Ion channels important to anesthetic and analgesic actions can exhibit nonspecific, low-affinity modulation by high concentrations of sufficiently water-soluble hydrocarbons. This molar water solubility hypothesis has been used to explain the anesthetic actions of conventional inhaled agents and endogenous metabolites.1–3 More recently, food-derived phytochemicals, such as the spearmint monoterpene L-carvone, have been shown to modulate anesthetic-sensitive N-methyl-d-aspartate receptors and voltage-gated sodium channels at high concentrations in vitro and to produce analgesia at high doses in sheep in vivo.4,5
A potential advantage of using natural food-derived livestock therapeutics is that many of these compounds have established concentrations that are generally regarded as safe for human consumption. If use of these therapeutics produces tissue drug concentrations below this established threshold, it might be possible to administer food-derived therapeutics to improve livestock welfare without concern for food residues or environmental contamination.
An injectable formulation of L-carvone has previously been shown to increase threshold responses to an electrical noxious stimulus in sheep for up to 5 hours when administered IM.5 One aim of the present study was to measure the plasma L-carvone concentrations associated with analgesia in sheep to estimate the drug therapeutic range. The second aim was to develop a pharmacokinetic model to describe the absorption, distribution, and elimination of IM L-carvone from sheep plasma. Third was to determine whether drug concentrations in sheep tissues were less than or equal to concentrations of L-carvone naturally found in mint used for human consumption or L-carvone concentrations allowed as food flavorings. Fourth and finally, this study aimed to evaluate the effects of L-carvone administration on hematological parameters and blood biochemistry in sheep.
Methods
This work was approved by the IACUC at the University of California-Davis (#22625). Six sheep, consisting of 5 wethers and 1 ewe, weighing 69 ± 5 kg (mean ± SD), and aged approximately 16 months, were studied and were the same animals used for a pharmacodynamics study of this same agent and dose following a 2-to-3-week drug washout period.5 Experiments were conducted using pairs of sheep in a 4-m2 study cubicle within a dedicated sheep barn. Animals always had access to hay and water.
The skin over the jugular furrow of each sheep was shaved, aseptically prepared, and desensitized with lidocaine, and a 16-gauge, 8.3-cm jugular catheter was placed percutaneously, sutured to the skin, and flushed with heparinized saline. Drug for injection was prepared in a sterile multidose vial by combining equal volumes of L-carvone (natural, 99%, food grade; Sigma-Aldrich) with a vehicle consisting of a 1:1 mixture of undenatured dehydrated ethanol (200 proof, USP, Kosher; Spectrum Chemical) and propylene glycol (≥ 99.5%, natural, Food Chemicals Codex food grade; Sigma-Aldrich). Freshly prepared solutions were drawn up through sterile 0.22-mm syringe-driven filter units (Millex-GS; Millipore), and 0.075 mL/kg of this 50% L-carvone formulation was injected into each semitendinosus muscle. Hence, a total of 0.15 mL/kg was administered IM to each sheep, equivalent to 71.6 mg/kg of L-carvone.
Blood was collected into EDTA tubes (Vacutainer; BD) immediately prior to drug administration (time = 0 minutes) and then 3, 6, 9, 12, 15, and 30 minutes and 1, 2, 3, 4, 5, 6, 8, and 24 hours after L-carvone injection. Blood was promptly centrifuged, plasma was pipetted into cryotubes, and samples were stored at −80 °C until analyzed. Additional blood was collected at 0 minutes, 120 minutes (2 hours), and 1,440 minutes (24 hours) into EDTA or serum collection tubes and promptly submitted to the Clinical Diagnostic Laboratory at the William R. Pritchard Veterinary Medical Teaching Hospital for a CBC and serum biochemistry panel. At the end of the study, sheep were deeply anesthetized with 5 mg/kg IV propofol and then euthanized with 0.5 mL/kg IV of a saturated KCl solution. Samples from the semitendinosus muscle (at the site of L-carvone injection), triceps muscle, liver, kidney, and omental fat were then collected and frozen at −80 °C until drug concentrations could be measured.
