In vivo confocal microscopy of the normal equine cornea and limbus

Objective  To describe morphologic features, pachymetry and endothelial cell density of the normal equine cornea and limbus by in vivo confocal microscopy.
Animals studied  Ten horses without ocular disease.
Procedure  The central and peripheral corneas were examined with a modified Heidelberg Retina Tomograph II and Rostock Cornea Module using a combination of automated and manual image acquisition modes. Thickness measurements of various corneal layers were performed and endothelial cell density determined.
Results  Images of the constituent cellular and noncellular elements of the corneal epithelium, stroma, endothelium, and limbus were acquired in all horses. Corneal stromal nerves, the subepithelial nerve plexus, and the sub-basal nerve plexus were visualized. Cells with an appearance characteristic of Langerhans cells and corneal stromal dendritic cells were consistently detected in the corneal basal epithelium and anterior stroma, respectively. Median central total corneal thickness was 835 μm (range 725–920 μm) and median central corneal epithelial thickness was 131 μm (range 115–141 μm). Median central endothelial cell density was 3002 cells per mm² (range 2473–3581 cells per mm²).
Conclusions  In vivo corneal confocal microscopy provides a noninvasive method of assessing normal equine corneal structure at the cellular level and is a precise technique for corneal sublayer pachymetry and cell density measurements. A resident population of presumed Langerhans cells and corneal stromal dendritic cells was detected in the normal equine cornea. The described techniques can be applied to diagnostic evaluation of corneal alternations associated with disease and have broad clinical and research applications in the horse.     

A Portable Blood Gas Analyzer for Equine Venous Blood

Evaluation of a portable blood gas analyzer, (StatPal II, Unifet, Inc, La Jolla, CA) was performed using tonometered solutions and equine blood. Samples were analyzed by the StatPal II and either an Instrument Laboratory IL1306 (Lexington, MA) or a Radiometer ABL50 blood gas analyzer (Radiometer America Inc., Westlake, OH). Comparison of the StatPal II and the IL1306 was done by analysis of 3 tonometered solutions (acidic, normal, and alkalotic) and 27 equine venous blood samples. Blood pH, Pco₂, Po₂, and [HCO₃^-] values were altered by IV infusion of 5% sodium bicarbonate or exercising the horses on a treadmill. Comparison of the StatPal II and the Radiometer was performed by analysis of 78 blood samples collected from Standard-bred horses before a race. Data were analyzed for the venous blood samples using a paired two-tailed Student's t test and Bland-Altman plots, with significance set at P < .05. The coefficients of variation for pH, Pco₂, Po₂, and [HCO₃^-] values of the tonometered solutions analyzed by the StatPal II ranged from 0.067% to 0.087%, 2% to 3.21 %, 1.21 % to 2.67%, and 0.267% to 0.828%, respectively. Comparison of the equine blood samples analyzed by the StatPal II and the IL1306 demonstrated statistically significant, but clinically irrelevant differences in pH, Pco₂, and Po₂, but not [HCO₃^-]. There were statistically significant, but clinically irrelevant differences between the StatPal II and the Radiometer for pH, Pco₂, and [HCO₃^-], but not for Po₂-It is concluded that the StatPal II provides reproducible and acceptable analysis of equine venous blood gas samples.     

Photoreceptor density of the domestic pig retina

The spatial distribution and densities of photoreceptors in seven whole-mounted porcine retinas were studied and maps illustrating photoreceptor topography were constructed. Total photoreceptor densities ranged from to 83 000 to 200 000 cells/mm², with a mean of 138 500 cells/mm². Cone densities ranged from 39 000 (area centralis) to 8500 cones/mm² (peripherally), with a mean of 16 400 cones/mm². Rod:cone ratios ranged from 3:1 centrally to 16:1 peripherally, with a mean ratio of 8:1. Averaged photoreceptor densities are greatest (166 000 cells/mm²) within the central inferior retina, and regional differences in rod:cone ratios were found. Cone densities are increased in a broad region dorsal to the optic disk, extending both nasally and temporally. This region is believed to represent the area centralis. Cone densities gradually decrease and taper towards the periphery and inferior retina as rod:cone ratios increase. In addition to the many anatomic and ultrastructural similarities to the human eye, this study illustrates similarities within the photoreceptor mosaic of these two species and supports the use of the pig retina as a model for human/animal research.     

