceptor agonists led to a minimal increase in mPAP as reported in ponies (Bettschart-Wolfensberger et al., 2005). Although the baseline values and three points in time differed significantly during anaesthesia in group VAT, there was no distinction amongst the groups.
The decrease in mPAP in group VAT during anaesthesia is most likely attributed to an isoflurane effect (Marshall et al., 1984), also the addition of vatinoxan slightly enhanced this effect, probably due to the adrenolytic effect on pulmonary vessels.
In both groups, the plasma concentrations of dexmedetomidine equally declined for 20 min after administering the initial bolus. Thereafter concentrations rose, reaching a steady-state after 60 min (Figure 3). Two potential scenarios might explain this observation. On the one, hand the induction of anaesthesia 20 min after bolus administration markedly influenced the dexmedetomidine plasma concentrations by increasing volume of distribution, leading to decreased dexmedetomidine plasma concentrations.
Dexmedetomidine potential influence on its own metabolism by decreasing hepatic blood flow might have led to the raised steadystate concentrations during the maintenance of anaesthesia (Dutta et al., 2000). On the other hand, the early distribution phase might have been ended leading to a decrease followed by an increase in dexmedetomidine plasma concentrations. Thus, plasma concentrations of dexmedetomidine were similar between groups. Similar observations have been prepared by Tapio, Raekallio, Mykkänen, Al-Ramahi, et al., (2019) where plasma concentrations of dexmedetomidine were similar in isoflurane anaesthetized horses treated with either medetomidine or medetomidine/vatinoxan. However, this is not in accordance with other reports where the addition of vatinoxan decreased plasma concentrations of various alpha-2 adrenoceptor agonists by alleviating their hemodynamic effects (Bennett et al., 2016;Honkavaara et al., 2012;Pypendop et al., 2016;Vainionpää et al., 2013). A possible reason for this observation might be the influence of the isoflurane concentration masking any difference. Interestingly, the vatinoxan plasma concentration was not affected by the anaesthesia induction and maintenance, which might lead to the assumption that hepatic metabolism and changes in volume distribution do not play a major role.
The 'ideal' dose ratio of medetomidine and vatinoxan was investigated in dogs but not in horses. In dogs under isoflurane anaesthesia a medetomidine/vatinoxan dosing ratio of 1:12.5 and lower <1:18 alleviated hemodynamic effects induced by medetomidine. Tapio, Raekallio, Mykkänen, Al-Ramahi, et al., (2019) used vatinoxan (140 µg kg −1 IV) as a bolus and a similar alpha-2 adrenoceptor agonist drug dose compared to our study, assuming that 7 µg kg −1 IV medetomidine equals 3.5 µg kg −1 IV dexmedetomidine, observing a marked hypotension after anaesthesia induction in vatinoxan treated horses. The same dose combination was investigated in standing horses, where medetomidine induced cardiovascular changes (bradycardia, increased systemic vascular resistance) could be alleviated by vatinoxan, indicating that inhalational anaesthetics will exacerbate vasodilation leading to hypotension . However, the comparability to this study is limited as vatinoxan was not given as CRI. In the present study, we used a slightly lower dose of vatinoxan bolus than previously reported leading to a dose ratio of 1:37 for the bolus. For the CRI of both drugs, we used a higher drug dose ratio with 1:6 based on pharmacokinetic studies in standing horses (Figure 3). In these studies, prevention of bradycardia, hypertension and gastrointestinal hypomotility by vatinoxan were seen when the plasma concentrations were around 100 ng ml −1 Vries et al., 2016). Based on this assumption we calculated our vatinoxan CRI (µg kg −1 h −1 ) with the clearance values obtained by Vries et al., (2016).
In the present study, the vatinoxan plasma concentration actually plateaued at approximately 100 ng ml −1 with a plasma concentration ratio between vatinoxan and dexmedetomidine of 1:20 from 60 to 240 min. Nevertheless, inotropic support was still needed to maintain MAP within acceptable limits. Therefore, one may assume that lower vatinoxan plasma concentrations might be preferable during anaesthesia. However, potential beneficial effects of vatinoxan on cardiovascular variables other than the arterial blood pressure might be reduced by decreasing vatinoxan dosages.
The restrictions imposed by the primary surgical study and the small number of horses are a limitation of the study, which might have contributed to some mitigation of differences in cardiovascular F I G U R E 4 Mean pulmonary artery pressure obtained from eleven horses (DEX n = 5; VAT n = 6) sedated with either dexmedetomidine 3.5 µg kg −1 intravenously (IV) followed by a CRI of 7 µg kg −1 h −1 (group DEX) or dexmedetomidine 3.5 µg kg −1 IV followed by a CRI of 7 µg kg −1 h −1 and vatinoxan 130 µg kg variables induced by vatinoxan. However, the surgical intervention in these horses represents a clinical anaesthesia. Sex distribution was not equal in this study which might lead to higher risk of bias.
However, stratified randomization was not possible due to the unpredictable availability of the horses and, therefore, their sex. The unaccountably high dexmedetomidine plasma concentrations in one horse and in another horse for two measurements most likely resulted from inadvertent use of the wrong catheter, thereby, further limiting the explanatory power of the remaining values. Another limitation is that the CO was measured by the lithium dilution technique (Linton et al., 2000), as it is a feasible and easy method for CO measurements with good accuracy. However, the influence on sensor voltage by various drugs might lead to erroneous results (Ambrisko et al., 2012(Ambrisko et al., , 2013. Dexmedetomidine had only a minor influence on sensor voltage in vitro (Ambrisko et al., 2013) and changes in sensor voltage could not be seen in vivo (Neudeck et al., 2018). The influence of vatinoxan on the lithium dilution technique has not been investigated, however, in the present study the sensor voltage was monitored, and no obvious changes could be seen in group VAT, which indicates an unlikely influence on the accuracy of CO measurements.

| CON CLUS ION
Our pilot study demonstrated peripheral and global perfusion parameters were better sustained in vatinoxan treated horses, indicated by a lower SVR, higher oxygen variables and superior tissue blood flow. Vatinoxan treatment increased the need for inotropic support. Taking these results into account vatinoxan implementation in the anaesthetic protocol can be beneficial for oxygenation and perfusion parameters. Furthermore, the concurrent administration of vatinoxan with dexmedetomidine as bolus followed by CRI in horses did not alter the plasma concentrations of dexmedetomidine.
However, more clinical studies are needed to optimize the dose ratio of dexmedetomidine and vatinoxan to avoid hypotension.

ACK N OWLED G EM ENT
We are grateful for the contribution from Lauri Vuorilehto, Medicine Hannover, Foundation, Hannover, Germany) for advice on the data processing of plasma concentrations.

CO N FLI C T O F I NTE R E S T
The authors have declared that there is no competing interests exist.

A N I M A L WE LFA R E A N D E TH I C S S TATE M E NT
The authors confirm that the ethical policies of the journal, as noted on the journal's author guidelines page, have been adhered to and the appropriate ethical review committee approval has been received.
The authors confirm that they have adhered to European standards for the protection of animals used for scientific purposes Germany (approval number 33.19-42502-04-16/2212; LAVES The Office for Consumer Protection and Food Safety of Lower Saxony).

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.