Materials and Methods

Twelve healthy Thoroughbreds were allocated to 1 of 2 groups to undergo P-TIVA or KMP-TIVA. The P-TIVA group included 5 mares and 1 stallion; the mean ± SD weight of these horses was 502 ± 44 kg (1,104 ± 97 lb; range, 442 to 566 kg [972 to 1,245 lb]), and mean age was 10.7 ± 6.2 years (range, 2 to 18 years). The KMP-TIVA group included 4 mares and 2 stallions; mean weight was 506 ± 31 kg (1,113 ± 68 lb; range, 454 to 540 kg [999 to 1,188 lb]), and mean age was 11.8 ± 5 years (range, 5 to 18 years). Food, but not water, was withheld from the horses for 12 hours before anesthesia. The horses were owned by the university and were cared for according to the principles of the Guide for the Care and Use of Laboratory Animals prepared by Rakuno Gakuen University. The Animal Care and Use Committee of Rakuno Gakuen University approved the study.

Each of the horses was premedicated with medetomidine^a (0.005 mg/kg [0.002 mg/lb], IV) via a 14-gauge, 13.3-cm catheter placed percutaneously into a jugular vein. Five minutes later, anesthesia was induced with midazolam^b (0.04 mg/kg [0.018 mg/lb], IV) and ketamine^c (2.5 mg/kg [1.14 mg/lb], IV). All horses were orotracheally intubated and breathed 100% oxygen. A loading dose of propofol^d (0.5 mg/kg [0.23 mg/lb], IV) was administered to horses in both groups. In the P-TIVA group, an IV infusion of propofol was started. In the KMP-TIVA group, a CRI of ketamine (1 mg/kg/h [0.45 mg/lb/h]) and medetomidine (0.00125 mg/kg/h [0.0006 mg/lb/h]) and an IV infusion of propofol were started; the CRI of the drug combination was administered by use of a syringe infusion pump,^e and the infusion of propofol was adjusted as needed to prevent a response to surgical stimulation. In both groups, propofol was administered via an infusion pump.^f The initial rates of propofol infusion were 0.2 mg/kg/min (0.09 mg/lb/min) during KMP-TIVA and 0.3 mg/kg/min (0.14 mg/lb/min) during P-TIVA, which were chosen on the basis of data from preliminary trials and published reports.^9,10,12,14 The rate of propofol infusion was increased by 0.025 mg/kg/min (0.011 mg/lb/min) when a purposeful (after stimulation) or spontaneous (nonstimulated) movement occurred; the infusion rate was increased in increments until the movement was no longer detected. A bolus of propofol (200 mg, IV) was administered when the movement was difficult to control by increasing the infusion rate of propofol. The rate of propofol infusion was decreased by 0.025 mg/kg/min when purposeful movement was not detected; the infusion rate was decreased in increments until the lowest effective level was reached.

After orotracheal intubation, the tube was connected to a large-animal circle system^g that incorporated a ventilator.^h Lactated Ringer's solutionⁱ was administered IV at a rate of 10 mL/kg/h (4.5 mL/lb/h) to all horses. Approximately 20 to 30 minutes after induction of anesthesia, the right carotid artery of each horse was relocated to a subcutaneous position as the horse lay in left lateral recumbency. In the P-TIVA and KMP-TIVA groups, drug infusions were stopped once surgery was completed and the horses were moved to a padded recovery box. After surgery, flunixin meglumine^j (1 mg/kg, IV) and penicillin G procaine (4 × 10⁶ U/horse, IM) combined with dihydrostreptomycin sulfate^k (5 g/horse, IM) were administered every 12 hours for 3 days. Qualities of anesthetic induction, transition to TIVA (the first 20-minute period of anesthesia), maintenance of anesthesia (> 20 minutes to the end of anesthesia [ie, cessation of drug infusions]), and recovery from anesthesia were categorized by use of a subjective scoring system (Appendix).

