Materials and Methods

Animals—Seventeen horses were enrolled in the study. Initial inclusion criteria for horses included signs of abdominal pain requiring general anesthesia for exploratory abdominal surgery, >2 years of age, hepatic and renal biochemical values within the reference range in samples obtained at the time of initial examination, and body weight obtained via an electronic scale immediately before anesthesia. For continued inclusion in the study, criteria were a period of lidocaine administration with a duration of 75 to 105 minutes during anesthesia, MAP ≥ 65 mm Hg immediately before and throughout the lidocaine infusion, and collection of at least 14 of 17 possible blood samples at predesignated time points for lidocaine analysis.

On the basis of these criteria, data from 6 horses were excluded (2 horses were euthanatized during surgery, 1 horse did not meet the minimum infusion duration, 1 horse exceeded the maximum infusion duration, and 2 horses did not have the minimum number of blood samples collected). The remaining 11 horses comprised the final study group and included 6 mares and 5 geldings (9 Quarter Horses, 1 Thoroughbred, and 1 Warmblood horse). Mean ± SD age was 10.3 ± 7.4 years, and mean body weight was 526 ± 40 kg.

The study protocol was reviewed and approved by the Colorado State University Animal Care and Use Committee. Because of the clinical nature of the study, the protocol was also reviewed and approved by the Director of the Veterinary Teaching Hospital before study initiation. This approval was granted on the basis of the history of safe and routine use of lidocaine administered IV to anesthetized horses at the veterinary teaching hospital. Permission to obtain blood samples for measurement of serum lidocaine concentrations was obtained from the attending clinician before a horse was enrolled in the study.

Initial examination—The 11 horses did not have access to feed (as reported by the owners at the time of initial examination) for a mean ± SD of 11.1 ± 1.6 hours. A physical examination, CBC count, and serum biochemical analysis were performed on all horses. In addition, abdominocentesis and a transrectal examination were performed on each horse at least 1 hour before they were anesthetized. A mean of 1.5 mL of a 2% solution of lidocaine hydrochloride^b was administered SC to facilitate abdominocentesis. Immediately before transrectal examination, approximately 40 mL of a 2% solution of lidocaine was administered per rectum to all horses to facilitate transrectal palpation.

Instrumentation—Before anesthetic induction, a 14gauge, 5-inch venous catheter^c was inserted in a jugular vein of each horse. This catheter was used exclusively for administration of fluids and lidocaine. Approximately 2 mL of a 2% solution of lidocaine was administered SC to facilitate placement of the venous catheter in the jugular vein. After anesthesia was induced, a 20-gauge, 2-inch catheter^d was percutaneously placed in a facial artery of each horse for measurement of blood pressure and to facilitate collection of blood samples for pH, blood gas analysis, PCV, TP, and serum concentrations of lidocaine. An ECG (base-apex lead) was recorded, and a nasopharyngeal temperature probe was also inserted. Polyionic fluids^e were administered IV to all horses at the discretion of the attending clinician before (5 to 15 L) and after anesthesia (4 mL/kg/h) and at the discretion of the anesthetist during anesthesia (≥10 mL/kg/h).

Anesthetic protocol—All horses were weighed and received medications (penicillin G potassium^f [20,000 U/kg, IV], gentamicin sulfate^g [6.6 mg/kg, IV], and flunixin meglumine^h [1.1 mg/kg, IV]) within 1 hour before anesthetic induction. Horses were administered xylazine hydrochlorideⁱ (mean ± SD, 0.27 ± 0.15 mg/kg, IV), followed 5 minutes later by IV administration of guaifenesin^j (8 horses; mean, 56 ± 33 mg/kg) or midazolam^k (3 horses; mean, 0.07 ± 0.03 mg/kg) and ketamine hydrochloride^l (mean, 1.9 ± 0.34 mg/kg) for induction of anesthesia. Because of the clinical nature of the study, induction drugs and doses were not standardized but instead were administered at the discretion of the anesthetist. Horses were orotracheally intubated and positioned in dorsal recumbency.

