Sheep are an important livestock species commonly treated in veterinary clinical practice. They are also commonly used for research, especially surgical and fetal-development research.1–3 Administration of opioids to pregnant females is of particular concern because the drugs may cross the placenta and affect the fetus or fetuses.4 For example, human neonates that were exposed to opioids during gestation may be born with signs of opioid withdrawal syndrome.5 In clinical and research settings, early recognition and appropriate management of pain are crucial to ensure animal welfare and prevent sequelae associated with untreated pain. Scoring systems to identify pain and the need for analgesia have been developed and validated for various species. Although a pain scoring system has yet to be validated for sheep, evidence suggests that multidimensional pain scoring systems are superior to a simple visual analog scale.4,6,7
Combinations of NSAIDs and opioids are commonly administered for perioperative analgesia, and buprenorphine and fentanyl are 2 commonly used opioids. Buprenorphine (a partial μ-receptor agonist with a long duration of action [eg, up to 10 hours in humans8]) and fentanyl (a short-acting full μ-opioid receptor agonist) are approximately 30 and 90 times as potent as morphine, respectively.9,10 Buprenorphine is most commonly administered by injection. Fentanyl is often administered by application of a TFP, which is designed to release the drug at a sustained rate for a prolonged period, thereby avoiding the extreme fluctuations in plasma drug concentrations and analgesia commonly observed when the drug is injected intermittently.
Extrapolation of drug doses and dosing intervals from other species to sheep can result in inadequate analgesia owing to differences in drug disposition and routes of administration. Administration of opioids to pregnant sheep is of particular concern because of the potential effect those drugs might have on the fetuses. The disposition and effects of buprenorphine on fetuses in utero are not completely understood, even in human medicine.5 Scientific literature regarding the application and efficacy of TFPs in pregnant sheep is scarce. Although the pharmacokinetics of the TFP has been described for sheep,11–16 factors that could potentially affect the pharmacokinetics, such as location of patch application, body composition, and pregnancy status, varied widely among those studies. Therefore, further research to compare the efficacy of a TFP with other common analgesic protocols used for sheep, such as serial injections of buprenorphine, is warranted.12,14,15 Additionally, the maternal and fetal effects of both buprenorphine and fentanyl have yet to be characterized for pregnant sheep.
The objectives of the study reported here were to compare the analgesic efficacy and fetal effects of perianesthetic application of a TFP between the dorsal borders of the scapulae with IM injection of buprenorphine in pregnant ewes. Of primary interest was whether the measured plasma drug concentrations were associated with analgesia, which was assessed by use of standardized pain and sedation scoring systems, and evaluation of intraoperative physiologic variables and inhalant-sparing effects. The effects of each drug on fetuses in utero were also characterized. We hypothesized that plasma drug concentrations adequate for analgesia would be achieved following administration of each drug protocol; that the analgesia induced by application of a TFP between the dorsal borders of the scapulae would be equivalent to that induced by IM injection of buprenorphine as determined on the basis of pain scores; that the plasma drug concentrations would be correlated with the pain scores, thus validating the pain scoring system used for sheep; and that both buprenorphine and fentanyl would cross the placenta and be detected in fetuses.
Materials and Methods
Animals
All study procedures were reviewed and approved by the Texas A&M University Institutional Animal Care and Use Committee (No. 2016-0248), and all sheep enrolled in the study were owned by the university. Twelve healthy pregnant Suffolk-crossbred ewes (age range, 2 to 4 years) underwent a surgical procedure for fetal catheterization and placement of a catheter in utero for collection of amniotic fluid samples. Each ewe was confirmed to be pregnant with a single fetus at 113 to 117 days of gestation on the basis of the observed breeding date and transabdominal ultrasonographic examination. Upon arrival at the research facility, each ewe was individually weighed and underwent a physical examination. Ewes were individually housed in indoor pens (2.5 × 2.5 m) within sight of the other study subjects.
Study design
For all ewes, anesthesia induction was designated as time 0 (T0). Each ewe was randomly assigned by means of an online automated randomization toola to receive buprenorphineb (0.01 mg/kg, IM in the neck, q 8 h for 48 hours beginning 1 hour prior to anesthesia induction [T–1]; n = 6) or fentanyl (a combination of TFPsc sufficient to deliver a dose of 2 μg of fentanyl/kg/h applied beginning 24 hours before anesthesia induction [T–24]; 6) for analgesia. To ensure that the investigator responsible for evaluating analgesia remained unaware of (was blinded to) the treatment group assignment for each ewe, a 10 × 10-cm adhesive bandaged (placebo patch) was applied between the dorsal borders of the scapulae of each buprenorphine-treated ewe at T–24; an identical bandage was placed over the TFP applied to each fentanyl-treated ewe. For each ewe prior to patch application, the area between the dorsal borders of the scapulae was clipped to remove wool and cleaned with isopropyl alcohol to remove dirt, excess lanolin, and other debris that could interfere with patch adherence and drug absorption (fentanyl-treated ewes) as described.4,13 The area was allowed to dry completely before the patch was applied. Self-adhering bandaging tapee and tubular nettingf were applied over the bandage patch for additional security to prevent the ewe from rubbing it off or potentially consuming it. Because the investigator responsible for administering the buprenorphine injections was not involved with assessing ewes for analgesia and was not present when the analgesia assessments were performed, the fentanyl-treated ewes did not receive placebo injections of saline (0.9% NaCl) solution.
Surgical instrumentation
Food but not water was withheld from each ewe for 24 hours before anesthesia induction for fetal catheterization. For each ewe, a 16-gauge, 7.6-cm-long catheterg was aseptically placed in a jugular vein and used for anesthesia induction and administration of IV fluids. Anesthesia was induced with a combination of midazolamh (0.35 mg/kg, IV) and ketamine hydrochloridei (4 mg/kg, IV). Following orotracheal intubation with an appropriately sized, cuffed endotracheal tube, anesthesia was maintained with isoflurane,j which was delivered in oxygen by use of a small animal anesthesia machine.k The ewe was positioned in dorsal recumbency, and administration of lactated Ringer solution (5 mg/kg/h, IV) was commenced.
For each ewe, standard surgical techniques were used to catheterize a cranial tibial artery of the fetus as described.16 Briefly, a ventral midline laparotomy was performed, the uterus was externalized and incised, and a fetal hind limb was exteriorized. An incision was made over the craniolateral aspect of the fetal hind limb. A catheter (polyvinyl chloride tubing; inner diameter, 0.08 cm; outer diameter, 0.13 cm; fetal catheter) was advanced from the cranial tibial artery into the abdominal aorta to the level of the diaphragm. Then, a custom amniotic catheter was made from tubing as described.17 Briefly, 1 end of the tubing was sutured to the inside of a hollow plastic golf ball with holes to prevent collapse of fetal membranes over the tubing and occlusion of the catheter.17 The catheter tubing was sutured to the skin over the tarsal region of the fetus so that the golf ball was floating inside the amniotic cavity.17 The fetal hind limb was returned to the uterus, and the uterus was closed in a routine manner around the fetal and amniotic catheters. In the right flank region of the ewe, a trocar was passed through the skin, subcutaneous tissues, musculature, and peritoneum into the abdominal cavity. The fetal and amniotic catheters were passed through the trocar, and the trocar was removed. The free ends of the fetal and amniotic catheters were placed in a cloth pouch that was sutured to the skin of the right flank region of the ewe near the catheter exit site. The ventral midline incision was closed in a routine manner.
All ewes received flunixin megluminel (1 mg/kg, IV) and tulathromycinm (2.5 mg/kg, SC) immediately before surgery. Following completion of all surgical procedures, administration of isoflurane in oxygen was discontinued, and the ewe was repositioned in sternal recumbency to recover from anesthesia. Each ewe was extubated when it was able to lift and maintain its head in an upright position, swallow, and fully control its tongue.
