Hemodynamic effects of dexmedetomidine in isoflurane-anesthetized cats

Veterinary Anaesthesia and Analgesia, 2011, 38, 555 567 doi:10.1111/j.1467-2995.2011.00663.x RESEARCH PAPER Hemodynamic effects of dexmedetomidine in isoflurane-anesthetized cats Bruno H Pypendop*, Linda S Barter*, Scott D Stanley & Jan E Ilkiw* *Department of Surgical and Radiological Sciences, School of Veterinary Medicine, University of California, Davis, CA, USA K. L. Maddy Equine Analytical Chemistry Laboratory, California Health and Food Safety Laboratory, Davis, CA, USA Correspondence: Bruno H Pypendop, Department of Surgical and Radiological Sciences, School of Veterinary Medicine, University of California, One Shields Avenue, Davis, CA 95616, USA. E-mail: bhpypendop@ucdavis.edu Abstract Objective To characterize the hemodynamic effects of dexmedetomidine in isoflurane-anesthetized cats. Study design Prospective experimental study. Animals Six healthy adult female cats weighing 4.6 ± 0.8 kg. Methods Dexmedetomidine was administered intravenously using target-controlled infusions to maintain nine plasma concentrations between 0 and 20 ng ml )1 in isoflurane-anesthetized cats. The isoflurane concentration was adjusted for each dexmedetomidine concentration to maintain the equivalent of 1.25 times the minimum alveolar concentration, based on a previous study. Heart rate, systemic and pulmonary arterial pressures, central venous pressure, pulmonary artery occlusion pressure, body temperature, and cardiac output were measured at each target plasma dexmedetomidine concentration. Additional variables were calculated. Arterial and mixed-venous blood samples were collected for blood gas, ph, and (on arterial blood only) electrolyte, glucose and lactate analysis. Plasma dexmedetomidine concentration was determined for each target. Pharmacodynamic models were fitted to the data. Results Heart rate, arterial ph, arterial bicarbonate concentration, mixed-venous PO 2, mixed-venous ph, mixed-venous hemoglobin oxygen saturation, cardiac index, stroke index, and venous admixture decreased following dexmedetomidine administration. Arterial blood pressure, central venous pressure, pulmonary arterial pressure, pulmonary arterial occlusion pressure, packed cell volume, PaO 2, PaCO 2, arterial hemoglobin concentration, mixed-venous PCO 2, mixed-venous hemoglobin concentration, ionized calcium concentration, glucose concentration, rate-pressure product, systemic and pulmonary vascular resistance indices, left ventricular stroke work index, arterial oxygen concentration, and oxygen extraction increased following dexmedetomidine administration. Most variables changed in a dexmedetomidine concentration-dependent manner. Conclusion and clinical relevance The use of dexmedetomidine as an anesthetic adjunct is expected to produce greater negative hemodynamic effects than a higher, equipotent concentration of isoflurane alone. Keywords blood gas, cardiovascular, cats, dexmedetomidine, isoflurane. Introduction Cats appear to be particularly sensitive to the cardiovascular depressant effects of inhaled anesthetics. For example, a study demonstrated that the cardiac output of healthy cats decreased by more than 50% 555

in response to halothane, whereas in dogs and humans, there was less than a 20% change at similar anesthetic concentrations (Dobkin & Fedoruk 1961; Ingwersen et al. 1988). For this reason, agents that decrease the requirements for inhaled anesthetics are of particular interest in cats. Agonists of alpha-2 adrenoceptors (alpha-2 agonists) have been shown to produce dose-dependent sedation, analgesia, decreased anesthetic requirements and cardiovascular depression (Maze & Tranquilli 1991). Studies in humans and in dogs suggest that low doses of alpha-2 agonists may produce the desired effects of decreased anesthetic requirements and analgesia, with minimal impact on cardiovascular function (Aantaa et al. 1990; Aho et al. 1992; Aantaa et al. 1997; Pascoe et al. 2006). Dexmedetomidine, the active stereoisomer in the racemic medetomidine, is a potent, selective alpha-2 agonist (Aantaa & Scheinin 1993). We have previously reported that dexmedetomidine reduced the minimum alveolar concentration (MAC) of isoflurane in cats, in a plasma concentration-dependent manner (Escobar et al. 2010a). However, limited data are available on the hemodynamic effects of dexmedetomidine in cats. Similarly, data on the cardiovascular effects of medetomidine in cats are sparse, and the dose-effect relationship is poorly characterized (Savola 1989; Golden et al. 1998; Lamont et al. 2001; Selmi et al. 2003). To the authors knowledge, only the effects of dexmedetomidine on heart rate and blood pressure in conscious cats have been reported (Selmi et al. 2003; Monteiro et al. 2009). In addition, the relationship between dexmedetomidine dose or plasma concentration and its cardiovascular effects has not been adequately characterized. The aim of this study was to characterize the hemodynamic effects of dexmedetomidine in isoflurane-anesthetized cats, and to examine the relationship between plasma dexmedetomidine concentration and hemodynamic effects. We hypothesized that dexmedetomidine would produce hemodynamic changes in a plasma concentrationdependent manner. Materials and methods Animals Six healthy adult female spayed domestic shorthair cats weighing 4.6 ± 0.8 kg (mean ± SD) were used in the study. The pharmacokinetics of dexmedetomidine and the MAC of isoflurane without dexmedetomidine, and with the target plasma dexmedetomidine concentrations used in this study had been determined in each cat in previous studies (Escobar et al. 2010a,b). Food was withheld from cats for 12 hours before the studies, but water was provided ad libitum. This study was approved by the institutional animal care and use committee at the University of California, Davis. Instrumentation and measurements Anesthesia was induced in an acrylic chamber using isoflurane in oxygen. After endotracheal intubation with a cuffed endotracheal tube, anesthesia was maintained with isoflurane in oxygen delivered via a coaxial Mapleson F circuit. Oxygen flow was set at 200 ml kg )1 minute )1. Ventilation was spontaneous throughout the study. A catheter was passed through the lumen of the endotracheal tube so that its tip was positioned close to the distal end of the tube. This catheter was connected to a Raman spectrometer (Ohmeda, UT, USA) for continuous measurement of end-tidal carbon dioxide and isoflurane concentrations. A 22-gauge, 25 mm catheter was placed in a medial saphenous vein for