Materials and Methods

Animals—Ten adult horses with a mean ± SD body weight of 551 ± 24 kg (1,212 ± 53 lb) and assessed as healthy on the basis of results of physical and lameness examinations were included in the study. The horses were from a group of horses that required sedation 7 times for bilateral carpal arthrocenteses as part of a separate study on intra-articular administration of polyglycan.^a All animal procedures were reviewed and approved by the Colorado State University Institutional Animal Care and Use Committee.

Procedures—For each sedation episode, horses were assigned by use of a randomization procedure (order of the treatments drawn from a bowl) to receive 1 of the following 6 sedation treatments: xylazine hydrochloride^b (0.4 mg/kg [0.18 mg/lb]; treatment X [n = 10]), xylazine (0.4 mg/kg) plus butorphanol tartrate^c (0.01 mg/kg [0.0045 mg/lb]; treatment XB [11]), xylazine (0.4 mg/kg) plus a lower dose of ketamine hydrochloride^d (0.1 mg/kg [0.045 mg/lb]; treatment XK1 [11]), xylazine plus a higher dose of ketamine (0.2 mg/kg [0.09 mg/lb]; treatment XK2 [11]), xylazine (0.4 mg/kg) plus butorphanol (0.01 mg/kg) plus the lower dose of ketamine (0.1 mg/kg; treatment XBK1 [10]), or xylazine (0.4 mg/kg) plus butorphanol (0.01 mg/kg) plus the higher dose of ketamine (0.2 mg/kg; treatment XBK2 [11]). Each horse was sedated by administration of at least 4 of the 6 treatments, and no horse received a specific treatment more than twice. All drugs were administered IV via an 18-gauge catheter that was previously placed aseptically in a jugular vein. For all treatments, xylazine or xylazine-butorphanol was administered first (time 0), followed 5 minutes later by administration of one of the doses of ketamine or saline (0.9% NaCl) solution.

Evaluation of response to stimuli—Various behaviors were evaluated by an investigator (EKC) who was unaware of the sedation treatment administered to each horse. Behaviors were evaluated 2 minutes before drug administration (−2 minutes [baseline]), 3 minutes after administration of xylazine or xylazine-butorphanol, and 5, 10, and 15 minutes after administration of the dose of ketamine or saline solution (time = 10, 15, and 20 minutes, respectively). The head position (height in centimeters from the floor to the poll) of each horse was measured at each time point prior to performing behavioral evaluations. Behavioral evaluations included response to abrupt noise (produced by clanging 2 metal rings together), response to insertion of a dental float in the oral cavity, and response to application of a 20-gauge needle to the flank area (ie, needle prick). Each behavioral response was subjectively scored by the same evaluator (EKC) in accordance with the following scoring system: 1 = no observable response, 2 = slight or barely perceptible response (eg, ear twitched or horse moved slightly), 3 = mild response (eg, horse calmly raised its head), 4 = moderate response (eg, horse raised its head briskly and ears were erect or laid back), or 5 = dramatic response (eg, horse was alert and moved, kicked, or bit). Response to pressure applied over the spine of the scapula was evaluated by recording the algometer^e pressure necessary to cause the horse to step away or shift weight from one forelimb to the other.

For each horse, the skin over the dorsal and lateral surfaces of both carpal joints (which had been previously clipped of hair) was aseptically prepared. Approximately 2 minutes after administration of a dose of ketamine or saline solution (time = 7 minutes), arthrocentesis was performed on one carpus, which was followed immediately by arthrocentesis of the other carpus. The investigator lifted the forelimb and held the carpus in a partially flexed position. A 20-gauge, 1.5-inch needle then was inserted into the middle carpal joint at a point just axial to the tendonous portion of the common digital extensor muscle. A 6-mL syringe was attached to the needle, and several milliliters of synovial fluid was aspirated from the joint. Three investigators performed the bilateral carpal arthrocenteses; all were unaware of the sedation treatment administered to each horse. One of these 3 investigators evaluated the response to arthrocenteses on the basis of the horse's movement, head position, and reaction to needle placement and fluid aspiration, using the same subjective 5-point numeric scale described for the behavioral responses. For each horse, a single overall score was assigned for the bilateral arthrocenteses. The investigator who performed each arthrocentesis procedure also had the option of having a twitch applied to the upper lip of any horse that was not sufficiently sedated; the decision to use or not to use a twitch was recorded.

