In horses, the perianesthetic mortality rate often exceeds 1/100 cases,1,2 which is far greater than mortality rates reported for dogs and cats (< 2/1,000)3,4 and humans (16/10,000).5 A third of these horses die or are euthanized because of complications associated with recovery from anesthesia (eg, fractures, myositis, or nerve damage).1,2 As a consequence, veterinary anesthesiologists are continually looking for new anesthetic and recovery techniques for horses that will improve anesthetic recovery and, thereby, decrease the perianesthetic mortality rate.
Currently, most horses anesthetized in the United States are first sedated with an α2-adrenoceptor agonist, with general anesthesia subsequently induced with a combination of a benzodiazepine (eg, midazolam) and a dissociative anesthetic (eg, ketamine).6 Anesthesia is then maintained with additional injectable drugs (eg, ketaminexylazine-guaifenesin) or with an inhalant anesthetic (eg, isoflurane) following endotracheal intubation. Horses recovering from inhalant anesthetics are particularly uncoordinated, and sedation with an α2-adrenoceptor agonist has been shown to improve recovery quality,7 with romifidine being superior to xylazine.8 Whereas poor anesthetic recovery can be attributed to the drugs administered (or not administered) immediately prior to recovery, it is likely that all drugs administered as well as numerous physiologic factors contribute to the quality of recovery.
In 2012, a nationwide shortage of benzodiazepines in the United States meant that many practitioners were unable to use midazolam as part of the routine anesthetic protocol for horses. Although horses can be anesthetized with a drug combination consisting of only an α2-adrenoceptor agonist and ketamine,9 many equine anesthesiologists prefer to also include a muscle relaxant such as a benzodiazepine to balance the muscle rigidity caused by ketamine. Propofol, a γ-aminobutyric acid–enhancing anesthetic, was considered as a replacement. Although propofol is a commonly used anesthetic agent in dogs and cats, it is rarely used alone in horses because of the unpredictable quality of induction10 and the high cost when administered as a sole induction agent. However, propofol has been shown to result in acceptable anesthetic induction quality in healthy horses used for research when combined with ketamine,11 and a propofol-ketamine combination was routinely used at the North Carolina State University Hospital for Animals without complications during the 2012 benzodiazepine shortage. Furthermore, in a retrospective analysis,12 client-owned horses anesthetized with various protocols including propofol and ketamine reportedly recovered rapidly and smoothly.
It is possible that the rapid and smooth recoveries following anesthesia with propofol and ketamine12 might be associated with the difference in duration of action between midazolam and propofol in horses. Midazolam has a surprisingly long half-life in horses, with a reported half-life > 400 minutes at doses typically administered for anesthesia.13 The half-life of commercially available propofol in horses is unknown; however, propofol in a micellar microemulsion has a half-life of 45 minutes.14 It is therefore reasonable to theorize that horses anesthetized with propofol or midazolam in which anesthesia was maintained for 1 hour would have relatively lower plasma propofol concentrations than plasma midazolam concentrations at the time of recovery. Residual midazolam could result in muscle weakness, which could contribute to a slower and poorer recovery.
The objective of the study reported here was to evaluate quality of recovery from general anesthesia in horses after induction with propofol and ketamine versus midazolam and ketamine. We hypothesized that horses would have better recovery scores and a more rapid recovery when general anesthesia was induced with propofol and ketamine than when it was induced with midazolam and ketamine. In addition, we hypothesized that after 1 hour of general anesthesia, the drug concentration at recovery, expressed as a percentage of the plasma concentration immediately following anesthetic induction, would be less with propofol than with midazolam.
Materials and Methods
Six adult horses (mean ± SD body weight, 529.2 ± 49.7 kg [1,164.2 ± 109.3 lb]; 3 mares and 3 geldings) were used in the study. Horses were deemed to be in good health on the basis of history and results of a complete physical examination. The North Carolina State University International Animal Care and Use Committee approved the study protocol.
The study was designed as a randomized crossover study with a washout period of at least 1 week between treatments. Horses were randomly assigned to a treatment sequence by means of a coin toss, which resulted in 5 of the 6 horses receiving midazolam first. The same anesthetist (MAJ) performed anesthesia in all experiments.
