Dose and time-dependent physiological effects of fentanyl in adult goats

Ethics statement and animals used.

All experimental protocols were reviewed and approved by the Medical College of Wisconsin (MCW) Institutional Case and Use Committee (IACUC) prior to initiation of experiments. MCW has an Animal Welfare Assurance with OLAW (D16–00064; A3102–01). Adult female (≥2 years of age) of various breeds were supplied by a local outbred herd and transported to the lab per USDA regulations. Weights and venous blood were sampled at intake for full blood panels and pregnancy testing. Ivermectin (0.2 mg/kg; SQ) and fenbendazole (5 mg/kg; PO) were administered for deworming and conditioning upon arrival and repeated 2 weeks later. All goats were supplied with water, hay and sweet feed pellets daily with ad libitum access at all times other than presurgical fasting (24 hours) or during experiments. Goats were maintained in the laboratory setting in a temperature-controlled (24°C) chamber on a 12-hour on (6 AM – 6 PM), 12-hour off light cycle. Data included herein are from animals 95633, 101766, 101767, 112098, 114093, 115819, 126115, 126116, 133253, 135953, 142730, and 142729 (12 total).

Surgical Instrumentation

The surgical methodology used herein was similar to that reported previously¹³. Briefly, animals were anesthetized with ketamine (15 mg/kg, IV) for intubation and subsequently maintained on 2%–3% isoflurane in 100% O₂ via mechanical ventilation. Under aseptic conditions, each animal underwent bilateral subcutaneous relocation of the carotid arteries to later allow repeated blood sampling and continuous HR/BP monitoring as part of our experimental protocol. EMG electrodes were implanted into two airway muscles: 1) the thyroarytenoid, a laryngeal muscle controlling vocal fold adduction/closure, and 2) the hyoglossus, a lateral sublingual muscle involved in swallowing and tongue movement. EMG electrodes were also implanted into the diaphragm, intercostal, and transverse abdominus muscles to assess activation patterns of respiratory pump and chest wall muscles.

Postoperative health assessments, including rectal body temperature, surgical incision healing etc., were performed daily by trained staff. Goats were administered antibiotics (Naxcel, 4 mg/kg, IM) and anti-inflammatory medication Flunixin Meglumine (2 mg/kg, IM) for 7 days, 3 days post-op respectively. ≥ 2 weeks post-surgery. Goats underwent unilateral carotid catheterizations and Indwelling catheters were flushed daily with heparinized saline (0.1%) and Naxcel (4 mg/kg, IM) injections were given daily thereafter.

Studies began ≥2 weeks after surgery following a 1-week acclimation period to the experimental set up. During the acclimation period and subsequent experimental protocol, goats were in a harness and tethered by rope to a stanchion. Each goat was fitted with a custom ventilatory mask equipped with a one-way breathing valve. The inspiratory port was connected to a calibrated pneumotachometer, and the expiratory port to a Tissot gasometer (OxiGraf, Sunnyvale, CA, United States) where expired air (expired O₂ (F_EO₂) and CO₂ (F_ECO₂)) were collected across 3–5 min intervals (spirometer) and analyzed to calculate metabolic rate (VO₂), CO₂ excretion (VCO₂), the respiratory exchange ratio (VCO₂/VO₂). During each study, a rectal temperature probe documented body temperature (°C).

Experimental Protocol

Prior to each experiment, an intravenous catheter was inserted into the jugular vein for administration of saline (0.9% NaCl) or fentanyl (50, 150, 250, or ≥500 μg/kg. Physiological measurements were continuously recorded during a 30-min control period and for 90 min following intravenous administration of either saline or fentanyl. In some cases, no-injection control conditions were substituted for animals that exhibited shivering at the onset of saline injection. IV injections were administered over a 2-min duration.