Statistical analysis—Data for each evaluation were analyzed as a randomized block design or a randomized block design with repeated measures over time, with horse as the block effect. The baseline value was included in the model as a covariate. Data were analyzed by use of commercially available software.^f When significant effects were detected, post hoc comparisons of the least squares means were performed by use of Tukey adjusted P values. Probability that a twitch was used was analyzed as a binary variable.^g Differences among sedation treatments were considered significant at values of P < 0.05.

Results

For 4 sedation episodes, a horse was not sufficiently sedated to allow carpal arthrocenteses (1 horse when administered treatment X, 1 horse when administered treatment XK2, and 1 horse when administered treatment X and that same horse when administered treatment XBK2). These 3 horses received additional xylazine and butorphanol to allow investigators to complete the procedures; data for these 4 episodes were not included in the statistical analysis.

Although all sedation treatments resulted in lowering of the head, there was no significant difference in head height among treatments (Table 1). There were no significant changes in response to noise among treatments (Table 2). Insertion of a dental float was significantly easier after XBK2 treatment (mean ± SE at 10 minutes, 1.5 ± 0.4), compared with the ease after treatments XK1 (2.4 ± 0.4) and XK2 (2.6 ± 0.4), whereas treatments X, XB, and XBK1 were associated with responses to insertion of the dental float that were intermediate (1.9 ± 0.3 to 2.2 ± 0.4) and that did not differ significantly from values for the other treatments (Table 3). Response to a needle prick on the flank was significantly lower after XB treatment (mean ± SE at 10 minutes, 1.7 ± 0.3), compared with the response after XK2 treatment (2.8 ± 0.4); all other treatments were associated with intermediate responses (1.8 ± 0.4 to 2.4 ± 0.3) that were not significantly different from the response for the XB or XK2 treatments (Table 4). Sedation with the XBK2 treatment was associated with tolerance of significantly greater algometer pressure (mean ± SE at 10 minutes, 6.5 ± 0.6 kg/cm²), compared with the response after the X treatment (4.0 ±0.5 kg/cm²); all other treatments were associated with intermediate responses (4.6 ± 0.5 kg/cm² to 5.0 ±0.5 kg/cm²) that were not significantly different from the response for treatments XBK2 or X (Table 5). There was no significant difference among treatments for the response to arthrocentesis (mean ± SE, 2.0 ± 0.3 to 2.8 ± 0.3; Figure 1). A twitch was applied for 11 of 65 (17%) arthrocenteses, but there was no difference among sedation treatments for the likelihood that a twitch would be required. There were no significant differences among treatments for behavioral responses at time points after the 10-minute evaluation, and behavioral evaluations were discontinued after the 20-minute time point because horses had returned to nearly normal behavior by that time.

Response (behavioral score) for carpal arthrocentesis in horses sedated by administration of xylazine hydrochloride (0.4 mg/kg [0.18 mg/lb]; treatment X [n = 10]; white bar with black dots), xylazine (0.4 mg/kg) plus butorphanol tartrate (0.01 mg/kg [0.0045 mg/lb]; treatment XB [11]; black bar with white dots), xylazine (0.4 mg/kg) plus butorphanol (0.01 mg/kg) plus a lower dose of ketamine (0.1 mg/kg [0.045 mg/lb]; treatment XBK1 [10]; vertical-striped bar), xylazine (0.4 mg/kg) plus butorphanol (0.01 mg/kg) plus a higher dose of ketamine (0.2 mg/kg [0.09 mg/lb]; treatment XBK2 [11]; black bar), xylazine (0.4 mg/kg) plus the lower dose of ketamine hydrochloride (0.1 mg/kg; treatment XK1 [11]; diagonal-striped bar), or xylazine plus the higher dose of ketamine (0.2 mg/kg; treatment XK2 [11]; horizontal-striped bar). All drugs were administered IV. For all treatments, xylazine or xylazine-butorphanol was administered first (time 0), followed 5 minutes later by administration of one of the doses of ketamine or saline (0.9% NaCl) solution; arthrocentesis was performed at time = 7 minutes. The behavioral score was assigned by use of a 5-point system (1 = no observable response, 2 = slight or barely perceptible response, 3 = mild response, 4 = moderate response, and 5 = dramatic response). Response scores did not differ significantly (P ≥ 0.05) among treatments.