The evening prior to each anesthetic event, horses were sedated with xylazine (0.25 mg/kg [0.11 mg/lb], IV) for IV jugular catheter placement. The catheter site was clipped and aseptically prepared, and a 14-gauge catheter was placed and secured with suture and elastic tape. In all horses, the left jugular vein was catheterized for the first anesthetic event and the right jugular vein was catheterized for the second.
Food but not water was withheld for a minimum of 12 hours prior to each anesthetic episode. On each procedure day, jugular catheter patency was verified, and horses were premedicated in their stall with xylazine (0.25 mg/kg, IV). Once noticeable sedation was achieved, the horses' hooves were cleaned, and their mouths were rinsed with water. Horses were then walked from their stall to a padded induction stall in a separate building, where an additional dose of xylazine (0.75 mg/kg [0.34 mg/lb], IV; total dose, 1.0 mg/kg [0.45 mg/lb], IV) was administered.
Once horses were considered maximally sedated (head below the level of the carpus and no response to aural stimulation), general anesthesia was induced with administration of either midazolam (0.1 mg/kg [0.045 mg/lb], IV) or propofol (0.5 mg/kg [0.23 mg/lb], IV) on the basis of the randomization scheme. Horses were then immediately given ketamine (3.0 mg/kg [1.36 mg/lb], IV). Doses of propofol, midazolam, and ketamine reflected the preferences of the anesthesiologists at our institution. Each anesthetic induction was videotaped with a digital camera hung above the induction stall, and induction time was recorded with a stopwatch. Induction time was calculated as the time from administration of induction drugs until the horse became laterally recumbent.
Once recumbent, horses were orotracheally intubated with a 26-mm (internal diameter) cuffed endotracheal tube. Hobbles were placed around the pastern regions of all 4 limbs, and horses were hoisted onto a foam mattress in the induction stall. For both anesthetic episodes, all horses were placed in right lateral recumbency because of stall size restrictions and for consistency. Once horses were positioned on the mattress, the right forelimb was pulled forward to avoid myopathy, and lubricant was placed in both eyes.
Immediately after horses were positioned on the mattress, a 12-mL venous blood sample was collected from the jugular vein catheter (following removal of a 12-mL wasted blood sample). Samples were immediately placed on ice until centrifuged; plasma was separated and stored frozen (–85°C) until analyzed.
General anesthesia was maintained for 60 minutes with isoflurane delivered in oxygen with a large animal anesthesia machine and rebreathing circuit. An initial oxygen flow rate of 10 L/min was used, but flow rate was decreased to 6 L/min after 10 minutes. Mechanical ventilation was initiated at 6 breaths/min with a tidal volume of 10 to 15 mL/kg (4.5 to 6.8 mL/lb) to maintain an end-tidal carbon dioxide concentration between 35 and 45 mm Hg. Anesthetic depth was maintained at stage III, plane 2 (palpebral reflex present and no nystagmus) in all horses throughout the anesthetic period. Stage III, plane 2 is the stage of anesthesia deemed appropriate to facilitate surgical procedures. Lactated Ringer solution was administered at a rate of approximately 10 mL/kg/h throughout each anesthetic episode.
Direct blood pressure, heart rate (by means of a lead II ECG), arterial oxygen saturation (by means of pulse oximetry), respiratory rate, vaporizer setting, and end-tidal carbon dioxide concentration (by means of capnography) were monitored with a multiparameter monitor throughout each anesthetic episode, and values were recorded every 5 minutes. For measurement of direct blood pressure, a 20-gauge catheter was placed in the left lateral metatarsal artery and connected to an electronic pressure transducer. The transducer was zeroed at the level of the right atrium and was recalibrated approximately 30 minutes after initial calibration. If mean arterial blood pressure was < 75 mm Hg, inotropic support was provided by means of dobutamine administered as a constant rate infusion (0.5 to 2 μg/kg/min [0.23 to 0.9 μg/lb/min]) to maintain mean arterial blood pressure between 75 and 85 mm Hg. Arterial blood gas analysis was performed immediately following arterial catheter placement and again 5 minutes prior to recovery with a blood sample obtained from the catheter in the lateral metatarsal artery. Fifteen minutes prior to the end of anesthesia, phenylephrine (50 mg) was administered into each nostril with an atomizer.