Baseline heart rate and respiratory rate were measured in all horses standing in stocks before any medication was administered. Once the horses were positioned in lateral recumbency, an 18-gauge catheter was placed in the dorsal third metatarsal artery. Arterial blood pressure was measured by connecting the catheter to a pressure transducer placed at the level of the left atrium (midsternum region) and zeroed in this position. An arterial blood sample was anaerobically collected from the catheter into a syringe containing heparin at 20-minute intervals; blood gas and pH analyses were performed immediately by use of a blood gas analyzer.^l Breathing was controlled via IPPV if the duration of apnea after induction of anesthesia was > 60 seconds. The ventilator was adjusted to maintain PaCO₂ at 40 to 50 mm Hg; the respiratory rate was set at 6 breaths/min, tidal volume at 15 mL/kg (6.8 mL/lb), and inspiration-to-expiration time of 1:2 or 1:3. During anesthesia, an apex-base lead ECG and heart rate and blood pressure values were recorded by the anesthetic monitoring system^m at 10 minutes after induction and 20-minute intervals thereafter. Duration of anesthesia was recorded as the interval from IV injection of midazolam and ketamine to discontinuation of the propofol infusion. Recovery from anesthesia began after the cessation of the infusion. During the recovery period, the times to extubation, first movement, attainment of sternal recumbency, and standing after the cessation of anesthesia were recorded. The quality of recovery was evaluated by use of a scoring system (Appendix). Observers were aware of the group allocation of each horse.

Statistical analysis—Data are reported as mean ± SD values. A repeated-measures ANOVA was used for comparison of changes in propofol infusion rate and cardiorespiratory data. A Mann-Whitney U test was also used to compare the infusion rate of propofol, quality of anesthesia (ie, the scores for anesthetic induction, transition to TIVA, and maintenance of anesthesia), and characteristics of recovery from anesthesia between groups. Differences were considered significant at a value of P < 0.05.

Results

Induction of anesthesia was smooth and excitement-free with adequate muscle relaxation and was subjectively scored as excellent in all horses. The interval from injection of ketamine-midazolam to recumbency was 1 to 2 minutes, and there were no limb movements or head shaking after the horses became recumbent. Within 2 minutes following administration of the loading dose of propofol, respiratory rate decreased and apnea was detected in all horses; IPPV was required for the duration of anesthesia. At 10 minutes after induction of anesthesia (prior to starting IPPV), PaCO₂ increased in all horses; in the KMP-TIVA group, the value ranged from 47 to 77 mm Hg, and in the P-TIVA group, the value ranged from 58 to 93 mm Hg. Hypoxemia developed in 3 horses undergoing KMP-TIVA (PaO₂ range, 27 to 77 mm Hg) and 4 horses undergoing P-TIVA (PaO₂ range, 26 to 64 mm Hg). Mean time after induction of anesthesia to start of IPPV was 21.0 ± 1.6 minutes and 23.8 ± 4.9 minutes in the KMP-TIVA and P-TIVA groups, respectively.

The transitions to infusion of propofol and the ketamine-medetomidine-propofol combination were smooth and subjectively scored as excellent in all horses. Both KMP-TIVA and P-TIVA provided satisfactory anesthesia for carotid artery translocation in the horses. Mean ± SD duration of anesthesia was 115 ± 17 minutes and 112 ± 11 minutes in horses undergoing KMP-TIVA and P-TIVA, respectively. The quality of maintenance of anesthesia was considered excellent in all 6 horses in the KMP-TIVA group and in 3 horses in the P-TIVA group. The other 3 horses undergoing P-TIVA required 2 to 3 bolus IV injections of propofol (200 mg/injection) to control limb movement during surgery (Table 1). Propofol infusion rate required to maintain a surgical plane of anesthesia via KMP-TIVA was significantly (P < 0.001) less than that associated with P-TIVA (Figure 1). Mean rate of propofol infusion was significantly (P = 0.004) less in horses undergoing KMP-TIVA (0.14 ± 0.02 mg/kg/min [0.064 ± 0.009 mg/lb/min]) than that in horses undergoing P-TIVA (0.22 ± 0.03 mg/kg/min [0.1 ± 0.013 mg/lb/min]).

