The effects of medetomidine on the action of vecuronium in dogs anaesthetized with halothane and nitrous oxide

ObjectiveTo quantify the effects of medetomidine on the onset and duration of vecuronium-induced neuromuscular blockade in dogs.Study designRandomized, prospective clinical study.AnimalsTwenty-four, healthy, client-owned dogs of different breeds, aged between 6 months and 10 years and weighing between 5.0 and 40.0 kg undergoing elective surgery.MethodsDogs were randomly allocated to two groups. Pre-anaesthetic medication in group M+ was intramuscular acepromazine (ACP) 25 μg kg⁻¹, morphine 0.5 mg kg⁻¹ and medetomidine 5 μg kg⁻¹. Group M− received ACP and morphine only, at the same dose rate. After induction with thiopental, anaesthesia was maintained with halothane in oxygen and nitrous oxide. End-tidal halothane concentration was maintained at 1.1%. Neuromuscular blockade was produced with intravenous vecuronium (50 μg kg⁻¹) and monitored using a train of four stimulus applied at the ulnar nerve. The times taken for loss and reappearance of the four evoked responses (twitches [T]) were recorded. Normal and nonparametric data were analysed with an independent t-test and Mann-Whitney's U-test, respectively.ResultsThe fourth twitch (T4) disappeared at similar times in each group: 107 ± 19; [72–132] (mean ± SD; [range]) seconds in M+ and 98 ± 17 [72–120] seconds in M− dogs. The first twitch (T1) was lost at 116 ± 15; [96–132] seconds in group M+ and 109 ± 19; [72–132] seconds in M−. The fourth twitch returned significantly earlier in M+ dogs: 20.8 ± 3.8 [14–28] minutes compared with 23.8 ± 2.7 [20–27] minutes (p = 0.032). The duration of drug effect (T4 absent) was significantly shorter (p = 0.027) in M+ (18.9 ± 3.7 minutes) compared with M− dogs (22.2 ± 2.9 minutes). The recovery rate (interval between reappearance of T1 and T4) was significantly more rapid (p = 0.0003) in medetomidine recipients (3.0 ± 1.2 versus 5.2 ± 1.3 minutes).Conclusion and clinical relevance Medetomidine 5 μg kg⁻¹ as pre-anaesthetic medication shortened the duration of effect of vecuronium in halothane-anaesthetized dogs and accelerated recovery, but did not affect the onset time. These changes are of limited clinical significance.     

The effect of nitrous oxide on halothane, isoflurane and sevoflurane requirements in ventilated dogs undergoing ovariohysterectomy

OBJECTIVE: To examine the effect of 64% nitrous oxide (N2O) on halothane (HAL), isoflurane (ISO) or sevoflurane (SEV) requirements in dogs undergoing ovariohysterectomy. STUDY DESIGN: Prospective, randomized, clinical trial. ANIMALS: Ninety, healthy dogs of (mean +/- SD) body weight 21.2 +/- 10.0 kg and age 17.8 +/- 22.8 months. MATERIALS AND METHODS: After premedication with acepromazine, hydromorphone and glycopyrrolate, anesthesia was induced with thiopental administered to effect. Dogs received one of six inhalant protocols (n = 15 group): HAL; HAL/N2O; ISO; ISO/N2O; SEV; or SEV/N2O. End-tidal CO2 was maintained at 40 +/- 2 mmHg with intermittent positive pressure ventilation (IPPV). Body temperature, heart rate, indirect systemic arterial blood pressures, inspired and end-tidal CO2, volatile agent, N2O and O2 were recorded every 5 minutes. The vaporizer setting was decreased in 0.25-0.5% decrements to elicit a palpebral reflex, and this level maintained. Statistical analysis included two-way anova for repeated measures with Bonferroni's correction factor and statistical significance assumed when p < 0.05. Percentage reduction in end-tidal volatile agent was calculated at 60 minutes after starting study. RESULTS: End-tidal HAL, ISO and SEV decreased when N2O was administered. Percentage reduction: HAL (12.4%); ISO (37.1%) and SEV (21.4%). Diastolic, mean and systolic blood pressures increased in ISO/N2O compared with ISO. Heart rate increased in ISO/N2O and SEV/N2O compared with ISO and SEV, respectively. Systolic, mean and diastolic blood pressures increased in SEV compared with HAL and ISO. Systolic, mean, diastolic blood pressures and heart rate increased in SEV/N2O and ISO/N2O compared with HAL/N2O. CONCLUSIONS: N2O reduces HAL, ISO and SEV requirements in dogs undergoing ovariohysterectomy. Cardiovascular stimulation occurred when N2O was used with ISO, less so with SEV and not with HAL     