Regional specialization of the Ganglion cell density in the retina of the Ostrich (Struthio camelus)

In this study, retinal whole-mount specimens were prepared and stained with 0.1% cresyl violet for the ganglion cell study in the Ostrich (Struthio camelus) . The total number, distribution, and size of these cells were determined in different retinal regions. The mean total number of ganglion cells (three retinas) was 1 435 052 with an average density of 652 cells/mm². The temporo – nasal area of the retina with high cell density were identified with the peak of 7525 cells/mm² in the central area. The size of most ganglion cells ranged from 113–403 µm², with smaller cells predominating along the temporo-nasal streak above the optic disc and larger cells comprising more of the peripheral regions. The average thickness of the retina was 196 µm. The central area was the thickest area (268.6 µm), whereas the peripheral area was the thinnest area. Thus, the specialization of ganglion cell densities, their sizes and the thickness of the retina support the notion that the conduction of visual information towards the brain from all regions of the retina is not uniform, and suggests that the temporo – nasal streak is the fine quality area for vision in ostriches.     

Comparison of bone healing by demineralized bone matrix and autogenous cancellous bone in horses

OBJECTIVE: The purpose of this study was to compare bone healing induced by equine demineralized bone matrix (DBM) to autogenous cancellous bone graft (ACB) or no graft (control) in a rib-defect model in horses. STUDY DESIGN: The osteogenic properties of ACB and DBM were evaluated in bilateral 19-mm circular defects created in the outer cortex of the 6th and 8th ribs of each horse. ANIMALS OR SAMPLE POPULATION: Eight mature horses. METHODS: Three rib defects in each horse were randomly treated with each of the 3 treatment groups, and the fourth rib defect received a random treatment. Rib sections, including the defects, were harvested 56 days after implantation and examined for bone mineral density, percent ash and calcium and graded for signs of radiographic and histological healing. RESULTS: All ribs were fractured at the defect site and were classified as nonunion fractures 56 days after implantation. There were no significant differences among groups in bone mineral density and signs of radiographic or histological healing. There was an increased volume of bone in control and ACB-treated sites compared with DBM-treated sites. Rib defects treated with ACB were significantly higher in percent ash and calcium than those treated with DBM. DBM elicited no inflammatory reaction, and remodeling occurred around the periphery and within vascular channels of the decalcified particles. CONCLUSION: DBM particles remodel from the periphery, which may explain the significantly lower percent ash, calcium, and bone when compared with ACB, because 2- to 4-microL pieces of DBM may act as space-occupying masses until completely mineralized. There was no evidence of enhanced healing associated with the use of DBM in this model. CLINICAL RELEVANCE: Particles of 2 to 4 mm DBM should not be used as an aid to fracture repair because particles of this size interfere with normal mineralization. However, our model of nonunion fracture healing may be useful in future studies.     

Determination of 'irritant' threshold concentrations for intradermal testing with allergenic insect extracts in normal horses

Abstract Sixteen healthy horses with no history of skin or respiratory disease were used for an intradermal testing (IDT) threshold study, in order to determine the concentrations of 13 commercial allergenic insect extracts most appropriate for IDT. Five dilutions of each extract were used, which included the manufacturer's recommended concentrations for equine IDT, plus one dilution higher and three lower than these standard concentrations. Allergens tested included caddisfly ( Trichoptera spp.), mayfly ( Ephemeroptera spp.), horsefly ( Tabanus spp.), deerfly ( Chrysops spp.), fire ant ( Solenopsis invicta ), black ant ( Camponotus pennsylvanicus ), cockroach mix ( Periplaneta americana and Blattella germanica ), mosquito ( Aedes aegypti ), house fly ( Musca domestica ), moth ( Heterocera spp.), flea ( Ctenocephalides canis / C. felis ), Culicoides variipennis and Culicoides nubeculosis. Two separate methods were used to calculate the allergen concentration for each insect extract where the normal horses, as a group, ceased to show false-positive ('irritant') reactions. 'Irritant' threshold concentrations were determined for 9/13 of these allergens, whereas the other 4 were undetermined due to either insufficient reactivity (flea, C. variipennis ) or excessive reactivity (black ant, moth) to the concentrations tested. Recommended concentrations for future use in equine patients with suspected insect hypersensitivity include: 125 pnu mL⁻¹ (mayfly); 250 pnu mL⁻¹ (caddisfly, horsefly, deerfly, fire ant, house fly); 500 pnu mL⁻¹ (cockroach); 1000 pnu mL⁻¹ (mosquito); and 1:10 000 w/v ( C. nubeculosis ).     