Citation: Journal of the American Veterinary Medical Association 238, 12; 10.2460/javma.238.12.1629

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Response (behavioral score) for carpal arthrocentesis in horses sedated by administration of xylazine hydrochloride (0.4 mg/kg [0.18 mg/lb]; treatment X [n = 10]; white bar with black dots), xylazine (0.4 mg/kg) plus butorphanol tartrate (0.01 mg/kg [0.0045 mg/lb]; treatment XB [11]; black bar with white dots), xylazine (0.4 mg/kg) plus butorphanol (0.01 mg/kg) plus a lower dose of ketamine (0.1 mg/kg [0.045 mg/lb]; treatment XBK1 [10]; vertical-striped bar), xylazine (0.4 mg/kg) plus butorphanol (0.01 mg/kg) plus a higher dose of ketamine (0.2 mg/kg [0.09 mg/lb]; treatment XBK2 [11]; black bar), xylazine (0.4 mg/kg) plus the lower dose of ketamine hydrochloride (0.1 mg/kg; treatment XK1 [11]; diagonal-striped bar), or xylazine plus the higher dose of ketamine (0.2 mg/kg; treatment XK2 [11]; horizontal-striped bar). All drugs were administered IV. For all treatments, xylazine or xylazine-butorphanol was administered first (time 0), followed 5 minutes later by administration of one of the doses of ketamine or saline (0.9% NaCl) solution; arthrocentesis was performed at time = 7 minutes. The behavioral score was assigned by use of a 5-point system (1 = no observable response, 2 = slight or barely perceptible response, 3 = mild response, 4 = moderate response, and 5 = dramatic response). Response scores did not differ significantly (P ≥ 0.05) among treatments.

Citation: Journal of the American Veterinary Medical Association 238, 12; 10.2460/javma.238.12.1629

• Download Figure
• Download figure as PowerPoint slide

Response (behavioral score) for carpal arthrocentesis in horses sedated by administration of xylazine hydrochloride (0.4 mg/kg [0.18 mg/lb]; treatment X [n = 10]; white bar with black dots), xylazine (0.4 mg/kg) plus butorphanol tartrate (0.01 mg/kg [0.0045 mg/lb]; treatment XB [11]; black bar with white dots), xylazine (0.4 mg/kg) plus butorphanol (0.01 mg/kg) plus a lower dose of ketamine (0.1 mg/kg [0.045 mg/lb]; treatment XBK1 [10]; vertical-striped bar), xylazine (0.4 mg/kg) plus butorphanol (0.01 mg/kg) plus a higher dose of ketamine (0.2 mg/kg [0.09 mg/lb]; treatment XBK2 [11]; black bar), xylazine (0.4 mg/kg) plus the lower dose of ketamine hydrochloride (0.1 mg/kg; treatment XK1 [11]; diagonal-striped bar), or xylazine plus the higher dose of ketamine (0.2 mg/kg; treatment XK2 [11]; horizontal-striped bar). All drugs were administered IV. For all treatments, xylazine or xylazine-butorphanol was administered first (time 0), followed 5 minutes later by administration of one of the doses of ketamine or saline (0.9% NaCl) solution; arthrocentesis was performed at time = 7 minutes. The behavioral score was assigned by use of a 5-point system (1 = no observable response, 2 = slight or barely perceptible response, 3 = mild response, 4 = moderate response, and 5 = dramatic response). Response scores did not differ significantly (P ≥ 0.05) among treatments.