The effects of nitrous oxide on recovery from isoflurane anaesthesia in dogs

O bjectives : To assess rate and quality of recovery from anaesthesia where isoflurane was delivered in oxygen or oxygen/nitrous oxide.
M ethods : Dogs anaesthetised with propofol were randomly allocated to receive isoflurane maintenance in either 100 per cent oxygen (group 1) or 66 per cent nitrous oxide (N₂O)/34 per cent oxygen (group 2). Time from end of anaesthesia to achieving sternal recumbency was recorded. Incidence of adverse behaviours (vocalisation, uncontrolled head movement and restlessness) were assessed. Recovery quality was recorded on a visual analogue scale (VAS) (anchored at 0 with "best possible" recovery and "did not recover" at 100 mm). Age, weight, gender, anaesthetic duration, mean vaporiser setting, VAS scores, recovery times, postoperative temperature and behavioural scores were compared (chi-squared test, Mann-Whitney U test or t -test as appropriate, significance P≤0·05).
R esults : Objective data from 54 dogs were analysed, only VAS data where the observer was unaware of treatment group were used (n=33). Recovery was faster in group 2 dogs (median 10 min [range 4 to 31] compared with 14 minutes [3 to 43] in group 1, P=0·049) with less restlessness (0 [0 to 4] compared with 2 [0 to 4] in group 1, P=0·013) and uncontrolled head movement (0 [0 to 4] compared with 1 [0 to 3] in group 1, P<0·001). However, VAS scores were not statistically different between groups (group 1: mean 39·4 mm [s.d. 24·0)]; group 2: 30·1 mm [25·9]; P=0·303).
C linical S ignificance : Addition of N₂O to isoflurane anaesthesia results in a lower incidence of adverse behaviour (for example restlessness) and marginally faster recovery.     

Cardiopulmonary effects of fentanyl-droperidol, nitrous oxide, and atropine sulfate in dogs.

The cardiopulmonary effects of droperidol-fentanyl, nitrous oxide, and atropine were evaluated in 12 adult male Beagle dogs. All dogs were surgically instrumented with a cardiac output thermistor and arterial and venous catheters and were prepared with a chronic tracheostomy. Each dog was used as its own control, and data obtained when dogs were nonanesthetized and nonmedicated were compared with data recorded after the test drugs were administered. The dogs were randomly allotted to 3 groups of 4 dogs each. Group I dogs were given droperidol-fentanyl alone intravenously (IV); group II dogs were given droperidol-fentanyl IV with 67% nitrous oxide; and group III dogs were given atropine sulfate intramuscularly followed by droperidol-fentanyl IV with 67% nitrous oxide. Minute volume was decreased in the 3 groups of dogs for 3 to 5 minutes after droperidol-fentanyl was injected. This resulted in respiratory and metabolic acidosis in all dogs, as indicated by increased arterial carbon dioxide tension, decreased pH, and increased base deficit. In addition, droperidol-fentanyl given alone caused a decrease in systolic pressure and a slight decrease in heart rate. Group 1 dogs were sensitive to auditory stimulation. Cardiovascular changes were not seen when nitrous oxide was added; however, analgesia and muscle relaxation were improved. Premedication with atropine sulfate resulted in increased cardiac output, heart rate, and diastolic pressure, and subsequent administration of droperidol-fentanyl with nitrous oxide caused a transient increase in mean arterial and systolic pressure. This last anesthetic regimen, along with assisted or controlled respiration, seems to provide an excellent anesthetic state with minimal cardiopulmonary depression.     

The pharmacokinetics of ketoconazole and its effects on the pharmacokinetics of midazolam and fentanyl in dogs