administration of lactated Ringer s solution at 3 ml kg )1 hour )1. A 5-F, 7.5-cm introducer (Introducer kit; Arrow International, PA, USA) was placed in a jugular vein. A 4-F, 75-cm thermodilution catheter (Thermodilution balloon catheter; Arrow International) was inserted through the introducer and positioned under fluoroscopic guidance with its distal port and thermistor in the pulmonary artery. This catheter was used for measurement of cardiac output, pulmonary arterial pressure, pulmonary arterial occlusion pressure, central venous pressure, and core body temperature and collection of mixed-venous blood samples (samples from the pulmonary artery). A 24- gauge, 9-cm catheter (Central venous catheterization kit; Arrow International) was inserted into a femoral artery using a surgical approach and a Seldinger technique; this catheter was used for measurement of arterial pressure and collection of arterial blood samples. Cats were placed in lateral recumbency. The electrocardiogram (lead II), heart rate, systolic arterial pressure, diastolic arterial pressure, mean arterial pressure, pulmonary artery pressure, and central venous pressure were continuously measured and recorded by use of a physiograph 556 Ó 2011 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesiologists, 38, 555 567

(Gould Instrument Systems, OH, USA) and acquisition software (Ponemah, version 4.7; Data Sciences International, MN, USA). All pressure transducers were calibrated against a mercury manometer before each experiment, and a value of zero was established at the level of the sternum. Prior to each measurement, expired gases were collected manually (10 ml collected during 4 10 respiratory cycles) for end-tidal isoflurane concentration (E Iso) determination, using an infrared analyzer (Medical Gas Analyzer LB1; Beckman Instruments, IL, USA). The analyzer was calibrated prior to each study with standards of known isoflurane concentration (0.5%, 1.5%, 2%, and 3%) and room air. This measurement was performed in triplicate and the mean value was calculated. Arterial and mixedvenous ph, PCO 2 and PO 2 were measured and bicarbonate concentration was calculated by use of a blood gas and electrolyte analyzer (ABL 700; Radiometer, Denmark) that corrected measurements on the basis of body temperature. In addition, blood sodium, potassium, chloride, ionized calcium, glucose, and lactate concentrations were measured in the arterial blood using the same analyzer. Arterial and mixed-venous blood hemoglobin concentration and hemoglobin oxygen saturation were measured by use of a hemoximeter (OSM3; Radiometer). Packed cell volume and total protein concentration were measured in arterial blood samples by microcentrifugation and refractometry, respectively. Dexmedetomidine concentration was measured in venous plasma. Cardiac output was determined, in triplicate, by use of a thermodilution technique and a cardiac output computer (COM-1; American Edwards Laboratories, CA, USA). Three ml of iced 5% dextrose were injected through the proximal port of the thermodilution catheter for each determination. The mean value of the three measurements was then calculated. Core body temperature was maintained between 38 and 39 C throughout the study by use of warm-water circulating blankets and forced-air blankets, as needed. Cardiac index, stroke index, rate-pressure product, systemic vascular resistance index, pulmonary vascular resistance index, left and right ventricular stroke work indices, arterial and mixed-venous oxygen concentrations, oxygen concentration, oxygen delivery, oxygen consumption, oxygen extraction ratio, alveolar-to-arterial PO 2 difference, and venous admixture were calculated by use of standard equations (Darovic 1995; Stoelting 1999; Lumb 2000). Experimental protocol After instrumentation, the E Iso was set at 1.25 times the individual s MAC that had been determined for each cat in a previous study (Escobar et al. 2010b). Dexmedetomidine hydrochloride (Dexdomitor; Pfizer Animal Health, NY, USA) was administered intravenously to target pseudo-steady state plasma concentrations of 0, 0.16, 0.31, 0.63, 1.25, 2.5, 5, 10, and 20 ng ml )1, using a target controlled infusion system consisting of a syringe pump (PHD 2000; Harvard Apparatus, MA, USA) and computer program (Rugloop I; Demed, Belgium). With this system, the central compartment was rapidly loaded to the desired concentration. The infusion rate was then updated every 10 seconds, as needed to maintain pseudo steady-state plasma concentrations, according to the equation: R ¼ C T V 1 ðk 10 þ k 12 e k21t Þ where R is the infusion rate, C T is the target plasma concentration, V 1 is the volume of the central compartment, t is the time, and k 10, k 12, and k 21 are the microrate constants. Individual pharmacokinetic data that had been determined for each cat in a previous study were used (Escobar et al. 2010b). Target plasma dexmedetomidine concentrations were administered in an increasing order to minimize experimental time. E Iso was reduced at each increasing dexmedetomidine target plasma concentration, in order to maintain equipotency at 1.25 MAC, according to the results of a previous study (Escobar et al. 2010a). Twenty minutes were allowed after each change of E Iso and target plasma dexmedetomidine concentrations for conditions to equilibrate. Samples of gas were then manually collected for measurement of E Iso. Heart rate, systolic, diastolic, and mean arterial pressures, central venous pressure, mean pulmonary artery pressure, and pulmonary artery occlusion pressure were recorded. Samples (1 ml) of arterial and mixed-venous blood were collected and immediately placed on ice until analyzed, which occurred within 15 minutes of collection. In addition, a sample of venous blood (2 ml) was collected from the proximal port of the thermodilution catheter. The blood was transferred to a tube containing ethylenediaminetetraacetic acid, immediately centrifuged at 3901 g for 10 minutes at 4 C, the plasma collected and frozen at )20 C for later dexmedetomidine concentration determination. Cardiac output was measured, the isoflurane and dexmedetomidine concentrations were adjusted, and measurements were repeated as Ó 2011 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesiologists, 38, 555 567 557