Citation: Journal of the American Veterinary Medical Association 238, 12; 10.2460/javma.238.12.1629

• Download Figure
• Download figure as PowerPoint slide

Discussion

The need to sedate horses for routine husbandry procedures (eg, tooth floating), diagnostic procedures (eg, endoscopy of the larynx and trachea), or interventional procedures (eg, laparoscopic surgery) is common in equine veterinary practice. Given the risks of general anesthesia and the associated recovery period, there is value in being able to safely provide chemical restraint and analgesia for these procedures in standing horses. Reportedly, 0.2% to 1.6% of recoveries from anesthesia result in serious or life-ending injuries, such as fractures.^3–5 In addition to being safer for patients, the standing position also facilitates performance of certain surgical procedures (eg, urethral extension or laparoscopic ovariectomy), is physiologically better for horses than is drug-induced recumbency, and is more time- and cost-effective.

With available α₂-receptor agents (such as xylazine, detomidine, and romifidine) and opioids (such as butorphanol and morphine), satisfactory sedation for procedures in standing animals can be achieved in most horses. However, most equine practitioners have experienced that some horses arouse from an apparently profound sedation with little warning and may cause harm to themselves or to those around them. It was suggested in 1 report⁶ that the sedation was inadequate in 11% of horses receiving detomidine, a drug that many would consider the most reliable of the α₂-receptor agents. Other authors report⁷ improvement in the degree of sedation when opioids are combined with α₂-receptor agonists, but there may also be occasional agitation or twitching. Some procedures have been aborted because a standing horse could not be appropriately sedated. Ataxia and respiratory compromise are also features of sedation with high doses of α₂-receptor agonists, and excitement may be observed with the IV administration of opioids.^7–10 A reduction in motility of the gastrointestinal tract may be detected following administration of α₂-receptor agonists and opioids.^11,12

Similar concerns are reported by bovine practitioners, who frequently have to deal with unruly cattle in circumstances that are less than ideal. As a means to address these concerns, a chemical restraint technique (commonly referred to as ketamine stun or ket stun) was developed.¹ It appears that the addition of ketamine (0.05 mg/kg, IV) significantly improves the quality of chemical restraint provided by xylazine and butorphanol in standing cattle, without adverse effects.¹

The concept of profound sedation or chemical restraint in standing horses is not new. Low doses of chloral hydrate, alone or in combination with xylazine, were once commonly used to provide heavy sedation in standing patients.¹³ The tendency for chloral hydrate to cause tissue necrosis if accidentally injected extravascularly, along with the advent of various α₂-receptor agonists, eventually diminished its use. More recently, ketamine has received attention as an analgesic in standing horses.^14–16 Behavioral effects were not detected after approximately 6 hours for an administration rate of 1.5 mg/kg/h (0.68 mg/lb/h or 750 mg/h for a 500-kg [1,100-lb] horse), but heart rate and respiratory rate were elevated in horses in that study.¹⁴ This is in contrast to results of another study¹⁶ in which authors reported no excitatory effects or changes in physiologic variables at slightly lower dosages ranging from 0.4 to 0.8 mg/kg/h (0.18 to 0.36 mg/lb/h or 200 to 400 mg/h for a 500-kg horse). Although an analgesic effect in response to hoof testers was not detected in the latter study,¹⁶ dose-dependent analgesia was reported by authors who used a nociceptive withdrawal reflex technique in ponies.¹⁵

Analysis of results of the study reported here revealed that the addition of the higher dose of ketamine to xylazine-butorphanol facilitated insertion of a dental float and enhanced tolerance to algometer pressure but did not decrease the response to noise, a needle prick, or carpal arthrocentesis. Interestingly, XK1 and XK2 treatments were associated with the least improvement in tolerance to insertion of the dental float (Table 3), which suggested that butorphanol was an important modifier of this behavior. Similarly, XK2 treatment was associated with the most pronounced response to a needle prick in the flank area, whereas XB treatment was associated with the lowest response to a needle prick, which again suggested that butorphanol played a major role in the analgesia to a needle prick and, perhaps, that ketamine actually negated any potential improvement in analgesia conveyed by xylazine (Table 4). When a benefit of ketamine was detected, the benefit appeared to be short-lived, with significant differences detected only at the 10-minute evaluation, which was 5 minutes after the administration of ketamine. The lower dose of ketamine did not appear to convey any significant benefit, compared with administration of xylazine or xylazine-butorphanol, for any of the procedures evaluated.