Plasma calibration standards were prepared by diluting L-carvone (Sigma Aldrich) with drug-free sheep plasma to concentrations ranging from 0.01 to 250 μg/mL. Calibration curves and negative control samples were prepared fresh for each quantitative assay. Quality control samples (sheep plasma fortified with analyte at 3 concentrations within the standard curve) were included as an additional check of accuracy.
L-carvone concentrations in tissue and plasma were measured using LC-MS-MS.4,5 Plasma samples (200 μL) were diluted with 150 μL of acetonitrile (ACN):1 M acetic acid (9:1, v:v) and 50 μL of methanol to precipitate proteins. The samples were vortexed for 2 minutes, refrigerated for 20 minutes, vortexed for an additional 1 minute, and centrifuged (3,102 X g) for 10 minutes at 4 °C prior to injection of 10 μL into an LTQ XL Orbitrap mass spectrometer (Thermo Scientific) coupled with a Waters Acquity HPLC. The system was operated at a resolution of 60,000 (M/ΔM, at full width at half maximum of the mass peaks) using positive electrospray ionization. Detection and quantification were conducted using full-scan accurate mass from 145 to 160 (m/z). The responses were plotted using a 20-ppm mass tolerance for L-carvone (151.11160 [m/z]). The spray voltage was 3,500 V, and the sheath and auxiliary gas were 45 and 20, respectively (arbitrary units). Chromatography employed an ACE 3 C18 10-cm X 1-mm 3-μm column (Mac-Mod Analytical) and a linear gradient of ACN in water with a constant 0.2% formic acid at a flow rate of 0.1 mL/min. The initial ACN concentration was held at 1% for 0.4 minutes and ramped to 99% over 9 minutes before re-equilibrating for 14 minutes at initial conditions.
Assay precision and accuracy were determined using quality control samples in replicates (n = 6). Accuracy is reported as percentage of nominal concentration and precision as percentage of relative SD. For the L-carvone, accuracy was 112% for 1.5 μg/mL, 106% for 10 μg/mL, and 114% for 100 μg/mL. Precision was 5% for 1.5 μg/mL, 2% for 10 μg/mL, and 2% for 100 μg/mL. The technique was optimized to provide a limit of quantitation of 0.01 μg mL−1 and a limit of detection of approximately 0.005 μg/mL for L-carvone.
For L-carvone measurements, the LC-MS-MS instrument response for all analytes was linear and gave correlation coefficients of 0.99 or better. The precision and accuracy of the assay were determined by assaying quality control samples in replicates (n = 6). Accuracy was reported as percentage of nominal concentration and precision as percentage of relative SD. For plasma, the precision was 2% for 0.3 ug/mL, 2% for 10 ug/mL, and 2% for 100 ug/mL, and accuracy was 99% for 0.3 ug/mL, 112% for 10 ug/mL, and 106% for 100 ug/mL. For tissue, precision was 7% for 300 ng, 3% for 2,000 ng, and 1% for 20,000 ng, and accuracy was 99% for 300 ng, 97% for 2,000 ng, and 95% for 20,000 ng.
For measurements of L-carvone in solid tissue, approximately 100 mm3 samples were weighed and placed into 7-mL Precellys hard-tissue homogenizing vials (Omni International), where they were homogenized at 4.5 m/s for 30 seconds in an Omni Bead Ruptor Elite tissue homogenizer (Omni International) with a 5-minute cooldown. A mixture of ACN:1 M acetic acid (1 mL; 9:1, v:v) and 50 μL of methanol was added, and the samples were shaken. Next, 500 μL of the homogenate was transferred to a microcentrifuge tube, and 1 mL of hexane was added. Tubes were subsequently rotated for 10 minutes and centrifuged at 17,000 XG for 5 minutes. Supernatant (300 μL) was transferred to an autosampler vial with insert, and 10 μL was injected into the LC-MS-MS system using the analytical method described above for plasma. Calibration samples were prepared by adding the working standard solutions to the Precellys vials at concentrations ranging from 0.1 to 30 μg/vial. For the L-carvone quality control samples, accuracy was 107% for 0.3 μg/vial, 100% for 2 μg/vial, and 101% for 20 μg/vial. Precision was 2% for 0.3 μg/vial, 1% for 2 μg/vial, and 1% for 20 μg/vial. The technique was optimized to provide a limit of quantitation of 0.1 μg/vial and a limit of detection of approximately 0.01 μg/vial.