described above until the last target plasma dexmedetomidine concentration had been studied. At the end of the experiment, the thermodilution catheter and introducer were removed, and a compressive bandage applied over the jugular vein for 15 minutes. The catheter in the femoral artery was removed, and the femoral artery and skin were sutured. Cats were allowed to recover. Meloxicam (0.2 mg kg )1 SC) was administered prior to recovery. Cats were observed continuously until they were able to walk, and once daily for 10 days for detection of complications. Plasma dexmedetomidine concentration determination Dexmedetomidine was quantitated in feline plasma by liquid chromatography-mass spectrometry analysis of protein-precipitated samples. The calibration standards were prepared as follows: stock solutions were made by dissolving 1.0 mg of dexmedetomidine hydrochloride standard (Orion Pharma, Finland) in 1.0 ml of methanol. Working solutions were prepared by dilution of the dexmedetomidine stock solution with methanol to concentrations of 10,000, 100 and 1 ng ml )1. Plasma calibrators were prepared by dilution of the working dexmedetomidine solutions with drug free feline plasma to concentrations of 0.25, 0.5, 1.0, 5.0, 10.0, 20.0, 50.0 and 100.0 ng ml )1. Calibration curves and negative control samples were prepared fresh for each quantitative assay. In addition, quality control samples (plasma fortified with analytes at concentrations mid-point of the standard curve) were routinely included as an additional check of accuracy. The concentration of dexmedetomidine in each sample was determined by the internal standard (detomidine-d3) (Frontier BioPharma, KY, USA) method using the peak area ratio and linear regression analysis. Quantitative analyses were performed on a triple quadrupole mass spectrometer (TSQ Vantage; Thermo Scientific, CA, USA) equipped with a heated electrospray ionization probe that was kept at 375 C. All analyses were performed in the positive ionization mode with a spray voltage set at 4000 V. The sheath, auxiliary and ion sweep gas used was nitrogen at 45, 30 and 2 arbitrary units, respectively. The system was operated in the selected reaction monitoring mode with argon as the collision gas at a pressure of 1.5 mtorr. The ion transfer tube was kept at 300 C while the scan time and width were 0.065 seconds and 0.1 mass to charge ratio, respectively. Data were processed using LCQuan software version 2.6 (Thermo Scientific). The triple quadrupole mass spectrometer was coupled with a turbulent flow chromatography system (Aria TLX-4; Thermo Fisher Scientific, MA, USA). The columns used (50 0.5 mm, 60 lm particle size, 60 Å pore size) were connected in tandem. Chromatographic separation employed a 100 2.1 mm, 3 lm, column (ACE C 18 ; Mac Mod, PA, USA) and a linear gradient of acetonitrile in water with a constant 0.2% formic acid at a flow rate of 0.35 ml minute )1. The acetonitrile concentration was held at 10% for 0.3 minutes, ramped up to 95% over 5.0 minutes. Prior to analysis, the plasma proteins, controls and calibrators were extracted by precipitation with 0.5 ml 9:1 acetonitrile:1 M acetic acid containing 10 ng ml )1 of internal standard, vortex mixed for 2.0 minutes, refrigerated for 30.0 minutes, followed by centrifugation (1864 g 3 minutes). The injection volumes were 50.0 ll. Detection and quantitation employed full scan liquid chromatography-mass spectrometry/mass spectrometry transitions of initial product ions for dexmedetomidine (mass to charge ratio 201.1). The response for the major product ion for dexmedetomidine (mass to charge ratio, 95.1) was plotted and peaks at the proper retention time integrated using LCQuan software. The concentration of dexmedetomidine in each sample (e.g., calibrators, quality control and unknowns) was determined by an internal standard method using the peak area ratio and linear regression analysis. The response for dexmedetomidine was linear and gave correlation coefficients (R 2 ) of 0.99 or better. The technique was optimized to provide a limit of quantitation at 0.1 ng ml )1. Values lower than the limit of quantitation were not used for analysis. Accuracy (percentage of nominal concentration) was 95% and 105% for 1 and 10 ng ml )1, respectively. Imprecision (percentage relative standard deviation) was 7% and 3% for 1 and 10 ng ml )1, respectively. Pharmacodynamic analysis Nonlinear least squares regression (WinNonlin 6.1; Pharsight, NC, USA) was performed on dexmedetomidine concentration-hemodynamic data. Simple and sigmoid effect maximum models with baseline effect were fitted to data demonstrating an increase 558 Ó 2011 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesiologists, 38, 555 567

following dexmedetomidine administration. Simple and sigmoid inhibitory maximum models were fitted to data demonstrating a decrease following dexmedetomidine administration. Model equations were: E c ¼ E 0 þ ðe max CÞ ðec 50 þ CÞ E c ¼ E 0 þ ðe max C c Þ ðec c 50 þ Cc Þ E c ¼ E 0 ði max CÞ ðic 50 þ CÞ E c ¼ E 0 ði max C c Þ ðic c 50 þ Cc Þ for the simple effect maximum, sigmoid effect maximum, simple inhibitory maximum, and sigmoid inhibitory maximum models, respectively, where E C is the effect at plasma dexmedetomidine concentration C, E 0 is the baseline effect (i.e. the effect of isoflurane alone), E max and I max are the maximum possible increase (Excitatory) or decrease (Inhibitory) in effect, respectively, EC 50 and IC 50 are the plasma dexmedetomidine concentration at which 50% of E max or I max, respectively, are produced, and c is the sigmoidicity factor. Observation of the residuals plot and use of Akaike s information criterion were used to select which model fitted the data best (Yamaoka et al. 1978). Statistical analysis A two-tailed paired t-test was used to compare baseline values (i.e. values without dexmedetomidine) and values at the highest plasma dexmedetomidine concentration, to determine if significant changes related to dexmedetomidine administration had occurred for each variable. Significance was set at p < 0.05. Pharmacodynamic analysis was only performed on the variables for which the t-test reached significance. Normal distribution of pharmacodynamic parameters was verified using the Shapiro Wilk test. Pharmacodynamic parameters are presented as weighted mean ± SE, unless specified otherwise. To improve the precision of the mean estimates, individual parameters used to calculate means were weighted by the reciprocal of the variance obtained by the non-linear regression procedure (Cooper & Weekes 1983). Plasma dexmedetomidine concentration at each target concentration in the six cats is reported as median (range). End-tidal isoflurane concentration at each target plasma dexmedetomidine concentration is reported as mean ± SD. The ratio of target to actual isoflurane concentration