Mechanical ventilation^m was initiated and anesthesia maintained with sevofluraneⁿ in oxygen. Vaporizer settings were adjusted by the anesthetist as needed; vaporizer settings were recorded at time points corresponding to intraoperative collection of blood samples for lidocaine analysis. Mean ± SD vaporizer setting was 3.1 ± 0.5 for the entire anesthetic period. The anesthetist maintained discretion for aspects of case management, including the use of inotropes and vasopressors to support arterial blood pressure. Use of these agents and any other pharmacologic agents (calcium gluconate, potassium chloride, morphine, dimethyl sulfoxide, or colloid supplements [eg, hyperimmune plasma]) administered during anesthesia by the anesthetist or at the request of the surgeon was recorded. Horses did not receive any other pharmacologic agents with known prokinetic effects.

Experimental protocol—Horses received a loading infusion of lidocaine (1.3 mg/kg [87 μg/kg/min], IV) during a 15minute period, followed immediately by a maintenance infusion of lidocaine (50 μg/kg/min) for an additional 60 to 90 minutes, resulting in a total infusion duration of 75 to 105 minutes. Lidocaine was administered by use of a calibrated syringe infusion pump.^o Administration was initiated approximately 10 to 15 minutes after induction of anesthesia (after insertion of instruments, establishment of study conditions, and collection of baseline data).

A sample (20 mL) of arterial blood was collected before (baseline), at the midpoint (7.5 minutes), and at completion (15 minutes) of the loading lidocaine infusion (time 0 was designated as the start of the lidocaine infusion). Arterial blood samples were collected at 15-minute intervals during the maintenance infusion, at 30-minute intervals for the first 2 hours after discontinuation of the lidocaine infusion, and then at 60-minute intervals for the subsequent 4 hours. When the arterial catheter could not be maintained (1 horse; 7 samples, all of which were in the period after lidocaine infusion), 20 mL of venous blood was collected from the uncatheterized jugular vein at the predesignated time points. In a previous study,^p investigators documented that there is no difference in lidocaine concentrations for samples obtained from various blood collection sites. After clotting, samples were centrifuged and the serum supernatant was collected, placed in tubes, and stored at –40°C until analysis.

Heart rate and rhythm, MAP, and nasopharyngeal temperature were recorded before lidocaine administration (baseline) and at each predesignated time point for blood sample collection throughout the loading and maintenance infusions of lidocaine. The MAP was determined directly by use of a calibrated pressure transducer^q that was calibrated to 0 at the presumed level of the heart base (point of shoulder). Three milliliters of arterial blood was collected in a heparinized syringe (needle and syringe were flushed with heparin [1,000 U/mL], and then the heparin was discarded) for pH, blood gas analysis^r (PaCO₂ and PaO₂), PCV measurement, and TP concentrations before and at 15 and 75 minutes during lidocaine administration and at 60 minutes after discontinuation of the lidocaine infusion. Samples were analyzed immediately; pH and blood gas values were corrected on the basis of body temperature. Although there was a presumed confounding influence of anesthetic medications, horses were continually observed for potential cardiovascular (eg, hypotension and dysrhythmias) and neurologic (eg, tremors and muscle fasciculation) manifestations associated with lidocaine toxicosis.

At the conclusion of the surgical procedure, the lidocaine infusion was discontinued and the endotracheal tube disconnected from the anesthetic breathing circuit. Horses were then transferred to a 4 × 4-m padded recovery stall where they were continuously observed. Horses typically recovered without assistance. Oxygen insufflation (15 L/min) via the endotracheal tube was maintained until horse movement precluded administration. Recovery quality was subjectively assessed by use of a scoring system described else-where.¹⁴ Briefly, a score of 1 was considered excellent and indicated a single coordinated effort to stand with minimal ataxia, and a score of 5 was considered unacceptably poor as indicated by multiple uncoordinated attempts to stand resulting in injury. Time intervals until sternal recumbency, standing posture, and extubation and the number of attempts to achieve sternal and standing postures were also recorded.