L-carvone concentration plasma data were analyzed using noncompartmental analysis with a commercially available pharmacokinetic software program (Phoenix Winnonlin, version 8.3; Certara). After performing noncompartmental analysis, pharmacokinetic modeling using a nonlinear mixed effect modeling approach with the Phoenix nonlinear mixed effect software program and concentration data was conducted. The first-order conditional estimation method with interaction was used in the model-building process. One- and 2-compartment models with and without lag times and with different error models (additive, multiplicative, Poisson, and mixed additive/multiplicative) were assessed using L-carvone concentration data. A diagonal variance-covariance matrix was used for the random effects. Visual analysis of the observed-versus-predicted concentration graphs, residual plots, Akaike information criterion, percentage of coefficient of variation, and −2 log likelihood (a measure of model deviance) were considered in assessing which model provided the best fit.
Plasma and tissue L-carvone concentration measurements and pharmacokinetic model parameters were summarized as medians and IQRs. Hematology and biochemistry data were summarized as the arithmetic mean ± SD; normal distribution of these variables was confirmed using Shapiro-Wilk tests, and data were analyzed using repeated-measures ANOVAs with comparisons of predrug values to 2-hour and 24-hour postdrug values made using Holm-Šidák tests (SPSS Statistics, version 29; IBM Corp). Differences were considered significant when P < .05.
Results
All sheep completed the study without evidence of any discomfort or lameness. One of the wethers (sheep #4) exhibited signs of sedation and was sternally recumbent—but awake—during the first 15 minutes following drug injection. Otherwise, all animals were intermittently standing, eating, and drinking throughout the study period. Following euthanasia, gross examination of the carcasses revealed no notable findings except for a 2.5-cm pale, wedge-shaped area of necrosis in the semitendinosus muscle at the site of injection in all animals (Supplementary Figure S1).
L-carvone concentrations measured in plasma and tissues are listed in Tables 1 and 2, and mean plasma concentrations over time are displayed in Figure 1. Although concentrations peaked within 12 to 15 minutes after administration, maximum L-carvone concentrations varied by a factor of 6.9 between individuals. Drug was still detectable in plasma even 24 hours later as it was in the local semimembranosus injection site in all animals and in the remote triceps muscle in half of the sheep studied. Low concentrations of L-carvone were present in the fat of all animals, but most did not have detectable concentrations present in the liver and kidney.
Concentrations of L-carvone (µg/mL) in sheep plasma measured over time.
L-carvone (µg mL−1) | Sheep #1 | Sheep #2 | Sheep #3 | Sheep #4 | Sheep #5 | Sheep #6 | Median (IQR) |
---|---|---|---|---|---|---|---|
0 min | ND | ND | ND | ND | ND | ND | 0 (0) |
3 min | 0.50 | 0.15 | 0.62 | 0.57 | 0.11 | 0.11 | 0.48 (0.32) |
6 min | 0.72 | 0.33 | 0.71 | 1.04 | 0.27 | 0.27 | 0.70 (0.30) |
9 min | 0.84 | 0.36 | 0.88 | 1.27 | 0.27 | 0.27 | 0.79 (0.41) |
12 min | 0.81 | 0.34 | 0.86 | 1.64 | 0.28 | 0.28 | 0.83 (0.40) |
15 min | 0.75 | 0.34 | 0.98 | 1.93 | 0.27 | 0.27 | 0.79 (0.51) |
30 min | 0.70 | 0.31 | 1.00 | 1.54 | 0.27 | 0.27 | 0.73 (0.53) |
60 min | 0.55 | 0.21 | 0.99 | 1.14 | 0.23 | 0.23 | 0.52 (0.58) |
120 min | 0.34 | 0.16 | 0.79 | 0.58 | 0.25 | 0.25 | 0.33 (0.24) |
180 min | 0.27 | 0.14 | 0.68 | 0.37 | 0.24 | 0.24 | 0.28 (0.11) |
240 min | 0.27 | 0.11 | 0.69 | 0.30 | 0.27 | 0.27 | 0.27 (0.09) |
300 min | 0.21 | 0.10 | 0.60 | 0.25 | 0.25 | 0.25 | 0.23 (0.06) |
360 min | 0.16 | 0.08 | 0.64 | 0.19 | 0.22 | 0.22 | 0.18 (0.07) |
480 min | 0.11 | 0.08 | 0.50 | 0.16 | 0.18 | 0.18 | 0.14 (0.07) |
1,440 min | 0.06 | < LOQ | 0.16 | 0.05 | < LOQ | < LOQ | 0.03 (0.06) |
IQR data presented as 75th to 25th percentiles.