KuKanich, B., Hubin, M. The pharmacokinetics of ketoconazole and its effects on the pharmacokinetics of midazolam and fentanyl in dogs. J. vet. Pharmacol. Therap . 33 , 42–49.
Ketoconazole inhibits the Cytochrome P450 3A12 (CYP3A12) metabolizing enzyme as well as the p-glycoprotein efflux pump. The extent and clinical consequence of these effects are poorly understood in dogs. The objective was to assess the pharmacokinetics of ketoconazole after single and multiple doses and the effect of multiple doses of ketoconazole on midazolam (a known CYP3A12 substrate) and the opioid fentanyl. Six greyhound dogs were studied. The study consisted of three phases. Phase 1 consisted of i.v. midazolam (0.23 mg/kg base) and fentanyl (15.71 μg/kg base). Phase 2 consisted of a single oral dose of ketoconazole (mean dose 12.34 mg/kg). Phase 3 consisted of i.v. midazolam (0.23 mg/kg) and fentanyl (10 μg/kg) after 5 days of oral ketoconazole (12.25 mg/kg/day). Ketoconazole significantly inhibited its own elimination with the mean residence time ( MRT ) increasing from 6.24 h in Phase 1 to 12.54 h in Phase 3. Ketoconazole significantly decreased the elimination of midazolam, as expected, with the MRT increasing from 0.81 to 1.49 h. The elimination of fentanyl was not significantly altered by co-administration of ketoconazole with the MRT being 3.90 and 6.35 h. The MRT was the most robust estimate of decreased drug elimination.     

Thiopental and halothane dose-sparing effects of magnesium sulphate in dogs

OBJECTIVE :To evaluate the effect of pre- and intraoperatively administered magnesium sulphate (MgSO(4)) on the induction dose of thiopental and of halothane for maintenance of anaesthesia in dogs undergoing ovariohysterectomy (OHE). STUDY DESIGN: Prospective, double-blind, randomized, placebo-controlled study. ANIMALS: Forty-six healthy, ASA physical status 1 dogs, scheduled for elective OHE. METHODS: The dogs were randomly assigned to receive a bolus of 50 mg kg(-1) MgSO(4) intravenously (IV), just before induction of anaesthesia, followed by a constant rate infusion (CRI) of 12 mg kg(-1) hour(-1) MgSO(4) intraoperatively (group Mg, n = 27) or a placebo bolus and CRI of 0.9% sodium chloride (NaCl) (group C, n = 19), approximately 30 minutes after premedication with acepromazine (0.05 mg kg(-1), intramuscularly, IM) and carprofen (4 mg kg(-1), subcutaneously, SC). Anaesthesia was induced with thiopental administered to effect and maintained with halothane in oxygen. End-tidal halothane (ET(hal)) was adjusted to achieve adequate depth of anaesthesia. Blood samples were obtained pre- and postoperatively for measurement of total serum magnesium concentration. RESULTS: The mean dose of thiopental was statistically lower (p < 0.0005) and the mean standardized ET(hal) concentration and end-tidal carbon dioxide partial pressure (Pe'CO(2)) areas under the curve were statistically smaller (p < 0.0005 and 0.014 respectively) in group Mg. Postoperatively the mean total serum magnesium concentration was statistically higher than the preoperative value (p < 0.0005) in group Mg, but not in group C. Nausea, associated with the MgSO(4) bolus injection, was observed in six dogs in group Mg, two of which vomited prior to induction of anaesthesia. CONCLUSIONS AND CLINICAL RELEVANCE: Magnesium sulphate administration reduced the induction dose of thiopental and ET(hal) concentration for maintenance of anaesthesia in dogs undergoing OHE. Observed side effects were nausea and vomiting.     

Comparison of two combinations of sedatives before anaesthetising pigs with halothane and nitrous oxide

Two groups of 21 three-month-old Landrace x Large White pigs were sedated with either azaperone (2 mg/kg), butorphanol (0.2 mg/kg) and ketamine (5 mg/kg) (group A), or detomidine (100 microg/kg), butorphanol (0.2 mg/kg) and ketamine (5 mg/kg) (group D) administered intramuscularly, before being anaesthetised with halothane, oxygen and nitrous oxide for a bilateral stifle arthrotomy. The pigs' heart rate, respiratory rate, mean arterial blood pressure, electrocardiogram, arterial oxygen saturation, arterial blood gases, and oesophageal and rectal temperature were measured while they were anaesthetised and five minutes after they were disconnected, and their recovery times and any complications were recorded. Both groups were well sedated. Their heart rate was unchanged during the period of anaesthesia but increased when they recovered. The respiratory rate, mean arterial blood pressure and rectal temperature were lower in group A than in group D (P<0.05). Mild respiratory acidosis developed during anaesthesia in both groups. Both groups recovered equally rapidly and complications were generally minor, though two pigs in group D appeared to develop malignant hyperthermia.     