After 60 minutes, isoflurane administration was discontinued, and horses were disconnected from the anesthesia machine. Horses were hoisted off the mattress and placed onto the induction stall floor in right lateral recumbency for recovery. At this time, a second 12-mL venous blood sample was collected and processed as described. Assisted ventilation was maintained with oxygen supplementation via a Hudson demand valve at a rate of 1 to 2 breaths/min until horses resumed spontaneous ventilation. At that time, the endotracheal tube was removed, and romifidine (0.02 mg/kg [0.009 mg/lb], IV) was administered.
Horses were allowed to recover unassisted, without halter or lead ropes. Recoveries were observed by the anesthetist outside of the recovery stall and videotaped with a digital camera hung above the recovery stall. The following variables were measured from outside of the recovery stall: time to first movement, time to sternal recumbency, time to standing, and number of attempts to stand.
Recovery videos were independently viewed at a later date by 3 graders (2 board-certified veterinary anesthesiologists [KMB and KMM] and a board-certified large animal surgeon [TP]) who were blinded to whether horses had received propofol or midazolam, and recovery quality was scored with a rubric validated for assessing recovery of horses from general anesthesia15 and a VAS. The recovery rubric assessed 11 behavioral and postural variables with a minimum possible combined score of 11 and maximum possible combined score of 100 and with lower scores indicating a better recovery (Supplementary Appendix A1, available at avmajournals.avma.org/doi/suppl/10.2460/javma.253.1.101). For the VAS, the left margin was labeled “worst recovery imaginable” and the right margin was labeled “best recovery imaginable”; potential scores ranged from 0 to 1, with higher scores indicating a better recovery.
Plasma concentrations of midazolam were determined by means of ultra-performance liquid chromatography with mass spectrometry. Calibration curves were prepared by fortifying blank equine plasma with stock solutions of midazolam hydrochloride dissolved in 100% methanol. Samples and standards were then prepared by adding 500 μL of plasma to 500 μL of 2% ammonium hydroxide in water in a glass tube and vortexing for 15 seconds. The sample mixture was added to supported liquid extraction cartridges, and a light vacuum was applied to initiate absorption. Two aliquots of 5 mL of ethyl acetate were added to the cartridges, which were allowed to sit for 5 minutes and then slowly eluted under light vacuum. The resulting eluate was placed in an evaporator and dried under a 20-psi stream of nitrogen for 40 minutes at 45°C. Samples were reconstituted in 300 μL of a 50:50 (vol/vol) water:acetonitrile mixture. All samples and standards were filtered with 0.22-μm injection vials. Volumes of 5 μL for samples and standards were used for liquid chromatography and mass spectrometry. A gradient was used and the initial mobile phase was 10mM ammonium acetate buffer (pH, 4.0) and 0.1% formic acid in acetonitrile (90:10 vol/vol) for the first 2 minutes. The mobile phase was then switched to a 10:90 (vol/vol) mixture of formic acid in acetonitrile for 1 minute. For the last 2 minutes of the run, the mobile phase was a 90:10 (vol/vol) mixture of formic acid in acetonitrile. Flow rate was maintained at 0.4 mL/min. The single quadrupole mass spectrometer was run in electrospray ionization mode. The selected ion recording used was 291 nm. Column temperature was maintained at 35°C (95°F), and sample temperature was maintained at 4°C (39.2°F). Separation was achieved with an ethylene-bridged hybrid column (1.7 μm, 2.1 ± 50 mm) and guard column. Standard curves were linear over a concentration range of 25 to 500 ng/mL, with an R2 ≥ 0.99. The lower limit of quantification was 25 ng/mL.
Plasma concentrations of propofol were determined by means of high performance liquid chromatography with fluorescence detection. Calibration curves were prepared by fortifying blank equine plasma with stock solutions of propofol dissolved in 100% methanol. Samples and standards were then prepared by adding 500 μL of plasma to 500 μL of distilled water in a glass tube and vortexing for 15 seconds. Samples were then subjected to solid phase extraction, following the same protocol described for midazolam. Samples were reconstituted in 250 μL of a 30:70 (vol/vol) water:acetonitrile mixture and transferred into injection vials. Volumes of 25 μL for samples and standards were used for liquid chromatography. The mobile phase was a 30:70 water:acetonitrile mixture with a flow rate of 1 mL/min for 5 minutes. The fluorescence detector was run with wavelengths of 276 nm and 310 nm for excitation and emission, respectively. Column temperature was maintained at 40°C (104°F). Separation was achieved with a phenyl column (3.5 μm, 4.6 × 150 mm) and guard column. Standard curves were linear over a concentration range of 100 to 10,000 ng/mL, with an R2 ≥ 0.99. The lower limit of quantification was 100 ng/mL.