LOQ = Limit of quantitation. ND = Not detected.
A concentration of 0 µg/mL was assumed for calculations of the median and IQR when L-carvone was ND or < LOQ.
Concentrations of L-carvone (µg/g tissue) in sheep measured over time.
L-carvone (µg g−1) | Sheep #1 | Sheep #2 | Sheep #3 | Sheep #4 | Sheep #5 | Sheep #6 | Median (IQR) |
---|---|---|---|---|---|---|---|
Semitendinosus | 94.3 | 43.9 | 6.22 | 110 | 138 | 110 | 102 (53) |
Triceps | ND | 4.31 | ND | 7.92 | 1.62 | < LOQ | 0.81 (3.64) |
Liver | < LOQ | 2.22 | ND | ND | ND | ND | 0 (0) |
Kidney | ND | ND | ND | ND | ND | ND | 0 (0) |
Fat | 1.60 | 0.87 | 3.42 | 1.69 | 1.34 | 1.47 | 1.54 (0.29) |
IQR data presented as 75th to 25th percentiles.
The semitendinosus sample was comprised of the drug injection site. A concentration of 0 µg/mL was assumed for calculations of the median and IQR when L-carvone was ND or < LOQ.
The final pharmacokinetic model was best described by a 2-compartment model with a lag time, parameterized with respect to clearance. A Poisson residual error model was used. Pharmacokinetic parameters (estimate and percentage of coefficient of variation for the fixed and random effects) for the model are listed in Tables 3 and 4. High interanimal variability in plasma L-carvone concentrations mirrors variability in the calculated pharmacokinetic parameters, such as the peak plasma concentration and terminal half-life, which varied by factors of 6.9 and 4.3, respectively. Although individual PK models fit data tightly, there was much greater data spread around the population models (Supplementary Figures S2 and S3). This is reflected by the relatively large percentage of coefficient of variation in the half-lives, volumes of distribution, and clearances for the population model (Table 4).
Select pharmacokinetic parameters for L-carvone generated from noncompartmental analysis following a single IM administration of 71.6 mg/kg to 6 sheep.
Parameter | Mean ± SD | Median | Range |
---|---|---|---|
Cmax (µg mL−1) | 0.72 ± 0.59 | 0.84 | 0.28–1.93 |
Tmax (min) | 14.5 ± 7.92 | 12.0 | 9.0–30.0 |
AUCinf (ug min mL−1) | 214.3 ± 251.3 | 232.3 | 95.5–744.1 |
AUC extrap (%) | 22.0 ± 13.2 | 20.0 | 11.2–47.1 |
Terminal t½ (min) | 412.5 ± 222.4 | 467.9 | 185.2–798.3 |
AUC = Area under the concentration curve. AUC extap = Percentage of the AUC extrapolate. AUCinf = Area under the concentration time curve from 0 to infinity. Cmax = Maximum concentration. t½ = Half-life. Tmax = Time of maximum concentration.
Mean is geometric mean.
Model estimated population mean values for L-carvone following a single IM administration of 71.6 mg/kg to 6 sheep.