The effects of ketoconazole and cimetidine on the pharmacokinetics of oral tramadol in greyhound dogs

Tramadol is administered to dogs for analgesia but has variability in its extent of absorption, which may hinder its efficacy. Additionally, the active opioid metabolite (M1) occurs in low concentrations. The purpose of this study was to determine if administration of oral tramadol with suspected metabolism inhibitors (ketoconazole, cimetidine) would lead to improved bioavailability of tramadol and M1. Six healthy Greyhounds were included. They were administered tramadol orally and intravenously, M1 intravenously, oral tramadol with oral ketoconazole and oral tramadol with oral cimetidine. Oral tramadol bioavailability was low (2.6%). Ketoconazole and cimetidine significantly increased tramadol bioavailability to 18.2% and 20.3%, respectively. The mean maximum plasma concentration of tramadol alone was 22.9 ng/ml, and increased to 109.9 and 143.2 ng/ml with ketoconazole and cimetidine, respectively. However, measured tramadol plasma concentrations were below the minimum concentration considered effective in humans (228 μg/ml). In all treatment groups, measured M1 concentrations (<7 μg/ml) were below concentrations associated with efficacy in humans. To conclude, tramadol and M1 concentrations were low and variable in dogs after oral dosing of tramadol, even in combination with cimetidine or ketoconazole, but effective concentrations in dogs have not been defined.     

Hemodynamic effects of nitrous oxide in isoflurane-anesthetized cats

OBJECTIVE: To determine the hemodynamic effects of nitrous oxide in isoflurane-anesthetized cats. ANIMALS: 12 healthy adult domestic shorthair cats. PROCEDURE: Cats were anesthetized by administration of isoflurane in oxygen. After instruments were inserted, end-tidal isoflurane concentration was set at 1.25 times the individual minimum alveolar concentration, and nitrous oxide was administered in a Latin-square design at 0, 30, 50, and 70%. Each concentration was administered for 25 minutes before measurements were obtained to allow for stabilization. Heart rate; systemic and pulmonary arterial pressures; central venous pressure; pulmonary artery occlusion pressure; cardiac output; body temperature; arterial and mixed-venous pH, PCO2, PO2, and hemoglobin concentrations; PCV; and total protein and lactate concentrations were measured before and during noxious stimulation for each nitrous oxide concentration. Arterial and mixed-venous bicarbonate concentrations and oxygen saturation, cardiac index, stroke index, rate-pressure product, systemic and pulmonary vascular resistance indices, left and right ventricular stroke work indices, arterial and mixed-venous oxygen contents, oxygen delivery, oxygen consumption, oxygen extraction ratio, alveolar-to-arterial oxygen difference, and venous admixture were calculated. RESULTS: Arterial pressure, central venous pressure, pulmonary arterial pressure, rate-pressure product, systemic and pulmonary vascular resistance indices, arterial PCO2, and PCV increased during administration of 70% nitrous oxide. Arterial and mixed-venous pH, mixed-venous PO2, and alveolar-to-arterial oxygen difference decreased during administration of 70% nitrous oxide. Results before and during noxious stimulation were similar. CONCLUSIONS AND CLINICAL RELEVANCE: Administration of 70% nitrous oxide to isoflurane-anesthetized cats resulted in improved arterial pressure, which was related to a vasoconstrictive effect.     

Comparison of the effects of thiopentone and propofol on the electrocardiogram of dogs

Sixty-four dogs were randomly assigned to receive either thiopentone or propofol and their electrocardiograms were recorded immediately before and shortly after they were anaesthetised. Thiopentone caused a marked increase in QT and JT intervals, a flattening of the T-wave and an increase in precordial QT dispersion. Propofol induced a less marked increase in QT and JT intervals, corrected for heart rate. Both agents induced an increase in heart rate and a decrease in heart rate variability, consistent with reduced vagal tone. Shortly after anaesthesia was induced, thiopentone affected ventricular repolarisation to a far greater extent than propofol, changes which suggest that it may be more likely to induce re-entrant ventricular arrhythmogenesis and could be associated with an increase in sympathetic tone. Propofol may therefore be more suitable than thiopentone for dogs with a susceptibility to ventricular arrhythmias or a long QT interval.     