Determination of serum concentrations of lidocaine—Serum concentrations of lidocaine were measured in duplicate by use of high-performance liquid chromatography with mass spectral detection. Briefly, reference, calibration, and test samples were pipetted into autosampler vials and vortexed for 5 to 10 seconds. A mixture containing acidic acetonitrile (acetonitrile:1M acetic acid [9:1]), including mepivicane (200 ng/mL; internal standard), was added to each sample vial. After addition of the internal standard, the contents of each vial were again mixed for 1 minute on a multipulse rack vortex mixer^s (speed, 60 to 70; pulse, 60). All samples were refrigerated for 30 minutes at 4°C and then centrifuged at 1,580 × g for 10 minutes at 4°C. Vials were transferred to an autosampler rack, and the supernatant was injected for analysis.

Quantitative analyses were performed by use of a mass spectrometer^t equipped with a liquid chromatograph system.^u The mobile phase was composed of a solvent mixture of acetonitrile with 0.05% TFA and water with 0.05% TFA. The liquid chromatograph pump provided a gradient of the acetonitrile from 30% to 90% during a period of 10.5 minutes at a flow rate of 0.9 mL/min.

The concentration of lidocaine in each sample was determined in duplicate by the internal standard method by use of the peak area ratio and linear regression analysis. Lidocaine response was linear and yielded a correlation coefficient of ≥0.99. The technique was optimized to provide a minimum limit of quantitation of 10 ng/mL for each analyte. The concentration of one of the metabolites of lidocaine metabolism, monoethylglycinexylidine, was periodically measured in each group of samples and was found to be negligible (< 10% of parent drug values); therefore, further analysis of the metabolite was not performed.

Pharmacokinetic analysis—Serum concentrations of lidocaine were evaluated by use of standard noncompartmental analysis. Maximum concentration of lidocaine and time of the maximum concentration of lidocaine were estimated from the data. Terminal half-life was calculated as 0.69/k_el, where k_el is the elimination rate constant, which was calculated as the slope of the terminal phase of the serum concentration curve that included a minimum of 3 points. Linear trapezoidal areas were used to calculate the AUC. Values were calculated for the AUC from 0 to 15 minutes, AUC from 15 minutes to the end of lidocaine infusion at 75 to 105 minutes, AUC from the end of lidocaine infusion to the end of sample collection at 6 hours after discontinuation of infusion, and AUC extrapolated to infinity. Other pharmacokinetic variables (eg, total body clearance, Vd_ss, and mean residence time) were determined with statistical software^v by use of standard noncompartmental equations.

Statistical analysis—Data were summarized as mean ± SD. Significance of time effects on outcome variables was assessed by use of a restricted maximum likelihood-based, mixed-effect model that included the fixed effect of time and a random effect of animal. Analysis was performed by use of a software option for fitting linear models.^w Contrasts (least square means) between the baseline value and values for all other time points were made when the overall effect of time was significant (P < 0.05).

Results

Gastrointestinal lesions identified and corrected at the time of surgery in the 11 horses of the study included 1 rectal tear, 1 strangulating (small intestinal) lipoma, 1 cecal impaction, 2 large-colon volvuluses (360°), and 6 large-colon displacements. Mean ± SD duration of anesthesia was 105 ± 21 minutes. There were no specific signs of lidocaine toxicosis (muscle fasciculations or hypotension) observed in any horse during the study. All horses survived to discharge from the hospital.

Lidocaine administration was discontinued after 75 minutes in 5 horses, 90 minutes in 3 horses, and 105 minutes in 3 horses. Lidocaine disposition was best described by use of a standard noncompartmental model because of insufficient frequency of initial sample collection after discontinuation of the infusion. The disposition pattern for lidocaine was determined for this group of horses (Figure 1). Mean ± SD baseline serum concentration of lidocaine was 8.8 ± 17.0 ng/mL. Lidocaine concentration increased to 2,032 ± 415 ng/mL by the end of the loading infusion period and ranged from 1,456 ± 385 ng/mL to 2,183 ± 263 ng/mL during the maintenance infusion. Serum concentrations returned to baseline values within 6 hours after discontinuation of the infusion, although the most notable decrease in concentration was within 30 minutes. The pharmacokinetic variables derived from measured lidocaine concentrations were summarized (Table 1).