Parameter | Estimate | CV (%) |
---|---|---|
Tlag (min) | 0.694 | 44.5 |
Ka (min−1) | 0.266 | 20.3 |
V/F (L kg−1) | 92.6 | 27.7 |
V2/F (L kg−1) | 40.1 | 68.8 |
Cl/F (mL min−1 kg−1) | 246.6 | 26.0 |
Cl2/F (mL min−1 kg−1) | 552.2 | 40.5 |
stdev0 | 0.0625 | 9.70 |
Ke (min−1) | 0.003 | 38.9 |
α (min−1) | 0.021 | 53.1 |
β (min−1) | 0.002 | 40.1 |
AUC (µg min mL−1) | 290.0 | 26.9 |
α t½ (min) | 33.7 | 53.1 |
β t½ (min) | 390.2 | 40.1 |
Ke t½ (min) | 260.6 | 39.0 |
Ka t½ (min) | 2.61 | 20.3 |
Between-subject variability (%CV) | ||
tLag | 0.122 | 36.0 |
Ka | 0.135 | 38.0 |
V | 0.457 | 76.1 |
V2 | 2.52 | 337.3 |
Cl | 0.418 | 72.0 |
Cl2 | 0.329 | 62.4 |
α and β = Slopes for the modeled equation. α t½ = Phase 1 half-life. β t½ = Phase 2 half-life. Cl = Clearance of drug from plasma. Cl2 = Clearance of drug from the peripheral compartment. CV = Coefficient of variation. F = Bioavailability. Ka = Rate of absorption. Ka t½ = Absorption half-life. Ke t½ = Elimination half-life. Lag = Lag time. stdev0 = Estimated residual SD for plasma data. tlag = Lag time between injection and first measurable plasma concentration. V = Central volume of distribution. V2 = Peripheral volume of distribution.
Clinical laboratory results (Tables 5 and 6) showed a mild increase in neutrophil count—although still within the reference range—and a statistically and clinically significant increase in creatine kinase and AST. These findings are consistent with muscle injury, presumably caused by injection site inflammation and myonecrosis. Other statistically significant parameter changes, as occurred with the 2-hour serum electrolytes, were transient, remained within reference ranges, and are of questionable clinical importance. Interestingly, blood glucose was always higher than the laboratory reference range, although this value decreased during experiments as sheep acclimated to the study conditions and were presumably less stressed.
Mean (± SD) of CBC results for 6 sheep before and 2 and 24 hours after IM L-carvone administration.
Parameter | Predrug | 2 hours postdrug | 24 hours postdrug |
---|---|---|---|
RBC (106 cells µL−1) | 11.36 ± 0.69 | 10.68 ± 0.61 | 11.55 ± 0.57 |
Hemoglobin (g dL−1) | 12.0 ± 0.7 | 11.2 ± 0.8 | 12.2 ± 0.4 |
Hematocrit (%) | 33.2 ± 2.4 | 31.0 ± 2.8 | 33.8 ± 0.8 |
MCV (fL) | 29.3 ± 1.6 | 29.0 ± 1.7 | 29.4 ± 1.3 |
MCH (pg) | 10.6 ± 0.5 | 10.5 ± 0.6 | 10.5 ± 0.5 |
MCHC (g dL−1) | 36.1 ± 1.1 | 36.2 ± 0.8 | 35.9 ± 0.7 |
WBC (µL−1) | 6,765 ± 1,365 | 8,443 ± 1,316a | 6,743 ± 871 |
Neutrophils (µL−1) | 2,276 ± 705 | 4,223 ± 1,112a | 2,706 ± 1,032 |
Lymphocytes (µL−1) | 4,139 ± 1,205 | 4,002 ± 831 | 3,809 ± 954 |
Monocytes (µL−1) | 231 ± 267 | 140 ± 139 | 136 ± 70 |
Eosinophils (µL−1) | 73 ± 59 | 47 ± 47 | 70 ± 40 |
Basophils (µL−1) | 34 ± 12 | 35 ± 35 | 26 ± 9 |
Platelets (µL−1) | 559,000 ± 160,000 | 515,000 ± 135,000 | 506,000 ± 131,000 |
Plasma protein (g dL−1) | 7.1 ± 0.4 | 7.1 ± 0.3 | 7.4 ± 0.3a |
Fibrinogen (mg dL−1) | 200 ± 100 | 200 ± 100 | 200 ± 100 |
Value with superscript is significantly (P < .05) different from predrug values.
No mean value for any parameter fell outside the reference range established by the diagnostic laboratory.
Mean (± SD) of chemistry panel results for 6 sheep before and 2 and 24 hours after IM L-carvone administration.