Anesthetic monitoring
An investigator who was blinded to the treatment group assignment of each ewe was responsible for maintaining anesthesia and monitoring the HR, hemoglobin oxygen saturation as measured by pulse oximetry, and esophageal temperaturen during the surgical procedure for all ewes. Blood pressure was measured noninvasively with an inflatable cuff,o the width of which was approximately 40% of the circumference of the metacarpal region. While a ewe was positioned in dorsal recumbency, the level of the shoulder joint was used as the reference point for the level of the right atrium.18 The blood pressure cuff was placed around the metacarpal region over the metacarpal artery and positioned at the level of the shoulder joint (ie, right atrium). The HR and MAP were monitored with an automated monitor,o which was calibrated prior to study initiation in accordance with the manufacturer's recommendation to verify readings at 0, 50, 100, 150, and 200 mm Hg. The respiratory rate, ETiso, and ETco2 were measured with a multiparameter monitorp that was calibrated before each surgical procedure with a calibration gas canister containing 5% carbon dioxide, 55% oxygen, 33% nitrous oxide, and 2% desflurane in accordance with the manufacturer's recommendation. The ETiso was manipulated to maintain an appropriate plane of anesthesia for the surgical procedure, which was determined on the basis of ocular position and an absence of palpebral reflexes and purposeful movement in response to surgical stimulation. Rescue analgesia (morphine [0.1 mg/kg, IV]) was planned if an increase in the delivered isoflurane concentration to twice its minimum alveolar concentration was insufficient to maintain an adequate plane of anesthesia. The time was recorded at anesthesia induction with ketamine and midazolam, orotracheal intubation, initiation and completion of surgery, extubation, and the point when the ewe was able to maintain a standing position following discontinuation of anesthesia.
Analgesic assessment and sample collection for ewes
For each ewe, the HR, respiratory rate, and rectal temperature were measured, and behavioral indicators of pain19 (Appendix 1) and sedation6 (Appendix 2) were assessed by use of composite scoring systems immediately prior to fentanyl or placebo patch application (T–24) and immediately before (T0) and at 4 (T4), 6 (T6), 8 (T8), 12 (T12), 24 (T24), and 48 (T48) hours after anesthesia induction. Pain and sedation assessments were performed by observers who were blinded to the treatment group assignment of each ewe. The pain score included assessment of comfort, movement, appetite, posture, and response to palpation of the surgical incision. At each assessment time, each of those 5 criteria was scored on a scale of 0 (normal) to 3 (severely abnormal behavior). The scores for each pain criterion were summed; thus, 15 was the maximum possible pain score at any assessment time. Rescue analgesia (morphine [0.1 mg/kg, IV]) was administered to any ewe with a pain score ≥ 8. Sedation was scored on the basis of the posture and demeanor of the ewe on a scale of 1 (alert) to 10 (very sedate). Water and feed consumption were also monitored and recorded.
From each ewe, a blood sample (5 mL) was collected by jugular venipuncture into a blood collection tube containing lithium heparin as an anticoagulant (lithium heparin tube) immediately before patch application (T–24), at T–1 (immediately before buprenorphine administration to buprenorphine-treated ewes), and at T0, 2 hours after anesthesia induction (T2), T4, T6, T8, T12, T24, and 36 hours (T36), 48 hours (T48), and 72 hours (T72) after anesthesia induction. Blood samples were also collected at 144 (T144) and 288 (T288) hours after anesthesia induction for some ewes. All blood samples were centrifuged immediately after collection. The plasma from each sample was harvested, placed in a cryovial, and stored frozen at −80°C until analyzed.
Fetal sample collection
Fetal HR and blood pressure were monitored, and blood and amniotic fluid samples were collected at T4, T8, T12, T24, T36, T48, and T72. Blood and amniotic fluid samples were also collected at T144 and T288 from some fetuses. At each sample acquisition time, the free ends of the fetal and amniotic catheters were cleaned with iodine scrub and alcohol, and waste blood or amniotic fluid was removed from the respective catheters. For each fetus and each sample acquisition time, an arterial blood sample (1.0 mL) was collected into a lithium heparin tube, and amniotic fluid (1.5 mL) was collected into a sterile polystyrene tube without any additives. An additional 0.5 mL of arterial blood was collected and immediately underwent arterial blood gas analysis.q The blood samples collected in the lithium heparin tubes were centrifuged immediately after collection. For each sample, the plasma was harvested, placed in a cryovial, and stored frozen at −80°C until analyzed. The fetal catheter was then connected to an automated data acquisition systemr that measured fetal HR and blood pressure. After each sample collection was complete, the fetal (arterial) catheter was flushed with heparinized saline solution and the amniotic catheter was flushed with saline solution.
Sample analysis
Liquid chromatography–mass spectrometry was used to determine buprenorphine and fentanyl concentrations in plasma and amniotic fluid samples. Mepivacaine was used as the internal standard for both buprenorphine and fentanyl assays. Information regarding assay methods and performance is detailed elsewhere (Supplementary Appendix S1, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.81.7.581).
Pharmacokinetic analysis
Noncompartmental analysess were used to estimate plasma pharmacokinetic parameters for fentanyl and buprenorphine in ewes and fetuses. Because multiple doses of buprenorphine were administered, the peak plasma concentration of the drug following injection of the first dose was recorded as the observed Cmax, and calculation of the elimination rate constant was based on data obtained after the last dose of the drug was administered.
Statistical analysis
All statistical analyses were performed with commercial statistical software.t Descriptive data were generated. The mean ± SD was used to summarize the results for all physiologic and intraoperative variables, pain and sedation scores, plasma and amniotic fluid drug concentrations, and pharmacokinetic parameters. Data distributions for continuous variables were assessed for normality by use of the Shapiro-Wilk test. Owing to the small number of animals, a 2-way ANOVA for repeated measures was used to evaluate outcomes of interest because it is robust to violations of the normality assumption. Fixed effects evaluated in each model included treatment (buprenorphine or fentanyl) and assessment time (time). The Sidak multiple comparison test was used when post hoc comparisons were necessary, and Sidak-adjusted values of P < 0.05 were considered significant. The Pearson correlation coefficient (r) was calculated to evaluate associations between plasma drug concentrations and specific physiologic variables.
Results
Sheep
All 12 ewes completed the study, and none of the ewes required rescue analgesia. No intraoperative complications were encountered.
Analgesic indices for ewes
The mean ± SD HR for the entire observation period for the buprenorphine-treated ewes (129 ± 22.2 beats/min) was significantly (P = 0.04) greater than that for the fentanyl-treated ewes (101.4 ± 22.4 beats / min). Mean feed consumption did not differ significantly between the buprenorphine-treated and fentanyl-treated ewes for the 2-day period before surgery or the 2-day period after surgery. However, mean feed consumption for the fentanyl-treated ewes was significantly (P = 0.04) greater than that for the buprenorphine-treated ewes on the day of surgery. Also, the mean feed consumption on the day of surgery and for the 2-day period after surgery was significantly less than that for the 2-day period before surgery for both the buprenorphine-treated and fentanyl-treated ewes (Figure 1).