Fentanyl was typically administered in randomized ascending dose order (50 to 750 μg/kg), with ≥48 h between studies to prevent cumulative effects of repeated fentanyl doses (see Table 1 in the companion manuscript). Doses were grouped into four categories: 50 μg/kg (n = 12), 150 μg/kg (n = 7), 250 μg/kg (n = 9), and ≥500 μg/kg (n = 8). All animals received the lowest fentanyl dose (50 μg/kg); however, not all animals received the maximal dose (750 μg/kg) due to inter-animal variability in fentanyl sensitivity or institutional animal care and use imposed dose limitation of the study. Sensitivity to fentanyl varied across animals. Naloxone was administered in emergent cases if an animal experienced apnea lasting ≥60 seconds, at which point AUA-mandated reversal criteria were met. In this cohort, three goats required emergent reversal with naloxone. Animals that received naloxone rescue were not advanced to higher fentanyl doses; thus, not all animals were exposed to every fentanyl dose.

Arterial blood gases (PaCO₂, PaO₂, O₂ saturation, and pH) were measured using a PRIME CCS blood gas analyzer (Nova Biomedical, Waltham, MA, United States). Samples were individually collected across a 30 second duration at predefined intervals every 15 min during the pre-injection baseline period, across 5 min immediately following injection (for a total of 7 samples), at 10- and 15-minutes post injection, and then every 15 min thereafter (Fig. 1A). Samples were corrected to barometric pressure and body temperature and used as an index of hypo- or hyperventilation. Upon completion of each study, blood samples were centrifuged (2,500 rpm for 15 min), and plasma was collected and frozen.

Heart rate (beats/min) and arterial blood pressure (mmHg) were measured using a calibrated pressure transducer. Systolic (peak) and diastolic (minimum) pressures were used to calculate mean arterial pressure (MAP; 1/3 systolic + 2/3 diastolic). All cardiovascular parameters were averaged over 5–15 min intervals, and transducer height was adjusted as needed to remain level with the heart if the animal changed posture (e.g., from standing to sternal position). The alveolar–arterial (A–a) O₂ gradient was calculated as PAO₂ − PaO₂, where PAO₂ = [0.209 × (barometric pressure − water vapor pressure)] − PaCO₂, following the method described by Meyer et al., 2010⁴². A–a gradient values were averaged over 5–15 min intervals.

Data were acquired and analyzed using Powerlab and LabChart software (ADInstruments, Colorado Springs, CO, United States). Recordings of inspiratory flow, HR/BP and respiratory muscle (EMG) activity were averaged in 30-second or 5-minute bins for baseline and post-injection periods (15 minutes each). Measures were collected over 1 s intervals and later averaged in either 30 s or 5 min bins for 15 min before and 15 min post-injection.

Diaphragmatic (DIA) EMG signals were digitally filtered, rectified, smoothed (400 ms), and integrated to obtain rate of rise (mV/sec) calculated using the “peak analysis” function in LabChart and binned in 1- or 15-minute intervals. EMG signals from hypoglossal, thyroarytenoid, intercostal, and abdominal muscles were high-pass filtered (180–190 Hz), rectified, smoothed (Bartlett window; 601 samples) and integrated to obtain mean integrated EMG activity on a breath-by-breath basis. Total Swallows, (identified as synchronous bursts of hyoglossus and thyroarytenoid activity), augmented breaths (sighs), and obstructive apneas (OA’s) were counted manually during offline review of EMG traces.

Following the 120-min experimental protocol, all instrumentation (including the mask, valve assembly, EMG leads, and catheter lines) were removed. Animals were fitted with a wearable BioHarness system (Biopac Systems Inc., Goleta, CA, United States) for measurement of 3 dimensional movement (vector magnitude; g) and transferred to a clear Plexiglas enclosure (3.5′ × 4′ × 6′) equipped with a high-definition video camera (Logitech Webcam C930e) to permit continuous recording of withdrawal-like behaviors including itching, biting, rearing and pawing (Fig. 1B). During this 4-h observation period, goats did not have access to food or water.