View Full Size

Mean ± SD serum concentration of lidocaine in 11 anesthetized horses with colic in which lidocaine hydrochloride was administered IV (loading infusion, 87 mg/kg/min for 15 minutes;maintenance infusion, 50 mg/kg/min for 60 to 90 minutes). Start of the loading lidocaine infusion was designated as time 0. Notice that the start (arrow) and end (arrowhead) of the maintenance infusion are indicated on the x-axis.

Citation: American Journal of Veterinary Research 67, 2; 10.2460/ajvr.67.2.317

• Download Figure
• Download figure as PowerPoint slide

Mean ± SD heart rate at baseline and during lidocaine administration ranged from 36 ± 1 beats/min to 43 ± 9 beats/min (Table 2). The MAP ranged from 74 ± 18 mm Hg to 89 ± 10 mm Hg. Significant differences were observed in nasopharyngeal temperature, which ranged from 35.1 ± 1.1°C to 36.1 ± 0.6°C. The overall vaporizer setting for sevoflurane was 3.1 ± 0.5% (3.6 ± 0.5% at the start of the loading infusion, 3.4 ± 0.2% at the start of the maintenance infusion, and 2.6 ± 0.3% at the conclusion of the maintenance infusion). Arterial pH, blood gas values, PCV, and TP concentrations were summarized (Table 3).

Dobutamine^x was administered IV periodically during sevoflurane-induced anesthesia in all horses except 2 (mean ± SD rate, 0.69 ± 0.58 μg/kg/min). Additional pharmacologic agents administered IV during surgery included ephedrine^y (0.06 mg/kg [1 horse]), morphine sulfate^z (0.03 mg/kg [1 horse]), ketamine (0.2 mg/kg [2 horses] and 0.3 mg/kg [1 horse]), 23% calcium gluconate^aa (23 mg/kg [1 horse] and 102 mg/kg [1 horse]), potassium chloride^bb (0.3 mEq/kg [1 horse]), 12% dimethyl sulfoxide^cc (0.44 g/kg [1 horse]), and hyperimmune plasma^dd (2 mL/kg [2 horses]). Additional drugs administered in transition to or during recovery included xylazine (0.09 to 0.20 mg/kg [4 horses]) and ketamine (0.2 to 0.5 mg/kg[3 horses]). Three horses did not receive any additional pharmacologic agents during anesthetic maintenance or transition to recovery.

Mean ± SD time until sternal recumbency and standing posture following discontinuation of anesthesia and lidocaine infusion for all horses was 31.5 ± 7.3 minutes and 42.4 ± 12.1 minutes, respectively. Horses were extubated 47.3 ± 11.8 minutes after disconnection from the anesthetic breathing circuit. Horses had a mean of 3.8 ± 3.9 attempts until they achieved sternal recumbency and 3.2 ± 2.4 attempts until they were able to stand. Mean quality of recovery was 3.1 ± 1.2; 1 horse had an excellent recovery, and 1 horse was considered to have an unacceptably poor recovery. This latter horse had signs of obturator nerve damage during recovery and required assistance by use of ropes on the head and tail to stand.

Discussion

The anesthetic management of these horses was generally uneventful. Although 1 horse needed assistance in the recovery period, the overall recovery quality was similar to that in other reports^15-17 for horses waking from inhalation anesthetics.

The disposition pattern of lidocaine obtained in the study reported here is comparable to that obtained for healthy horses administered lidocaine in accordance with a similar dosing regimen¹³ in that serum concentrations increased rapidly with lidocaine administration and returned to baseline values by 6 hours after lidocaine administration was discontinued. The fact that baseline serum lidocaine values were >0 is likely explained by the absorption of lidocaine administered SC for abdominocentesis and insertion of a venous catheter. There may also have been some contribution from the systemic absorption of lidocaine administered per rectum to facilitate transrectal palpation.