Mean ± SD percentage of feed consumption for 12 healthy pregnant ewes for the 2-day period before (preoperative period), day of (surgery day), and 2-day period after (postoperative period) a surgical procedure was performed in which the fetus was instrumented with an arterial catheter and a catheter was placed in utero for collection of amniotic fluid samples. Ewes were randomly assigned to receive buprenorphine (0.01 mg/kg, IM in the neck, q 8 h for 48 hours beginning 1 hour prior to anesthesia induction [T–1]; black bars; n = 6) or fentanyl (a combination of TFPs sufficient to deliver a dose of 2 μg of fentanyl/kg/h applied between the dorsal borders of the scapulae 24 hours prior to anesthesia induction [T–24]; white bars; 6) for perioperative analgesia. Each ewe was confirmed pregnant with a single fetus at 113 to 117 days of gestation prior to study enrollment. *Within a period, the mean value for each group differs significantly (P < 0.05) from that for the preoperative period. †Within a period, the mean value differs significantly (P < 0.05) between the 2 treatment groups.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Mean ± SD percentage of feed consumption for 12 healthy pregnant ewes for the 2-day period before (preoperative period), day of (surgery day), and 2-day period after (postoperative period) a surgical procedure was performed in which the fetus was instrumented with an arterial catheter and a catheter was placed in utero for collection of amniotic fluid samples. Ewes were randomly assigned to receive buprenorphine (0.01 mg/kg, IM in the neck, q 8 h for 48 hours beginning 1 hour prior to anesthesia induction [T–1]; black bars; n = 6) or fentanyl (a combination of TFPs sufficient to deliver a dose of 2 μg of fentanyl/kg/h applied between the dorsal borders of the scapulae 24 hours prior to anesthesia induction [T–24]; white bars; 6) for perioperative analgesia. Each ewe was confirmed pregnant with a single fetus at 113 to 117 days of gestation prior to study enrollment. *Within a period, the mean value for each group differs significantly (P < 0.05) from that for the preoperative period. †Within a period, the mean value differs significantly (P < 0.05) between the 2 treatment groups.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Mean ± SD percentage of feed consumption for 12 healthy pregnant ewes for the 2-day period before (preoperative period), day of (surgery day), and 2-day period after (postoperative period) a surgical procedure was performed in which the fetus was instrumented with an arterial catheter and a catheter was placed in utero for collection of amniotic fluid samples. Ewes were randomly assigned to receive buprenorphine (0.01 mg/kg, IM in the neck, q 8 h for 48 hours beginning 1 hour prior to anesthesia induction [T–1]; black bars; n = 6) or fentanyl (a combination of TFPs sufficient to deliver a dose of 2 μg of fentanyl/kg/h applied between the dorsal borders of the scapulae 24 hours prior to anesthesia induction [T–24]; white bars; 6) for perioperative analgesia. Each ewe was confirmed pregnant with a single fetus at 113 to 117 days of gestation prior to study enrollment. *Within a period, the mean value for each group differs significantly (P < 0.05) from that for the preoperative period. †Within a period, the mean value differs significantly (P < 0.05) between the 2 treatment groups.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
The mean ± SD total pain score over the entire observation period was numerically greater for the buprenorphine-treated ewes (2.3 ± 1.3) than for the fentanyl-treated ewes (1.6 ± 0.4) but did not differ significantly (P = 0.06) between the 2 groups. For the buprenorphine-treated ewes, the mean total pain score was significantly greater than that at baseline (T–24) at T4, T12, T24, T36, and T48. For the fentanyl-treated ewes, the mean total pain score did not differ significantly from baseline at any time during the observation period (Figure 2). The mean total pain score for the buprenorphine-treated ewes (4.0 ± 3.2) was significantly (P = 0.01) greater than that for the fentanyl-treated ewes (1.0 ± 1.3) at T4 but did not differ significantly between the 2 groups at any other time during the observation period.
Mean ± SD pain scores over time for the buprenorphine-treated (black bars) and fentanyl-treated (gray bars) ewes described in Figure 1. The pain score included assessment of comfort, movement, appetite, posture, and response to palpation of the surgical incision and was recorded immediately prior to fentanyl or placebo (buprenorphine-treated ewes) patch application (T–24) and immediately before (T0) and at 4 (T4), 6 (T6), 8 (T8), 12 (T12), 24 (T24), and 48 (T48) hours after anesthesia induction. At each assessment time, each of the 5 criteria was scored on a scale of 0 (normal) to 3 (severely abnormal behavior). The scores for each pain criterion were summed; thus, 15 was the maximum possible pain score at any assessment time. *Within a time, mean value for the buprenorphine-treated ewes differs significantly (P < 0.05) from that at baseline (T–24). †Within a time, the mean value differs significantly (P < 0.05) between the 2 treatment groups. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Mean ± SD pain scores over time for the buprenorphine-treated (black bars) and fentanyl-treated (gray bars) ewes described in Figure 1. The pain score included assessment of comfort, movement, appetite, posture, and response to palpation of the surgical incision and was recorded immediately prior to fentanyl or placebo (buprenorphine-treated ewes) patch application (T–24) and immediately before (T0) and at 4 (T4), 6 (T6), 8 (T8), 12 (T12), 24 (T24), and 48 (T48) hours after anesthesia induction. At each assessment time, each of the 5 criteria was scored on a scale of 0 (normal) to 3 (severely abnormal behavior). The scores for each pain criterion were summed; thus, 15 was the maximum possible pain score at any assessment time. *Within a time, mean value for the buprenorphine-treated ewes differs significantly (P < 0.05) from that at baseline (T–24). †Within a time, the mean value differs significantly (P < 0.05) between the 2 treatment groups. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Mean ± SD pain scores over time for the buprenorphine-treated (black bars) and fentanyl-treated (gray bars) ewes described in Figure 1. The pain score included assessment of comfort, movement, appetite, posture, and response to palpation of the surgical incision and was recorded immediately prior to fentanyl or placebo (buprenorphine-treated ewes) patch application (T–24) and immediately before (T0) and at 4 (T4), 6 (T6), 8 (T8), 12 (T12), 24 (T24), and 48 (T48) hours after anesthesia induction. At each assessment time, each of the 5 criteria was scored on a scale of 0 (normal) to 3 (severely abnormal behavior). The scores for each pain criterion were summed; thus, 15 was the maximum possible pain score at any assessment time. *Within a time, mean value for the buprenorphine-treated ewes differs significantly (P < 0.05) from that at baseline (T–24). †Within a time, the mean value differs significantly (P < 0.05) between the 2 treatment groups. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
The mean ± SD sedation score over the entire observation period did not differ significantly between the buprenorphine-treated (1.0 ± 1.0) and fentanyl-treated (0.57 ± 0.5) ewes. Also, the mean sedation score did not change significantly after analgesic administration for either treatment group.
Other ewe variables
The mean HR (P < 0.001), MAP (P < 0.001), ETco2 (P = 0.003), and ETiso (P < 0.001) for the buprenorphine-treated ewes were significantly greater than those for the fentanyl-treated ewes during anesthesia and surgery (Table 1). However, the mean values for all variables were within acceptable ranges for anesthetized sheep.20
Mean ± SD values for select cardiorespiratory variables during anesthesia for 12 healthy anesthetized pregnant ewes that were randomly assigned to receive buprenorphine (0.01 mg/kg, IM in the neck, q 8 h for 48 hours beginning 1 hour prior to anesthesia induction [T–1]; n = 6) or fentanyl (a combination of TFPs sufficient to deliver a dose of 2 μg of fentanyl/kg/h applied between the dorsal borders of the scapulae 24 hours prior to anesthesia induction [T–24]; 6) as perioperative analgesia for a surgical procedure during which the fetus was instrumented with an arterial catheter and a catheter was placed in utero for collection of amniotic fluid samples.
| Variable | Buprenorphine-treated ewes | Fentanyl-treated ewes | P value |
|---|---|---|---|
| MAP (mm Hg) | 85.0 ± 4.79 | 72.7 ± 8.28 | < 0.001 |
| HR (beats/min) | 113 ± 5.37 | 91.3 ± 4.75 | < 0.001 |
| ETco2 (%) | 43.5 ± 1.68 | 40.3 ± 2.98 | 0.003 |
| Spo2 (%) | 93.8 ± 1.61 | 94.5 ± 1.39 | 0.300 |
| ETiso (%) | 1.42 ± 0.28 | 1.16 ± 0.23 | < 0.001 |
Each ewe was confirmed pregnant with a single fetus at 113 to 117 days of gestation prior to study enrollment. Values of P < 0.05 were considered significant.
Spo2 = Hemoglobin oxygen saturation as measured by pulse oximetry.
Treatment group did not have a significant effect on surgery duration. The mean ± SD surgery duration was 96.5 ± 4.2 minutes for all 12 ewes. The mean durations between discontinuation of isoflurane administration and extubation and between extubation and the point at which the ewe attained a standing position were significantly longer for buprenorphine-treated ewes than for fentanyl-treated ewes by a mean ± SD of 24.33 ± 10.62 minutes (P = 0.03) and 30.60 ± 11.63 minutes (P = 0.01), respectively (Figure 3). During recovery from anesthesia, 4 of the 6 fentanyl-treated ewes exhibited excitatory behavior that consisted of vocalizing, paddling, head jerking, and multiple unsuccessful attempts to stand. Light manual restraint was sufficient to prevent the ewes from injuring themselves until the excitatory phase passed.