was pooled for all cats at all plasma dexmedetomidine concentrations, and is presented as mean ± SD. Baseline values are presented as mean ± SD. Results Plasma dexmedetomidine concentrations were 0 (0 0), 0 (0 0), 0.07 (0 0.28), 0.21 (0.15 0.25), 0.32 (0.18 0.37), 0.84 (0.52 1.49), 2.98 (1.36 3.96), 8.64 (7.57 13.69), and 21.40 (19.86 30.63) ng ml )1 for the 0, 0.16, 0.31, 0.63, 1.25, 2.5, 5, 10, and 20 ng ml )1 target plasma dexmedetomidine concentrations, respectively. Endtidal isoflurane concentration was 2.58 ± 0.17, 2.31 ± 0.18, 2.20 ± 0.21, 2.18 ± 0.16, 1.93 ± 0.20, 1.74 ± 0.30, 1.12 ± 0.43, 0.85 ± 0.41, and 0.60 ± 0.25% for the 0, 0.16, 0.31, 0.63, 1.25, 2.5, 5, 10, and 20 ng ml )1 target plasma dexmedetomidine concentrations, respectively. Target/ actual isoflurane concentration was 1.00 ± 0.04. Baseline values for all study variables are presented in Table 1. Baseline values and values measured at the highest plasma dexmedetomidine concentration were not significantly different for body temperature, total protein concentration, arterial hemoglobin oxygen saturation, mixedvenous bicarbonate concentration, blood sodium, chloride and potassium concentrations, blood lactate concentration, right ventricular stroke work index, mixed-venous oxygen concentration, oxygen delivery, oxygen consumption, and alveolar-to-arterial PO 2 difference. Heart rate (Fig. 1), arterial ph, arterial bicarbonate concentration, mixed-venous PO 2, mixed-venous ph, mixed-venous hemoglobin oxygen saturation, cardiac index (Fig. 2), stroke index, and venous admixture decreased following dexmedetomidine administration. Systolic arterial pressure, diastolic arterial pressure, mean arterial pressure (Fig. 3), central venous pressure, pulmonary arterial pressure, pulmonary artery occlusion pressure, packed cell volume, PaO 2, PaCO 2, arterial hemoglobin concentration, PvCO 2, mixed-venous hemoglobin concentration, blood ionized calcium concentration, blood glucose concentration, ratepressure product, systemic vascular resistance index (Fig. 4), pulmonary vascular resistance index, left ventricular stroke work index, arterial oxygen Ó 2011 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesiologists, 38, 555 567 559

Table 1 Mean ± SD baseline values (i.e. values measured before administration of dexmedetomidine) for the study variables, obtained in six cats Variable Mean ± SD Variable Mean ± SD HR 172 ± 21 Hbv 10.6 ± 1.3 SAP 87 ± 10 Sodium 150 ± 0.5 DAP 54 ± 7 Potassium 3.9 ± 0.4 MAP 67 ± 7 Chloride 124 ± 3 CVP 8 ± 3 Calcium 1.23 ± 0.13 MPAP 18 ± 2 Glucose 123 ± 25 PAOP 12 ± 3 Lactate 1.2 ± 0.1 Temperature 38.4 ± 0.3 CI 1.48 ± 0.43 PCV 27 ± 6 SI 0.52 ± 0.13 TP 5.0 ± 0.5 RPP 14,982 ± 3044 PaO 2 457 ± 55; 61 ± 7 SVRI 3300 ± 653 PaCO 2 40.4 ± 3.7; 5.4 ± 0.5 PVRI 345 ± 97 pha 7.321 ± 0.048 LVSWI 0.387 ± 0.160 Arterial HCO 3 19.8 ± 0.7 RVSWI 0.070 ± 0.034 SaO 2 99.8 ± 0.3 CaO 2 16.0 ± 1.6 Arterial Hb 10.9 ± 1.1 Cv O 2 12.0 ± 2.4 Pv O 2 73 ± 21; 9.7 ± 2.8 DO _ 2 63.5 ± 17.1 Pv CO 2 46.4 ± 3.9; 6.2 ± 0.5 V_ O 2 15.1 ± 3.2 phv 7.288 ± 0.044 O 2 extraction 0.25 ± 0.08 Mixed-venous HCO 3 21.0 ± 0.9 P(A-a)O 2 216 ± 56; 28.8 ± 7.5 Sv O 2 82.3 ± 6.6 Q_ s = Q _ t 0.15 ± 0.05 HR: heart rate (bpm); SAP: systolic arterial pressure (mmhg); DAP: diastolic arterial pressure (mmhg); MAP: mean arterial pressure (mmhg); CVP: central venous pressure (mmhg); MPAP: mean pulmonary artery pressure (mmhg); PAOP: pulmonary artery occlusion pressure (mmhg); Temperature: body temperature ( C); PCV: packed cell volume (%); TP: total plasma protein concentration (g dl )1 ); PaO 2 : arterial partial pressure of oxygen (mmhg; kpa); PaCO 2 : arterial partial pressure of carbon dioxide (mmhg; kpa); Arterial HCO 3 : arterial bicarbonate concentration (meq L )1 ); SaO 2 : arterial hemoglobin oxygen saturation (%); Arterial Hb: arterial hemoglobin concentration (g dl )1 ); Pv O 2 : mixed-venous partial pressure of oxygen (mmhg; kpa); Pv CO 2 : mixed-venous partial pressure of carbon dioxide (mmhg; kpa); Mixed-venous HCO 3 : mixed-venous bicarbonate concentration (meq L)1 ); Sv O 2 : mixed-venous hemoglobin oxygen saturation (%); Mixed-venous Hb: mixed-venous hemoglobin concentration (g dl )1 ); Sodium: arterial sodium concentration (mm); Potassium: arterial potassium concentration (mm); Chloride: arterial chloride concentration (mm); Calcium: arterial ionized calcium concentration (mm); Glucose: arterial glucose concentration (mg dl )1 ); Lactate: arterial lactate concentration (mm); CI: cardiac index (L minute )1 m )2 ); SI: stroke index (ml beat )1 kg )1 ); RPP: rate-pressure product (bpm mmhg); SVRI: systemic vascular resistance index (dynes seconds cm )5 m )2 ); PVRI: pulmonary vascular resistance index (dynes seconds cm )5 m )2 ); LVSWI: left ventricular stroke work index (cj beat )1 kg )1 ); RVSWI: right ventricular stroke work index (cj beat )1 kg )1 ); CaO 2 : arterial oxygen concentration (ml dl )1 ); Cv O 2 : mixed-venous oxygen concentration (ml dl )1 ); D O 2 : oxygen delivery (ml minute )1 ); V_ O 2 : oxygen consumption (ml minute )1 ); O 2 extraction: oxygen extraction ratio; P(A-a)O 2 : Alveolar-to-arterial difference in partial pressure of oxygen (mmhg; kpa); Q_ s = Q _ t : venous admixture. concentration, and oxygen extraction ratio increased following dexmedetomidine administration. Pharmacodynamic parameters are presented in Tables 2 and 3. For most variables the models tested did not adequately fit the data from all cats. The models tested adequately fitted the data from two cats or less for pulmonary artery occlusion pressure, PaO 2, PaCO 2, arterial ph, arterial bicarbonate concentration, PvO 2,PvCO 2, blood calcium concentration, rate-pressure product, stroke index, and venous admixture. Pharmacodynamic parameters are not presented for these variables, but the median (range) maximum observed change is presented in Table 4. Discussion In this study, intravenous administration of dexmedetomidine produced hemodynamic effects in isoflurane-anesthetized cats, in a plasma concentration-dependent manner. The main hemodynamic effects related to dexmedetomidine administration were a decrease in heart rate and cardiac index, and an increase in systemic vascular resistance index. This is in agreement with findings in other species, both in the conscious and anesthetized states (Lawrence et al. 1996, 1997; Bettschart-Wolfensberger et al. 2005; Kutter et al. 2006; Rioja et al. 560 Ó 2011 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesiologists, 38, 555 567