During anesthesia, heart rate and MABP were not significantly different between groups (Table 2). Mean heart rate was maintained between 31 and 41 beats/min and MABP remained between 91 and 131 mm Hg during both treatments. Hypercarbia and hypoxemia detected at the earlier stage of anesthesia were improved by IPPV in all horses. From the 30-minute time point, mean PaO₂ was maintained between 289 and 487 mm Hg during both treatments, and mean PaCO₂ was maintained between 42 and 48 mm Hg. An increase in arterial pH occurred in both groups during the earlier stage of anesthesia.

View Full Size

Mean ± SD propofol infusion rates administered alone to 6 horses undergoing P-TIVA (closed circles) or in combination with CRI of ketamine (1 mg/kg/h [0.45 mg/lb/h]) and medetomidine (0.00125 mg/kg/h [0.0006 mg/lb/h]) in 6 horses undergoing KMP-TIVA (open circles). Value significantly (*P < 0.05; †P < 0.01) greater than the KMP-TIVA value at this time point.

Citation: Journal of the American Veterinary Medical Association 228, 8; 10.2460/javma.228.8.1221

• Download Figure
• Download figure as PowerPoint slide

After cessation of anesthesia, the times to extubation, first movement, sternal recumbency, and standing, and the number of attempts to stand were not significantly different between KMP-TIVA and P-TIVA groups (Table 1). The quality of recovery from KMP-TIVA was assessed as good, whereas the quality of recovery from P-TIVA was assessed as good or satisfactory; between groups, there was a significant (P = 0.019) difference in the recovery score.

Discussion

Medetomidine is an α₂-adrenoceptor agonist that activates α₂ receptors located in the spinal cord and brainstem, resulting in sedation and analgesia.^16,17 Medetomidine, although similar to xylazine and detomidine, was chosen for use in the present study because it is reported to induce more profound behavioral and neurochemical effects.^2,18-20 In horses, as in other species, medetomidine induces profound sedative and analgesic effects at smaller doses than xylazine^18,19 and detomidine.^20,21 Compared with xylazine administered IV at a dose of 1 mg/kg and detomidine administered IV at doses of 20 to 40 μg/kg (9.1 to 18.2 μg/lb), equipotent sedative doses of medetomidine administered IV range from 5 to 10 μg/kg (2.3 to 4.5 μg/lb).^18–20,22 Furthermore, the pharmacokinetics of medetomidine in horses make it suitable for use as an infusion; an appropriate rate of infusion results in a constant level of sedation and steady plasma medetomidine concentrations.²³ In addition, it has been reported^2,12,13 that propofol infusion rates required to maintain anesthesia decreased when combined with CRI of medetomidine in ponies. The authors of those reports on ponies and of the present study speculated that medetomidine could provide more profound sedation and analgesia in horses, compared with xylazine and detomidine.

Induction of anesthesia with ketamine after sedation with α₂-adrenoceptor agonists is common in equine practice. Ketamine, a dissociative anesthetic agent, induces analgesia by antagonizing N-methyl-D-aspartate receptors.²⁴ In horses, ketamine provides slow, controlled, and calm attainment of recumbency. Furthermore, the sympathomimetic action of ketamine²⁵ minimizes the development of bradycardia and hypotensive effects associated with drugs used as sedatives.²⁶ Ketamine decreases the minimum alveolar concentration of halothane in horses²⁷ and the minimal infusion rate of propofol in ponies^9,11; hence, ketamine was selected for use in the present study. Midazolam is a centrally acting skeletal muscle relaxant that selectively depresses the transmission of impulses at the interneurons of spinal cord, brainstem, and subcortical regions of the brain.^28,29 Midazolam was chosen because of its profound muscle relaxant and hypnotic effects, which are essential for smooth induction of anesthesia.³⁰ In the present study, the ketamine-midazolam induction of anesthesia was considered excellent; the rate of induction was not so fast as to cause injury, nor was it particularly slow with subsequent ataxia and resistance from the horse.^30,31 On the basis of these reasons, we chose to premedicate the horses with medetomidine followed by induction of anesthesia with ketamine and midazolam. Propofol was not used for induction of anesthesia in all horses (especially the P-TIVA group) because the quality of propofol-associated induction is reported to be unpredictable and unsatisfactory in horses.^5,7,13 Also, the volume of propofol required for induction of anesthesia is too large to enable rapid IV injection in horses.