Statistical analysis
Because previous reports suggested that horses recover from anesthesia more quickly following induction with propofol, an a priori power analysis was performed on the basis of an expected shorter recovery time following induction with propofol and ketamine. Power analysis was performed with a computerized statistical programa and the following estimates from an experienced anesthesiologist (LPP): routine recovery time of 45 ± 10 minutes, a decrease in recovery time of 15 minutes would be considered clinically relevant, α = 0.05, and power = 80%. The power analysis resulted in a required sample size of 6.
Data analysis was performed with multiple computer software programs.b,c On the basis of data distribution and small sample sizes, results of blood gas analyses, recovery variables, and drug concentration data were analyzed with nonparametric tests. Differences between groups and within a group over time were tested with the Wilcoxon matched-pairs signed rank test.
Scores for the recovery rubric and VAS were analyzed with a mixed-effects ANOVA for crossover designs that included a random reviewer effect to model the variability across graders. Additionally, the intraclass correlation coefficient was calculated to assess agreement among graders. Heart rate, mean arterial pressure, and mean isoflurane vaporizer settings were analyzed with a mixed-effects ANOVA for crossover designs that included a time effect to account for the repeated measurements over time. For all analyses, values of P ≤ 0.05 were considered significant.
Results
Number of attempts to stand was significantly lower when horses received propofol (median, 2; range, 1 to 3) than when they received midazolam (median, 7.5; range, 3 to 16). For all 3 graders, the recovery rubric score was significantly (P < 0.001) lower (ie, better recovery quality) when horses received propofol than when they received midazolam (Figure 1). Similarly, the VAS score was significantly higher (ie, better recovery quality) for all 3 graders when horses received propofol than when they received midazolam. The intraclass correlation coefficient for agreement among graders was 0.82 for recovery rubric scores and 0.84 for VAS scores. No significant differences between treatments (propofol vs midazolam) were identified for induction time, time to first movement, time to sternal recumbency, or time to standing.
Median recovery rubric (RR) and VAS scores for recovery quality assigned by 3 blinded reviewers to 6 healthy horses premedicated with xylazine (1.0 mg/kg [0.45 mg/lb], IV) in which anesthesia was induced with midazolam (0.1 mg/kg [0.045 mg/lb], IV) or propofol (0.5 mg/kg [0.23 mg/lb], IV) followed by ketamine (3.0 mg/kg [1.36 mg/lb], IV) and maintained with isoflurane for 60 minutes. Horses received each of the 2 treatments 1 week apart in a randomized crossover design. For the recovery rubric, a higher number indicates a poorer recovery. For the VAS, a lower number indicates a poorer recovery. *Value was significantly (P < 0.05) different from the value for the midazolam treatment.
Citation: Journal of the American Veterinary Medical Association 253, 1; 10.2460/javma.253.1.101
No significant differences in physiologic variable were identified between treatments or over time within treatments. All horses required dobutamine to maintain a mean arterial pressure between 75 and 85 mm Hg, and all horses had normal sinus rhythm throughout the anesthetic period. Mean ± SD isoflurane vaporizer setting when horses received propofol (2.15 ± 0.04%) was not significantly (P = 0.32) different from mean setting when horses received midazolam (2.30 ± 0.06%). Plasma drug concentrations and results of arterial blood gas analyses were summarized (Tables 1 and 2).