The influence of anesthesia on lidocaine concentration and, to a lesser extent, disposition in horses has been assessed.^1,13,18 In 1 study,¹⁸ investigators reported higher but more variable plasma lidocaine concentrations (30 to 4,230 ng/mL) in healthy anesthetized horses, compared with concentrations obtained for unanesthetized horses. In another study,¹ similar serum concentrations were measured in anesthetized surgical patients (1,060 ± 600 ng/mL) administered one half of the lidocaine infusion dose of awake postoperative patients (1,000 ± 520 ng/mL); investigators in that study subsequently recommended the lower dose for horses undergoing anesthesia. In another study¹³ conducted by our laboratory group, a clear influence of anesthesia on lidocaine disposition was identified and attributed primarily to a decrease in cardiac output and hepatic blood flow in anesthetized horses, compared with values for awake horses (Vd_ss and total body clearance were 0.40 ± 0.09 L/kg and 15 ± 3.3 mL/min/kg in anesthetized horses and 0.79 ± 0.16 L/kg and 29 ± 7.6 mL/min/kg in awake horses).¹³

We hypothesized that lidocaine disposition would be further influenced by the systemic effects of gastrointestinal tract disease in anesthetized horses and cause increases in serum concentrations and further decreases in the drug distribution and clearance pattern, resulting in a potential increase in the risk of drug toxicosis. Surprisingly, we found the opposite to be true in this population of horses because lidocaine concentrations were lower and, in turn, distribution and total body clearance values were higher (mean ± SD Vd_ss, 0.7 ± 0.31 L/kg; total body clearance, 25 ± 3 mL/min/kg), compared with values for healthy anesthetized horses (Vd_ss, 0.40 ± 0.09 L/kg; total body clearance, 15 ± 3.3 mL/min/kg) reported elsewhere.¹³ Values reported here are similar to those obtained in awake horses in that same study.¹³

It is probable that despite anesthesia, horses with gastrointestinal tract disease maintained a higher circulating blood volume and cardiac output than anesthetized healthy horses studied in controlled laboratory conditions.¹³ In another study,¹⁹ investigators reported that anesthetized horses receiving a low dose of endotoxin (0.1 μg/kg) to mimic coliclike conditions maintained a higher cardiac index than control horses. Although surgical intervention was deemed necessary on the basis of the overall clinical evaluation of horses in our report, objective clinical variables (eg, heart rate, MAP, and pH) and blood values (eg, PCV and TP concentration) within the respective reference ranges suggest mild to moderate systemic disease.²⁰ Blood lactate concentrations were not measured in the group of horses in the study reported here, which may have provided additional evidence to support the disease sta-^21-23 Horses in our study were aggressively treated by IV administration of fluids and other pharmacologic agents, which likely contributed to improved cardiovascular performance.^24,25 The perianesthetic use of α₂-adrenoceptor agonists, such as xylazine, which substantially reduces cardiac output in horses, was also minimized.^26,27

Other factors that reduce lidocaine clearance include withholding of food and hepatic or renal dys-function.²⁸ Hepatic and renal biochemical values were within the reference range for horses, and duration for withholding of food (ie, lack of access to feed) reported by owners was similar to that reported in another study¹³ of healthy anesthetized horses. The loading and maintenance infusion dosages used in most of the aforementioned studies and the study reported here have been based on the recommendation of a study^a in which investigators reported beneficial effects in horses with gastrointestinal tract disease after administration of this dose of lidocaine. Hence, it is unlikely that administration in accordance with this protocol explains the observed pharmacokinetic differences.¹³ However, the shorter mean duration for lidocaine administration and resulting lower cumulative dose in 8 of 11 horses in our study may have contributed to differences.

On the basis of results for this population of horses with gastrointestinal lesions requiring surgical intervention, analysis of our findings suggests that modification of the lidocaine dose is not necessary. However, we caution against extrapolating these results to all horses requiring gastrointestinal tract surgery because the influence of severe disease on drug disposition remains unclear.