Both KMP-TIVA and P-TIVA induced clinically acceptable anesthesia in the horses. Transition to infusion of KMP-TIVA or P-TIVA was without complications. Maintenance of anesthesia via KMP-TIVA was considered more satisfactory than that achieved via P-TIVA because 3 horses in the P-TIVA group required additional propofol injections to prevent limb movement during surgery. Furthermore, the propofol requirements in horses undergoing KMP-TIVA were significantly less than in horses undergoing P-TIVA; in the KMP-TIVA group, the CRI of ketamine and medetomidine resulted in 30% reduction in the propofol requirement. Results of previous studies^8,14 have indicated that the rate of propofol infusion required to maintain a surgical plane of anesthesia in horses was 0.18 to 0.15 mg/kg/min (0.081 to 0.068 mg/lb/min). Flaherty et al⁹ reported that the administration of ketamine as a CRI (2.4 mg/kg/h [1.09 mg/lb/h]) has a sparing effect on the requirement for propofol (0.12 mg/kg/min [0.055 mg/lb/min]) in ponies undergoing surgery. In addition, Bettschart-Wolfensberger et al^2,12,13 reported that mean propofol infusion rates to maintain anesthesia ranged from 0.06 to 0.1 mg/kg/min (0.027 to 0.045 mg/lb/min) when combined with the administration of medetomidine as a CRI (0.0035 mg/kg/h [0.0016 mg/lb/h]). In the present study, the propofol infusion rate required to maintain a surgical plane of anesthesia in horses was 0.14 mg/kg/min when combined with ketamine (1.0 mg/kg/h) and medetomidine (0.00125 mg/kg/h) as a CRI. This was significantly less than the infusion rate of propofol (0.22 mg/kg/min) required to maintain anesthesia with propofol alone. The sparing effect on the requirement of propofol might be caused by a combination of anesthetic and analgesic effects provided by ketamine^24,32 and medetomidine.^17,16

In the horses of our study, KMP-TIVA and P-TIVA were associated with respiratory depression and apnea resulting in hypoventilation, hypercarbia, and hypoxemia during the early stage of anesthesia. Intermittent positive-pressure ventilation was initiated approximately 20 minutes after induction of anesthesia to successfully treat these adverse effects. Considering that the horses were breathing 100% oxygen, the development of marked hypoxemia in the early stage of anesthesia is an important complication. In anesthetized horses breathing > 95% inspired oxygen concentration, PaO₂ values that were lower than expected were suggested to be the result of hypoventilation and ventilation-perfusion abnormalities caused by anesthesia-induced recumbency, pulmonary shunting, and diffusion abnormalities in addition to development of apnea.³³ Hypoxemia was similarly reported in horses during TIVA^8,14,34 and highlighted the importance of IPPV. Other authors also reported mean PaCO₂ values as high as 103 mm Hg⁸ and 16 kPa⁹ during propofol infusion in horses. Matthews et al¹⁴ detected hypoventilation and hypoxemia during detomidine-propofol anesthesia in horses and suggested that the initial decrease in respiratory rate and significant increase in PaCO₂ from baseline values may have been related to the bolus of propofol administered for induction of anesthesia or to hoisting and positioning in dorsal recumbency. It is possible that drug interaction between propofol and medetomidine³⁵ may have exacerbated the respiratory compromise in the horses of the present study. There are similar reports of apnea early in the anesthetic period¹⁴ and after the start of propofol infusion^8,10 in horses. Therefore, the development of respiratory depression and apnea must be considered prior to the administration of propofol in the horse.