Figure 1 Observed (symbols) and modeled (lines) changes in heart rate with increasing plasma dexmedetomidine concentration in five cats anesthetized with isoflurane. Modeled effects were obtained by the use of a sigmoid inhibitory maximum model. Figure 2 Observed (symbols) and modeled (lines) changes in cardiac index with increasing plasma dexmedetomidine concentration in six cats anesthetized with isoflurane. Modeled effects were obtained by the use of a sigmoid inhibitory maximum model. Figure 3 Observed (symbols) and modeled (lines) changes in mean arterial pressure with increasing plasma dexmedetomidine concentration in four cats anesthetized with isoflurane. Modeled effects were obtained by the use of a sigmoid effect maximum model. Figure 4 Observed (symbols) and modeled (lines) changes in systemic vascular resistance index with increasing plasma dexmedetomidine concentration in five cats anesthetized with isoflurane. Modeled effects were obtained by the use of a sigmoid effect maximum model. 2006). While the effects of dexmedetomidine on heart rate and blood pressure in conscious cats have been reported (Selmi et al. 2003; Monteiro et al. 2009), to the authors knowledge, neither the effects on global hemodynamics, nor the relationship between dose or plasma concentration and effects have previously been reported. The results of this study should be interpreted in view of five main limitations. 1) The hemodynamic effects of dexmedetomidine were studied in isoflurane-anesthetized cats. While the results are therefore relevant to the use of dexmedetomidine as an anesthetic adjunct, which is the context addressed by this and our previous two studies on dexmedetomidine in cats (Escobar et al. 2010a,b), the effects reported here should not be extrapolated to the use of dexmedetomidine alone. Indeed, these effects are due to the interaction between isoflurane and dexmedetomidine, and while the direction of most changes observed in the present study would likely be identical with dexmedetomidine alone, some effects may be absent without the addition of isoflurane, and the magnitude of the effects would likely be different in conscious cats. Isoflurane dosedependently produces cardiovascular effects (Hodgson et al. 1998), some of which (e.g. effect on cardiac output) are similar to those of dexmedetomidine and are therefore likely to augment them, while others (e.g. effect on systemic vascular resistance) are opposite to those of dexmedetomidine and may therefore decrease them. 2) Dexmedetomidine was administered intravenously using a targetcontrolled infusion technique. Although this technique has been reported for the administration of dexmedetomidine to human patients (Fragen & Ó 2011 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesiologists, 38, 555 567 561

Table 2 Weighted mean ± SE pharmacodynamic parameters for variables that increased following dexmedetomidine administration Variable E 0 EC 50 E max Gamma N SAP 87 ± 6 1.12 ± 0.09 102 ± 17 1.45 ± 0.42 4 DAP 57 ± 3 1.12 ± 0.10 80 ± 10 1.71 ± 0.37 4 MAP 73 ± 4 1.21 ± 0.09 86 ± 12 1.74 ± 0.37 4 CVP 10 ± 2 0.73 ± 0.60 5 ± 0.4 4 PAP 17 ± 1 1.99 (0.88 6.96)* 8 (7 12)* 1.76 ± 0.28 5 PCV 32 ± 3 1.29 ± 0.38 10 ± 3 3.28 ± 0.47 4 Arterial Hb 9.7 (9.3 12.0)* 1.46 ± 0.72 4.2 ± 0.6 1.31 ± 0.30 5 Mixed-venous Hb 9.9 ± 0.5 1.32 ± 0.54 4.9 ± 0.4 1.25 ± 0.20 5 Glucose 151 ± 23 0.26 ± 0.19 168 ± 28 3 SVRI 3062 ± 457 1.21 ± 0.45 10,770 ± 2577 1.67 ± 0.35 5 PVRI 429 ± 74 1.29 ± 2.03 987 ± 92 3 LVSWI 0.343 ± 0.026 0.61 ± 0.23 0.536 ± 0.060 1.27 ± 1.46 3 CaO 2 14.3 (13.7 17.4)* 1.33 ± 0.59 5.8 ± 0.8 1.44 ± 0.27 5 O 2 extraction 0.26 ± 0.05 0.83 ± 0.03 0.08 ± 0.03 1.63 ± 0.36 3 *Values reported as mean (range) because they were not normally distributed. E 0 : effect in the absence of dexmedetomidine (unit of the variable); EC 50 : plasma dexmedetomidine concentration at which 50% of E max is produced (ng ml )1 ); E max : maximum possible increase in effect (unit of the variable); Gamma: sigmoidicity factor. Missing gamma values indicate that a simple effect maximum model best fitted the data for that variable. N is the number of cats in which the model fitted the data adequately. E 0 values may be different from baseline values (Table 1) because they were estimated by the models. See Table 1 for remainder of key. Table 3 Weighted mean ± SE pharmacodynamic parameters for variables that decreased following dexmedetomidine administration Variable E 0 IC 50 I max Gamma N HR 165 ± 11 0.75 ± 0.13 47 ± 7 1.34 ± 0.24 5 CI 1.49 ± 0.19 0.45 ± 0.54 0.55 (0.36 1.57)* 2.24 (1.65 10.00)* 6 Mixed-venous ph 7.290 ± 0.018 0.90 ± 0.27 0.080 ± 0.006 3 SvO 2 80.5 (80.4 92.4)* 0.37 ± 0.15 8.1 ± 4.3 1.19 ± 0.82 3 *Values reported as mean (range) because they were not normally distributed. E 0 : effect in the absence of dexmedetomidine (unit of the variable); IC 50 : plasma dexmedetomidine concentration at which 50% of I max is produced (ng ml )1 ); I max : maximum possible decrease in effect (unit of the variable); Gamma: sigmoidicity factor. The missing gamma value for mixed-venous ph indicates that a simple inhibitory maximum model best fitted the data for that variable. N is the number of cats in which the model fitted the data adequately. E 0 values may be different from baseline values (Table 1) because they were estimated by the models. See Table 1 for remainder of key. Fitzgerald 1999; Tanskanen et al. 2006; Hammer et al. 2009; Kunisawa et al. 2009, 2010; Tsai et al. 2010), it is not representative of current clinical veterinary practice. In this study, target-controlled infusions were used in order to rapidly establish and maintain stable plasma concentrations. Use of a continuous rate infusion, while more representative of current clinical practice, is indeed expected to result in changing plasma concentrations for a long time, which would unduly prolong experimental time. 3) With the exception of the two highest targeted concentrations, actual plasma dexmedetomidine concentrations were lower than the corresponding target concentration in all cats. At the two highest targeted concentrations, actual plasma dexmedetomidine concentration was equal or higher than the corresponding target in two and five cats, respectively. Target-controlled infusions are highly dependent on pharmacokinetics; therefore, actual concentrations will be close to target concentrations only if the model used is a good predictor of the disposition of the drug. In this study, we used individual pharmacokinetics in an attempt to limit variability due to individual differences in dexmedetomidine disposition. However, the pharmacokinetic model used was derived from intravenous administration of 10 lg kg )1 of dexmedetomidine as a 5 minute infusion in 562 Ó 2011 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesiologists, 38, 555 567