Mean duration between anesthesia induction and orotracheal intubation (black), between orotracheal intubation and initiation of surgery (dark pink), between initiation and completion of surgery (teal), between completion of surgery and extubation (dark purple), and between extubation and the animal attaining a standing position (light purple) for the buprenorphine-treated and fentanyl-treated ewes described in Figure 1. *Bracketed duration differs significantly (P < 0.05) from that for the buprenorphine-treated ewes. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Mean duration between anesthesia induction and orotracheal intubation (black), between orotracheal intubation and initiation of surgery (dark pink), between initiation and completion of surgery (teal), between completion of surgery and extubation (dark purple), and between extubation and the animal attaining a standing position (light purple) for the buprenorphine-treated and fentanyl-treated ewes described in Figure 1. *Bracketed duration differs significantly (P < 0.05) from that for the buprenorphine-treated ewes. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Mean duration between anesthesia induction and orotracheal intubation (black), between orotracheal intubation and initiation of surgery (dark pink), between initiation and completion of surgery (teal), between completion of surgery and extubation (dark purple), and between extubation and the animal attaining a standing position (light purple) for the buprenorphine-treated and fentanyl-treated ewes described in Figure 1. *Bracketed duration differs significantly (P < 0.05) from that for the buprenorphine-treated ewes. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Fetal variables
The mean HR (P = 0.79), MAP (P = 0.49), and arterial blood pH (P = 0.18), Pco2 (P = 0.21), and Po2 (P = 0.70) during the entire observation period did not differ significantly between fetuses of buprenorphine-treated ewes and fetuses of fentanyl-treated ewes (Table 2). For all 12 fetuses, all measured variables were indicative of viability at all data acquisition times.17 There was a significant moderate negative correlation between fetal HR and fetal plasma drug concentration for both buprenorphine (P = 0.021) and fentanyl (P = 0.019; Supplementary Figure S1, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.81.7.581).
Mean ± SD values for select cardiorespiratory variables over the entire observation period for the fetuses of the buprenorphine-treated and fentanyl-treated ewes described in Table 1.
| Variable | Fetuses of buprenorphine-treated ewes | Fetuses of fentanyl-treated ewes | P value |
|---|---|---|---|
| MAP (mm Hg) | 44.9 ± 10.56 | 41.43 ± 7.16 | 0.49 |
| HR (beats/min) | 171.1 ± 16.97 | 160.7 ± 9.53 | 0.79 |
| pH | 7.37 ± 0.07 | 7.35 ± 0.14 | 0.18 |
| Pco2 (mm Hg) | 45.61 ± 8.54 | 46.17 ± 8.8 | 0.21 |
| Po2 (mm Hg) | 12.87 ± 3.86 | 12.82 ± 4.72 | 0.70 |
For each fetus, HR and MAP were measured and arterial blood samples were collected for arterial blood-gas evaluation at 4 (T4), 8 (T8), 12 (T12), 24 (T24), 36 (T36), 48 (T48), and 72 (T72) hours after anesthesia was induced in the ewe.
See Table 1 for remainder of key.
Pharmacokinetic analysis
For the buprenorphine-treated ewes, plasma buprenorphine concentrations varied greatly among ewes and over time. The plasma buprenorphine concentration over time was plotted for each buprenorphine-treated ewe in conjunction with the mean ± SD plasma buprenorphine concentration over time for the fetuses of those ewes (Figure 4). The lowest plasma buprenorphine concentration measured in a ewe was 0.15 ng/mL. The mean plasma buprenorphine concentration for the ewes was significantly greater than that for the fetuses at T4 and T8 but did not differ significantly between the ewes and fetuses at any other time. Within individual ewes, the ratio of fetal plasma buprenorphine concentration to ewe plasma buprenorphine concentration ranged from 0.05 to 0.15 throughout the observation period. Pharmacokinetic parameters for buprenorphine were summarized for both ewes and fetuses (Table 3). Pharmacokinetic parameters for buprenorphine in individual ewes are provided elsewhere (Supplementary Table S1, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.81.7.581). The plasma buprenorphine concentration remained fairly stable over time for 3 of the 6 fetuses, and the λz could not be calculated for those 3 fetuses. Therefore, values for λz and other pharmacokinetic parameters that are dependent on λz (ie, t1/2λ, AUC0–∞, and AUC%extrap) represent the mean for only 3 fetuses.
Plasma buprenorphine concentration over time for the 6 buprenorphine-treated ewes described in Figure 1 (solid lines) and mean ± SD plasma buprenorphine concentration over time for the fetuses of those ewes (dashed line). A blood sample for determination of plasma buprenorphine concentration was obtained at T–24, T–1, T0, 2 hours (T2), T4, T6, T8, T12, T24, 36 hours (T36), T48, and 72 hours (T72) after anesthesia induction from each ewe and at T4, T8, T12, T24, T36, T48, and T72 from each fetus. Notice the plasma buprenorphine concentration varied greatly within individual ewes and among ewes. The mean plasma buprenorphine concentration for the ewes was significantly (P < 0.05) greater than that for the fetuses at T4 and T8. See Figures 1 and 2 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Plasma buprenorphine concentration over time for the 6 buprenorphine-treated ewes described in Figure 1 (solid lines) and mean ± SD plasma buprenorphine concentration over time for the fetuses of those ewes (dashed line). A blood sample for determination of plasma buprenorphine concentration was obtained at T–24, T–1, T0, 2 hours (T2), T4, T6, T8, T12, T24, 36 hours (T36), T48, and 72 hours (T72) after anesthesia induction from each ewe and at T4, T8, T12, T24, T36, T48, and T72 from each fetus. Notice the plasma buprenorphine concentration varied greatly within individual ewes and among ewes. The mean plasma buprenorphine concentration for the ewes was significantly (P < 0.05) greater than that for the fetuses at T4 and T8. See Figures 1 and 2 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Plasma buprenorphine concentration over time for the 6 buprenorphine-treated ewes described in Figure 1 (solid lines) and mean ± SD plasma buprenorphine concentration over time for the fetuses of those ewes (dashed line). A blood sample for determination of plasma buprenorphine concentration was obtained at T–24, T–1, T0, 2 hours (T2), T4, T6, T8, T12, T24, 36 hours (T36), T48, and 72 hours (T72) after anesthesia induction from each ewe and at T4, T8, T12, T24, T36, T48, and T72 from each fetus. Notice the plasma buprenorphine concentration varied greatly within individual ewes and among ewes. The mean plasma buprenorphine concentration for the ewes was significantly (P < 0.05) greater than that for the fetuses at T4 and T8. See Figures 1 and 2 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Mean ± SD values for select pharmacokinetic parameters for buprenorphine and fentanyl determined for the buprenorphine-treated and fentanyl-treated ewes described in Table 1 and the fetuses of those ewes.
| Buprenorphine | Fentanyl | |||
|---|---|---|---|---|
| Parameter | Ewes | Fetuses | Ewes | Fetuses |
| tmax (h) | 5 ± 3 | 6 ± 2 | 17 ± 22 | 13 ± 11 |
| Cmax (ng/mL) | 5.1 ± 6.5 | 0.1 ± 0.0 | 3.8 ± 2.3 | 0.6 ± 0.2 |
| λz (h–1) | 0.014 ± 0.008 | 0.005 ± 0.003* | † | ‡ |
| t1/2λz (h) | 65.8 ± 33.5 | 197.3 ± 86.6* | 0.033 ± 0.011§ | 26.8 ± 15.8 |
| AUC0–last (h·ng/mL) | 148.3 ± 122.9 | 9.2 ± 5.3 | 160.7 ± 102.0 | 21.0 ± 11.5 |
| AUC0–∞ (h·ng/mL) | 186.0 ± 114.6 | 28.1 ± 6.3* | 235.1 ± 58.2§ | 34.2 ± 11.4 |
| AUC%extrap (%) | 28.6 ± 24.1 | 49.6 ± 15.3* | 7.6 ± 6.4§ | 22.6 ± 21.1 |
A blood sample (5 mL) was collected at T–24 and T–1; immediately before anesthesia induction (T0) and at 2 hours (T2), T4, 6 hours (T6), T8, T12, T24, T36, T48, and T72 after anesthesia induction from each ewe; and at T4, T8, T12, T24, T36, T48, and T72 from each fetus. Blood samples were also collected at 144 (T144) and 288 (T288) hours after anesthesia induction from some ewes and their fetuses. Liquid chromatography–mass spectrometry was used to determine the plasma buprenorphine or fentanyl concentration in all samples. Pharmacokinetic parameters were determined by noncompartmental analyses.
The λz could not be calculated for 3 of the 6 fetuses of buprenorphine-treated ewes; therefore, the values for λz and other parameters dependent on λz (ie, t1/2λz, AUC0–∞, and AUC%extrap) represent the mean ± SD for only 3 fetuses.