Parameter | Predrug | 2 hours postdrug | 24 hours postdrug |
---|---|---|---|
Sodium (mmol L−1) | 144 ± 1 | 149 ± 2a | 145 ± 1 |
Potassium (mmol L−1) | 5.1 ± 0.7 | 4.5 ± 0.6a | 4.7 ± 0.3 |
Chloride (mmol L−1) | 107 ± 2 | 110 ± 2a | 109 ± 2 |
Bicarbonate (mmol L−1) | 21 ± 2 | 23 ± 0a | 23 ± 2 |
Phosphorus (mg dL−1) | 5.7 ± 0.5 | 5.8 ± 0.4 | 5.1 ± 0.8 |
Calcium (mg dL−1) | 10.7 ± 0.4 | 10.7 ± 0.3 | 11.0 ± 0.3 |
BUN (mg dL−1) | 27 ± 3 | 28 ± 3 | 27 ± 4 |
Creatinine (mg dL−1) | 0.7 ± 0.1 | 0.7 ± 0.1 | 0.7 ± 0.1 |
Glucose (mg dL−1) | 110 ± 17b | 95 ± 14b | 87 ± 7.2a,b |
Serum protein (g dL−1) | 6.9 ± 0.4 | 6.8 ± 0.3 | 7.2 ± 0.4a |
Albumin (g dL−1) | 3.9 ± 0.1 | 3.9 ± 0.2 | 4.1 ± 0.2a |
Globulin (g dL−1) | 3.0 ± 0.3 | 2.9 ± 0.2 | 3.1 ± 0.2 |
AST (IU L−1) | 222 ± 104 | 267 ± 92a | 563 ± 128a,b |
CK (IU L−1) | 197 ± 86 | 3,362 ± 1,886a,b | 5,019 ± 3,836a,b |
ALP (IU L−1) | 176 ± 52 | 168 ± 59 | 172 ± 52 |
GGT (IU L−1) | 57 ± 8 | 58 ± 8 | 62 ± 10a |
SDH (IU L−1) | 44 ± 27 | 45 ± 23 | 55 ± 35 |
Total bilirubin (mg dL−1) | 0.1 ± 0 | 0.1 ± 0 | 0.1 ± 0 |
Discussion
Intramuscular administration of L-carvone in an ethanol-propylene glycol vehicle is associated with reasonably rapid increases in plasma drug concentration, peaking at about 15 minutes, consistent with previous reports of analgesia onset in sheep by this timepoint.5 Slow redistribution of L-carvone from the central-to-peripheral body compartments, described by the α half-life, is responsible for most of the initial decline in concentration after this time. The elimination half-life is approximately 6.5 hours, signifying that L-carvone metabolism and excretion is relatively slow compared to many conventional opioid analgesics in sheep, including as alfentanil,6 buprenorphine,7 morphine,8 oxycodone,9 and tramadol.10 Since pain requiring analgesics is usually not transient, having a therapeutic with a long half-life and slow clearance may be advantageous. However, a slow terminal half-life, coupled with the large volume of distribution and high lipid solubility of L-carvone,11 suggests that the duration of effect might become progressively prolonged with repeated dosing as has been noted to occur with infusions of very lipid-soluble opioids having a long terminal half-life and relatively large steady-state distribution volume.12
The median peak plasma L-carvone concentrations in sheep were close to 10−5 M. In vitro electrophysiology studies4 show that similar concentrations inhibit currents from N-methyl-d-aspartate receptors and voltage-gated sodium channels by only 5% to 8%. Given these modest effects, 1 of 3 possibilities seems likely: (1) plasma drug concentrations underestimate the L-carvone concentrations present at the effect site tissues during corresponding analgesic testing times; (2) in vitro responses of nonovine ion channels underestimate in vivo drug responses in sheep; or (3) there may be multiple other—and as yet undefined—targets besides glutamate receptors and sodium channels that contribute to L-carvone analgesic actions. Analgesic effects in sheep are reported to last up to 5 hours.5 This corresponds to a median L-carvone concentration of 0.23 µg/mL (1.5 µM) and establishes a likely lower bound for the therapeutic drug level, assuming minimal pharmacokinetic-pharmacodynamic hysteresis.13 Much higher L-carvone plasma concentrations exceeding 1 µg/mL in 1 sheep were also associated with sedation, suggesting a possible secondary use for this agent at higher doses.