Median (range) plasma concentrations of midazolam and propofol in 6 healthy horses premedicated with xylazine (1.0 mg/kg [0.45 mg/lb], IV) in which anesthesia was induced with midazolam (0.1 mg/kg [0.045 mg/lb], IV) or propofol (0.5 mg/kg [0.23 mg/lb], IV) followed by ketamine (3.0 mg/kg [1.36 mg/lb], IV) and maintained with isoflurane for 60 minutes.
| Variable | Midazolam | Propofol |
|---|---|---|
| Plasma drug concentration (μg/mL) | ||
| Induction | 0.25 (0.18–0.54) | 0.78 (0.61–1.57) |
| Recovery | 0.08 (0.06–0.09) | 0.13 (0.11–0.14) |
| Recovery concentration as percentage of induction concentration | 28.67 (14.25–51.30)* | 16.76 (9.17–20.6)* |
Horses received each of the 2 treatments 1 week apart in a randomized crossover design.
Values were significantly (P < 0.05) different between treatments.
Median (range) arterial blood gas variables for the horses in Table 1.
| Midazolam | Propofol | |||
|---|---|---|---|---|
| Variable | Induction | Recovery | Induction | Recovery |
| pH | 7.457 (7.420–7.498) | 7.514 (7.415–7.528) | 7.459 (7.416–7.504) | 7.490 (7.440–7.526) |
| Paco2 (mm Hg) | 40.85 (35.10–44.60) | 40.10 (34.20–43.00) | 41.55 (37.10–45.80) | 40.00 (34.70–41.60) |
| Pao2 (mm Hg) | 377.5 (289.0–498.0) | 461.0 (389.0–584.0)* | 460.0 (312.0–534.0) | 514.5 (462.0–583.0) |
| Base excess | 5.0 (0.0–9.0) | 6.0 (3.0–11.0) | 6.0 (2.0–7.0) | 7.0 (2.0–9.0) |
| HCO3−− (mmol/L) | 29.00 (24.80–32.40) | 29.65 (27.60–33.90) | 29.55 (25.80–30.90) | 30.05 (26.60–32.00) |
| Total CO2 (mmol/L) | 30.00 (27.00–34.00) | 31.00 (29.00–35.00) | 31.00 (27.00–32.00) | 31.50 (28.00–33.00) |
| Spo2 (%) | 100.0 (100.0–100.0) | 100.0 (100.0–100.0) | 100.0 (100.0–100.0) | 100.0 (100.0–100.0) |
| Lactate (mmol/L) | 0.865 (0.500–1.130) | 1.150 (0.870–1.450)* | 0.960 (0.800–1.550) | 1.170 (1.000–1.460) |
Significantly (P < 0.05) different from value recorded at induction for the same treatment.
Spo2 = Oxygen saturation determined by pulse oximetry.
Discussion
Results of the present study supported our hypothesis that recovery quality was better when general anesthesia was induced with propofol and ketamine than when it was induced with midazolam and ketamine.
Anesthetic recovery is a complex, multifactorial process, and there is no uniform consensus regarding what constitutes a “good” recovery for horses. Among other criteria, recovery time, number of attempts to stand, and degree of calmness are commonly part of the assessment. However, no universally accepted scoring system for quality of anesthetic recovery in horses is available. For this reason, we evaluated a number of subjective and objective variables in the present study. Scoring systems based on simple numerical scales are concise and easy to use; however, they do not allow for the evaluation of individual components of the recovery period.16,17 In contrast, multifactorial scoring systems divide the recovery score into various categories that are graded separately, with scores for each category summed to obtain a total overall recovery score.18 In the present study, we chose a 100-point recovery rubric and a VAS to assess anesthetic recovery, in an attempt to include both subjective and objective variables. The 100-point scale was based on a previously described system15 and allowed for the assessment of 11 categories contributing to overall recovery quality. Scores for this scale range from 11 to 100, with 11 being the best possible recovery and 100 being the worst. The VAS is a commonly used subjective scoring method that has been documented for evaluation of anesthetic recovery in horses and is considered to be reliable.19,20 Whereas subjective scoring methods are inherently somewhat unreliable because of interrater variability, variation between graders in the present study was minimal, and recovery scores were consistent. Although there are no universally agreed upon cutoff values for interpreting intraclass correlation coefficients, a previous publication21 cited values between 0.75 and 1.0 to be excellent. Because coefficients for the recovery rubric and VAS scores in the present study were 0.82 and 0.84, respectively, we concluded that agreement among graders was excellent.