Mean ± SD not calculated for λz because the duration between drug administration and the first observed plasma fentanyl concentration (tlag) was 0 hours for 5 of the 6 fentanyl-treated ewes and 26 hours for the remaining ewe.
Mean ± SD not calculated for λz because the tlag was 0, 32, and 28 hours for 3, 1, and 2 fetuses, respectively.
The λz could not be calculated for 2 of the 6 fentanyl-treated ewes; therefore, the values for parameters dependent on λz (ie, t1/2λz, AUC0–∞, and AUC%extrap) represent the mean ± SD for only 4 ewes.
AUC0–last = Area under the concentration-time curve from time 0 to the last measurable concentration.
One of the 6 fentanyl-treated ewes had an atypical plasma concentration–time curve; therefore, the data for that ewe and its fetus were reported separately (Supplementary Table S2, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.81.7.581) and were excluded from all mean calculations for the fentanyl-treated group. For the remaining 5 ewes and 5 fetuses in the fentanyl-treated group, the mean plasma fentanyl concentration over time was plotted (Figure 5). The lowest plasma fentanyl concentration measured in a ewe was 0.9 ng/mL. The mean plasma fentanyl concentration for ewes was significantly greater than that for fetuses at T4, T8, T12, T24, T36, and T48. Within individual ewes, the ratio of fetal plasma fentanyl concentration to ewe plasma fentanyl concentration ranged from 0.14 to 0.27 throughout the observation period. Pharmacokinetic parameters for fentanyl were summarized for both ewes and fetuses (Table 3).
Mean ± SD plasma fentanyl concentration over time for 5 of the 6 fentanyl-treated ewes (solid line) described in Figure 1 and the fetuses (dashed line) of those ewes. One fentanyl-treated ewe had an atypical plasma concentration–time curve; therefore, the data for that ewe and its fetus were excluded from all mean calculations for the fentanyl-treated group. *Within a time, the mean value for the ewes is significantly (P < 0.05) greater than that for the fetuses. See Figures 1 and 4 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Mean ± SD plasma fentanyl concentration over time for 5 of the 6 fentanyl-treated ewes (solid line) described in Figure 1 and the fetuses (dashed line) of those ewes. One fentanyl-treated ewe had an atypical plasma concentration–time curve; therefore, the data for that ewe and its fetus were excluded from all mean calculations for the fentanyl-treated group. *Within a time, the mean value for the ewes is significantly (P < 0.05) greater than that for the fetuses. See Figures 1 and 4 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Mean ± SD plasma fentanyl concentration over time for 5 of the 6 fentanyl-treated ewes (solid line) described in Figure 1 and the fetuses (dashed line) of those ewes. One fentanyl-treated ewe had an atypical plasma concentration–time curve; therefore, the data for that ewe and its fetus were excluded from all mean calculations for the fentanyl-treated group. *Within a time, the mean value for the ewes is significantly (P < 0.05) greater than that for the fetuses. See Figures 1 and 4 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Amniotic fluid drug concentrations
Successful acquisition of amniotic fluid samples varied among ewes. For each of the 12 ewes, there was at least 1 time point when an amniotic fluid sample could not be obtained. Buprenorphine was detected in amniotic fluid samples from all 6 buprenorphine-treated ewes. However, fentanyl was detected in amniotic fluid samples from only 3 of the 6 fentanyl-treated ewes. The amniotic fluid buprenorphine and fentanyl concentrations over time for individual ewes are available elsewhere (Supplementary Table S3, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.81.7.581). The mean amniotic fluid buprenorphine concentration over time for all 6 buprenorphine-treated ewes and the mean amniotic fluid fentanyl concentration over time for the 3 fentanyl-treated ewes with detectable concentrations of the drug were plotted (Figure 6).
Mean ± SD amniotic fluid buprenorphine (A) and fentanyl (B) concentrations for the buprenorphine-treated and fentanyl-treated ewes, respectively, described in Figure 1. An attempt was made to collect an amniotic fluid sample from each ewe at the same times fetal blood samples were collected; however, for each of the ewes, there was at least 1 time point when an amniotic fluid sample could not be successfully obtained. Additionally, fentanyl was detected in amniotic fluid samples from only 3 of the 6 fentanyl-treated ewes. Therefore, the number of ewes that contributed to the mean varies among times and does not exceed 3 for the fentanyl-treated group. Notice that the y-axis scale differs between the 2 panels. See Figures 1 and 4 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Mean ± SD amniotic fluid buprenorphine (A) and fentanyl (B) concentrations for the buprenorphine-treated and fentanyl-treated ewes, respectively, described in Figure 1. An attempt was made to collect an amniotic fluid sample from each ewe at the same times fetal blood samples were collected; however, for each of the ewes, there was at least 1 time point when an amniotic fluid sample could not be successfully obtained. Additionally, fentanyl was detected in amniotic fluid samples from only 3 of the 6 fentanyl-treated ewes. Therefore, the number of ewes that contributed to the mean varies among times and does not exceed 3 for the fentanyl-treated group. Notice that the y-axis scale differs between the 2 panels. See Figures 1 and 4 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Mean ± SD amniotic fluid buprenorphine (A) and fentanyl (B) concentrations for the buprenorphine-treated and fentanyl-treated ewes, respectively, described in Figure 1. An attempt was made to collect an amniotic fluid sample from each ewe at the same times fetal blood samples were collected; however, for each of the ewes, there was at least 1 time point when an amniotic fluid sample could not be successfully obtained. Additionally, fentanyl was detected in amniotic fluid samples from only 3 of the 6 fentanyl-treated ewes. Therefore, the number of ewes that contributed to the mean varies among times and does not exceed 3 for the fentanyl-treated group. Notice that the y-axis scale differs between the 2 panels. See Figures 1 and 4 for remainder of key.
Citation: American Journal of Veterinary Research 81, 7; 10.2460/ajvr.81.7.581
Discussion
The buprenorphine (0.01 mg/kg, IM, q 8 h for 48 hours beginning at T–1) and fentanyl (a combination of TFPs sufficient to deliver a dose of 2 μg of fentanyl/kg/h applied between the dorsal borders of the scapula at T–24) protocols administered to the healthy pregnant ewes of the present study both provided adequate perioperative analgesia, as evidenced by subjective assessment of animal comfort, movement, appetite, posture, and response to palpation of the surgical incision (ie, the total pain score), and resulted in plasma drug concentrations that exceeded the assumed minimum effective concentration necessary for clinical efficacy. No adverse effects were observed for any of the ewes. Compared with the buprenorphine protocol, the fentanyl protocol appeared to induce more profound analgesia (on the basis of assessment of the mean total pain scores for each treatment group) and decreased the requirement for isoflurane during the surgical procedure (ie, had a greater inhalant-sparing effect).
Pain is a multidimensional phenomenon that involves sensory and affective components, and an ideal pain assessment should evaluate those dimensions. A pain scoring system has yet to be validated for sheep. In the present study, the antinociceptive properties of the buprenorphine and fentanyl protocols administered were assessed by use of a composite pain scoring system that was based on scoring systems previously used in sheep and that fit the conditions of the study.4,6,7,21 The pain scoring system used for this study collectively evaluated indices of behavior, physiology, and productivity to detect and assess the intensity of pain. Use of a pain scoring system has limitations, such as interobserver variation and some degree of subjectivity in assigning scores. To minimize those factors, the ewes of the present study were acclimatized to the study environment, baseline pain scores were obtained, and the same observer assigned scores to all ewes throughout the observation period. Objective measures such as vital parameters and feed consumption were also measured.
Both analgesic protocols evaluated in the present study resulted in plasma drug concentrations that exceeded the minimum effective plasma concentration associated with analgesia reported in the human medical literature.15,22 To our knowledge, therapeutic plasma buprenorphine and fentanyl concentrations have not been established for sheep or other ruminant species. The fentanyl protocol appeared to induce more profound and consistent analgesia than did the buprenorphine protocol in the pregnant sheep of the present study. Similar findings have been reported in nonpregnant ewes.13 For the fentanyl-treated ewes of the present study, the mean total pain score did not differ from the baseline (presurgical) mean total pain score at any time during the postsurgical observation period. In contrast, for the buprenorphine-treated ewes, although the total pain scores were never high enough to warrant rescue analgesia during the postsurgical observation period, they were significantly greater than the baseline total pain score at multiple times. Additionally, the mean total pain score for the buprenorphine-treated ewes was significantly greater than that for the fentanyl-treated ewes at T4 (4 hours after anesthesia induction). That finding might have been a reflection of a greater analgesic effect associated with the fentanyl protocol or associated with a ceiling effect and the partial agonist activity of buprenorphine.