Although ataxia was not prospectively evaluated and scored, it should be mentioned that a number of horses in each group were described as ataxic or wobbly after drug administration. The number of horses described as ataxic or wobbly was 3 of 10 for treatment X, 1 of 11 for treatment XB, 1 of 10 for treatment XBK1, 5 of 11 for treatment XBK2, 1 of 11 for treatment XK1, and 3 of 11 for treatment XK2. These numbers suggest that the higher dose of ketamine was associated with a greater tendency to cause noticeable ataxia. Arthrocenteses were delayed for up to 10 minutes in 4 horses for the XBK2 treatment as well as in 1 horse for the XBK1 treatment because those particular horses were considered to be too ataxic or were reluctant to remain standing on 3 legs for the arthrocentesis procedures. However, even 2 horses that did not receive ketamine (1 for the X treatment and 1 for the XB treatment) were reluctant to stand on 3 legs. Increasing the depth of sedation is not necessarily linked to better performance of a procedure, most notably because of difficulties in handling heavily sedated horses for certain procedures.^h All arthrocenteses were eventually successfully performed once ataxia had subsided, usually within 4 minutes (but always within 10 minutes) after administration of a dose of ketamine or saline solution.

The dosages of xylazine and butorphanol (0.4 and 0.01 mg/kg, IV, respectively) in the present study were selected to mimic clinical circumstances. Although the suggested dosages^b,c may be considerably higher (up to 1.1 mg/kg [0.5 mg/lb], IV, for xylazine and up to 0.1 mg/kg, IV, for butorphanol) for procedures in standing horses, many equine veterinarians prefer to start with lower dosages prior to administering any additional drugs. The ketamine dosages were extrapolated from the dosages used in other studies of analgesic testing in standing unsedated horses in which no physiologic or behavioral complications were reported^14,16 as well as from a report¹ of ketamine use in standing cattle. It is possible that results may have differed if higher dosages of xylazine, butorphanol, or ketamine had been used.

Butorphanol is typically described as primarily a visceral analgesic.¹⁷ Ketamine is more often associated with superficial or somatic analgesia.¹⁸ However, the analgesic effects of ketamine have not been fully characterized. Although ketamine significantly reduces nociceptive behavior in mice subjected to tonic or persistent pain in the form of a formalin test, it does have pronociceptive properties in mice subjected to low-intensity heat stimulation, which is considered a test of phasic or acute, transient pain.¹⁹ Analysis of the data in that report¹⁹ suggests that ketamine has no analgesic effect on phasic (acute or transient) pain in somatic tissues, but it may alleviate tonic (persistent) pain after inflammation. Presumably, the brief noxious insults applied to the horses of the study reported here did not result in inflammation or tonic pain, which may be a reason that ketamine had minimal benefit for this method. Interestingly, the addition of ketamine to xylazine without butorphanol (treatments XK1 and XK2) actually appeared to worsen the horses' tolerance to insertion of a dental float (Table 3) and to a needle prick on the flank (Table 4), compared with results for treatments X or XB, although the difference was significant only for a needle prick on the flank for treatment XK2. In healthy cats subjected to a thermal stimulus, ketamine (2 mg/kg, IV) was associated with apparent hyperalgesia.ⁱ Although ketamine may not provide analgesia for acute, transient pain, it is not clear why it might cause apparent hyperalgesia in some situations.

We concluded that on the basis of results of the study reported here, the addition of a subanesthetic dose of ketamine to xylazine plus butorphanol facilitated insertion of a dental float and enhanced tolerance to algometer pressure but did not decrease response to noise or to carpal arthrocentesis. Ketamine administered after xylazine without butorphanol did not convey any advantage and appeared to worsen response to a needle prick on the flank. Small doses of ketamine added to xylazine and butorphanol may be useful for augmenting sedation and restraint for certain procedures in horses, but further evaluation is needed to determine exactly which clinical procedures might be facilitated by the addition of ketamine to sedation regimens.