In the present study, both the recovery rubric and the VAS indicated that horses recovered better following induction with propofol, compared with recovery quality following induction with midazolam. One possible explanation for the better overall recovery when horses received propofol is the comparatively prolonged elimination half-life of midazolam. A recent study13 investigating the pharmacokinetics of midazolam in horses found a terminal elimination half-life of > 400 minutes after IV administration. Whereas the termination of clinical effects of midazolam is thought to be due to redistribution, the long half-life suggests the possibility that some of these effects would persist into the recovery period. For instance, horses administered midazolam at a dose of 0.1 mg/kg, IV, were noted to be ataxic for up to 90 minutes.13 Because the same dose was used in the present study and horses were anesthetized for 60 minutes, we suggest that it was likely that residual midazolam effects contributed to the poorer recovery. Similar findings were reported in a previous study22 that compared various induction drug combinations in horses anesthetized with isoflurane. In that study, pronounced ataxia was noted in horses that received ketamine and midazolam, especially after periods of anesthesia of < 75 minutes.22 Whereas the elimination half-life of the propofol used in the present study is unknown, the elimination half-life of propofol in a unique vehicle is approximately 45 minutes.14 Furthermore, a study23 that investigated propofol and ketamine infusions in horses indicated that during recovery, horses achieved sternal recumbency when propofol plasma concentrations were between 0.8 and 1.1 μg/mL.23 In the present study, propofol plasma concentrations were between 0.1 and 0.14 μg/mL prior to extubation. As such, we suggest that it is unlikely that any clinical effects of propofol were present during the recovery period. Considering that a greater percentage of midazolam was still present during the recovery period in study horses, these findings would support our assertion that the prolonged effects of midazolam likely contributed to the observed differences in recovery scores.
When scores for the individual categories of the recovery rubric were examined, strength, number of attempts to stand, balance and coordination, and movement to stand were the categories that most contributed to the difference in the overall recovery quality between midazolam and propofol for horses of the present study. Similarly, these categories describe the expected effects of midazolam administration in horses (eg, muscle weakness and incoordination) and likely can be attributed to the longer elimination half-life of midazolam, compared with that of propofol. Also of interest was the difference between the 2 treatments for the number of attempts to stand. For midazolam, the number of attempts ranged from 3 to 16, whereas the range for propofol was 1 to 3. These results suggested that with midazolam, the recovery characteristics may vary from patient to patient when horses are allowed to recover from anesthesia unassisted. Given the already unpredictable nature of anesthetic recovery in horses, the inconsistent effects for midazolam are undesirable.
Unexpectedly, when horses received propofol and ketamine in the present study, they did not have a significantly shorter recovery time, compared with recovery time when horses received midazolam and ketamine. One possible explanation for this finding was that horses were sedated with romifidine prior to recovery and likely remained recumbent until the effects of the romifidine dissipated. The sedative effects of romifidine are dose-dependent in both extent and duration of action. A previous study24 indicated that romifidine administered to horses at doses of 0.04 to 0.08 mg/kg (0.018 to 0.036 mg/lb), IV, resulted in sedation for 40 to 80 minutes. In the present study, the dose of romifidine was only 0.02 mg/kg, so it was reasonable to expect a shorter duration of sedation.
Although evaluation of induction quality was not a specific aim of the present study, no adverse events were noted during the induction period and, subjectively, induction quality was similar between the propofol and midazolam treatment. This finding was consistent with previous reports11 of induction quality following administration of ketamine and propofol. Additionally, no difference in induction times were noted between treatments in the present study.
Although plasma lactate concentration did increase significantly following induction with midazolam in the present study, values were within reference limits, and plasma lactate concentration was not significantly different between treatments. Similarly, the arterial partial pressure of oxygen increased significantly following induction with midazolam, but again no significant difference was noted between treatments, and values were within the range expected for mechanically ventilated, anesthetized horses breathing isoflurane in oxygen.
Heart rate did not differ between the midazolam and propofol treatments in the present study. In horses, benzodiazepines rarely produce changes in heart rate,25 and in a previous study,13 no changes in heart rate were seen over a 60-minute period in horses that received midazolam IV. Furthermore, whereas 1 study reported26 transient tachycardia after induction of anesthesia with propofol, values for heart rate returned to baseline within 5 minutes. Because heart rate was recorded during the induction period in the present study, it is possible that changes went undetected. Interpretation of blood pressure values was limited in this study as all horses required dobutamine support during both anesthetic periods. The total doses of dobutamine administered to each horse was not quantified in this study.