For the ewes of the present study, the standardized multidimensional (total) pain scores were negatively correlated with plasma drug concentration. That finding substantiates the pain scoring system used in this study as a method for detecting and assessing pain in sheep.
Cardiovascular variables during anesthesia for all ewes of the present study remained within reference ranges reported for anesthetized pregnant ewes of other studies,23,24 which supported our conclusion that the buprenorphine and fentanyl protocols administered provided adequate analgesia during the surgical procedure. Nevertheless, the mean HR, MAP, ETco2, and ETiso for the fentanyl-treated ewes were significantly lower than the corresponding mean values for the buprenorphine-treated ewes, which led us to conclude that the fentanyl protocol induced more profound analgesia. These results were in contrast to those of a similar study13 in which the intraoperative cardiorespiratory variables did not differ significantly between sheep that received buprenorphine (0.01 mg/kg, IM, q 6 h for 60 hours after surgery) and those that received fentanyl (a combination of TFPs sufficient to deliver a dosage of 2 μg of fentanyl/kg/h applied to the lateral aspect of the left antebrachium 12 hours before anesthesia induction) as analgesia for a tibial osteotomy procedure. Unfortunately, the plasma drug concentrations were not measured and reported for the sheep of that study.13 In the present study, the TFPs were applied to the fentanyl-treated ewes at T–24, and the Cmax for fentanyl was observed 26 hours after TFP application. Thus, it seems likely that the intraoperative plasma fentanyl concentrations for the ewes of the present study were higher than those for the sheep of the other study,13 which may explain the apparently conflicting results regarding differences in intraoperative cardiorespiratory variables between buprenorphine-treated and fentanyl-treated sheep for the 2 studies.
In the present study, the MAP of ewes was measured by an indirect method. It is well established that blood pressure measured by indirect methods can differ substantially from that measured by a direct method. However, we believed indirect measurement of MAP was acceptable in this study owing to the consistency of body position, anesthetic protocol, and surgical procedure among ewes and the fact that the same machine was used to monitor blood pressure for all ewes. Therefore, we believe that, although the indirect MAPs reported for the ewes of this study might not accurately reflect the actual (direct) MAP, differences in the mean MAP observed between buprenorphine-treated and fentanyl-treated ewes are valid because the measurement error should have been consistent for all ewes.
A constant rate infusion of fentanyl has a minimum alveolar concentration–sparing effect in isoflurane-anesthetized sheep.24 A similar phenomenon was observed for the ewes that received fentanyl by a transdermal patch in the present study (ie, the fentanyl-treated ewes required less isoflurane during surgery than did the buprenorphine-treated ewes). The amount of inhalant administered is positively associated with the time required to clear the drug from systemic circulation. Thus, decreasing the amount of inhalant administered results in a quicker recovery from anesthesia (ie, shorter durations between discontinuation of inhalant administration and extubation and between extubation and the animal attaining a standing position), which is generally beneficial to the patient. Four of the 6 fentanyl-treated ewes of the present study exhibited excitatory behavior during anesthesia recovery, which required light manual restraint; nevertheless, the recovery quality was considered clinically acceptable for all ewes. Excitement following TFP application and injection of fentanyl, buprenorphine, and other opioids has been reported in ruminants,1 and practitioners should be aware of this. The mechanism of that excitement is activation of opioid receptors in dopaminergic neurons, and administration of tranquilizers can be considered to counteract that effect.25
For the buprenorphine-treated ewes of the present study, therapeutic plasma concentrations of the drug were achieved within 2 hours after the initial injection and were maintained throughout the observation period. The variation observed in the Cmax and tmax for the buprenorphine-treated ewes was expected and consistent with IM administration of the drug to cats.26 We were unable to find other reports of plasma drug concentrations following IM injection of buprenorphine to sheep. However, when buprenorphine (6 μg/kg) was administered IV to non-pregnant sheep, the maximum antinociceptive effect was observed 45 minutes after administration and nociception returned to baseline levels at 240 minutes after administration, at which time the mean plasma buprenorphine concentration was 189 pg/mL.27 The lack of parallelism between that study27 and the present study precludes comparisons between the 2 studies. The same dose of buprenorphine (0.01 mg/kg) as that used in the present study was administered IM to nonpregnant sheep of another study,13 and although the investigators of that study did not measure plasma buprenorphine concentrations, the physiologic variables and clinical responses for the sheep suggested that therapeutic concentrations of the drug were achieved. In yet another study,28 a plasma buprenorphine concentration of 0.1 ng/mL was considered the minimum threshold for therapeutic benefit in sheep that received a sustained-release formulation of the drug.
Comparison of the tmax for a drug among studies is difficult, particularly when it is determined by noncompartmental analysis, because the tmax is based on observed plasma drug concentrations, which are dependent on the duration between drug administration and sample acquisition time. Unfortunately, we were unable to identify any parallel studies performed in sheep with which we could compare the tmax values for buprenorphine and fentanyl observed for the sheep of the present study.
The plasma buprenorphine concentration varied greatly within and among the 6 buprenorphine-treated ewes of the present study. It is currently unknown to what extent pregnancy affects the volume of distribution and thus the elimination rate of buprenorphine, but the pregnant state of the ewes of this study might have contributed to the observed variation in plasma buprenorphine concentration. We assumed that initiation of the buprenorphine protocol at T–1 would be sufficient to ensure that therapeutic plasma concentrations of the drug would be present during the surgical procedure. However, it is possible that the difference in isoflurane requirement between buprenorphine-treated and fentanyl-treated ewes was influenced by the rates of absorption and activity onset for buprenorphine. Also, plasma buprenorphine concentration might not be directly correlated with the antinociceptive effect. Therefore, buprenorphine dosing protocols are often based on observation of clinical effects.26
For the fentanyl-treated ewes of the present study, plasma fentanyl concentrations that exceeded concentrations associated with analgesia were achieved within 24 hours and were maintained for 72 hours after TFP application. Results of a human study29 indicate that a mean serum fentanyl concentration of 0.63 ng/mL (range, 0.23 to 1.18 ng/mL) was sufficient to achieve adequate postoperative analgesia in most opioid-naïve patients. In sheep that underwent multilevel lumbar intervertebral fusion, a minimum dose of 2 μg of fentanyl/kg/h and plasma fentanyl concentrations > 1 ng/mL were considered sufficient for postoperative analgesia even though the animals were not assessed with a standardized or formal pain scoring system and assessment for signs of pain was highly subjective.30 In an experimental study14 in which sheep underwent a unilateral tibial osteotomy, transdermal administration of fentanyl (mean dosage, 2.05 μg of fentanyl/kg/h for 72 hours) provided adequate perioperative analgesia as determined on the basis of pain scores; unfortunately, plasma fentanyl concentrations were not reported for the sheep of that study. In a study7 similar to the present study, a TFP was applied to the medial thigh region close to the groin area of 10 pregnant ewes. The mean dose of fentanyl administered was 1.4 ± 0.2 μg/kg/h, and a plasma fentanyl concentration of 0.5 ng/mL was consistent with analgesia, as determined by a standardized pain scoring system.7 In the present study, the lowest plasma fentanyl concentration measured in a fentanyl-treated ewe was 0.9 ng/mL, which was greater than the plasma concentration associated with analgesia in that other study.7 However, the dose of fentanyl transdermally administered to the ewes of the present study (2 μg/kg/h) was approximately 43% greater than that administered to the ewes of the other study.7
The observed tmax was 26 hours for the fentanyl-treated ewes of the present study. This finding supported the recommendation that a TFP be applied 12 to 36 hours before surgery.7,29 The highly lipophilic nature of fentanyl facilitates its absorption following transdermal administration. However, the reported distribution and plasma concentrations of fentanyl vary greatly following transdermal administration of the drug to sheep.7,12,14 The anatomic site of TFP application can affect drug absorption because permeability of the skin is affected by its composition, temperature, blood flow, and physical integrity.31 The stratum corneum of the epidermis is highly lipophilic and acts as a second depot for drug absorption. In sheep, the thickness of the stratum corneum ranges from 2 to 30 μm and is dependent on the hydration status of the animal, both of which can result in variation in drug absorption.15
In the present study, TFPs (or sham patches) were applied between the dorsal borders of the scapulae of each ewe because the temperature of the skin and pressure on the patch at that location would remain fairly constant during the observation period regardless of the animal's position. Also, patches applied at that location could not be easily manipulated by the sheep. Application of TFPs at the same anatomic site in nonpregnant sheep resulted in pharmacokinetic parameters similar to those reported in the present study.15 Pharmacokinetic parameters for fentanyl following TFP application to sheep at other anatomic locations vary considerably. For sheep of other studies7,12,14 in which the TFP was applied to the antebrachium, the drug absorption rate was faster, but the plasma fentanyl concentrations were lower, compared with those for the fentanyl-treated sheep of the present study. Differences in the absorption and elimination rates for fentanyl following application of TFPs at different anatomic locations are likely the result of differences in skin composition. In sheep, body fat is deposited along the back, and adipose tissue sequesters lipophilic drugs. That might be the reason absorption and elimination rates for fentanyl following TFP application between the dorsal borders of the scapulae were longer than those following TFP application to the antebrachium.