Arrhythmogenicity of dopamine and its effects on hemodynamics in dogs under halothane anesthesia

The arrhythmogenicity of dopamine, its effects on cardiac function, hemodynamics, and diuresis under halothane anesthesia were evaluated in dogs. The induction time of arrhythemias and the effect of arrhythmias on cardiac function, hemodynamics, and diuresis were determined after infusion of dopamine for 30-min period at increasing doses of 3, 5, 7, 10, and 15 micrograms/kg/min. The results were as follows. 1. Arrhythmia induction percentage was 28.6% at 5 micrograms/kg/mn, 42.9% at 7 micrograms/kg/min, 25% at 10 micrograms/kg/min, and 41.7% at 15 micrograms/kg/min. The induction time of arrhythemias (sec) was 459 at 5 micrograms/kg/min, 332 at 7 micrograms/kg/min, 152 at 10 micrograms/kg/min, and 279 at 15 micrograms/kg/min. No arrhythmias were present at 3 micrograms/kg/min. 2. Heart rate and myocardial oxygen consumption was increased in the arrhythmia-induced group compared to the non-arrhythmia-induced group. 3. Myocardial contractility, mean aortic pressure, mean pulmonary arterial pressure, and diuresis increased dose-dependently in the non-arrhythmia-induced group; however, these measures were increased in the arrhythmia-induced group without regard to dose.     

Hepatic effects of halothane, isoflurane or sevoflurane anaesthesia in dogs

The effects of halothane, isoflurane and sevoflurane anaesthesia on hepatic function and hepatocellular damage were investigated in dogs, comparing the activity of hepatic enzymes and bilirubin concentration in serum. An experimental study was designed. Twenty-one clinically normal mongrel dogs were divided into three groups and accordingly anaesthetized with halothane (n = 7), isoflurane (n = 7) and sevoflurane (n = 7). The dogs were 1-4 years old, and weighed between 13.5 and 27 kg (18.4 +/- 3.9). Xylazine HCI (1-2 mg/kg) i.m. was used as pre-anaesthetic medication. Anaesthesia was induced with propofol 2 mg/kg i.v. The trachea was intubated and anaesthesia maintained with halothane, isoflurane or sevoflurane in oxygen at concentrations of 1.35, 2 and 3%, respectively. Intermittent positive pressure ventilation (tidal volume, 15 ml/kg; respiration rate, 12-14/min) was started immediately after intubation and the anaesthesia lasted for 60 min. Venous blood samples were collected before pre-medication, 24 and 48 h, and 7 and 14 days after anaesthesia. Serum level of aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP) and gamma-glutamyltransferase (GGT), lactate dehydrogenase (LDH GGT) activities and bilirubin concentration were measured. Serum AST, ALT and GGT activities increased after anaesthesia in all groups. In the halothane group, serum AST and ALT activities significantly increased all the time after anaesthesia compared with baseline activities. But in the isoflurane group AST and ALT activities increased only between 2 and 7 days, and in the sevoflurane group 7 days after anaesthesia. GGT activity was increased in the halothane group between 2 and 7 days, and in the isoflurane and sevoflurane groups 7 days after anaesthesia. All dogs recovered from anaesthesia without complications and none developed clinical signs of hepatic damage within 14 days. The results suggest that the use of halothane anaesthesia induces an elevation of serum activities of liver enzymes more frequently than isoflurane or sevoflurane from 2 to 14 days after anaesthesia in dogs. The effects of isoflurane or sevoflurane anaesthesia on the liver in dogs is safer than halothane anaesthesia in dogs.     

Comparison of the hemodynamic effects of halothane alone and halothane combined with epidurally administered morphine for anesthesia in ventilated dogs

The hemodynamic effects of 1.5 minimal alveolar concentration of halothane alone (1.6% end-tidal) and 1.5 minimal alveolar concentration of halothane (1.1% end-tidal concentration) combined with epidurally administered morphine were compared during controlled ventilation in 10 dogs used on 2 occasions and randomly allocated to 2 groups. Arterial blood pressure, cardiac index, stroke volume, left ventricular work, and pulmonary arterial pressure were significantly (P less than 0.05) higher in dogs of the morphine-treated group before administration of morphine. After epidural administration of morphine (0.1 mg/kg of body weight diluted in 0.26 ml of saline solution/kg), hemodynamic changes were not observed, and the aforementioned variables remained significantly (P less than 0.05) higher than values in dogs of the halothane only group. Compared with halothane (1.6%) alone, the reduction in halothane end-tidal concentration (1.1%) associated with epidurally administered morphine is beneficial in maintaining hemodynamic function.     