The concentration of drug that could be considered safe for consumption in an animal product is directly proportional to the acceptable daily intake and inversely related to amount of edible product estimated to be consumed by an individual, also known as the food consumption factor.14 However, L-carvone is a natural component of many foods and fragrances and is generally recognized as safe in the US.15 The European Food Safety Authority found the highest aggregate human exposure to L-carvone to be 1.87 mg/kg/d without evidence of toxicological effects.16 Using this highest exposure, and assuming a human body weight of 60 kg and a food consumption factor of 0.3 for muscle meat,14 maximum L-carvone consumption would be 374 mg/d. Even if all meat were from the injection site area, this would require a person to eat more than 3.5 kg of mutton per day. If noninjected muscle, then this same person would need to consume more than 450 kg of mutton per day, for which any postprandial side effects would likely be unrelated to the L-carvone itself.
Although the current drug formulation has a pharmacokinetic profile appropriate for an IM analgesic, the local tissue injury caused by injection is not ideal. Myonecrosis could be directly from the L-carvone. However, tissue injury may be from the excipients used to increase drug solubility in water and plasma. Pharmaceutical formulations containing ethanol or propylene glycol can damage the tissues into which they are injected. This occurs from both a direct irritation and from creation of severely hyperosmolar solutions.17,18 If these inactive ingredients are indeed the cause of muscle injury, then generally-recognized-as-safe formulations that decrease or eliminate ethanol and/or propylene glycol should be explored.19 Alternatively, it may be possible to administer this or a different L-carvone formulation by different parenteral or enteral routes to achieve good drug absorption without risk of tissue injury.
There are a couple of important limitations to this study. First, the variance of pharmacokinetic model parameters was relatively high. This was due to the small sample size of 6 sheep as well as sparse sampling at late time points. The β-slope of the model was much shallower than anticipated when the experiments were designed, and additional samples between 8 and 24 hours should have been collected to better characterize this second phase and refine estimates of the volume and clearances of the peripheral compartment. Second, local inflammation caused by the L-carvone formulation could have altered local blood flow, local pH, and cell and tissue permeability to drug movement and thus affected absorption kinetics.20 A nonirritating water-soluble formulation with otherwise similar physical properties might therefore exhibit very different peak times and maximum plasma concentrations than observed in the present study.
In summary, analgesic doses of L-carvone formulated in ethanol-propylene glycol administered IM exhibit reasonably rapid absorption and initial distribution but with a very long terminal half-life. After 24 hours, concentrations of L-carvone in sheep tissues—except for the actual injection site itself—are detectable but low relative to exposure from other natural and processed foodstuffs. Although no systemic adverse effects were noted, the current L-carvone formulation causes myonecrosis, and a different preparation or administration route is needed prior to adoption of this agent for clinic use.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org.
Acknowledgments
The authors thank Paige Condy and Courtney Snell for technical assistance.
Disclosures
Dr. Brosnan is named as the inventor on a patent, assigned to the University of California, regarding methods of use of terpenoids (including L-carvone) as injectable analgesics.
No AI-assisted technologies were used in the generation of this manuscript.
Funding
This work was supported by the USDA National Institute of Food and Agriculture (2022–67015-37082).
ORCID
R. J. Brosnan https://orcid.org/0000-0002-0508-6363
References
- 1.↑
Steffey EP, Brosnan RJ, Mama KR. Inhalation anesthetics. In: Lamont LA, Grimm KA, Robertson SA, et al, eds. Veterinary Anesthesia and Analgesia. 6th ed. Wiley-Blackwell; 2024:489–525.
- 2.
Brosnan RJ, Pham TL. Anesthetic-sensitive ion channel modulation is associated with a molar water solubility cut-off. BMC Pharmacol Toxicol. 2018;19(1):57. doi:10.1186/s40360-018-0244-z
- 3.↑
Brosnan RJ, Pham TL. Hydrocarbon molar water solubility predicts NMDA vs. GABAA receptor modulation. BMC Pharmacol Toxicol. 2014;15:62.