A potential limitation of the present study was that the attending anesthesiologist (MAJ) was aware of which treatment horses received; however, it was unlikely that this knowledge affected objective measurements of recovery (ie, recovery times and number of attempts to stand). It is possible that subjective criteria, such as depth of anesthesia, may have been influenced. However, the anesthetist adjusted anesthetic concentrations as needed to be as consistent as possible. On the other hand, graders for subjective assessments of recovery quality (recovery rubric and VAS) were blinded to the treatment. A second limitation of the present study was that no invasive procedures were performed during the anesthetic period. Thus, this study may not be representative of horses recovering from anesthesia following a painful surgical procedure. Horses that experience postoperative pain despite appropriate analgesia and sedation, may be more inclined to attempt to stand before the effects of the anesthetic have dissipated and thus have a poorer recovery.27 Thus, we cannot speculate as to how recovery quality would differ between midazolam and propofol in situations with a painful stimulus. Nonetheless, horses are routinely anesthetized for diagnostic procedures (eg, MRI and CT), and in these circumstances, we suggest that propofol may provide benefits during the recovery period compared with midazolam. A third limitation was that with our randomization scheme, 5 of the 6 horses were administered midazolam first. Whereas it is possible that horses that are repeatedly anesthetized do recover better after multiple anesthetic events,28,29 the 1 horse that received propofol first had better recovery scores during the first anesthetic event. Although the randomization order was followed as generated, it is possible this affected recovery quality. A final limitation was the sampling technique used. For all horses in the present study, the same catheter was used for drug delivery and blood sample collection. We acknowledge that this may have affected the plasma concentrations obtained, although a 12-mL waste sample was removed from each catheter prior to sample collection. Furthermore, no plasma samples were analyzed prior to drug administration. The results from the present study suggested that both the induction drug duration of action and the total procedure time should be factored into the decision process for choosing drugs for equine patients. Additional study, including that involving patients, is indicated.
Acknowledgments
Presented in abstract form at the American College of Veterinary Anesthesia and Analgesia Scientific Meeting, Washington DC, September 2015.
ABBREVIATIONS
| VAS | Visual analog scale |
Footnotes
Statistical Solutions LLC, Cottage Grove, Wis.
GraphPad Prism, version 5.00 for Windows, GraphPad Software Inc, La Jolla, Calif.
SAS 9.3 for Windows, SAS Institute Inc, Cary, NC.
References
1. Johnston GM, Eastment JK, Wood J, et al. The Confidential Enquiry into Perioperative Equine Fatalities (CEPEF): mortality results of phases 1 and 2. Vet Anaesth Analg 2002;29:159–170.
2. Dugdale AH, Obhrai J, Cripps PJ. Twenty years later: a singlezcentre, repeat retrospective analysis of equine perioperative mortality and investigation of recovery quality. Vet Anaesth Analg 2016;43:171–178.
3. Brodbelt DC, Pfeiffer DU, Young LE, et al. Risk factors for anaesthetic-related death in cats: results from the Confidential Enquiry into Perioperative Small Animal Fatalities (CEPSAF). Br J Anaesth 2007;99:617–623.
4. Dyson DH, Maxie MG, Schnurr D. Morbidity and mortality associated with anesthetic management in small animal veterinary practice in Ontario. J Am Anim Hosp Assoc 1998;34:325–335.
5. Pignaton W, Braz JR, Kusano PS, et al. Perioperative and anesthesia-related mortality: an 8-year observational survey from a tertiary teaching hospital. Medicine 2016;95:e2208.
6. Bettschart-Wolfensberger R. Horses. In: Grimm KA, Lamont LA, Tranquilli WJ, et al, eds. Veterinary anesthesia and analgesia: the fifth edition of Lumb and Jones. Ames, Iowa: Wiley-Blackwell, 2015;897–885.
7. Santos M, Fuente M, Garcia-Iturralde P, et al. Effects of alpha-2 adrenoceptor agonists during recovery from isoflurane anaesthesia in horses. Equine Vet J 2003;35:170–175.