In the present study, the mean fetal plasma buprenorphine concentration remained fairly constant throughout the observation period without any obvious peak. The ratio of fetal plasma buprenorphine concentration to ewe (maternal) plasma buprenorphine concentration ranged from 2% to 8% in the present study, which is similar to that reported for human patients.32 The ratio of fetal plasma fentanyl concentration to maternal plasma fentanyl concentration ranged from 12% to 25% in this study, which was lower than that (50% to 80%) reported in other studies.7,11 However, the fetal plasma fentanyl concentrations observed in the present study were similar to those reported in those other studies.7,11 The lower ratios observed in this study relative to those other studies7,11 were likely a function of greater maternal plasma fentanyl concentrations rather than the lower maternal-to-fetal transfer of fentanyl.
The mean fetal HR, MAP, and arterial blood gas variables observed in the present study were similar to those reported for the fetuses of sevoflurane-anesthetized ewes of another study.33 Moreover, those variables did not differ significantly between fetuses of buprenorphine-treated ewes and fetuses of fentanyl-treated ewes, which suggested that the fetuses remained stable and viable throughout the experiment.33 The negative correlation between fetal HR and fetal plasma drug concentrations for both buprenorphine and fentanyl observed in the present study was consistent with findings reported in the human medical literature,34,35 which suggested that both analgesics have an effect on fetal cardiovascular function when administered to pregnant patients.
Drug availability, costs, and logistics of administration are important considerations when perioperative analgesic regimens are developed for veterinary patients. In the United States, buprenorphine (schedule III) and fentanyl (schedule II) are both classified as controlled substances. This indicates that both drugs are viewed as having an abuse potential, which can affect their availability. In the present study, the cost of the fentanyl protocol ($57.89/ewe) was approximately a quarter of the cost of the buprenorphine protocol ($241.47/ewe). In a clinical situation, the fentanyl protocol will require more skin preparation and bandaging materials for TFP application (ie, analgesic administration) than the buprenorphine protocol, but the animals have to be restrained for analgesic administration only once rather than multiple times, which should decrease animal stress.
Results of the present study indicated that buprenorphine (0.01 mg/kg, IM, q 8 h for 48 hours beginning at T–1) and fentanyl (a combination of TFPs sufficient to deliver a dose of 2 μg of fentanyl/kg/h applied between the dorsal borders of the scapulae at T–24) protocols both provided adequate perioperative analgesia to pregnant ewes. However, compared with the buprenorphine protocol, the fentanyl protocol appeared to provide more profound postoperative analgesia, decreased the requirement for isoflurane while the ewe was anesthetized, and was associated with a shorter anesthesia recovery time. Administration of the fentanyl protocol also required less animal restraint and was less expensive than the buprenorphine protocol. Both drugs were detected in the fetuses of the treated ewes and had similar effects on fetal HR and MAP. Additional research is necessary to further elucidate the pharmacokinetics of buprenorphine and fentanyl following administration to pregnant patients. Finally, the standardized multidimensional pain scores for the ewes of this study were negatively correlated with plasma drug concentrations, which suggested that the pain scoring system used is a reliable method for detecting and assessing pain in sheep.
Acknowledgments
Supported by the Michael E. DeBakey Institute, Department of Veterinary Physiology and Pharmacology, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University. No third-party funding or support was received in connection with this study or the writing or publication of the manuscript.
The authors thank Drs. Carly A. Patterson and Barbara Gastel for assistance with manuscript preparation.
ABBREVIATIONS
| AUC0–∞ | Area under the concentration-time curve from time 0 to infinity |
| AUC%extrap | Area under the concentration-time curve from the last measured time extrapolated to infinity and expressed as a percentage of the total area under the concentration-time curve |
| Cmax | Maximum serum concentration |
| ETco2 | End-tidal carbon dioxide concentration |
| ETiso | End-tidal isoflurane concentration |
| HR | Heart rate |
| λz | Terminal rate constant |
| MAP | Mean arterial pressure |
| t1/2λz | Terminal half-life |
| TFP | Transdermal fentanyl patch |
| tmax | Time to maximum concentration |
Footnotes
Research Randomizer, Urbaniak GC, Plous S. Available at: randomizer.org. Accessed Jan 15, 2017.
Par Pharmaceuticals Co, Spring Valley, NY.
Mylan Pharmaceuticals, Morgantown, WV.
Elastikon, Johnson & Johnson, New Brunswick, NJ.
Tape Pet Flex, Andover, Salisbury, Mass.
Stretch Net N84, Nich Marketers, Gulf Breeze, Fla.
Jorgensen, Loveland, Colo.
Akorn, Lake Forest, Ill.
Zetamine, VetOne, Boise, Idaho.
Fluriso, VetOne, Boise, Idaho.
Matrx Model 3000, Midmark, Orchard Park, NY.
Banamine, Merck Animal Health, Summit, NJ.
Draxxin, Zoetis, Parsippany, NJ.
Datascope Passport 2, Mindray, Mahwah, NJ.
Cardell Veterinary Monitor, 9401 BP, Midmark, Dayton, Ohio.
Datex Capnomac Ultima Monitor, Instrumentarium Corp, Helsinki, Finland.
iSTAT model 300A, Abbott Inc, Princeton, NJ.
PowerLab 8/30 model ML870, AD Instruments Inc, Colorado Springs, Colo.
Phoenix WinNonLin, version 8.0.0.3176, Certara, Princeton, NJ.
Prism 7, GraphPad Software Inc, La Jolla, Calif.
References
1. Pape D, Madry H. The preclinical sheep model of high tibial osteotomy relating basic science to the clinics: standards, techniques and pitfalls. Knee Surg Sports Traumatol Arthrosc 2013;21:228.–236.
2. Wilson SE, Cudd TA. Focus on: the use of animal models for the study of fetal alcohol spectrum disorders. Alcohol Res Health 2011;34:92–98.
3. Washburn SE, Tress U, Lunde ER, et al. The role of cortisol in chronic binge alcohol-induced cerebellar injury: ovine model. Alcohol 2013;47:53–61.
4. Musk GC, Murdoch FR, Tuke J, et al. Thermal and mechanical nociceptive threshold testing in pregnant sheep. Vet Anaesth Analg 2014;41:305–311.
5. Chavan NR, Ashford KB, Wiggins AT, et al. Buprenorphine for medication-assisted treatment of opioid use disorder in pregnancy: relationship to neonatal opioid withdrawal syndrome. AJP Rep 2017;7:e215–e222.
6. Stasiak KL, Maul D, French E, et al. Species-specific assessment of pain in laboratory animals. Contemp Top Lab Anim Sci 2003;42:13–20.
7. Musk GC, Catanchin CSM, Usuda H, et al. The uptake of transdermal fentanyl in a pregnant sheep model. Vet Anaesth Analg 2017;44:1382–1390.
8. Tigerstedt I, Tammisto T. Double-blind, multiple-dose comparison of buprenorphine and morphine in postoperative pain. Acta Anaesthesiol Scand 1980;24:462–468.
9. Mather LE. Clinical pharmacokinetics of fentanyl and its newer derivatives. Clin Pharmacokinet 1983;8:422–446.
10. Johnson RE, Fudala PJ, Payne R. Buprenorphine: considerations for pain management. J Pain Symptom Manage 2005;29:297–326.