Characterization of muscarinic receptor subtype mediating contraction and relaxation in equine coronary artery in vitro

In coronary arterial rings isolated from horses, 10-^-8-10^-6 mol/l acetylcholine (ACh) induced concentration-dependent contractions which were potentiated by the removal of endothelium and by pretreatment with I,-nitro-arginine (LNAG) or methylene blue (MB). Relatively lower concentrations of Ach 10-¹⁴-10-⁸ mol/l) induced relaxation when the coronary rings were contracted by phenylephrine (PE). ACh-induced contractions in the coronary rings without endothelium were competitively inhibited by each muscarinic subtype selective antagonist in the following order of potency: 4-diphenylacetoxy-N-methylpiperidine methiodide (4-DAMP) > pirenzepine ≥ parafluoro-hexahydrosiladiphenidol (pFHHSiD) > methoctramine. ACh-induced relaxation in the rings with endothelium was inhibited by LNAG or MB, and by each selective antagonist in the following order of potency: 4-DAMP < pFHHSiD ≥ pirenzepine ≥ methoctramine. These results suggest that the ACh-induced contraction and relaxation in equine coronary arteries are mediated mainly by an M₃-receptor located on the smooth muscle cells and endothelial cells, respectively, and that the stimulation of the M₃-receptor on the endothelial cells liberates nitric oxide.     

Effect of the 5-lipoxygenase inhibitor, fenleuton, on antigen-induced neutrophil accumulation and lung function changes in horses with chronic obstructive pulmonary disease

The leukotrienes (LT) LTD₄ and LTB₄ have been shown to cause bronchoconstriction and neutrophil accumulation, respectively, in horse lungs. Such changes are characteristic of the equine allergic respiratory disease, chronic obstructive pulmonary disease (COPD). To further investigate the role of these putative mediators in the pathogenesis of equine COPD the effect of a 5-lipoxygenase inhibitor, fenleuton, on antigen-induced changes in horses with this condition has been examined. Six horses with COPD underwent a series of four antigen challenges, one month apart, with placebo pre-treatment on three occasions and fenleuton (4 days oral dosing 5 mg/kg) pre-treatment on one occasion. Three horses received fenleuton prior to the second challenge and three horses received the drug prior to the fourth antigen challenge. Changes in radiolabelled neutrophil distribution, lung function and peripheral leucocyte counts were monitored on each occasion for 7 h following the start of antigen challenge. Antigen challenge caused an increase in radioactive counts over the lungs and a decrease in peripheral leucocyte count. Neither response was affected by fenleuton pre-treatment. Mean maximal changes in pleural pressure (ΔP_plmax) and respiratory rate were also unaffected by fenleuton pre-treatment. However, in the two horses which responded to antigen-challenge with a particularly marked increase in ΔP_plmax (>15 cm H₂O), prior administration of fenleuton reduced the response by 64 and 63%. These results suggest that 5-lipoxygenase inhibitors warrant further investigation as bronchodilators in equine COPD.     

EFFECT OF WETNESS LEVEL ON THE SUITABILITY OF WET GAUZE AS A SUBSTITUTE FOR SUPERFLAB® AS A BOLUS MATERIAL FOR USE WITH 6 MV PHOTONS

Despite the availability of commercial tissue equivalent bolus material, wet gauze has an application in radiation therapy to provide superior conformance to irregular contours. Wet gauze bolus has the potential to reduce air gaps between the bolus and surface, which could decrease surface dose if sufficiently large to disrupt electronic equilibrium. Wet gauze bolus is often fabricated and wetness judged qualitatively. We assessed the effect of specific gauze wetness levels, quantified in terms of physical density, at various field sizes with respect to their effectiveness as bolus material compared with Superflab^®. For large fields, >7 × 7 cm² in this study, wet gauze sponges with a physical density of 1.02 g/cm³ performed essentially identical to Superflab^®; at a smaller field size the wet gauze was slightly less effective, likely due to the heterogeneity of the gauze-water matrix. Gauze that was wetter, with a physical density of 1.2 g/cm³, or less wet, with a physical density of 0.75 g/cm³ was not as effective either due to enhanced photon absorption in the wetter sponges, or less effective establishment of electronic equilibrium in the less wet sponges. The presence of an air gap under Superflab^® led to reduced surface dose, especially for small fields and large air gaps. Thus, if Superflab^® use leads to poor contact with the skin, wet gauze having a physical density of 1.02 g/cm³ can be used as a substitute. Judging the water content of wet gauze subjectively is not acceptable as over- or under wetness can lead to decreased effectiveness of the bolus material.     