Table 4 Median (range) maximum observed change from baseline and maximum observed change indexed to baseline measurements following dexmedetomidine administration for selected variables Variable Direction of change Maximum observed change Maximum observed change (% baseline measurement) PAOP Increase 4 (2 8) 33 (13 86) PaO 2 Increase 72 (1 129); 10 (0.1 17) 15 (0.2 36) PaCO 2 Increase 2.8 (0 7.5); 0.4 (0 1.0) 7 (0 19) Arterial ph Decrease 0.064 (0.047 0.108) 0.9 (0.6 1.5) Arterial HCO 3 Decrease 2.1 (1.7 3.8) 11 (8 18) Pv O 2 Decrease 11 (0 50); 1 (0 7) 17 (0 44) Pv CO 2 Increase 10.3 (4 12.4); 1.4 (0.5 1.7) 23 (9 27) Calcium Decrease 0.2 (0.09 0.4) 17 (8 27) RPP Increase 9478 (1994 16,560) 63 (14 145) SI Decrease 0.14 (0.04 0.44) 29 (8 57) Qs= _ Qt _ Decrease 0.10 (0.05 0.16) 66 (50 73) See Table 1 for key. isoflurane-anesthetized cats. It is likely that the disposition of dexmedetomidine is plasma concentration-dependent, due to the dose-dependent effect of alpha-2 agonists on the cardiovascular system reported in the literature and observed in this study (Lawrence et al. 1996; Golden et al. 1998; Pypendop & Verstegen 1998; Talke et al. 2000; Lamont et al. 2001). A single dose pharmacokinetic study may therefore not allow adequate characterization of the disposition of the drug over a wide dose range. Hydroxylation in the liver has been reported to be the main pathway for dexmedetomidine elimination (Salonen 1991), and it is likely that liver blood flow decreases following administration of this drug (Lawrence et al. 1996). The effect of dexmedetomidine-induced decrease in cardiac output on its pharmacokinetics has been reported in humans (Dutta et al. 2000). Of note, actual plasma dexmedetomidine concentrations in this study were close to or even higher than their respective targets for the two highest target concentrations, which are close to the peak concentration seen in the pharmacokinetic study. The pharmacokinetic model therefore appears appropriate for predicting the disposition of dexmedetomidine when high, but not low plasma concentrations, are produced. An additional factor likely contributing to the underestimation of clearance, and possibly volume of distribution, of dexmedetomidine in the pharmacokinetic study, is the arbitrary decision to maintain isoflurane concentration at 0.7 MAC without dexmedetomidine. Neither the pharmacokinetics of dexmedetomidine nor its effect on the minimum alveolar concentration of isoflurane had been previously reported in cats, precluding better adjustment of isoflurane concentration. The peak plasma dexmedetomidine concentration in that study was approximately 18 ng ml )1 ; according to our study on the effects of dexmedetomidine on the MAC of isoflurane (Escobar et al. 2010a), that dexmedetomidine concentration in combination with the selected isoflurane concentration represents the equivalent of approximately 3.5 MAC. Isoflurane concentration was therefore likely above the equivalent of 1.25 times the minimum alveolar concentration, the concentration used in the present study, for a large part of the pharmacokinetic study. Isoflurane has been shown to affect drug pharmacokinetics in cats, resulting in decreased clearance and volume of distribution (Thomasy et al. 2005; Pypendop et al. 2008), and it is likely that the effect is dose-dependent. Therefore, the higher relative isoflurane concentration in the pharmacokinetic study would likely have resulted in underestimation of clearance and volume of distribution of dexmedetomidine at the lower isoflurane concentration used in this study, and this may have contributed in the resulting overall lower than expected plasma dexmedetomidine concentrations. 4) In addition to the lower than targeted concentrations, there was a large variability in plasma dexmedetomidine concentrations between individuals, as illustrated by their wide range at each target. This is similar to studies on ketamine in dogs (Boscan et al. 2005; Solano et al. 2006), but contrary to studies on lidocaine and gabapentin in cats (Pypendop & Ilkiw 2005b,a; Reid et al. 2010) conducted in our laboratory using similar methods. The large variability Ó 2011 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesiologists, 38, 555 567 563

in actual plasma dexmedetomidine concentration at each target among individuals precludes the direct comparison of these dexmedetomidine concentrations using inferential statistics. However, it does not affect pharmacodynamic modeling, since the actual concentrations in each individual were used for that analysis. Nevertheless, the models tested did not adequately fit the data from all cats for all variables. In most cases, when an adequate fit could not be obtained, it was related to a very small magnitude of change in that particular individual, or to the lack of a ceiling effect. With the latter, neither E max or I max, nor EC 50 or IC 50 could be determined. The lack of a ceiling effect is most likely due to the fact that the plasma dexmedetomidine concentration was not high enough for the effect to plateau. The range of plasma dexmedetomidine concentrations was however wide enough for complete characterization of the concentration-effect relationship for most variables of primary interest. 5) Lastly, while an effort was made to maintain equipotent anesthetic concentrations by adjusting isoflurane concentration in relation to the expected plasma dexmedetomidine concentration based on a previous study conducted in the same cats, and in which the same methods of dexmedetomidine administration were used (Escobar et al. 2010a), plasma dexmedetomidine concentrations in the present study tended to be higher than in the previous study, particularly at the higher plasma concentrations. Our baseline values are therefore truly 1.25 times the individual s MAC of isoflurane, while at the high target plasma dexmedetomidine concentrations, cats received more than 1.25 MAC equivalent. Although this is unlikely to have changed the overall shape of the plasma concentration-effect relationship, direct comparison with baseline effects should be restricted to plasma dexmedetomidine concentrations of 0.4 ng ml )1 or less, since up to that value, very similar concentrations were produced in both the MAC study and the study reported here. Dexmedetomidine decreased heart rate in a concentration-dependent manner. Alpha-2 agonists have been shown to