8. Woodhouse KJ, Brosnan RJ, Nguyen KQ, et al. Effects of postanesthetic sedation with romifidine or xylazine on quality of recovery from isoflurane anesthesia in horses. J Am Vet Med Assoc 2013;242:533–539.
9. Muir WW, Skarda RT, Milne DW. Evaluation of xylazine and ketamine hydrochloride for anesthesia in horses. Am J Vet Res 1977;38:195–201
10. Mama KR, Steffey EP, Pascoe PJ. Evaluation of propofol for general anesthesia in premedicated horses. Am J Vet Res 1996;57:512–516.
11. Wagner AE, Mama KR, Steffey EP, et al. Behavioral responses following eight anesthetic induction protocols in horses. Vet Anaesth Analg 2002;29:207–211.
12. Posner LP, Kasten JI, Kata C. Propofol with ketamine following sedation with xylazine for routine induction of general anaesthesia in horses. Vet Rec 2013;173:550.
13. Hubbell JA, Kelly EM, Aarnes TK, et al. Pharmacokinetics of midazolam after intravenous administration to horses. Equine Vet J 2013;45:721–725.
14. Boscan P, Rezende ML, Grimsrud K, et al. Pharmacokinetic profile in relation to anaesthesia characteristics after a 5% micellar microemulsion of propofol in the horse. Br J Anaesth 2010;104:330–337.
15. Clark-Price SC, Posner LP, Gleed RD. Recovery of horses from general anesthesia in a darkened or illuminated recovery stall. Vet Anaesth Analg 2008;35:473–479.
16. Matthews NS, Hartsfield SM, Mercer D, et al. Recovery from sevoflurane anesthesia in horses: comparison to isoflurane and effect of postmedication with xylazine. Vet Surg 1998;27:480–485.
17. Ray-Miller WM, Hodgson DS, McMurphy RM, et al. Comparison of recoveries from anesthesia of horses placed on a rapidly inflating-deflating air pillow or the floor of a padded stall. J Am Vet Med Assoc 2006;229:711–716.
18. Donaldson LL, Dunlop GS, Holland MS, et al. The recovery of horses from inhalant anesthesia: a comparison of halothane and isoflurane. Vet Surg 2000;29:92–101.
19. Hubbell JAE. Recovery from anaesthesia in horses. Equine Vet Educ 1999;11:160–167.
20. Vettorato E, Chase-Topping ME, Clutton RE. A comparison of four systems for scoring recovery quality after general anaesthesia in horses: comparison of recovery quality scoring systems. Equine Vet J 2010;42:400–406.
21. Cicchetti DV. Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychol Assess 1994;6:284–290.
22. Gangl M, Grulke S, Caudron I, et al. Comparison of thiopentone/guaifenesin, ketamine/guaifenesin and ketamine/midazolam for the induction of horses to be anaesthetised with isoflurane. Vet Rec 2001;149:147–151.
23. Nolan A, Reid J, Welsh E, et al. Simultaneous infusions of propofol and ketamine in ponies premedicated with detomidine: a pharmacokinetic study. Res Vet Sci 1996;60:262–266.
24. England GC, Clarke KW, Goosens L. A comparison of the sedative effects of three alpha 2-adrenoceptor agonists (romifidine, detomidine and xylazine) in the horse. J Vet Pharmacol Ther 1992;15:194–201.
25. Muir WW, Sams RA, Huffman RH, et al. Pharmacodynamic and pharmacokinetic properties of diazepam in horses. Am J Vet Res 1982;43:1756–1762.
26. Frias AF, Mársico F, Gómez de Segura IA, et al. Evaluation of different doses of propofol in xylazine pre-medicated horses. Vet Anaesth Analg 2003;30:193–201.
27. Young SS, Taylor PM. Factors influencing the outcome of equine anaesthesia: a review of 1,314 cases. Equine Vet J 1993;25:147–151.
28. Matthews NS, Hartsfield SM, Sanders EA, et al. Repetitive injectable anesthesia in a 27-year-old horse. Cornell Vet 1993;83:219–225.
29. Valverde A, Black B, Cribb NC, et al. Assessment of unassisted recovery from repeated general isoflurane anesthesia in horses following post-anesthetic administration of xylazine or acepromazine or a combination of xylazine and ketamine. Vet Anaesth Analg 2013;40:3–12.