Upon the completion of all experimental protocols, animals were euthanized via intravenous administration of ketamine/xylazine (2.5 mL; 24:1), followed by phenytoin/phenobarbital (B-euthanasia; 97.5 mg/kg). Brains were harvested within 10 min of euthanasia, rapidly frozen using 2-methylbutane and dry ice, and stored at −80 °C for future analyses.

Statistical Analysis

All physiological measurements obtained following fentanyl exposure were compared to a Saline or no-injection control study. Shivering and/or artifact were removed prior to analysis. Data was analyzed using one- or two-way repeated-measures ANOVA, with mixed-effects models applied when data were missing due to technical limitations. Dunnett’s or other appropriate post hoc tests were performed as indicated (GraphPad Prism 9). Statistical significance was defined as p < 0.05 unless otherwise specified. Complete statistical details are provided in the Results section and corresponding figure legends.

Fentanyl dose-dependently increases the magnitude and duration of respiratory depression.

IV fentanyl led to a shift from a standing posture to laying and commonly resulted in obstructive or central apneas (a cessation of breathing for ≥2 respiratory cycles) during or immediately following fentanyl injection (Appendix 1). The duration of initial apnea increased as fentanyl dose increased (Appendix 1A), where apneas were mostly obstructive (55–88%) compared to central (13–40%) apnea (Appendix 1B). Fentanyl transiently reduced minute ventilation (V̇_I) ~20–50% (Fig. 2A), though at the lowest dose (50 μg/kg) V̇_I exceeded control values ~3 minutes (min) post-injection following a transient reduction. However, at the higher doses (150 – 500+ μg/kg), V̇_I remained below baseline up to 15 min post-injection (Fig. 2A) with the largest reduction (~50%) at the 500+ μg/kg dose. Breathing frequency (F_B; Fig. 2B) decreased by ≥40% at doses ranging from 250 to 500+ μg/kg with the greatest reduction at the 500+ μg/kg dose (~51% of control). Suppression of F_B persisted for ≥14 min post-injection across all doses (50 – 500+ μg/kg). In contrast to F_B, tidal volume (V_T) mostly increased after fentanyl injection for ≥10 min (Fig. 2C), with the greatest increase in V_T was observed at the lowest test dose (50 μg/kg) similar to previous reports¹³.

Fentanyl elicits dose-dependent hypoventilation and sustained hypoxemia via impaired pulmonary gas exchange

Each fentanyl dose tested induced some degree of hypoxemia and hypercapnia, with the largest changes in arterial blood gases occurring at the highest fentanyl doses (Fig. 3). However, the duration of the changes in arterial O₂ and CO₂ showed a time-dependent divergence. Mild hypercapnia (+3.7–3.9 mmHg) was present with lower doses (50–150 μg/kg) of fentanyl, with more pronounced hypoventilation (+7–10 mmHg) at higher doses (250–500+ μg/kg; Fig. 3A). Concomitant decreases in pH (Fig. 3C) were consistently evident during periods of hypoventilation. Except for the highest fentanyl dose, PaCO₂ returned to baseline by 30 min post-injection. At the lower fentanyl doses PaCO₂ was transiently decreased and pH transiently increased indicating periods of hyperventilation-induced hypocapnia (Fig. 3A). Fentanyl elicited time- and dose-dependent reductions in PaO₂ (Fig. 3B) and O₂ saturation (Fig. 3D). Hypoxemia was evident within 2 min post-injection, where the largest reductions in PaO₂ and O₂ saturation occurred during the period of hypoventilation (3–10 min post-injection). Although PaO₂ appeared to partially normalize 15 min post-injection, thereafter there was sustained hypoxemia lasting 30–90 min post-injection where the largest doses had the greatest effects (Fig. 3B,D). Similarly, the change in the Alveolar to arterial (A–a) O₂ gradient was calculated relative to baseline (ΔA-a O₂ gradient) and compared across all doses tested (Fig. 4). 500+ μg/kg fentanyl elicited the largest increase in the A-a O₂ gradient (+13 mmHg) within the immediate post-injection period. However, 250–500+ μg/kg fentanyl led to the greatest overall increase (~24 mmHg; Fig. 4), which was sustained at least 90 min post-injection. These data indicate that high doses of fentanyl elicit an initial hypoventilation but sustained hypoxemia likely due to impaired pulmonary gas exchange.