cause decreases in heart rate by increasing baroreceptor responses to increases in blood pressure, decreasing sympathetic tone, and/or increasing parasympathetic tone (Badoer et al. 1983; Harron et al. 1985; Devcic et al. 1994; Xu et al. 1998; Vayssettes-Courchay et al. 2002; Penttila et al. 2004). In this study, the model predicts that, on average, the lowest heart rate observed at high dexmedetomidine concentration would be approximately 118 beats minute )1, which would be considered mild to moderate bradycardia in anesthetized cats. Cardiac index decreased with increasing plasma dexmedetomidine concentration. This change appears in large part related to the effect on heart rate. However, stroke index also decreased, although the changes could not be modeled. Direct effects of alpha-2 agonists on myocardial contractility are unlikely (Flacke et al. 1990); the decrease in stroke index is most likely related to a combination of increase in afterload, due to the increase in systemic vascular resistance, and decreased sympathetic tone (Bloor et al. 1992). Systemic and pulmonary vascular resistance indices increased due to the effect of alpha-2 agonists on the vascular smooth muscle (Hyman & Kadowitz 1985; Ruffolo 1985; Duka et al. 2000; Talke et al. 2001; Willems et al. 2001; Gornemann et al. 2007, 2009). This vasoconstrictive effect leads to the increase in systemic, and likely pulmonary, arterial blood pressure (Bloor et al. 1992). Central venous pressure and pulmonary arterial occlusion pressure increased, most likely as a result of decreased cardiac output (Sheriff et al. 1993). Rate-pressure product is an indirect estimate of myocardial oxygen consumption (White 1999). At the highest dexmedetomidine concentration, rate-pressure product was higher than baseline, although the plasma concentration-effect relationship could not be modeled. The increase indicates that the relative increase in systolic blood pressure exceeds the relative decrease in heart rate. Since myocardial oxygen consumption is linked to myocardial work, increased myocardial oxygen consumption is also suggested by the increase in myocardial work illustrated by the changes in left ventricular stroke work index. The importance of this increase in myocardial oxygen consumption may have been partly offset by the increase in arterial oxygen concentration, related to the increase in hemoglobin concentration, which prevented oxygen delivery from decreasing in the face of decreased cardiac output. However, it is possible that imbalances between oxygen delivery and consumption occurred at the local level. Indeed, PvO 2 and SaO 2 decreased, and oxygen extraction ratio increased with increasing plasma dexmedetomidine concentrations, indicating that despite the lack of statistical change in oxygen delivery and consumption, the ratio of delivery to consumption decreased. Venous admixture was lower than baseline at the highest 564 Ó 2011 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesiologists, 38, 555 567

dexmedetomidine concentration. This is possibly due to the effect of decreased cardiac output on intrapulmonary shunting (Lynch et al. 1979), and contributed to the increase in PaO 2. PCO 2 was higher than baseline at the highest plasma dexmedetomidine concentration in both arterial and mixed-venous blood, indicating that dexmedetomidine, at least when high doses are administered, worsens the respiratory depressant effect of isoflurane. The increase in PCO 2 resulted in a decrease in blood ph. In addition, arterial bicarbonate concentration also decreased, contributing to the decrease in arterial ph. Blood glucose concentration increased with dexmedetomidine administration; this effect is likely due to a combination of an inhibition of insulin secretion by alpha-2 agonists (Angel et al. 1988) and the administration of dextrose in water for cardiac output measurement. Balanced anesthesia is described as the concurrent administration of several drugs so that no single drug is given in a dose sufficient to produce toxicity during or after surgery, to improve operating conditions and patient safety, and reduce the side-effects of anesthesia. Balanced anesthesia is often advocated to decrease the requirements for inhalational anesthetics, and thereby limit the cardiovascular depression they induce (Bailey et al. 2000). In a previous study, we have shown that dexmedetomidine decreased the requirements for isoflurane in a plasma concentration-dependent manner (Escobar et al. 2010a). However, the study reported here suggests that the use of dexmedetomidine in combination with a lower concentration of isoflurane does not provide hemodynamic benefits, and would likely worsen global hemodynamics compared to the use of a higher, equipotent dose of isoflurane alone. Indeed, according to our study on the effects of dexmedetomidine on the MAC of isoflurane, a plasma dexmedetomidine concentration of 0.4 ng ml )1 would produce, on average, a 20% decrease in MAC (i.e. a decrease of clinically relevant magnitude). According to our pharmacokinetic data, that plasma concentration would be achieved using a loading dose of 0.2 lg kg )1 followed by a constant rate infusion of 0.15 lg kg )1 hour )1 (Escobar et al. 2010b). While at that concentration, the mean model predicts that heart rate would only be 8% lower and mean arterial pressure 15% higher (arguably a benefit) than with isoflurane alone, cardiac index is predicted to decrease by 16%, and systemic vascular resistance index to increase by 48%, both effects being, at least in theory, undesirable. Therefore, it appears that the use of dexmedetomidine as an anesthetic adjunct in cats will worsen inhalant-induced cardiovascular depression, despite the decrease in anesthetic requirements. This is in disagreement with findings in dogs, in which an intravenous loading dose of 0.5 lg kg )1 followed by a constant rate intravenous infusion of 0.5 lg kg )1 hour )1 produced close to a 20% decrease in the MAC of isoflurane, with minimal cardiopulmonary effects (Pascoe 2005). In conclusion, in isoflurane-anesthetized cats, dexmedetomidine produces hemodynamic effects in a concentration-dependent manner. These effects are mainly characterized by a decrease in heart rate and cardiac output, and an increase in blood pressure and systemic and pulmonary vascular resistances. At plasma concentrations producing decreases in MAC of clinically relevant magnitude, dexmedetomidine administration is expected to result in a lower cardiac output and higher systemic vascular resistance than a higher, equipotent concentration