- 4.↑
Brosnan RJ, Ramos K, Aguiar AJA, Cenani A, Knych HK. Anesthetic pharmacology of the mint extracts l-carvone and methyl salicylate. Pharmacology. 2022;107(3–4):167–178. doi:10.1159/000520762
- 5.↑
Brosnan RJ, Cenani A, Costa LR, Condy P, Snell C. Analgesic effect of the mint terpenoid l-carvone in sheep. Vet Anaesth Analg. 2023;50(5):459–465. doi:10.1016/j.vaa.2023.06.004
- 6.↑
Ilkiw JE, Benthuysen JA, McNeal D. Comparative study of the pharmacokinetics of alfentanil in rabbits, sheep, and dogs. Am J Vet Res. 1991;52(4):581–584. doi:10.2460/ajvr.1991.52.04.581
- 7.↑
Hakomaki H, Kokki H, Lehtonen M, et al. Pharmacokinetics of buprenorphine in pregnant sheep after intravenous injection. Pharmacol Res Perspect. 2021;9(2):e00726.
- 8.↑
Sloan PA, Mather LE, McLean CF, et al. Physiological disposition of i.v. morphine in sheep. Br J Anaesth. 1991;67(4):378–386. doi:10.1093/bja/67.4.378
- 9.↑
Kinnunen M, Kokki H, Hautajarvi H, et al. Oxycodone pharmacokinetics and fetal exposure after intravenous or epidural administration to the ewe. Acta Obstet Gynecol Scand. 2018;97(10):1200–1205. doi:10.1111/aogs.13378
- 10.↑
Bortolami E, Della Rocca G, Di Salvo A, et al. Pharmacokinetics and antinociceptive effects of tramadol and its metabolite O-desmethyltramadol following intravenous administration in sheep. Vet J. 2015;205(3):404–409. doi:10.1016/j.tvjl.2015.04.011
- 11.↑
Griffin S, Wyllie SG, Markham J. Determination of octanol-water partition coefficient for terpenoids using reversed-phase high-performance liquid chromatography. J Chromatogr A. 1999;864(2):221–228. doi:10.1016/S0021-9673(99)01009-2
- 12.↑
Shafer SL, Varvel JR. Pharmacokinetics, pharmacodynamics, and rational opioid selection. Anesthesiology. 1991;74(1):53–63. doi:10.1097/00000542-199101000-00010
- 13.↑
Louizos C, Yanez JA, Forrest ML, Davies NM. Understanding the hysteresis loop conundrum in pharmacokinetic/pharmacodynamic relationships. J Pharm Pharm Sci. 2014;17(1):34–91.
- 14.↑
Riviere JE. Tissue residues and withdrawal times. Comparative Pharmacokinetics. Iowa State Press; 1999:308–318.
- 15.↑
Substances Generally Recognized as Safe In: DHHS, ed. 21CFR18260: Synthetic Flavoring Substances and Adjuvants. United States Food and Drug Administration; 1989.
- 16.↑
European Food Safety Authority. Scientific opinion on the safety assessment of carvone, considering all sources of exposure. EFSA J. 2014;12(7):3806.
- 17.↑
Rasmussen F. Tissue damage at the injection site after intramuscular injection of drugs. Vet Sci Commun. 1978;2:173–182. doi:10.1007/BF02291447
- 18.↑
Doenicke A, Nebauer AE, Hoernecke R, Mayer M, Roizen MF. Osmolalities of propylene glycol-containing drug formulations for parenteral use. Should propylene glycol be used as a solvent? Anesth Analg. 1992;75(3):431–435. doi:10.1213/00000539-199209000-00020
- 19.↑
Boquet MP, Wagner DR. Injectable formulations of poorly water-soluble drugs. In: Williams RO, Watts AB, Miller DA, eds. Formulating Poorly Water Soluble Drugs. Springer; 2012:209–242.
- 20.↑
Martinez MN, Greene J, Kenna L, et al. The impact of infection and inflammation on drug metabolism, active transport, and systemic drug concentrations in veterinary species. Drug Metab Dispos. 2020;48(8):631–644. doi:10.1124/dmd.120.090704