Fentanyl consistently reduced heart rate (HR) and increased mean arterial blood pressure (MAP) without affecting metabolic rate and body temperature¹³. Higher doses of IV fentanyl tended to reduce HR (Fig. 5A) but consistently increased MAP (Fig. 5B) across all doses. The somewhat ambiguous effect of fentanyl on HR likely resulted from a sub-population of goats that showed increased HRs in response to fentanyl and another with decreased HR consistent with variable responses to fentanyl in this species¹³. , High doses of IV fentanyl exerted minimal effects on oxygen consumption (ΔV̇O₂; Fig. 6A) and rates of CO₂ production (V̇CO₂; data not shown). Body temperature increased progressively throughout study periods during saline control studies and the 50–150 μg/kg fentanyl studies. In contrast, fentanyl at the highest doses (250–500+ μg/kg) attenuated this temperature increase (Fig. 6B).

Fentanyl induces airway and pump muscle activation and inhibits swallowing.

A consequence of rapid IV fentanyl injection is fentanyl-induced respiratory muscle rigidity (FIRMR) which is also known as Wood Chest Syndrome (WCS). Similar to previous data in goats¹³, high doses of IV fentanyl increased integrated EMG activity (μV; % control) across most of the muscles assayed (Figs. 7 + 8). Consistent with the effects of high dose fentanyl to increase V_T, the rate of rise of integrated diaphragm (DIA) EMG activity was increased after fentanyl (compared to saline) throughout the post-injection period with the largest and most sustained increases observed at the 150–250 μg/kg dose without differences across fentanyl dose (P=0.2708; Fig. 7). IV fentanyl also increased integrated thyroarytenoid (laryngeal constrictor), intercostal and abdominal muscle activity, with the magnitude and duration varying by muscle group (Fig. 8A-D). Typical rhythmic activity was observed in the intercostal and abdominal muscles prior to fentanyl, but thereafter fentanyl evoked a rapid and robust increase in tonic integrated thyroarytenoid (Fig. 8B), intercostal (Fig. 8C) and transversus abdominus (Fig. 8D) EMG activity immediately post-injection reaching 200–300% of control (baseline) at the highest doses. Thyroarytenoid and intercostal EMG activity was typically greater after the highest doses of fentanyl (250 and 500+ μg/kg) compared to 50 μg/kg fentanyl, where the 500+ μg/kg fentanyl dose often had greater effects compared to all other doses (see Fig. 8 legend). However, integrated hyoglossus muscle activity was unchanged relative to baseline or IV saline conditions (data not shown). Under saline conditions, swallow frequency (Fig. 9A) remained stable across the duration of the study (~30–35 swallows/bin). However, following fentanyl administration across all doses, there was a significant reduction in swallowing (≤10 swallows/time bin) which was sustained at least 90 min post-injection. In contrast, fentanyl had no consistent effect on the occurrence of augmented breaths (sighs; Fig. 9B) or in the occurrence of obstructive apneas across all doses and the full study period (Fig. 9C).

Dose-dependent acute withdrawal-like behaviors with high doses of fentanyl

There were dose-dependent increases in cumulative locomotor activity ≥90 min post-injection (Fig. 10A). Quantification of withdrawal-like behavior via cumulative movement (acceleration in any direction; g) and total behavioral response (sum total of bouts of itching, biting, rearing and pawing; Fig. 10B) were time- and dose-dependent. When compared to saline treated animals, fentanyl at all doses produced larger cumulative movement and total behavioral responses, with the largest effects seen at the highest doses (Fig. 10A-B). Thus, acute withdrawal-like behaviors were evident and dose-dependent.