Plasma pharmacokinetics of ranitidine HCl in adult horses

Plasma pharmacokinetics of ranitidine HCl were investigated after intravenous (i.v.) and oral (p.o.) administration of 2.2 mg/kg drug to six healthy adult horses. Concentrations of ranitidine were determined using normal-phase, high-performance liquid chromatography. Plasma concentrations of ranitidine HCl declined from a mean of 5175 ng/mL at 5 min to 37 ng/mL at 720 min after i.v. administration. A three-exponent equation, C_p= A₁· e^–k1t+ A₂· e^–k2t+ A₃· e^–k3t, best described data for all horses. Mean values for model-independent values calculated from the last quantifiable time point were: apparent volume of distribution (Vd_ss) = 1.07 L/kg; area under the curve ( AUC ) = 231,000 ng · min/mL; area under the moment curve ( AUMC ) = 26,900,000 ng · min²/mL; mean residence time ( MRT ) = 113 min; and clearance (Cl) = 9.8 mL/min.kg. Following p.o. administration, a two-exponent equation, C_p= A₁· e^–k1t+ A₂· e^–k2t, best described the data for five horses; data for the remaining horse were best described by a three-exponent equation. Mean values of pharmacokinetic values from the p.o. study include: AUC = 59,900 ng · min/mL; AUMC = 10,600,000 ng · min²/mL; mean absorption time ( MAT ) = 58.9 min; T _max= 99.2 min; C _max= 237 ng/mL; and F = 27%.     

Prenatal Development of the Human Epicardiac Ganglia

The aim of this study was to determine the developmental anatomy of intrinsic cardiac ganglia with respect to epicardiac ganglionated nerve plexus in the human fetuses at different gestation stages. Twenty fetal hearts were investigated applying a technique of histochemistry for acetylcholinesterase to visualize the epicardiac neural ganglionated plexus with its subsequent examinations on total (non-sectioned) hearts. Most epicardiac ganglia embodied multilayered neurons and were oval in shape, but some ganglia involved neurons lying in one layer or had the irregular appearance because of their extensions along inter-ganglionic nerves. The mean ganglion area of fetuses at gestation stages of 15–40 weeks was 0.03 ± 0.008 mm². The largest epicardiac ganglia, reaching in area 0.4 mm², were concentrated on the dorsal surface of both atria. The particular fused or "dual" ganglia were identified at the gestation stages of 23–40 weeks, but they composed only 2.3 ± 0.7% of all found epicardiac ganglia. A direct positive correlation was determined between the fetal age and the ganglion area (mm²) as well as between the fetal age and the number of inter-ganglionic nerves. The revealed appearance of epicardiac ganglia in the human fetuses at 15–40 weeks of gestation confirms their prenatal development and presumable intrinsic remodelling.     

The use of compound 48/80 as a positive control in equine intradermal allergy testing

The macroscopic and microscopic effects following the intradermal injection of the mast cell degranulator compound 48/80 were investigated in horses. It was found that a 1000 μg mL⁻¹ solution of compound 48/80 could be used as an alternative to histamine as a positive control in equine intradermal allergy tests. An inflammatory cell infiltration was noted on histological examination, following compound 48/80 injection. This inflammatory cell pattern was similar to that noted for Type-I hypersensitivity.     

Disposition of human drug preparations in the horse. III. Orally administered alclofenac

Concentrations of the non-steroidal anti-inflammatory drug (NSAID) alclofenac were determined by a sensitive high performance liquid chromatographic procedure in plasma and urine of horses following oral administration of a dose of 3 g. In plasma, alclofenac was present in detectable concentrations for 72 h. The plasma disposition in individual horses was best described by a bi-compartmental model with two successive rate constants ka₁= 0.05 ± 0.06 h^-1 and ka₂= 0.06 ± 0.01 h^-l. Alclofenac half-lives t _½ and t _1/2β were 1.0 ± 0.8 h and 6.9 ± 1.5 h, respectively. Maximal concentrations (38.9 ± 16.2 μg/ml) were obtained after 8.5 ± 2.4 h. Alclofenac was detected in urine for at least 48 h after dosing. The percentage of the dose excreted as unchanged alclofenac in 12 h was very low (0.68 ± 0.19%), total (free + conjugated) alclofenac accounted for 2.16 ± 0.55% of the dose.     