Acute cardiovascular effects and pharmacokinetics of carvedilol in healthy dogs

OBJECTIVE: To determine acute cardiovascular effects and pharmacokinetics of carvedilol in healthy dogs. ANIMALS: 14 mature healthy Beagles. PROCEDURE: 12 dogs were anesthetized with morphine and alpha-chloralose. Catheters were placed in the aorta, left ventricle, and right atrium to record systemic and pulmonary pressures and determine vascular resistance and cardiac output. Electrocardiograms (leads I, aVF, and V3) were recorded to determine electrocardiographic changes. Variables were measured before and after IV injection of incremental doses of carvedilol (cumulative doses, 10, 30, 70, 150, 310, and 630 microg/kg of body weight; n = 6) or vehicle alone (6). Pharmacokinetic analysis was performed, using 2 conscious dogs given 160 microg of carvedilol/kg as a single IV injection. RESULTS: Heart rate and velocity of fiber shortening at zero load (Vmax) increased slightly but significantly from baseline values at doses of carvedilol > or = 310 microg/kg and 10 microg/kg, respectively. Carvedilol did not affect systemic and pulmonary pressures or vascular resistances. Intravenous administration of approximately 150 microg of carvedilol/kg resulted in a plasma carvedilol concentration of approximately 100 ng/ml. Mean elimination half-life was 54 minutes, half-life of distribution was 3.5 minutes, and volume of distribution was 2,038 ml/kg. CONCLUSIONS AND CLINICAL RELEVANCE: The therapeutic plasma concentration of carvedilol in humans is 100 ng/ml. At that plasma concentration in dogs, the reduction in afterload and positive inotropic effect that we observed would be beneficial for treating heart failure and minimizing the cardiotoxic effects of doxorubicin.     

Cardiovascular and respiratory effects of propofol administration in hypovolemic dogs.

Cardiopulmonary effects of propofol were studied in hypovolemic dogs from completion of, until 1 hour after administration. Hypovolemia was induced by withdrawal of blood from dogs until mean arterial pressure of 60 mm of Hg was achieved. After stabilization at this pressure for 1 hour, 6 mg of propofol/kg of body weight was administered IV to 7 dogs, and cardiopulmonary effects were measured. After blood withdrawal and prior to propofol administration, oxygen utilization ratio increased, whereas mean arterial pressure, mean pulmonary arterial pressure, central venous pressure, pulmonary capillary wedge pressure, cardiac index, oxygen delivery, mixed venous oxygen tension, and mixed venous oxygen content decreased from baseline. Three minutes after propofol administration, mean pulmonary arterial pressure, pulmonary vascular resistance, oxygen utilization ratio, venous admixture, and arterial and mixed venous carbon dioxide tensions increased, whereas mean arterial pressure, arterial oxygen tension, mixed venous oxygen content, arterial and mixed venous pH decreased from values measured prior to propofol administration. Fifteen minutes after propofol administration, mixed venous carbon dioxide tension was still increased; however by 30 minutes after propofol administration, all measurements had returned to values similar to those measured prior to propofol administration.     

The effect of a 50% inspired mixture of nitrous oxide on arterial oxygen tension in spontaneously breathing horses anaesthetised with halothane

The effect of nitrous oxide (N₂O) on arterial partial pressure of oxygen (P_aO₂) was evaluated in 20 adult horses anaesthetised with halothane. A fresh gas flow rate of 20ml/kg/min, comprising a 1:1 N₂O/oxygen (O₂) mixture, was supplied via the rotameter flowmeters of an anaesthetic machine to a large animal breathing system. The horses breathed spontaneously from the circuit immediately after endotracheal intubation. Ten horses were subsequently positioned in lateral recumbency and ten in dorsal recumbency. A further twenty adult horses were anaesthetised with halothane and acted as controls; halothane in 20mls/kg/min of O₂ being supplied to the same breathing system. Fifty percent NO caused significant decreases in P_aO₂ for horses in lateral and dorsal recumbency. However when administered to horses in lateral recumbency it did not promote arterial hypoxaemia. There was a higher risk of intraopera- tive arterial hypoxaemia (P_aO₂ < 8.6kPa) associated with its use in spontaneously breathing horses in dorsal recumbency. Arterial hypoxaemia occurred in all horses during the first fifteen minutes of recovery but when N₂O was discontinued, halothane in oxygen supplied to the breathing circuit for five minutes at a flow rate of 20ml/kg/minute was sufficient to ensure that diffusion hypoxia did not occur. The magnitude of the hypoxaemia was not signficantly different between the groups. The time taken to adopt sternal recumbency was significantly shorter in the horses that had received N₂O.