However, in our study, both methods of TIVA provided anesthesia with well-maintained MABP values accompanied by no change or an increase in heart rate in adult horses receiving IPPV. Intermittent positive-pressure ventilation sometimes aggravates cardiovascular depression because of the decrease in preload.³⁶ It has been suggested that propofol may be associated with increased sympathetic tone,⁵ and the sympathomimetic action of ketamine²⁵ is reported to minimize the bradycardia and hypotensive effects associated with drugs used as sedatives.²⁶ High blood pressure values are often associated with a light plane of anesthesia; however, arterial blood pressure spikes quickly with surgical stimuli, whereas we detected a steady increase in blood pressure over time in the horses of our study. Peripheral vasoconstriction caused by medetomidine may also be a factor contributing to increased blood pressure. In the horses used in the present study, KMP-TIVA and P-TIVA caused minimal effects on heart rate and MABP; values were maintained within acceptable range (compared with values associated with inhalant anesthesia) during a surgical plane of anesthesia. This finding may have important implications in the development of TIVA techniques because the maintenance of arterial blood pressure is a key factor in maintaining adequate muscle perfusion. This finding requires further investigation and detailed measurements of muscle blood flow and cardiovascular function.

Recovery from anesthesia is influenced by many factors including individual variability, development of hypotension during anesthesia, type of surgical procedure, duration of anesthesia, external stimuli, and use of sedatives as adjunctive drugs.^36,37 In our study, recovery from P-TIVA was thought to be slow, compared with the recovery rate after KMP-TIVA, because the horses required approximately 90 minutes longer before they could stand after a 2-hour period of anesthesia. We have no explanation for this difference. Although the main advantage of propofol is the rapid recovery associated with its use, the slow recovery we detected was similar to that reported previously in horses in which anesthesia was maintained by use of propofol only. When propofol was administered at 0.18 ± 0.04 mg/kg/min, the time to standing after 61 ± 19 minutes of anesthesia was 67 ± 29 minutes¹⁴; when administered at 0.15 mg/kg/min, the time to standing after 76 ± 1 minutes of anesthesia was 90.6 ± 35.6 minutes, and at 0.25 mg/kg/min (0.114 mg/lb/min), the time to standing after 73 ± 1 minutes of anesthesia was 80.3 ± 32.8 minutes.⁸ In another study³⁴ of TIVA in horses, time to standing ranged from 33 ± 8 minutes to 69 ± 27 minutes after 66 ± 6 minutes to 73 ± 3 minutes of xylazine and ketamine anesthesia. On comparison of these data, it was suggested that the longer reported recovery times with propofol anesthesia may have been caused by longer duration of anesthesia and surgery or the fact that horses remained undisturbed in a darkened, padded recovery room with no stimulus to stand.¹⁴

It is likely that breed also plays a role in recovery from propofol anesthesia because studies (including our study and others^8,14) that used Thoroughbreds, Quarter Horses, or other horses of similar weight revealed longer times of recovery, compared with times reported for ponies.^2,9,11–13 Accumulation of propofol in tissues could also have occurred because there was a 30% reduction in propofol requirement in the horses undergoing KMP-TIVA, which probably resulted in the difference in recovery times between the 2 TIVA groups in the present study. Future investigations should evaluate ketamine and medetomidine for TIVA in horses because of the possibility that the overall quality of anesthesia might be better and recoveries faster without propofol. In our study in horses, there was no significant difference in recovery time between the P-TIVA and KMP-TIVA groups but the quality of recovery from KMP-TIVA was judged to be better than recovery from P-TIVA. It was suggested that the ketamine-medetomidine drug infusion provided an improvement in the quality of recovery from propofol anesthesia. The quality of recoveries in the horses of the present study was good to satisfactory, whereas other TIVA protocols have been associated with excellent to good recovery scores.^8,34 However, in our study, the duration of anesthesia was approximately 2 hours and surgery was also performed, unlike the other studies^8,34 in which anesthesia was maintained for approximately 1 hour and surgery was not performed.

Our results suggest that premedication with medetomidine, induction of anesthesia with ketamine and midazolam, and maintenance of anesthesia via KMP-TIVA or P-TIVA provide controllable anesthesia of satisfactory quality in horses during which heart rate and MABP remain within acceptable limits. However, IPPV was required to treat depression of respiratory function and apnea caused by propofol administration. The ketamine-medetomidine infusion provided a sparing effect on propofol requirement for maintaining a surgical plane of anesthesia in horses. Because of the presently high cost of propofol, the requirement for IPPV to treat hypoxemia and hypoventilation, and long times to recovery from anesthesia, it seems unlikely that either of these drug protocols will replace commonly used TIVA techniques in horses.