of isoflurane alone. Acknowledgements This study was funded by the Center for Companion Animal Health, School of Veterinary Medicine, University of California, Davis. The authors are grateful to Orion Pharma for providing dexmedetomidine analytical standard. References Aantaa R, Scheinin M (1993) Alpha 2-adrenergic agents in anaesthesia. Acta Anaesthesiol Scand 37, 433 448. Aantaa R, Kanto J, Scheinin M et al. (1990) Dexmedetomidine, an alpha 2-adrenoceptor agonist, reduces anesthetic requirements for patients undergoing minor gynecologic surgery. Anesthesiology 73, 230 235. Aantaa R, Jaakola ML, Kallio A et al. (1997) Reduction of the minimum alveolar concentration of isoflurane by dexmedetomidine. Anesthesiology 86, 1055 1060. Aho M, Erkola O, Kallio A et al. (1992) Dexmedetomidine infusion for maintenance of anesthesia in patients undergoing abdominal hysterectomy. Anesth Analg 75, 940 946. Angel I, Bidet S, Langer SZ (1988) Pharmacological characterization of the hyperglycemia induced by alpha- 2 adrenoceptor agonists. J Pharmacol Exp Ther 246, 1098 1103. Badoer E, Head GA, Korner PI (1983) Effects of intracisternal and intravenous alpha-methyldopa and clonidine Ó 2011 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesiologists, 38, 555 567 565

on haemodynamics and baroreceptor heart rate reflex properties in conscious rabbits. J Cardiovasc Pharmacol 5, 760 767. Bailey PL, Egan TE, Stanley TH (2000) Intravenous opioids anesthetics. In: Anesthesia,Vol. 1 (5th edn). Miller RD (ed.). Churchill Livingstone, Philadelphia, PA, USA. pp. 273 376. Bettschart-Wolfensberger R, Freeman SL, Bowen IM et al. (2005) Cardiopulmonary effects and pharmacokinetics of i.v. dexmedetomidine in ponies. Equine Vet J 37, 60 64. Bloor BC, Frankland M, Alper G et al. (1992) Hemodynamic and sedative effects of dexmedetomidine in dog. J Pharmacol Exp Ther 263, 690 697. Boscan P, Pypendop BH, Solano AM et al. (2005) Cardiovascular and respiratory effects of ketamine infusions in isoflurane-anesthetized dogs before and during noxious stimulation. Am J Vet Res 66, 2122 2129. Cooper RA, Weekes AJ (1983) Univariate descriptive analysis. In: Data, Models and Statistical Analysis. Cooper RA, Weekes AJ (eds). Rowman and Littlefield Publishers, Lanham, MD, USA. pp. 49 76. Darovic GH (1995) Hemodynamic Monitoring: Invasive and Noninvasive Clinical Application. WB Saunders Company, Philadelphia, PA, USA. Devcic A, Schmeling WT, Kampine JP et al. (1994) Oral dexmedetomidine preserves baroreceptor function and decreases anesthetic requirements of halothane-anesthetized dogs. Anesthesiology 81, 419 430. Dobkin AB, Fedoruk S (1961) Comparison of the cardiovascular, respiratory and metabolic effects of methoxyflurane and halothane in dogs. Anesthesiology 22, 355 362. Duka I, Gavras I, Johns C et al. (2000) Role of the postsynaptic alpha(2)-adrenergic receptor subtypes in catecholamine-induced vasoconstriction. Gen Pharmacol 34, 101 106. Dutta S, Lal R, Karol MD et al. (2000) Influence of cardiac output on dexmedetomidine pharmacokinetics. J Pharm Sci 89, 519 527. Escobar A, Pypendop BH, Siao KT et al. (2010a) Effect of dexmedetomidine on the minimum alveolar concentration of isoflurane in cats. J Vet Pharmacol Ther 2011, doi: 10.1111/j.1365-2885.2011.01301.x. [Epub ahead of print] Escobar A, Pypendop BH, Siao KT et al. (2010b) Pharmacokinetics of dexmedetomidine in isoflurane-anesthetized cats. Proceedings of the 16th International Veterinary Emergency & Critical Care Symposium, San Antonio, TX. pp. 672 (abstract). Flacke JW, Flacke WE, Bloor BC et al. (1990) Hemodynamic effects of dexmedetomidine, an alpha 2-adrenergic agonist, in autonomically denervated dogs. J Cardiovasc Pharmacol 16, 616 623. Fragen RJ, Fitzgerald PC (1999) Effect of dexmedetomidine on the minimum alveolar concentration (MAC) of sevoflurane in adults age 55 to 70 years. J Clin Anesth 11, 466 470. Golden AL, Bright JM, Daniel GB et al. (1998) Cardiovascular effects of the alpha2-adrenergic receptor agonist medetomidine in clinically normal cats anesthetized with isoflurane. Am J Vet Res 59, 509 513. Gornemann T, von Wenckstern H, Kleuser B et al. (2007) Characterization of the postjunctional alpha 2C-adrenoceptor mediating vasoconstriction to UK14304 in porcine pulmonary veins. Br J Pharmacol 151, 186 194. Gornemann T, Villalon CM, Centurion D et al. (2009) Phenylephrine contracts porcine pulmonary veins via alpha(1b)-, alpha(1d)-, and alpha(2)-adrenoceptors. Eur J Pharmacol 613, 86 92. Hammer GB, Sam WJ, Chen MI et al. (2009) Determination of the pharmacodynamic interaction of propofol and dexmedetomidine during esophagogastroduodenoscopy in children. Paediatr Anaesth 19, 138 144. Harron DW, Riddell JG, Shanks RG (1985) Effects of azepexole and clonidine on baroreceptor mediated reflex bradycardia and physiological tremor in man. Br J Clin Pharmacol 20, 431 436. Hodgson DS, Dunlop CI, Chapman PL et al. (1998) Cardiopulmonary effects of anesthesia induced and maintained with isoflurane in cats. Am J Vet Res 59, 182 185. Hyman AL, Kadowitz PJ (1985) Evidence for existence of postjunctional alpha 1- and alpha 2-adrenoceptors in cat pulmonary vascular bed. Am J Physiol 249, H891 H898. Ingwersen W, Allen DG, Dyson DH et al. (1988) Cardiopulmonary effects of a halothane/oxygen combination in healthy cats. Can J Vet Res 52, 386 391. Kunisawa T, Nagata O, Nagashima M et al. (2009) Dexmedetomidine suppresses the decrease in blood pressure during anesthetic induction and blunts the cardiovascular response to tracheal intubation. J Clin Anesth 21, 194 199. Kunisawa T, Nagashima M, Hanada S et al. (2010) Awake intubation under sedation using target-controlled infusion of dexmedetomidine: five case reports. J Anesth 24, 789 792. Kutter AP, Kastner SB, Bettschart-Wolfensberger R et al. (2006) Cardiopulmonary effects of dexmedetomidine in goats and sheep anaesthetised with sevoflurane. Vet Rec 159, 624 629. Lamont LA, Bulmer BJ, Grimm KA et al. (2001) Cardiopulmonary evaluation of the use of medetomidine hydrochloride in cats. Am J Vet Res 62, 1745 1749. Lawrence CJ, Prinzen FW, de Lange S (1996) The effect of dexmedetomidine on nutrient organ blood flow. Anesth Analg 83, 1160 1165. Lawrence CJ, Prinzen FW, de Lange S (1997) Hemodynamic and coronary vascular effects of dexmedetomidine in the anesthetized goat. Acta Anaesthesiol Scand 41, 830 836. Lumb AB (2000) Nunn s Applied Respiratory Physiology (5th edn). Butterworth-Heinemann, Oxford, UK. 566 Ó 2011 Association of Veterinary Anaesthetists and the American College of Veterinary Anesthesiologists, 38, 555 567