AN INNOVATIVE TECHNIQUE FOR DISPLAYING THREE DIMENSIONAL RADIOGRAPHIC ANATOMY OF SYNOVIAL STRUCTURES IN THE EQUINE DISTAL LIMB

Radiography is the most commonly applied imaging modality in equine practice and forms an essential part of the diagnostic work-up of lame horses. Radiographic signs of musculoskeletal pathology are frequently localized at sites of soft tissue attachment, which are often not clearly visible on radiographs. Different lesions carry different prognoses and require a variety of treatments, and a good knowledge of the position of the synovial structures in the distal limb of the horse is essential for practitioners in the interpretation of radiographs. This study describes a new technique for creating three-dimensional (3D) models of the synovial structures and superimposing them onto radiographs for the purpose of teaching radiographic anatomy. A set of standard radiographs was acquired of the metacarpophalangeal and the distal interphalangeal joints of a fresh cadaver leg while the leg was positioned in a material-testing machine to mimic the weight-bearing horse. Computed tomography of the same regions was performed after injection of negative contrast medium into the joints. 3D reconstructions of the joints were created using grayscale thresholding and polynomial surface meshing in Mimics. The resulting 3D reconstructions were superimposed on top of the radiographs using Adobe^© Photoshop^© CS3 Extended, thus allowing the visualization of the joint anatomy in relation to the bone on all projections. The main advantage of this technique is that it allows synovial structures to be visualized on radiographs where they are normally indistinct, which will serve as a teaching aid for anatomy.     

Sedative effects of propofol in horses

Objective  We hypothesized that propofol can produce rapidly-reversible, dose-dependent standing sedation in horses.
Study design  Prospective randomized, blinded, experimental trial.
Animals  Twelve healthy horses aged 12 ± 6 years (mean ± SD), weighing 565 ± 20 kg, and with an equal distribution of mares and geldings.
Methods  Propofol was administered as an intravenous bolus at one of three randomized doses (0.20, 0.35 and 0.50 mg kg⁻¹). Cardiovascular and behavioral measurements were made by a single investigator, who was blinded to treatment dose, at 3 minute intervals until subjective behavior scores returned to pre-sedation baseline values. Continuous data were analyzed over time using repeated-measures anova and noncontinuous data were analyzed using Friedman tests.
Results  There were no significant propofol dose or temporal effects on heart rate, respiratory rate, vertical head height, or jugular venous blood gases (pH_v, P_vO₂, P_vCO₂). The 0.35 mg kg⁻¹ dose caused mild sedation lasting up to 6 minutes. The 0.50 mg kg⁻¹ dose increased sedation depth and duration, but with increased ataxia and apparent muscle weakness.
Conclusions and clinical relevance  Intravenous 0.35 mg kg⁻¹ propofol provided brief, mild sedation in horses. Caution is warranted at higher doses due to increased risk of ataxia.     

Occlusion of the Middle Cerebral Artery: a New Method of Focal Cerebral Ischemia in Rats

The study in Wistar rats attempted to improve the occlusion technique of the middle cerebral artery (MCA) as a precise method for initiating stroke. In a first part it was necessary to study the exact anatomy of blood vessels of the brain in seven rats of 170-410 g body weight by corrosion cast. The lengths and diameters of defined locations of the blood vessels were measured. The temporary as well as the permanent methods were refined or replaced. The first one was completed in main training the physiological blood flow after temporary occlusion, while the permanent occlusion was performed by positioning a silicone cap in the MCA. A filament guide was introduced from the common carotid artery (CCA) via internal carotid artery (ICA) to guide the silicon cap at the branch of the MCA. Histological sections of the brain of rats showed 24 h after the permanent occlusion a reproducible infarct in the brain. This area corresponded very well with the supply of the MCA. The new occlusion method with a silicon cap was compared with the occlusion methods of CCA route and external carotid artery (ECA) route. The total infarct volume was significantly larger in the CCA route and ECA route groups than in the silicon cap group (means: CCA route 261 mm³; ECA route 191 mm³ vs. 128 mm³ silicon cap group; P  < 0,05). It could be demonstrated that the new silicon cap occlusion technique imitates the pathological situation of a cerebral infarct in man. Moreover it is less invasive for the animals and more precise and reproducible regarding the infarcted area in comparison to the other occlusion methods. Based on anatomical measurements of the blood vessels the described silicon cap method can be recommended for rats of a body weight between 340–370 g.     