11. Heikkinen EM, Kokki H, Heikkinen A, et al. Foetal fentanyl exposure and ion trapping after intravenous and transdermal administration to the ewe. Basic Clin Pharmacol Toxicol 2017;120:195–198.
12. Heikkinen EM, Voipio HM, Laaksonen S, et al. Fentanyl pharmacokinetics in pregnant sheep after intravenous and transdermal administration to the ewe. Basic Clin Pharmacol Toxicol 2015;117:156–163.
13. Ahern BJ, Soma LR, Boston RC, et al. Comparison of the analgesic properties of transdermally administered fentanyl and intramuscularly administered buprenorphine during and following experimental orthopedic surgery in sheep. Am J Vet Res 2009;70:418–422.
14. Ahern BJ, Soma LR, Rudy JA, et al. Pharmacokinetics of fentanyl administered transdermally and intravenously in sheep. Am J Vet Res 2010;71:1127–1132.
15. Jen KY, Dyson MC, Lester PA, et al. Pharmacokinetics of a transdermal fentanyl solution in Suffolk sheep (Ovis aries). J Am Assoc Lab Anim Sci 2017;56:550–557.
16. Washburn SE, Sawant OB, Lunde ER, et al. Acute alcohol exposure, acidemia or glutamine administration impacts amino acid homeostasis in ovine maternal and fetal plasma. Amino Acids 2013;45:543–554.
17. MacKay EE, Washburn KE, Padgett AL, et al. Pharmacokinetics of tulathromycin in fetal sheep and pregnant ewes. J Vet Pharmacol Ther 2019;42:373–379.
18. Aarnes TK, Hubbell JA, Lerche P, et al. Comparison of invasive and oscillometric blood pressure measurement techniques in anesthetized sheep, goats, and cattle. Vet Anaesth Analg 2014;41:174–185.
19. Otto KA, Steiner KH, Zailskas F, et al. Comparison of the postoperative analgesic effects of buprenorphine and piritramide following experimental orthopaedic surgery in sheep. J Exp Anim Sci 2000;41:133–143.
20. Riebold TW. Ruminants. In: Grim KA, Lamont LA, Tranquilli WJ, et al, eds. Veterinary anesthesia and analgesia: the fifth edition of Lumb and Jones. 5th ed. Ames, Iowa: John Wiley & Sons Inc, 2015;920–922.
21. Murdoch FR, Maker GL, Nitsos I, et al. Intraperitoneal medetomidine: a novel analgesic strategy for postoperative pain management in pregnant sheep. Lab Anim 2013;47:66–70.
22. Zullian C, Lema P, Lavoie M, et al. Plasma concentrations of buprenorphine following a single subcutaneous administration of a sustained release formulation of buprenorphine in sheep. Can J Vet Res 2016;80:250–253.
23. Mohamadnia AR, Hughes G, Clarke KW. Maintenance of anaesthesia in sheep with isoflurane, desflurane or sevoflurane. Vet Rec 2008;163:210–215.
24. Funes FJ, del Granados M, Morgaz J, et al. Anaesthetic and cardiorespiratory effects of a constant rate infusion of fentanyl in isoflurane-anaesthetized sheep. Vet Anaesth Analg 2015;42:157–164.
25. Smith JS, Mochel JP, Borts DJ, et al. Adverse reactions to fentanyl transdermal patches in calves: a preliminary clinical and pharmacokinetic study. Vet Anaesth Analg 2018;45:575–580.
26. Steagall PV, Pelligand L, Giordano T, et al. Pharmacokinetic and pharmacodynamic modelling of intravenous, intramuscular and subcutaneous buprenorphine in conscious cats. Vet Anaesth Analg 2013;40:83–95.
27. Nolan A, Livingston A, Waterman AE. Investigation of the antinociceptive activity of buprenorphine in sheep. Br J Pharmacol 1987;92:527–533.
28. Walkowiak KJ, Graham ML. Pharmacokinetics and antinociceptive activity of sustained-release buprenorphine in sheep. J Am Assoc Lab Anim Sci 2015;54:763–768.
29. Gourlay GK, Kowalski SR, Plummer JL, et al. Fentanyl blood concentration-analgesic response relationship in the treatment of postoperative pain. Anesth Analg 1988;67:329–337.
30. Christou C, Oliver RA, Rawlinson J, et al. Transdermal fentanyl and its use in ovine surgery. Res Vet Sci 2015;100:252–256.
31. Lane ME. The transdermal delivery of fentanyl. Eur J Pharm Biopharm 2013;84:449–455.
32. Nanovskaya T, Deshmukh S, Brooks M, et al. Transplacental transfer and metabolism of buprenorphine. J Pharmacol Exp Ther 2002;300:26–33.
33. Lépiz ML, Sayre R, Sawant O, et al. Maternal and fetal effects of dexmedetomidine infusion in pregnant ewes anesthetized with sevoflurane. Am J Vet Res 2017;78:1255–1263.
34. Ngamprasertwong P, Dong M, Niu J, et al. Propofol pharmacokinetics and estimation of fetal propofol exposure during mid-gestational fetal surgery: a maternal-fetal sheep model. PLoS One 2016;11:e0146563.
35. Gaiser RR, McHugh M, Cheek TG, et al. Predicting prolonged fetal heart rate deceleration following intrathecal fentanyl/bupivacaine. Int J Obstet Anesth 2005;14:208–211.
Appendix 1
Description of the standardized scoring system used to assess healthy pregnant ewes for signs of pain at predetermined times before and after a surgical procedure was performed during which the fetus was instrumented with an arterial catheter and a catheter was placed in utero for collection of amniotic fluid samples.
| Score | ||||
|---|---|---|---|---|
| Variable | 0 | 1 | 2 | 3 |
| Comfort | Awake and alert; interested in surroundings | Dull; little interest in surroundings; rises readily when approached | Depressed; not interested in surroundings; lethargic; ears drooped; not chewing cud, teeth grinding | Recumbent; no response when approached; fixed look (staring) or eyes half closed; little response when gently prodded |
| Movement | Standing or recumbent and readily rises when approached; full weight bearing | Does not get up promptly when approached but able to stand | Stands and walks with assistance | Recumbent and unable to stand |
| Palpation of wound or surgical site | Normal posture; no response to palpation of abdominal wound | Slight tucking of the abdomen; slight flinching of the skin and abdominal muscles and turning of head when the wound is gently palpated | Moderate tucking of the abdomen; moderate flinching of the skin and abdominal muscles and animal tries to walk away when the wound is gently palpated | Severe tucking of abdomen; abdominal muscles very tense and animal guards wound (kicks and attempts to walk away) |
| Appetite | Eating and drinking normally; ruminating with normal rumen sounds | Some decrease in food and water intake; ruminating | Minimal food and water intake; quiet and infrequent ruminations | Not eating or drinking and no ruminations |
| Posture when standing | Normal posture when standing and walking | Slightly abnormal posture when standing and walking | Extremely abnormal posture when standing and walking | Statue-like posture with little to no walking; obvious withdrawal from interactions with surroundings |
| Posture when in sternal recumbency | Alert; head up | Head down | Abnormal position with at least 1 limb partially extended | Abnormal position with at least 1 limb fully extended |
| Posture when in lateral recumbency | — | — | Head up | Head down |
At each assessment time, the scores for all variables were summed to calculate the total pain score; therefore, at any assessment time, the maximum possible total pain score was 15. Only 1 score was assigned for posture depending on the position of the ewe at the time the score was assigned.
— = Not defined.
Appendix 2
Description of scoring system used to assess the extent of sedation for the ewes described in Appendix 1.
| Score | Posture and demeanor |
|---|---|
| 0 | Standing, alert, normal behavior |
| 1 | Standing, alert, reduced head and ear movements |
| 2 | Standing, slight head drop |
| 3 | Standing, moderate head drop |
| 4 | Standing, severe head drop, ataxia |
| 5 | Standing, severe head drop, severe ataxia (stumbling) |
| 6 | Sternal recumbency, head up |
| 7 | Sternal recumbency, unable to support head |
| 8 | Lateral recumbency, occasional attempts to obtain sternal recumbency |
| 9 | Lateral recumbency, uncoordinated head and limb movements |
| 10 | Lateral recumbency, no movements |