EVALUATION OF THE EXTIRPATED EQUINE EYE USING B-MODE ULTRASONOGRAPHY

B-mode ultrasonography using a 7.5 megahertz (MH^z) transducer was performed on 95 equine cadaver eyes submerged in a water bath. Identification and measurement of the anterior to posterior diameters of the intraocular structures and of the axis bulbi was done. The dorsal ventral axis of the eye was also measured. The eyes were then frozen, sectioned in a saggital plane and the same measurements were made using calipers. Two sets of measurements were then compared statistically.
With the exception of the cornea and the dorsal ventral axis measurements of male horses there was no statistically significant difference in the measurements obtained by the two methods.
The use of B-mode ultrasonography at a frequency of 7.5 MH^z to examine the architecture of the equine eye is suggested as a valuable clinical tool. Biometry of the intraocular structures at this frequency may be reasonably accurate.     

IRIDIUM-192 INTERSTITIAL BRACHYTHERAPY OF EQUINE SARCOID

Twenty-three equine sarcoids in 22 horses were treated, using an afterloading technique, with iridium-192 (¹⁹²Ir) interstitial radiation therapy. Previously unsuccessful treatment in 14 horses included surgical excision, cryosurgery, hyperthermia, and nonspecific immunotherapy. Twenty tumors were located in the skin of the head, and three were in the skin of the extremities. Before implantation, surgical debulking was done in six tumors, and hyperthermia (43°C, 30 min) was done in three tumors. Total implantation time varied from 4 to 14 days to give radiation doses of 52–93 Gy. The criteria for assessment of therapy were tumor response, tumor-free incidence at one year, and complications. Follow-up times ranged from six to 41 months (mean, 20 months). All tumors of the head and two tumors of the extremities had complete response to therapy. One sarcoid of the extremity had partial response to treatment. Tumor recurence was seen in the head of one horse. Of 16 horses followed for one year, the tumor-free incidence was 94% (15/16). Complications were severe local necrosis in two horses and deep infection in one horse. It was concluded that ¹⁹²Ir interstitial brachytherapy was an effective treatment of recurrent and difficult-to-manage equine sarcoid.     

The Effects of Sample Size on the Outcome of Ovarian Tissue Cryopreservation

Cryopreservation of ovarian tissue is known to affect follicular survival. Several variables may be responsible for this. Little attention has focused on the effect of the size of the fragment to be cryopreserved. This study was conducted to assess the effect of the size of the tissue on follicular histology after freezing with 1,2-propanediol. Histological evaluations were performed of control and cryopreserved tissue. Fragments were cut 10 × 3 × 2 mm³ (2 mm group) or 10 × 3 × 4 mm³ (4 mm group). Percentages of normal follicles in control fragments cut into 2 and 4 mm slices were 56% and 34%, respectively. The relative risks to obtain normal follicles in the 2 mm and the 4 mm fragments after cryopreservation were 0.63 and 0.47, respectively. Freezing reduced follicle survival to a significantly greater extent in the larger tissue fragments. There is an increased risk of damage to primary and primordial follicles, when the tissue slices are cut with all dimensions larger than 2 mm.     

The actions of medetomidine may not be mediated exclusively by α2-adrenoceptors in the equine saphenous vein

Spirals of endothelially denuded equine saphenous vein were used to study the pre- and post-junctional effects of medetomidine in vitro . The pD₂ values were calculated for noradrenaline (6.7 /pm 0.1), phenylephrine (5.6 /pm 0.1), BHT 920 (6.2 /pm 0.2) and UK 14304 (5.7 /pm 0.2). Medetomidine produced a biphasic response, with a pD₂1 of 8.2 /pm 0.1 and a pD₂2 of 5.7 /pm 0.1 in the equine saphenous vein ( n = 6 ). Prazosin (10⁻⁷ m) significantly shifted the second phase of the medetomidine concentration-response curve to the right (pD₂1 was 8.1 /pm 0.2 and pD₂2 was 5.0 /pm 0.2, P < 0.05). Rings of equine saphenous vein were electrically stimulated to investigate the pre-junctional effects of medetomidine. Increasing concentrations of the α₂-adrenoceptor agonist BHT 920 reduced the response to electrical stimulation in a concentration dependent manner to a maximum of 40 /pm 5%. whereas medetomidine (0.1-100 nm) caused a concentration dependent enhancement to a maximum of 490 /pm 150%. These results suggest α₁- and α₂-adrenoceptors are functional in the equine saphenous vein, but that medetomidine is not acting exclusively as an α₂-adrenoceptor agonist.